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
|
|---|---|---|---|---|---|
Staphylococcus aureus is a microorganism that causes serious diseases in the human being . This microorganism is able to escape the phagolysosomal pathway , increasing intracellular bacterial survival and killing the eukaryotic host cell to spread the infection . One of the key features of S . aureus infection is the production of a series of virulence factors , including secreted enzymes and toxins . We have shown that the pore-forming toxin α-hemolysin ( Hla ) is the S . aureus–secreted factor responsible for the activation of the autophagic pathway and that this response occurs through a PI3K/Beclin1-independent form . In the present report we demonstrate that cAMP has a key role in the regulation of this autophagic response . Our results indicate that cAMP is able to inhibit the autophagy induced by Hla and that PKA , the classical cAMP effector , does not participate in this regulation . We present evidence that EPAC and Rap2b , through calpain activation , are the proteins involved in the regulation of Hla-induced autophagy . Similar results were obtained in cells infected with different S . aureus strains . Interestingly , in this report we show , for the first time to our knowledge , that both EPAC and Rap2b are recruited to the S . aureus–containing phagosome . We believe that our findings have important implications in understanding innate immune processes involved in intracellular pathogen invasion of the host cell .
Autophagy is a cellular process in response to stress , which is activated when cells are subjected to nutrient limitation , high temperatures , oxidative stress , accumulation of damaged organelles , or infection with certain pathogens [1] . When autophagy is activated , various cellular constituents , including long-lived proteins , cytoplasmic organelles , and some microorganisms , are encapsulated by the phagophore , a growing cistern that finally closes generating the autophagosome lined by two membranes . These vesicles intersect with the endosomal compartment , generating the amphisome , which finally fuses with lysosomes to become autolysosomes , where sequestered cellular components are digested and essential molecules are recycled back to the cytoplasm [2] . Genetic studies in yeast have led to the discovery of several Atg ( autophagy related ) genes , many of which have mammalian orthologs [3] . Atg12-Atg5 and the Atg8 systems are key components of the autophagic pathway . Atg5 interacts covalently with Atg12 and noncovalently with the multimeric protein Atg16 . The microtubule-associated protein 1 light chain 3 ( MAP1-LC3/Atg8/LC3 ) is cleaved at its C terminus by Atg4 to form LC3-I , which is covalently conjugated to phosphatidylethanolamine to generate LC3-II . LC3-II is formed where the Atg12-Atg5-Atg16 complex is localized and it remains associated with autophagosomes , even with mature autophagosomes/autolysosomes although at a lesser degree [4] , [5] . Two main mechanisms involved in the regulation of the classical autophagy pathway have been described . One of them involves the serine/threonine kinase , mammalian target of rapamycin ( mTOR ) , which inhibits autophagy and functions as a sensor for cellular energy and amino acid levels [3] , [6] . The other one is through phosphatidylinositol-3-kinase ( PI3K ) Class III , which plays an important role in the activation of the autophagic pathway , acting as a positive regulator . Class III PI3K and its human ortholog hVps34 interact with Beclin 1 and p150 myristoylated kinase , activating some of the Atg proteins involved in the autophagic pathway [7] . More recently , a new kind of autophagic pathway independent of mTOR and rapamycin has been revealed [8] . Rubinsztein and coworkers demonstrated that autophagy can be induced by lowering intracellular inositol or inositol 1 , 4 , 5-trisphosphate ( IP3 ) levels , in a mTOR-independent form [8] , [9] . Consistently , Kroemer and collaborators have shown that genetic knockdown or pharmacological inhibition of the IP3 receptor ( IP3R ) induces autophagy [10] . Interestingly , it has been recently shown that IP3R represses autophagy through Bcl-2-mediated sequestration of Beclin 1 [11] , thus linking IP3R with initial steps of the autophagic pathway . Cumulative evidence indicates that autophagy is involved in the defense against several pathogen microorganisms [1] , [12] , [13] . Upon autophagy induction , intracellular bacteria such as Streptococcus pyogenes , Mycobacterium tuberculosis , and Salmonella are sequestrated within autophagosomes which then fuse with lysosomes to eliminate the intruder [13] . However , some pathogens like Coxiella burnetti and Legionella pneumophila benefit from autophagy and generate a replicative niche with autophagic features where the bacteria can actively replicate [12] . Other bacteria like Shigella flexneri and Listeria monocytogenes can escape from the phagosomes into the cytoplasm , where they multiply and generate actin tails to disseminate from the host cell to neighboring cells [12] . Staphylococcus aureus is a microorganism that causes serious diseases in humans . S . aureus has been classically considered an extracellular pathogen , but numerous studies have now shown that S . aureus can infect various types of non-professional phagocytic cells such as keratinocytes , fibroblasts , endothelial , and epithelial cells , leading to host cell death [14]–[17] S . aureus is able to escape the phagolysosomal pathway and , in this way , increase the intracellular bacterial survival and killing of the eukaryotic host cell , spreading the infection [18] . A previous study has shown that after the infection , S . aureus localizes to autophagosomes and inhibits its maturation and fusion with lysosomes [18] . Interestingly , S . aureus was not able to replicate in autophagy-deficient Atg5−/− mouse embryonic fibroblasts , indicating that S . aureus requires autophagy activation for replication , subsequent escape from autophagosomes into the cytoplasm , and S . aureus–induced host cell death [18] . We have previously demonstrated that α-hemolysin ( Hla ) , which is a pore-forming toxin secreted by S . aureus [19] , [20] , is the factor responsible for the autophagic response induced by this pathogen in the host cell [21] . In addition , we have shown that this autophagic response induced by Hla occurs via a “non canonical" pathway , which is independent of Beclin1 and PI3K , but requires Atg5 [21] . Also , we have demonstrated that the LC3-labeled vesicles generated by the toxin are non-acidic and non-degradative compartments , indicating that somehow the toxin impedes the maturation of these autophagic structures [21] . In the present study , we have analyzed the signaling mechanisms that regulate the Hla-induced autophagy . We have identified some of the molecular components involved in this pathway and determined their participation in the regulation of the autophagic response generated by the toxin .
We have previously shown that the autophagic response induced by the toxin α-hemolysin ( Hla ) was not suppressed by the classical autophagy inhibitors 3-mehtyladenine or wortmannin , suggesting that this process occurs independently of PI3Kinase activation [21] . Thus , we were interested in determining whether other pathways might be involved in the regulation of this “non-canonical" autophagic response . In a recent publication , it was shown that cAMP-dependent protein kinase A ( PKA ) activation by cAMP is able to inhibit the autophagy pathway through LC3 phosphorylation [22] ( Figure S1A ) . In order to determine if this pathway regulates the autophagic activation induced by alpha-toxin , we analyzed this process in stable transfected CHO cells overexpressing GFP-LC3 . The protein LC3 is an autophagic marker present in eukaryotic cells as a soluble form ( LC3-I ) and a membrane-associated form ( LC3-II ) . When autophagy is activated , LC3-I is conjugated to phosphatidylethanolamine to generate LC3-II , which localizes to autophagosomal membranes [5] . CHO cells overexpressing GFP-LC3 were incubated in complete medium , with or without the toxin or subjected to starvation conditions , in the absence or presence of dbcAMP , a permeable cyclic AMP ( cAMP ) analog . As shown in Figure 1A , cAMP caused a marked inhibition in the autophagic response induced by the toxin , as indicated by a decrease in LC3-positive vesicles ( Panel f ) . In contrast , and to our surprise , just a little decrease in starvation-induced autophagy was observed ( Panel e ) . The quantification of the percentage of cells presenting LC3 puncta upon incubation in the different conditions is shown in the right panel . Next , we performed a Western blot assay to analyze the processing of LC3 . CHO GFP-LC3 cells were incubated in complete medium , with or without the toxin , or subjected to starvation , with or without dbcAMP and in the presence ( Figure 1B , lower panel ) or the absence ( Figure 1B , upper panel ) of bafilomycin A1 , an inhibitor of the H+ ATPase pump and autophagosome/lysosome fusion . In agreement with the results mentioned above , a decreased level of lipidated LC3 protein was detected in cells incubated with Hla in the presence of dbcAMP , even with bafilomycin A1 ( Figure 1B ) . A quantification of the intensity of the bands is shown in Figure 1C . To corroborate these results , endogenous LC3 was also detected after the different conditions and similar results were obtained ( Figure S2A ) . In order to address whether PKA participates in cAMP-dependent inhibition of the toxin response , we incubated the cells with H89 , a PKA inhibitor , in the presence or absence of dbcAMP ( Figure 2A ) . As shown in Figure 2A , H89 was unable to revert the autophagy inhibition induced by dbcAMP in cells treated with the toxin ( Panel c ) . In addition , we overexpressed a dominant-negative PKA regulatory subunit mutated in both sites A and B of the cAMP-binding domain [23] . This dominant negative mutant of the regulatory subunit of PKA cannot be activated by cAMP . Similar to the results obtained with H89 , overexpression of this PKA inactive mutant was unable to antagonize the inhibitory effect of dbcAMP in the autophagic response induced by alpha toxin ( data not shown ) . These results indicate that PKA is not participating in the cAMP-dependent inhibition of Hla-induced autophagy . In addition , H89 alone did not affect the autophagic response in either conditions ( Figure 2A , Panels e and f ) . The quantification of the percentage of cells presenting LC3 puncta upon treatment with the different conditions is depicted in Figure 2B . To verify the activity of H89 , we analyzed the phosphorylation of CREB , which is a PKA substrate , in cells stimulated with cAMP in the presence or absence of H89 . As shown in Figure 2C , H89 decreased the phosphorylation of CREB , confirming that the PKA inhibitor is active in our system . These results indicate that cAMP is acting via a PKA-independent pathway . cAMP has traditionally been thought to act exclusively through PKA , but at present , cAMP is also known to directly regulate ion channels and the ubiquitous protein EPAC ( exchange protein activated by cAMP ) , a cAMP-regulated effector that is a guanine nucleotide exchange factor ( GEF ) for the low molecular weight GTPase , Rap [24] ( Figure S1A ) . 8-pCPT-2′-O-Me-cAMP ( 8-pCPT-cAMP ) is a cAMP analog that specifically activates EPAC . Recently published studies demonstrate that 8-pCPT-cAMP is a useful tool to assess atypical actions of cAMP that are PKA-independent [25] . Thus , we next assessed the effect of 8-pCPT-cAMP on the autophagic response induced either by the toxin or by starvation . Similar to dbcAMP , 8-pCPT-cAMP was able to abolish the autophagic response upon α-hemolysin-treatment but did not substantially affect starvation-induced autophagy ( Figure 3A ) . Thus , cAMP-induced EPAC activation seems to be sufficient to inhibit the autophagic response generated by Hla . To ascertain the participation of the Rap exchange factor EPAC in the Hla-induced autophagic pathway , CHO cells were cotransfected with GFP-LC3 and Myc-EPAC wt or the Myc-ΔEPAC mutant , a constitutively active GEF mutant that maximally activates Rap even in the absence of cAMP stimulation [26] . Subsequently , the cells were incubated in complete medium ( with or without 10 µg/ml of α-hemolysin ) or under starvation conditions , a physiological inducer of autophagy . As shown in Figure 3B , overexpression of either EPAC wt ( left panels ) or its active mutant ( ΔEPAC , right panels ) was able to inhibit the Hla-induced autophagic response , but had no inhibitory effect in the autophagy induced by starvation . The quantification of the percentage of cells presenting LC3 puncta upon incubation in the different conditions is shown in Figure 3C . As control , CHO cells were cotransfected with GFP-LC3 and RFP-vector and treated as described above . As expected , both autophagy inducers ( i . e . , starvation and rapamycin ) , as well as treatment with the toxin , caused the typical LC3 punctate distribution ( Figure S3A and S3B ) . Thus , taken together , these results indicate that EPAC regulates the autophagy induced by the toxin and when activated is able to inhibit this autophagic response . In order to corroborate whether the EPAC pathway is responsible for the regulation of autophagy activation induced by Hla , we analyzed a downstream component of this pathway , the small GTPase Rap2b . For this purpose , CHO cells were cotransfected with RFP-LC3 and GFP-Rap2b wt or the GFP-Rap2b ΔAAX mutant , which cannot be lipidated because it has a deletion in its C-terminal motif , losing both membrane localization and activity . It has been shown that deletion of the CAAX motif abolishes plasma membrane localization and compromises the biological function of many Ras-related GTPases [27] . The transfected cells were incubated in complete medium ( in the presence or the absence of 10 µg/ml of α-hemolysin ) , under starvation conditions , or with 50 ng/µl of rapamycin , a pharmacological inducer of autophagy ( Figure 4A ) . As control , CHO cells were cotransfected with RFP-LC3 and GFP-vector and treated as described above . As expected , both autophagy inducers ( i . e . , starvation and rapamycin ) , as well as treatment with the toxin , caused the typical LC3 punctate distribution ( Figure S3C and S3D ) . As shown in Figure 4A , overexpression of Rap2b wt was able to inhibit the autophagic response induced by the toxin , but had no effect in the autophagy activated by starvation or rapamycin . In contrast , overexpression of the mutant Rap2b ΔAAX did not affect toxin-induced autophagy , indicating that Rap2b participates in the pathway that regulates the autophagy induced by α-hemolysin preventing this autophagic response . Of note , overexpression of Rap2b ΔAAX decreased the autophagy induced by starvation and rapamycin , suggesting that Rap2b might be a common link between both the “non-canonical" autophagy pathway induced by the toxin and the classical autophagic pathway induced by starvation . The quantification of the percentage of cells presenting LC3 puncta subjected to different treatments is shown in Figure 4B . To corroborate these results , we performed a Western blot assay to analyze the processing of LC3 . CHO cells were transfected with Rap2b wt or Rap2b ΔAAX and incubated in complete medium ( with or without the toxin ) with rapamycin or subjected to starvation conditions in the presence or absence of bafilomycin A1 . In agreement with the results mentioned above , a decreased level of lipidated LC3 protein was detected in cells overexpressing Rap2b wt and incubated with Hla , whereas no effect was observed in cells overexpressing the inactive mutant Rap2b ΔAAX ( Figure 4C ) . A quantification of the bands intensities is shown in Figure 4D . To determine whether depletion of Rap2b affects the starvation- or Hla-iduced autophagy , we used a Rap2b siRNA . HeLa cells were cotransfected with GFP-LC3 and Rap2b siRNA or with an irrelevant siRNA , and then they were incubated in complete medium ( in the presence or the absence of 10 µg/ml of α-hemolysin ) with rapamycin or under starvation conditions ( Figure 5A ) . As shown in Figure 5A , knockdown of Rap2b was able to decrease the autophagic response induced by starvation and rapamycin , but did not affect the autophagy activated by the toxin . The quantification of the percentage of cells presenting LC3 puncta subjected to the different treatments is shown in Figure 5B . The effective knockdown of Rap2b was determined by Western blot as shown in Figure 5C . To corroborate these results , we performed a Western blot assay to analyze the processing of LC3 . HeLa cells were transfected with Rap2b siRNA or irrelevant siRNA and subjected to the different treatments as indicated above . As shown in Figure 5D , a decreased level of lipidated LC3 protein was detected in cells transfected with Rap2b siRNA and incubated with rapamycin or under starvation conditions , but the levels were not affected in cells treated with the toxin . A quantification of the intensity of the bands is shown in Figure 5E . Next , we were interested in addressing whether cAMP was able to inhibit the Hla-induced autophagy even in cells overexpressing Rap2b ΔAAX . Thus , CHO cells were cotransfected with RFP-LC3 and GFP-Rap2b ΔAAX , and then they were treated with Hla or starvation medium in the presence or absence of cAMP ( Figure 6A ) . Our results indicate that cAMP was unable to inhibit the autophagy induced by the toxin in cells overexpressing the Rap2b negative mutant , indicating that the inhibitory effect of cAMP requires a functional Rap2B ( Figure 6A ) . CHO cells cotransfected with RFP-LC3 and GFP-vector and treated as above displayed the expected autophagic response induced by Hla or by starvation ( data not shown ) . The quantification of the percentage of cells presenting LC3 puncta upon incubation in the different conditions is shown in Figure 6B . Taken together , these results clearly indicate that Rap2b is a key participant in the regulation of autophagy induced by α-hemolysin , and it is likely that this small GTPase needs to be inactivated to allow this autophagic response . Rap2b is known to produce an increase of cytoplasmic calcium through a rise in IP3 [28] . Calpains are a family of cysteine-proteases that are activated by intracellular calcium . When calpains are activated , they are able to cleave Atg5 , inhibiting autophagy in basal conditions [29] . In order to demonstrate the participation of calpains in this pathway we used calpeptin , a calpains inhibitor . CHO cells overexpressing GFP-LC3 were incubated with dbcAMP in the presence or the absence of calpeptin . Our results indicate that while cAMP alone was able to inhibit the Hla-induced autophagic response , calpeptin-preincubation prevented its inhibitory effect ( Figure 7A , Panels k and l ) . These results suggest that during the autophagic response induced by Hla calpains are inhibited . The quantification of the percentage of cells presenting LC3 puncta upon treatment with the different conditions is shown in Figure 7B . In addition , the processing of LC3 was analyzed by Western blot . CHO cells were preincubated with dbcAMP , calpeptin , or calpeptin+dbcAMP , and then they were incubated in complete medium with or without the toxin . In agreement with the results mentioned above , calpeptin-preincubation prevented the inhibitory effect of dbcAMP in the Hla-induced autophagy , whereas a decreased level of lipidated LC3 protein was detected in cells preincubated with dbcAMP alone and treated with Hla ( Figure 7C ) . A quantification of the bands' intensities is shown in Figure 7C ( lower panel ) . Thus , these results suggest that the activation of calpains by cAMP leads to the inhibition of the Hla-induced autophagy . Therefore , calpains might act as negative regulators of this particular form of toxin-induced autophagic response . We have previously shown that a population of internalized S . aureus recruits GFP-LC3 to their containing phagosomes and that this recruitment was dependent on the production of α-hemolysin [21] . To corroborate the participation of the above pathway in the regulation of autophagy induced by the toxin , we used different S . aureus strains: a wild-type strain ( wt ) , a mutant deficient for α-hemolysin ( Hla− ) , and the Hla ( − ) mutant complemented with an α-hemolysin plasmid ( Hla ( − ) +pHla ) . GFP-LC3 CHO cells were preincubated with dbcAMP and infected for 4 h with the different S . aureus strains . Cells were then incubated with TOPRO , a DNA marker , to label the bacteria . As shown in Figure 8A , right panels , cAMP caused inhibition of autophagy induced either by S . aureus wt or by the complemented Hla ( − ) mutant . As control , autophagy activation by the wt and the complemented Hla ( − ) mutant strains in the absence of dbcAMP is also depicted ( Figure 8A , left panels ) . The quantification of the percentage of infected cells presenting LC3 puncta upon treatment with or without dbcAMP is shown in Figure 8B . We have previously shown that autophagy inhibition decreases bacterial replication [21] , so next we determined whether cAMP affects intracellular bacterial grown . For this purpose , CHO cells were infected for 3 h with the wt strain of S . aureus , the mutant deficient for α-hemolysin ( Hla− ) , or the Hla ( − ) mutant expressing an α-hemolysin plasmid . After the infection , cells were lysed and plated for colony forming units ( CFU ) quantification ( Figure 8C ) . Interestingly , a marked decrease in bacterial replication was observed after treatment with cAMP , corroborating that autophagy is necessary for efficient bacterial replication and demonstrating that elevated levels of cAMP negatively affects S . aureus intracellular growth . In order to corroborate the participation of EPAC and Rap2b in this autophagy regulation , CHO cells were cotransfected with GFP-LC3 and myc-EPAC wt or myc-ΔEPAC ( Figure 9A ) ; or with RFP-LC3 and GFP-Rap2b wt or GFP-Rap2b ΔAAX ( Figure 10A ) . Cells were then incubated with TOPRO , to label the bacteria . As shown in Figure 9A , overexpression of EPAC wt ( upper panels ) or its active mutant ΔEPAC ( lower panels ) was sufficient to suppress the autophagic response induced by S . aureus wt and the complemented Hla ( − ) mutant ( Hla ( − ) +pHla ) . The quantification of the percentage of infected cells overexpresing EPAC presenting LC3 puncta is shown in Figure 9B . As control , CHO cells were cotransfected with GFP-LC3 and RFP-vector and infected as described above . As expected , both the wt strain and the complemented Hla ( − ) mutant caused the typical LC3 recruitment ( Figure S4 ) . Consistently , overexpression of Rap2b was also able to inhibit the autophagy induced by the bacteria ( Figure 10A , upper panels ) . In contrast , overexpression of the negative mutant Rap2b ΔAAX had not effect in the autophagic response induced by S . aureus ( Figure 10A , lower panels ) . The quantification of the percentage of infected cells overexpresing Rap2b presenting LC3 puncta is shown in Figure 10B . Strikingly , a remarkable recruitment of both EPAC and Rap2b to the bacteria-containing phagosome was observed ( insets in Figures 9A and 10A ) . The quantification of the percentage of bacteria decorated with EPAC or Rap2b upon the infection is shown in Figure 9C and 10C . These data indicate that approximately 40%–50% of the bacteria-containing phagosomes showed association with either EPAC or Rap2B . The association of both proteins to the S . aureus phagosomal compartment was only partly decreased in phagosomes containing the Hla-deficient strain . For this reason , we determined the localization of endogenous EPAC or Rap2b . CHO cells were infected with the different S . aureus strains and EPAC or Rap2b were detected by indirect immunofluorescense ( Figure S5 ) . Interestingly , we observed that the Hla-deficient strain was unable to recruit either EPAC or Rap2b , suggesting that the partial recruitment of the overexpressed EPAC and Rap2b by the Hla ( − ) strain is likely due to the excess of molecules present in the transfected cells . Next , we analyzed the colocalization between LC3 and endogenous EPAC or Rap2b . For this purpose , CHO GFP-LC3 cells were infected as above and the proteins were detected by indirect immunofluorescense ( Figure S6 ) . Interestingly , neither EPAC nor Rap2b colocalized with LC3 , suggesting that the population of bacteria that recruit EPAC/Rap2b does not recruit LC3 or that both proteins are differentially recruited on time . Taken together , these results clearly confirm that the identified molecular components involved in the autophagic pathway induced by the purified toxin also participate in the autophagic response upon infection with S . aureus , negatively regulating the autophagic response exerted by this pathogen . The results shown above clearly indicate that cAMP and Rap2b are negative regulators of the autophagic activation induced by the toxin and by the infection with S . aureus . In order to determine how the bacteria and the toxin are able to regulate this autophagic pathway , we analyzed the levels of active Rap2b present in the cells . For this purpose , HeLa cells were incubated with the toxin or infected for 4 h with S . aureus wt strain or the mutant deficient in Hla . Then , the cells were lysed and GTP-bound Rap2b was pulled down with immobilized Ral-GDS-RBD , a cassette that is able to bind GTP-Rap proteins [30] . The amount of protein pulled down was determined by Western blot assay , using a polyclonal Rap2b antibody . As shown in Figure 11B , the levels of active Rap2b decreased in response to Hla treatment or when the cells were infected with S . aureus wt strain . However , no differences in the total amount of cellular Rap2b in the cells subjected to the different conditions were observed ( Figure 11A ) . These results suggest that the toxin induces a decrease in the amount of active Rap2b , likely by decreasing the levels of intracellular cAMP , to allow autophagy induction , a response that favors pathogen intracellular survival as previously demonstrated . A model indicating the two cAMP-dependent pathways involved in the control of autophagy and the effect of different molecules used in this study on the autophagic response induced by Hla and by S . aureus is shown in Figure S1B .
In previous studies we have demonstrated that S . aureus induces autophagy activation in infected cells [21] . We presented evidence indicating that S . aureus uses the α-hemolysin to activate the autophagic response , generating LC3-positive vesicles that are unable to mature , interrupting the autophagic flux [21] . These autophagic vesicles , which are non-acidic and non-degradative compartments , are likely used by the bacteria as a replicative niche , escaping then toward the cytoplasm to subsequently infect neighboring cells . We have also shown that the autophagic response induced by Hla does not occur by the classical pathway of autophagy activation . The toxin uses an alternative form to induce autophagy , which is independent of the PI3K/Beclin1 complex but dependent on the autophagic protein Atg5 [21] . In the present study , we have shown that cAMP plays a key role in the Hla-induced autophagic response . Indeed , our results indicate that this response is strongly suppressed by cAMP treatment . cAMP is a classical PKA activator that participates in several cellular process [31] . Recently , it has been demonstrated that PKA participates in autophagy regulation induced by the rapamycin-mediated inactivation of the TOR pathway [22] . In that report , Charleen Chu and coworkers have shown that , upon activation by cAMP , PKA is able to phosphorylate LC3 . This phosphorylation in LC3 prevents autophagy activation , suggesting cAMP as a possible regulator of the autophagic pathway induced by rapamycin [22] . In addition , it has been also shown in the budding yeast S . cerevisiae that elevated levels of Ras/PKA activity prevented the autophagy activity induced by either nitrogen starvation or by the pharmacological inducer rapamycin [32] . It was proposed that this signaling pathway is controlling an activity required during the early stages of the autophagic pathway . However , in our work , we have demonstrated by employing the widely used PKA inhibitor H89 [33] ( Figure 2 ) , and by overexpressing a dominant negative mutant of the regulatory subunit of PKA [23] ( data not shown ) , that PKA does not seem to participate in cAMP inhibition of the autophagic pathway induced by the toxin . For many years , cAMP signaling was solely associated with PKA . However , novel cAMP sensors have come to light and they regulate many physiological processes either in concert with PKA or by themselves . It is known that cAMP is able to stimulate the cAMP-activated guanine exchange factor EPAC , which specifically turns on the monomeric G protein Rap [24] , [34] , [35] . EPAC proteins are known to control a range of diverse effectors and to regulate several pivotal processes . Here , we have shown that EPAC and its effector Rap2b participate in the regulation of Hla-induced autophagy . We have demonstrated that the direct activation of EPAC by cAMP or the overexpression of EPAC/Rap2b is sufficient to inhibit the autophagy response induced by the toxin ( Figures 3 , 4 , and 6 ) . Interestingly , Rubinsztein and collaborators [28] have shown that drugs that signal via the imidazoline type 1 receptor ( I1R ) , such as clonidine , act by reducing cAMP levels . The compounds enhanced A53T α-synuclein clearance and decreased toxic protein aggregation by activating an m-TOR independent autophagy . In addition , these I1R agonists signal via cAMP/Epac/Rap2b/PLC . When activated by cAMP , EPAC in turn activates Rap2b , which , through PLCε and an increase in the cytoplasmic levels of IP3 , induces exit of calcium from the endoplasmic reticulum . Rise in intracytosolic Ca2+ activates the calcium-dependent cysteine-protease calpains [36] . Indeed , it was shown that calpain inhibitors and siRNA knockdown of either calpain 1 or calpain 2 increased LC3-labeled autophagosomes [28] . Likewise , Junying Yuan and coworkers have shown that flurispirene , a compound that inhibits calcium flux , activates autophagy [29] . These authors have demonstrated that inactivation of calpain 1 , which in turn is able to cleave Atg5 , leads to activation of autophagy by increasing the levels of the Atg5-Atg12 complex required for LC3-lipidation . Consistently , with both reports , we have shown that the inhibition of calpains by the inhibitor calpeptin was sufficient to revert cAMP inhibition of the autophagy induced by Hla ( Figure 7 ) . Our results demonstrate , to our knowledge for the first time , that this signaling pathway participates in the regulation of the Hla-induced autophagic response and suggest that the toxin likely controls this pathway to allow autophagy activation , which is beneficial to the bacteria [18] , [21] . Additionally , a role for Atg5 in apoptosis has also been demonstrated . Hans-Uwe Simon and collaborators have identified a truncated form of Atg5 of 24 kDa in human neutrophils and Jurkat cells that is generated following different stimuli [37] . They concluded that Atg5 is cleaved by calpain 1 and 2 . They also showed that cells overexpressing Atg5 are more sensitive to cell death induced by different apoptotic stimuli and that the silencing of Atg5 reduces this cell death . Interestingly , this truncated Atg5 translocates from cytoplasm to mitochondria and causes cytochrome c release . The truncated form of Atg5 binds to Bcl-xl and may inactivate the Bcl-xl anti-apoptotic activity , promoting apoptotic cell death . These results clearly represent a link between autophagy and apoptosis through the calpain-mediated Atg5 cleavage [37] . Thus , it is tempting to speculate that S . aureus inhibits the calpain-mediated Atg5 cleavage to avoid apoptotic cell death and to favor autophagy activation , which is known to promote bacterial replication and bacterial survival [18] . To confirm the participation of this cAMP/EPAC/Rap2b pathway in the bacterial infection process , we used different S . aureus Hla positive and null strains . We have demonstrated that preincubation with cAMP was also able to inhibit the autophagy activation induced by S . aureus . Similar results were obtained when Rap2b wt and EPAC wt or its constitutively active mutant ΔEPAC were overexpressed . Interestingly , both EPAC and Rap2b were markedly recruited to the membrane of the bacteria-containing phagosome ( Figure 9 and Figure 10 ) . Aronoff and collaborators have demonstrated that following treatment of alveolar macrophages with prostaglandin E2 , EPAC-1 changes its localization from tubular membranes to the nuclear envelope and late phagosomes [38] . Since it has been shown that 8-pCPT , via EPAC , inhibits H2O2 production and bacterial killing , we propose that EPAC is recruited to the S . aureus phagosomal membrane to suppress its microbicidal capacity , inhibiting the killing of this intraphagosomal pathogen [38] , [39] . To the best of our knowledge , our studies are the first to demonstrate the recruitment of EPAC and Rap2b to a pathogen-containing compartment . Additionally , we have observed that those phagosomes that recruit EPAC are not labeled by LC3 ( Figure S6 ) . This is consistent with our model that EPAC ( and Rap2b ) act as an inhibitory molecule of the autophagic response induced by the Hla toxin . Further studies will be necessary to determine whether EPAC recruitment affects pathogen survival . We believe that our findings have important implications in understanding innate immune processes . Experiments are under way in our laboratory to determine how EPAC modulation controls the microbicidal capacity of different bacterial-containing phagosomes . Additionally , our results suggest that S . aureus keeps autophagy under tight control by downregulating levels of active Rap2b ( Figure 11 ) , likely to maintain appropriate levels of the Atg5-Atg12 conjugate ( Figure S2B ) . Since cAMP through EPAC activation is able to activate Rap2b , we determined the intracellular cAMP levels by RIA as described in Materials and Methods . We have observed a decrease in cAMP level in cells treated with the toxin or infected with S . aureus wt strain ( data not shown ) . Inhibition of calpain activity as a result of reductions in intracellular Ca2+ might be part of the signal that leads to the activation of autophagy machinery by increasing the levels of a key autophagy signaling molecule such as Atg5 . Indeed , we have previously shown that Atg5 is an absolute requirement for the toxin-activated autophagic response [21] . The evidence presented in this report indicates that the complex interplay between cAMP and Ca2+ , known to be involved in the control of many cellular processes , may also expand to the regulation of a pathogen-induced autophagic response . We have previously shown that a high concentration of BAPTA-AM ( 30 µM ) , an intracellular calcium chelator , is able to avoid the Hla-induced autophagy [21] . This concentration of BAPTA-AM allows the compound to cross the plasma membrane and the membrane of some organelles , chelating not only cytosolic but also intravesicular calcium [40] . Interestingly , we have found that a lower concentration of BAPTA-AM ( 5 µM ) , which chelates only the cytosolic calcium [41] , allows the autophagic response induced by the toxin ( data not shown ) . It is known that a certain level of intracellular calcium is necessary for autophagosome formation [42] , but we believe that an excess of cytosolic calcium leads to activation of the calpains proteases , which in turn could arrest the autophagic response induced by Hla . Thus , is likely that the intracellular calcium concentration is tightly regulated upon infection of S . aureus . Given the fact that S . aureus is a microorganism that causes serious diseases such as pneumonia , endocarditis , osteomyelitis , and wound infections , we believe that knowledge of the signal transduction mechanisms involved in the autophagy response and how these mechanisms enhance the intracellular survival of S . aureus is of seminal importance . The present findings will contribute to our understanding of the molecular mechanisms used by S . aureus to survive in infected cells , a key step in Staphylococcal pathogenicity .
α-MEM and D-MEN cell culture media and fetal calf serum were obtained from Invitrogen , Argentina ( Buenos Aires , Argentina ) . H89 was purchased from LC Laboratory ( Massachusetts , USA ) . A polyclonal rabbit anti-LC3 antibody was purchased from Sigma ( Buenos Aires , Argentina ) . The anti-myc antibody , the anti-Rap2b antibody , and Rap2b siRNA were purchased from Santa Cruz Biotechnology ( Buenos Aires , Argentina ) . All the other reagents were from Sigma ( Buenos Aires , Argentina ) . The anti-phosphoCREB was kindly provided by Dr . Verónica García ( Facultad de Ciencias Exactas , UBA , Buenos Aires , Argentina ) . pEGFP-LC3 was kindly provided by Dr . Noboru Mizushima ( The Tokyo Metropolitan Institute of Medical Science , Japan ) . The insert encoding the LC3 protein was subcloned into the red fluorescent protein vector ( pRFP , kindly provided by Dr . Philip Stahl , Washington University ) . Briefly , the insert from pEGFP-LC3 was cut with the Bgl II and EcoRI restriction enzymes and subcloned in the corresponding restriction sites of pRFP vector . The pGFP-Rap2B wt and pGFP-Rap2B ΔAAX were kindly provided by Dr . Mauro Torti ( University of Pavia , Pavia , Italy ) . The plasmids pCMV myc-Epac wt and pCMV myc-ΔEpac were kindly provided by Dr . Omar A . Coso ( IFIBYNE , Facultad de Ciencias Exactas , UBA , Buenos Aires , Argentina ) . CHO cells , an ovary hamster cell line , were grown in α-MEM supplemented with 10% FCS , streptomycin ( 50 µg/ml ) , and penicillin ( 50 U/ml ) . HeLa human epithelial cells were grown in D-MEM supplemented with 10% FCS , streptomycin ( 50 µg/ml ) , and penicillin ( 50 U/ml ) . For some experiments cells were incubated in starvation medium EBSS ( Earle's balanced salt solution ) . Stably transfected CHO cells overexpressing pEGFP-LC3 were used . For some experiments CHO cells were transiently cotransfected with pRFP-LC3 and pGFP-Rap2B wt or pGFP-Rap2B ΔAAX; or with pGFP-LC3 and pCMV myc-Epac wt or myc-ΔEpac . Cells were cotransfected using Lipofectamine 2000 ( Invitrogen ) , according to the manufacturer's instructions . Stably transfected CHO cells overexpressing pEGFP were used as control . Transfected CHO cells were incubated for 4 hours ( h ) with 10 µg/ml of α-hemolysin from S . aureus ( Sigma Aldrich; Buenos Aires , Argentina ) . Cells were fixed and analyzed by confocal fluorescence microscopy . For infection experiments , S . aureus strains , wt ( 01016 ) , the mutant deficient for α-hemolysin ( Hla− ) ( 01017 ) , or the Hla ( − ) mutant complemented with an α-hemolysin plasmid ( 01018 ) were grown overnight at 37°C in 5 ml of LB broth with appropriate antibiotics . Bacteria were resuspended in infection medium containing 10% FCS and 20 mM HEPES , at an OD650 of 0 . 4 ( ∼4×108 CFU ) . Bacteria were diluted to achieve a multiplicity of infection ( moi ) of 10∶1 ( bacteria∶cell ) in the infection medium . Transfected CHO cells were analyzed by fluorescence microscopy using an Olympus Confocal FV1000 ( Japan ) and processed with the program FV10-ASW 1 . 7 . In some experiments , to visualize the pathogen , bacterial DNA was labeled with 45 nM TOPRO in Mowiol . CHO GFP-LC3 cells were preincubated 30 min with 1 mM N6 , 2′-O-DIBUTYRYLADENOSINE 3′:5′-CYCLIC ( dbcAMP; Sigma; Buenos Aires , Argentina ) or 10 µM 8-pCPT-2′-O-Me-cAMP-AM ( 8-pCPT; BioLog; Bremen , Germany ) , and then they were treated with 10 µg/ml of α-hemolysin for 4 h or incubated in starvation medium for 2 h in the presence of the drugs . Cells were fixed and analyzed by confocal microscopy . CHO and HeLa cells were incubated under different conditions and lysed with sample buffer . Protein samples of a total cell lysate were run on a 10% polyacrylamide gel and transferred to Hybond-ECL ( Amersham ) nitrocellulose membranes . The membranes were blocked for 1 h in Blotto ( 5% non-fat milk , 0 . 1% Tween 20 , and PBS ) , washed twice with PBS and incubated with a primary antibody anti-LC3 and a peroxidase-conjugated secondary antibody ( Jackson Immuno Research , 211-032-171 ) . Anti-tubulin ( Jackson Immuno Research ) was used as a loading control . The corresponding bands were detected using an enhanced chemiluminescence detection kit from Healthcare ( Amersham , RPN2109 ) and the band was detected using Fujifilm LAS-4000 . CHO GFP-LC3 cells were preincubated 30 min with 10 µM H89 in the presence or absence of 1 mM db cAMP , and then they were treated with 10 µg/ml of α-hemolysin for 4 h or incubated in starvation medium for 2 h in the presence of the inhibitor . Cells were fixed and analyzed by confocal microscopy . CHO GFP-LC3 cells were preincubated 30 min with 10 µM calpeptin , in the presence or the absence of 1 mM dbcAMP . Subsequently , they were treated with 10 µg/ml of α-hemolysin for 4 h or incubated in starvation medium for 2 h . Cells were fixed and analyzed by confocal microscopy . CHO cells were infected for 3 h with the wt strain of S . aureus , the mutant deficient for α-hemolysin ( Hla− ) , or the Hla ( − ) mutant expressing an α-hemolysin plasmid . Infected cells were washed with PBS and lysed in water at 4°C . Lysates were diluted with PBS , plated on LB agar and incubated for 12 h at 37°C . Colonies were counted on the plate with dilutions yielding 50–100 visible colonies as previously determined [21] . HeLa cells were incubated for 4 h in the presence or absence of 10 µg/ml Hla , or they were infected for 4 h with S . aureus wt strain or Hla ( − ) strain . Afterwards , cells were lysed in a GST pull-down buffer ( 200 mM NaCl , 2 . 5 mM MgCl2 , 1% [v/v] Triton X-100 , 10% glycerol , 1 mM phenylmethylsulfonyl fluoride , a protease inhibitor mixture [Pepstatin , Leupeptin and Trypsin inhibitor] , and 50 mM Tris-HCl , pH 7 . 4 ) by sonication on ice ( two times for 30 s ) and used immediately . Glutathione-sepharose beads were washed twice with the GST pull-down buffer and incubated with bacterial lysates containing GST-Ral-GDS-RBD for 1 h at 4°C under constant rocking . Beads were washed twice with PBS and once with GST pull-down buffer and used immediately . Twenty µl of glutathione-sepharose containing 10 µg of the appropriate fusion protein was added to cell lysates in a total volume of 0 . 6 ml and incubated by rotation at 4°C for 1 h . The resin was recovered by centrifugation at 4°C ( 5 min at 10 , 000 rpm ) and washed three times with ice-cold GST pull-down buffer [25] . The resin-bound fractions were resolved by SDS-PAGE , and cellular GTP-Rap2b levels were analyzed by immunoblotting as described earlier , using a primary antibody anti-Rap2b and a peroxidase-conjugated secondary antibody ( Jackson Immuno Research ) . CHO cells were incubated 4 h with 10 µg/ml of α-hemolysin or infected 4 h with the wt strain of S . aureus ( 01016 ) or the mutant deficient for Hla ( 01017 ) . After incubations , cells were placed on ice , washed with PBS , and 0 . 7 ml cold 100% ethanol was added to each well . Cells were scraped and transferred to tubes ( Eppendorf ) , sonicated twice for 2 min , and heated for 5 min at 95 C° to destroy endogenous proteins . Afterwards , the samples were centrifuged 5 min at 10 , 000 rpm . Supernatants were dried and kept at −20 C° . Samples were diluted in 200 µl of 50 mM sodium acetate buffer ( pH 6 . 0 ) . Unknown samples and standards were acetylated and assayed by RIA using the method described by Del Punta et al . [43] . The interassay and intraassay variations of coefficients were lower than 10% . | Staphylococcus aureus is a microorganism that causes serious infectious diseases such as pneumonia , endocarditis , osteomyelitis , and wound infections . This pathogen can infect various types of non-professional phagocytic cells and after internalization is able to escape the phagolysosomal compartment towards the cytoplasm , where it actively replicates . Subsequently , the eukaryotic host cell is killed to spread the infection . Besides the clinical importance of this microorganism , the molecular mechanisms of S . aureus infection are not completely understood . S . aureus induces an autophagic response in infected cells , which is beneficial for bacterial replication and cell killing . We have previously shown that Hla is responsible for this autophagy activation . We found that the Hla-induced autophagic response occurs by a “non-canonical" pathway independent of PI3K/Beclin1 complex but dependent on Atg5 . Here we show that cAMP has a key role in the regulation of Hla-induced autophagic response . cAMP , through EPAC/Rap2b and via calpain activation , inhibits S . aureus–induced autophagy . Additionally , we show that EPAC and Rap2b are recruited to the S . aureus–containing phagosome . Our study contributes to the understanding of the molecular mechanisms used by S . aureus to survive , a key step in Staphylococcal pathogenicity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"staphylococci",
"biology",
"microbiology",
"bacterial",
"pathogens"
]
| 2012 | cAMP and EPAC Are Key Players in the Regulation of the Signal Transduction Pathway Involved in the α-Hemolysin Autophagic Response |
G protein-coupled receptors ( GPCRs ) control cellular signaling and responses . Many of these GPCRs are modulated by cholesterol and polyunsaturated fatty acids ( PUFAs ) which have been shown to co-exist with saturated lipids in ordered membrane domains . However , the lipid compositions of such domains extracted from the brain cortex tissue of individuals suffering from GPCR-associated neurological disorders show drastically lowered levels of PUFAs . Here , using free energy techniques and multiscale simulations of numerous membrane proteins , we show that the presence of the PUFA DHA helps helical multi-pass proteins such as GPCRs partition into ordered membrane domains . The mechanism is based on hybrid lipids , whose PUFA chains coat the rough protein surface , while the saturated chains face the raft environment , thus minimizing perturbations therein . Our findings suggest that the reduction of GPCR partitioning to their native ordered environments due to PUFA depletion might affect the function of these receptors in numerous neurodegenerative diseases , where the membrane PUFA levels in the brain are decreased . We hope that this work inspires experimental studies on the connection between membrane PUFA levels and GPCR signaling .
Cellular membranes host functional membrane domains ( “lipid rafts’’ ) rich in proteins and cholesterol ( CHOL ) [1] . Many G protein-coupled receptors ( GPCRs ) and cognate G proteins are found in these domains [2] , and numerous reports have suggested that CHOL is involved in GPCR function [3–7] . Moreover , impaired CHOL homoeostasis and raft disruption have been linked to different neurodegenerative diseases [2 , 8] , where GPCRs play a pivotal role . However , the mechanism driving the partitioning of GPCRs to their native functional CHOL-rich environments is still not well understood . Polyunsaturated fatty acids ( PUFAs ) such as docosahexaenoic acid ( DHA , 22:6 ( n-3 ) ) are likewise key membrane components of brain cells [9] . PUFAs esterify to phospholipids together with a saturated chain to form a hybrid lipid . Intriguingly , despite their disordered nature , hybrid lipids are found in raft extracts [10–12] , and they also partition surprisingly well to cholesterol-rich ordered membrane regions [13] . However , raft PUFA levels are reduced in various neuropsychiatric and mental disorders [14] including Alzheimer’s [10] and Parkinson’s diseases [11] . This lack of PUFAs could thus affect GPCR function . In fact , experiments have shown that DHA-containing lipids enhance the function of the prototypical GPCR rhodopsin [15–17] , which simulation studies have explained to take place as a result of the high conformational flexibility of DHA chains . This provides hybrid lipids with high affinity for the rough surface of GPCRs , [18–21] further promoting protein–protein interactions [22] . We recently reported the high affinity of DHA for the adenosine A2A receptor ( A2AR ) [23] , a GPCR with an important role in the central nervous systems , where different antagonists of A2AR have shown promising neuroprotective effects [24 , 25] . Membrane CHOL is also known to closely interact with A2AR [7 , 26–28] , modulating its function [29] and ligand binding properties [7] . The partitioning of A2AR into ordered membrane domains [30] is therefore quite expected , though the mechanism rendering it possible has been suggested to be complex [31] . Moreover , given the numerous factors affecting protein partitioning [32] and the limited ability of model systems to capture in vivo behavior [33] , it is not surprising that the role of PUFAs in A2AR partitioning remains to be investigated . Given the central role of GPCRs in cell signaling , unlocking how DHA interacts with GPCRs is the key to understanding why GPCR function is impaired in severe brain diseases associated with a lowered membrane DHA level . Here , we studied the role of PUFAs in the partitioning of GPCRs into CHOL-rich ( raft-like ) liquid-ordered ( Lo ) and CHOL-depleted liquid-disordered ( Ld ) phases . Combining all-atom and coarse-grained molecular dynamics ( MD ) simulations with free energy calculations , we demonstrate for A2AR that in the absence of DHA , corresponding to brain tissue of diseased individuals , partitioning to the Ld phase is energetically favored . However , in membranes including DHA-containing hybrid lipids , corresponding to brain tissue of healthy individuals , DHA drives A2AR to partition to the Lo phase , as a favorable structural arrangement of DHA around A2AR minimizes the structural perturbations therein . Furthermore , based on our studies on a number of distinct membrane proteins , we demonstrate that the observed effect of DHA could be limited to rough helical multi-pass membrane proteins , which include GPCRs .
We calculated the free energy of transfer of A2AR between Lo and Ld phases in the coarse-grained ( CG ) scheme using the non-polarizable Martini 2 . 2 model [34–36] . First , we embedded the protein in an Lo phase membrane containing distearoylphosphatidylcholine ( DSPC , Fig 1B ) , 20 mol% CHOL ( Fig 1E ) , and different concentrations of stearoyldocosahexaenoylphosphatidylethanolamine ( SDPE , Fig 1D ) with a polyunsaturated DHA chain , see Fig 1A . In line with lipidomics experiments , DHA was paired with the PE head group . [37] Next , we mutated Lo-forming DSPC to Ld-forming dioleoylphosphatidylcholine ( DOPC , Fig 1C ) in a set of simulations and extracted the free energy change ΔGLoProt→LdProt using the free energy perturbation approach . Here , a coupling parameter λ has a value of 0 for DSPC and 1 for DOPC . Then , we obtained ΔGLo→Ld by repeating this calculation in the absence of the protein . As discussed in Section B . 7 in the S1 File , it is possible that the experimentally observed microscopic phase separation in this DOPC/DSPC/CHOL mixture [38] is associated by a fairly large line tension and hence only takes place in membranes larger than those currently within the reach of MD simulations . This limits us from studying protein partitioning in DOPC/DSPC/CHOL mixtures with coexisting domains . Nevertheless , the lipid chain order parameters , shown in Fig 2A as a function of λ , demonstrate a smooth transition between distinct Lo and Ld phases in both sets of the simulations . We therefore believe that our approach is able to capture the physical properties of the coexisting phases in isolation . Further analyses shown in Section B . 1 in the S1 File also support this view . Following the thermodynamic cycle depicted in Fig 1F , we carried on to extract the free energies of transfer as ( ΔGLoProt→LdProt−ΔGLo→Ld ) . Additionally , we also used a more realistic composition—based on the tie lines measured for the DOPC/DPPC/CHOL mixture—where the DSPC/DOPC ratio was 2 . 3 in the Lo phase , and then reversed to 1/2 . 3 in the Ld phase . For further details on our computational approach , the system compositions , and the simulation parameters , see Methods and the S1 File . The free energy of transfer of A2AR as a function of SDPE concentration is shown as a solid line in Fig 3A . Strikingly , the free energy of transfer changes sign at the SDPE concentration of ∼8 mol% . This highlights that for dilute concentrations of SDPE , A2AR partitions to the Ld phase . However , at higher SDPE concentrations the picture changes completely and the protein favors partitioning to the Lo phase . Concluding , the data provide compelling evidence that the presence of SDPE , and therefore DHA , makes A2AR compatible with the Lo phase . Fig 4A shows the 2D radial distribution functions ( RDFs ) of all lipid chain types around A2AR in the Lo phase with 4 mol% of SDPE . The data are extracted from well-equilibrated membranes in the CG scheme . Fig 4A demonstrates that A2AR is fully coated by SDPE with polyunsaturated DHA forming the first solvation shell , followed by the saturated acyl chain of SDPE and CHOL in the second shell . The formation of these shells is illustrated in the movie at DOI:10 . 6084/m9 . figshare . 5903881 . With increasing SDPE concentration , the right tail of the RDF peaks of all lipids extends further away from the protein ( see Fig . E in the S1 File ) , indicating that the A2AR surface becomes saturated with DHA . Interestingly , Fig 4A shows that CHOL penetrates the shell formed by the saturated chains of SDPE , and occupies annular binding sites , in line with experimental and computational studies on CHOL–A2AR interaction [26–28] . Indeed , cholesterol finds the suggested binding sites in the absence ( Fig 4B ) but also in the presence of ( Fig 4C ) an SDPE shell . These lipid shells around A2AR are dynamic . Lipids exchange in the time scale of ∼100 ns , as evidenced by the decay time constants found through double exponential fits to the contact data , shown in Table B in the S1 File ( see also Section B . 2 in the S1 File ) . For the Lo phase , the rates of SDPE and CHOL exchange increase as SDPE concentration increases . In the Ld phase , the SDPE corona dissolves ( see Fig . G in the S1 File ) . This lack of a tightly-bound SDPE shell leads to higher SDPE and CHOL exchange rates . Similarly , CHOL exchange rates are also higher in the absence of SPDE . These findings demonstrate that the formation of an SDPE shell also affects the dynamics of CHOL association by stabilizing the neighborhood of A2AR . Concluding , the strong affinity of DHA to interact with A2AR leads to coating of the protein by SDPE lipids . DHA is in contact with the protein , whereas the saturated chains favor interactions with CHOL . Partitioning of a membrane protein to either the Lo or the Ld phase is driven by the mutual structural compatibility between the protein and the lipids forming the membrane phase . Possible parameters describing this compatibility include hydrophobic mismatch , the conformational entropy of the protein , and perturbation of lipid chain order . We evaluated the contribution of all these factors in the CG scheme . Membrane thickness is shown in Fig . K in the S1 File as a function of distance from protein surface . The presence of SDPE has a clear effect on the thickness . Based on the mattress model [39] and using the hydrophobic mismatch parameter from Ref . [32] and the hydrophobic thickness of A2AR from the OPM database [40] , we estimate that hydrophobic mismatch contributes to the free energy of transfer by approximately 1 . 8 kJ/mol , favoring the Ld phase . However , the presence of 8 mol% of SDPE has an insignificant effect on this value , indicating that negating hydrophobic mismatch is not the mechanism through which SDPE shifts partitioning of A2AR towards the Lo phase . Notably , this conclusion is insensitive to the value of the hydrophobic mismatch parameter , which might be different between experiment and our simulation model . Next , we evaluated whether the SDPE corona promotes protein flexibility , hence resulting in a favorable entropic contribution for partitioning to the Lo phase in the presence of SDPE . We plot the residue-wise root mean squared fluctuations ( RMSF ) of the protein structure in both the Lo and Ld phases in Fig . L in the S1 File . Curiously , in the absence of SDPE , the average RMSF value is slightly higher in the Lo phase . However , at 8 mol% of SDPE the average RMSF becomes larger in the Ld phase than in the Lo phase ( see inset in Fig . L in the S1 File ) . This suggests that the entropic contribution due to the presence of SDPE actually promotes A2AR partitioning to the Ld phase and hence acts against the observed effect of SDPE on the free energy of transfer . Moreover , we note that the omitted lipid entropies also likely play a role on partitioning . How about protein-induced changes in lipid acyl chain order ? The outer layer of the SDPE corona around A2AR is formed by the saturated stearic acid chains of SDPE ( see Fig 4A ) . This layer is likely more compatible with the Lo phase than the rough surface of A2AR . This idea is indeed backed up by results from CG systems , which show that the effects of SDPE on membrane properties are reduced in the presence of A2AR and vice versa ( see Fig 2A ) . We note here that the CG approach is not well-suited to fully characterize acyl chain order . Therefore , we also studied the effects of SDPE and A2AR on membrane order in all-atom detail . To this end , we fine-grained selected coarse-grained Lo phase systems and carried out all-atom simulations using the CHARMM36 force field [41 , 42] as described in Methods . The averaged stearic acid chain order parameters from both all-atom and coarse-grained simulations of the Lo phase membranes are shown in Fig 2B . It is evident that both A2AR and SDPE lower the average order of the membrane . However , at an SDPE concentration of 8 mol% , the presence of A2AR actually increases membrane order , and this observation holds for both all-atom and coarse-grained schemes . The explanation to this behavior is that when both SDPE and A2AR are present , the DHA–A2AR interactions shield the order-lowering effects of both SDPE and A2AR . Importantly , the values from coarse-grained and atomistic simulations are in the same ballpark . The spatial variation of membrane order due to the presence of A2AR is studied in detail in Section B . 3 in S1 File . To conclude , in the Lo phase , the association of the flexible DHA chains and the rough surface of A2AR weakens their perturbations on membrane ( acyl chain ) order . Previous simulations and experiments have demonstrated the favorable interactions of DHA and GPCRs , including A2AR and dopamine D2 receptor ( D2R ) [18 , 20–23 , 43] . Here , we systematically studied four distinct membrane protein types—one β-barrel and three α-helical structures with 1 , 2 , or 7 transmembrane passes , including A2AR as a representative GPCR . The proteins are 1 ) the transmembrane domain of the human receptor tyrosine kinase ( ErbB1 , PDB id: 2M0B ) , a single helix; 2 ) a dimer formed by two Glycophorin A peptides [44] ( GpA dimer , PDB id: 1AFO ) ; 3 ) A2AR ( PDB id: 3EML ) [45] , a heptahelical bundle employed in the CG free energy calculation; and 4 ) the voltage-dependent anion channel ( VDAC , PDB id: 3EMN ) [46] , a β-barrel . These proteins are depicted in the middle column of Fig 5 . Notably , the lengths of the hydrophobic spans of the helical proteins were all equal to 3 . 2 nm [40] , so this factor cannot lead to differences in lipid–protein interactions . However , the β-barrel is substantially thinner at 2 . 3 nm . We simulated these proteins in membranes comprised of lipids , whose chains’ unsaturation level was varied ( chains with 0 , 1 , 2 , or 6 double bonds per chain ) . We evaluated how the lipids solvated the proteins in these membranes using unbiased all-atom simulations together with the CHARMM36 force field [41 , 42] . We paired all lipid chains with a PC head group in order to study only the effect of lipid chains . The final structures of the simulated systems are shown in the rightmost column of Fig 5 . The details are given in Methods and in Section A . 4 in the S1 File . The RDFs of the fatty acid chains around the proteins were determined after full lipid mixing had taken place . It is evident from these RDFs ( see leftmost column of Fig 5 ) that the non-GPCR proteins ( here ErbB1 , GpA dimer , and VDAC ) do not show any clear preference for DHA . Meanwhile , A2AR , as a representative example of GPCRs , interacts mostly with the DHA chain of SDPE , and the saturated chain of SDPE again forms an outer layer of the lipid corona that is in contact with the protein . This observation , in agreement with the results of CG simulations ( Fig 4A and our earlier study [23] ) , suggests that DHA adapts to the rough surface of A2AR . Protein roughness ( i . e . the degree of irregularity of a protein surface ) [47] is known to correlate with its propensity to interact with small molecules [48] . Therefore , it has been used to predict binding sites at the protein surface [49] . Importantly , surface roughness is a general feature of GPCRs [50] and explains the preferential interaction of the flexible and kinked DHA chain with A2AR [19] . The fact that a smoother β-barrel ( VDAC ) surface is not solvated by DHA is in favor of this view . Since this phenomenon is also not observed for proteins with a smaller number of helices ( ErbB1 and GpA dimer ) , its origin likely lies in the preference of DHA for the creviced tertiary structure instead of the helical secondary structure . To further quantify the DHA adaptation onto the A2AR surface , we calculated the mean number of the residues in the helical TM region of A2AR that were in the vicinity ( <0 . 3 nm ) of a lipid chain in the fine-grained simulations . We found a systematic increase: +10% for the membrane with 4 mol% of SDPE and +15% for the membrane with 8 mol% of SDPE as compared to the SDPE-free case . This effect was not dependent on the chosen cutoff , as values of +7% and +17% were calculated for a cutoff of 0 . 4 nm . While this calculation clearly shows that DHA chains adapt better to the A2AR surface , a further and more systematic study on the effects of the surface topology and the amino acid content therein on DHA–protein interactions is required in the future to verify our findings . The favorable interaction between flexible DHA chains and GPCR surfaces is highlighted in Fig 6 , which shows representative configurations sampled in the fine-grained all-atom simulations , where a DHA chain has adapted its conformation to the rough protein surface and entered a crevice on the A2AR surface ( see Fig 6A ) , or penetrated into the helical bundle of A2AR ( see Fig 6B ) . Concluding , hybrid lipids with a DHA ( or likely other PUFA ) chain and a saturated chain seem to be favored by GPCRs , and this is likely due to the rough surface of the transmembrane region in these multi-helical proteins . Based on the observation that the DHA–protein interaction is characteristic for proteins with multi-pass helical bundles , we extended our free energy of transfer calculation in the CG scheme to two additional proteins of this kind . We also note that the effects for other membrane protein types might be similar in the Martini scheme as many proteins seem to interact favorably with PUFAs [51] , likely due to unbalanced entropic and enthalpic contributions to this interaction . However , based on our all-atom simulations , we abstain from studying the free energies of transfer for protein types without multiple TM helices . D2R is linked to many neurological and psychiatric disorders [52] associated with lowered PUFA levels [10 , 11 , 14] . The DHA–D2R interaction was recently demonstrated by us [23] . We also considered the brain-associated glucose transporter GLUT1 , whose function is also dependent on PUFAs [53 , 54] . While GLUT1 is not a GPCR , it also has a multi-pass structure consisting of 12 helices . We estimated the free energies of transfer for all three proteins—D2R , GLUT1 , and A2AR—in the absence and in the presence ( 16 mol% ) of SDPE and hence DHA . We also note that while the phase-separation of the commonly used lipid mixtures in the Martini model is complete and the phase boundaries are sharp [55] , experiments report less distinct compositions between the Lo and Ld phases [56] . We therefore considered both the situation mimicking complete separation ( such as above ) , as well as a more realistic situation in which the Lo phase had a realistic DSPC/DOPC ratio of 2 . 3 , which is reversed during the mutation into an Ld phase ( see Methods and Section A . 2 in S1 File ) . The free energies of transfer for all three proteins are shown in Fig 3B . The effect of SDPE is clearly demonstrated for all proteins . Moreover , while the absolute values are smaller in the membranes with more realistic compositions , the change of sign , i . e . the change in the favored phase changes consistently upon the addition of SDPE . This behavior is in line with the two phases now being less distinct , as demonstrated by the order parameters shown in Fig . D in the S1 File . Moreover , the strong association of D2R and GLUT1 with DHA ( see Fig . F in the S1 File ) is again responsible for the effect—similar to what was observed for A2AR ( see Fig 4A ) . It is also worth pointing out that while we paired DHA with a PE head group ( to form SDPE ) , the calculations performed with SDPC instead of SDPE show an almost equal effect on protein partitioning ( see Section B . 8 in S1 File ) . Concluding , the SDPE-induced partitioning to the Lo phase is reproduced across three multi-helical brain-associated proteins—two of which are GPCR neuroreceptors—whose function is compromised by changes in membrane DHA levels . This effect is also consistently observed with less distinct and more realistic phase compositions .
Using multi-scale simulations and free energy calculations , we showed that a small amount of SDPE , a DHA-containing hybrid lipid , enhances A2AR partitioning to the Lo phase . Without DHA , the protein favors partitioning to the Ld phase instead . The change in this behavior stems from the rough surface of A2AR that favors interacting with DHA and , presumably , also with other PUFAs over saturated chains . This interaction leads to a well-organized SDPE corona where the DHA chains face the receptor , while the saturated chain of SDPE in the outer layer of the corona interacts with CHOL and saturated phospholipid chains in the Lo phase . Through this mechanism , the perturbations of the flexible DHA chains and the rough receptor surface on the Lo phase are largely eliminated . The striking finding made in this work is that the lipid corona could play a decisive role in the partitioning of membrane proteins . We showed that this holds true for A2AR used in this work as a prototypical GPCR . The additional results strongly suggest that the same conclusion holds for helical multi-pass proteins such as GPCRs with rough surfaces , yet not for other protein topologies with smoother surfaces . We acknowledge that while coarse-grained models are designed to capture the correct trends , the absolute free energy values should be taken with caution . Still , our values are in line with [32] if not smaller than ( compare the data for WALP23 in Methods with Ref . [55] ) the values obtained with the Martini model exploiting different free energy techniques . We discuss other possible methodological limitations in detail in Section B . 7 in the S1 File . Our results suggest that small concentrations of lipids not included in model membranes might have drastic effects on the partitioning behavior of membrane proteins studied in vivo , and can explain why raft-associated proteins partition to the Ld phase in phase-separated giant unilamellar vesicles [33] . Further , the present simulation results are in line with experiments suggesting that other structural features such as post-translational modifications , protein surface roughness , and hydrophobic mismatch modulate the affinity of membrane proteins for lipid rafts [32] . Given that the solvation of a GPCR by a DHA-containing hybrid lipid is based on a layer where DHA stands next to the protein surface and saturated chains occupy the outermost shell of the protein , this arrangement can increase the raft affinity of the GPCR protein in three ways: it provides the protein with non-covalently bound saturated lipid anchors , it complements the surface roughness of the protein , and with an appropriate choice of the saturated chain in the hybrid lipids , hydrophobic mismatch can be reduced . The concentration of DHA in raft membrane domains in the brain of healthy subjects is ∼7 mol-% [11] . Assuming a protein area coverage similar to that in red blood cells [57] and an average protein and lipid area of 10 nm2 and 0 . 7 nm2 , respectively , the protein-to-lipid ratio would be approximately 1 to 50 per leaflet . With an SDPE content of ∼14 mol-% , and considering that the membrane has two leaflets , the estimated protein-to-SDPE ratio is 1 to 13 . Strikingly , the saturation of the A2AR surface in our simulations takes place around this protein-to-SDPE ratio ( see Figs . G and H in the S1 File ) . Hence , this consideration suggests that in the brain tissue of healthy subjects the DHA concentration is sufficiently large to favor the partitioning of A2AR to ordered regions with structural similarity to the Lo phase . However , one has to keep in mind that our simplified model membranes do not capture either the heterogeneity or leaflet asymmetry of membranes in the brain , which can fine-tune the partitioning behavior of proteins . Moreover , the membranes considered in this study are planar , yet GPCRs with high intrinsic curvature are also sorted by curvature [58] , and the DHA corona might have an effect therein . Studies of asymmetry or curvature are beyond this work , yet might need to be taken into account when experimental validation for our findings is sought . Then what happens if the DHA level is decreased ? It is known that the level of DHA in the brain of people suffering from neurodegenerative diseases is substantially reduced [10 , 11 , 59] . It is tempting to speculate that the reduced DHA content would alter the partitioning of A2A or D2 receptors , displacing them from CHOL-rich domains to disordered regions , compromising GPCR signaling . It has been shown that cholesterol binds to GPCRs such as beta-2-adrenergic receptor in an allosteric manner [6] , affecting its conformational distribution , thus the concern of compromised GPCR signalling due to a lowered DHA level is justified . In brief , the effect observed in the present study on partitioning has implications on health . While DHA is promising in the treatment of neurodegenerative diseases [60] , the mechanism behind this protective effect , despite rendering membranes more fluid , remains elusive . In our earlier study [23] , we showed that the formation of A2AR homo- and hetero-oligomers with the dopamine D2 receptor is decreased when the DHA levels are reduced . In the current work , we postulate that DHA-containing lipids have a dual role in preventing neurodegenerative diseases by lipid–protein interactions: 1 ) they can influence raft partitioning , therefore indirectly 2 ) modulating key aspects of the GPCR biology , such as protein oligomerization . The proper function of these oligomeric and mutually regulatory receptor units in a suitable lipid environment is essential for the properly functioning healthy brain . Our findings could explain some of the beneficial effects of DHA-based therapies previously shown for certain brain disorders [61] .
We embedded A2AR ( PDB id: 3EML [45] ) to an Lo membrane consisting of DSPC and 20 mol% CHOL . Next , varying amounts of DSPC was replaced by the hybrid lipid SDPE with a saturated ( C18:0 ) and a polynsaturated ( DHA ) chain . The protein and the lipids were modeled in the coarse-grained ( CG ) scheme using the non-polarizable Martini 2 . 2 model [34–36] together with the elastic network for A2AR [62] . Next , DSPC was transformed into DOPC , resulting in the change of membrane phase from Lo to Ld . This process was performed as an alchemical transformation using the dual topology paradigm with 27 windows . We verified the change in phase thoroughly ( see Section A . 1 in the S1 File ) , and validated our approach using the 27-residue WALP peptide that favored the Ld phase ( free energy of transfer of 17 . 2±1 . 0 kJ/mol ) , in line with eperiments and simulations [55] . The associated free energy changes were estimated by the Bennett acceptance ratio ( BAR ) method [63] implemented in the gmx bar tool of GROMACS , and the free energy of transfer was obtained as ΔGLoProt→LdProt−ΔGLo→Ld where the two terms correspond to this phase change in the presence and absence of the protein . To study the generality of the effect of SDPE on the partitioning of helical multi-pass membrane proteins , we considered two additional brain-associated cases , with relation to DHA—dopamine D2 receptor ( D2R ) and glucose transporter GLUT1 ( PDB id: 4PYP [64] ) , whose free energies of transfer were calculated in the absence of SDPE and in the presence of 16 mol%SDPE . The systems were set up identically to the ones containing A2AR , and the same equilibration and simulation protocols were followed . In the simulations , performed using GROMACS v5 . 0 . x [65] , the recently suggested “New-RF” simulation parameters [66] were employed . See Section A . 2 in the S1 File for further details . Finally , the free energies of transfer were also calculated for A2AR , D2R , and GLUT1 in the absence of and in the presence of 16 mol% SDPE in membranes whose compositions mimicked those of coexisting phases in model membranes ( see Table A in the S1 File ) . To study how DHA affects the adaptation of the protein into the membrane , we fine-grained the well-equilibrated CG systems containing 0 , 4 , and 8 mol% SDPE into all-atom resolution using the backward tool [67] . Additionally , we simulated membranes with identical lipid ratios yet in the absence of the protein as a control . All all-atom simulations , performed using GROMACS v5 . 0 . x [65] , employed the CHARMM36 force field [41 , 42] . The last 150 ns of 200 ns simulations was used in the analyses . The default input parameters provided by CHARMM-GUI were used [68] . See Section A . 3 in the S1 File for further details . We studied whether certain protein types are more prone to be solvated by DHA in all-atom detail . To this end , we simulated four structurally different transmembrane proteins: 1 ) the transmembrane domain of the human receptor tyrosine kinase ( ErbB1 , PDB id: 2M0B ) , a single helix; 2 ) a dimer formed by two Glycophorin A peptides [44] ( GpA dimer , PDB id: 1AFO ) ; 3 ) A2AR ( PDB id: 3EML ) [45] , a heptahelical bundle employed in the CG free energy calculation; and 4 ) the voltage-dependent anion channel ( VDAC , PDB id: 3EMN ) [46] , a β-barrel . These proteins were embedded in a lipid bilayer consisting of equimolar concentrations of CHOL , dipalmitoyl-phosphatidylcholine ( DPPC , two saturated chains; di-16:0 ) , DOPC ( two monounsaturated chains; di-18:1 ) , dilinoleoyl-phosphatidylcholine ( DLiPC , two diunsaturated chains; di-18:2 ) , and stearoyl-docosahexaenoyl-phosphatidylcholine ( SDPC , one saturated 18:0 chain and one polyunsaturated 22:6 ( DHA ) chain ) . The input structures for GROMACS were generated using the CHARMM-GUI Membrane Builder [68] , and the systems were simulated for 4 μs using the input parameters provided by CHARMM-GUI [68] . The last 500 ns were used in the analyses . See Section A . 4 in the S1 File for further details . | Our current picture of cellular membranes depicts them as laterally heterogeneous sheets of lipids crowded with membrane proteins . These proteins often require a specific lipid environment to efficiently perform their functions . Certain neuroreceptor proteins are regulated by membrane cholesterol that is considered to be enriched in ordered membrane domains . In the brain , these very same domains also contain a fair amount of polyunsaturated fatty acids ( PUFAs ) that have also been discovered to interact favorably with many receptor proteins . However , certain neurological diseases—associated with the inadequate functioning of the neuroreceptors—seem to result in the decrease of brain PUFA levels . We hypothesized that this decrease in PUFA levels somehow inhibits receptor partitioning to cholesterol-rich domains , which could further compromise their function . We verified our hypothesis by an extensive set of computer simulations . They demonstrated that the PUFA–receptor interaction indeed leads to favorable partitioning of the receptors in the cholesterol-rich ordered domains . Moreover , the underlying mechanism based on the shielding of the rough protein surface by the PUFAs seems to be exclusive for multi-helical protein structures , of which neuroreceptors are a prime example . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"protein",
"interactions",
"simulation",
"and",
"modeling",
"membrane",
"proteins",
"cellular",
"structures",
"and",
"organelles",
"thermodynamics",
"g",
"protein",
"coupled",
"receptors",
"research",
"and",
"analysis",
"methods",
"lipids",
"proteins",
"transmembrane",
"receptors",
"cell",
"membranes",
"free",
"energy",
"cholesterol",
"physics",
"biochemistry",
"signal",
"transduction",
"biochemical",
"simulations",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology"
]
| 2019 | Reduced level of docosahexaenoic acid shifts GPCR neuroreceptors to less ordered membrane regions |
In somatic cells of female placental mammals , one of the two X chromosomes is transcriptionally silenced to accomplish an equal dose of X-encoded gene products in males and females . Initiation of random X chromosome inactivation ( XCI ) is thought to be regulated by X-encoded activators and autosomally encoded suppressors controlling Xist . Spreading of Xist RNA leads to silencing of the X chromosome in cis . Here , we demonstrate that the dose dependent X-encoded XCI activator RNF12/RLIM acts in trans and activates Xist . We did not find evidence for RNF12-mediated regulation of XCI through Tsix or the Xist intron 1 region , which are both known to be involved in inhibition of Xist . In addition , we found that Xist intron 1 , which contains a pluripotency factor binding site , is not required for suppression of Xist in undifferentiated ES cells . Analysis of female Rnf12−/− knockout ES cells showed that RNF12 is essential for initiation of XCI and is mainly involved in the regulation of Xist . We conclude that RNF12 is an indispensable factor in up-regulation of Xist transcription , thereby leading to initiation of random XCI .
X chromosome inactivation ( XCI ) in placental mammals is random with respect to the parental origin of the X chromosome that undergoes inactivation , during early embryonic development [1] . In contrast , in marsupials and mouse extra-embryonic tissues XCI is imprinted . Imprinted XCI always targets the paternally inherited X chromosome ( Xp ) , and is initiated during the early cleavage divisions [2] , [3] , [4] . In the inner cell mass ( ICM ) of the mouse blastocyst , the inactive X chromosome is reactivated , after which random XCI is initiated around 5 . 5 days of embryonic development . In mouse , two non-coding X-linked genes , Xist and Tsix , play a central role in the random XCI mechanism . Upon initiation of XCI , Xist is up-regulated on the future inactive X chromosome ( Xi ) , and the transcribed RNA spreads along the X in cis , directly and indirectly recruiting chromatin modifying enzymes acting to establish the Xi [5] , [6] , [7] . Tsix is a negative regulator of Xist; the Tsix gene overlaps with Xist but is transcribed in the anti-sense direction [8] , [9] . Random XCI is a stochastic process in which each X chromosome has an independent probability to become inactivated [10] , [11] . Initiation of XCI is thought to be regulated by X-encoded activators and autosomally encoded inhibitors [11] , [12] . With two active X chromosomes , female cells will have a concentration of XCI activators two-fold higher than male cells , sufficiently different to drive XCI in female cells only . Rapid down-regulation of XCI activator genes in cis , after initiation of XCI on either one of the X chromosomes , prevents initiation of XCI on the second X chromosome . XCI inhibitors are involved in maintaining a threshold for XCI to occur . So far , several XCI inhibitors have been identified , acting through different mechanisms , in mouse . YY1 and CTCF act as positive regulators of Tsix , by binding the DXpas34 Tsix regulatory element [13] . The pluripotency factors OCT4 , SOX2 and NANOG were proposed to regulate XCI by binding to intron 1 of Xist and suppressing Xist expression directly [14] . OCT4 and SOX2 have also been implicated in the positive regulation of Tsix and Xite , the latter being an enhancer of Tsix [15] . These findings indicate that several proteins and pathways act in concert to suppress Xist transcription and to block Xist RNA spreading in cis . XCI activators could act by activation of Xist , but also by suppression of negative regulators of Xist such as Tsix and the Xist intron 1 region . Recently , we identified RNF12 ( RLIM ) as the first X-linked activator of XCI [16] . This E3 ubiquitin ligase is involved in regulation of LIM-homeodomain transcription factors and telomere length homeostasis , through degradation of LDB1 and TRF1 , respectively [17] , [18] . Previously , we found that additional transgenic copies of the Rnf12 gene encoding this protein resulted in induction of XCI on the single X in transgenic male cells , and on both X chromosomes in a high percentage of female cells . XCI was also affected in Rnf12+/− ES cells supporting a dose-dependent role for RNF12 in activation of XCI . In the present study , we aimed to dissect the role of RNF12 in XCI , and we obtained evidence that RNF12 regulates XCI in trans , by activation of the Xist promoter . In addition , the generation and analysis of Rnf12−/− ES cells indicated that RNF12 is required for the XCI process and appears to be involved in XCI mainly by activation of Xist . The results reinforce that RNF12 is a key player in regulation of the XCI process .
XCI is regulated by several cis elements , and Rnf12 is located in close proximity to Xist ( ∼500 kb ) . Therefore , we aimed to test whether all the activity of RNF12 is mediated in trans . Our previous studies showed that Rnf12+/− female ES cells induce XCI in a reduced number of ES cells . Here , we rescued 129/Sv/Cast/Ei ( 129/Cas ) polymorphic Rnf12+/− female ES cells by introducing a 129 BAC ( RP24-240J16 ) construct covering Rnf12 . RT-PCR analysis followed by RFLP detection confirmed expression of the transgenic copies of Rnf12 ( Figure 1A ) . Xist RNA-FISH analysis , to detect the Xist coated inactive X chromosome ( Xi ) in day 3 differentiated transgenic ES cell lines with one additional copy of Rnf12 , shows that XCI was restored to wild type level ( Figure 1B ) . In line 20 , with 5 transgenic copies of Rnf12 the percentage of cells with one or two Xi's is even more pronounced , supporting a dose dependent role of RNF12 in XCI ( Figure 1B , 1C ) . XCI is skewed in wild type 129/Cas female ES cells towards inactivation of the 129 X . This is due to the presence of different X-linked cis elements ( Xce ) that affect random choice [19] . RT-PCR detecting a length polymorphism was used to distinguish Xist emanating from either the 129 or the Cas alleles . We observed that skewed XCI is more pronounced in the Rnf12+/− cells , as compared to XCI in wild type cells at day 3 of differentiation ( Figure 1D ) . This could be caused by selection against cells inactivating the wild type X chromosome , which would result in complete loss of RNF12 from these cells . However , RNF12 possibly is not essential for cell survival , also of differentiated cells , so that selection against cells inactivating the wild type X chromosome might point to a role for RNF12 in maintaining Xist expression . In the rescued cell lines , Xist was up-regulated from both alleles at day 3 of differentiation ( Figure 1D ) . This result demonstrates that RNF12 activates XCI in trans . One possible mechanism for regulation of XCI by RNF12 , might be a direct interaction with Xist RNA to target chromatin components . However , examination of day 3 differentiated female cells by immunocytochemistry detecting RNF12 , together with the Polycomb protein SUZ12 which accumulates on the Xi [20] , [21] , excludes this possibility ( Figure 2A ) . Interestingly , we noticed that the RNF12 staining intensity was much more dynamic in female compared to male cells ( Figure 2B , Figure S1 ) . Also , in female cells , a SUZ12 coated Xi appeared mainly in cells with low RNF12 staining ( Figure 2A , Figure S2 , and data not shown ) . Immunostaining of differentiating female ES cells indicated a negative correlation between expression of RNF12 and NANOG , although expression was not completely mutually exclusive ( Figure 2C ) . To analyze this in more detail , we targeted an Rnf12 promoter-mCherry construct into ES cells , also harboring a knock-in GFP transgene in the Nanog and Oct4 loci . We analyzed expanded individual clones and pooled clones and obtained similar results . FACS analysis , prior to differentiation and at different time points after differentiation of these double transgenic ES cell lines , showed a negative correlation between RNF12-mCherry and NANOG-GFP expression , but not for RNF12-mCherry and OCT4-GFP ( Figure 2D , 2E , Figure S3 ) . Our findings therefore suggest specific counteracting regulatory roles for RNF12 and NANOG in XCI , which might include an inhibitory effect of NANOG on Rnf12 transcription . Interestingly , NANOG has been implicated in the regulation XCI by direct suppression of Xist in ES cells , and Xist suppression in the ICM of the developing blastocyst corresponds with up-regulation of NANOG expression [22] . Therefore , mutual exclusive expression of RNF12 and NANOG may be required for initiation of XCI . Recently , the first intron of Xist has been identified as a region involved in recruitment of three pluripotency factors , OCT4 , NANOG and SOX2 [14] . It was shown that down-regulation of Nanog and Oct4 , through gene ablation , resulted in an increase in Xist expression , and initiation of XCI in male cells . Interestingly , the intron 1-mediated suppression of XCI was suggested to directly act on Xist , without involvement of Tsix . To study if RNF12 might regulate XCI by interfering with binding of pluripotency factors to the intron 1 region of mouse Xist , we removed 1 . 2 kb of Xist intron 1 including all reported NANOG , OCT4 and SOX2 binding sites by homologous recombination with a BAC targeting construct , without disturbing the integrity of the Xist transcript . Targeted clones were screened by PCR amplification of a targeted RFLP ( BsrgI ) in female F1 2-1 , 129/Cas polymorphic ES cells , which was confirmed by Southern blotting , followed by Cre mediated loop-out of the kanamycin/neomycin resistance cassette ( Figure 3A , Figure S4 ) . Xist RNA FISH at different time points of differentiation of several Xistintron1+/− ES cell lines indicated that XCI is initiated with the same kinetics as in wild type cells , and showed that the intron 1 region is not required for repression of Xist in undifferentiated ES cells or early during initiation of XCI ( Figure 3B , 3C , and Figure S4G ) . Nevertheless , Xist specific RT-PCR , detecting a length polymorphism distinguishing 129 and Cas Xist , showed enhanced skewing at day 3 of differentiation towards 129 Xist expression , suggesting a role for the intron 1 region in suppressing Xist at later stages of differentiation , when NANOG , OCT4 and SOX2 are expressed at a lower level ( Figure 3D ) . To test an involvement of the intron 1 region in RNF12-mediated activation of XCI , we introduced an Rnf12 BAC transgene into the Xistintron1+/− ES cell lines . Additional copies of Rnf12 resulted in induction of Xist , even in undifferentiated ES cells ( Figure 3E , 3F , 3I ) , confirming our previous findings [16] . However , allele specific RT-PCR did not point to an increased preference for expression of the mutated or wild type allele , in undifferentiated ES cells ( Figure 3G , 3H ) , indicating that RNF12-mediated action on XCI does not require the Xist intron 1 region ( Figure 3J ) . At day 3 of differentiation , in several cell lines , we found higher expression of Cas Xist in Rnf12 transgenic Xistintron1+/− cells compared to Xistintron1+/− only cells . We attribute this finding to an increase in the percentage of cells with two Xist clouds . We conclude that the Xist intron 1 region is not essential for suppression of XCI in undifferentiated ES cells , but may play a role later during differentiation . Furthermore , RNF12-mediated activation of XCI is independent from the Xist intron 1 region . RNF12 could regulate XCI through activation of Xist or suppression of Tsix , or both . Previously , we analyzed Xist transgenic male ES cell lines with a BAC RP24-180B23 integration covering Xist only [16] , or a BAC RP23-338B22 sequence containing both Xist and Tsix ( Figure 4A ) . These male transgenic ES cell lines also contained 16 copies of an ms2 bacteriophage repeat sequence located in exon 7 of the endogenous Xist gene , allowing separate detection by RNA-FISH of autosomal versus endogenous Xist spreading [23] . Differentiation of transgenic male ES lines containing the Xist-Tsix transgene resulted in expression of Xist from the autosomal integration site in cell lines containing multicopy integrations . Autosomal spreading of Xist in these cell lines is most likely due to accumulation of enough Xist RNA to silence at least one copy of Tsix , allowing spreading of Xist in cis . Integration of truncated transgenes that lack Tsix would facilitate this process [16] . This also explained autosomal Xist spreading in BAC RP-24-180B23 single copy male transgenic ES cell lines upon differentiation , because Tsix is not covered by this BAC [16] . We used two of these , Xist only , BAC RP-24-180B23 ES cell lines to introduce 129 BAC RP24-240J16 transgenes covering Rnf12 , and found Xist spreading on the single endogenous X ( Figure 4B and 4C ) , confirming previous results . We also found a significant increase in the number of cells with autosomal Xist spreading , indicating that RNF12 activates XCI through Xist . Next , we introduced an Rnf12 transgene ( BAC RP24-240J16 ) in a single copy Tsix male transgenic ES cell line that lacks transgenic Xist ( BAC RP23-447O10 ) . These double transgenic ES cell lines contain a Cas X chromosome which allowed RFLP mediated discrimination of endogenous ( Cas ) and transgenic ( 129 ) Tsix . Analysis of these cell lines indicated that transgenic over-expression of RNF12 does not lead to down-regulation of Tsix , as measured by qPCR and by RNA-FISH examining the relative number of Tsix pinpoint signals ( Figure 4D , 4E , 4G ) . Interestingly , allele specific RT-PCR indicated that endogenous Tsix ( Cas ) is even down-regulated in samples with higher Xist expression , indicating Xist-mediated silencing of Tsix in cis ( Figure 4F ) . Taken together , these results indicate that Xist and not Tsix is the functionally most important downstream target of RNF12 . We previously found that the rate of initiation of XCI is reduced in differentiating female Rnf12+/− ES cells , compared to wild type ES cells [16] . The RNF12 protein level in these Rnf12+/− female cells is equal to that in male cells [16] , but XCI is still occurring at a higher rate than in male cells . This indicated the presence of additional X-encoded XCI activators , but did not exclude the possibility that RNF12 is essential for XCI . To address this point , we generated Rnf12−/− female ES cells by targeting the wild type Cas Rnf12 allele in Rnf12+/− ES cells ( Figure 5A ) . Correct targeting was confirmed by RT-PCR , showing loss of a targeted RFLP located in exon 5 of Rnf12 ( Figure 5B ) . The presence of two X chromosomes in these Rnf12−/− female ES cells was ascertained by X chromosome DNA FISH analysis and amplification of an RFLP in the Xist gene ( Figure 5C , and data not shown ) . Western blotting analysis confirmed the absence of RNF12 protein in the knockout cells ( Figure 5D ) . RT-PCR and qRT-PCR of pluripotency associated genes and differentiation markers gave information that differentiation of the Rnf12−/− ES cells was not different from that of wild type ES cells ( Figure 5E , 5F and Figure S5 ) . However , Xist RNA FISH analysis showed that differentiating Rnf12−/− ES cells only sporadically initiate XCI ( Figure 5G , 5H and 5I ) . QPCR analysis confirmed that Xist is not detectably up-regulated when measured for a population of Rnf12−/− cells upon differentiation . Moreover , DNA-FISH detecting a whole chromosome X paint probe at day 7 and 10 of differentiation excluded X chromosome loss ( Figure S5 ) . The few Rnf12−/− cells that initiated XCI appeared in clusters , suggesting clonal expansion of a few cells that initiated XCI ( Figure S5 ) . We therefore conclude that RNF12 is an essential factor in XCI . Evidently , the Rnf12−/− knockout cells present the possibility to study control of gene expression by RNF12 . Therefore , we next performed micro-array expression analysis comparing day 3 differentiated Rnf12−/− and wild type cells . We found that Xist was the only gene that was subject to differential regulation , showing pronounced down-regulation ( Figure 5J ) . Interestingly , none of the known downstream targets of RNF12 appeared affected in our analysis . This may be due to our ES cell differentiation system resulting in a mixed population of cells at different stages of differentiation . In addition , the 3-day-time span allowed in our studies for cell differentiation may have prevented detection of effects on downstream targets which are expressed at later stages of differentiation . Nevertheless , our results indicate that the main function of RNF12 at this early stage of differentiation concerns the regulation of XCI . The observed dependency of Xist transcription on RNF12 might be effectuated by RNF12 acting through the Xist promoter . To test this , we expressed Xist promoter luciferase reporter constructs , both transiently and stably , in wild type female and Rnf12−/− ES cell lines and differentiated these cells for 3 days . The results revealed an unequivocal correlation between RNF12 expression and luciferase expression ( Figure 5K ) . Our results therefore demonstrate that RNF12 activates the Xist promoter , although this does not exclude a role for other cis regulatory sequences , further away from the Xist promoter , in RNF12-mediated activation of XCI .
Here , we present evidence that RNF12 is an essential activator of random XCI . RNF12 acts in trans on the Xist promoter , in differentiating mouse ES cells , to activate Xist transcription , leading to Xist RNA cloud formation and spreading of the silencing complex over the future inactive X chromosome in cis . Although our results show that RNF12 acts in trans , it is to be expected that the close proximity of the Rnf12 gene to the Xist locus , taken together with the dose-dependent action of RNF12 , is quite crucial for well-tuned regulation of XCI . Such proximity most likely facilitates rapid down-regulation of Rnf12 in cis upon initiation of XCI , leading to a lower nuclear RNF12 content , thereby preventing inactivation of the second X chromosome . Whole genome expression analysis suggests that the major function of RNF12 in ES cells is its regulation of Xist RNA expression , hence XCI . This is a very surprising finding , as RNF12 has been implicated in many other biological pathways . Apparently , in the present cell differentiation system , loss of expression of RNF12 does not cause a deviation from the wild type differentiation process to such an extent that it affects gene expression other than that of Xist . However , also based on our studies we do not exclude a function for RNF12 at later stages of cell differentiation , or in mouse development . In addition , redundant pathways or proteins such as RNF6 , a close homologue of RNF12 , may prevent full phenotypic expression of loss of RNF12 . However , RNF12 exerts a predominant role in targeting Xist , as evidenced by our observation that Xist is largely silenced in the RNF12 deficient cells . While our manuscript was under review , Shin et al . ( 2010 ) published a paper suggesting that RNF12 might be required in particular for imprinted XCI in mice [24] . Remarkably , that study included the observation that RNF12 depletion did not prevent initiation of random XCI in a significant percentage of Rnf12−/− ES cells derived from mouse blastocysts . This discrepancy with our findings might be explained by experimental differences , such as differences concerning the design of the knockout , the genetic background of the ES cells , or the cell derivation and culture procedures . Differences in cell differentiation protocols have been shown to have a pronounced impact on the XCI process [25] . Also , ES cells derived from embryos with a different genetic background could express XCI activators and XCI inhibitors at different levels , allowing XCI in either a lower or a higher percentage of Rnf12−/− cells . Future studies comparing the two independently generated Rnf12−/− ES cell lines will yield useful information about these points . Although our observations provide evidence that RNF12 is an essential factor for the XCI process to occur in differentiating ES cells , we anticipate that other XCI activators act in parallel , and might independently regulate Xist or Tsix , or both . Dosage compensation mechanisms in species such as D . melanogaster and C . elegans also involve multiple factors and pathways , possibly leading to increased fidelity of these mechanisms [26] . In such a mechanism involving multiple factors , RNF12 would be the dose-dependent factor that is required to exceed the cumulative threshold limit to proceed towards initiation of XCI . It is feasible that female Rnf12−/− cells sometimes do initiate XCI ( Figure 6A ) , as a consequence of the stochasticity of the process . This would be compatible with a mechanism , in which the combined total activity of all putative XCI activators exclusive of RNF12 is just below or around the threshold to initiate XCI . Interestingly , Xist cloud formation is also sporadically found in male cells , but in contrast to female Rnf12−/− cells , this represents a lethal condition and will be selected against . Our studies indicate that RNF12 participates in Xist promoter activation , through an action which requires the presence of the minimal promoter . Although the direct protein target ( s ) of RNF12 remain elusive , its reported E3 ubiquitin ligase activity [17] would be compatible with RNF12 targeting an inhibitor of Xist transcription through proteasome-mediated degradation . This does not exclude that RNF12 might be involved , in addition or alternatively , in activation of a transcription factor driving Xist expression through positive regulation of transcription . Furthermore , RNF12 could be involved in regulation of cis-regulatory sequences other than the Xist promoter , yet to be identified and further away from the Xist locus . Selection against cells inactivating the X chromosome containing the wild type allele of Rnf12 in the heterozygous Rnf12+/− ES cells could point to a continued requirement for Rnf12 in maintaining Xist expression , following the early stages of differentiation . From the fact that male Rnf12−/Y knockout male mice are viable [24] , it can be concluded that RNF12 deficiency is compatible with survival of differentiated cells in which XCI does not play any role . Hence , it would be difficult to explain the observed selection against cells inactivating the wild type X chromosome in the heterozygous Rnf12+/− ES cells by loss of any possible function of RNF12 independent of XCI . If RNF12 would be required for maintaining Xist expression and XCI , the cells inactivating the wild type allele and becoming deficient in RNF12 can be expected to lose Xist expression and to reactivate the Xi . In contrast , cells inactivating the X chromosome containing the mutated allele , keeping one functional allele of Rnf12 , will be able to maintain Xist expression and XCI . In a population of cells this will lead in skewed XCI of the mutated allele . In fact , such a mechanism might also be relevant to explain the reported defect in imprinted XCI resulting from an Rnf12 mutation [24] . Imprinted XCI involves activation of Xist on the Xp , and the observed phenotype concerns lack of this imprinted XCI of the Xp when the mutant Rnf12 allele is inherited from the mother . It was observed that no female embryos were born , inheriting a mutated Rnf12 allele from either a Rnf12−/− or a Rnf12+/− mother in crosses with wild type males . In contrast , the mutated allele was transmitted to male offspring . Maternal storage of RNF12 in the oocyte was proposed to play a crucial role in imprinted silencing of the Xp in the early embryo [24] . Rnf12 is at a 46 cM distance of the centromere , so that it can be expected that many haploid oocytes generated by the first meiotic division ( the reduction division ) of Rnf12+/− oocytes , which occurs at the time of ovulation , will contain both wild type and Rnf12 mutated alleles , as a consequence of meiotic recombination . Hence , we anticipate that there will be ongoing expression of Rnf12 in a high percentage of oocytes transmitting the mutated Rnf12 allele , until fertilization triggers meiotic division II . The recombined wild type and mutant alleles which are present within one haploid oocyte , will be exposed to the same maternal storage of RNF12 . Taken together with the observation that Rnf12+/− oocytes did not give rise to female offspring carrying the mutant allele , whereas female offspring carrying the wild type allele were obtained at the expected mendelian ratio from these oocytes [24] , this argues against a predominant role for maternal storage in imprinted XCI . Rather , we favor the hypothesis that continued transcription of Rnf12 throughout ovulation and after fertilization is required for sustained expression of RNF12 , activation of Xist from the Xp , and maintenance of the inactive Xp . Future research will be required to address this hypothesis . Our results indicate a negative correlation between NANOG and RNF12 expression . NANOG and the other pluripotency factors OCT4 and SOX2 have been shown to be recruited to the Xist intron 1 region in undifferentiated ES cells , and were proposed to play a role in Tsix independent suppression of Xist [14] . In this regulatory mechanism , ablation of Tsix did not result in up-regulation of Xist in undifferentiated ES cells , and Tsix was not required for repression of Xist located on the inactivated paternal X chromosome in the inner cell mass . This pointed to an important role for recruitment of NANOG , OCT4 and SOX2 to Xist intron 1 in suppression of Xist in ES and ICM cells [14] . However , the present findings show that the intron 1 region is dispensable , in silencing the XCI process in undifferentiated ES cells . Deletion of Xist intron 1 caused an effect , but only in the form of skewing of XCI , which was notable at later stages of differentiation . Interestingly , a previous study analyzing an Xist mutant allele that lacks the intron 1 region but leaves the Xist promoter intact , also did not show up-regulation of the mutated allele in undifferentiated ES cells [27] . Although these latter results support our findings , they should be interpreted with caution because the selection cassette was still present in the cells analyzed by Marahrens et al . [27] . Like for the role of RNF12 , this points to the presence of additional mechanisms , involved in suppression of XCI . Tsix and Xite are the most likely candidate genes taking part , and the combined action of these repressive mechanisms may be sufficient to suppress Xist . However , even with all the repressive elements in place RNF12 can induce Xist expression and XCI in undifferentiated ES cells [16] . This points towards another mechanism involved in Xist suppression , in which the nuclear concentration of the XCI activator may be too low in undifferentiated ES cells and ICM cells to allow Xist expression and initiation of XCI , even in the absence of repressive elements such as the intron 1 region . Future research should clarify whether these mechanisms indeed act synergistically in silencing the XCI process . The negative correlation of RNF12 and NANOG expression that we report could reflect the differentiation state of the ES cells , and does not necessarily entail a cross-regulatory role for these proteins . Nevertheless , NANOG and other pluripotency factors are also recruited to the Rnf12 promoter in ES cells , where it might be involved in down-regulation of Rnf12 ( Figure 6B ) [28] , which opens the intriguing possibility that NANOG might also be implicated in regulation of the initiation of XCI through suppression of Rnf12 . This highlights the complexity of the overall mechanism and the interconnection of the different players involved in XCI , but also reinforces the predominant role of RNF12 in this process .
ES cells were grown in standard ES medium containing DMEM , 15% foetal calf serum , 100 U ml−1 penicillin , 100 mg ml−1 streptomycin , non-essential amino acids , 0 . 1 mM β-mercaptoethanol , and 1000 U ml−1 LIF . To induce differentiation , ES cells were split , and pre-plated on non-gelatinised cell culture dishes for 60 minutes . ES cells were then seeded in non-gelatinised bacterial culture dishes containing differentiation medium to induce embryoid body ( EB ) formation . EB-medium consisted of IMDM-glutamax , 15% foetal calf serum , 100 U ml−1 penicillin , 100 mg ml−1 streptomycin , non-essential amino acids , 37 . 8 µl l−1 monothioglycerol and 50 µg/ml ascorbic acid . EBs were plated on coverslips 1 day prior to harvesting , and allowed to grow out . For the Rnf12 rescue experiments , an Ampicilin-Puromycin resistance cassette was inserted in the backbone of BAC RP24-240J16 by homologous recombination in bacteria . The modified BAC was electroporated in to female heterozygous Rnf12+/− cells [16] , and colonies were picked after 8–10 days of Puromycin selection , expanded and differentiated . BAC copynumber was determined by qPCR , and transgene specific expression was determined by allele specific RT-PCR , as described previously [16] . To generate the female homozygous Rnf12 −/− ES cell line , the previously generated Rnf12+/− ES cell line was targeted with an Rnf12 BAC targeting construct containing an Ampicilin-Puromycin cassette disrupting the open reading frame of Rnf12 . To generate this targeting construct , targeting arms were PCR amplified using primers GCCTTCGAACATCTCTGAGC , GAGCCGGACTAATCCAAACA , cloned into pCR-BluntII-TOPO ( Invitrogen ) , and linearized with NheI to introduce an Ampicilin-Puromycin cassette from pBluescript . The targeting cassette was inserted in a Cast/Ei Rnf12 BAC RP26-81P4 by homologous recombination in bacteria , and the resulting construct was used to target specifically the Cast/Ei X chromosome of the Rnf12 +/− ES cell line . Colonies were selected under Neomycin and Puromycin selection , and the absence of Rnf12 expression was confirmed by Western analysis . To generate the Xist intron 1 deletion , a BAC targeting construct was generated by homologous recombination , replacing intron 1 by a floxed Neomycin cassette . Targetting arms were PCR amplified using primers 5′Forw:CATCAGGCTTGGCAGCAAGT , 5′R: CCTTGTTGGTCCAGACGACTATT and 3′Forw: CCAGACCAGGTCTTTGTATGCA , 3′Rev: GTGCTCCTGCCTCAAGAAGAA . Correctly targeted clones were identified by allele specific RFLP analysis using primers CAGTGGTAGCTCGAGCCTTT and CCAGAAGAGGGAGTCAGACG , followed by BsrGI digestion . The Neomycin cassette was removed by transient transfection with a CrePAC vector and selection with puromycin . The final cell lines were verified by Southern blotting . To generate the Rnf12 promoter cherry reporter cell lines , the Rnf12 promoter was PCR amplified using previous described primers [29] , and cloned into pCR-BluntII-TOPO and sequence verified . The Rnf12 promoter was then released from pCR-BluntII-TOPO by digestion with SacI and KpnI , and blunt cloned into an AseI-BamHI fragment from pmCherry-N1 ( Clonetech ) , thereby replacing the pCMV promoter of pmCherry-N1 with the Rnf12 promoter . The resulting construct was used to electroporate in Oct-GFP and Nanog-GFP ES cell lines . Both pooled cell lines and single colonies were expanded , and cherry expression was analysed by FACS analysis using a BD FACSAria apparatus . The Xist promoter was amplified using primers: TCCCAAGGTATGGAGTCACC , and GGAGAGAAACCACGGAAGAA , and cloned into pGL3-basic vector . As a control , the promoter less pGL3-basic vector was transfected . Stable pooled cell lines of wild type or Rnf12 −/− ES cells were generated by co-transfection with a puromycin or hygromycin selection vector . Expression of Luciferase was determined using the Bright-Glo luciferase assay system ( Promega ) and measured using a Promega luminometer . Results were normalized to the amount of protein present in the cell lysate measured by nanodrop , and copynumber of Xist promoter integration determined by qPCR . qRT-PCR using primers detecting luciferase ( TCTAAGGAAGTCGGGGAAGC and CCCTCGGGTGTAATCAGAAT ) confirmed the results obtained . For transient luciferase experiments , cells were co-transfected using the Xist reporter constructs and a control Renilla construct , using Lipofectamine 2000 . Luciferase activity was measured using the Dual Glo luciferase system ( Promega ) . Xist RNA-FISH was performed as described [11] , [16] . Immunofluorescence was performed using standard procedures . RNF12 and NANOG were detected using a mouse anti- RNF12 antibody ( 1∶250 , Abnova ) , and a rabbit anti-NANOG antibody ( 1∶100 , SC1000 , Calbiochem ) . ImageJ software was used to measure staining intensities; at least 100 cells were measured for each indicated time point , and background correction was performed . Western blotting was performed as previously described [16] . RNA was isolated using Trizol reagent ( Invitrogen ) using manufacturers instructions . DNAse treatment was performed , and cDNA was prepared using SuperScriptII ( Invitrogen ) , using random hexamers . qRT-PCR was performed using a Biorad thermocycler , using primers described in Table S1 . Results were normalized to Actin , using the ΔCT method . Whole genome wide expression analysis of female wild type and Rnf12−/− ES cells differentiated for 3 days was performed with Affymetrix Mouse Genome 430 2 . 0 Arrays . Differentially expressed genes were identified using Limma ( Bioconductor package ) in R software . | In all placental mammals , the males have only one X chromosome per diploid genome , as compared to the females who have two copies of this relatively large chromosome , carrying more than 1 , 000 genes . Hence , the evolution of the heterologous XY sex chromosome pair has resulted in an inevitable need for gene dosage compensation between males and females . This is achieved at the whole-chromosome level , by transcriptional silencing of one of the two X chromosomes in female somatic cells . Initiation of X chromosome inactivation ( XCI ) is regulated by X-encoded activators and autosomally encoded suppressors controlling Xist gene transcription . Spreading of Xist RNA in cis leads to silencing of one of the X chromosomes . Previously , we obtained evidence that the X-encoded E3 ubiquitin ligase RNF12 ( RLIM ) is a dose-dependent XCI activator . Here , we demonstrate that RNF12 exerts its action in trans and find that RNF12 regulates XCI through activation of transcription from the Xist promoter . Furthermore , analysis of female Rnf12−/− knockout ES cells shows that RNF12 is essential for initiation of XCI and that loss of RNF12 resulted in pronounced and exclusive down-regulation of Xist . It is concluded that RNF12 is an indispensable factor in Xist transcription and activation of XCI . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"developmental",
"biology",
"molecular",
"cell",
"biology",
"genetics",
"biology",
"evolutionary",
"biology",
"genetics",
"and",
"genomics"
]
| 2011 | RNF12 Activates Xist and Is Essential for X Chromosome Inactivation |
Acute renal failure is one of the most serious complications of envenoming resulting from Crotalus durissus terrificus bites . This study evaluated the relevance of hyperuricemia and oxidative stress and the effects of allopurinol and probenecid in renal dysfunction caused by direct nephrotoxicity of C . d . terrificus venom . Hematocrit , protein , renal function and redox status were assessed in mice . High ratio of oxidized/reduced glutathione and hyperuricemia induced by C . d . terrificus venom were ameliorated by both , allopurinol or probenecid , but only allopurinol significantly reduced the lethality caused by C . d . terrificus venom . The effectiveness of probenecid is compromised probably because it promoted hypercreatinemia and hypocreatinuria and worsed the urinary hypo-osmolality in envenomed mice . In turn , the highest effectiveness of allopurinol might be due to its ability to diminish the intracellular formation of uric acid . Data provide consistent evidences linking uric acid with the acute renal failure induced by C . d . terrificus venom , as well as that this envenoming in mice constitutes an attractive animal model suitable for studying the hyperuricemia and that the allopurinol deserves to be clinically evaluated as an approach complementary to anti-snake venom serotherapy .
Envenoming resulting from snake bite is recognized nowadays as one of the major neglected public health issue within poor communities living in the rural areas of several countries throughout the world . Because of serious misreporting , the true worldwide burden of snake bite is not known , but according to conservative country estimates used to calculate the regional estimates , Brazil had a fourth estimated number of envenomings annually ( 30 , 000 ) [1] . In Brazil , bites by snakes of genus Crotalus were responsible for 7 . 7% of these , but accidents involving this genus originate the highest mortality ( about 2% ) among registered snake bites [2] . Among the species of this genus , Crotalus durissus terrificus is the most frequently implicated in these accidents [3] . Because it is well vascularized , the kidney is a particularly vulnerable organ to venom toxicity [4] . In fact , the most serious complication of Crotalus snake bite is acute renal failure ( ARF ) [5] . The prospective survey of 100 cases of Crotalus bites followed from hospitalization to death or discharge in São Paulo , Brazil , revealed a high prevalence of ARF ( 29% ) in the first 72 hours after the bite with the case fatality ratio of 10% [5] . Two primary events of the direct nephrotoxic effect of the C . d . terrificus venom ( vCdt ) in mice are the oxidative stress in renal tissue and an incidence of 100% of hyperuricemia over an incidence of 60% of hypercreatinemia [6]–[7] . Although hyperuricemia has also been observed in human accidents with C . d . terrificus [8] and others species of snakes [9] , this parameter has not received any special attention as a relevant factor in the etiology of ARF , mainly because according to the recommendation of AKIN ( Acute Kidney Injury Network ) the prominent criteria to identify ARF should be the detection of changes in absolute values of serum creatinine , plasma urea and urinary volume [5] , [10]–[12] . However , it is well known that in many situations ARF is associated with a rise in plasma uric acid as a result of both increased generation and decreased excretion [13] . A marked hyperuricemia , currently defined as uricemia higher than the maximum value of normal range [14]–[15] , occurs with intrarenal urate crystal deposition leading to ARF [16]–[17] , but rats with uricemia near three times higher than the minimum value of normal range ( mild hyperuricemia ) [14]–[15] , as that caused in mice by vCdt [6]–[7] , also develop systemic hypertension , interstitial renal disease , afferent arteriolopathy , increased renin expression [18]–[19] and glomerular hypertrophy [20] . Experimentally , raising uric acid in rats can induce these dysfunctions via a crystal-independent mechanism [21]–[25] . It has also been reported that mild hyperuricemia in rats can induce oxidative stress in the kidney [15] and within the endothelial [21] and vascular smooth muscle cells [26] , as well as in adipocytes [27] . However , paradoxically the uric acid has been considered as circulating antioxidant which potently reacts with superoxide anion , peroxynitrite , chelates iron-based radicals [28] and prevents the oxidative inactivation of extracellular superoxide dismutase [29] . High plasma levels [30] or the infusion of uric acid [31] increased plasma antioxidant activity in humans . Regarding the controversy about the effects of uric acid on redox status , the involvement of uric acid and oxidative stress in ARF induced by vCdt is an attractive reason to investigate this envenomation . Moreover , clinical investigations have established that antivenoms are highly effective in the neutralisation of toxins responsible for systemic effects , but among the fatal cases of Crotalus bites in Brazil 5% are patients treated with antivenom [32] . Therefore , to know more about the role of uric acid in ARF and to highlight potential complementary agents for the treatment of envenoming by Crotalus snake bites , this study evaluated the effects of uricostatic ( allopurinol ) and uricosuric ( probenecid ) drugs on renal function ( hematocrit , protein , osmolality , creatinine , uric acid and urea ) and oxidative stress ( oxidized [GSSG] over reduced [GSH] glutathione index and content of malondialdehyde [MDA] ) in mice inoculated with vCdt .
The conducts and procedures involving animal experiments were approved by the Butantan Institute Committee for Ethics in Animal Experiments ( License number CEUAIB 717/2010 ) in compliance with the recommendations of the National Council for the Control of Animal Experimentation of Brazil ( CONCEA ) . 1 . 0 mg of lyophilized venom ( provided by the Instituto Butantan ) was suspended in 1 . 0 mL of sterile phosphate buffered saline ( PBS ) ( Na2HPO4 . 7H2O , 19 . 3 g/L; NaH2PO4 . H2O , 3 . 9 g/L; NaCl , 8 . 77 g/L; pH 7 . 4 ) , under mild mixing , for 10 min , at 4°C and , then , centrifuged at 10 , 192× g for 20 min at 4°C . The pellet was discarded and the supernatant was aliquoted and stored at −20°C at a maximum time of one week and administered intraperitoneally ( ip ) at a dose of 1 . 024 µg venom/20 g body mass ( 80%LD50 ) in a maximum volume of 0 . 2 mL [6]–[7] . The same lot of venom was used throughout this study . Allopurinol ( 4-Hydroxypyrazolo[3 , 4-d]pyrimidine ) ( Sigma , USA ) was dissolved in 1 M NaOH at a concentration of 50 mg/mL and subsequently diluted ( 1∶5 ) in PBS just before administration by gavage per oral ( po ) at a dose of 2 mg/20 g body mass in a maximum volume of 0 . 2 mL . Probenecid ( p-[Dipropylsulfamoyl]benzoic acid ) ( Sigma , USA ) was dissolved in 1 M NaOH at a concentration of 600 mg/mL and subsequently diluted ( 1∶5 ) in PBS just before administration po at a dose of 24 mg/20 g body mass in a maximum volume of 0 . 2 mL . Adult male Swiss mice , weighing 18–20 g , provided by the Animal Facility of the Instituto Butantan , were maintained in polyethylene cages ( inside length×width×height = 56×35×19 cm ) with food and water ad libitum , in a container with controlled temperature of 25°C , relative humidity of 65 . 3±0 . 9% and 12 h ∶12 h photoperiod light∶ dark ( lights on at 6:00 am ) . Animals were divided into eight groups , which received: ( i ) 0 . 2 mL PBS , ip ( control ip ) ; ( ii ) 0 . 2 mL PBS , po ( control po ) ; ( iii ) 0 . 2 mL PBS , ip and after 2 h , 0 . 2 mL PBS , po ( control ip+po ) ; ( iv ) 2 mg allopurinol in 0 . 2 mL PBS per 20 g body mass , po ( NL ) ; ( v ) 24 mg probenecid in 0 . 2 mL PBS per 20 g body mass , po ( PB ) ; ( vi ) 1 . 024 µg venom in 0 . 2 mL PBS per 20 g body mass , ip ( 80%LD50 ) ( vCdt ) ; ( vii ) 80%LD50 vCdt , ip and after 2 h , 2 mg allopurinol in 0 . 2 mL PBS per 20 g body mass , po ( vCdt+NL ) ; ( viii ) 80%LD50 vCdt , ip and after 2 h , 24 mg probenecid in 0 . 2 mL PBS per 20 g body mass , po ( vCdt+PB ) . Immediately after treatments , each group was placed in appropriate metabolic cages for urine collection , which was performed 24 h after venom injection . Pooled urine was centrifuged at 2 , 564× g , for 5 min , at 4°C; the supernatant was stored at −80°C , for the appropriate procedures . Immediately after urine collection , animals were anesthetized for blood and kidneys collection . It was monitored at intervals of one hour for 24 h , starting after administration of venom or vehicle or drugs alone . The animals were anesthetized with xylazin ( Calmiun , Agener União , Brazil ) 0 . 1% and ketamine ( Cetamin , Syntec , Brazil ) 1% ( ip , 0 . 1 mL/10 g body mass ) . Then , the blood was collected with heparinized Pasteur pipette after scission in right axillary plexus . The thoracic cavity was opened to perform cardiac perfusion with 50 mM PBS , over a period of 5 min at a flow rate of 8–10 mL/min . Immediately after perfusion , kidneys were removed , frozen in dry ice and stored for a maximum period of 10 days , at −80°C , until the use in the appropriate procedures . Measurement of hematocrit was made in duplicate of individual samples , in micro-hematocrit capillary tubes , centrifuged at 3 , 000 rpm for 5 min , at room temperature ( centrifuge HT model H240 ) . For plasma obtainment , blood was centrifuged individually at 5 , 232× g for 5 min , at 4°C . Total protein was measured , photometrically ( Bio-Tek Power Wave XR spectrophotometer ) , at 630 nm , in triplicates , in samples of plasma ( diluted 500-fold ) and pool of urine ( diluted 75-fold ) , by the method of Bradford [33] , using a Bio-Rad protein assay reagent ( Hercules , USA ) . Protein contents were extrapolated by comparison with standard curves of bovine serum albumin ( BSA ) in the same diluent . The measurements were performed as described by Marinho et al . [34] . Briefly , osmolality was determined in triplicates of 10 µL with a cryoscopic osmometer ( Osmette II Fisher ) and creatinine , uric acid , and urea were photometrically quantified in triplicate of individual plasma and pooled urine samples . Oxidative stress was evaluated on renal cortex and medulla from dissected kidneys stored at −80°C . GSSG and GSH were fluorometrically measured as described by Yamasaki et al . [6] . For this purpose , cortex and medulla were homogenized in 0 . 1 M phosphate buffer , with 0 . 005 M EDTA , pH 8 . 0 ( PBEDTA ) plus 5 . 26% HPO3 ( 0 . 1 g tissue/1 . 5 mL PBEDTA plus 0 . 4 mL 25% HPO3 ) , at 800 rpm for 3 min . These homogenates were ultracentrifuged at 100 , 000× g for 30 min . The pellet was discarded and the supernatant was immediately used . All steps were carried out at 4°C . MDA was assayed based on the method described by Selmanoglu et al . [35] . The reaction solution was prepared with 90 µL of 8 . 1% SDS; 675 µL of 20% acetic acid solution ( pH 3 . 5 adjusted with NaOH ) ; and 675 µL of 0 . 8% aqueous solution of thiobarbituric acid . 90 µL of tissue homogenates ( 10% [0 . 1 g/mL] prepared in 1 . 15% KCl ) was added to this solution and then the volume was made up to 1 . 8 mL with distilled water . This mixture was kept in a 98°C dry bath for about 1 h and subsequently centrifuged at 2 , 500× g for 10 min at 4°C . The supernatant was measured at 532 nm . The 1 , 1 , 3 , 3-tetraethoxypropane was used as a standard for MDA . The kidneys were fixed in 4% paraformaldehyde solution in 0 . 1 M NAOH and 0 . 1 M sodium tetraborate , processed in the routine histological processes , 10-µm-thick sagittal sections were prepared and stained with hematoxylin and eosin for light microscopy examination . Data are shown as mean ± standard error of the mean ( SEM ) and were analyzed using GraphPad Prism™ software packages . Regression analyses were performed to obtain standard curves of protein , 1 , 1 , 3 , 3-tetraethoxypropane , GSH and GSSG . One-way analysis of variance ( ANOVA ) followed , when differences were detected , by the Newman-Keuls test was performed to compare values among groups . Lethality data were analyzed by the two-sided Fisher's exact test . In all the calculations , a minimum critical level of p<0 . 05 was set .
Considering that various checked controls ( ip , po and ip+po ) presented statistically similar values for mortality , creatinine , uric acid and urea , the group that received vehicle ( PBS ) by ip and po routes ( group ip+po ) , both routes of administration of venom and drugs , was adopted as control . Table 1 shows that the lethality of 80%LD50 of vCdt ( 43% ) at 24 h was about the theoretically expected ( 40% ) and did not differ statistically from that of the vCdt group treated with probenecid ( 25% ) . The lethality rate at 24 h was null in all control groups and in those treated with allopurinol or probenecid alone , as well as significantly reduced about 58% ( from 43% to 18% ) in vCdt group treated with allopurinol . Most deaths occurred between 14–16 h and fewer between 16–18 h in envenomed mice . In envenomed mice treated with probenecid most deaths occurred between 16–18 h while in envenomed mice treated with allopurinol occurred between 18–20 h . Figure 1 shows that hematocrit and plasma osmolality did not differ among all examined groups . Allopurinol or vCdt alone , or the association of vCdt with probenecid caused hypercreatinemia , in comparison with control ( ip+po ) , but envenomed animals treated with allopurinol had a mitigation of this hypercreatinemia . The rise of uricemia induced by vCdt was normalized by allopurinol and probenecid . Compared with control , allopurinol alone caused hypouremia . Allopurinol associated with vCdt caused hyperproteinemia . Figure 2 shows that urinary urea was not affected by the treatments under study . Relative to control , urinary osmolality and urinary content of uric acid were diminished by vCdt . Allopurinol and probenecid alone caused a rise in urinary osmolality , but only the treatment of envenomed animals with allopurinol restored this parameter to a level of control group . Urinary content of uric acid in envenomed animals was normalized only by treatment with probenecid . The association of probenecid and vCdt caused hypocreatinuria . Allopurinol alone caused hyperproteinuria . Figure 3 shows that the pattern of changes on GSH , GSSG and GSSG/GSH ratio caused by all the treatments under study was similar between the renal cortex and medulla , in comparison with control ( ip+po ) . GSH was not altered by any treatment under study . GSSG and GSSG/GSH ratio were increased by vCdt . Probenecid diminished GSSG/GSH ratio in non-envenomed and both allopurinol and probenecid normalized GSSG and GSSG/GSH ratio in envenomed mice . Probenecid and vCdt did not affect MDA in the renal cortex but slightly diminished MDA in the renal medulla . Allopurinol decreased MDA in the renal cortex and medulla in envenomed and markedly in normal healthy mice . Histopathological changes such as edema , fibrosis and tubular necrosis were observed in envenomed mice , corroborating previous findings about direct nephrotoxic effects of this venom [5] , [32] , [36] . These changes were predominant in the cortex . The kidneys of animals treated with allopurinol or probenecid have much less intense and less numerous alterations , with an appearance that resembles that of the control animals . Some of these histological aspects were illustrated in Figure 4 .
ARF induced by vCdt through direct renal action in mice promotes histological changes , hyperuricemia , hypercreatinemia , increased renal GSSG/GSH ratio with unaltered lipidic peroxidation , hypo-osmolality , decreased excretion of uric acid in urine and death after a time course higher than 14 h . Epidemiological data compiling reports of snakebites in Brazil during 100 years show that the average time elapsed between snakebites in humans and the medical attendance is generally less than 6 hours [37] . In the case of envenomation by rattlesnake , urinary changes do not usually occur before the 12th hour [2] and the available data on the distribution of the number of attendances , according to the average time elapsed between the accident and the attendance is about 67% ( <3 h ) , 14% ( 3–6 h ) , 11% ( >6 h ) , 8% ( unknown ) [38] . In the present study allopurinol and probenecid administered 2 h after vCdt fully restore uricemia and renal GSSG/GSH ratio and ameliorate histopathological changes caused by this venom . Additionally , probenecid restores uricosuria , while allopurinol restores the normal levels of urinary osmolality in envenomed mice . Above all , allopurinol significantly decreases the lethality of vCdt . This is the first study correlating agents capable of reducing renal uric acid with the ARF induced by animal venoms . A typical property of allopurinol is to decrease the concentration of uric acid and urates relatively insoluble in tissues , plasma and urine . Allopurinol blocks the formation of uric acid , reducing its synthesis by competitive inhibition of xanthine oxidase [39] . Allopurinol is rapidly and extensively metabolised to oxypurinol , and its hypouricemic efficacy is due very largely to this metabolite [39] . Thus , its beneficial effect on envenomation by vCdt should be associated to the cell lysis caused by this venom , which contributes to the formation of uric acid and consequent deposit of urate . On the other hand , probenecid is an inducer of uric acid excretion in urine , without influence on its formation [40] , acting as an inhibitor of an organic anion transport exchanger that blocks the entry of uric acid into the cells [41]–[42] . Allopurinol is approved by the US Food and Drug Administration for a dose up to 800 mg/day and is available as a low-cost generic drug [43] . Allopurinol at the same dose used in the present study and administered orally for 7 days possess potent hypouricemic effect in mice [44] . Probenecid at the same dose and route used in the present study accelerates the uric acid excretion in mice [45] . A single dose of allopurinol by the same route and thirty times lower than that used here is effective as uricostatic [46] , while a fiftieth part of a single dose of probenecid by the same route used in the present study promotes a consistent uricosuric response in human subjects [47] . The same dose of allopurinol used here , but administered by intra-arterial infusion , totally abolishes the vascular oxidative stress [40] , while about a half of this dose administered by the same route improves peripheral vasodilator capacity and blood flow in humans with chronic heart failure [48] . In the present study , despite having played its excretory activity in envenomed mice , probenecid did not provide the same beneficial effect of allopurinol against the lethality ( it tended to reduce mortality , but without statistical significance ) . Furthermore , a rise of the antioxidant uric acid in the plasma paradoxically induced increased GSSG/GSH ratio and unaltered level of MDA in kidneys of mice envenomed by vCdt; and allopurinol and probenecid were both efficient in restoring this GSSG/GSH ratio to normal value . Other studies have reported that simvastatin [6] and lipoic acid [7] also ameliorated the GSSG/GSH ratio in the renal tissue , but these drugs did not affect the lethality caused by vCdt in mice . Thus , only the reduction of this index of oxidative stress was not responsible for the reduction of the lethality of vCdt . Furthermore , the antioxidant effect of simvastatin [6] , lipoic acid [7] , allopurinol and probenecid , as well as the lethality of vCdt could not be related simply to a uric acid-associated crystal-dependent mechanism . Therefore , what are the additional beneficial effects of allopurinol in this envenomation ? Allopurinol could be protective due to the blocking of xanthine oxidase-associated oxidants [49] as opposed to the simple uric acid-lowering effect of probenecid , since the pathway of xanthine oxidase , which leads to the synthesis of uric acid , can form reactive oxygen species such as MDA [35] , which can attack a wide variety of cellular components . In fact , the present study shows that besides reducing the uric acid , allopurinol was active on MDA-mediated lipid peroxidation in the kidney . However , we did not observe any significant changes in lipid peroxidation expressed by MDA level in envenomed mice as compared with healthy controls . Allopurinol at a dose twice lower has reported to be more effective than probenecid in improving endothelial function in patients with congestive heart failure , despite equivalent lowering of uric acid [40]; and in the present study the ratio between the doses of allopurinol and probenecid was 1∶12 . It is known that uric acid itself may cause endothelial dysfunction , which requires intracellular uptake of uric acid ( as noted by the ability of probenecid to block the effects of uric acid on vascular cells ) [21] . In this regard , allopurinol may be more effective at lowering intracellular uric acid levels when cellular production is high such as observed in heart failure [40] . These findings suggest that increased cellular production of uric acid is an important cause of the hyperuricemia induced by vCdt . Furthermore , uric acid , while being an antioxidant in the extracellular environment , has direct pro-oxidative effects once it gains entry into cells [30] . When uric acid reacts with oxidants such as peroxynitrite , it generates both radicals and alkylating species as it degrades peroxynitrite [50] . In the extracellular environment , these substances may dissipate into the circulation , but these substances are highly likely to be reactive with local constituents in the intracellular environment . In addition , uric acid may also be more likely to function as an antioxidant in a hydrophilic environment ( such as present in the extracellular environment ) as opposed to the primarily hydrophobic intracellular environment [51] . On the other hand , it is well documented that probenecid reduces the uric acid excretion rate at a low dose and accelerates it at a high dose , showing the so-called paradoxical effect [45] . The differential determinants of the low efficiency of probenecid compared to allopurinol against the lethality of vCdt could be associated with this paradoxical effect and/or with effects that occur only with the administration in envenomed animals . In fact , this last kind of effect of allopurinol is only the hyperproteinemia , while probenecid promotes hypercreatinemia , hypocreatinuria and aggravation of the urinary hypo-osmolality in envenomed mice . Based on this framework of evidences , the most likely mechanisms of differential action of allopurinol and probenecid on the lethality induced by vCdt are sketched in Figure 5 . In conclusion , the evidences presented here support the hypothesis proposed by Yamasaki et al . [6] that the hyperuricemia is involved in the early stages of ARF induced by direct nephrotoxic action of vCdt . Data shows that this envenoming constitutes an attractive animal model suitable for studying the hyperuricemia and that the therapeutic intervention with allopurinol at an early stage can prevent or recover its renal effects and especially prevent its lethality in mice . While one must be cautious in extrapolating animal models to human disease , this study provides a consistent evidence linking uric acid with the ARF induced by vCdt and should stimulate clinical trials to address whether allopurinol may contribute to anti-snake venom serotherapy . | In Brazil , among registered snake bites , those by the genus Crotalus originate the highest mortality rate . The rattlesnake Crotalus durissus terrificus is the most frequently implicated in these accidents . The kidney is a particularly vulnerable organ to the venom of this rattlesnake . In fact , the most serious complication of Crotalus snake bite is the renal dysfunction , and among the fatal cases of Crotalus bites in Brazil 5% are patients treated with antivenom . The hyperuricemia has been observed in human accidents with snake venoms , but this parameter has not received any special attention as a relevant factor in the etiology of renal dysfunction caused by these venoms . This study examined the effects of treatments with low-cost and low-risk uricostatic ( allopurinol ) and uricosuric ( probenecid ) drugs on the envenomation by C . d . terrificus , showing that allopurinol and probenecid mitigated certain nephrotoxic effects , as well as the survival of envenomed mice was improved through the effects of allopurinol on reduction of oxidative stress and intracellular formation of uric acid . This new knowledge provides consistent evidences linking uric acid with the renal dysfunction induced by rattlesnake bites and that the allopurinol deserves to be clinically evaluated as an approach complementary to anti-snake venom serotherapy . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"herpetology",
"public",
"health",
"medicine",
"biochemistry",
"public",
"health",
"and",
"epidemiology",
"biology",
"zoology",
"toxic",
"agents",
"toxicology"
]
| 2011 | Allopurinol Reduces the Lethality Associated with Acute Renal Failure Induced by Crotalus durissus terrificus Snake Venom: Comparison with Probenecid |
Trachomatous trichiasis is one of the leading causes of preventable blindness worldwide . A relatively simple surgery can spare vision . Although this surgery is usually performed free of charge in endemic regions , multiple studies indicate that surgical refusal is common . Prior studies have attempted to examine these reasons , although they generally rely on patient recall months to years after the surgery was offered . This study set out to determine major decision-making factors at the time of refusal . In addition , this study looked for ways to help increase surgical uptake by targeting modifiable factors . We used a combination of focus groups , interviews with community health workers , and individual interviews with trichiasis patients who refused surgery to understand their decision-making . We found that several factors influenced surgical refusals , including misconception regarding recovery time , inability to find a post-surgical caregiver , and the time of year of the surgical campaign . Fear of the surgery itself played a minimal role in refusals . Trichiasis patients refuse surgery for many reasons , but a large percentage is due to lack of information and education , and is , therefore , modifiable within the structure of a surgical outreach project . To address this , we developed a “frequently asked questions” ( FAQ ) document aimed at community health workers , which may have helped to decrease some of the misconceptions that had led to prior refusals .
Trachoma is an eye disease caused by the bacterium Chlamydia trachomatis . It is common in areas of the world that lack access to health care and clean water . Repeated infections cause recurrent conjunctivitis , which then leads to significant scarring of the eyelid . This scarring causes contraction of the inner surface of the eyelid , pulling the eyelashes inward to rub against the eye in a process known as “trichiasis” . The resulting irritation and damage to the cornea makes trachomatis trichiasis one of the leading causes of preventable blindness in the world [1] . According to the World Health Organization ( WHO ) , nearly 2 million people worldwide have vision loss due to trichiasis , with a further 190 million at risk and living within trachoma-endemic regions[2] . As part of their “SAFE” strategy for the elimination of blinding trachoma , WHO recommends surgery for the correction of trichiasis; this procedure has the ability to relieve pain from corneal trauma and prevent further vision loss . In addition , recent data from Ethiopia suggest that trichiasis surgery improves quality of life irrespective of post-surgery visual acuity improvement[3] . This surgery can be performed in an outpatient setting in less than half an hour , and is often performed free of charge to patients in areas with high need , with costs typically underwritten by donors . Despite the availability of free services , anecdotal and published evidence indicate that surgical refusal rates remain high [4 , 5] . This suggests that important drivers besides cost may affect the decision-making of trichiasis patients . Better understanding of these drivers can help tailor future interventions and large-scale trachoma elimination programs , making them more effective for both patients and donors alike . The goal of this project was to identify why people with trichiasis refuse free , corrective surgery . Reasons for refusal were obtained using interviews with community health workers ( CHWs ) ( many of whom had close relationships with their villagers and were knowledgeable regarding their reasons for not showing up ) and with the individuals who refused surgery . Although literature exists on patient satisfaction years after this procedure , [4 , 6–8] there is a lack of information obtained from patients at the time of their decision whether to undergo surgery . By conducting interviews on the same day that patients refused surgery , this study has the ability to examine patient reasoning with less recall bias .
The PRET Surgical Trial parent study was approved by the Johns Hopkins Medical Institutions and Wake Forest School of Medicine Institutional Review Boards and the National Institute for Medical Research in Tanzania . Participants who were interviewed regarding their reasons for surgical refusal ( which constituted minimal harm and had no “intervention” group ) provided implied oral consent in their native language ( Kiswahili or Kimakonde ) at the time when they were asked if they would like to talk with the interview team . The PRET Surgical Trial was a randomized controlled trial designed to investigate the performance of a novel surgical instrument , the TT clamp , [9] developed specifically for use in trichiasis surgery . It compared one of the WHO-endorsed procedures , the bilamellar tarsal rotation procedure ( BLTR ) , [10] performed using the standard method ( two hemostats ) to BLTR using the TT clamp . The study was conducted in Mtwara Region , southern Tanzania , which is a trachoma-endemic area . Surgeries took place between June and October 2009 . Results from the primary study were previously reported[11] . The PRET trial identified study participants at the village level through a multistep process . Trained district-level health personnel , accompanied by study team members , visited each study village on two separate occasions before the surgical outreach , once for an informational session and then again for a screening day . During the informational session , both village leaders and CHWs were familiarized regarding the process of TT treatment and were mobilized to assist on the upcoming screening and surgery days . In Mtwara region , CHWs are lay people without formal health education who act as point people for many different health initiatives and interventions . Village leaders and CHWs were tasked with spreading the information among their constituents . On the subsequent screening day , all residents with any eye problem were invited to receive an evaluation by a trained nurse , who determined a diagnosis and recorded patient demographics in a screening log . Villagers who were identified as having trichiasis at the screening were advised to return on a specific date for free , corrective surgery . Surgeries were performed by surgeons with experience performing trichiasis surgery in this region at a central health dispensary , which generally served several villages within a few-mile radius . Upon arrival at the surgery site , all eligible trichiasis-surgery patients were invited to participate in the PRET Surgical Trial . If they were interested , they were consented and then randomized to receive surgery using either the standard BLTR procedure or BLTR with the new instrument . Patients undergoing surgery who either were ineligible or declined consent to participate in the trial received surgery using the standard BLTR method . The names of all patients receiving surgery were recorded in a surgical log , which was maintained separately from the screening log for each village . The refusals ancillary study ran concurrently with the PRET Surgical Trial from 7 July through 18 August 2009 , which encompassed two focus groups held in villages in Masasi district and 19 surgical days in Mtwara Rural district . We first conducted two focus groups with trichiasis patients in Masasi district , in villages where surgeries had been performed two weeks prior . The goal of these discussions was to determine major themes of surgical refusal to help guide day-of-surgery individual interviews in Mtwara Rural district . Subsequently , during trichiasis surgical camps in Mtwara Rural district , we identified subjects we considered “refusals” because they screened positive for trichiasis but did not present for surgery . For the time period of this study , 575 individuals were identified as having trichiasis through village screening days , of which 464 underwent corrective surgery . Out of the 111 subjects who did not receive surgery , five had presented as instructed to the health dispensary , but PRET staff determined they had been inappropriately screened and did not have trichiasis , leaving a total of 106 refusals ( 18 . 4% ) . The count of refusals was determined by comparing the screening log to the surgical log for that village . We determined the names of refusals from cross-referencing the logbooks and then met with the associated CHWs to learn information regarding any known reasons for the absences . Using a convenience sample , we then conducted in-person interviews with patients who lived within a short walking distance of the health dispensary . We conducted interviews using an open-ended question format; patients initially were asked simply why they had not presented for surgery . Subsequently , their answers were explored with particular emphasis on themes that had emerged during the Masasi focus groups ( such as anecdotal experiences of friends and family with trachoma surgery , and their understanding of why a surgery was necessary and how it differed from other ways to manage trichiasis ) . A PRET study team member fluent in Kiswahili and English , and a community health worker who could provide translation between Kiswahili and Kimakonde ( the local language ) when needed , conducted the interview discussions , translating the findings into English for the visiting research team member , who then guided further questioning . Interviewers were instructed to maintain a non-judgmental style and were told that the goal of the interviews was to determine reasons for refusal , not to attempt to convince patients to consent to surgery . However , for ethical reasons , during the course of the interview if the participant expressed interest in undergoing the procedure , he/she was offered surgery . Patients who initially refused surgery were still counted as refusals for this ancillary study , even if they decided to undergo surgery after their interview . Using the results from focus groups , CHWs , and individual interviews , we prepared a frequently asked questions ( FAQ ) sheet and distributed it to CHWs at informational meetings between district teams and village leadership , with the goal of enabling village leaders to better educate their community regarding trichiasis surgery before the surgical outreach began .
Among the patients interviewed in Masasi district who had not received surgery , almost half ( four of nine ) reported confusion regarding the date that the surgical team was to be present . This sub-group expressed interest in having the procedure at a later date . For the rest , several themes emerged . The most common reason given for not undergoing surgery was anecdotal poor experiences from acquaintances and family , some of whom had previously undergone trichiasis surgery , and some of whom had clearly undergone other types of eye surgery ( e . g . cataract surgery ) . There was a widespread belief that the surgery itself would cause more problems . Others believed that because the pain associated with their trichiasis could be managed non-surgically ( with medications and epilation ) , surgery was not necessary and would offer no additional benefits . Finally , several people who declined surgery discussed fear of the pain associated with surgery as a major deterring factor . In the focus group of people who had initially refused , but then later requested and received surgery , fear and skepticism were the predominant reasons mentioned for initial refusal . Many admitted that they wanted to see how others in the village did after surgery , and when they saw the rapid recovery of their neighbors , they wanted the surgery for themselves . Members of this group thought that better education surrounding the surgery would lead to higher participation rates . Interestingly , several people in this group said that after their surgery they have encouraged others to get the surgery , citing its benefits and short recovery time . This was in contrast to reports from the group that continued to refuse surgery , where several members insisted that they had heard negative things about the procedure and recovery process from villagers who had undergone surgery . For those patients in Mtwara Rural who did not present for surgery , information obtained through in-person interviews and conversations with CHWs revealed several trends . For in-person interviews , the most common reason given for refusing surgery was lack of a post-operative caregiver . This rationale was augmented by the widespread belief that the recovery period from surgery was up to six months , during which time the patient would be unable to cook , help with household chores , or farm . However , even those who understood that the period of incapacitation was very short encountered difficulties , with numerous women reporting that they were unable to find anyone to cook for them for even one night . These findings were roughly mirrored in the responses given by CHWs , although workers were more likely to state that it was the perceived long recovery period ( rather than a lack of caregiver ) that prompted the refusal . PRET surgery recruitment in Mtwara Rural district coincided with cassava harvest . Many of the refusals who were unable to be interviewed in person were out working on their farms . Although this precluded the patients themselves from being interviewed , CHWs reported this season to be a significant factor in patient refusals . A less common , but not infrequent reason for refusal was lack of understanding regarding the need for trachoma surgery if medication was already giving effective pain relief . Many people with trichiasis either apply topical antibiotics or epilate their eyelashes to provide relief from the pain of their condition . The problem with both of these methods is that while they temporarily relieve the pain , they do not address the underlying problem ( eyelashes rubbing against the surface of the eye ) and may not prevent blindness [12] . In the absence of accurate information about the etiology of their disorder and how surgery could correct the problem , many villagers were choosing medication over surgery . The procedure to correct trichiasis was described as upasuaji mdogo ( literal translation: small surgery ) , which led many people to associate it with other types of surgery with which they had personal or family experiences . A number of anecdotes that subjects recounted in their decision to refuse surgery clearly involved other procedures , most often relating to childbirth or abdominal complaints . One person interviewed clearly stated she had been confused by the word “upasuaji” until it was further explained . No one interviewed individually gave fear of pain or fear of the surgery itself as a reason for refusal . In contrast , CHWs attributed some refusals to fear , although it was still not a leading reason for refusal . For the purpose of further exploring the issues surrounding fear of surgical pain , non-refusal patients were informally interviewed while waiting their turn for surgery . They reported that their trichiasis was so painful that they didn’t mind temporary surgical pain if it would bring them more permanent relief . Fear of the surgery among this group tended to manifest itself as fear of the more long-term consequences , such as the inability to see afterwards or inability to care for their family for many weeks or months . In two cases , villagers refused the surgery because they were already blind and no longer experienced pain from the in-turned lashes . These subjects felt that , because the surgery would not substantially restore the vision they had already lost , they would not benefit from the procedure . Summaries of reasons for surgical refusal are given in Tables 1 and 2 . A comparison of findings in focus groups and in individual interviews is given in Table 3 . When it became apparent that one of the primary drivers of high refusal rates was lack of understanding around trachoma , trichiasis , and the surgery , we created a frequently asked questions sheet ( FAQ , see S1 and S2 Appendix ) and distributed it to CHWs at the informational meetings between the district team and the village leadership . We designated the CHWs to serve as the primary educators of potential surgical candidates because we felt the candidates would have more trust in members of their own village than in district and study team members , who were not local to the community . We felt that if the CHWs were properly educated about the procedure and why it was necessary , they could share their knowledge to potential patients and to anyone else who had questions or concerns . The FAQ addressed issues such as how trichiasis leads to blindness , how the surgery is more effective than medication , the length of the recovery time , and the fact that the surgery does not involve the eyeball , only the eyelid . The FAQ was distributed for the remainder of the PRET recruitment period following the end of the refusals study . Reports from the field indicated that it was successful in reducing the number of refusals and increasing health worker knowledge about the procedure , although no formal comparison of refusal rates was performed .
This study found that important misinformation exists that limits people from deciding to undergo trichiasis surgery . These findings are applicable not only to trichiasis surgery programs , but to a broad range of health-related activities implemented at the community level , particularly in developing country settings where access to medical information is limited . In-person interviews with trichiasis patients and discussions with CHWs provided important insights into how inaccurate information and societal pressures can impact the effectiveness of a well-intentioned health care initiative . One key addressable concern was the pervasive belief that trichiasis surgery required a prolonged recovery time , which was a significant deterrent to undergoing surgery . Many believed that their recovery would take months , and that during that time they could not work . In a society that survives on subsistence farming , a months-long recovery period could have substantial economic consequences . This concern was especially apparent during the timing of this study , which coincided with cassava harvest season . In actuality , the recovery time is remarkably short , with patients cleared to remove their bandages the morning after surgery , and able to farm within a few days , even while stitches are still in place . Misconceptions about the recovery period highlighted a general paucity of experience with the formal health care system as well as a specific lack of information regarding this particular surgery . Through interviewing the CHWs ( who met with district team personnel prior to the screening days ) , it became clear that they wanted to convince their neighbors to have surgery , but were not armed with enough knowledge to counteract misconceptions . It was this finding that prompted the idea of an educational tool that could be implemented for future surgical screenings in an effort to reduce the rate of refusals . Roughly 75% of PRET trial participants were middle-aged or elderly women; these women are traditionally charged with most of the household tasks and are unable to spend time being incapacitated after surgery . Even among people who understood that the recovery period is short , a substantial proportion indicated that even a few days being unable to work and provide for their families was not feasible , and some indicated that they had no one to help them for even one night . 70–80% of trichiasis patients require bilateral surgery and , hence , would be unable to see for the night of surgery due to the bandaging . For women living alone , this can make surgery unattainable . Unfortunately , these single women are among the most in need , since they likely would have fewer options for care if they become blind from trichiasis . While these rural communities are closely-knit , this finding highlights the fact that stronger support systems are needed to ensure surgical services for those most in need . One method is to provide a structure wherein surgical services are offered again two weeks later , such that the patient can have one eyelid corrected at a time . Research into other methods that would provide better support structures for these most at risk individuals is needed . An interesting contrast between the focus group interviews in Masasi district and the individual interviews in Mtwara Rural district was the number of anecdotes highlighting poor health outcomes of family members among villagers in Masasi . In this area , trichiasis surgery has been provided for some time , and many people in the focus groups mentioned the bad experiences of friends or family as a reason why they were refusing surgery ( e . g . , my uncle had this surgery and went blind; my mother had this done and she got worse ) . In Mtwara Rural , an area that was generally surgery-naïve at the time of this study , no personal eye surgery anecdotes were given as reasons for refusal . It seems that in areas where no one has had the surgery before , there is little opportunity for the potential bad experiences of others to play a powerful role in limiting future surgical uptake . This study highlights the importance of good initial surgical outcomes in programmatic settings , as they are likely to set the standard for future surgical participation in the community . Hence , we strongly urge programs to take all possible steps to provide the highest quality surgery possible . This study is unique not only in the fact that it interviewed refusals at the time of surgery but also because it surveyed a population for which data also exist regarding satisfaction with trichiasis surgery and reported barriers to surgical uptake . At the follow-up visit two years after surgery , members of the PRET study staff interviewed nearly 500 participants in the Mtwara Rural district regarding their perceptions of the surgery and their outcomes . They found that 86% of patients were “very satisfied” overall , and that 83% felt that surgery had improved their quality of life . In addition , 97% said that they would recommend this surgery to others [6] . This is an optimistic finding , given the weight that anecdotal experience had during our focus groups in Masasi Rural district . There is also concern that as these regions become more familiar with this type of surgery , there is more opportunity for surgical recipients to dissuade their family and neighbors based on poor prior experience . Moving forward , it may be informative to assess current programs in this region to see how improved perceptions about surgery have changed surgical uptake over time . At the two-year visits , PRET surgical trial participants also were asked to recall if anything made it difficult to obtain the surgery . Only 11% of those surveyed reported facing challenges in obtaining the surgery ( although significantly more women reported difficulties than men ) , and the most common reasons reported were farming , cooking , and not being able to find a suitable caregiver [6] . These reasons , elicited two years post-surgery , are consistent with our findings on the day surgery is offered and serve to reinforce that these are some of the main barriers to overcome to increase rates of surgical uptake . This study has some important limitations . Due to the design of the study and the logistics of interviewing refusals , in-person interviews were only able to be obtained with those living within a short distance of the surgical site . The effect of this bias cannot be measured , but is presumed to underestimate the effect that travel distance might have played on a patient’s decision not to undergo surgery . A previous study noted that lower surgical refusal rates correspond to shorter distances between patients’ homes and the site of surgery[13] , which suggests that some of our patient refusals may have attributed their decision not to undergo surgery to long travel distances . This study also looks at a single region in rural Tanzania , and so conclusions about family/caregiver relationships , the role of women in household tasks , and concerns related to specific crop harvests cannot necessarily be generalized to other settings . However , subpar utilization of available health services is not an issue specific to Tanzania or to trichiasis surgery programs . In reality , it is encountered worldwide , and may be a major impediment to improving health outcomes in areas with low health care literacy and unfamiliarity with health care infrastructure . Lack of patient education and mistrust of the health care system has been implicated in refusals for interventions as diverse as lumbar punctures[14] , hospital births and cesarean sections [15–17] , cataract surgery [18] , malaria prophylaxis [19 , 20] , and general surgical procedures [21] . This study suggests that the educational deficits surrounding these interventions can be lessened if programs and providers undertake enhanced training of CHWs and village leaders . Based on our findings , we expect that addressing these knowledge gaps has the potential to make large-scale interventions more effective . | Many international health interventions in resource-limited settings involve treatments that are fully funded by donor organizations and , therefore , available for free to those in need . Trachoma , a bacterial eye disease that can cause blindness if left untreated , has frequently been the target of surgical outreach programs . However , even when the surgery is offered for free , a substantial number of eligible patients refuse surgery . This study explored why eligible patients did not present for surgery . When findings indicated that many patients did not understand the intervention and feared that it would be a major procedure or involve a long recovery time , we created a frequently asked questions document to address these misconceptions directly . This document was used at the surgical recruitment stage to better educate patients and community health workers , and it likely helped increase turnout for surgery . This study has implications beyond just eye surgery , and suggests steps that other global health interventions may want to consider to identify and address barriers uptake . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"and",
"health",
"sciences",
"sociology",
"tropical",
"diseases",
"geographical",
"locations",
"social",
"sciences",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"health",
"care",
"bacterial",
"diseases",
"human",
"families",
"tanzania",
"eye",
"diseases",
"neglected",
"tropical",
"diseases",
"eyes",
"patients",
"africa",
"blindness",
"infectious",
"diseases",
"head",
"people",
"and",
"places",
"visual",
"impairments",
"anatomy",
"ophthalmology",
"biology",
"and",
"life",
"sciences",
"ocular",
"system",
"cataract",
"surgery",
"trachoma",
"ophthalmic",
"procedures"
]
| 2018 | Why do patients refuse trichiasis surgery? Lessons and an education initiative from Mtwara Region, Tanzania |
In budding yeast , transcriptional silencing , which is important to regulate gene expression and maintain genome integrity , requires silent information regulator ( Sir ) proteins . In addition , Rtt106 , a histone chaperone involved in nucleosome assembly , functions in transcriptional silencing . However , how transcriptional silencing is regulated during mitotic cell division is not well understood . We show that cells lacking Dia2 , a component of the SCFDia2 E3 ubiquitin ligase involved in DNA replication , display defects in silencing at the telomere and HMR locus and that the F-box and C-terminal regions of Dia2 , two regions important for Dia2's ubiquitylation activity , are required for proper transcriptional silencing at these loci . In addition , we show that Sir proteins are mislocalized in dia2Δ mutant cells . Mutations in Dia2 and Rtt106 result in a synergistic loss of silencing at the HMR locus and significant elevation of Sir4 proteins at the HMR locus , suggesting that silencing defects in dia2Δ mutant cells are due , at least in part , to the altered levels of Sir4 at silent chromatin . Supporting this idea , we show that SCFDia2 ubiquitylates Sir4 in vitro and in vivo . Furthermore , Sir4 binding to silent chromatin is dynamically regulated during the cell cycle , and this regulation is lost in dia2Δ mutant cells . These results demonstrate that the SCFDia2 complex is involved in transcriptional silencing , ubiquitylates Sir4 , and regulates transcriptional silencing during the cell cycle .
Chromatin structure governs a host of cellular processes , including gene expression , DNA replication , and DNA repair [1] . In higher eukaryotic cells , chromatin is classified into two major forms , euchromatin and heterochromatin , based on cytological staining . Euchromatin , the less dense and gene-rich form of chromatin , is associated with active gene transcription , whereas heterochromatin has a more compact structure and is associated with gene silencing . Though poor in genes , heterochromatin is important for development , centromere formation and the maintenance of genome integrity [2]–[4] . Therefore , it is important to understand how heterochromatin is formed and inherited during S phase of the cell cycle . The budding yeast , Saccharomyces cerevisiae , has three heterochromatin-like loci: telomeres , the HM cryptic mating type loci ( HMR and HML ) and ribosomal DNA ( rDNA ) repeats . The initiation and maintenance of silent chromatin in budding yeast require Silent Information Regulator ( Sir ) proteins . While silencing at telomeres and the HM loci is regulated by many of the same factors , including Sir2 , Sir3 and Sir4 , only Sir2 is required for rDNA silencing . A stepwise model has been proposed for silent chromatin formation [5] , [6] . For instance , at the HMR locus , Sir1 and Sir4 are recruited to the E silencer , a DNA sequence element containing binding sites for the origin-recognition complex ( ORC ) and transcription factors Rap1 and Abf1 , through protein-protein interactions . Sir4 then recruits Sir2 , a NAD+-dependent histone deacetylase , which deacetylates lysine residues on histones H3 and H4 , including histone H4 lysine 16 ( H4K16 ) . This leads to the recruitment and binding of Sir3 and Sir4 to the adjacent nucleosome , as Sir3 and Sir4 bind hypoacetylated histones with higher affinity . This cycle of histone deacetylation and Sir protein binding to hypoacetylated nucleosomes leads to the spread of Sir proteins across the entire silent chromatin domain [5]–[7] . Despite the fact that protein factors and histone modifications involved in silent chromatin formation and maintenance in budding yeast are different from those in mammalian cells , this mechanism of step-wise formation of silent chromatin is likely to be conserved in higher eukaryotic cells [5] , [8] . Importantly , despite advances made in understanding chromatin structure and transcriptional silencing , how silent chromatin is inherited and maintained during S phase of the cell cycle remains elusive . During S phase of the cell cycle , nucleosomes ahead of the replication fork are disassembled to facilitate access of DNA replication machinery to DNA . Immediately following DNA replication , replicated DNA is reassembled into nucleosomes using both newly-synthesized histones and parental histones in a process called DNA replication-coupled nucleosome assembly . It is known that deposition of newly-synthesized H3–H4 requires histone H3–H4 chaperones , including CAF-1 , Asf1 and Rtt106 [9] . Various studies in budding yeast indicate that these histone chaperones function in two parallel pathways in transcriptional silencing: an Asf1 dependent pathway and a CAF-1 dependent pathway [10] , [11] . For instance , asf1Δ or rtt106Δ cells exhibit reduced silencing at both telomeres and the HM loci when combined with mutations in Cac1 , the large subunit of CAF-1 [10]–[13] . In addition , Sir proteins are mislocalized in cells lacking both Rtt106 and CAF-1 [11] . On the other hand , rtt106Δ asf1Δ double mutant cells do not exhibit synergistic silencing defects , suggesting that Rtt106 and Asf1 function in the same genetic pathway in transcriptional silencing [13] . These results support the idea that nucleosome assembly factors are important for proper formation and inheritance of silent chromatin structure in budding yeast [6] . Dia2 is an F-box containing protein that serves as the substrate recognition component of a SCF ( Skp1/Cullin/F-box protein ) ubiquitin E3 ligase . Cells lacking Dia2 exhibit gross chromosomal rearrangements and sensitivity to cytotoxic agents , indicative of a role for Dia2 in maintaining genome integrity [14]–[16] . Furthermore , Dia2 has been shown to be important during DNA replication [15]–[17] . The F-box domain in each of 11 known F-box containing proteins in budding yeast interacts with the SCF component Skp1 , enabling interactions between the ubiquitylation machinery and substrate [18] . In addition to the F-box domain , Dia2 has two additional important domains: a tetratricopeptide repeat ( TPR ) domain at the N-terminus involved in mediating Dia2's interaction with replisome components [14] , [17] and a leucine rich repeat ( LRR ) region at the C-terminus . The LRR domain in other F-box containing proteins is known to be involved in substrate binding [19] . In a genetic screen designed to identify genes that function in parallel to RTT106 in silencing at the HMR locus , we discovered that DIA2 , when deleted , enhances rtt106Δ silencing defects at the HMR locus . Structure-function studies revealed that the Dia2 F-box and LRR regions are important for transcriptional silencing . Furthermore , both Sir3 and Sir4 are mislocalized in dia2Δ cells , and Sir4 binding to the HMR locus is significantly elevated in dia2Δ rtt106Δ mutant cells . In addition , we show that Sir4 is ubiquitylated in yeast cells in a Dia2-dependent manner and that Sir4 levels on chromatin are cell cycle regulated , and this regulation is lost in dia2Δ mutant cells . Therefore , we suggest that the SCFDia2 E3 ligase functions in transcriptional silencing , in part through the regulation of Sir4 ubiquitylation .
We identified RTT106 in a screen for genes that function in parallel with PCNA in silencing at the HMR silent mating type locus [13] . Using a similar approach , we set out to identify genes that functioned in parallel with RTT106 in transcriptional silencing . Briefly , we used the synthetic genetic array ( SGA ) approach [20] , [21] to combine the rtt106Δ single mutant containing the HMR::GFP reporter gene with each of ∼4700 yeast deletion mutants . The HMR::GFP reporter contains the green fluorescent protein ( GFP ) integrated at the HMR silent mating type locus within the a1 gene; thus , GFP is silenced . Once double mutants containing the HMR::GFP reporter gene were selected , flow cytometry was used to identify those genes from the collection of mutants that when combined with rtt106Δ , resulted in a significant elevation in the percentage of GFP expressing cells . We identified five genes ( CAC1 , CAC2 , SIR1 , ARD1 , and DIA2 ) that enhanced the silencing defects of rtt106Δ cells . Cac1 and Cac2 are two subunits of the histone chaperone CAF-1 , and Sir1 is necessary for initiation of silent chromatin formation at the HM loci [22] , [23] . Deletion of SIR1 , CAC1 , or CAC2 is known to enhance the silencing defects of rtt106Δ mutant cells [11] , [13] . In addition , ARD1 is predicted to have a role in transcriptional silencing [24] , [25] . These results affirm that our screen was effective for identifying factors that enhance silencing defects of rtt106Δ cells . Dia2 is the F-box containing protein of the SCFDia2 ubiquitin E3 ligase involved in DNA replication , and it may also have a role in transcriptional regulation [16] , [26] . Therefore , we decided to focus our studies on Dia2 . To confirm our results , we deleted DIA2 from our standard genetic background ( W303 ) and assessed transcriptional silencing at the HMR locus using the HMR::GFP reporter . Cells defective for transcriptional silencing at the HMR locus express GFP and exhibit a rightward shift in the flow cytometry profile as observed for sir3Δ cells ( Figure 1A ) . Mutating DIA2 in rtt106Δ cells resulted in a rightward shift compared to single mutant cells ( Figure 1A ) , indicating an increase in the percentage of GFP expressing cells , which was quantified in Figure 1B . Moreover , dia2Δ single mutant cells exhibited an elevated percentage of cells expressing GFP compared to wild-type cells , suggesting a role for Dia2 in HMR silencing . To validate the flow cytometry analysis , the percentage of cells expressing GFP was also determined using fluorescence microscopy . Among the strains tested , a similar trend was observed using both methods of determining the percentage of cells expressing GFP ( Figure S1 ) . To confirm the effect of dia2Δ on HMR silencing , we determined how the loss of Dia2 affected the expression of the silenced a1 gene at the HMR locus . RNA was collected from single and double mutant cells of the alpha mating type , reverse transcribed , and cDNA analyzed using real-time PCR . Upon normalizing the expression of a1 to ACT1 , dia2Δ cells exhibited elevated a1 gene expression compared to wild-type cells ( Figure 1C ) . Furthermore , dia2Δ rtt106Δ had an even greater a1 gene expression level compared to either the dia2Δ or rtt106Δ single mutant . These results are consistent with the idea that DIA2 and RTT106 function in parallel to regulate transcriptional silencing at the HMR locus . Transcriptional silencing at telomeres and the HMR locus utilize similar mechanisms [6] . Moreover , it has been shown that Pof3 , the homolog of Dia2 in S . pombe , is required for maintaining telomere length and telomeric silencing [27] . We , therefore , determined whether Dia2 in S . cerevisiae had a role in telomeric silencing using cells containing the reporter gene , URA3 , integrated at the left arm of telomere VII . When plated on media containing 5-fluoroorotic acid ( FOA ) that is toxic to cells that express URA3 , wild-type cells survive because of telomeric silencing of the URA3 gene . Cells with telomeric silencing defects ( such as sir3Δ , used as a control ) exhibit growth defects in media containing FOA [24] . We found that dia2Δ cells exhibited defects in telomeric silencing compared to wild-type cells ( Figure 1D ) . We confirmed these observed telomeric silencing defects using RT-PCR to assess the expression of YFR057W , a gene found at the telomere on the right arm of chromosome VI and known to be silenced via a Sir-mediated mechanism [28] . Compared to wild-type and rtt106Δ cells , dia2Δ cells exhibited higher YFR057W gene expression , suggesting a telomeric silencing defect ( Figure 1E ) . Interestingly , deletion of RTT106 in dia2Δ mutant cells did not significantly increase expression of YFR057W , suggesting that deletion of RTT106 does not exacerbate the telomeric silencing defect of dia2Δ cells . These results demonstrate that Dia2 is required for efficient silencing at both the HMR locus and telomeres and suggest that Dia2 functions in a pathway parallel to Rtt106 in transcriptional silencing at the HMR locus , but not at telomeres . In addition to RTT106 , mutations in ASF1 , HIR1 and CAF-1 also result in silencing defects [12] , [13] . We , therefore , tested how loss of DIA2 affected silencing at the HMR locus in the absence of each of these H3–H4 histone chaperones . Each double mutant , dia2Δ asf1Δ , dia2Δ cac1Δ or dia2Δ hir1Δ , had a higher percentage of cells expressing GFP compared to the respective single mutants ( Figure 2A ) . These results suggest that Dia2 impacts transcriptional silencing at the HMR locus in a pathway parallel to each of the known H3–H4 histone chaperones . In addition to defects in transcriptional silencing , histone chaperone mutants are sensitive to DNA damaging agents [10] , [29] , [30] . Moreover , dia2Δ cells also exhibit sensitivity to a number of DNA damaging agents [15] , [16] . To determine whether the parallel action observed for DIA2 and H3–H4 histone chaperones extended beyond their function in transcriptional silencing , we tested the growth and DNA damage sensitivity of dia2Δ cells containing a mutation at each H3–H4 histone chaperone . The DNA damaging agents camptothecin ( CPT ) and methyl methanesulfonate ( MMS ) were used to assess genetic interactions in response to DNA damage . Synthetic interactions in growth and DNA damage sensitivity were observed for dia2Δ with each of the histone chaperone mutants tested ( Figure 2B , 2C and Figure S2 ) . The most dramatic effects were observed in dia2Δ asf1Δ and dia2Δ cac1Δ rtt106Δ mutants , with the interaction between DIA2 and ASF1 being the most pronounced . Furthermore , dia2Δ cac1Δ rtt106Δ cells grew slower on regular growth media ( YPD ) than dia2Δ and cac1Δ rtt106Δ cells and showed an increased sensitivity to CPT over dia2Δ , cac1Δ rtt106Δ and dia2Δ rtt106Δ cells . Taken together , these genetic interactions provide support for a role for Dia2 and the SCFDia2 complex in processes linked to nucleosome assembly . Post-translational modifications on newly synthesized histones work in concert with histone chaperones to regulate nucleosome assembly [9] . For instance , histone H3 lysine 56 acetylation ( H3K56Ac ) , catalyzed by the histone acetyltransferase Rtt109 , is important for regulating the interaction between histones and the histone chaperones CAF-1 and Rtt106 , and thus , H3K56Ac is a critical regulator of histone deposition [29] , [31] . H3 N-terminal tail acetylation , catalyzed by Gcn5 and Rtt109 , also serves as an important regulator of nucleosome assembly [32] . Finally , acetylation of histone H4 lysines 5 , 8 and 12 ( H4K5 , 8 , 12Ac ) , catalyzed by Hat1 and Elp3 , has also been implicated in nucleosome assembly [33] , [34] . Given the observed genetic interactions between dia2Δ and the H3–H4 histone chaperones involved in nucleosome assembly , we , therefore , determined how mutations in these important histone lysine residues affected the growth and CPT sensitivity of dia2Δ cells . For each histone mutant , the acetylated lysine residues were mutated to arginine to mimic the unacetylated state . We observed that dia2Δ cells showed significant growth defects and increased sensitivity towards CPT when combined with H3K56R , H4K5 , 12R and H3K9 , 14 , 18 , 23 , 27R ( Table 1 ) . Notably , we were unable to construct the dia2Δ H4K5 , 8 , 12R mutant via plasmid shuffling , suggesting that dia2Δ exhibited a synthetic lethal interaction with mutations at H4 lysines 5 , 8 and 12 ( Table 1 and Figure 2D ) . To confirm this result , wild-type and dia2Δ cells expressing wild-type histones H3–H4 from a uracil ( URA3 ) containing plasmid were transformed with either wild-type or mutant forms of H3-H4K5 , 8 , 12 on a plasmid with a histidine ( HIS ) selection marker . The wild-type H3–H4 histone plasmid ( URA ) could not be lost in dia2Δ H4K5 , 8 , 12R cells ( as no growth was observed when these cells were plated on FOA medium ) , whereas this plasmid was readily lost in dia2Δ or dia2Δ H4K8R cells ( Figure 2D ) . This suggests that Dia2 functions in parallel with acetylation of H4 lysine residues 5 , 8 and12 in maintaining cell viability . Together , these genetic analyses provide further evidence supporting the idea that the SCFDia2 E3 ligase has a role in a process linked to nucleosome assembly . Dia2 contains three primary functional domains: an N-terminal tetratricopeptide repeat ( TPR ) region , an F-box domain and a C-terminal leucine rich repeat ( LRR ) domain ( Figure 3A ) . The TPR region is important for Dia2's localization to the replisome [14] , [17] . The F-box domain is required for Dia2 ubiquitylation activity , as the F-box , in general , facilitates interactions with other SCFDia2 components [15] , [16] . The LRR region is proposed to be an interaction motif for substrates of the SCFDia2 complex [17] . To gain further insight into Dia2's role in transcriptional silencing , we deleted each of the three domains of Dia2 , expressed these dia2 mutants in dia2Δ cells and analyzed transcriptional silencing at the silent HMR locus and the telomere . Expression of Dia2 in dia2Δ cells restored the silencing at the HMR locus , as the percentage of cells expressing GFP was similar to wild-type cells ( Figure 3B ) . Expression of the Dia2-TPRΔ domain mutant also restored HMR silencing . Interestingly , deletion of the Dia2 F-box ( F-boxΔ ) or LRR domain ( LRRΔ ) did not rescue the HMR silencing defect . These results suggest that the Dia2 F-box and LRR domains are indispensable for silencing at the HMR locus . Similar to the effects observed at the HMR locus , the expression of full-length Dia2 in dia2Δ cells containing the URA3 reporter at telomere VIIL rescued the telomere silencing defects of dia2Δ cells to that of wild-type cells , and expression of Dia2 lacking the TPR region ( TPRΔ ) resulted in an almost complete rescue of dia2Δ telomere silencing defects ( Figure 3C ) . In contrast , expression of Dia2 lacking the F-box ( F-boxΔ ) or the LRR region ( LRRΔ ) was unable to rescue the telomere silencing defects of dia2Δ cells . While the expression of the TPRΔ and F-boxΔ mutants was similar to that of full length Dia2 in dia2Δ cells ( Figure S3 ) , expression of the Dia2 LRRΔ mutant was much less than the other mutants ( data not shown ) , most likely due to instability of the shortened form of the protein . Therefore , we made five additional Dia2 mutants , each with deletion of approximately 75 amino acids of the LRR domain: LRR ( amino acids 347–424 ) Δ , 425–502Δ , 503–580Δ , 581–658Δ , and 653–737Δ . Expression of these five mutants was , for the most part , similar to full length Dia2 and other Dia2 mutant forms ( Figure S3 ) . Importantly , dia2Δ cells expressing each of the LRR mutants exhibited defects in HMR ( Figure 3D ) and telomere silencing ( Figure 3E ) similar to dia2Δ cells transformed with empty vector . Together , these results suggest that silencing defects displayed in cells expressing dia2 mutants lacking the LRR or F-box are unlikely due to reduced expression of these mutant proteins . Instead , the F-box domain , essential for Dia2's role in protein ubiquitylation , and LRR region , predicted to be important for substrate recognition , are indispensable for SCFDia2's role in transcriptional silencing , suggesting that SCFDia2's role in silencing is likely mediated through its ability to ubiquitylate a substrate involved in transcriptional silencing . Sir proteins serve as structural components of yeast silent chromatin [6] . Sir3 and Sir4 form four to five foci at the nuclear periphery , which reflects the clustering of yeast telomeres [35]–[37] . Thus , we tested whether loss of Dia2 affected the localization of Sir3 and Sir4 , and therefore , silent chromatin structure , using fluorescence microscopy . As reported , wild-type cells expressing Sir3-GFP or GFP-Sir4 formed foci at the nuclear periphery [11] , [37] . While some dia2Δ cells had Sir3-GFP or Sir4-GFP foci patterns similar to wild-type cells ( Figure 4A , dia2Δ , panel 1 ) , a considerable percentage of dia2Δ mutant cells lost proper localization of Sir3 ( Figure 4B ) and Sir4 ( Figure 4C ) in dia2Δ or dia2Δ rtt106Δ cells . In the case of Sir3-GFP foci , some mutant cells had large areas of fluorescence without distinct foci ( Figure 4A , dia2Δ panels 2 and 3 ) . A commonly observed Sir4 pattern was multiple small foci ( >8 ) that were relatively non-distinct ( Figure 4A , dia2Δ panel 2 ) , in addition to cells containing areas of fluorescence with no distinct foci ( Figure 4A , dia2Δ , panel 3 ) . Because the dia2Δ mutation did not affect the protein levels of Sir3 and Sir4 to a significant degree ( Figure S4 ) , the mis-localization of Sir3 and Sir4 observed in dia2Δ mutant cells is likely due to changes in telomeric chromatin structure . To determine whether the localization of Sir3-GFP and Sir4-GFP was dependent upon particular Dia2 domains , we expressed the Dia2 mutants lacking specific domains ( see Figure 3 ) in dia2Δ cells . Expression of full-length Dia2 or Dia2 TPRΔ restored the percentage of cells containing wild-type Sir3-GFP or Sir4-GFP foci closer to the percentage observed for wild-type cells , whereas cells expressing dia2 mutants lacking the F-box ( F-boxΔ ) or a portion of the LRR domain [LRR ( 425–502Δ ) and LRR ( 581–658Δ ) ] did not ( Figure 4D–4E and data not shown ) . Thus , the F-box and LRR regions are important for proper localization of Sir3 and Sir4 , consistent with the idea that Dia2's role in transcriptional silencing is dependent upon its ubiquitylation activity . To further analyze how the dia2Δ mutation affects the binding of Sir proteins to chromatin , chromatin immunoprecipitation ( ChIP ) assays were performed in unsynchronized wild-type , dia2Δ , dia2Δ rtt106Δ and control cells ( sir3Δ ) using antibodies against Sir4 , Sir2 , and Sir3 . ChIP DNA was analyzed using real-time PCR with primers amplifying different positions at the HMR locus or the telomere region on the right arm of chromosome VI ( Tel-VIR ) . Consistent with published results , Sir protein binding at silent chromatin loci was lost in sir3Δ cells , and more Sir proteins bound to silent chromatin than the corresponding active chromatin loci tested at both the HMR locus and telomere ( Figure 5 and Figure S5 ) [11] . Compared to wild-type cells , dia2Δ cells exhibited a slight elevation in Sir4 binding at HMR silent chromatin ( a1 gene , HMR silent ) ( Figure 5A ) , whereas dia2Δ rtt106Δ cells had a much larger and significant increase in Sir4 binding at the silent HMR locus than wild-type and dia2Δ cells ( Figure 5A ) . Similarly , we also observed significantly more Sir2 binding at the HMR locus in dia2Δ rtt106Δ mutant cells compared to wild-type and dia2Δ mutant cells ( Figure 5B ) . In contrast , Sir3 binding to the silent HMR locus was not altered to a significant degree in either dia2Δ or dia2Δ rtt106Δ cells when chIP was performed using an antibody against endogenous Sir3 ( Figure S5A ) or performed using IgG beads in strains in which Sir3 was tagged with a tandem affinity purification ( TAP ) tag ( Sir3-TAP ) ( Figure 5C ) . Notably , the dia2Δ or dia2Δ rtt106Δ mutation did not affect mRNA levels of Sir4 or Sir2 ( Figure S6 ) , suggesting that the increase in Sir4 and Sir2 proteins at the HMR locus is not likely due to increased gene transcription in mutant cells . In addition , the overall protein levels of Sir3 and Sir4 were not altered to a detectable degree in cell extracts ( Figure S4 ) . Therefore , the observed elevation in Sir4 and Sir2 at the silent HMR locus in dia2Δ cells is not likely due to increased steady-state levels of Sir proteins and is likely due to elevated Sir2 and Sir4 levels on chromatin . Sir protein binding was also assessed at telomeric silent chromatin . No significant change in Sir4 binding was detected at telomeric silent chromatin ( Tel-VIR , silent ) in dia2Δ cells compared to wild-type cells ( Figure 5D ) . Sir4 binding to the telomeric silent chromatin in dia2Δ rtt106Δ cells was not increased significantly compared to wild-type ( p-value for three independent experiments is 0 . 1 ) . Thus , levels of Sir4 proteins are unlikely altered at telomeres in dia2Δ rtt106Δ mutant cells compared to wild-type cells . These results are consistent with the result in Figure 1 showing that deletion of RTT106 synergistically increases the silencing defect of dia2Δ cells at the HMR locus , but not the telomere . Sir2 ( Figure 5E ) and Sir3 ( Figure S5B and S5C ) binding to telomeric silent chromatin was not altered to a detectable degree in either dia2Δ or dia2Δ rtt106Δ cells compared to wild-type cells . Taken together , it is likely that mutations in Dia2 alter the chromatin binding of Sir4 and Sir2 , but not Sir3 , in dia2Δ rtt106Δ cells , and these imbalanced alterations in Sir protein binding to silent chromatin may contribute to the silencing defects observed for dia2Δ and dia2Δ rtt106Δ cells , especially at the silent HMR locus . Overexpression of Sir4 is known to result in silencing defects , most likely due to disruptions in Sir protein stoichiometry [38] , [39] . To further test whether the effect of dia2Δ on silencing is due to altered Sir4 protein levels on chromatin , we investigated the effects of exogenous expression of Sir4 on telomeric silencing , as transcriptional silencing at telomeres is perturbed more easily than that at the HMR locus [40] . Wild-type and dia2Δ cells were transformed with a centromere plasmid for expression of Sir4 under the control of its own promoter . RNA was then extracted for quantitative RT-PCR analysis of YFR057W expression . While ectopic expression of Sir4 in wild-type cells did not result in obvious changes in expression of YFR057W , Sir4 expression in dia2Δ cells resulted in over a two-fold increase in YFR057W expression compared to dia2Δ cells transformed with empty vector ( Figure 5F ) . This suggests that telomeric silencing in dia2Δ cells is more sensitive to changes in Sir4 levels than that of wild-type cells . These results are consistent with the idea that altered Sir4 levels at chromatin contribute to the silencing defects observed in dia2Δ cells . Since dia2Δ rtt106Δ cells exhibited elevated Sir4 binding at silent chromatin , we hypothesized that SCFDia2 may ubiquitylate Sir4 . To test this idea , we first determined whether cell extracts prepared from cells expressing MYC-Dia2 or MYC-Dia2 F-boxΔ integrated at the endogenous DIA2 locus ( Figure 6A , left panel ) could ubiquitylate Sir4 purified from dia2Δ cells using TAP purification . Briefly , Flag–ubiquitin ( Flag-Ub ) , E1 , E2 ( Cdc34 ) and the respective whole cell extract ( E3 ) were incubated with Sir4-CBP ( calmodulin binding peptide , part of the TAP tag ) . As negative controls , reaction mixtures containing all components except E1 , E3 ( cell extract ) or substrate were assembled . Following ubiquitylation , Sir4 was pulled down using calmodulin beads , and ubiquitylated species were detected via Western blot using antibodies against the Flag epitope . More Sir4-associated ubiquitylated species were detected in the reactions using the extracts from full length Dia2 than from those using extracts from cells expressing Dia2 lacking the F-box domain ( Figure 6A , right panel ) . No ubiquitylated species were detected in control reactions lacking E1 , E3 or Sir4 ( substrate ) . Thus , SCFDia2 ubiquitylates Sir4 and/or a Sir4-associated protein in vitro . To test this idea further , the SCFDia2 complex was purified from yeast cells expressing MYC-Dia2 and MYC-Dia2 F-boxΔ and used to ubiquitylate Sir4 purified from yeast cells as described in Figure 6A . More ubiquitylated species were detected in reactions containing purified SCFDia2 ( Dia2 ) than those containing the complex purified using the Dia2 F-box mutant ( F-boxΔ ) ( Figure 6B ) . Taken together , these experiments indicate that SCFDia2 ubiquitylates Sir4 , or a protein associated with Sir4 , in vitro , and this ubiquitylation depends on the F-box domain of Dia2 . To provide additional evidence that SCFDia2 ubiquitylates Sir4 , we tested whether Sir4 bound to SCFDia2 in vitro . Beads containing the MYC-Dia2 complex purified for in vitro ubiquitylation reactions in Figure 6B or control beads were incubated with two different amounts of in vitro translated 35S-methionine labeled Sir4 . After the beads were washed , bound 35S-Sir4 was detected via autoradiography , and the presence of the MYC-Dia2 was detected via Western blot using antibodies against the Myc epitope . Compared to control reactions , more 35S-Sir4 signal was detected in samples containing the MYC-Dia2 complex ( Figure 6C , compare lanes 3–4 to lanes 1–2 of lower panel ) despite the fact that more MYC antibodies ( IgG ) could be detected by CBB staining in control samples ( Figure 5C , middle panel ) than samples containing Myc-Dia2 ( Figure 5C , upper panel ) These data further support the idea that Sir4 is a substrate of the SCFDia2 complex . Next , we asked whether Sir4 is ubiquitylated in vivo and whether this ubiquitylation depends on Dia2 using our published procedures [41] . Briefly , Sir4-TAP was purified from wild type or dia2Δ cells transformed with a plasmid expressing HA-tagged ubiquitin ( HA-Ub ) . Following TAP purification , ubiquitylated species were detected by Western blot with antibodies against the HA epitope . In wild-type ( Sir4-TAP ) cells , ubiquitylated protein species co-purified with Sir4 , and these immunoprecipitated species were specific as no HA signal could be detected in controls ( a strain without Sir4-TAP but containing HA-Ub or a strain with Sir4-TAP but no HA-Ub ) ( Figure 6D , lanes 1 and 2 , respectively ) . Importantly , these ubiquitylated species were notably reduced in dia2Δ cells compared to wild-type cells , despite equal levels of purified Sir4 ( Figure 6D , compare lane 4 to lane 3 ) . These results suggest that the SCFDia2 complex ubiquitylates Sir4 in vivo . Interestingly , we observed similar amounts of Sir2 co-purified with Sir4 in wild-type cells and dia2Δ cells , suggesting that the Sir2-Sir4 interaction is not affected to a detectable degree in dia2Δ mutant cells ( Figure S7 ) . To further confirm that Sir4 is ubiquitylated in vivo , we purified ubiquitylated species using a two-step purification procedure . First , Sir4-TAP was purified from wild-type and dia2Δ mutant cells expressing either HA-tagged ubiquitin or His-HA tagged ubiquitin ( Figure 6E ) . The associated ubiquitylated species were then purified under denaturing conditions using Ni-NTA beads that bound to His-ubiquitin . A band with a similar size to Sir4 was detected from wild-type cells expressing His-HA ubiquitin , but not from dia2Δ cells expressing His-HA tagged ubiquitin . In addition , little signal was detected from cells expressing HA-ubiquitin ( lanes 1 and 2 ) , suggesting that co-purification of Sir4 with ubiquitin under denatured conditions was specific ( Figure 6E ) . Together , these data strongly support the conclusion that Sir4 is a substrate for SCFDia2 . It has been shown that heterochromatin is dynamically regulated during S phase of the cell cycle in mammalian cells . During mitosis , phosphorylation of histone H3 serine 10 by Aurora B kinase displaces HP1 from heterochromatin [42] , [43] . In S . pombe , this dynamic loss of HP1 protein from heterochromatin is proposed to facilitate transcription of siRNA required for the re-establishment of heterochromatin during S phase [44] . Therefore , we asked whether Sir4 proteins at chromatin are also dynamically regulated during mitotic cell division and whether this regulation depends on Dia2 . Briefly , wild-type and dia2Δ mutant cells were arrested at G1 using α-factor and then released into the cell cycle . Cells were collected at 0 , 30 and 60 minutes following release for analysis of DNA content by flow cytometry ( Figure 7B ) , for ChIP assays using Sir4 antibodies ( Figure 7C ) and for analysis of telomere gene expression by RT-PCR ( Figure 7D ) . Flow cytometry analysis indicated that both wild-type and dia2Δ mutant cells collected at 0 , 30 and 60 minutes were predominantly at G1 , S and G2/M phase of the cell cycle , respectively ( Figure 7B ) . In wild-type cells , Sir4 binding at telomeric silent chromatin was significantly reduced as the cells proceeded from G1 to S and G2/M phases ( Figure 7C ) . In contrast , Sir4 binding was not altered at telomeric silent chromatin in dia2Δ cells as the cells progressed through the cell cycle . To determine whether the observed changes in Sir4 binding affect silencing at telomeres , RT-PCR was performed to assess the expression of the telomere gene , YFR057W , as the cells progressed through the cell cycle . Consistent with Figure 1E , dia2Δ cells exhibited higher expression of YFR057W at all time points compared to wild-type cells ( data not shown ) . When normalized against the expression of YFR057W at G1 , we observed that there was an increase in gene expression of the telomere gene , YFR057W as cells progressed from G1 through to G2/M phase in wild-type cells ( Figure 7D ) . Interestingly , only minor fluctuations in YFR057W expression were observed in dia2Δ cells . Together , these results suggest that Sir4 proteins at silent chromatin in budding yeast are regulated during the cell cycle , and this regulation is dependent on Dia2 . Given our observation of SCFDia2's role in Sir4 ubiquitylation , we suggest that Sir4 ubiquitylation , mediated by SCFDia2 , regulates the binding of Sir4 at silent chromatin during the cell cycle .
The SCFDia2 complex is known to have a role in DNA replication [16] . However , how SCFDia2 functions in DNA replication is not clear . Our genetic analysis suggests that SCFDia2 may function in DNA replication-coupled nucleosome assembly . First , we show that the dia2Δ mutation exhibits synthetic defects in growth and sensitivity towards DNA damaging agents when combined with mutations in CAF-1 , Asf1 and Rtt106 , histone H3–H4 chaperones known to be involved in replication-coupled nucleosome assembly . Second , we show that the dia2Δ mutation exhibits synthetic defects with mutations at lysine residues of histones H3 and H4 known to be involved in the regulation of replication-coupled nucleosome assembly . Surprisingly , the dia2Δ mutation is synthetic lethal with mutations at H4 lysine residues 5 , 8 and 12 . Acetylation of these lysine residues occurs on newly synthesized H4 and is conserved from yeast to human cells [9] . Interestingly , DIA2 shares many genetic interactions with RTT101 [45] , another ubiquitin ligase functioning in DNA replication [41] , [46] . Genetic analysis using the epistatic miniarray profile ( E-MAP ) approach indicates that Rtt101 functions in the same genetic pathway as Rtt109 , the histone H3 lysine 56 acetyltransferase known to be involved in DNA replication-coupled nucleosome assembly [29] , [31] , [47] . These results , combined with ours presented here , suggest that the SCFDia2 ubiquitin E3 ligase may function in nucleosome assembly . Further investigation is needed to determine Dia2's exact role in this process . Using RT-PCR and reporter genes integrated at the HMR locus and telomeres , we show that Dia2 is needed for efficient transcriptional silencing at both the HMR locus and telomeres . Cells with defects in nucleosome assembly are known to affect transcriptional silencing [11] , [12] , [48] . Because genetic studies presented here suggest that Dia2 has a role in nucleosome assembly , it is possible that Dia2 impacts silencing through its role in nucleosome assembly . However , we have presented several lines of evidence supporting the idea that SCFDia2 impacts transcriptional silencing , at least partly through ubiquitylation of Sir4 . First , we show that the role of SCFDia2 in transcriptional silencing depends on the Dia2 functional domains ( the F-box and LRR ) involved in protein ubiquitylation . Second , SCFDia2 ubiquitylates Sir4 in vitro and in vivo and interacts with Sir4 in vitro . Third , in wild-type cells , the levels of Sir4 at the silent telomere locus are reduced when cells proceed from G1 to S and G2/M phase of the cell cycle , and this regulation is not observed in dia2Δ mutant cells . Importantly , the reduction of Sir4 at telomere silent chromatin correlates with increased transcription of YFR057W , a gene located near a telomere , during the cell cycle . Thus , we suggest that SCFDia2 functions in transcriptional silencing in budding yeast , at least partly through ubiquitylation of Sir4 . It is not unprecedented for an E3 ubiquitin ligase to be involved in transcriptional silencing . In fact , several ubiquitin ligases and hydrolases have been shown to have roles in transcriptional silencing in budding yeast and other organisms . In S . pombe , the Rik1-Cul4 E3 ligase complex is important for the recruitment of the Clr4 histone methyltransferase for RNAi-mediated heterochromatin formation . Deletion of the Rik1-Cul4 complex results in silencing defects [49]–[51] . In budding yeast , Rtt101 , a cullin protein of a ubiquitin E3 ligase complex , is important for telomeric silencing [52] . Rad6 , involved in ubiquitylation of H2B , also impacts transcriptional silencing in budding yeast [53] , [54] . Finally , Sir4 is known to interact with two ubiquitin hydrolases , Ubp3 and Dot4/Ubp10 [55] , [56] . Ubp10 is proposed to regulate transcriptional silencing by deubiquitylating H2B [57] . How Ubp3 is involved in silencing is still unknown . Interestingly , loss of Upb3 results in improved silencing at telomeres and the HM loci ( HML ) [55] , the opposite effect as that of loss of Dia2 . How does Sir4 ubiquitylation affect transcriptional silencing ? Protein ubiquitylation , in general , is known to mediate two distinct functions . First , protein ubiquitylation marks proteins for degradation by the 26S proteasome . Second , protein ubiquitylation can also regulate protein-protein interactions [58] , [59] . Interestingly , we did not detect significant changes in the steady-state levels of Sir4 in dia2Δ cells , suggesting that Sir4 ubiquitylation by SCFDia2 is not likely involved in regulating the steady-state level of Sir4 . However , we did observe a significant increase in Sir4 proteins at the HMR silent locus in dia2Δ rtt106Δ double mutant cells . It has been reported that β-catenin levels on chromatin , but not steady state levels , are regulated by ubiquitylation through a protein complex containing the histone acetyltransferase component TRRAP and Skp1 [60] . Therefore , it is possible that a SCF ubiquitin ligase can regulate protein levels on chromatin . We have shown previously that Rtt106 binds Sir4 [11] . While the functional significance of the Rtt106-Sir4 interaction is not clear , it is possible that this interaction regulates Sir4 binding to silent chromatin . This could explain why loss of Rtt106 leads to aberrant accumulation of Sir4 at silent chromatin in dia2Δ rtt106Δ mutant cells . Therefore , we suggest that Sir4 ubiquitylation by SCFDia2 regulates Sir4 levels at chromatin , which in turn regulates silencing . Supporting this idea , we show that elevation of Sir4 levels using a centromere plasmid , while having no apparent effect on telomeric silencing in wild-type cells , reduces telomeric silencing in dia2Δ mutant cells . It has been observed that heterochromatin proteins such as HP1 are dynamically regulated during mitotic cell division . For instance , in human cells , phosphorylation of serine 10 of histone H3 ( H3S10ph ) during mitosis reduces the binding affinity of HP1 towards H3K9me3 , which propels the dissociation of HP1 from heterochromatin [42] . In S . pombe , dissociation of the sequence homolog of HP1 , Swi6 , from heterochromatin via H3S10ph results in transcription of siRNA during S phase . This , in turn , helps to maintain heterochromatin during S phase of the cell cycle [44] . These results highlight the fact that heterochromatin in S . pombe and mammalian cells is dynamically regulated during mitotic cell division . It was previously unknown whether silent chromatin in budding yeast was also regulated during S phase of the cell cycle . We observed that Sir4 binding to telomeric silent chromatin was significantly reduced as cells progressed from G1 to S and G2/M phase of the cell cycle . Concomitant with the reduction of Sir4 binding , the transcription of the telomere gene , YFR057W , increased when cells entered S phase . These results demonstrate that budding yeast silent chromatin is also dynamically regulated during S phase of the cell cycle . Because HP1 and histone modifications equivalent to H3K9me3 and H3S10ph are not present in budding yeast , we propose that perhaps SCFDia2-mediated ubiquitylation of Sir4 serves as a mechanism to regulate Sir4 proteins and silent chromatin structure during the cell cycle . Further investigation is warranted to address such a role for the SCFDia2 complex and Sir4 ubiquitylation . In summary , our studies reveal a role for the SCFDia2 E3 ligase in transcriptional silencing . In addition , we show that the SCFDia2 E3 ligase binds and ubiquitylates Sir4 . Furthermore , Dia2 is required for the regulation of Sir4 binding to chromatin during S phase of the cell cycle . These studies reveal a novel mechanism by which yeast silent chromatin is regulated during S phase of the cell cycle .
All yeast strains , except those for the SGA screen , were derived from W303 ( leu2-3 , 112 ura3-1 , his3-11 , trp1-1 , ade2-1 can1-100 ) and constructed using standard methods and can be found listed in Table S1 . The synthetic genetic array method to screen 4 , 700 viable yeast deletion mutants has been previously described in detail , along with the assay used to screen specifically for mutants that exhibit defects in HMR silencing [13] , [21] . Detailed methods for the screen can be found in Text S1 . Plasmids for Dia2 and Sir4 expression were constructed using standard methods in the vector , pRS313 . Oligos used to construct the mutant Dia2 plasmids , as well as those used for analysis of mRNA expression and ChIP DNA using real-time PCR , are listed in Table S2 . Telomeric silencing was analyzed as described previously [13] . Briefly , yeast cells containing the URA3 gene at the left arm of chromosome VII ( URA3-VIIL ) were plated onto the indicated media in a 10 fold series dilution with a starting OD600 of 6 . 0 for growth on 5-fluoroorotic acid ( FOA ) and OD600 0 . 6 on media not containing FOA . Images were taken after four days of incubation at 30°C . Assays for silencing at the HMR silent mating type locus were performed as described [13] in cells containing a GFP reporter integrated at the silent HMR locus . Cells were grown at 25°C to OD600 0 . 6–0 . 8 and washed three times with synthetic complete ( SC ) –TRP media . The percentage of cells expressing GFP was determined using flow cytometry with the GFP populations in wild-type and sir3Δ cells as standards . Cells in which Sir3 or Sir4 were tagged with GFP were grown in YPD or selective media ( SC-HIS ) at 25°C and collected at OD600 0 . 6–0 . 8 . Cells were washed three times with SC-TRP media and analyzed using a Zeiss fluorescence microscope . Images were taken with z-stack images captured at every 0 . 3 µm , and one z-stack image was shown in Figure 4A . At least 100 cells of each genotype were counted from at least two independent experiments , and the percentage of cells exhibiting foci similar to wild-type cells was reported . ChIP assays were performed as described [11] . Cells were first fixed with 1% formaldehyde and then quenched with glycine . Cells were collected and homogenized using glass beads . Chromatin DNA was sheared to an average size of 0 . 5 to 1 kb by sonication and immunoprecipitated with specific antibodies against the protein of interest . The co-precipitated DNA was analyzed by real-time PCR using primers whose sequences are listed in Table S1 . To analyze Sir4 levels during the cell cycle , cells were arrested for 3 hours with α-factor . After washing away α-factor with cold water three times , cells were released into fresh medium and collected at different time points for analysis of DNA content , gene expression and ChIP assays as described above . Ubiquitylation assays were performed as described [41] . Briefly , Flag-ubiquitin , E1 , E2 , E3 ( whole cell extracts prepared from cells expressing full length MYC-Dia2 or MYC-Dia2 F-boxΔ or purified MYC-Dia2 ( full length and F-boxΔ ) . More details for both the in vitro and in vivo ubiquitylation assays can be found in Text S1 . | Heterochromatin is important for the maintenance of genome stability and regulation of gene expression . Heterochromatin protein 1 ( HP1 ) , a protein that binds to histone H3 methylated at lysine 9 ( H3K9me3 ) at heterochromatin loci in mammalian cells , is dynamically regulated during the cell cycle by phosphorylation of histone H3 serine 10 ( H3S10ph ) . Compared to mammalian cells , transcriptional silencing at budding yeast silent chromatin requires silent information regulator ( Sir ) proteins , and H3K9me3 and H3S10ph are not present in budding yeast . Therefore , it is not known whether and how silent chromatin in budding yeast is regulated during the cell cycle . Here , we show that the SCFDia2 ubiquitin E3 ligase complex regulates transcriptional silencing . We show that SCFDia2 ubiquitylates Sir4 , a structural component of yeast silent chromatin , and that Sir4 levels decrease during the cell cycle in a Dia2-dependent manner . Concomitant with the reduction of Sir4 at telomeric silent chromatin during the cell cycle , the expression of a telomere-linked gene increases . Therefore , we propose that transcriptional silencing at budding yeast silent chromatin is regulated during the cell cycle , in part by SCFDia2-mediated Sir4 ubiquitylation on chromatin . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"biochemistry",
"genetics",
"biology",
"genetics",
"and",
"genomics"
]
| 2012 | The SCFDia2 Ubiquitin E3 Ligase Ubiquitylates Sir4 and Functions in Transcriptional Silencing |
Receptor-targeted lentiviral vectors ( LVs ) can be an effective tool for selective transfer of genes into distinct cell types of choice . Moreover , they can be used to determine the molecular properties that cell surface proteins must fulfill to act as receptors for viral glycoproteins . Here we show that LVs pseudotyped with receptor-targeted Nipah virus ( NiV ) glycoproteins effectively enter into cells when they use cell surface proteins as receptors that bring them closely enough to the cell membrane ( less than 100 Å distance ) . Then , they were flexible in receptor usage as demonstrated by successful targeting of EpCAM , CD20 , and CD8 , and as selective as LVs pseudotyped with receptor-targeted measles virus ( MV ) glycoproteins , the current standard for cell-type specific gene delivery . Remarkably , NiV-LVs could be produced at up to two orders of magnitude higher titers compared to their MV-based counterparts and were at least 10 , 000-fold less effectively neutralized than MV glycoprotein pseudotyped LVs by pooled human intravenous immunoglobulin . An important finding for NiV-LVs targeted to Her2/neu was an about 100-fold higher gene transfer activity when particles were targeted to membrane-proximal regions as compared to particles binding to a more membrane-distal epitope . Likewise , the low gene transfer activity mediated by NiV-LV particles bound to the membrane distal domains of CD117 or the glutamate receptor subunit 4 ( GluA4 ) was substantially enhanced by reducing receptor size to below 100 Å . Overall , the data suggest that the NiV glycoproteins are optimally suited for cell-type specific gene delivery with LVs and , in addition , for the first time define which parts of a cell surface protein should be targeted to achieve optimal gene transfer rates with receptor-targeted LVs .
Cell entry as first step in the viral replication cycle is initiated by the attachment of virus particles to distinct cell surface proteins . While many viral receptors have been identified , there is only limited knowledge available about the molecular requirements that cell surface proteins have to fulfill to act as entry receptors and why they have been chosen during viral evolution [1] . Paramyxoviruses encode two envelope proteins required for cell entry , the receptor attachment protein and the fusion protein ( F ) which mediates fusion of the viral and cellular membranes upon receptor contact . Three types of attachment proteins can be distinguished , the hemagglutinin-neuraminidase ( HN ) , the hemagglutinin ( H ) and the glycoprotein ( G ) , which in contrast to the others has no hemagglutinating function . All attachment proteins are type II membrane proteins with a membrane proximal stalk domain and a propeller-like head domain [2] . While HN proteins use sialic acid as receptor , morbillivirus H and henipavirus G recognize proteinaceous receptors . Due to this and its separated attachment and fusion functions , the measles virus ( MV ) H protein has been the first viral attachment protein that was successfully engineered to use a cell surface protein of choice for entry instead of its natural receptor [3] . While this approach suggested a high flexibility in receptor usage for MV , it was also of applied relevance for the engineering of tumor–specific oncolytic viruses [4] and when combined with pseudotyping for the generation of cell-type specific lentiviral vectors ( LVs ) . With LVs as a major tool , gene therapy has developed to one of the most important technologies in modern medicine for the treatment of monogenetic diseases as well as various cancer types [5–7] . LVs mediate stable long-term expression and integration of transgenes into the genome of transduced cells . The commonly used LVs for therapeutic applications are pseudotyped with either the glycoprotein G of the vesicular stomatitis virus ( VSV ) or the envelope ( Env ) proteins of γ-retroviruses such as murine leukemia virus ( MLV ) or , more recently , the baboon retrovirus [8] . Since the use of all these glycoproteins result in LVs with a broad cellular tropism allowing gene transfer into a variety of cell types , further modifications in vector design have been established to restrict gene transfer into the cell type relevant for a given application . The concept of engineering vector particle entry relies on targeting the particles to a cell surface protein of choice which is then used as entry receptor [9 , 10] . By picking surface proteins that are selectively expressed in a particular cell type , gene transfer can be restricted to this cell type . Natural receptor usage is destroyed through mutating specific residues in the attachment protein and the desired receptor usage is achieved through displaying a polypeptide ( targeting domain ) exhibiting high affinity for the targeted receptor . So far , the envelope glycoproteins from three different viruses have been successfully engineered to generate such receptor-targeted LVs . While all available receptor-targeted LVs work in principle , they also have certain disadvantages making their broad application and translation into the clinic difficult . LVs pseudotyped with receptor-targeted Sindbis virus glycoproteins have been developed for a large variety of cell types but are limited by the non-covalent linkage of the targeting domain , the requirement for low pH membrane fusion triggering provided by efficient endocytosis of the targeted receptor as well as insufficient selectivity , which often has been compensated by the use of cell-type specific promoters [11–13] . A similar large number of cell surface proteins has been targeted by engineered MV glycoproteins , resulting in LVs exhibiting high selectivity for their target cells even when combined with the strong and ubiquitously active spleen focus forming virus ( SFFV ) promotor [10 , 14] . MV-based receptor-targeted LVs , however , can only be produced at moderate titers and are susceptible towards neutralizing antibodies induced by MV vaccination , thus preventing multiple dosing in patients . While the latter problem may be circumvented by engineered Tupaia paramyxovirus ( TPMV ) glycoproteins , this system turned out to be too inefficient in vector production [15] . To address the disadvantages of the MV- and TPMV-based systems , while keeping their high selectivity , we generated here for the first time receptor-targeted glycoproteins derived from a henipavirus . Nipah virus ( NiV ) is naturally harbored by fruit bats and can cause fatal illness in humans with respiratory and encephalitic symptoms [16] . As no vaccination programs against NiV exist , neutralizing antibodies in the human population are vanishingly small . Previous studies showed that LVs can be pseudotyped with the NiV glycoproteins resulting in vector stocks with high titers , although different cytoplasmic tail truncations were found to be optimal [17–19] . These results suggest that the NiV glycoproteins may be more suited for the generation of receptor-targeted LVs than those of MV or TPMV . Here we show , that LVs pseudotyped with the engineered NiV glycoproteins delivered genes to their target cells as selectively as MV-based receptor-targeted LVs but could be produced at substantially higher titers . Notably , receptor-targeted NiV-LVs turned out to be highly sensitive towards the position of their binding site on the targeted receptor , with membrane-proximal positions being preferred over membrane-distal ones .
Previous studies showed that LVs can be pseudotyped with the NiV glycoproteins resulting in vector stocks with high titers [17–19] . While these studies showed that successful pseudotyping is achieved by cytoplasmic tail truncation of the fusion protein F , only Witting et al . ( 2013 ) [19] found that the glycoprotein G had to be truncated as well , while others came to different conclusions [17–19] . In order to generate targeted NiV-LVs , we therefore first tested the previously described cytoplasmic tail variants of NiV G and F ( Fig 1A ) . As target receptor we chose the human epithelial cell adhesion molecule ( EpCAM ) , a putative marker of early tumor cells [20] , and fused the EpCAM-specific DARPin Ac1 [21] to the ectodomain of the G protein cytoplasmic tail variants . All nine combinations of the resulting GEpCAM and F ( Fig 1A ) variants were assessed for their ability to mediate transfer of the gfp gene into CHO-EpCAM cells . CHO cells lack the natural NiV receptors ephrin-B2 and ephrin-B3 and are not susceptible for NiV . Gene transfer into CHO-EpCAM cells must therefore have been mediated by entry via human EpCAM . All NiV-LVs tested showed transduction of CHO-EpCAM cells ( Fig 1B ) . The highest titers were obtained when the cytoplasmic tail of the F protein was truncated by 22 residues and that of the GEpCAM protein by 33 or 34 amino acids , with a slight but not significant advantage for Gc∆34 , resulting in unconcentrated titers of 4-5x105 t . u . /ml . The differences in titers were not due to differences in the cell surface expression levels of the GEpCAM variants , since HEK-293T cells transfected with plasmids encoding the three different constructs showed similar high surface expression levels ( Fig 1C; S1 Fig ) . Next , incorporation of the GEpCAM variants into LV particles was investigated . Particles pseudotyped with all nine combinations of GEpCAM , Gc∆33EpCAM or Gc∆34EpCAM with F , Fc∆22 or Fc∆25 were produced , normalized by particle numbers and analyzed via Western blot analysis . As controls , we also produced LV stocks pseudotyped with untruncated but His-tagged G/F ( GHis-LV ) or with Gc∆34His/Fc∆22 ( NiVwt-LV ) . All G and F variants were incorporated into vector particles ( Fig 1D ) . When correlated to the intensities of the p24 signals , Gc∆33EpCAM and Gc∆34EpCAM showed substantially higher incorporation levels than GEpCAM and GHis carrying full-length C-tails ( S2 Fig ) , which corresponds well to the higher vector titers observed for these constructs . The combination Gc∆34EpCAM and Fc∆22 was used for further vector productions . As the ratio of G to F in the viral envelope can influence virus-cell fusion and thus LV entry , we aimed to determine the optimal ratio of plasmids . Ratios of plasmids ranging from a tenfold excess of F- over G-encoding plasmid to a tenfold excess of G over F were tested . Overall , higher amounts of the Gc∆34EpCAM encoding plasmid reduced LV titers ( Fig 1E ) . The optimal ratio was determined to be a five-fold excess of the Fc∆22 encoding plasmid , which was used for all further LV productions . As the Gc∆34EpCAM protein described above still recognizes the natural NiV receptors ephrin-B2 and B3 , we aimed at completely restricting gene transfer to hEpCAM+ cells by destroying the natural receptor usage of NiV-G in the next step . For this purpose , six point mutations , previously described to influence the natural receptor recognition and cell entry of Nipah or Hendra virus [22–24] , were introduced into Gc∆34EpCAM individually or in combinations . All mutated residues localize in the contact area of NiV-G and ephrin-B2 ( Fig 2A ) [25 , 26] . For all Gc∆34EpCAM proteins with a single point mutation the position and type of mutation is indicated in their respective designation . The three variants carrying combined mutations were termed as follows: Gc∆34EpCAMmut2 . 1 includes mutations E501A+W504A , Gc∆34EpCAMmut2 . 2 mutations Q530A+E533A , and Gc∆34EpCAMmut4 mutations E501A , W504A , Q530A and E533A . U87-MG cells , a glioblastoma cell line known to be highly susceptible for NiV infections , was used to determine ephrin-B2/B3 receptor usage by the LVs pseudotyped with the mutated Gc∆34EpCAM proteins . To control that the mutations in Gc∆34EpCAM did not affect the fusion-helper function of G but selectively inhibited recognition of ephrin-B2/B3 , LVs were assessed for their ability to transduce ephrin-B2/B3 negative CHO-EpCAM cells . As expected , LVs pseudotyped with Gc∆34 and Fc∆22 transduced U87-MG cells but not CHO-EpCAM cells , while Gc∆34EpCAM carrying LVs transduced both cell types ( Fig 2B ) . Of the six single mutations three showed a significant negative impact on gene delivery into U87-MG cells ( Fig 2B , black bars ) , with the mutation at position 533 ( E533A ) being most effective . Combining Q530A with E533A showed a slight advantage over the combination of E501A with W504A or the single mutations . None of the mutations significantly influenced the transduction of CHO-EpCAM cells ( Fig 2B , white bars ) . To clearly demonstrate that the mutations in Gc∆34EpCAM interfere with binding to the natural NiV receptors , the ability to bind soluble recombinant ephrin-B2 and ephrin-B3 was analyzed by a flow cytometry-based binding assay . Interestingly , even though mutant E533A and Q530A+E533A ( mut2 . 2 ) showed only very low levels of transduction , binding of ephrin-B2 , but not of ephrin-B3 , could still be detected ( Fig 2C and 2D ) . Only binding of ephrin-B3 was clearly impaired indicating that LV entry into U87-MG is mediated by binding of NiV-G to ephrin-B3 ( Fig 2D ) . Importantly , when combining the four mutations in construct Gc∆34EpCAMmut4 , binding of recombinant ephrin-B2 dropped to background levels as well ( Fig 2C ) . Therefore , this mutant was used for all further experiments to ensure entry and binding to be ephrin-B2/B3 independent . To further characterize the NiVmutEpCAM-LV particles , their size was determined by single nanoparticle tracking ( NanoSight ) . The average particle size peaked at 138 nm ( ± 4 . 4 ) ( Fig 2E ) . This is a slight increase over NiVwt-LV ( Gc∆34/Fc∆22 pseudotyped ) , which showed a peak size of 130 nm ( ± 3 . 2 ) , probably due to fusion of the DARPin to the NiV-G protein . For comparison , VSV-LV , vector particles pseudotyped with the glycoprotein G of vesicular stomatitis virus , had a diameter of 116 nm ( ± 3 . 2 ) ( Fig 2E ) . Electron microscopy of concentrated NiVmutEpCAM-LV stocks revealed numerous particles exhibiting the typical morphology of HIV-1 core particles . On their surface , a homogeneous high-density layer of spike proteins was readily detectable ( Fig 2F ) . To assess the selectivity of gene transfer mediated by NiVmutEpCAM-LV , CHO-K1 cells which are negative for both , EpCAM and ephrin-B2 , CHO-ephrin-B2 and CHO-EpCAM cells were transduced either individually or in mixed cultures , where EpCAM+ and EpCAM- cells are cultivated in close contact . VSV-LV transduced all these cell lines equally efficient . NiVwt-LV selectively transduced CHO-ephrin-B2 cells in single and mixed culture , while NiVmutEpCAM-LV selectively transduced only CHO-EpCAM cells ( Fig 3A ) . On receptor-negative cells at most 0 . 1% GFP-positive cells were detectable . This was also the case for CHO-ephrin-B2 cells , demonstrating that natural receptor usage had been completely ablated by the introduced point mutations ( Fig 3A ) . The selective gene transfer mediated by NiVmutEpCAM-LV was stable over a period of 30 days of cultivating the transduced cells ( Fig 3B ) . GFP protein transfer can therefore be ruled out and integration of the gfp reporter gene into the host cell chromosomes assumed . To verify cell entry via the targeted receptor , NiVwt-LV or NiVmutEpCAM-LV were incubated with increasing amounts of the entire extracellular domain of human ephrin-B2 , ephrin-B3 , human EpCAM or murine EpCAM as a control at 4°C for 1 h prior to transduction of target cells . For NiVwt-LV , a dose-dependent decrease in the transduction of CHO-ephrin-B2 cells with increasing concentrations of recombinant ephrin-B2 was documented , whereas human or murine EpCAM did not influence transduction ( Fig 4A ) . Ephrin-B3 , which is known to bind to the same site on the NiV G protein as ephrin-B2 , also competed with gene transfer via ephrin-B2 but at much higher concentrations . NiVmutEpCAM-LV , in contrast , was only competed by human EpCAM but neither by murine EpCAM nor ephrin-B2 or -B3 ( Fig 4B ) . To examine if cell entry of the NiV-pseudotyped LVs was pH-independent , as described for Nipah virus [27] , acidification of endosomes was blocked with bafilomycin A1 . VSV-LV mediated gene transfer was strongly reduced by bafilomycin A1 which corresponds to its well established pH-dependent cell entry [28] ( Fig 4C ) . In contrast , the transduction by NiVwt-LV was not affected . Surprisingly , gene transfer mediated by NiVmutEpCAM-LV was even enhanced upon bafilomycin A1 treatment ( Fig 4C ) . These data demonstrate that NiVmutEpCAM-LV enters cells pH-independently . Next , we asked if other surface proteins can be targeted by engineered NiV glycoproteins as well . Human CD8 , a marker for cytotoxic T cells , human CD20 , a marker for B cells , and Her2/neu , a marker for breast cancer , were targeted as described previously for MV glycoprotein pseudotyped LVs [29–31] . As targeting ligands , we used DARPin 9 . 29 for Her2/neu [30] , and single chain antibodies ( scFv ) specific for CD8 [29] or CD20 [31] . Each G variant was expressed at the cell surface to the same level as that of the corresponding MV-HcΔ18mut variant , respectively ( Fig 5A; S3 Fig ) . In addition , they were incorporated into the LV particles with a tendency for higher incorporation levels of the DARPin- over the scFv-displaying G proteins ( Fig 5B; S4 Fig ) . Several batches of all four NiV-based and the corresponding MV-based receptor-targeted vectors were generated and titrated on cell lines expressing the targeted receptor . Interestingly , we observed 10-100-fold higher titers for NiVmutEpCAM-LV , NiVmutCD20-LV , and NiVmutCD8-LV compared to their MV-based counterparts ( Fig 5C and 5D ) . Notably , NiVmutHer2-LV was an exception . Here , the titer was reduced 30-fold when compared to its MV-based counterpart ( Fig 5C and 5D ) . Importantly , concentrating the vector particles did not impair transduction efficiency with recovery rates of 80 . 8 to 96 . 9% ( Fig 5E ) . When supernatant from packaging cells cultivated in ten T175 flasks was concentrated down to 0 . 6 ml , titers of above 108 t . u . /ml were obtained . To identify potential reasons for the increased titers of receptor-targeted LVs based on the NiV glycoproteins , particle numbers in all concentrated vector stocks were determined by single nanoparticle tracking analysis . All NiV-based LV stocks contained between three- to five-fold higher particle numbers than the corresponding MV-based LV stocks ( S5 Fig ) . In 108 particles , all the NiV pseudotyped LV stocks contained more transducing units , than the corresponding MV-based LVs ( Fig 5F ) . This holds true especially for the NiV-based CD8 and CD20 targeted LVs which exhibited more than 10-fold higher values ( Fig 5F ) . Thus , the higher gene transfer activity observed for the NiV-based receptor-targeted LVs must be due to a mixture of two aspects: On the one hand , more particles are released during production and on the other hand , the particles are more active . Due to the vaccination against measles virus , MV-based LVs become neutralized by human serum to at least some extent even though receptor-targeted MV-LVs exhibit some level of protection [32] . Since there is no vaccination against NiV and the outbreaks were limited to a few cases in Malaysia , Bangladesh and India [33] , there should be no neutralizing antibodies present in humans . To cover the widest range of human serum donors , intravenous immunoglobulin ( IVIG; Intratect ) , which contains serum from many thousand donors , was incubated with NiVwt-LV , NiVmutEpCAM-LV , MVEpCAM-LV and VSV-LV at increasing concentrations prior to the transduction of target cells . MVEpCAM-LV showed a dose-dependent decrease in transduction rates and was completely neutralized at 10 μg/ml of IVIG . In contrast , VSV-LV , NiVwt-LV and also NiVmutEpCAM-LV were resistant against IVIG at all concentrations used and must thus be at least 10 , 000-fold less sensitive against human immunoglobulin than the corresponding MV-based vector ( Fig 6 ) . These results suggest that receptor-targeted vectors based on NiV glycoproteins will not be neutralized when injected into humans . To prove that the established system does not only show selective transduction in cell lines but does also transfer genes into primary cells , human peripheral blood mononuclear cells ( PBMC ) were chosen as targets . Freshly isolated PBMC were activated for three days and transduced with NiVmutCD8-LV . As controls , cells were transduced with VSV-LV or NiVwt-LV . GFP expression was followed over a period of 5 , 10 and 17 days . VSV-LV transduced both cell fractions , CD8+ and CD8- ( Fig 7A , top right diagram ) . NiVwt-LV was unable to transduce any cell type present in human PBMC ( Fig 7A ) . NiVmutCD8-LV , in contrast , selectively transduced the CD8+ cells at high efficiency . The gene transfer mediated by NiVmutCD8-LV into CD8+ cells was stable for at least 17 days ( Fig 7B ) . Notably , PBMC transduced by VSV-LV showed a significant decrease in GFP+ cells over this period . This was most likely due to cells that had not integrated the GFP gene but taken up GFP as protein . Thus , NiVmutCD8-LV cannot only transduce selectively CD8+ populations in human PBMC but also ensures stable gene expression over time . To investigate the underlying mechanism resulting in the reduced titers of Her2/neu-targeted NiV-based LV , surface expression of the Gc∆34Her2mut4 protein on HEK-293T producer cells was assessed first , as this is a critical step for incorporation into budding vector particles . Surface expression levels of Gc∆34Her2mut4 and the corresponding MV-derived Hc∆18mutHer2 did not differ and were slightly enhanced compared to Gc∆34His , which is most likely due to a better recognition of the His-tag when displayed on top of the DARPin ( Fig 8A ) . Next , the ability of Gc∆34mutHer2mut4 to bind recombinant Her2/neu was investigated . For this purpose , HEK-293T cells transfected with plasmids encoding Gc∆34Her2mut4 , Hc∆18mutHer2 , or Gc∆34His were incubated with Her2/neu and subsequently stained for Her2/neu binding . As expected , Gc∆34His and mock transfected cells showed no binding of Her2/neu . In contrast , the NiV- and MV-based Her2/neu-specific constructs showed an identical binding efficiency of recombinant Her2/neu ( Fig 8B ) . To quantitatively compare the amounts of Gc∆34Her2mut4 and Hc∆18mutHer2 in LV particles , we applied different amounts of particles from several independently generated batches of vector stocks to Western blot analysis making use of the His-tag in both proteins for detection . There were reproducibly higher amounts of G than of H protein present in each batch and dilution analyzed ( Fig 8C and 8D ) . At the highest dilution , H was detectable in only one sample . On the average of all dilutions and vector batches analyzed , there was 3 . 14 ± 0 . 66 ( n = 8 ) fold more Gc∆34Her2mut4 than Hc∆18mutHer2 incorporated . Since Gc∆34Her2mut4 was able to bind Her2/neu and was incorporated into LV particles at even higher levels than Hc∆18mutHer2 , we tested next , if the vector particles had lost their gene transfer activity due to an increased endocytosis and subsequent degradation by endo-lysosomal proteases . For this purpose , endosomal acidification of SK-OV-3 cells ( positive for Her2/neu and ephrin-B3 ) was blocked by bafilomycin A1 . As shown before , transduction by VSV-LV was inhibited whereas transduction by NiVwt-LV was not influenced ( Fig 8E ) . Interestingly , gene transfer by NiVmutHer2-LV was rather enhanced by bafilomycin A1 by a factor similar to that seen for NiVmutEpCAM-LV ( Figs 8E and 4C ) . Thus , the defect in transduction of the Her2/neu specific NiV-LV was likely not caused by loss of particles to acidified endosomal compartments . DARPin 9 . 29 had been used in NiVmutHer2-LV as this targeting domain was found to be the best for the MV-glycoprotein-based system [30] . To test , if the targeting domain may make a difference , DARPin 9 . 29 was exchanged against five different Her2/neu specific DARPins ( 9 . 01 , 9 . 16 , 9 . 26 , H14R , G3 ) [34 , 35] and one trastuzumab-derived scFv ( 4D5++ ) [36] . All Gc∆34mut4 fusion proteins were expressed on the surface of HEK-293T producer cells , with a tendency for the 9 . 29 DARPin to mediate higher and the scFv 4D5++ to mediate lower expression levels ( Fig 8F ) . Particle incorporation levels for most of the new G variants were in the same range as that for the 9 . 29 displaying Gc∆34mut4 , with the only exception for the scFv 4D5++ ( Fig 8G; S8 Fig ) . Notably , DARPins H14R and G3 as well as the scFv 4D5++ mediated gene transfer activities comparable to or even higher than that of MVHer2-LV ( Fig 8H ) . Enhanced binding to Her2/neu could be excluded as being causative , since Gc∆34mut4 proteins fused to these targeting ligands were similar or rather less efficient in binding Her2/neu ( S9 Fig ) . Remarkably , the two DARPins as well as the scFv bind to domain IV of Her2/neu , the most membrane-proximal domain , whereas the four other DARPins , which resulted in low functional titers , bind to domain I [34 , 35 , 37] . To further investigate a potential preference of the NiV glycoproteins for membrane-proximal binding sites , we targeted NiV-LV particles to two further receptors with large extracellular parts , the human c-kit receptor ( CD117 ) and the glutamate receptor 4 ( GluA4 ) . CD117 was targeted by displaying its natural ligand stem cell factor ( SCF ) , and GluA4 by a recently selected DARPin , in each case via a ( G4S ) 3 linker on Gc∆34mut4 . CD117 is a tyrosine kinase receptor composed of five Ig-like domains of which the first two domains supported by domain III form the SCF binding site [38] . The crystallized extracellular part of CD117 forms a rigid structure which projects the SCF binding site away from the cell membrane by about 120 Å [38 , 39] . GluA4 is a typical channel protein with an amino-terminal domain ( ATD ) reaching up to 120 Å away from the cell membrane . As alternative receptors , we moved domains I-III of CD117 by about 70 Å and the ATD of GluA4 by about 50 Å closer to the cell membrane by fusing them to the CD4 transmembrane domain , respectively . All receptors were stably expressed in human fibrosarcoma HT1080 cells . The Gc∆34mut4 variants as well as the receptors were readily detected at the cell surface by flow cytometry with a tendency for the shortened receptors to reach lower surface expression levels ( Fig 9A and 9B; S10 Fig ) . For CD117 this was verified by using recombinant SCF instead of the CD117-specific antibody to detect surface expression ( Fig 9C; S10 Fig ) . The titers of NiVmutCD117-LV and of NiVmutGluA4-LV particles were in a similar low range as that of NiVmutHer2-LV when added to cells expressing the unmodified receptors ( Fig 9D ) . Titers increased by at least 20-fold on cells expressing the shortened receptors ( Fig 9D ) , thus supporting the idea of more efficient cell entry via membrane proximal receptors .
Here we describe successful engineering of the NiV glycoproteins for LV pseudotyping and receptor targeting , which allowed us to rapidly generate a large series of glycoprotein variants attaching to a variety of cell surface proteins and assessing cell entry . For pseudotyping , distinct truncations in G ( Gc∆33 and Gc∆34 ) and F protein ( Fc∆22 ) were found to be optimal . Our data are thus in line with those of Witting et al ( 2013 ) for G protein , and with those of Palomares et al . ( 2012 ) for F protein . For both , F and G , the enhanced titers correlated well to an enhanced incorporation into LV particles , suggesting steric hindrance as likely explanation for the need for cytoplasmic tail truncations . In the second engineering step we eliminated use of the natural NiV receptors ephrin-B2 and ephrin-B3 by introducing point mutations into the Gc∆34EpCAM protein . For identifying the most effective mutations , we relied on the G protein structure and previously identified contact residues [22 , 23 , 25 , 26] . Yet , this turned out to be challenging , since the picomolar affinity of G for ephrin-B2 is among the strongest viral envelope-receptor interactions known [40 , 41] . Accordingly , we found that mutations E533A and W504A , the previously identified key residues for receptor attachment [22 , 23] , were not sufficient to destroy ephrin-B2 binding completely , either individually or in combination ( E533A/Q530A and E501A/W504A ) . However , combining both double mutations ultimately diminished binding to a level below detection . Importantly , transduction via the targeted EpCAM receptor was unimpaired by these mutations . Off-target transduction tested in CHO cells overexpressing ephrin-B2 was barely detectable and at least 1000-fold reduced when compared to the transduction of CHO cells overexpressing EpCAM . Since ephrin-B2 is widely expressed in the organism , including microvascular endothelial cells [42] , having achieved complete abrogation of LV particle attachment to ephrin-B2 is an important step towards efficient in vivo gene delivery with receptor-targeted LVs . NiV and MV enter cells by pH-independent membrane fusion at the cell membrane . EpCAM is known to be rapidly internalized upon antibody binding [21 , 43] . It is therefore likely that also binding of EpCAM-targeted vector particles induces internalization of EpCAM together with the bound particle . Interestingly , bafilomycin A1 enhanced gene delivery by NiVmutEpCAM-LV but , as expected , substantially reduced that mediated by VSV-LV , which is known to rely on pH-dependent entry [28] . Bafilomycin A1 is a selective inhibitor of the V-ATPase preventing the influx of protons into endosomes [44 , 45] . Thus , in the presence of bafilomycin A1 endocytosed NiVmutEpCAM-LV particles are less degraded by pH-dependent endo-/lysosomal proteases and can therefore enter the cytoplasm via fusion of the LV envelope with the endosomal membrane more efficiently [46] . Notably , a similar observation has been made for Her2/neu targeted MV-LVs using chloroquine as inhibitor [47] . Although being more unspecific , chloroquine also neutralizes the low pH in endosomes . It is well conceivable that in both settings more particles can escape the endosomes by membrane fusion and then contribute to the observed enhanced gene delivery rates . An unexpected observation of our study was the behavior of the Her2/neu-targeted NiV-LV . In contrast to NiV-LVs targeted to CD8 , EpCAM , or CD20 , it was substantially reduced in mediating gene transfer lagging behind its MV-based counterpart by about 30-fold . Changes in cell surface expression , particle incorporation as well as Her2/neu binding could be excluded as being causative . Also blocking proteolytic degradation after potential endocytosis did not restore gene transfer activity to similar levels as that of the MV-based vector particles . Among a panel of six further Her2/neu binding domains , however , two DARPins and a scFv were identified that mediated substantially higher titers now being in the same range or even exceeding those of the corresponding MV-based LVs . Vector particles displaying these Her2/neu binding domains being active in mediating transduction neither contained more G protein nor were they more active in binding Her2/neu . Strikingly , their binding sites invariably localize to the membrane proximal domain IV of Her2/neu , while those of the four binding domains mediating low transduction rates localize to the membrane distal domain I ( Fig 10A ) . Her2/neu is known to exist mainly in the so called “open” conformation in which the extracellular domains are straightened up and thus oriented almost perpendicularly to the cell membrane [48] . Thus , vector particles binding to the membrane distal domain I of Her2/neu will be about 80 Å further away from the cell membrane than those displaying a domain IV-specific targeting domain . The SCF binding site on CD117 ( about 120 Å ) and the ATD of GluA4 are similarly far away from the cell membrane [38 , 39] . Indeed , titers of the NiV-LV particles targeted to these receptors were in a similar low range as that of NiVmutHer2-LV , but increased by at least 20-fold when we moved domains I-III of CD117 and the ATD of GluA4 closer to the membrane ( Fig 10A ) . In contrast to CD117 , GluA4 and Her2/neu , the natural NiV receptor ephrin-B2 is a transmembrane protein with a single , rather small extracellular domain that brings the bound virus particle similarly close to the cell membrane as domain IV of Her2/neu . This holds true also for EpCAM and CD20 , which both mediated efficient entry of the targeted NiV-LV particles ( Fig 10A ) . The extracellular part of CD8α is composed of a single immunoglobulin ( Ig ) -like domain linked to a thin stalk sequence of 47 residues . Although the 3D structure of the stalk is not available , it is assumed to be highly flexible , thus also allowing a close proximity of vector particles having attached to the Ig-like domain [49] . Taken together , we thus have experimental evidence from three different receptors that changing the distance of the attachment site relative to the plasma membrane makes a huge difference in particle entry . Thereby , it was irrelevant if we altered the distance by receptor engineering ( CD117 , GluA4 ) or by displaying targeting domains that bind to more membrane-proximal epitopes ( Her2/neu ) . We can therefore conclude , with only some uncertainty for CD8 , that gene delivery mediated by NiV-LVs at high efficiency requires binding of cell surface receptors close to the cell membrane within a maximal distance of about 50 Å . Receptor attachment in distances clearly beyond this results in a substantially reduced gene delivery efficiency , most likely due to inefficient or absent membrane fusion . How can we imagine that the distance between attachment site and cellular membrane makes such a huge difference for pH-independent membrane fusion mediated by the NiV glycoproteins ? It is important to realize that NiV-LV particles are completely covered with glycoproteins ( Fig 2F ) . Thus upon cell attachment , a rigid scaffold will be formed between viral and cellular membranes by numerous glycoprotein-receptor contacts . These trigger conformational changes in G and F , which then projects the fusion peptide on top of a long coiled-coil structure , the heptad repeat A ( HRA ) , towards the cell membrane [2] . In its fully extended , so called prehairpin intermediate state , F can cover a maximal distance of 210 Å between the viral and the cellular membrane [50] . Although the structure of the prehairpin intermediate has so far only been modeled for parainfluenza virus 5 ( PIV5 ) , we can assume a very similar distance for the NiV F protein , since structure and size of HRA and HRB ( adjacent to the transmembrane domain ) are well conserved among paramyxoviruses [51 , 52] and the recently crystallized prefusion form of NiV F exhibits an overall similar size as that of PIV5 [53] . With G protein being slightly bigger in size than F , and receptor attachment sites being located on top of the globular heads of G [25 , 26] , we estimate the natural receptor binding site being about 120 Å away from the viral membrane . The conformational change in F can then cover an additional distance of up to 90 Å . Any distance beyond that would not allow insertion of the fusion peptide into the cell membrane . The distances we determined here for the G-receptor pairs which were inefficient in mediating vector particle entry were indeed above 90 Å . The most likely explanation for our observations thus is an incompatible architecture of the fusion protein with rigid receptors that expose binding sites for NiV more than 90 Å away from the cell membrane ( Fig 10B ) . While most of the paramyxoviruses use sialic acid as receptor and can thus choose between many attachment sites exposed at various distances from the cell membrane , Henipa- and morbilliviruses using protein receptors must have adapted to receptors bringing them so close to the cell membrane that the distance between both viral and cellular membrane can be covered by their F protein . Supporting this model , a study analyzing a panel of chimeric CD46-CD4 proteins to function as MV receptors demonstrated that putting the MV binding domains of CD46 on top of the complete CD4 molecule ( four extracellular Ig domains ) strongly reduced membrane fusion [54] . While this fits nicely to our observations for the NiV glycoproteins , targeting MV-pseudotyped LVs to the membrane distal domain of Her2/neu did not affect gene delivery ( Fig 5C ) [30] . A prominent difference between MV-LVs on one hand and NiV-LVs as well as MV on the other is the level of incorporated glycoproteins . NiV-LVs are completely covered with glycoproteins ( Fig 2F ) as it is the case for MV particles [55] . MV-LV particles , in contrast , contained on average more than three-fold less H than NiV-LVs G protein . Thus , LV particles pseudotyped with MV glycoproteins bind to cells via very few or even single receptor contacts , which leaves them more flexibility to take a position within an optimal distance to the cell surface for membrane fusion . This may well help MV-LVs to better compensate when being bound to a membrane distal domain of a receptor . For NiV-LVs , in contrast , this may not be as easily possible since they form many receptor contacts resulting in a much more rigid complex between virus particle and target cell . Moreover , the henipavirus G proteins are unique among all paramyxoviruses , including MV , in forming covalently linked tetramers ( dimers-of-dimers ) [56] . This could further contribute to a more rigid receptor-attachment protein complex for NiV than for MV , which in turn results in higher sensitivity towards membrane-distal receptor attachment . In summary , the data presented in this manuscript imply that for the engineering of cell-type-specific LVs , binding domains should be used bringing the particles within a close distance to the cell membrane . By applying this to NiV-LVs , important progress in the engineering of cell-type specific LVs has been made . Titers of these vectors are substantially enhanced compared to vectors pseudotyped with engineered MV glycoproteins . The reasons for this could be allocated to an increased number of particles released from packaging cells which is most likely due to the intrinsic budding capability of the NiV glycoproteins [57] . Second , the particles are more active in delivering the packaged gene which is likely the consequence of the substantially higher glycoprotein density of NiV-LV particles compared to MV-LVs . Since NiV-LVs can be produced at titers exceeding 106 t . u . /ml , they will better qualify for scale up and GMP production—an important requirement for applying receptor-targeted LVs in clinical settings .
The plasmid pHL3-Ac1 coding for truncated and mutated MV Hc∆18mut protein and for a ( G4S ) 3 linker ( L3 ) between H and the His-tagged DARPin Ac1 was generated by inserting the PCR-amplified coding sequence of the EpCAM specific DARPin Ac1 [21] from pQE30ss_Ac1_corr into the backbone of plasmid pHL3-HRS3opt2#2 [58] via SfiI/NotI . All plasmids encoding Nipah virus G protein variants were derived from pCAGGS-NiV-G [59] . The coding sequence for the Ac1 targeting domain was fused to the C-terminus of the G protein reading frame by PCR amplification of each fragment and simultaneously introducing a common AgeI restriction site , which was used for ligation resulting in plasmid pCAGGS-NiV-GEpCAM . All other targeting domains ( DARPins or scFv ) were exchanged via AgeI/NotI , resulting in the corresponding expression plasmids encoding Gc∆34 fused to the targeting domain . Truncations of the G protein cytoplasmic tail were introduced by PCR of the G protein reading frame and insertion of the PCR fragments into pCAGGS-NiV-GEpCAM resulting in plasmids pCAGGS-NiV-Gc∆33EpCAM and pCAGGS-NiV-Gc∆34EpCAM . The His-tagged GHis and Gc∆34His proteins were generated by PCR amplification from pCAGGS-NiV-G . The fragments were cloned via PacI/NotI restriction into the plasmid backbone of pCAGGS-NiV-GEpCAM resulting in pCAGGS-NiV-GHis and pCAGGS-NiV-Gc∆34His , respectively . Mutations interfering with natural receptor recognition were introduced into the NiV-Gc∆34EpCAM protein coding sequence by site-directed mutagenesis . Each mutation was generated by amplification of two fragments carrying the designated mutation with homologous regions at the mutation site . These fragments were fused and amplified by a flanking primer pair . Resulting fragments were cloned into pCAGGS-NiV-Gc∆34EpCAM via RsrII/AgeI , generating the plasmids pCAGGS-NiV-Gc∆34EpCAMmut . For the generation of the NiV-F variants , the coding sequences for Fc∆22 [60] and Fc∆25 were amplified from pCAGGS-NiV-F [59] and cloned via PacI/SacI restriction into the plasmid backbone of pCAGGS-NiV-G resulting in the plasmids pCAGGS-NiV-Fc∆22 and pCAGGS-NiV-Fc∆25 . AU1 tagged NiV-F variants used for Western blot analysis of vector particles were generated by amplifying the NiV-F variants from pCAGGS-NiV-F and simultaneously adding the AU1 tag C-terminally . The resulting PCR fragments were cloned via PacI/SacI into the backbone of pCAGGS-NiV-G , resulting in the plasmids pCAGGS-NiV-F-AU1 , pCAGGS-NiV-Fc∆22-AU1 , and pCAGGS-NiV-Fc∆25-AU1 . For sequences of primers used for the PCR reactions see S1 Table . HEK-293T ( ATCC CRL-11268 ) , U87-MG ( ATCC HTB-14 ) and CHO-K1 ( ATCC CCL-61 ) cells were grown in Dulbecco’s modified Eagle’s medium ( DMEM ) ( Sigma-Aldrich , Munich , Germany ) supplemented with 10% fetal calf serum ( FCS ) ( Biochrom , Berlin , Germany ) and 2 mM L-glutamine ( Sigma-Aldrich , Munich , Germany ) . SK-OV-3 ( ATCC HTB-77 ) cells were grown in McCoy’s A5 medium ( Sigma-Aldrich , Munich , Germany ) supplemented with 10% FCS and 1% L-glutamine . Raji ( ATCC CCL-86 ) as well as Molt4 . 8 cells were grown in RPMI 1640 ( Biowest , Nuaillé , France ) supplemented with 10% FCS and 2 mM L-glutamine . The cell lines CHO-EpCAM [61] and CHO-ephrin-B2 were derived from CHO-K1 cells ( ATCC CCL-61 ) and cultivated in the same medium in presence of 10 μg/ml puromycin ( Thermo Fisher Scientific , Waltham , USA ) . For generation of CHO-ephrin-B2 cells , the gene encoding human ephrin-B2 was amplified from pCAGGS-EB2 [62] and cloned into a lentiviral transfer vector resulting in the bicistronic plasmid pS-ephrin-B2-IRES-puro-W . CHO-K1 cells were transduced with LV particles having packaged the ephrinB2-IRES-puro sequence and were selected using 10 μg/ml puromycin for 2 weeks . HT1080-CD117 and HT1080-GluA4 cells were derived from HT1080 cells ( ATCC CCL-121 ) . For this , the coding sequence for human CD117 was amplified via PCR from pCMV6-XL4-cKIT ( OriGene Technologies , Rockville , USA ) and that of GluA4 from pk0002-Imyc [63] ( kindly provided by Kari Keinänen ) , thereby adding an N-terminal myc tag . PCR fragments were then cloned into the backbone of pS-ephrin-B2-IRES-puro-W ( BamHI/SpeI ) resulting in the bicistronic plasmids pS-CD117-IRES-puro-W and pS-GluA4-IRES-puro-W , respectively . For the shortened receptor versions , the coding sequence for domains I-III of human CD117 , and of the amino terminal domain ( ATD ) of murine GluA4 were fused to that of the CD4 transmembrane domain , respectively , and cloned into the backbone of pS-ephrin-B2-IRES-puro-W ( BamHI/SpeI ) resulting in pS-CD117short-IRES-puro-W and pS-GluA4short-IRES-puro-W . HT1080 cells were transduced with LVs having packaged the receptor encoding constructs and selected using 10 µg/ml puromycin for 2 weeks . Primary PBMC were isolated from human buffy coats purchased from the German blood donation center ( DRK-Blutspendedienst Hessen , Frankfurt ) . PBMC were activated for 72 h in RPMI 1640 supplemented with 10% FCS , 2 mM L-glutamine , 0 . 5% streptomycin/penicillin , 25 mM HEPES ( Sigma-Aldrich , Munich , Germany ) , 100 U/ml interleukin-2 ( R&D Systems , Minneapolis , USA ) , 1 μg/ml CD3 antibody ( clone: OKT3 , eBioscience , San Diego , USA ) and 1 μg/ml CD28 antibody ( clone: CD28 . 2 , eBioscience , San Diego , USA ) . Following transduction , cells were cultivated in the same medium without OKT3 antibody and CD28 antibody . Vector particles were generated by transient transfection of HEK-293T cells using polyethylenimine ( PEI ) . Twenty-four hours before transfection , 2 . 5x107 cells were seeded into a T175 flask . On the day of transfection , the cell culture medium was replaced by 10 ml DMEM with 15% FCS and 3 mM L-glutamine . The DNA mix was prepared by mixing 35 μg of total DNA with 2 . 3 ml of DMEM without additives . For initial experiments ( Fig 1B and 1D ) 1 . 35 μg of plasmid DNA encoding NiV-G wildtype or truncation mutants was mixed with 4 . 04 μg plasmid DNA encoding the NiV-F variants , 14 . 5 μg of the packaging plasmid pCMV∆R8 . 9 [64] and 15 . 2 μg transfer vector pSEW encoding green fluorescent protein ( GFP ) as reporter [65] . Following optimization of G to F ratios , 0 . 9 μg of plasmid encoding G protein variants were mixed with 4 . 49 μg plasmid coding for F variants . The amounts of packaging plasmid and transfer vector remained unchanged . LVs pseudotyped with the VSV glycoprotein G were generated by co-transfecting cells with 6 . 13 μg pMD2 . G ( kindly provided by Didier Trono , Lausanne , Switzerland ) , 11 . 38 μg pCMV∆R8 . 9 and 17 . 5 μg pSEW . The transfection reagent mix was prepared by adding 140 μl of 18 mM PEI solution in H2O to 2 . 2 ml DMEM without additives . This solution was combined with the DNA mix , vortexed , incubated for 20 minutes at room temperature and added to the HEK-293T cells , resulting in DMEM with 10% FCS , 2 mM L-glutamine in total . 24 h later , the medium was replaced by DMEM with 10% FCS , 2 mM L-glutamine . At day two post transfection , cell supernatants containing the vector particles were passed through a 0 . 45 μm pore size filter . If needed , vector particles were purified by centrifugation at 4500 rpm for 24 h over a 20% sucrose cushion . The pellet was resuspended in phosphate-buffered saline ( PBS ) . For transduction , 8x103 of CHO-EpCAM and SK-OV-3 cells or 2x104 Molt4 . 8 and Raji cells were seeded into 96-well-plate and transduced on the next day . When needed , cells were pre-incubated for 30 min at 37°C with medium containing different concentrations of bafilomycin A1 ( Santa Cruz Biotechnology , Dallas , USA ) before LVs were added . For titration , cells were transduced with at least four serial dilutions of vector particles . After 72 h , the percentage of GFP-positive cells was determined by flow cytometry and the transducing units per milliliter ( t . u . /ml ) were calculated by selecting the dilutions showing linear correlation between dilution factor and number of GFP-positive cells . For transduction of primary PBMC , cells and vector were spinfected by centrifugation at 850xg at 32°C for 90 minutes . Percentages of GFP-positive cells were determined by flow cytometry at the indicated days post transduction . For electron microscopy , concentrated NiVmutEpCAM-LV particles were adsorbed to glow discharged formovar coated 200-mesh nickel grids for 5 min , washed three times with H2O and contrasted with 2% aqueous uranyl acetate ( Merck , Darmstadt , Germany ) for 10 s . Samples were analyzed with the EM109 transmission electron microscope ( Zeiss , Jena , Germany ) . Flow cytometry analysis was performed on the MACSQuant Analyzer 10 ( Miltenyi Biotec , Bergisch Gladbach , Germany ) . For surface expression experiments of Nipah virus G constructs , HEK-293T cells were transfected with the corresponding expression plasmid . After 48 h , adherent cells were detached with PBS-EDTA solution and subsequently washed in 800 µl FACS washing buffer ( PBS , 2% FCS , 0 . 1% NaN3 ) , and incubated with a phycoerythrin ( PE ) -conjugated mouse anti-His antibody ( clone GG11-8F3 . 5 . 1 , Miltenyi Biotec , Bergisch Gladbach , Germany , dilution 1:100 ) in FACS washing buffer . Human EpCAM was detected by an Allophycocyanin ( APC ) labeled mouse anti-EpCAM antibody ( clone HEA-125 , Miltenyi Biotec , Bergisch Gladbach , Germany , dilution 1:100 ) . CD117 and CD117short expression was detected by staining with PE-coupled CD117 antibody ( clone: 104D2; 1:100; BioLegend , San Diego , USA ) . Expression of myc-tagged GluA4 and GluA4short was detected by staining with PE-coupled anti-myc antibody ( clone 9B11; 1:100; Cell Signaling Technology , Danvers , USA ) . Primary PBMC were transferred into FACS washing buffer , washed twice and CD8 expression was detected by a human APC-conjugated anti-CD8 antibody ( clone RPA-T8 , 1:100 , BD Biosciences , San Jose , USA ) . After two additional washing steps , cells were resuspended in 100 µl PBS containing 1% formaldehyde . Data were analyzed using FCS Express version 4 . 0 ( De Novo Software , Glendale , USA ) . 1 . 4x106 HEK-293T cells were seeded into one well of a 6-well plate and transfected on the next day with 1 . 92 µg plasmid DNA coding for the different NiV-G constructs . 48 h later , cells were detached and 1x105 cells were washed with FACS washing buffer , incubated with 1 µg of the Fc-tagged extracellular domain of human ephrin-B2 , ephrin-B3 or Her2/neu ( R&D Systems , Minneapolis , USA ) for 1 h at 4°C , washed again and subsequently stained with a FITC-tagged anti-human Fc antibody ( 1:100 , SouthernBiotech , Birmingham , USA ) . Samples were analyzed by flow cytometry . NiVmutEpCAM-LV , MVEpCAM-LV , VSV-LV and NiVwt-LV were pre-incubated for 1 h at 4°C with different amounts of soluble extracellular domains of human or murine EpCAM ( Sino Biological , Beijing , China ) or human ephrin-B2 and B3 ( R&D Systems . , Minneapolis , USA ) , respectively . Then , CHO-EpCAM or CHO-ephrin-B2 cells were transduced with pre-incubated LVs before GFP expression was analyzed after 72 h by flow cytometry . CHO-EpCAM and CHO-ephrin-B2 cells were transduced at a multiplicity of infection ( MOI ) of 0 . 4 with NiVmutEpCAM-LV , MVEpCAM-LV , VSV-LV and NiVwt-LV that have been pre-incubated with serial dilutions of intravenous immunoglobulins ( IVIG , Intratect , Biotest , Dreieich , Germany ) for 2 h at 37°C . After 72 h , the percentage of GFP-positive cells was determined by flow cytometry . Particle size and concentration of LVs was determined using the NanoSight NS500 instrument ( Malvern Instruments , Worcestershire , UK ) . Concentrated vector stocks were diluted in degassed PBS to contain between 1x107 and 1x109 particles/ml and measured five times for 90 s at 25°C . NTA2 . 3 software ( Malvern Instruments , Worcestershire , UK ) was used for particle identification , size analysis and determination of particle concentration . Concentrated vector particles were denatured by incubation with 2x urea sample buffer ( 5% sodium dodecyl sulfate , 8 mM urea , 200 mM Tris-HCl , 0 . 1 mM EDTA , 0 . 03% bromphenol blue , 2 . 5% dithiothreitol , pH 8 . 0 ) for 10 minutes at 95°C , separated by gel electrophoresis on 10% sodium dodecyl sulfate-polyacrylamid electrophoresis gels , and blotted onto nitrocellulose membranes ( GE Healthcare , Freiburg , Germany ) . Blots were incubated with mouse anti-His ( clone 27E8 , 1:1 , 000; Cell Signaling Technology , Danvers , USA ) for detection of His-tagged NiV and MV glycoproteins , mouse anti-p24 ( clone 38/8 . 7 . 47 , 1:1 , 000; Gentaur , Aachen , Germany ) for detection of the LV core protein p24 or goat anti-AU1 antibody for detection of AU1-tagged NiV F protein ( 1:1 , 000; Thermo Fisher Scientific , Waltham , USA ) . Subsequently , horseradish peroxidase conjugated secondary antibodies ( 1:2 , 000; DakoCytomation , Hamburg , Germany ) were used and signals were detected using the ECL Plus Western Blotting Detection System ( Thermo Fisher Scientific , Waltham , USA ) . Released photon units were then quantified using the IVIS Spectrum ( PerkinElmer , Waltham , USA ) . To determine the photon intensities of bands corresponding to the glycoproteins and to p24 , areas corresponding to the respective bands were manually defined using the Living Image 4 . 3 . 1 software ( PerkinElmer , Waltham , USA ) . Background activities were identified by the negative control samples and subtracted from the glycoprotein and p24 signals . Buffy-oats obtained from anonymous blood donors were purchased from the German blood donation center . | Pseudotyping of lentiviral vectors ( LVs ) with glycoproteins from other enveloped viruses has not only often been revealing in mechanistic studies of particle assembly and entry , but is also of practical importance for gene delivery . LVs pseudotyped with engineered glycoproteins allowing free choice of receptor usage are expected to overcome current limitations in cell-type selectivity of gene transfer . Here we describe for the first time receptor-targeted Nipah virus glycoproteins as important step towards this goal . LV particles carrying the engineered Nipah virus glycoproteins were substantially more efficient in gene delivery than their state-of-the-art measles virus-based counterparts , now making the production of receptor-targeted LVs for clinical applications possible . Moreover , the data define for the first time the molecular requirements for membrane fusion with respect to the position of the receptor binding site relative to the cell membrane , a finding with implications for the molecular evolution of paramyxoviruses using proteinaceous receptors for cell entry . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"flow",
"cytometry",
"gene",
"transfer",
"plasmid",
"construction",
"mutation",
"membrane",
"fusion",
"dna",
"construction",
"molecular",
"biology",
"techniques",
"gene",
"delivery",
"glycoproteins",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"molecular",
"biology",
"cell",
"membranes",
"spectrophotometry",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"gene",
"expression",
"and",
"vector",
"techniques",
"biochemistry",
"cytophotometry",
"point",
"mutation",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"glycobiology",
"spectrum",
"analysis",
"techniques"
]
| 2016 | Receptor-Targeted Nipah Virus Glycoproteins Improve Cell-Type Selective Gene Delivery and Reveal a Preference for Membrane-Proximal Cell Attachment |
Paramyxoviruses cause a wide variety of human and animal diseases . They infect host cells using the coordinated action of two surface glycoproteins , the receptor binding protein ( HN , H , or G ) and the fusion protein ( F ) . HN binds sialic acid on host cells ( hemagglutinin activity ) and hydrolyzes these receptors during viral egress ( neuraminidase activity , NA ) . Additionally , receptor binding is thought to induce a conformational change in HN that subsequently triggers major refolding in homotypic F , resulting in fusion of virus and target cell membranes . HN is an oligomeric type II transmembrane protein with a short cytoplasmic domain and a large ectodomain comprising a long helical stalk and large globular head domain containing the enzymatic functions ( NA domain ) . Extensive biochemical characterization has revealed that HN-stalk residues determine F specificity and activation . However , the F/HN interaction and the mechanisms whereby receptor binding regulates F activation are poorly defined . Recently , a structure of Newcastle disease virus ( NDV ) HN ectodomain revealed the heads ( NA domains ) in a “4-heads-down” conformation whereby two of the heads form a symmetrical interaction with two sides of the stalk . The interface includes stalk residues implicated in triggering F , and the heads sterically shield these residues from interaction with F ( at least on two sides ) . Here we report the x-ray crystal structure of parainfluenza virus 5 ( PIV5 ) HN ectodomain in a “2-heads-up/2-heads-down” conformation where two heads ( covalent dimers ) are in the “down position , ” forming a similar interface as observed in the NDV HN ectodomain structure , and two heads are in an “up position . ” The structure supports a model in which the heads of HN transition from down to up upon receptor binding thereby releasing steric constraints and facilitating the interaction between critical HN-stalk residues and F .
The Paramyxoviridae are membrane-enveloped negative-sense single-stranded RNA viruses that infect animals and humans often resulting in significant disease and mortality . Most paramyxoviruses enter cells at neutral pH by fusing their envelope with the plasma membrane of a target cell thereby releasing a ribonucleoprotein complex into the cytoplasm . Paramyxovirus fusion is typically mediated by two glycoproteins on the surface of virions: a trimeric fusion protein , F , with type I viral fusion protein characteristics , and a receptor binding protein variously named HN , H , or G depending on the virus and protein functionality [1] . Viruses with hemagglutinin-neuraminidase ( HN ) attachment proteins use sialic acid as a receptor and include parainfluenza virus 5 ( PIV5 ) , Newcastle disease virus ( NDV ) , mumps virus , human parainfluenza viruses ( hPIV1-4 ) , and Sendai virus . HN proteins have at least three functions: ( 1 ) they bind sialic acid receptors on glycoproteins and gangliosides at the cell surface ( hemagglutinin activity ) . This function is thought to play an important role in timing and initiating virus-cell fusion . ( 2 ) HN proteins are neuraminidases , which catalyze the hydrolysis of glycosidic linkages on terminal sialic acid residues thus destroying the receptor . Neuraminidase activity likely plays a crucial role removing sialic acid from viral and cellular conjugates during assembly and budding [2] . ( 3 ) A function common to paramyxovirus attachment proteins is to lower the activation barrier of the F protein , presumably through direct interaction , thereby triggering a major refolding event in F from a metastable prefusion form [3] , [4] , [5] to a highly stable post-fusion form [6] , [7] , [8] . The merging of viral and target cell membranes is coupled with this structural rearrangement [9] . PIV5 HN is comprised of a short cytoplasmic tail ( residues 1–17 ) at its N-terminus , a transmembrane domain ( residues 18–36 ) , and a large ectodomain ( residues 37–565 ) . The ectodomain is composed of a helical stalk and a large globular head containing the hemagglutinin/neuraminidase ( NA ) active site ( s ) . The X-ray crystal structures of the globular head domain of NDV , hPIV3 , PIV5 , measles virus ( MeV ) , Hendra virus ( HeV ) , and Nipah virus ( NiV ) reveal a six-bladed beta-propeller fold typical among neuraminidases [10] , [11] , [12] , [13] , [14] , [15] , [16] , [17] . For PIV5 , a sialyllactose receptor-bound globular head domain structure revealed that sialyllactose binds in the center of the beta propeller [17] . These structures , as well as biochemical data [18] , reveal that PIV5 HN forms covalent dimers ( via Cys 111 ) that further assemble into a non-covalent dimer of dimers via the stalk domain [19] with possible contribution from the cytoplasmic tail and transmembrane domains [20] . Recent crystal structures of both the NDV-HN ectodomain ( including stalk residues 79–115 and head domains ) and the isolated stalk domain of PIV5 HN ( residues 56–108 ) reveal a four-helix bundle ( 4HB ) stalk with robust hydrophobic core packing [16] [21] . Additional studies have shown that attachment proteins of other paramyxoviruses have stalks consistent with a 4HB structure [22] , [23] , [24] , [25] . A substantial body of evidence including point mutations , additions of glycan moieties , chimeras , insertions , and truncations has established that residues in the stalk affect fusion promotion [21] , [24] , [26] , [27] , [28] , [29] , [30] , [31] , [32] , [33] , [34] , [35] , [36] . For NDV HN and several paramyxoviruses that utilize protein receptors , stalk mutations that block fusion have also been shown to disrupt coimmunoprecipitation with F [34] , [27] , [37] , [36] . The crystal structure of the NDV HN ectodomain [16] revealed an interface between the HN heads and the upper portion of one side of the stalk 4HB , in a conformation we refer to as “4-heads-down” ( Fig . 1A ) . Interestingly , the NDV heads largely obscure the region of the stalk implicated by mutagenesis as forming direct interactions with F on two sides of the 4HB . The PIV5 HN stalk structure [21] revealed a similar 4HB and potential site for F interactions , but since that structure did not include the HN head domains , evidence for PIV5 HN head-stalk interactions has been lacking . We recently showed that the head domains of PIV5 HN are completely dispensable for fusion activation . A construct lacking the head domains ( “headless HN” ) activates fusion at levels comparable to wild-type ( wt ) HN [38] . Together , these results suggested a model for fusion activation in which the ‘4 heads down’ HN represents a prefusion conformation that restricts or blocks F interactions with the stalk and that deletion of the head domains allows F interactions that promote membrane fusion [38] . The NDV ectodomain structure also revealed a novel organization of the four head domains relative to one another as compared to those previously observed in crystal structures of other paramyxovirus attachment protein head domains [16] . Regardless of the absence or presence of the stalk domain in the expression constructs used to solve NDV- , hPIV3- , and PIV5-HN head domain structures , the head domains for each of these viruses formed a similar dimeric arrangement , except for a single low pH form of the NDV HN protein [12] . A dimer-of-dimers ( DOD ) tetramer was observed in a previous PIV5 HN ectodomain crystal structure , and a similar arrangement was observed in NDV neuraminidase domain crystals ( Figs . 1B , S1 ) [12] , [14] , [17] . Unlike , the NDV HN 4-heads-down conformation , the original PIV5 tetramer arrangement places N-termini relatively close together on one side of the tetramer such that a stalk could extend down from the center toward the viral membrane . This tetrameric arrangement could represent a post-receptor binding state for HN , which we refer to as the “4-heads-up” conformation . A composite model of this hypothetical conformation can be generated using the separately determined PIV5 HN head domain and 4HB stalk structures as shown in Fig . 1B . NDV HN heads can adopt different arrangements in different crystal forms , and electron microscopy data indicates that the heads of NDV and PIV5 HN can adopt various conformations relative to each other and the stalk [16] , [38] , [19] . It is therefore likely that HN head domains can adopt different conformations on the virion surface as well , and these structural rearrangements may be linked to fusion activation . For MeV H it has been proposed that conversion between two head arrangements , form I and II , is coupled to fusion activation [39] . Based on available structural and functional data , we proposed a potentially general model for paramyxovirus fusion activation in which the head domains of the receptor-binding protein move from the “down position” to the “up position” upon receptor binding to expose residues in the stalk that are the trigger of F protein activation [38] . However , outside of the NDV HN ectodomain structure , no structural information on potential head-stalk interactions has been available for other members of the paramyxovirus family that could support the generality of this model . Here , we report the structure of the PIV5 HN ectodomain ( residues 61–565 ) , which adopts a hybrid state compared to the previously observed 4-heads-down and 4-heads-up conformations . The PIV5 HN heads in the down position form an analogous interaction with the 4HB stalk as observed in the NDV HN ectodomain structure [16] , indicating that the formation of this interaction is common to HN proteins of different viruses . The two other PIV5 heads in this structure adopt a “heads up” conformation with one subunit exhibiting a fully helical extension to the stalk 4HB , consistent with a composite structural model ( Fig . 1B ) of this conformation . A novel DOD interface is observed in the hybrid state and the ability of the full-length HN to form this DOD arrangement is supported by the ability of engineered Cys mutations to form disulfide bonds . As a hybrid between the 4-heads-up and 4-heads-down conformations , the PIV5 structure is consistent with a dynamic head-stalk interaction , in which neuraminidase domain dimers form mobile structural units flexibly linked to the 4HB stalk . Overall , these results demonstrate that two different paramyxovirus HN proteins , from NDV and PIV5 , can adopt a conformational state in which the receptor-binding head domains interact with and obscure stalk residues implicated in F protein activation , supporting the hypothesis that this represents a general regulatory feature of the HN-dependent paramyxovirus entry mechanism , potentially applicable to the broader paramyxovirus family as well .
The expression construct used previously to determine the X-ray crystal structures of PIV5 HN heads contained the entire ectodomain including the full-length stalk ( residues 37–565 ) . However , it has been noted that protease cleavage occurs between residues 55 and 56 of the PIV5 HN ectodomain when expressed in baculovirus infected insect cell culture [19] . Therefore , to boost protein purity and yield in this expression system , a construct comprising residues 56–565 of the PIV5 HN ectodomain was used in this study . HN was purified , concentrated , and mixed with a slight molar excess of sialyllactose immediately prior to setting up crystallization trials . Upon optimizing conditions , HN crystallized in the I4 space group and diffracted X-rays to ∼2 . 5 angstroms . The structure was solved by molecular replacement using PIV5 HN structures of the head domain and isolated stalk 4HB . Crystallographic data and refinement statistics are summarized in Table 1 . The positions of the head domains relative to the stalk in this structure represent a novel configuration for a paramyxovirus attachment protein , that is a hybrid of the 4-heads-down [16] and 4-heads-up [17] conformations ( Fig . 1C–E ) . Two disulfide-linked heads are in the up position , while two are in the down position ( 2-heads-up/2-heads-down ) , and the structure exhibits a novel non-covalent DOD interface between one head in the up position and one head in the down position ( hereafter , the DOD interfaces observed in the 4-heads-up and 2-heads-up/2-heads-down conformations will be referred to as DOD1 and DOD2 , respectively ) ( Figs . 1B–C , S1 ) . Additionally , an interface is observed between residues in two helices on one side of the stalk and one head in the down position similar to that in the NDV ectodomain structure ( Fig . 1C ) . Although sialylactose was included in the crystallization experiments , we do not have any direct structural evidence that its inclusion induced the conformational arrangement observed . Contiguous density linking a stalk helix and head domain is also observed ( Fig . 1F ) . The stalk of chain B extends beyond the 4HB core , which ends at residue L101 , forming an extended alpha helix through N110 . This extended helix also corresponds to the positioning of the head of this chain higher than its covalent partner and is consistent with the 4-heads-up conformation , where one head of each covalent dimer extends higher than its partner leading to a ∼30° offset between covalent dimers across the DOD1 interface . Of note , although D105 and E108 of chain B are within an extended helix , they project toward the central 4HB axis and may disrupt the coiled coil from propagating beyond L101 ( Fig . 1F ) . Accordingly , residues 102–110 of chain A/E form a non-structured linker ( residues 103–106 are disordered ) . Interestingly , chains A and B cross at C111 where they form a disulfide bond and short antiparallel β-sheet . Residues 113–117 form a partially helical strand connecting with the head domain at I118 . Therefore , residues 56–101 form the 4HB stalk that is followed by a linker region ( residues 102–117 ) , part of which can extend from the 4HB stalk as an isolated helix ( residues 103–110 ) , that eventually connects to the globular head domain at I118 ( Fig . 2A ) . Each of the four globular head domains overlays with good agreement with the search model ( PDB ID: 1Z4X ) ( ≤0 . 210 RMSD over ∼404 atoms , Fig . 2B ) . The only notable exception is the position of the active site loop containing H188 and the conserved D187 in the two heads that form the novel DOD interface ( discussed below ) . Electron density was also observed for the stalk of HN . The final model aligns well with the previously solved isolated PIV5 HN stalk structure ( residues 56–105 , 0 . 627 RMSD over 143 atoms ) [21] including a relatively non-supercoiled region in the upper portion of the stalk comprised of an 11-mer repeat that transitions to a supercoiled region comprised of a heptad repeat in the lower half of the observed stalk ( Fig . 2C ) . A novel dimer-of-dimers interface lacking pseudo-tetrameric symmetry is observed between one head of one disulfide-linked dimer in the up position ( chain A ) and another ( chain D ) in the down position ( DOD2 , Fig . 1C , 3A ) . This ∼2-fold symmetric interface buries ∼1 , 139 Å2 ( total ) of solvent-accessible surface area and includes 16–17 residues from each of the two heads . The interface is generally hydrophobic in nature with 31 hydrophobic interactions and seven electrostatic interactions including six hydrogen bonds ( Fig . 3B ) . For comparison , DOD2 is larger than DOD1 observed in the 4-heads-up conformation , which has a total buried surface area of ∼846 Å2 ( using similar methodology ) . However DOD2 is much smaller than the dimer interface ( ∼3919 Å2 , chains A/B ) , which is slightly expanded here compared to that previously observed due to the more complete model . We probed the functional significance of this interface by introducing cysteine mutations into a mammalian expression construct of HN containing a C111A mutation that eliminates wt covalent dimer formation . Reducing and non-reducing SDS-PAGE analysis following immunoprecipitation of radiolabeled protein revealed that the R273C point mutation and mutant pairs D282C/S253C and E258C/N276C are capable of extensive disulfide bond formation manifested as dimer bands on non-reducing SDS-PAGE gels ( Fig . 3C–D ) . At the DOD2 interface , R273 residues have a Cα-Cα distance of 7 . 14 Å , D282/S253 residues have a Cα-Cα distance of 6 . 63 Å and E258/N276 residues have a Cα-Cα distance of 7 . 44 Å , consistent with the observed disulfide bond formation . Due to the lack of 4-fold symmetry in the head domains , the two cysteine mutations not paired in a disulfide bond at DOD2 can potentially form disulfide bonds with neighboring HN tetramers leading to higher order covalent structures . Indeed , higher molecular weight species are observed at the top of the non-reducing gel . However , this band is also observed in the mock-transfected lane making it unlikely that these species are HN ( Fig . 3D ) . Although the R273C single mutant is capable of disulfide bond formation , no dimer is observed when R273C is combined with T372C despite overexpression of this double mutant compared to wt . This result is likely due to a preference for intrachain vs interchain disulfide bond formation , as these residues are also adjacent in the monomer , with an intrachain Cα-Cα distance of 7 . 2 Å . Interestingly , HN mutations harboring the successfully engineered disulfide bonds are each expressed on the surface of cells at levels similar to wt , and they are fusogenic at 75–85% wt levels in a cell-cell fusion assay ( Fig . 3E–F ) . Finally , we note that while DOD2 is formed within individual tetramers in this structure , a crystal packing interface equivalent to DOD1 is formed between tetramers in the crystal lattice . In the original crystal structure of PIV5 HN [17] , we noted that the DOD1 packing interactions between dimers must occur both within and between tetramers , forming an extended ribbon throughout the crystal lattice . In the current crystal form , the DOD1 packing interactions between tetramers further points to the favorable nature of this interface . The possible significance of this observation is discussed further in supporting information ( Text S1 and Fig . S2 ) . The recent crystal structure of the NDV HN ectodomain revealed an interface between head domains and the stalk ( Fig . 1A ) , which also overlaps a stalk surface implicated in direct interactions with F . Interestingly , mutations in a corresponding region of the PIV5 HN stalk also affect fusion and neuraminidase activities ( Fig . 4A ) , and it has been predicted that PIV5 HN would form a similar head-stalk interaction [21] . A stalk/head interface is indeed observed in the current structure , which resembles that of NDV HN in that covalently-linked heads are in the down position allowing one head to form a similarly sized interface with the stalk ( buried solvent-accessible surface area = 1 , 294 and 1 , 185 Å2 for PIV5 and NDV using similar methods , respectively ) ( Fig . 4B–C ) . Overall , there are 15 residues from two stalk helices and 20 residues from the head that form the interface . Like the stalk/head interface in NDV HN , the PIV5 HN interface is primarily hydrophobic with 28 hydrophobic interactions and 5 electrostatic interactions including two hydrogen bonds and two salt bridges . While 18 of the residues that comprise this interface in PIV5 and NDV HN align based on primary sequence alignment , none of the interactions between these residues are identical ( Fig . 4C ) . Additionally , the angle of the heads relative to the stalk differs between the two molecules by ∼30° ( Fig . 4D ) . To investigate if mutations at or near the stalk/head interface would affect fusion , we made single point mutations of residues lying within both the stalk and head domains . Flow cytometry revealed that the point mutants were expressed at the cell surface at levels within ∼2-fold of wt ( Fig . 5A ) , except for the D398L mutation . Within the stalk region , only V81T and L85Q were significantly impaired for cell-cell fusion ( Fig . 5B ) , while mutations Y77A , T89A and T96A were fusion competent . Interestingly , Y77A is fusion active whereas a mutation that introduces a glycan at this position ( N77 mut ) is not [21] . Of the eight mutants within the head domain , only one ( D398L ) showed reduced fusion activity , but this also corresponded to a significant reduction in D398L expression levels . The mutational data further support the significance of this stalk region in F activation and fusion , given the functional effects of the V81T and L85Q mutations . As the headless PIV5 HN stalk is active in fusion , mutations of the head domain that disrupt the head-stalk interaction may not be defective in fusion activation , but may mimic the headless HN stalk activity . To investigate whether the V81T and L85Q mutations affect F activation directly , these were introduced into the headless PIV5 HN stalk construct and examined . Even though surface expression for the mutant stalks was reduced only ∼50% , fusion promotion was completely inhibited ( Fig . 5C ) . Thus , these data suggest that the V81T and L85 mutations cause direct interference with the HN stalk-F interface and do not reduce fusion by affecting HN head-stalk mechanics . The previous PIV5 HN crystal structures of the head domain were solved using various conditions including native , receptor-bound ( sialic acid and sialyllactose ) , and inhibitor-bound ( DANA ) at either pH 7 . 0 or 8 . 0 [17] . There was good agreement between the structures , and no conformational change was observed upon sialyllactose or DANA binding ( sialic acid could be converted to DANA by HN as observed with other neuraminidases ) . However , these ligand-bound structures were obtained by soaking soluble ligands into native crystals , and preformed crystal contacts may prevent movement that might otherwise occur . As mentioned above , each of the HN heads in the current structure superpose with good agreement with the previously solved structures ( Fig . 2B ) . However , a notable exception is the position of the loop containing H188 in chains A and D that interact to form the DOD2 interface . In these chains , H188 points directly into the active site whereas in the other heads of this structure , and previously solved structures , H188 is pointing up and away from the active site ( Fig . 6A–B ) . This loop has been observed to be flexible as residues 186/7-190 were disordered in two of the previous PIV5 HN head structures . Furthermore , this loop has been observed in unique conformations in a low-pH NDV Kansas HN and NDV Ulster HN crystal structures [12] , [40] . Interestingly , the movement of H188 in the current structure is correlated with electron density observed in the active site of each chain . In the active site of the chains not involved in the DOD2 interface ( chains B and C , Fig . 6A ) , there was no interpretable electron density except for spherical density that was modeled as a sulfate ion coordinated by R405 , R495 , and Y523 . However , in the heads of the DOD2 interface ( chains A and D ) additional electron density was observed near the alternately oriented H188 . Part of this density appears consistent with the lactose moiety of α ( 2 , 3′ ) -sialyllactose potentially generated during crystallization , given the enzymatically permissive pH ( pH = 6 . 6 ) of the crystallization conditions ( Fig . 6B–D ) . However , the crystallization buffer includes ∼0 . 5 M ammonium sulfate , and high salt concentrations are known to inhibit NA activity [2] . Fitting lactose to the density clearly does not satisfy the density that protrudes deepest into the active site ( Fig . 6B–D ) ; however , significant negative difference density is associated with modeling the ‘sialic acid’ portion of α ( 2 , 3′ ) -sialyllactose into this density ( not shown ) . Due to the ambiguity associated with ligand identification and conformation , the final model does not include atoms fit to this active-site density .
Here we observe that the PIV5 HN ectodomain can form a head-stalk interaction similar to the arrangement in the structure of the NDV HN [16] , providing further evidence that this conformational state could regulate F activation across different members of the paramyxovirus family . We further observed that the PIV5 HN protein can adopt a hybrid conformation consisting of 2-heads-down and 2-heads-up , with pairs of dimeric heads moving as unified structural units . This hybrid conformational state demonstrates that the PIV5 HN tetramer has inherent flexibility and potential for asymmetry in the head region dimer-of-dimers , consistent with the possibility that sialic acid receptor binding could reorient head domain dimers , exposing the stalk region for F engagement and activation . This hybrid structure also provides a concrete model for the fully helical head-stalk connection in the hypothesized 4-heads-up conformational state . The PIV5 HN ectodomain structure identifies a novel dimer-of-dimer interface , DOD2 , which is readily formed in the intact HN , as revealed by cysteine mutagenesis and disulfide bond formation . Engineered disulfide bonds bridging the PIV5 HN DOD2 interface formed readily , consistent with the ability of the NA domains to access the observed 2-heads-up/2-heads-down arrangement in intact HN . Interestingly , covalent linkage at the DOD2 interface does not impair HN's ability to activate fusion . It would not be possible for all of the heads to simultaneously be in the down or up positions when the heads are crosslinked via this interface , yet all of the mutants that form disulfide bridges are fusion competent . These results suggest that either exposure of one stalk site could result in F activation , or alternatively that exposure of both stalk sites , but not full formation of a 4-heads-up conformation , is sufficient for F activation . The mutant proteins may be active in the observed 2-heads-up/2-heads-down hybrid state or may be able to adopt a state where none of the heads engage the stalk after receptor binding , thereby activating F . Movement of one pair of heads to the up position may be sufficient to trigger fusion , as recent data suggests for measles virus H activation [41] . The cysteine mutants provide further evidence for the dynamic and flexible nature of HN head arrangements . An unexpected observation from the current structure is a conformational change within the active site of the two HN heads comprising the DOD2 interface . Early crystal structures of NDV and hPIV3 revealed a pliable active site within HN monomers that could switch between sialic acid binding and catalytic activity , however , no conformational changes were observed in the hPIV3 structure beyond the active site upon ligand binding [12] , [14] . In addition , no structural changes were observed in PIV5 HN comparing the non-liganded , inhibited , and receptor bound structures [17] . However , in the current structure , the position of the loop containing H188 is altered in the chains that form the DOD2 interface , with the H188 side chain pointing into the active site , as compared to the other chains and previously determined HN crystal structures . Electron density is observed near H188 in this altered conformation , which appears consistent with the binding of a partially disordered carbohydrate moiety at an alternate active site location also not previously observed . The unique conformation of H188 suggests that it might play a role in receptor hydrolysis or in facilitating alternative receptor binding modes . The H188 loop is directly adjacent to the loop containing residues 221–223 at the DOD2 interface . Small movements of this loop within chains A and D compared to B and C may indicate that subtle movements caused by formation of DOD2 could impact active site residues ( or vice versa ) . Local chemical environments within the crystals , NA domain conformations and/or dynamics within the lattice and other parameters may play a role in the visualization of this extra density within only two of the active sites . Contiguous electron density linking the stalk and head of one HN subunit in the current PIV5 HN structure supports the possibility that HN can adopt the modeled 4-heads-up conformation , where the 4HB stalk is located in the center and directly beneath the four heads ( Fig . 1B , S1 ) . The current structure establishes the height of the heads relative to the stalk in the 4-heads-up model , assuming that the negative charge from D105 and E108 side chains can be accommodated in two opposing helical extensions beyond the 4HB . If the membrane proximal portions of the HN and prefusion F protein stalks , for which structural information is not available , are assumed to be fully helical , modeling of the F/HN interaction with HN in the 4-heads-up conformation suggests that steric restrictions could still affect F/HN interactions . For example , F access might be limited to fully extended heads ( ‘defined’ by the helix extension of residues 103–110 , Fig . 1D ) and to the more open gap between dimers ( Fig . 7A–C ) . The ∼30° tilt between dimers in the 4 heads up model lowers the height of two of the heads , potentially blocking F interactions from these angles ( Fig . 1D ) , unless the N-terminal region of the stalk is flexible and can extend further from the membrane surface . In the absence of such stalk extension , steric restrictions could limit the stoichiometry of F trimer to HN tetramers to 2∶1 or 1∶1 interactions . Steric access restrictions would explain why HN would not activate F in the 4-heads-down conformation , where two sides of the stalk are at least partially accessible . This HN-F model also spatially aligns corresponding residues identified in MeV H implicated in direct F contacts ( Fig . 7A ) [37] . Residues in the PIV5- and NDV-HN stalk implicated in direct interaction with F also map to approximately the right height for the contact point assumed in this model [37] . Curiously , mutations in the PIV5 HN stalk that are exposed and disrupt only fusion map to a lower regions of the 4HB . This may be because the PIV5 HN stalk mutations introduced carbohydrate chains [38] and the mutations may block fusion activation by steric hindrance , disrupting F-interactions from a distance . Single point mutations V81T and L85Q narrow down the putative PIV5-HN interacting region ( Fig . 5 ) and this region for PIV5-HN aligns well with the analogous region proposed for NDV-HN and measles virus-H [37][16] . Electrostatic surface maps also provide additional support for the general features of this HN-F model . A band of negative charge in the HN stalk that corresponds to the presumed F-interaction region aligns with charged regions of F ( Fig . 7D ) . Finally , the model suggests why the glycoprotein spikes of paramyxoviruses appear to be approximately the same height ( “parallel-head” model ) [42] in electron micrographs [1] versus the “staggered-head” model [27] , [37] that fits better with available biochemical and structural data ( i . e . , F is as tall as HN when HN is in the 4-heads-down conformation , but HN is taller in the 4-heads-up conformation ) . Much biochemical and mutagenesis evidence spanning a variety of paramyxoviruses has established the HN stalk as the key site for triggering F , but with potentially conflicting mechanistic data on the structures and role of the receptor-binding head domains . Experiments involving disulfide crosslinking of MeV H head dimers have indicated that movement between heads across the dimer interface may be important for fusion [43] . However , a similar analysis of head dimers in NDV and hPIV3 indicated that movement at the dimer interface is not required [44] . It has been proposed that subtle changes at the head dimer interface may trigger larger changes , perhaps propagating to a dimer-of-dimers interface [45] . Different crystal forms of MeV H suggested that rearrangement of dimers in the tetramer could possibly trigger F by also affecting the 4HB stalk [39] . Disulfide bond stabilization of the central region of the morbillivirus H protein stalk is inhibitory , and fusion is recovered upon reduction of the disulfides [23] , [41] , [26] . These authors suggest an unwinding of the supercoiled region of the stalk may activate F . This model shares several features in common with a membrane-proximal stalk-extension model that we have discussed previously [16] . Although morbillivirus and henipavirus entry mechanisms may differ from other paramyxoviruses , potentially in the strength and timing of H/G and F interactions , a role for the attachment protein stalk regions in F activation is a common underlying feature . We suggest that inhibitory head-stalk domain interactions across the broader paramyxovirus family could occur , providing a unifying model for regulating F activation . For morbillivirus and henipavirus family members , receptor-dependent alterations in head-stalk interactions could allow F full engagement of the stalk site necessary for fusion activation , despite the potential formation of pre-fusion H/G and F complexes . Additional experiments are necessary to establish whether morbillivirus and henipavirus entry mechanisms share the key elements of this stalk-exposure model . In summary , while many specific details of the F/HN interaction remain to be elucidated , the present structure provides further evidence that paramyxovirus HN head domains interact with the 4HB stalk and that movement from a ‘heads-down’ to a ‘heads-up’ conformation allows F access to critical fusion-promoting residues in the stalk . Notably , this ‘stalk-exposure’ model fits with the “provocateur” mechanism of fusion protein activation [46] , rather than a “clamp model” ( reviewed in [47] ) consistent with available data from paramyxoviruses that utilize HN attachment proteins . It remains possible that other paramyxoviruses promote fusion differently , although the common role of the attachment glycoprotein stalk domain in F activation indicates that additional mechanistic similarities will be conserved across the virus family .
Vero cells and 293T cells were maintained in Dulbecco's modified Eagle medium ( DMEM ) supplemented with 10% fetal bovine serum ( FBS ) . BHK-21F cells were grown in DMEM containing 10% FBS and 10% tryptose phosphate broth . BSR-T7/5 cells were grown in DMEM containing 10% FBS , with 500 µg/ml G418 added every third passage . Hi5 insect cells were maintained in Express 5 serum free medium ( Gibco ) supplemented with 10% GlutaMax ( Gibco ) and Sf9 insect cell lines ( for generating baculovirus stocks ) were maintained in SF900 II medium containing 10% FBS . The construct for PIV5 HN ectodomain ( 56–565 ) expression has been described previously [19] . Antibody specific for HN was polyclonal antibody ( PAb ) R471 , raised in rabbits against the purified HN ectodomain expressed by a recombinant baculovirus in insect cells . Hi5 insect cells were infected ( moi = 2 ) with a recombinant baculovirus stock containing the PIV5-HN ectodomain ( 56–565 ) construct and harvested 65 hr post infection . Protein was purified from the supernatant by affinity chromatography using Ni-NTA agarose ( Qiagen ) and was >90% pure by SDS-PAGE and Coomassie brilliant blue staining analysis . The S- and His-tags were cleaved from the expressed protein as previously described [21] . Purified protein was buffer exchanged into 10 mM Tris pH 7 . 4 , 50 mM NaCl and concentrated to ∼10 mg/mL , and a 1 . 2 molar excess of sialyllactose ( Sigma ) to HN monomers was added immediately prior to setting up crystallization trials . Initial crystals were obtained with the Crystal Screen HT ( Hampton ) by the hanging drop vapor diffusion method using a Mosquito ( TTP LabTech ) at the High Throughput Analysis Lab ( Northwestern University , Evanston ) . After optimization , crystals were grown at room temperature by the sitting drop vapor diffusion method over a reservoir solution containing 1 . 6 M ammonium sulfate , 0 . 1 M MES monohydrate pH 6 . 5 , 10% v/v 1 , 4-dioxane . Drops consisted of protein and precipitant at a 2∶1 ratio . The crystals were flash frozen in liquid nitrogen using the crystallization condition diluted with glycerol to 20% as the cryoprotectant solution . A native dataset was collected at the Life Sciences Collaborative Access Team ( LS-CAT ) beamline at the Argonne National Laboratory Advanced Photon Source and processed to 2 . 5 Å using HKL2000 [48] . The monomeric neuraminidase ( PDB ID: 1Z4X ) and 4HB stalk domains ( PDB ID: 3TSI ) of PIV5 HN were used as search models for molecular replacement to determine initial phases in the I4 spacegroup . Four NA domain monomers and the 4HB stalk domain were found in the asymmetric unit . Subsequent model building , structure refinement , and validation were performed with Coot [49] , PHENIX Refine [50] and MolProbity [51] , respectively . Use of TLS parameters ( as recommended by Phenix ) and individual B-factors during late stages of refinement helped to lower the Rfree values , however , the use of non-crystallographic symmetry restraints increased Rfree values . The data collection and final refinement statistics are shown in Table 1 . The atomic coordinates and structure factors have been deposited in the Protein Data Bank , www . pdb . org ( PDB ID: 4JF7 ) . A previously described pCAGGS-HN expression construct harboring the PIV5 ( W3A ) HN gene was used [19] . Mutants in pCAGGS HN were constructed as described previously [21] . IP of HN muts from transfected 293T cells was performed as described previously [21] . Surface expression of HN mutants in transfected 293T cells was performed as described previously [21] . Syncytia formation was measured in BHK-21 cells co-transfected with PIV5 F and HN mutants as described previously [21] . Levels of fusion between Vero cells co-transfected with pCAGGS F , pCAGGS HN , and pT7 luciferase and BSR-T7/5 cells expressing T7 RNA polymerase were quantitated using a previously described Luciferase reporter assay [21] . | Paramyxoviruses comprise a large family of significant pathogens including Newcastle disease virus ( NDV ) , parainfluenza viruses 1-5 ( PIV1-5 ) , respiratory syncytial virus , the highly transmissible measles virus , and the emerging and deadly Nipah and Hendra viruses . Five paramyxoviruses are U . S . Department of Health and Human Services and U . S . Department of Agriculture “select agents , ” and prevention and/or treatment of these viruses is a public health priority . Paramyxoviruses infect host cells through the concerted action of a “mushroom-shaped” receptor binding protein ( HN , H , or G ) and fusion protein ( F ) on the viral surface . However , despite numerous biochemical and structural insights , many details remain unknown about how these proteins interact and the mechanism by which the interaction triggers membrane fusion . Here we present the X-ray crystal structure of the PIV5 HN ectodomain comprised of a large fragment of the stalk and complete head domains . The structure reveals a unique conformation that is a hybrid of that seen in previous NDV ectodomain and PIV5 attachment protein head domain structures . A high-resolution view of the different orientations that head domains can adopt combined with recent biochemical data suggest a simple mechanism for paramyxovirus fusion . These new insights will help guide vaccine and inhibitor discovery efforts for paramyxoviruses . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"biochemistry",
"proteins",
"virology",
"protein",
"structure",
"host-pathogen",
"interaction",
"biology",
"microbiology"
]
| 2013 | Structure of the Parainfluenza Virus 5 (PIV5) Hemagglutinin-Neuraminidase (HN) Ectodomain |
The N6-methyladenosine ( m6A ) modification is the most prevalent internal RNA modification in eukaryotes . The majority of m6A sites are found in the last exon and 3’ UTRs . Here we show that the nuclear m6A reader YTHDC1 is essential for embryo viability and germline development in mouse . Specifically , YTHDC1 is required for spermatogonial development in males and for oocyte growth and maturation in females; Ythdc1-deficient oocytes are blocked at the primary follicle stage . Strikingly , loss of YTHDC1 leads to extensive alternative polyadenylation in oocytes , altering 3’ UTR length . Furthermore , YTHDC1 deficiency causes massive alternative splicing defects in oocytes . The majority of splicing defects in mutant oocytes are rescued by introducing wild-type , but not m6A-binding-deficient , YTHDC1 . YTHDC1 is associated with the pre-mRNA 3’ end processing factors CPSF6 , SRSF3 , and SRSF7 . Thus , YTHDC1 plays a critical role in processing of pre-mRNA transcripts in the oocyte nucleus and may have similar non-redundant roles throughout fetal development .
More than one hundred different RNA modifications are known in eukaryotes [1] . N6-methyladenosine ( m6A ) is the most prevalent internal modification in eukaryote mRNAs , occurring in transcripts of approximately one third of genes in human and mouse [2–4] . Globally , m6A is enriched in the 3' most exons , long internal exons , and 5’ untranslated regions ( UTRs ) [5–9] . In addition to mRNAs , m6A is also present in long non-coding RNAs such as Xist , small nuclear RNAs , and ribosomal RNAs [10–12] . The m6A RNA modification is widely conserved among eukaryotes including yeast , flies , and plants [13–17] . Generation of m6A is catalyzed by a multi-component methyltransferase ( m6A writer ) consisting of methyltransferase-like 3 ( METTL3 ) , methyltransferase-like 14 ( METTL14 ) , and Wilm’s tumor associated protein ( WTAP ) [8 , 18–20] . m6A is a reversible modification and two m6A demethylases have been identified: fat mass and obesity-associated protein ( FTO ) and alkB homolog 5 ( ALKBH5 ) [21 , 22] . Readers of the m6A mark preferentially bind to m6A and elicit downstream functions . Five mammalian m6A readers contain the YTH ( YT521-B homology ) domain: YTHDF1 , 2 , 3 and YTHDC1 , 2 [6 , 23–26] . YTHDF1 , 2 , and 3 are cytoplasmic [6 , 23] . YTHDC1 localizes to the nucleus in cultured mammalian somatic cells [27 , 28] , whereas YTHDC2 is cytoplasmic in meiotic spermatocytes [29–33] . The m6A modification occurs preferentially at the conserved RRACH motif ( R: G or A; H: A , C , or T ) [34] . The YTH domain is an RNA-binding motif [35] and crystal structural studies reveal that the YTH domain of YTHDC1 selectively binds to m6A in the consensus motif [24 , 25] . In addition to the five YTH domain-containing m6A readers , a number of RNA-binding proteins lacking a YTH domain are m6A readers: IGF2BP proteins [36] , FMR1 [37] , the translation initiation factor eIF3 complex [38] , HNRNPA2B1 [39 , 40] , HNRNPC [41] , and HNRNPG [42] . The HNRNP family members are considered “indirect” m6A readers , because m6A alters the local RNA structure to facilitate their binding to m6A [4 , 40–42] . m6A functions in key RNA metabolic processes . m6A regulates gene expression [5 , 6] , mRNA stability [23 , 43] , translation efficiency [44 , 45] , alternative splicing [15 , 16 , 46] , and cytoplasmic mRNA turnover [23 , 47] . m6A is also involved in a number of developmental processes . In yeast , m6A formation occurs only during meiosis and is catalyzed by IME4 , which is the sequence homologue of mammalian METTL3 and induces meiosis [13 , 14] . m6A modulates alternative splicing of Sxl ( sex lethal ) transcript and thus sex determination in Drosophila [15 , 16] . m6A is abundant on the long non-coding RNA Xist and promotes Xist-mediated gene silencing during X-inactivation [10] . Inactivation of Mettl3 in mouse or IME4 in Drosophila leads to embryonic lethality , demonstrating an essential role for m6A in lineage differentiation [48 , 49] . Mouse Mettl3 is required for spermatogonial development and spermatogenesis [50 , 51] . Disruption of the m6A demethylase gene Alkbh5 causes male infertility in mouse [22] , whereas YTHDC2 is required for spermatogenesis and oogenesis in mouse [26 , 31–33] . YTHDF2-mediated clearance of maternal transcripts promotes zygotic genome activation in zebrafish [52] . Mouse YTHDF2 regulates maternal transcript dosage and is essential for female fertility [53] . In addition , knockdown studies have uncovered a role of m6A in zebrafish development [20] , circadian rhythm [54] , cell reprogramming [7 , 49 , 55] , and miRNA biogenesis and effects [39 , 56] . Therefore , m6A plays important roles in a large number of developmental processes . We previously identified YTHDC1 as a meiotic chromatin-associated protein in a proteomic screen [57] . YTHDC1 ( initially referred to as YT521-B ) changes alternative splicing patterns in a concentration-dependent manner [27] and localizes to nuclear speckles , which contain active transcription sites [28] . Tyrosine phosphorylation of YTHDC1 regulates its intra-nuclear localization , thereby modulating its effects on alternative splicing [58] . YTHDC1 promotes exon inclusion by recruitment of serine/arginine-rich ( SR ) splicing factor 3 ( SRSF3 ) , a pre-mRNA splicing factor [46] . YTHDC1 facilitates nuclear export of m6A-containing mRNAs through SRSF3 and NXF1 [59] . Although these studies in cultured cells have provided important insights into the function of YTHDC1 , its requirement during development is unknown . In addition , the biological function of accumulation of m6A sites in 3’ UTRs remains mysterious . Here , we report that YTHDC1 is essential for embryonic development in the mouse . Using a conditional inactivation approach , we find that YTHDC1 is required for survival of spermatogonia in males and controls postnatal oocyte development in females . Strikingly , in addition to alternative splicing defects , loss of YTHDC1 causes widespread alternative polyadenylation in oocytes . Importantly , YTHDC1 is associated with SR proteins and pre-mRNA 3’ end processing factors .
We examined expression of YTHDC1 in adult mouse tissues using polyclonal antibodies raised against an N-terminal region of mouse YTHDC1 ( S2A Fig ) . Western blot analysis showed that YTHDC1 was expressed in multiple adult mouse tissues including brain , testis , and ovary , with an apparent molecular weight of ~120 kDa ( Fig 1A ) . High levels of YTHDC1 were present in postnatal oocytes , MII eggs , and pre-implantation embryos , and low levels in germinal vesicle ( GV ) stage oocytes ( Fig 1B ) . The increase in YTHDC1 protein abundance between the GV oocyte stage and MII egg suggests that YTHDC1 is encoded by a dormant maternal mRNA that is recruited during oocyte maturation . Immunostaining showed that YTHDC1 localized to the nucleus in postnatal oocytes and pre-implantation embryos , with the increase in staining between the GV oocyte and MII egg being consistent with the immunoblotting results ( Fig 1C ) . The diffuse cytoplasmic signal of YTHDC1 and its increased abundance in MII oocytes suggest that Ythdc1 is under translational control , possibly in preparation for zygotic activation at the two-cell stage . The nuclear localization of YTHDC1 is consistent with a previous finding that it is associated with chromatin [57] . Notably , in postnatal day ( PND ) 5 and 12 oocytes , transcription is active and YTHDC1 is nuclear . In adult testis ( S1 Fig ) , YTHDC1 is nuclear in spermatogonia , spermatocytes , and round spermatids , which are transcriptionally active . However , YTHDC1 is absent in elongating and elongated spermatids , which are transcriptionally inactive due to nuclear condensation ( S1 Fig ) . Therefore , the nuclear localization of YTHDC1 in cells with active transcription suggests that it is involved in co-transcriptional and/or post-transcriptional regulations . To determine the physiological function of Ythdc1 , we generated a Ythdc1 floxed ( conditional ) allele , referred to as Ythdc1fl , by gene targeting in embryonic stem ( ES ) cells ( S2B Fig ) . Ythdc1fl/fl mice were healthy and fully fertile . We next crossed Ythdc1fl/fl mice to Actb-Cre mice , which express Cre ubiquitously , to obtain mice with a Ythdc1 null allele ( Ythdc1+/- ) [60] . Cre-mediated excision of the floxed exons removes the YTH domain and causes a frameshift in the resulting Ythdc1 mutant transcript ( S2B Fig ) . Intercrosses of Ythdc1+/- mice did not produce any Ythdc1-/- pups , suggesting that Ythdc1 is essential for embryonic development ( S2C Fig ) . To determine the time of developmental failure , we genotyped fetuses recovered from intercrosses of Ythdc1+/- mice at embryonic day 8 . 5 ( E8 . 5 ) , E9 . 5 , and E11 . 5 . No Ythdc1-/- embryos were found at E11 . 5 . Out of 9 embryos at E8 . 5 and out of 33 embryos at E9 . 5 , only one Ythdc1-/- embryo was found at each time point ( S2C Fig ) . Resorbed embryos were found at E8 . 5 through E11 . 5 and expected to be homozygous mutants based on their Mendelian distribution ( S2C Fig ) . These results show that YTHDC1 is indispensable for embryo development past early post-implantation stages . To bypass the embryonic lethality resulting from Ythdc1 deficiency , we used Ddx4-Cre to inactivate Ythdc1 specifically in the germline to generate Ythdc1fl/- Ddx4-Cre ( referred to as Ythdc1cKO ) mice ( Figs 2 and 3A ) . All subsequent studies were conducted with Ythdc1fl/- Ddx4-Cre ( cKO ) mice unless noted otherwise . Ddx4-Cre expression begins at ~E15 in both male and female germ cells but differs in the developmental stage of onset due to the sexual dimorphism in the timing of meiotic entry [61] . In males , Ddx4-Cre is expressed in mitotic germ cells including spermatogonia prior to meiosis , whereas in oocytes , Ddx4-Cre expression occurs only after meiotic entry ( Fig 3A ) . Ythdc1cKO mice were viable and grossly normal . Seminiferous tubules from newborn ( PND0 ) Ythdc1cKO males contained prospermatogonia ( Fig 2A ) , which lacked YTHDC1 as determined by immunostaining ( S3 Fig ) . Tubules from PND8 Ythdc1cKO males contained substantially fewer spermatogonia than those from control Ythdc1fl/+ or Ythdc1fl/- males ( Fig 2B ) . However , testes from PND25 and adult Ythdc1cKO males lacked any germ cells including mitotic spermatogonia and exhibited a Sertoli-cell-only phenotype ( Fig 2C and 2D ) , demonstrating that Ythdc1 is required for development of spermatogonia and male fertility . In contrast to the absence of germ cells in adult Ythdc1 cKO testis , oocytes were present in ovaries from 8-week-old Ythdc1fl/- Ddx4-Cre ( cKO ) females ( Fig 3 ) . Wild-type adult ovaries contained follicles at different developmental stages , including primary , secondary , and antral follicles ( Fig 3B and 3C ) . However , Ythdc1fl/- Ddx4-Cre ovaries lacked secondary or antral follicles , indicating that oocyte development was blocked at the primary follicle stage , which is characterized by one layer of granulosa cells surrounding the oocyte ( Fig 3B ) . Histological analysis of ovaries from older Ythdc1fl/- Ddx4-Cre females ( 6-month and beyond ) showed a complete loss of oocytes . Western blot analysis confirmed that YTHDC1 protein was absent in oocytes collected from Ythdc1fl/- Ddx4-Cre ovaries ( S4A Fig ) . As expected , a nuclear immunofluorescent signal of YTHDC1 was not detected in Ythdc1 mutant oocytes ( S4B Fig ) . These results confirm the specificity of our YTHDC1 antibody and the complete depletion of YTHDC1 in Ythdc1fl/- Ddx4-Cre oocytes . Because expression of Ddx4-Cre begins at the pachytene stage of meiotic prophase I during fetal development , it is not clear whether the observed defects in Ythdc1fl/- Ddx4-Cre postnatal ovaries were due to the requirement of YTHDC1 at embryonic or postnatal stages . To investigate whether postnatally expressed YTHDC1 is required for oocyte development , we used Zp3-Cre to inactivate Ythdc1 in oocytes postnatally ( Fig 3A ) . Zp3-Cre is expressed in developing oocytes around postnatal day 3 ( Fig 3A ) [62] . We found that Ythdc1fl/- Zp3-Cre ovaries exhibited similar defects in folliculogenesis as observed in Ythdc1fl/- Ddx4-Cre ovaries–blockade at the primary follicle stage ( Fig 3C ) . We next performed mating tests of three Ythdc1fl/- Zp3-Cre females and three wild-type littermate control females . At the age of 8 weeks , each female was housed with one wild-type male for two months . The three control females produced two litters each ( 6 . 8 ± 1 . 4 pups/litter ) , whereas none of the three Ythdc1fl/- Zp3-Cre females produced any offspring . Taken together , these genetic studies demonstrate that YTHDC1 plays an essential role in postnatal oocyte development . We were able to retrieve oocytes from ovaries of Ythdc1fl/- Ddx4-Cre females at the ages of 3–6 weeks by poking . However , the number of oocytes retrieved from Ythdc1fl/- Ddx4-Cre females was only 10% ( n = 3 ) that of wild-type littermates . In addition , the GV oocytes from Ythdc1fl/- Ddx4-Cre females were not able to resume meiosis in vitro . In contrast to the smooth appearance of wild-type germinal vesicle ( GV ) stage oocytes , Ythdc1-deficient oocytes contained one or two prominent granules in the cytoplasm ( Fig 4A ) . Such granules were not observed in wild-type oocytes . These granules stained positive with Sytox green , which recognizes both DNA and RNA , suggesting a nucleic acid content ( Fig 4B ) . When double stained with both DAPI and Sytox green , the nuclei of Ythdc1-deficient oocytes were positive for both stains , whereas the granules only retained the Sytox green stain , indicating that the granules contained RNA but not DNA ( Fig 4B ) . To our knowledge , such large RNA granules have not been observed before . The appearance of large cytoplasmic RNA granules indicates severe defects in RNA metabolism in oocytes in the absence of YTHDC1 . It is possible that , like P granules , incorrectly processed RNAs are sequestered in these novel RNA granules in oocytes in the absence of YTHDC1 . Knockdown of YTHDC1 in HeLa cells causes acute nuclear accumulation of mRNAs within hours [59] . We did not observe nuclear accumulation of RNAs in postnatal Ythdc1-deficient oocytes , possibly because inactivation of Ythdc1 begins at E15 ( Fig 3A ) , weeks prior to our analysis . To investigate the consequences of Ythdc1 deficiency on the oocyte transcriptome , we performed RNA-seq analysis of oocytes collected from 6-week-old wild-type and Ythdc1fl/- Ddx4-Cre females ( S1 Table ) . With a FDR cutoff of 0 . 01 , a total of 4933 transcripts showed differential expression: 2656 transcripts were up-regulated and 2277 transcripts down-regulated in Ythdc1-deficient oocytes compared with control oocytes , indicating that the transcriptome in Ythdc1-deficeint oocytes was dramatically altered ( S5A Fig and S2 Table ) . Validation of 10 randomly selected differentially abundant transcripts by real-time PCR confirmed the RNA-seq findings ( S5B Fig ) . Gene Ontology ( GO ) analysis identified a number of significantly altered biological processes for both up-regulated and down-regulated transcripts , with regulation of transcription as the most significantly affected process ( S5C Fig ) . Because YTHDC1 affects alternative splicing in cultured somatic cells [27] , we next analyzed the oocyte RNA-seq data to systematically identify local splicing variants ( LSVs ) between wild-type and Ythdc1-deficient oocytes using the MAJIQ package [63] . We identified a total of 2937 significant LSVs ( q < 0 . 05 ) with ΔPSI ( difference in percent spliced in ) > 0 . 2 ( Fig 5A and S3 Table ) . These LSVs affected 1966 genes , involved 10 , 266 exons , and included differential retention of 500 introns . Of the 1966 genes with LSVs , 34% ( 659 genes ) were differentially expressed between wild-type and Ythdc1-deficient oocytes ( up-regulated , 245 genes; down-regulated , 414 genes ) . According to GO analysis , these changes affect genes involved in multiple fundamental biological processes , including chromatin modification , regulation of transcription , mRNA processing , and regulation of translation ( S6 Fig ) . We designed RT-PCR assays to validate different types of MAJIQ-identified splicing events: exon inclusion/skipping , intron retention , and splicing in 3’ UTRs . Using GV-stage oocytes from wild-type and Ythdc1fl/- Ddx4-Cre mice , RT-PCR analysis confirmed 90% ( 9 of 10 tested ) of LSVs involving internal exons ( Fig 5B and 5C ) . For example , as illustrated in the gene track view , the second exon ( part of the 5’UTR ) of Tmem2 was partially skipped in wild-type but not in Ythdc1-deficient oocytes ( Fig 5B ) . Two LSVs affecting Jam2 and Spata7 , respectively , involved nearly complete exon skipping in wild-type oocytes but partial exon skipping in Ythdc1-deficient oocytes . Exon skipping in two genes ( Dner and Rad1 ) was complete in Ythdc1-deficient oocytes but partial in wild-type . Four LSVs in Enpp5 , Hip1r , Rap1a , and Parp6 resulted in partial exon skipping in Ythdc1-deficient oocytes whereas no skipping occurred in wild-type . There was no apparent preference for the directionality of exon skipping in regard to genotype . In total , our validation results suggest that most LSVs predicted by MAJIQ are true splicing events . LSVs involving introns or 3’ UTRs were more complex . We tested nine intron-retention LSVs predicted by MAJIQ and confirmed six of these ( 67% ) by RT-PCR ( Fig 6A and 6B ) . Among three confirmed LSVs , two transcripts ( Dnpep and Mcph1 ) showed intron retention preferentially in wild-type oocytes , whereas one mRNA ( Phf1 ) retained introns preferentially in Ythdc1-deficient oocytes . We examined 13 MAJIQ-predicted LSVs involving splicing within 3’ UTRs and validated three of these ( 23% ) by RT-PCR ( Fig 6C and 6D ) : Ifnar1 , Abl2 , and Ikzf5 . Coincidentally , all of these transcripts were associated with longer 3’ UTRs in wild-type oocytes . Although MAJIQ was not designed for the analysis of changes in introns and 3’ UTR length as pointed out by the MAJIQ authors , we were able to validate most of the MAJIQ-predicted LSVs involving introns and some of the 3’ UTR splicing events . We further examined the 9 LSVs involving exon inclusion/skipping using GV-stage oocytes from 6-week-old Ythdc1fl/- Zp3-Cre and wild-type females . All 9 LSVs validated in Ythdc1fl/- Ddx4-Cre oocytes ( Fig 5 ) were also confirmed in Ythdc1fl/- Zp3-Cre oocytes ( S7 Fig ) . Collectively , our results show that inactivation of YTHDC1 in oocytes causes severe defects in mRNA splicing . The majority of m6A sites are present in the 3’ most exons , raising the possibility that m6A may play a role in regulating 3’ UTR length [5 , 9] . Many genes produce transcripts with 3’ UTRs of different lengths due to usage of alternative polyadenylation ( APA ) sites and 3’ UTRs contain sites for microRNAs and RNA-binding proteins . Thus , the 3’ UTR of a particular mRNA regulates its translation and subcellular localization [64 , 65] . For instance , transcripts in brain exhibit extensive lengthening of 3’ UTRs due to APA [66] . We systematically analyzed the 3’ UTR length of wild-type versus Ythdc1-deficient oocytes using the ROAR algorithm [67] . The ROAR program identifies alternative polyadenylation using standard RNA-seq data by measuring the reads upstream ( pre ) and downstream ( post ) of the annotated polyadenylation site ( PAS ) ( Figs 7A and 6B ) . ROAR analysis of our oocyte RNA-seq data revealed 1210 alternative polyadenylation ( APA ) events in 864 genes between wild-type and Ythdc1 mutant oocytes ( cutoff , p value < 0 . 05; Fig 7A and S4 Table ) . Some genes had more than one differential APA event . Overall , 709 APA events ( ROAR < 1 ) resulted in higher levels of the longer isoform ( longer 3’ UTR ) in Ythdc1-deficient oocytes , whereas 501 APA events ( ROAR > 1 ) was associated with higher levels of the shorter isoform in the mutant ( Fig 7A ) . We chose 8 transcripts with predicted longer 3’ UTRs in the Ythdc1 mutant for RT-PCR validation ( Fig 7B ) . These 8 transcripts were not differentially expressed between wild-type and mutant oocytes . Our validation strategy involved one PCR assay ( termed PRE ) that amplified both short and long transcripts and a second PCR assay ( termed POST ) that was specific for the long isoform . The POST RT-PCR assay for Arl5a produced a stronger signal from Ythdc1-deficient versus wild-type oocytes , indicating that the former contained a higher level of the long isoform . Of the eight transcripts tested , seven ( 88% ) ( Arl5a , Ddx21 , Noc3l , Rybp , Scamp1 , Slc11a2 , and Slc25a51 ) preferentially produced the longer isoform in Ythdc1-deficient oocytes due to APA , whereas , one transcript ( Frs2 ) did not exhibit detectable differences in isoform prevalence using this assay ( Figs 7C and 6D ) . These results demonstrate extensive alternative polyadenylation in Ythdc1-deficient oocytes . Previous findings in brain tissue [9] have shown that five of seven gene transcripts with APA defects contained known m6A sites in the last exons: Arl5a , Ddx21 , Noc3l , Slc11a2 , and Slc25a51 , implicating m6A in regulation of alternative polyadenylation . To examine the effect of m6A on splicing in oocytes , we collected oocytes from wild-type and Ythdc1fl/- Ddx4-Cre ovaries at postnatal day 12 ( PND12 ) , when oocytes are still transcriptionally active . We evaluated alternative splicing of the nine transcripts for which we had identified splicing defects in GV stage oocytes from 6-week-old Ythdc1fl/- Ddx4-Cre mice ( Fig 5 ) and found that all nine transcripts showed similar splicing defects in PND12 Ythdc1-deficient oocytes ( Fig 8 , first two lanes of center panel ) . However , there were notable differences for two transcripts: Rad1 and Tmem2 . The Rad1 two-exon-skipping isoform was detected in PND12 mutant oocytes ( Fig 8 ) but not in 6-week-old mutant oocytes ( Fig 5 ) . Similarly , the Tmem2 spliced short isoform was present in PND12 mutant oocytes but absent in 6-week-old oocytes . These data suggest that the short isoforms of Rad1 and Tmem2 were degraded during the long period between cessation of transcription at PND20 and the time point of analysis at 6 weeks-of-age . To investigate if alternative splicing defects in Ythdc1-deficient oocytes could be rescued by supplying YTHDC1 , we used transcriptionally active PND12 oocytes . We injected PND12 Ythdc1-deficient oocytes with in vitro transcribed wild-type or mutant Ythdc1 mRNA , followed by overnight culture . The Ythdc1 mutant mRNA contains two missense mutations ( W377A , W428A ) that completely abolish the m6A binding activity of YTHDC1 [24] . We quantified the FLAG-YTHDC1 protein levels in the nucleus of injected Ythdc1-deficient oocytes by immunofluorescence and confocal microscopy and found no difference in YTHDC1 protein levels between oocytes ( 4 oocytes each ) injected with wild-type and mutant Ythdc1 mRNAs . We found that exon skipping and exon inclusion defects in six transcripts were rescued in Ythdc1-deficient oocytes injected with wild-type but not mutant Ythdc1 mRNA , whereas no difference was observed for the remaining three genes ( Dner , Enpp5 and Jam2 ) ( Fig 8 ) . These rescue experiments suggest that the majority of alternative splicing defects in Ythdc1-deficient oocytes is m6A-dependent . However , we cannot rule out the possibility that the failure of mutant YTHDC1 ( ( W377A , W428A ) to rescue might be caused by reduced RNA-binding activity or instability , independent of m6A . To elucidate the mechanism by which YTHDC1 affects alternative polyadenylation , we investigated potential interactions of YTHDC1 with pre-mRNA 3’end cleavage and polyadenylation factors by co-immunoprecipitation ( Fig 9 ) . Cleavage factor Im ( CFIm ) and cleavage stimulating factor ( CSTF ) are two multi-protein complexes that bind to upstream sequence elements ( USE ) and downstream sequence elements ( DSE ) around the PAS , respectively [68 , 69] . We found that YTHDC1 was associated with CPSF6 , one of the four subunits of the CFIm complex ( Fig 9A ) . This result is consistent with previous reports identifying CPSF6 among proteins co-immunoprecipitated with human YTHDC1 in 293T cells [46] . However , YTHDC1 did not interact with NUDT21 , another subunit of the CFIm complex ( Fig 9A ) . In addition , YTHDC1 was not associated with cleavage stimulating factors CSTF1 or CSTF2 by co-transfection and co-IP assays . Interestingly , knockdown of Cpsf6 induces widespread use of proximal PAS , resulting in 3’ UTR shortening [70 , 71] . Moreover , a mutation in the Medaka Cpsf6 gene causes 3’ UTR shortening in developing embryos and a defect in primordial germ cell migration [72] . YTHDC1 interacts with the SR splicing factors SRSF3 and SRSF7 ( Fig 9B ) . SRSF3 and SRSF7 couple RNA processing with mRNA export through association with the nuclear mRNA export factor NXF1 [73] . SRSF3 and SRF7 bind to the last exons and regulate polyadenylation in an opposing manner . Knockdown of SRSF3 leads to 3’ UTR shortening , whereas depletion of SRSF7 results in 3’ UTR lengthening [73] . In conclusion , these results support a model in which YTHDC1 regulates alternative polyadenylation through interaction with the 3’end processing machinery .
Here , we report that the nuclear m6A reader YTHDC1 is essential for mouse embryogenesis and germline development , and describe a critical role of YTHDC1 in orchestrating m6A-dependent processing of pre-mRNA transcripts in oocytes . Our studies implicate YTHDC1 in the choice of polyadenylation sites , which determines the length of 3’ UTRs . The 3’ UTR contains target sites for microRNAs and many RNA-binding proteins . Therefore , lengthening or shortening of 3’ UTR would predictably have profound effects on translation efficiency , transcript stability , and subcellular transcript localization [64 , 65 , 68 , 69] . Precise translational control of maternal transcripts is especially critical during oocyte maturation , due to lack of transcription during this prolonged stage . We find that loss of YTHDC1 in oocytes results in alternative polyadenylation and thus altered 3’ UTR length in more than 800 genes . To date , YTHDC1 is the only m6A reader that has been demonstrated to regulate 3’ UTR length . Triple knockdown of three m6A writer components ( METTL3 , METTL14 , and WTAP ) in human A549 cells changes the usage of proximal versus distal polyadenylation sites with some switching to proximal sites and others switching to distal sites , demonstrating a critical role for m6A in regulation of 3’ UTR length [9] . In addition , ALKBH5 , an m6A demethylase , regulates 3’UTR length in male germ cells [74] . How the m6A signal is relayed to the 3’ end processing machinery is unknown . A number of multi-protein complexes participate in pre-mRNA 3’ end cleavage and polyadenylation , including cleavage factor Im ( CFIm ) , cleavage and polyadenylation specificity factor ( CPSF ) , cleavage stimulating factor ( CSTF ) , and poly ( A ) -binding proteins [68 , 69] . We find that YTHDC1 forms complexes with components of the 3’ end processing machinery: CPSF6 ( a CFIm component ) , SRSF3 , and SRSF7 ( Fig 9 ) . These factors bind to the 3’ UTR around the PAS . Specifically , the CFIm binds to the UGUA motif upstream of the PAS . Knockdown of each of these factors in cell culture causes a shift in PAS usage , resulting in APA . Knockdown of CPSF6 favors usage of proximal PAS [70 , 71] , whereas knockdown of SRSF7 causes preferential usage of distal PAS [73] . Our data support a model in which YTHDC1 recognizes m6A in the last exons of pre-mRNA transcripts and orchestrates the choice of polyadenylation sites through interactions with 3’end processing factors . YTHDC1 may recruit these factors to the 3’ UTRs or sequester them in the nucleoplasm through interactions , resulting in opposing APA patterns . Alternatively , these factors may compete for binding to YTHDC1 . In addition , SRSF3 and SRSF7 link alternative polyadenylation with nuclear export through interaction with NXF1 [73] . Therefore , it is conceivable that , through interaction with SRSF3 and SRSF7 , YTHDC1 may couple m6A in alternatively polyadenylated transcripts with nuclear export . In cultured cells , YTHDC1 facilitates binding of m6A-modified nuclear transcripts to SRSF3 and NXF1 and mediates nuclear export [59] . Our study reveals an essential role for YTHDC1 in development of both the embryo and the germline . Proteins ( writers , readers , and erasers ) involved in establishment , recognition , and erasure of m6A sites in mRNAs play important roles in development and fertility in mouse . Loss of the key m6A writer enzyme METTL3 causes early post-implantation lethality with defects in lineage priming [49] . m6A mainly reduces mRNA stability in embryonic stem cells and pre-implantation embryos and its loss leads to a failure in termination of naïve pluripotency during lineage specification [49] . Conditional inactivation of Mettl3/Mettl14 reveals their essential role in spermatogenesis [50 , 51] . Alkbh5-deficient mice are viable but exhibit impaired spermatogenesis with increased apoptosis of meiotic spermatocytes [22] . ALKBH5-mediated m6A demethylation affects mRNA export . The cytoplasmic m6A reader YTHDC2 interacts with the meiosis-specific protein MEIOC [29 , 30] . Ythdc2-deficient mice are viable but sterile due to a failure in meiotic progression [26 , 31–33] . YTHDC2 together with MEIOC promotes translation efficiency of its target transcripts but decreases their mRNA abundance . In addition , YTHDC2 modulates the level of m6A-enriched transcripts in germ cells , which is required for progression through meiosis [32] . Ythdf2 deficiency causes incomplete penetrance of lethality and female-specific infertility [53] . Similar to Mettl3 , inactivation of Ythdc1 is embryonic lethal , showing that loss of YTHDC1 is not compensated for by other m6A readers . The lack of compensation is not entirely surprising , given that , to date , YTHDC1 is the only known m6A reader in the nucleus . By conditional inactivation of Ythdc1 in the germline , we find that YTHDC1 is essential for fertility in both males and females . Specifically , YTHDC1 is required for development of mitotic spermatogonia in males and oocyte growth in females . Because of loss of spermatogonia in Ythdc1fl/- Ddx4-Cre males , different Cre drivers will be needed to examine its role in meiotic spermatocytes and post-meiotic round spermatids in future studies . Strikingly , the mouse mutant phenotypes of three m6A readers YTHDC1 , YTHDC2 , and YTHDF2 are different , suggesting non-redundant functions . YTHDC2 is required for meiotic progression in both sexes but is dispensable for viability [26 , 31–33] . YTHDF2 is partially necessary for viability and specifically required for female fertility and oocyte competence [53] . Here we find that YTHDC1 is essential for viability and is required for spermatogonial development in males and oocyte growth in females . About 200 transcripts were upregulated in Ythdf2-deficient MII oocytes [53] . The overlap between the upregulated transcripts in Ythdf2-deficient oocytes and Ythdc1-deficient oocytes is significant ( 1 . 48-fold enrichment , p = 0 . 0004 ) , suggesting that YTHDC1 may play a later role in oocyte competence . In Ythdc1fl/- Ddx4-Cre or Ythdc1fl/- Zp3-Cre females , germ cells progress through the prophase of meiosis I , however , oocyte development is blocked at the primary follicle stage . This blockade is similar to the oocyte growth arrest in females lacking GDF9 , a key TGFβ receptor ligand [75] . Several early studies using cultured cells show that YTHDC1 is involved in alternative splicing of internal exons in a dosage–dependent manner [27 , 28 , 58] . YTHDC1 localizes to so-called YT bodies in the nucleus that contain active transcription sites . Tyrosine phosphorylation of YTHDC1 regulates its solubility in the nucleus and its effect on alternative splicing . Structural demonstration of YTHDC1 as an m6A reader raises a possible connection between m6A and alternative splicing [24 , 25] . Among the YTH domain proteins , only YTHDC1 contains a selective binding pocket for the nucleotide preceding the m6A nucleotide [25] . Two pre-mRNA splicing factors SRSF3 and SRSF10 competitively bind to YTHDC1 [46] . It was proposed that YTHDC1 promotes exon inclusion by recruiting SRSF3 while blocking SRSF10 binding to target transcripts [46] . In this study , we find that YTHDC1 regulates mRNA splicing in oocytes . In addition , loss of YTHDC1 leads to formation of large novel cytoplasmic RNA-containing granules in the oocyte cytoplasm , which may contain aberrantly processed transcripts . Furthermore , the m6A-binding activity of YTHDC1 is required for rescue of alternative splicing defects in Ythdc1-deficient oocytes . Collectively , these in vitro and in vivo studies demonstrate the critical role of YTHDC1 in the regulation of alternative splicing , apparently in an m6A-dependent manner . A number of studies show that m6A is a determinant of mRNA stability and turnover in the cytoplasm [23 , 47 , 76] . YTHDC1 facilitates nuclear export of m6A-containing mRNAs through its interaction with SRSF3 and thus regulates their cytoplasmic abundance [59] . YTHDC2 regulates the levels of m6A-containing transcripts in meiotic germ cells [32] . Binding by YTHDF2 causes redistribution of bound mRNAs to RNA degradation sites [23] . In addition , YTHDF2 regulates maternal mRNA clearance in both zebrafish and mouse [52 , 53] . In contrast with the established role of m6A in mRNA turnover , the role of m6A in splicing has been a point of contention in the field . Some studies in ES cells conclude that m6A in nascent transcripts has a minor role in splicing , even though Mettl3 inactivation in ES cells affects 3% of ~12 , 000 alternative cassette exons [47 , 49] . It is possible that Mettl3 inactivation may have more pronounced effect on splicing in differentiated cells . Indeed , Mettl3 inactivation in male germ cells affects splicing [51] . The differentially spliced genes in Ythdc1-deficient oocytes significantly overlap with the differentially spliced genes in Mettl3-deficient testes ( S8 Fig ) . Study of Alkbh5-deficient spermatogenic cells also supports a role of m6A in the regulation of splicing [74] . Therefore , the extent of effect on splicing by m6A most likely varies in different cell types and developmental stages .
Mice were maintained and used for experimentation according to the protocol approved by the Institutional Care and Use Committee of the University of Pennsylvania . The GST-YTHDC1 ( aa 3–109 ) fusion protein ( S2A Fig ) was expressed in E . coli using the pGEX4T-1 vector and affinity purified with glutathione Sepharose 4B ( GE Healthcare ) . Rabbits were immunized with recombinant protein , yielding antisera UP2410 and UP2411 ( Cocalico Biologicals Inc . ) . For western blotting and immunofluorescence , antibodies were affinity purified against the GST fusion protein . The Ythdc1 targeting construct was designed to insert two tandem copies of loxP-flanked hygromycin phosphotransferase-thymidine kinase ( HyTK ) cassettes into Ythdc1 intron 4 , and a loxP site into intron 9 ( S2B Fig ) . Genomic fragments were amplified from the Ythdc1-containing BAC clone RP24-567O8 by PCR with high-fidelity Taq DNA polymerase . The targeting construct was confirmed by sequencing . ClaI-linearized targeting construct was electroporated into V6 . 5 mouse embryonic stem ( ES ) cells , and ES cells were cultured in media containing 120 μg/ml hygromycin B . Of 368 hygromycin-resistant ES clones screened by long-range PCR , three clones were homologously targeted . Two positive clones ( 1A6 and 3D6 ) were expanded and electroporated with the Cre-expressing plasmid pOG231 , followed by culture in media containing 2 μM ganciclovir . Ninety-six clones were screened for removal of the HyTK cassette and presence of loxP sites flanking Ythdc1 exons 5–9 ( S2B Fig ) , resulting in seven positive clones . Two ( 1A6H10 and 3D6G7 ) Ythdc1fl/+ ES clones were injected into blastocysts . The resulting chimeric mice transmitted the Ythdc1 floxed allele through the germline . Heterozygous ( Ythdc1+/- ) animals were produced by mating Ythdc1fl/+ with Actb-Cre mice [60] . Mice with conditional deletion of Ythdc1 were obtained from the intercrosses of Ythdc1fl/+ with Ddx4-Cre or Zp3-Cre mice [61 , 62] . The resulting Ythdc1fl/+ Cre males were crossed with Ythdc1fl/fl females , yielding Ythdc1fl/- Cre mice with germline-specific inactivation . Offspring were genotyped by PCR of genomic DNA with the following primers: wild-type ( 396 bp ) and Ythdc1 floxed allele ( 473 bp ) , CTTCCAGCAGGAATGAGTGC and GGCAATAAATAGCCCCAAAA; Ythdc1- ( deletion ) ( 426 bp ) , GATATCTTCTCTGATTCATGCG and GGCAATAAATAGCCCCAAAA; Ddx4-Cre ( 240 bp ) , CACGTGCAGCCGTTTAAGCCGCGT and TTCCCATTCTAAACAACACCCTGAA; Zp3-Cre ( 220 bp ) , CCCAGATTCTGATCGTTGGT and CAGGTTCTTGCGAACCTCAT . Full-grown , germinal vesicle ( GV ) -intact oocytes , metaphase II ( MII ) eggs , fertilized eggs and preimplantation embryos were collected as previously described [77 , 78] . GV oocytes were cultured in Chatot-Ziomek-Brinster ( CZB ) medium [79] containing 2 . 5 μM milrinone ( Sigma , St . Louis , MO , USA ) to inhibit GV breakdown [80]; MII eggs were cultured in CZB medium and fertilized eggs/embryos cultured in KSOM [81] . GV oocytes were collected by poking of ovaries or enzymatic digestion . For enzymatic digestion , ovaries were dissected out and placed in Ca2+-Mg2+-free CZB medium containing 1 mg/ml collagenase ( #LS004196 , Worthington Biochemical Corp ) and 0 . 2 mg/ml DNase I ( Sigma #DN-25 ) in 35 mm petri dish . Each ovary was chopped into 4–5 pieces . Enzymatic digestion was carried out at 37°C for 40 min . Ovaries were pipetted up and down several times using a P1000 pipette to facilitate cell dissociation . Oocytes free of follicle cells were transferred and washed with three drops of CZB medium before further analysis . Equal numbers of GV oocytes , metaphase I ( MI ) eggs , MII eggs , fertilized eggs and embryos were lysed in 2xSDS loading buffer ( Sigma ) . Lysates were separated by 10% SDS-PAGE gel electrophoresis and proteins transferred to PVDF membrane ( Amersham ) . For western blot analysis of adult mouse tissues , tissue samples were collected from 8-week-old adult mice and 20 μg of protein lysate per tissue analyzed per lane . The following antibodies/antisera were used for western blotting: rabbit anti-YTHDC1 affinity-purified antibody ( this study ) ; mouse anti-TUBB antibody ( T4026 , Sigma ) , mouse monoclonal ACTB ( Clone AC-15 , Sigma-Aldrich ) . Immuno-detection was performed using horseradish peroxidase-conjugated secondary antibodies and ECL prime reagents ( Amersham ) according to the manufacturer’s instructions . For immunofluorescence , oocyte , egg or embryo samples were fixed in 2 . 5% paraformaldehyde for 40 min at room temperature . Cells were permeabilized for 15 min in PBS containing 0 . 2% Triton X-100 , blocked in PBS containing 0 . 2% IgG-free BSA and 0 . 01% Tween-20 for 30 min ( blocking solution ) , and then incubated with the rabbit anti-YTHDC1 affinity-purified antibody for 1 h at room temperature . After four 15-min washes in blocking buffer , samples were incubated for 1 h with appropriate Cy5-conjugated secondary antibody ( Jackson ImmunoResearch ) . After three additional 15-min washes in blocking buffer , the samples were mounted in Vectashield mounting solution with Sytox green ( Vector Laboratories ) . Images were captured by a Leica TCS SP laser-scanning confocal microscope . Immunofluorescence analysis in testis was performed as previously described [82] . Briefly , adult or neonatal testes were fixed in 4% formaldehyde for 3–4 h and processed for sectioning in a cryostat . Testicular sections were immunostained with anti-YTHDC1 and anti-SP10 antibodies [83] . FITC- or Texas red-conjugated secondary antibodies were used . Slides were mounted in VectaShield solution with DAPI ( Vector Laboratories ) . Images were captured with an ORCA digital camera ( Hamamatsu Photonics ) on a Leica DM5500B microscope . Oocytes were collected from ovaries of 6-week-old wild-type or Ythdc1fl/- Ddx4-Cre females by needle poking . Oocytes from Ythdc1fl/- Ddx4-Cre females with gross abnormal morphology were excluded from studies . Total RNA was extracted from 25 oocytes per library using PicoPure RNA isolation kit with on-column genomic DNA digestion according to the manufacturer’s instruction ( Thermo Fisher Scientific ) . As a normalization control , each sample was spiked in with 0 . 2 pg synthesized Renilla luciferase mRNA before extraction . RNA-seq libraries were constructed by using Ovation RNA-seq system V2 ( NuGEN ) followed by Ovation Ultralow Library system ( DR Multiplex System , NuGEN ) . Reverse transcription of total RNA was primed with a pool of primers that hybridize either to the 5’ portion of the poly ( A ) sequence or randomly across the transcript . Per genotype , three biological replicate libraries were constructed . RNA-seq libraries were pooled and sequenced by three 150-bp paired-end runs on mid-output flow cells on the NextSeq 550 system ( Illumina ) ( S1 Table ) . RNA-seq data are available under the NCBI/SRA number: SRP116737 . Oocyte RNA-seq data were mapped using the RNA-Seq aligner STAR . The STAR genome was generated using the mouse mm10 genome assembly ( Genome Reference GRCm38 ) . STAR was run with the parameter—clip3pAdapterSeq GATCGGAAGAGCACACGTCTGAACTCCAGTCAC . The SAM files from STAR were converted to BAM format using samtools view , reads were sorted by name using samtools sort , and separate lanes were merged into one file using samtools merge [84] . The number of reads in each genomic feature was quantified with HTSeq using the intersection-strict overlap setting [85] . Differential abundance between Ythdc1fl/- and Ythdc1-deficient oocytes was then analyzed using the R package DESeq2 on the HTSeq count files with default settings and an FDR cutoff of 0 . 01 [86] . Gene Ontology analysis was performed using the bioinformatics analysis resource database DAVID 6 . 8 [87] . Separate lists of differentially expressed up-regulated and down-regulated genes ( all , FDR < 0 . 01 , Fold change ≥ 2 , and mean expression ≥100 ) were uploaded , with the Genbank_accession identifier selected and mus_musculus as the specified organism . A custom background list was supplied consisting of all genes with at least one read observed in any genotype or replicate in our RNA-seq libraries . Analysis of units of LSVs between Ythdc1fl/- and Ythdc1-deficient oocytes was performed using the MAJIQ software package [63] . MAJIQ v0 . 9 . 2 was run using the GRCm38 mm10 reference genome . Default settings were used for quantifying LSVs in the oocyte RNA-seq data and for the ΔPSI analysis . Please note that MAJIQ was not designed to identify alternative polyadenylation from RNA-seq data [63] . Alternative polyadenylation ( APA ) analysis was performed with the ROAR Bioconductor package in R [67] . The package’s general workflow was followed . GV oocyte RNA-seq reads were mapped to the mm9 ( NCBI37 ) genome using the RNA STAR aligner . ROAR was run using an annotation database of polyadenylation sites from the PolyADB version 2 [88] . The ratio of shorter to longer isoforms referred to as m/M ratio was computed for each sample using the counts of mapped reads and the lengths of the transcript’s PRE and POST portions as defined using a multiple APA annotation file . The ratio of the m/M ( WT ) to the m/M ( KO ) yielded the ratio of a ratio ( ROAR ) values , which were used to identify shifts in polyadenylation site usage . 3’ UTR lengthening or shortening was called when a Fisher test for all sample pairings returned nominal p-values < 0 . 05 . We analyzed the RNA-seq data from the SRSF3 and SRSF7 knockdown experiments by Muller-McNicoll et al . [73] using the ROAR algorithm and reached the same conclusions on opposing changes on 3’ UTR length , validating the ROAR algorithm . Oocytes were collected from 6-week-old wild-type or Ythdc1fl/- Ddx4-Cre or Ythdc1fl/- Zp3-Cre females by needle poking of ovaries . Total RNA was extracted from oocytes using the PicoPure RNA Isolation Kit with on-column genomic DNA digestion ( Thermo Fisher Scientific ) and reverse transcribed by Superscript II reverse transcriptase ( Invitrogen ) using random hexamers . The resulting cDNA was quantified by real-time PCR on an ABI Prism 7000 thermocycler ( Applied Biosystems ) using Power SYBR Green Master Mix ( Thermo Fisher Scientific ) . The following gene transcripts were tested: Psat1 , Grhl3 , Cnnm1 , Nupr , Zfp711 , Lrp1b , Rgn , Trps1 , Tnip3 , Piwil1 ( Oligonucleotide primer sequences in S5 Table ) . PCR parameters: 95˚C , 15 sec; 60˚C , 60 sec; 40 cycles . Each sample was analyzed in duplicates . Quantification was normalized to the endogenous Actb using the comparative Ct method ( ABI Prism 7700 Sequence Detection System , Applied Biosystems ) . For validation of local splicing variants , each PCR assay was optimized individually ( S5 Table ) . The PCR cycles varied for each LSV , depending on transcript abundance . For APA validation , two pairs of PCR primers were designed: PRE and POST ( before and after the polyadenylation site ) ( S5 Table ) . All assays used an amount of cDNA equivalent to one oocyte per PCR reaction . RT-PCR band quantification was performed using ImageJ . Testes and ovaries were prepared for histological analysis by fixation in Bouin’s solution ( Sigma Aldrich ) , followed by serial dehydration and paraffin infiltration and embedding . Serial sections were cut at 8 μm thickness , adhered to glass slides , and dried overnight . Slides were de-paraffinized with xylene and re-hydrated . Slides were then stained with hematoxylin , rinsed , and exposed to 0 . 1 M ammonia before staining with eosin . The slides were then dehydrated and mounted with Permount mounting media ( Fisher Scientific ) . Images were taken on a DM5500B microscopy platform with a DFC450 camera ( Leica Microsystems ) . Wild-type mouse Ythdc1 coding region was amplified from bulk mouse testis cDNA samples by PCR . The double mutation ( W377A , W428A ) was introduced by PCR-based mutagenesis by mutating codons 377 ( TGG ) 428 ( TGG ) to 377 ( GCG ) 428 ( GCG ) , resulting in W377A W428A amino acid changes . The entire coding region was cloned into the pcDNA3 . 1 vector for in vitro transcription . Plasmids pcDNA3 . 1-wt-Ythdc1 and pcDNA3 . 1-Ythdc1- W377A W428A were verified by sequencing and linearized before in vitro transcription . Capped mRNAs were made by in vitro transcription with T7 mScript mRNA production System ( CellSCRIPT ) according to the manufacturer’s instructions . Following in vitro transcription , template DNAs were digested by adding RNase-free DNase , and synthesized mRNA was purified by MEGAclear kit ( Ambion ) . A single mRNA band of the expected size was observed for each RNA sample on a 1% formaldehyde denaturing gel . Synthesized RNA was stored in aliquots at -80°C . GV oocytes were collected from postnatal day 12 Ythdc1-deficient ovaries by enzymatic digestion ( collagenase ) . Oocytes were microinjected with approximately 5 pl of wild-type or mutant Ythdc1 mRNA in water as previously described [89] . PND12 wild-type oocytes were mock injected with water as controls . Following microinjection , oocytes were returned to CZB medium with 2 . 5 μM milrinone and cultured overnight , followed by RNA extraction and reverse transcription . For LSV rescue experiments , each PCR assay was performed using an amount of cDNA equivalent to 0 . 5 oocyte . PCR bands were quantified using the Image J software . The rescue experiments were performed two times . The FLAG-YTHDC1 expression construct was made by cloning the full-length mouse Ythdc1 coding sequence into a pcDNA6 vector containing a previously inserted 3xFLAG sequence 5’ of the cloning site . V5 tagged constructs for Srsf7 , Cstf1 , Cstf2 , Cpsf6 , and Nudt21 were all produced by subcloning RT-PCR products amplified from bulk mouse testis cDNAs using the pcDNA 3 . 1/V5-His TOPO TA Expression Kit ( Invitrogen , K4800 ) . Srsf3 cDNA was cloned into a pcDNA6 vector containing a Myc tag 5’ of the cloning site to express Myc-SRSF3 . The HEK 293T cells were cultured in Dulbecco’s Modified Eagle Medium ( DMEM ) supplemented with 10% fetal bovine serum ( FBS ) , 1% penicillin/streptomycin , and 1% L-glutamine in a 37°C humidified incubator at 5% CO2 . Cells were co-transfected using Lipofectamine 2000 ( Invitrogen ) and cultured in Opti-mem media for 36 h . Transfected cells were harvested with RIPA buffer ( 10 mM Tris-HCl , pH 8 , 1 mM EDTA , 0 . 5 mM EGTA , 1% Triton X-100 , 0 . 1% SDS , 140 mM NaCl , 0 . 1% sodium deoxycholate ) supplemented with 1 mM phenylmethylsulfonyl fluoride ( PMSF ) . Cells were lysed by Dounce homogenization and solubilized by rocking at 4°C for 30 min . Following centrifugation at 16 , 100 x g for 25 min , lysate supernatants were incubated with 1 mg/ml RNase A at room temperature for 30 min . Lysates were cleared by centrifugation at 16 , 100 x g for 20 min and incubated with Protein G agarose beads ( Invitrogen , 15920010 ) for 1 h . After pre-clearing , either anti-FLAG ( F-3165 , Sigma ) , anti-V5 ( R96025 , ThermoFisher ) , or anti-Myc ( 631206 , Clontech ) was incubated with cell lysates rotating overnight at 4°C . Equilibrated Protein G agarose beads were added to the lysates and incubated for 1 h . Immunoprecipitated protein complexes were washed three times with RIPA buffer supplemented with PMSF . Beads and respective input lysates were boiled with 2x SDS sample buffer for 5 min prior to SDS-PAGE and immunoblotting to nitrocellulose membranes . | The N6-methyladenosine ( m6A ) modification , one type of RNA methylation , is the most abundant internal RNA modification in eukaryote messenger RNAs . m6A is specifically recognized by RNA-binding reader proteins . Here we report an essential role of the nuclear m6A reader , YTHDC1 , in embryo development and fertility . In particular , YTHDC1 is required for oocyte growth and maturation . YTHDC1-deficient oocytes exhibit massive defects in alternative splicing , which can be rescued by introducing into mutant oocytes wild-type , but not m6A-binding-deficient , YTHDC1 . Strikingly , loss of YTHDC1 causes extensive alternative polyadenylation in oocytes , resulting in altered 3’ UTR length . YTHDC1 interacts with the pre-mRNA 3’end processing factors CPSF6 , SRSF3 , and SRSF7 . Thus , YTHDC1 is a key nuclear factor in the processing of pre-mRNA transcripts . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"reproductive",
"system",
"3'",
"utr",
"messenger",
"rna",
"germ",
"cells",
"alternative",
"splicing",
"developmental",
"biology",
"oocytes",
"untranslated",
"regions",
"molecular",
"biology",
"techniques",
"embryos",
"research",
"and",
"analysis",
"methods",
"embryology",
"artificial",
"gene",
"amplification",
"and",
"extension",
"animal",
"cells",
"gene",
"expression",
"molecular",
"biology",
"biochemistry",
"rna",
"ovaries",
"polyadenylation",
"cell",
"biology",
"ova",
"nucleic",
"acids",
"rna",
"processing",
"anatomy",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"polymerase",
"chain",
"reaction"
]
| 2018 | Nuclear m6A reader YTHDC1 regulates alternative polyadenylation and splicing during mouse oocyte development |
The balance of global integration and functional specialization is a critical feature of efficient brain networks , but the relationship of global topology , local node dynamics and information flow across networks has yet to be identified . One critical step in elucidating this relationship is the identification of governing principles underlying the directionality of interactions between nodes . Here , we demonstrate such principles through analytical solutions based on the phase lead/lag relationships of general oscillator models in networks . We confirm analytical results with computational simulations using general model networks and anatomical brain networks , as well as high-density electroencephalography collected from humans in the conscious and anesthetized states . Analytical , computational , and empirical results demonstrate that network nodes with more connections ( i . e . , higher degrees ) have larger amplitudes and are directional targets ( phase lag ) rather than sources ( phase lead ) . The relationship of node degree and directionality therefore appears to be a fundamental property of networks , with direct applicability to brain function . These results provide a foundation for a principled understanding of information transfer across networks and also demonstrate that changes in directionality patterns across states of human consciousness are driven by alterations of brain network topology .
Current large-scale initiatives are attempting to construct a map of the structural and functional network connections in the brain [1 , 2] . One critical goal of these initiatives is to understand the mechanism by which local and functionally specialized neural activity becomes globally integrated to achieve efficient brain function [3–5] . Neural oscillations may represent one mechanism of what is sometimes referred to as “information flow” between segregated neural nodes [6–9] . However , in order to understand the principles of information transfer across networks , the mechanisms of directionality between the oscillations of interacting nodes need to be elucidated . There have been a number of computational studies on the relationship of network structures , local dynamics , and directional connectivity [10–13] . More recently , a causal relationship between global brain network topology and the dynamics of corticocortical interactions has been postulated [14 , 15] . Emerging empirical data and computational models suggest that the relative location of neuronal populations in large-scale brain networks might shape the neural dynamics and the directional interactions between nodes , which implies a significant influence of global topology on local dynamics and information flow [16–21] . For example , a study analyzing the electroencephalogram ( EEG ) recorded from human volunteers demonstrated that if a brain region is topologically more accessible to other brain regions , then it has a larger variability in its local activity [16] . As another example , a magnetoencephalogram ( MEG ) study showed that variability in the MEG sources determines the direction of information flow between local brain regions [17 , 18] . These studies provide empirical evidence of a direct influence of brain network topology on variability of local brain activity and directionality in brain networks . In addition , computational models and simulation studies of global brain networks have revealed that hub nodes ( i . e . , nodes with extensive connections ) have a significant influence on the local node dynamics and the direction of information flow in normal and pathological brains [19–21] . For example , Stam et al . showed in a model that the phase lead/lag relationship between local node dynamics is correlated with the degree of the node [19] . However , these past studies all describe special cases without analytical or direct empirical support; a general mechanism that links global network topology , local node dynamics and information flow has yet to be identified . In the current study we address an important prerequisite to understanding this general mechanism by identifying the relationship of topology , local dynamics and directionality . The directionality of interactions between nodes was studied through the modulated phase lead/lag relationship of coupled oscillators in general network models , large-scale anatomical brain network models and empirically-reconstructed networks from high-density human EEG across different states of consciousness ( Fig 1 ) . Analytical , computational and empirical results demonstrate definitively that the node degree ( i . e . , the number of connections to other nodes ) defines both the directionality between local node dynamics and the amplitude of the oscillations at that node . Importantly , the directionality is shown to result from inhomogeneous interactions of local dynamics and can be differentiated from the conventional observation of directed physical connections .
The central purpose of this study was to identify a general relationship of network topology , local node dynamics and directionality in inhomogeneous networks . We proceeded by constructing a simple coupled oscillatory network model , using a Stuart-Landau model oscillator to represent the neural mass population activity at each node of the network ( see Materials and Methods , and S1 Text for details ) . The Stuart-Landau model is the normal form of the Hopf bifurcation , which means that it is the simplest model capturing the essential features of the system near the bifurcation point [22–25] . The Hopf bifurcation appears widely in biological and chemical systems [24–33] and is often used to study oscillatory behavior and brain dynamics [25 , 27 , 29 , 33–36] . We first ran 78 coupled Stuart-Landau models on a scale-free model network [37 , 38]—that is , a network with a degree distribution following a power law—where coupling strength S between nodes can be varied as the control parameter . The natural frequency of each node was randomly drawn from a Gaussian distribution with the mean at 10 Hz and standard deviation of 1 Hz , simulating the alpha bandwidth ( 8-13Hz ) of human EEG , and we systematically varied the coupling strength S from 0 to 50 . We also varied the time delay parameter across a broad range ( 2~50ms ) , but this did not yield a qualitative difference in the simulation results as long as the delay was less than a quarter cycle ( < 25 ms ) of the given natural frequency ( in this case , one cycle is about 100 ms since the frequency is around 10Hz ) . The simulation was run 1000 times for each parameter set . Subsequently , the directionality between all local node dynamics was measured using the directed phase lag index ( dPLI ) , which calculates the phase lead and lag relationship between two oscillators ( see Materials and Methods for detailed definition ) [19] . dPLI between two nodes a and b , dPLIab , can be interpreted as the time average of the sign of phase difference ϕa*−ϕb* . It will yield a positive/negative value if a is phase leading/lagging b , respectively . dPLI was used as a surrogate measure for directionality between coupled oscillators [19] . Without any initial bias , if one node leads/lags in phase and therefore has a higher/lower dPLI value than another node , the biased phases reflect the directionality of interaction of coupled local dynamics . dPLI was chosen as the measure of analysis because its simplicity facilitated the analytic derivation of the relationship between topology and directionality . However , we note that we also reach qualitatively similar conclusions with our analysis of other frequently-used measures such as Granger causality ( GC ) and symbolic transfer entropy ( STE ) ( see S1 Text and S1 Fig for the comparison ) [39–41] . Fig 2A–2C demonstrates how the network topology is related to the amplitude and phase of local oscillators . Fig 2A shows the mean phase coherence ( measure of how synchronized the oscillators are; see Materials and Methods for details ) [42] for two groups of nodes in the network: 1 ) hub nodes , here defined as nodes with a degree above the group standard deviation ( green triangles , 8 out of 78 nodes ) ; and 2 ) peripheral nodes , here defined as nodes with a degree of 1 ( yellow circles , 33 out of 78 nodes ) . When the coupling strength S is large enough , we observed distinct patterns for each group . For example , at the coupling strength of S = 1 . 5 , which represents a state in between the extremes of a fully desynchronized and a fully synchronized network ( with the coherence value in the vicinity of 0 . 5 ) , the amplitudes of node activity are separated into two groups—hub nodes , with larger amplitudes , and peripheral nodes , with smaller amplitudes ( Fig 2B ) . More strikingly , the phase lead/lag relationship is clearly differentiated between the hub and peripheral nodes: hub nodes phase lag with dPLI <0 , while the peripheral nodes phase lead with dPLI >0 ( Fig 2C ) . Fig 3 shows the simulation results in random and scale-free networks , which represent two extreme cases of inhomogeneous degree networks . This figure clearly demonstrates that larger degree nodes lag in phase with dPLI <0 and larger amplitude ( see S2 Fig for various types of networks: scale free , random , hierarchical modular and two different human brain networks ) even at the coupling strength S = 1 . 5 , where the separation of activities between hub nodes and peripheral nodes just begins to emerge . To explain these simulation results , we utilized Ko et al . ’s mean-field technique approach to derive the relationships for the coupled Stuart-Landau oscillators with inhomogeneous coupling strength , which in turn can be applied to inhomogeneous degree networks by interpreting inhomogeneous coupling strength as inhomogeneous degree for each oscillator [43] . We then proceeded to identify the relationships between network topology ( node degree ) , node dynamics ( amplitude ) and directionality between node dynamics ( dPLI ) ( see S1 Text for complete derivation ) . The analytical results demonstrate that , for the Stuart-Landau oscillators with the same natural frequencies and inhomogeneous coupling , when the coupling strength between oscillators is sufficiently high and the delay time given as constant between them is sufficiently small , ( KjR˜ ) 2 can only have a monotonically increasing relationship with respect to ( rj* ) 2 as shown in Fig 2D . Here Kj corresponds to the average coupling strength to oscillator j , and is interpreted as the degree of node j , kj , times the coupling strength S ( Kj≈ kjS ) , and R˜ is the order parameter ( sum of all oscillators: see S1 Text for details ) . Therefore , the following relationship holds: if Ka>Kb , then ra*>rb* . ( 1 ) In other words , nodes with higher degrees naturally have larger amplitudes . The analytic results also demonstrate the following: if ra*>rb* , then tanϕa*-Φ+β<tan ( ϕb*-Φ+β ) ( 2 ) Accordingly , if ( ϕa*−Φ+β ) ∈[−π2 , π2] , then ϕa*<ϕb* , for tan ( x ) is monotonically increasing function of x for x∈[−π2 , π2] . Here ϕa* and ϕb* are the phase of node a and b , respectively , Ф is the average phase across all nodes and β is the time delay . The inequality states that if the amplitude of node a is larger than that of node b , then it follows that the phase of node a is smaller than the phase of node b . Thus , a will phase lag b . Therefore , given two nodes a and b with their degrees ka>kb , our results show that the amplitudes and phases will be ra*>rb* and ϕa*<ϕb* , respectively . By definition , dPLIab ( defined as the time average of the sign of phase difference ϕa*−ϕb* ) will have a negative value . In short , higher-degree nodes have larger amplitude and phase lag ( dPLI<0 ) , while lower-degree nodes have smaller amplitude and phase lead ( dPLI>0 ) . The inequalities for node degree k , amplitude r* and phase ϕ* mathematically explain how the degree of a network node is related to the amplitude and phase of oscillation . We note that we have also repeated the same analysis with the coupled Kuramoto model , which is the canonical model capturing the dynamics of the oscillator network with only a single phase variable for each oscillator [6 , 25 , 33 , 44 , 45] ( see S1 Text for its relationship to more complex models ) , and found it yields the same result: higher degree nodes phase lag with dPLI <0 ( see S1 Text for the analytical derivation and S3 Fig for the simulation result ) . In the next section , our analytic studies for two extreme cases of inhomogeneous networks of Gaussian ( random ) and power-law ( scale-free ) degree distributions will be applied to complex human brain networks . The network topology of the human cortex consists of primary hubs in the posterior-parietal region with most peripheral nodes located in the frontal region [46–48] . We predict that this archetypical topology gives rise to the characteristic amplitude topography and directionality pattern observed in the human brain . To test this hypothesis , we simulated human brain networks for both conscious and unconscious ( i . e . , uncoupled ) states . An anatomical network from diffusion tensor imaging ( DTI ) was used as the underlying network for the model oscillators [47] . Each network node represents one of 78 cortical regions and two nodes were considered connected if the probability of fiber connections exceeded a statistical criterion . The anatomical network has the following properties: 1 ) small-world network , 2 ) scale-free degree distribution with an exponential cut-off , 3 ) higher degree nodes are mostly distributed in the parietal and occipital lobes , whereas the lower degree nodes are located in the frontal lobe . Alpha-band neural oscillations were simulated with 78 coupled Stuart-Landau models on the anatomical network . In order to study the effect of changing the brain network topology , we also perturbed the anatomical network in proportion to the degree of the nodes . Therefore , the hub structures were preferentially disrupted , which is consistent with empirical observations of the behavior of the human brain during anesthetic-induced unconsciousness [49] . In mathematical terms , the preferential disruption of hub nodes is given by multiplying 1/gγ factor to the coupling strength S in eqs ( 2 ) and ( 3 ) ( see Materials and Methods ) . Here g is the degree for each node , and γ is the perturbation strength . Higher values of γ generate stronger perturbations of the node . For γ = 1 , the network becomes homogeneous with the coupling strength S for a node normalized by its degree: S/g . Otherwise , if γ >1 , the coupling term S/gγ will be smaller for a node with high degree producing a larger perturbation effect for such a node . Therefore , an excessive perturbation of γ >>1 will yield an inverse hub-periphery structure . Fig 4A and 4C clearly demonstrate a negative correlation between node degree and dPLI ( Spearman correlation coefficient = - 0 . 61 , p<0 . 01 ) and positive correlation between node degree and amplitude of oscillators ( Spearman correlation coefficient = 0 . 92 , p<0 . 01 ) at coupling strength S = 3 . As predicted , higher degree nodes have higher amplitude and stronger incoming directionality than lower degree nodes ( dPLI<0 ) . Fig 4B and 4D show that after the perturbation ( γ = 1 ) , the correlations among node degree , amplitude and dPLI disappear . The homogenized network does not produce any biases in the directionality and amplitude distribution in the modeled brain . Fig 4E and 4F present the relationship between the node degree of the anatomical brain network and dPLI of the alpha oscillators as a ring plot . In the anatomical brain network , the parietal-occipital regions have the higher node degrees ( presented as dense and dark connections ) . Notably , the left and right precuneus in the parietal region have the highest node degrees ( denoted with red arrows in Fig 4E ) , while the lower node degrees are mostly distributed in the frontal region . The functional network strongly correlates with the anatomical network . Accordingly , the two precuneus regions have the largest negative dPLI values , playing a role as the strongest target of directionality , and the typical overall network topology produces the dominant directionality from frontal region ( as source; red color ( dPLI>0 ) in the ring in Fig 4E ) to the parietal-occipital region ( as sink; blue color ( dPLI<0 ) in the ring in Fig 4E ) . However , after perturbing the heterogeneous human network to a homogeneous functional network topology , the typical patterns in amplitude and directionality are neutralized ( presented as the same green color ( dPLI~0 ) in the ring in Fig 4F ) . In summary , this simulation of normal and perturbed human brain networks clearly demonstrates that the typical topology of the anatomical brain network shapes the spatial distribution of node amplitude and the characteristic directionality patterns . Furthermore , the perturbed network topology with preferential hub disruption produces homogenized patterns in amplitude and directionality across the network . To test whether or not these results depend on the given network , we repeated the same analysis with another human anatomical network , which is based on 66 parcels of the cerebral cortex [46] , and observed qualitatively similar results ( see S4 Fig ) . In order to verify the theoretical predictions of the directionality and amplitude patterns in human brain networks before and after perturbation , we analyzed empirical EEG data collected from human volunteers in states of consciousness ( eyes closed , at rest ) and anesthetic-induced unconsciousness . Since anesthesia primarily disrupts hub structures in the human brain network [49] , we predicted that the directionality toward the hub nodes would be preferentially disrupted , which would manifest in the empirical data as a disruption of front-to-back directionality between primary peripheral nodes in frontal region and primary hub nodes in posterior-parietal regions . 64-channel EEG was recorded continuously from 7 healthy human volunteers during consciousness and sevoflurane-induced unconsciousness; 5-minute artifact-free epochs were analyzed ( see Materials and Methods for the details on the EEG experiment ) . Recorded data were referenced to the vertex . After the experiment , EEG data were re-referenced to an average reference , and data from the vertex channel was calculated , yielding a total of 65 EEG data channels for analysis . Graph theoretic network analysis was applied to construct functional brain networks from the EEG . Phase lag index ( PLI ) , a measure of phase locking between two signals , was calculated between all combinations of EEG channels , and channel pairs constituting the top 30% of PLI values , a threshold at which the results match well with those of model network , were chosen as the functional connections of the network [50] . The directionality was estimated for each channel by calculating the average dPLI between a given channel and each of the remaining 64 EEG channels . Because anesthesia causes a large spectral change in EEG during the transition from consciousness to unconsciousness , we examined 6 frequency bands ( delta: 0 . 5–4Hz , theta: 4–8Hz , alpha: 8–13 Hz , beta: 13–25Hz , gamma: 25–55Hz and the whole band: 0 . 5–55Hz ) and their respective functional networks . Our analysis demonstrated that: ( 1 ) the theoretical predictions made from computational human brain models regarding the relationship between node degree and dPLI are supported by patterns observed in empirical EEG networks recorded from waking and unconscious states ( in Fig 5A and 5B ) ; ( 2 ) The functional brain network of the whole frequency band ( 0 . 5–55Hz ) is highly correlated with the node degree distribution found in the anatomical brain network model . The majority of hub nodes were located in the posterior-parietal region in both the anatomical network and the functional EEG network . In the waking state , the high-degree nodes were mainly distributed in the back part of the brain ( upper row in Fig 5B ) , while in the unconscious state , this pattern was completely disrupted ( upper row in Fig 5B ) ; ( 3 ) The alpha band ( 8–13HZ ) EEG network that has been the focus of our computational simulations demonstrates a dominant front-to-back directionality in the brain during the conscious state ( eyes closed ) , with frontal dPLI > 0 and posterior dPLI < 0 ( the 2nd row in Fig 5B ) [51 , 52] , which was neutralized in the unconscious state . This neutralized directionality in the EEG network supports the results of our simulation in which we preferentially perturbed hub nodes ( the 3rd row in Fig 5A and 5B ) ; ( 4 ) The correlation between node degree ( of the whole band , 0 . 5–55Hz ) and directionality ( of the alpha band , 8–13HZ ) changes significantly across states . The strong negative correlation observed during the conscious state ( Spearman correlation coefficient of -0 . 76 ( p<0 . 01 ) ) disappears during the unconscious state ( Spearman correlation coefficient of -0 . 04 ( p<0 . 01 ) ) . These correlations are consistent with the theoretical predictions from the analytical solution and simulations . However , the correlation between node degree and amplitude for the EEG network differs from the models ( non-significant Spearman correlation coefficient of 0 . 266 ( p = 0 . 1 ) for the conscious state ) . The lack of significance is potentially due to a distortion of the scalp EEG recording as the signals pass through the skull , which may cause a deviation from the model prediction . MEG would be more appropriate to study the correlation of amplitude and node degree in the whole brain network .
In this study , we provide a general relationship for how network topology ( node degree ) determines the directionality ( phase lead/lag relationship ) and local dynamics ( amplitude of oscillator ) using the mean-field approximation . Simple oscillatory models ( Kuramoto/Stuart-Landau models ) were used first to simulate the global network dynamics and to find the mathematical relationship among node degree , local dynamics and directionality ( defined by phase lead/lag ) . We have shown that the directionality arises naturally from the topology of the underlying network . The hub nodes phase lag: they act as a sink that is driven by connected nodes . The non-hub peripheral nodes phase lead: they are sources and drive the connected nodes . This finding may be counterintuitive , as network hubs could be regarded as “control centers” that serve as the source of outflowing information . The present results suggest , by contrast , that hub nodes with high degree may “attract” information from peripheral nodes . The consistently phase-lagging nature of the high-degree hub node may allow for the inputs of spatially and functionally distinct peripheral nodes to converge and be integrated , a critical feature for optimal network function . Network topology also predicts the local dynamics , defined here by the amplitude of an oscillation in the case of the Stuart-Landau model; high degree hub nodes are associated with oscillations of larger amplitude and low degree peripheral nodes are associated with oscillations of smaller amplitude . There have been several important studies exploring the effect of brain network topology on the local and global dynamics of the brain . De Haan et al . simulated normal and diseased brain activities based on a neural mass model of the anatomical network . They found that the hub regions are associated with the highest level of activity and that excessive neuronal activity at the hub may lead to degeneration in Alzheimer’s disease [20] . Stam et al . simulated how network structure affects the phase lead/lag relationship between brain regions in a realistic brain network model [19] . Nicosia et al . showed in a network model that if two nodes are symmetrically located within a given network topology , the dynamics of the nodes will be fully synchronized even at a significant distance [53] . Angelini et al . measured Granger causality for the Kuramoto model on networks and demonstrated that inflow/outflow ratio changes depend on the degree of each node [54] . However , despite these recent empirical and computational model studies , there has been no general explanatory mechanism linking global topology , local node dynamics and directionality between interacting nodes based on mathematical derivation . The strength of our analysis lies in its simplicity and generality . The models we employed are simple enough to analyze extensively yet succeed in capturing the essential features of dynamic behavior of the network related to the emergence of directionality . More complex models are difficult to analyze due to the abundance of equations and parameters , rendering analytic solutions difficult except for very special cases . The models used in this study are rich enough in their behavior yet simple enough to analyze and analytically calculate . Another advantage is the generality of the models: they are representative of many other oscillating systems so that the results from these models will be widely applicable . Furthermore , the analytical results are independent of the type of network , as long as the network is inhomogeneous in terms of connections . Expressing the central relationship quantitatively , when coupling strength S between oscillators is sufficiently weak , any system of interacting oscillators can be considered to interact only with its phases , and the Kuramoto model is the first-order approximation for such phase-only interacting oscillators . When the coupling term is stronger so that the amplitude equations must be considered , the Stuart-Landau model equation holds its generality because it is the normal form of the Hopf bifurcation . The Hopf bifurcation is one of the most frequently appearing mechanisms in models generating oscillatory behavior , as in the case of the Wilson-Cowan model , the Fitzhugh-Nagumo model and the Morris-Lecar model , among other numerous examples . One can gain general insights about the behavior of more complex interacting oscillator models by analyzing such generalized models . Assertions regarding the applicability of findings derived from these simple models are substantiated by a number of successful predictions . First , we simulated the oscillator models in a human anatomical network and demonstrated that anterior-to-posterior directionality arises due to a network structure in which posterior regions contain more hub nodes than anterior regions . This simulated result was confirmed with empirically-reconstructed human brain networks derived from high-density EEG recordings , demonstrating again that the anterior-to-posterior directionality occurs because of the posterior-hub structure . When this hub structure is perturbed , the directionality was eliminated in the model on the simulated neuroanatomical network . When consciousness was lost after administration of the anesthetic sevoflurane in human volunteers , anterior-to-posterior directionality was similarly eliminated with the disappearance of the posterior-hub structure . Application of this principle could have relevance to clinical conditions in which hub structure may be damaged or dramatically reorganized . Altered information flow has been reported in network-altering conditions such as Alzheimer’s disease , schizophrenia , and epilepsy [5 , 20 , 55 , 56] . Our findings not only explain why information flow changes across different brain states , but also could ultimately contribute to treating such disorders by modulating the directionality of node interactions using brain stimulation techniques . There are a number of limitations to this study . The first relates to the relationship of phase lead/lag measures and information flow . Although it can be asserted that causal events lead and resultant effects lag ( simply by virtue of the temporal constraints on cause-effect relationships ) , the converse assertion that every lead/lag relationship reflects a causal influence does not hold . In other words , an appropriate phase lead/lag relationship is a necessary but not sufficient condition for the kinds of interactions that are associated with information transfer . As such , we have conducted parallel analyses with other measures ( GC and STE ) based on distinct theoretical frameworks ( linear regression and information theory , respectively ) ( S1 Fig ) . These metrics were found to parallel the measure of dPLI , thus supporting the general interpretation that our studies of directionality may , indeed , provide a foundation for future studies related to information transfer in the brain . The second limitation is that our analysis focused primarily on the 10 Hz oscillation , but this reflects our choice to investigate brain networks . The analysis can easily be applied to other frequency bandwidths , yielding similar results as long as the time delay is sufficiently small compared to the time of one oscillation . Third , our results do not reflect short-term stationary brain network behavior such as metastability [57] . Furthermore , in a long-term time scale , the brain network structure itself will change via mutual influences between network topology and local dynamics as the brain matures [58] . The time scale of our study lies in between these two extreme limits , where the functional connectivity can reflect underlying structural connectivity yet the effect of local dynamics on the network structure can be disregarded . Fourth , the Kuramoto and Stuart-Landau models are the normal forms of complex oscillator models . Thus , the results of the coupled oscillator networks—as well as the data from our EEG experiments—describe large-scale temporal and spatial behavior , i . e . , network dynamics that are relatively long-term and macroscopic . As such , our simple models and the analytical results may not explain fine-scale neuronal firing relationships and the short-term dynamics of complex local connections such as the influence of a common source with different time delays . Further work is warranted to test whether the current findings apply to finer-scale dynamics . Fifth , in the empirical data test , we analyzed a functional brain network reconstructed from scalp EEG , which reflects the anatomical brain network with less spatial fidelity than the simulated network [13 , 59] . Therefore , instead of examining the one-to-one correspondence between the functional networks of the empirical data and of the model , we investigated the correlation patterns among node degree , amplitude and dPLI in the EEG network and the model network . Finally , we used a simple exponential function to achieve a preferential disruption of hub structure in the simulation of anesthetic effects on the brain network . The study of more realistic perturbation functions would be an interesting future investigation to simulate diverse anesthetic effects in the brain . In conclusion , the topological property of node degree determines local dynamics such as the amplitude of an oscillation , as well as directionality between interacting nodes . This relationship , derived from simple oscillator models , was applied successfully to complex brain network models generated computationally or reconstructed empirically . The high-degree/high-inflow relationship predicted the behavior of human brain networks across multiple states of consciousness . These findings may provide clarity to future studies of information transfer as the complexity of the human brain connectome becomes more fully elucidated . Furthermore , the analytical mechanism provided and general relationships identified have the potential to advance network science across numerous disciplines .
In order to study the general relationships among topology , node amplitude and directionality between interacting nodes in a network , we used a simple oscillatory model , the Stuart-Landau model . The Stuart-Landau model is defined as the following: z˙jt = λj+iωj-zjt2zjt+S∑k = 1NKjkzk ( t-τjk ) , j = 1 , 2 , … , N . ( 3 ) Here , zj ( t ) is the complex variable describing the state of jth oscillator . The eq ( 3 ) can be separated into two variables: r˙jt = λj-rjt2rjt+S∑k = 1NKjkrkcosθkt-τjk-θjt , ( 4 ) θ˙jt = ωj+S∑k = 1NKjkrkrjsinθkt-τjk-θjt , j = 1 , 2 , … , N . ( 5 ) rj ( t ) is the amplitude of the signal the oscillator j produces at time t . λj ( t ) is a parameter governing the amplitude and we set all λj ( t ) for j = 1 , 2 , … , N equal for our simulations so that the differences in the amplitude between oscillators can only come from the coupling term in each equation . Also , we note that when all the amplitudes are set equal to each other and do not change , the eqs ( 4 ) and ( 5 ) reduce to the phase-only equation , which is the Kuramoto model . In this respect the Stuart-Landau model can be considered as the generalized model of the Kuramoto model , with the inclusion of the amplitude equation . Descriptions of the Kuramoto model itself , the relationship between the Kuramoto model , the Stuart-Landau model and more complex neural mass models , and the derivation from Wilson-Cowan model to Stuart-Landau/Kuramoto models are included in the S1 Text . For the functional connection in the network , we use two types of phase coherence measures; ( 1 ) mean phase coherence ( PC ) and ( 2 ) phase lag index ( PLI ) . PC is a measure of mean phase synchronization , which can be directly calculated from the model oscillators’ phases . On the contrary , PLI measures nonzero phase lead/lag relationships , which mitigates the effects of choice of reference and of volume conduction in EEG analysis . The mean phase coherence between two oscillators j and k in a network is defined as: PCjk = 1T∑t = 1TeiΔθjkt , ( 6 ) where Δθjk ( t ) is the phase difference . For complete phase synchronization , PCjk has 1 , and 0 for completely desynchronized case [42] . For each node j , we can calculate PCj as the averaged value of PCjk for all other nodes k . Such averaged mean phase coherence for each hub/non-hub node with respect to all other nodes in the coupled Stuart-Landau oscillator network is demonstrated in Fig 2A . PLI was used to define the functional connectivity in the EEG network [50] . We use a Hilbert transform to extract the instantaneous phase of the electroencephalogram from each channel and calculate the phase difference Δθjk ( t ) between channels i and j , where Δθij ( t ) = θi ( t ) -θj ( t ) , t = 1 , 2 , …n , n is the number of samples within one epoch . PLIij between two nodes i and j is then calculated using eq ( 7 ) : <Display_Math>PLIij = signΔθij ( t ) , 0 ≤ PLIij ≤1 . ( 7 ) Here , the sign ( ) function yields: 1 if Δθij ( t ) >0; 0 if Δθij ( t ) = 0; and -1 if Δθij ( t ) <0 . The mean < > is taken over all t = 1 , 2 , … , n . If the instantaneous phase of one signal is consistently ahead of the other signal , the phases are considered locked , and PLIij ≈ 1 . However , if the signals randomly alternate between a phase lead and a phase lag relationship , there is no phase locking and PLIij ≈ 0 . To determine the phase-lead/phase-lag relationship between channels , we calculate dPLI between nodes i and j using eq ( 8 ) [19]: dPLIij = signΔθij ( t ) , -1 ≤ dPLIij ≤1 . ( 8 ) Here , again the sign ( ) function yields: 1 if Δθij ( t ) >0; 0 if Δθij ( t ) = 0; and -1 if Δθij ( t ) <0 . The mean < > is again taken over all t = 1 , 2 , … , n . Therefore , if on average , node i leads node j , 0< dPLIij ≤1; if node j leads node i , -1≤ dPLIij <0; and if there is no phase-lead/phase-lag relationship between nodes , dPLI = 0 . In this study , dPLIi for a node i can be defined as the average of dPLIij for all other nodes j . For the purpose of brevity , each time we denote dPLI of a node i in the Results section , we are referring to dPLIi for the node i . All the parameters for the models are set accordingly to simulate alpha oscillations in the brain . For both models , the natural frequencies of the oscillators in our simulation are given as a Gaussian distribution to simulate alpha with mean at 10 Hz and standard deviation 1 , making ωj around 10∙2π rad/s . Time delay is ( a ) given an identical value between 2ms and 50 ms for all edges ( for model networks as well as Gong et al . ’s and Hagmann et al . ’s human brain networks ) , or ( b ) given proportional to the physical distances for each edges with propagation speed of between 5 to 10m/s ( for Gong et al . ’s human brain network ) [60 , 61] . In the simulation , however , the difference in the propagation speed or time delay does not provide any qualitative differences in the results , as long as the resulting time delay is less than the time of a quarter cycle for the natural frequency ( in the simulation , the time for one cycle is 100 ms for the given frequency of 10 Hz , thus it is less than 25 ms ) . The coupling strength between the oscillators is increased from 0 to 50 . For the Stuart-Landau model , the amplitude parameter λj is given identically for all oscillators with a value of 2 . For all simulations , we also added a Gaussian white noise ξj ( t ) of vanishing mean and standard deviation of 2 to each oscillator's equation to test the robustness of our results against random fluctuations . With each model , we produce a times series of length 10 , 000 for each run of the simulation , and then take the latter half of the time series for the measurement . The sampling rate of the time series is 1 , 000Hz , making the length of the produced time series 10s containing approximately 100 cycles of oscillation . For a given parameter set , measurement is averaged over at least 1 , 000 runs of the simulation . For the simulations on the random networks and the scale-free networks , a new network is generated for each run . To test the role of hub structure on the node amplitude and directionality between interacting nodes , we perturbed the topology of human brain network by preferentially disrupting hub structures . The perturbation factor 1/gγ is multiplied to the coupling strength S in eqs ( 4 ) and ( 5 ) : r˙jt = λj-rjt2rjt+Sgγ∑k = 1NKjkrkcosθkt-τjk-θjt , ( 9 ) θ˙jt = ωj+Sgγ∑k = 1NKjkrkrjsinθkt-τjk-θjt , j = 1 , 2 , … , N . ( 10 ) Here g is the degree for each node and γ is the perturbation factor . By multiplying 1/gγ , the effective coupling strength S/gγ depends on the degree of each node . Thus , the higher the value γ is , the stronger the perturbation of the hub . For γ = 1 , the coupling term in each equation is normalized with respect to the degree of the node , thus the network topology become homogeneous . Consequently , it does not provide any asymmetric dynamics between the hub and peripheral nodes . If γ >1 , the original hub nodes are excessively perturbed such that the original hub-periphery relations are reversed . Oscillator models were run over both random and scale-free networks with size 78 , 100 , and 1000 , respectively . We used the Gilbert algorithm for producing a random network with the parameter of G ( N , ( 1+ε ) log ( N ) /N ) , where N is the number of nodes , and ε is an arbitrary small number , such that the resulting network is connected . We use Catanzaro et al . ’s algorithm to make a randomly connected network with scale-free node degree distribution given a priori [62] . The slope of the degree distribution was set to -2 . 2 . The size of the network does not result in qualitative differences . The human brain network was constructed from diffusion tensor imaging ( DTI ) of 80 young adults [47] . The network is consisted of 78 parcels of the cerebral cortex . Another human brain network by Hagmann et al . [46] , which is based on 66 parcels of the cerebral cortex , was used for the simulation , with qualitatively similar results . The node degree , amplitude and dPLI for each node were calculated in EEG networks constructed from a 64-channel EEG dataset . First , each 5 min epoch of EEG data for both states ( waking and anesthetic-induced unconsciousness ) was segmented into 10 sec epochs for pseudo-stationary state . The node degree , amplitude and dPLI for individual are the averages over all the segmented data . For each segmented dataset , the band pass filter was applied for the six frequency bands . Band-pass filtering with the fifth-order Butterworth filter was applied to EEG forward and backward , correcting the potential phase shifting after band-pass filtering ( “butterworth . m” , and “filtfilt . m” in Matlab; MathWorks , Natick , MA ) . For each frequency band , the PLI was calculated for all pairs of EEG channels and the adjacency matrix was constructed with the top 30% of PLI connections through searching for the best-fit to the simulation and robust threshold . Node degree for each channel was computed from the binary network , which counts the number of links connected to a node . The amplitude was calculated from mean power spectrum density . For power spectrum density , a Hamming window and a modified periodogram were used for each 10 sec EEG segment ( in “pwelch . m” , in Matlab ) . dPLI for a channel was computed with averaged dPLI between the given channel and the other all EEG channels . Consequently , for a 5 min long EEG epoch , we can have the node degrees , amplitudes , and dPLIs for all 64 EEG channels . The spearman correlation coefficient was used for evaluating the correlations among node degree , amplitude and dPLI of the 64 channels ( “corr . m” in Matlab ) . The results from S1 Text can be summarized as follows: for Kuramoto oscillators and Stuart-Landau oscillators with inhomogeneous coupling strength between them , the oscillators with larger average coupling strength phase lag behind those with smaller average coupling strength , given the same natural frequencies , small enough constant time delays and sufficiently strong coupling strengths between them . For Stuart-Landau oscillators , we also show that the oscillators with larger average coupling strength have larger amplitude oscillations . We utilized Ko et al . ’s mean-field technique to derive these results , and applied them to inhomogeneous degree networks as an approximation: the inhomogeneous coupling strength of each oscillator was interpreted as the inhomogeneous degree of each oscillator [43] . For simulations , we expanded our conditions further: we used a Gaussian distribution for natural frequencies of the oscillators and distance-varying time delays between the oscillators for Gong’s anatomical network . We also added a Gaussian-noise to each oscillator’s equation to test the robustness . The simulation results confirmed that the central relationship of degree , node dynamics and directionality ( i . e . , higher degree nodes have larger amplitudes and phase lag behind lower degree nodes ) still holds firmly . | Current brain connectome projects are attempting to construct a map of the structural and functional network connections in the brain . One goal of these projects is to understand how network organization determines local functions and information transfer patterns , which is essential to achieve higher cognitive brain functions . Because of the limitation of constructing all brain maps for all cognitive states , finding a general relationship of global topology , local dynamics and the directionality of information transfer in a network is crucial . In this study , we show that inter-node directionality arises naturally from the topology of the network . Analytical , computational , and empirical results all demonstrate that network nodes with more connections ( i . e . , higher degree ) lag in phase , while lower-degree nodes lead . Our mathematical analysis allowed us to predict the directionality patterns in general model networks as well as human brain networks across different states of consciousness . These findings may provide more straightforward approaches to dissecting how directionality between interacting nodes is shaped in complex brain networks , providing a foundation for understanding principles of information transfer . Furthermore , the underlying mathematical relationship between node connections and directionality patterns has the potential to advance network science across numerous disciplines . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | General Relationship of Global Topology, Local Dynamics, and Directionality in Large-Scale Brain Networks |
IAPP , a 37 amino-acid peptide hormone belonging to the calcitonin family , is an intrinsically disordered protein that is coexpressed and cosecreted along with insulin by pancreatic islet β-cells in response to meals . IAPP plays a physiological role in glucose regulation; however , in certain species , IAPP can aggregate and this process is linked to β-cell death and Type II Diabetes . Using replica exchange molecular dynamics with extensive sampling ( 16 replicas per sequence and 600 ns per replica ) , we investigate the structure of the monomeric state of two species of aggregating peptides ( human and cat IAPP ) and two species of non-aggregating peptides ( pig and rat IAPP ) . Our simulations reveal that the pig and rat conformations are very similar , and consist of helix-coil and helix-hairpin conformations . The aggregating sequences , on the other hand , populate the same helix-coil and helix-hairpin conformations as the non-aggregating sequence , but , in addition , populate a hairpin structure . Our exhaustive simulations , coupled with available peptide-activity data , leads us to a structure-activity relationship ( SAR ) in which we propose that the functional role of IAPP is carried out by the helix-coil conformation , a structure common to both aggregating and non-aggregating species . The pathological role of this peptide may have multiple origins , including the interaction of the helical elements with membranes . Nonetheless , our simulations suggest that the hairpin structure , only observed in the aggregating species , might be linked to the pathological role of this peptide , either as a direct precursor to amyloid fibrils , or as part of a cylindrin type of toxic oligomer . We further propose that the helix-hairpin fold is also a possible aggregation prone conformation that would lead normally non-aggregating variants of IAPP to form fibrils under conditions where an external perturbation is applied . The SAR relationship is used to suggest the rational design of therapeutics for treating diabetes .
The Islet Amyloid Polypeptide/IAPP ( also known as amylin ) is coexpressed and cosecreted with insulin by pancreatic islet β-cells [1]–[3] and acts as a synergistic partner of insulin to limit after-meal glucose excursions [4] . IAPP belongs to the calcitonin ( CT ) family of peptides ( See Table S1 in Text S1 ) [5]–[7] . The CT peptides function as hormones and are distributed in various peripheral tissues ( the endocrine pancreas in the case of IAPP ) , and play important biological roles including reducing nutrient intake ( IAPP ) , decreasing bond resorption ( Calcitonin ) and vasodilatation ( CGRP/calcitonin gene-related peptide ) . These functions are fulfilled through hormone-receptor agonism in which the CT peptide binds to a signal transduction membrane protein complex and thereby induces cell response in peripheral tissues . In the particular case of IAPP , this peptide binds to multiple amylin-specific ( AMY ) receptor complexes [1] , [7]–[10] . CT peptides show strong sequence homology in their two terminal regions ( residues 1–19 and 30–37 ) , and shares three features: [7] an N-terminal disulfide bond , an N-terminal amphipathic region , and C-terminal amidation . In addition to playing an important physiological role as a CT hormone , IAPP can also play a pathological role . Indeed , the 37-residue long human form of IAPP is the major protein constituent of pancreatic islet amyloid deposits found in 95% of Type 2 Diabetes ( T2D ) patients [11]–[14] . Interestingly , the IAPP peptide is found in a number of animal species , with a few point mutation differences , yet not all of these animals develop T2D . The development of T2D appears to be directly linked with the inherent aggregation propensity of the peptide . A recent bioinformatics study [15] ranked the aggregation propensities of the IAPP variants using the AGGRESCAN program [16] , with the pig sequence emerging as the least aggregation prone and the puffer fish as the most aggregation prone . Figure 1 lists the sequences of four IAPP species that we will consider in this study , and their aggregation propensities . Human and Cat ( as well as Monkey and Dog ) IAPPs all have high aggregation propensities , and all these species can develop T2D . Tellingly , species ( such as rodents and pigs ) that are well-known to tolerate excessive food intake without obvious health ramifications , have low aggregation propensities and are not known to develop Type II Diabetes . Transgenic rats , on the other hand , possessing the human variant of IAPP , spontaneously develop T2D when placed on a high calorie , sedentary diet [13] , [17]–[21] . In all IAPP variants ( Figure 1 ) , the two terminal parts are conserved ( namely residue 1–16 and residue 30–37 , which we will refer to as “conserved region I and II” , respectively ) . These are the very regions that are conserved in all CT family peptides and play important biological functions , with conserved region I activating the receptor and conserved region II binding in an antagonistic manner [22] . The mutations that differentiate the different IAPP forms occur in the middle region ( residues 18–29 , hereafter referred to as “the mutation region” ) and are responsible for the different aggregation propensities . The importance of this mid-region in governing aggregation is further highlighted by a familial form of T2D found in Japan that involves a single point mutation in this middle region ( S20G ) . The S20G mutant aggregates more rapidly than its human wild type hIAPP counterpart [23] , [24] , and leads to β-cell death and early onset T2D [25] , [26] . The above observations suggest a link between IAPP aggregation and the β-cell apoptosis occurring in T2D . Further support for a toxic role of IAPP aggregates include 1 ) the observation that human hIAPP amyloids play a deleterious role in transplanted islet tissue [27]–[30] and that aggregates of synthetic hIAPP induce apoptotic β-cell death in vitro [31]–[33] . 2 ) The recent experimental observation of two parallel pathways leading to β-cell apoptosis by hIAPP: one involving extrinsic death signals triggered by extracellular hIAPP aggregates [34] , [35] , and an intrinsic endoplasmic reticulum ( ER ) stress pathway linked to the presence of intracellular hIAPP aggregates [36]–[40] . 3 ) Experimental evidence of a membrane-damaging effect of hIAPP aggregates leading to β-cell dysfunction [41] , [42] . 4 ) The fact that blocking aggregation of hIAPP through interaction with non-aggregating hIAPP mutants [43]–[46] as well as small molecules ( e . g . EGCG [47] , tetracycline [48] and resveratrol [49] ) can reduce hIAPP-induced toxicity . A summary of a putative scheme embodying the functional and pathological roles of IAPP is given in Figure 2 . The fact that some IAPP variants aggregate while others do not raises the question as to whether the small sequence differences between species ( Figure 1 ) could lead to structural differences in the monomeric forms of these peptides that would lead one species to favor aggregate prone conformations . This question has been examined both experimentally and computational ( by our group and by others ) through a study of the rat and human IAPP sequences . The study of IAPP is particularly challenging because the peptide is intrinsically disordered , in other words , it does not populate a single well-defined three-dimensional structure , but rather interconverts between a number of co-existing conformations . While not readily amenable to traditional ensemble-averaging experimentally ( for instance CD studies [50]–[53] simply report that both human and rat IAPP variants are globally disordered and mainly adopts random coiled structures ) , recent experimental advances in studying IDPs have revealed subtle differences between IAPP variants . For instance , NMR studies looking at secondary chemical shifts [54]–[60] showed that IAPP variants are partially disordered but that the N-terminal part adopts helical structure , and that this structuring is significantly more pronounced in rIAPP than in hIAPP . A study combining data from CD , fluorescence dye binding , atomic force microscopy/AFM and electron microscopy/EM [61] showed that hIAPP adopts two distinct conformers containing both β-sheet and α-helix structural motifs . More recently , ion-mobility mass spectrometry studies [62] , [63] and 2D-IR spectroscopy analysis [64] , [65] showed evidence of the presence of β-structure in hIAPP monomeric and oligomeric samples . We recently explored the conformational space sampled by hIAPP and rIAPP using replica exchange molecular dynamics simulations ( REMD ) with an Amber force field ff96 and an implicit solvent ( igb5 ) . Our simulations showed that both peptides were flexible , and populated over 10 structural families ( see supporting material of ref . [62] ) even when using a very large dissimilarity measure ( i . e . Cα-RMSD cutoff of 3 Å ) . Despite this flexibility , we were able to identify some very interesting structural similarities and differences between these two sequences . Whereas rIAPP was found to populate only helix-coil conformations , hIAPP populated both helix-coil conformations and β-rich conformations including a helix-hairpin and an extended β-hairpin . The structures obtained from simulation were a good match for the collision cross-sections obtained experimentally using ion-mobility mass spectrometry [62] , [63] . Using GBSA implicit solvent model and Amber99SB force field , Murphy et al identified partially structured conformational states of the hIAPP monomer [66] . Using an explicit solvent and the Gromos96 53a6 force field , Reddy et al . independently obtained similar structures for both rIAPP and hIAPP , consistent with their 2D-IR data [64] , [65] . Using the modified TIP3P water model and the CHARMM27 force field , Liang et al . [67] probed sequence-induced differences in structural stability between hIAPP and rIAPP from preformed monomer to pentamer , which is based on strand-loop-strand scaffold . Their simulations showed rIAPP adopt less β-sheet-rich structure and a disturbed U-shaped topology than hIAPP . In the present paper , we perform an extensive investigation of the conformational space of two additional IAPP variants , the non-amyloidogenic pig variant ( pIAPP ) and the amyloidogenic cat variant ( cIAPP ) . Additionally , we extend our simulations of the rat and human forms to match the simulations lengths ( 600 ns/replica ) that we use in this study . To our knowledge , there are no published studies of the pig and cat IAPP structures . By enlarging our dataset of IAPP variants and using available peptide-activity data , we are now able to formulate a novel structure-activity relationship that rationalizes the dual function ( pathological and physiological ) of IAPP .
The secondary structure propensities for the four peptides are shown in Figure 3 . While all peptides show a high fraction of turn and coil structures ( from 0 . 49 for hIAPP to 0 . 57 for rIAPP ) , consistent with the natively disordered structural nature of IDPs , there are nonetheless some striking differences between the amyloidogenic ( human , cat ) and non-amyloidogenic ( rat , pig ) sequences . In particular , while helicity is present in each peptide , the degree of helicity is much more pronounced for the non-amyloidogenic sequences ( ∼0 . 4 for pIAPP and rIAPP vs . ∼0 . 1 for cIAPP and hIAPP ) . The trend for sheet structure is reversed , with low β-sheet content ( ∼0 . 03 ) for pIAPP and rIAPP and significantly larger content ( ∼0 . 4 ) for cIAPP and hIAPP . The location of the secondary structure elements on a per-residue level is shown in Figure 4 . Overall , the two non-amyloidogenic sequences ( pIAPP and rIAPP ) share similar secondary structural profiles , while the amyloidogenic sequences ( cIAPP and hIAPP ) share a different pattern . For the non-amyloidogenic sequences , conserved region I ( residues 1–16 ) consists primarily of helical structure , whereas the mutation region ( residues 18–29 ) and conserved region II ( residues 30–37 ) consist primarily of turns and coils , with a modest amount of helix and sheet structure . For the amyloidogenic sequences , the conserved region I ( residues 1–16 ) shows both helical and sheet elements , with the sheet contribution much more pronounced . The mutation region ( residues 18–29 ) now show a turn at residues 18–22 , linking the N-terminal ( residues 5–18 ) strand to a C-terminal ( residues 22–33 ) strand ( the latter located in conserved region II ( residues 30–37 ) ) , indicative of the presence of β-hairpin population . From our clustering analysis ( described in the Methods section ) , a large number of diverse structural families were identified . The centroid structures of the top 15 most populated structural families ( ≥1% of total structure population ) from the last 100 . 0 ns of simulation are shown in Table S3 of Text S1 for each IAPP sequence . The structural families were then further merged into several super structural families based on similarity in the molecular topology . A representative structure and the abundance for each super structural family are presented in Figure 5 . The non-amyloidogenic pIAPP and rIAPP structural ensembles contain two super families: a helix-coil super family ( structures A and C in Figure 5 ) and a helix-hairpin super family ( structures B and D in Figure 5 ) . In the helix-coil fold , the peptide adopts a short turn-coil ( residues 1–7 ) , a short helix ( residues 8–17 ) and a long turn-coil ( residues 18–37 ) . In the helix-hairpin fold , the peptide adopts a short turn-coil ( residues 1–7 ) , a short helix with a kink ( residues 8–17 ) and a short β-hairpin close to the N-terminal ( β-strands: residues 25–27 and 33–36 , loop: residues 28–31 ) . The helix-coil supper family ( ∼87% ) is more abundant than the helix-hairpin super family ( ∼13% ) . Although pIAPP differs from rIAPP by 9 out of 37 residues , their tertiary structure ensembles are very similar ( as are their secondary structural profiles , as seen in Figure 4 ) . The cIAPP and hIAPP structural ensembles contain three super families: a β-hairpin super family ( structures E and H in Figure 5 ) ( a β-strand 9–17 , turn 18–22 , and another β-strand 23–33 ) , a helix-coil super family ( structures F and I in Figure 5 ) , and a helix-hairpin super family ( structures G and J in Figure 5 ) . The first two super families are very similar to the two super families adopted by pIAPP and rIAPP , but occur in different abundance . The difference is particularly striking in the case of the helix-coil super family , with a population of ∼28% for cIAPP and ∼15% for hIAPP , dramatically less than the ∼87% seen for pIAPP and rIAPP . The helix-hairpin super family population is modest for all sequences ( ∼25% for cIAPP , ∼9% for hIAPP , and ∼13% for pIAPP and rIAPP ) . The β-hairpin fold is only seen for the amyloidogenic sequences , and occurs with large population ( ∼47% for cIAPP and ∼75% for hIAPP ) . Although the cIAPP structural ensemble is quantitatively similar to that of hIAPP , we identified a number of quantitative differences . In particular , the population of the hairpin super family is ∼28% less for cIAPP than for hIAPP , and the population of helix-coil super family of cIAPP is correspondingly larger ( by ∼13% ) . Subtle differences ( for instance , a shift or difference in the strand length ) are observed ( see Table S3 of Text S1 in which the top 15 structural families are shown ) . Peptide solubility is an important component in the aggregation process . We computed the GBSA solvation energy for each of the IAPP super families ( Table 1 ) . The absolute GBSA solvation energies of these ionic peptides are large ( <−554 kcal/mol ) as a result of the charges ( +4 of pIAPP and +3 of rIAPP , cIAPP and hIAPP ) carried by the four peptides . Based on the order of the GBSA solvation energy , we find that the β-hairpin is the least soluble motif , the helix-coil the most soluble with the helix-hairpin lying in the middle . When the relative GBSA of the four IAPP variants is considered ( using hIAPP as the zero scale reference ) , the order of solubility is rIAPP ( −97 . 1 kcal/mol ) >pIAPP ( −90 . 7 ) >cIAPP ( −74 . 3 ) >hIAPP ( 0 . 0 kcal/mol ) . This order correlates with the aggregation ability order of the four IAPP variants , with the non-amyloidogenic sequences being more soluble than the amyloidogenic ones . Since IAPP is an IDP , and , as such , populates partially structured conformers , it is important to consider the structural flexibility of the conformations identified in simulation . Our structural ensembles enable us to directly characterize this feature for the folds of the four IAPP variants by calculating their RMSF ( see Methods section ) . These results are reported in Figure 6 . The helix-coil fold of all four IAPP variants show much smaller structural fluctuation in the N-terminal part ( residues 1–17 , where the helix is located ) ( RMSF of ∼5 Å ) than in of the C-terminal part ( residues 18–37 ) , with an approximately linear increase from ∼5 Å to ∼20 Å . The flexibility of the N-terminal region may be required for the hormone function of IAPP ( i . e . interacting multiple membrane receptors ) . In the case of the helix-hairpin fold of the four IAPP variants , the N-terminal part ( residues 1–22 ) has comparable fluctuations to the same region in the helix-rich fold , but the C-terminal part ( residues 30–37 ) is slightly more rigid than the corresponding part in the helix-coil fold ( by ∼2 Å ) . In the case of the β-hairpin fold seen only in the amyloidogenic cIAPP and hIAPP sequences , the N-terminal part ( residues 1–22 ) has comparable fluctuations to the same region of their helix-coil fold , but the C-terminal part ( residues 23–37 ) is significant more rigid than the corresponding part of the helix-coil fold ( by ∼5 Å ) .
All IAPP sequences play the same physiological role in reducing post-meal blood glucose [2] , however , some sequences are capable of aggregating into pathological structures . In this paper , we used all-atom REMD simulations coupled with an implicit solvent model to thoroughly sample the conformations adopted by two amyloidogenic sequences of IAPP ( cat and human ) and two non-amyloidogenic sequences ( rat and pig ) . We wished to examine whether structural similarities existed between non-amyloidogenic and amyloidogenic forms of IAPP that could explain the dual functional/pathological roles that certain IAPP variants play . The similarity in functions suggests a possible similarity in structure . On the other hand , the fact that a few point mutations lead to enhanced aggregation tendencies suggests that these mutations may lead to dramatic conformational changes at the monomeric level , shifting the population from functional to pathological . Our simulations revealed that all four peptides populated helix-coil and helix-hairpin conformations , but that the amyloidogenic sequences populated in addition a β-hairpin conformation . The helix-coil structure was the dominant fold for the non-amyloidogenic structures , and the second dominant fold for the amyloidogenic sequences ( after the hairpin ) . We propose that this fold corresponds to the physiologically relevant fold . The helical region in this fold is located in the N-terminal region , a region conserved in all CT peptides . NMR studies on rat and human variants support the presence of helicity in the N-terminal region [54] , [56] , [59] , [68] . We found that this helical region was the most rigid region of all folds . Intriguingly , the C-terminal ( turn/coil ) region , corresponding to conserved region 2 , is the most flexible part . Both conserved regions 1 and 2 are involved in receptor binding , and it is possible that this dual rigid/flexible architecture may play a role in enabling IAPP to bind to multiple AMY1–3 receptors [7] , [22] . Given the lower abundance ( ∼22% ) of the helix-coil conformation adopted by amyloidgenic peptides ( cIAPP and hIAPP ) relative to that ( ∼87% ) of non-amyloidogenic peptides ( pIAPP and rIAPP ) , we would expect hIAPP and cIAPP to have slightly reduced normal hormone function relative to pIAPP and cIAPP . Indeed , Young et al . [2] have shown in a preclinical rat study that hIAPP has slightly lower binding affinities for amylin , CGRP , and calcitonin receptors , and induces slightly weaker responses in isolated muscle ( Table 2 ) . In contrast , the β-hairpin structure , present only in the amyloidogenic sequences , is a possible candidate for an amyloid-competent structure . β-hairpin structures have been found in molecular dynamics simulations in a number of amyloidogenic peptides , most notably in fragments of the Alzheimer Amyloid Aβ peptide [69]–[83] and the prion protein [84] , [85] . Simulations of oligomerization of the Aβ ( 25–35 ) peptide indicate that hairpins play an important role in initiating aggregation and in stabilizing the growing front of the fibril [83] . This may also be the case for IAPP . The hairpin structure that we find in simulation shares important similarities with the structure of hIAPP in the context of a fibril . The solid state NMR [86] structure of the hIAPP fibril consists of a strand-loop-strand topology ( related to the strand-turn-strand hairpin by a 90° rotation ) , with the loop ( residue 18–27 ) located at the turn region of our hairpin . By having the correct strand placement ( as in the fibril ) , the hairpin structure could facilitate nucleation and subsequent fibril growth . Further support for the notion of this structure as a key player in aggregation comes from the work of Kapurniotu and co-workers who identified through fragment binding affinity studies a number of “hot-spot” regions responsible for inter-peptide interactions in aggregation [87] . These hot-spots correlate with the β-strand regions of the hairpin seen in our simulations [62] . The β-hairpin structure is less soluble and less flexible than the helix-coil fold based on our analysis of the GBSA solvation energy and Cα-RMSF , features that make it a good candidate for β-sheet formation . Indeed , this increased rigidity might contribute to the fast association of β-hairpins into β-sheet rich oligomers . Our recent dimer simulations of hIAPP and rIAPP [63] support this picture: the β-rich monomers have strong tendencies to form β-rich dimers , while the helix-coil rich monomers form , if anything , only loosely bound , disordered complexes with much lower binding energies . Formation of β-rich hIAPP dimers was also found in atomistic , explicit solvent simulations [65] and both hIAPP dimers with moderate β-content and rIAPP dimers with no β-content were observed in a Hamiltonian-Temperature-REMD simulations using a coarse grained protein force field ( OPEP ) [88] . The notion of a hairpin as an important player in the aggregation process can explain a number of experimental observations . For instance , within this framework , the increased aggregation rates of the S20G IAPP mutation [25] , [26] can be explained by an increased propensity to form a β-turn ( glycines are turn promoters and residue 20 that is involved in this mutation lies right in the turn region of our hairpin ( residues 18–22 ) ) . In addition , the observed inhibition effects of a non-aggregating form of hIAPP with two N-methylations at positions G24 and I26 ( hIAPP-GI ) could be explained by blocking inter-strand β-sheet formation [43] , [89] , [90] . In other words , the β-strand of hIAPP-GI could bind to the β-strand of hIAPP , forming a complex that now has a face with exposed N-methyl groups that is unable to hydrogen bond with another hIAPP peptide , thus blocking further growth . cIAPP is less amyloidogenic than hIAPP [15] and , consistent with our hypothesis of a role of the hairpin in facilitating aggregation , our structural data shows that cIAPP has slightly lower β-sheet content ( β-hairpin ) than hIAPP . We speculate that the decrease in hairpin population is due to the S29P mutation that differentiates cat from human . Proline is indeed known to be a β-sheet breaker . Along similar lines , other disorder-promoting substitutions ( e . g . P , R and K ) [91] may further lead to a diminished propensity for hairpin formation , eventually leading , in the case of rIAPP ( with key mutations A25P , S28P and S29P ) and pIAPP ( S20R and N31K ) to the complete disappearance of the β-hairpin population . It is interesting to note that the drug PRAMLINTIDE ( symlin ) with same sequence as hIAPP , but containing the 3 proline substitutions of rIAPP , shows a very weak tendency to aggregate [58]and has proven to be an efficacious agent that takes over the physiological role of hIAPP , acting as a synergistic partner to insulin [2] , [92] . Interestingly , we find that all four sequences adopt a helix-hairpin fold , although in lower amounts than the helix-coil fold ( for pIAPP and rIAPP ) and hairpin ( for hIAPP and cIAPP ) . This fold is more soluble than the β-hairpin fold , but less soluble that the helix-coil fold . We speculate that this motif may act as an on-pathway intermediate leading to β-rich conformations ( β-hairpin or β-sheet ) and thus amyloid fibrils under certain conditions ( e . g . the presence of an interface , peptide-peptide interaction , solvent effect etc . ) . Indeed , rIAPP , although commonly thought of as a non-aggregating species , has been observed to aggregate into fibrils under specific non-physiological conditions [93]–[95] . Indeed β-sheet-rich fibrils of rIAPP were seen to form at a liquid-solid interface [93] , mixing rIAPP with hIAPP lead to a templating of rIAPP onto hIAPP fibrils [94] , and placing rIAPP in Tris-HCl buffer and sonicating lead to fibril formation [95] . We note that it is also plausible , as proposed by Miranker and coworkers [54] , Eisenberg and coworkers [57] and Raleigh and coworkers [96] , that early oligomerization may be initiated by helix-helix association , with β-structure emerging later in the aggregation process . It is of course difficult to tell whether experimentally observed helix-rich oligomers are on or off-pathway to fibrils . Likewise , the hairpins that we see in simulation may be on-route to fibrils formation , or may as well lead to off-pathway aggregates . However , small hairpin oligomers , even if not directly on-route to fibril formation , may play an important role in toxicity . Small oligomers are also increasingly being associated with hIAPP cytotoxicity [35] , [37] . In particular , membrane pores , formed by small amyloidogenic oligomers , have been suggested as a means of toxicity of hIAPP [97] . These pores can be formed by helical conformers of hIAPP , as supported by experiments on the 1–19 fragment of IAPP . Indeed , the human and rat IAPP ( 1–19 ) fragments can adopt helical conformation in membrane mimics [56] and have been shown to be toxic to cells , with hIAPP ( 1–19 ) being more toxic than rIAPP ( the latter differ by an H to R substitution ) [98] . We note that it is possible that IAPP pores can also be formed from β-rich conformation , as in the case of the β-rich annular-like channel proposed for Aβ and other amyloid peptides [99] , [100] , based on molecular dynamics simulations , atomic force microscopy and channel conductance measurements . A similar β-rich annular-like channel model for hIAPP has recently been proposed using molecular dynamics simulations [101] . The hairpin structures that we observe are reminiscent of the cylindrin structures discovered by Eisenberg and co-workers , cylindrical barrels of β-hairpins that may interact with membranes and constitute a generic architecture for toxic amyloid oligomers [102] . There is at present no experimental data for a cylindrical barrel model for hIAPP , but such a structure is plausible given the observed β-hairpin in simulations and the observation of β-barrel type of ion channel for other amyloid systems . Finally , another toxicity mechanism of hIAPP may be associated with membrane fragmentation due to the growth of amyloid fibrils [103] . In summary , there are compelling genetic , biochemical , cellular and animal data to support both natural biological functions for hIAPP and a toxic role of hIAPP leading to β-cell death [14] , [34] , [37] , [104]–[107] . Combining these functional data with our structural models of four IAPP variants , we put forth the structure-activity relationship ( SAR ) that the helix-coil conformations are responsible for the normal hormone function of IAPP; and that β-rich conformations of IAPP may be linked to β-rich aggregation and contribute , along with other mechanisms , to the toxicity of IAPP . While the former might be realized by binding of the helix-coil conformers to AMY receptor , the latter might be due to the formation of toxic β-rich oligomers and amyloid fibrils leading to β-cell death . This SAR scheme is summarized in Figure 7 . Our SAR can give insights into the rational design of drugs to combat Type II Diabetes , with drugs that either destabilize the pathological conformations and/or promote the formation of the physiologically active conformations . These drugs could come in the form of small molecules , or be peptide based . Ideally , one could design an IAPP variant that retains the functional role of hIAPP , but does not have the same tendency as hIAPP to misfold and aggregate ( such as pramlintide ) , and that furthermore inhibits the aggregation of wild type hIAPP . This drug would not only enhance insulin-sensitivity ( like pramlintide ) , but also preserve β-cells by preventing aggregation .
Because experiments are typically performed around 300 K , our data analysis was focuses on the replica at 300 K . The convergence was rigorously checked by a block analysis: the total 600 . 0 ns sampling at 300 K was equally divided into six blocks , and structural properties was calculated for each block . For the four sets of REMD simulations , a good convergence was found during the last half of the trajectory ( see for example , the secondary and tertiary structure data of the four IAPP variants in Table S2 of Text S1 ) . Thus , the standard deviations of the structural properties presented in the main text were calculated from the last three blocks ( i . e . the last 300 . 0 ns ) . The STRIDE program of Frishman and Argos [120] was used to obtain secondary structure propensities . For tertiary structure analysis , the structural ensembles from simulations were classified into structural families using the GROMACS clustering protocol [121] , in which the structure similarity metric is based on a pair wise Cα- RMSD ( root mean square deviation ) cutoff of 3 . 0 Å , the neighboring structures are identified for every structure using the RMSD similarity cutoff , the structure having the most neighbors ( called as the centroid structure ) is removed together with its neighbors to form a structure family , and the process continued for the remaining structures until all structures have been assigned into the structural families . The centroid structure serves as the representative structure of the structural family . For example , the centroid structures of the top 15 populated structural families ( ≥1% of total structure population ) from the last 100 . 0 ns for each IAPP sequence are shown in Figure S3 . Next , the structural families were further merged into three super-families ( helix-coil , helix-hairpin and β-hairpin ) based on the secondary structure type of both the N-terminal part ( residues 1–17 ) and the C-terminal part ( residues 18–37 ) : First , it belongs to a helix-rich fold if the N-terminal part contains more helix than β-sheet , otherwise it belongs to β-hairpin super family; Second , a helix-rich fold belongs to the helix-hairpin super family if the C-terminal part contains more than four sheet-residues coil ( i . e . a minimal 2∶2 β-hairpin ) , otherwise it belongs to the helix-coil super family . Dynamic fluctuations of each residue was characterized by calculating the Root Mean Square Fluctuation ( RMSF ) of its Cα atom from the structural ensemble . Because the N-terminal part ( residues 1–17 ) of these IAPP peptides is more rigid than its C-terminal part ( residues 18–37 ) , a superposition of the C-terminal part was carried out prior to the RMSF calculation . Also because the four IAPP variant contains two or three types of folds ( helix-coil , helix-hairpin and β-hairpin ) , the RMSF was calculated separately for each one . The absolute solubility of the four IAPP variants can be estimated from their solvation free energies . When the solute conformational entropies of the IAPP variants are comparable , the relative solvation free energy can be estimated from the relative GBSA solvation energy . A recent benchmark study [122] has shown that GBSA models give reliable results when only the relative solvation energy is considered . We obtained the statistics of GBSA solvation energy for each IAPP fold from its last 300 . 0 ns of simulation data at 300 K . | IAPP , a 37 amino-acid peptide hormone belonging to the calcitonin family , is an intrinsically disordered peptide produced along with insulin by pancreatic islet β-cells in response to meals . In its functional form , IAPP acts as a synergic partner of insulin to reduce blood glucose . IAPP can , however , also play a pathological role , contributing to Type II diabetes ( T2D ) . Knowledge of the structural nature of the physiological and pathological forms of IAPP will facilitate the rational design of novel drugs for therapeutic treatment of T2D . However , because IAPP does not fold to a single structure , but rather co-exists between multiple functional ( and toxic ) structures , it is extremely challenging for experimental methods to gain detailed structural information . Using a computational approach , we were able to obtain detailed structures of four IAPP variants and propose a novel structural hypothesis for the two opposing roles of this peptide . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| []
| 2013 | Structural Similarities and Differences between Amyloidogenic and Non-Amyloidogenic Islet Amyloid Polypeptide (IAPP) Sequences and Implications for the Dual Physiological and Pathological Activities of These Peptides |
The influenza viruses contain a segmented , single-stranded RNA genome of negative polarity . Each RNA segment is encapsidated by the nucleoprotein and the polymerase complex into ribonucleoprotein particles ( RNPs ) , which are responsible for virus transcription and replication . Despite their importance , information about the structure of these RNPs is scarce . We have determined the three-dimensional structure of a biologically active recombinant RNP by cryo-electron microscopy . The structure shows a nonameric nucleoprotein ring ( at 12 Å resolution ) with two monomers connected to the polymerase complex ( at 18 Å resolution ) . Docking the atomic structures of the nucleoprotein and polymerase domains , as well as mutational analyses , has allowed us to define the interactions between the functional elements of the RNP and to propose the location of the viral RNA . Our results provide the first model for a functional negative-stranded RNA virus ribonucleoprotein complex . The structure reported here will serve as a framework to generate a quasi-atomic model of the molecular machine responsible for viral RNA synthesis and to test new models for virus RNA replication and transcription .
The influenza A viruses belong to the family Orthomyxoviridae and are genetically and antigenically heterogeneous . They are responsible for annual epidemics of respiratory disease and represent an important public-health problem [1] . All viral subtypes can be found in their natural reservoir , that comprises several wild avian aquatic and terrestrial species . From this reservoir , influenza viruses can occasionally infect mammalian species , including man , by either gene reassortment with already established mammalian viruses or by direct adaptation [2] , and thus produce a pandemic . Since 1997 , transmissions of avian H5N1 influenza viruses to humans have originated hundreds of highly pathogenic infections and generated fears for a new pandemic of unprecedented impact [2] , [3] . The recent transmission of swine H1N1 influenza viruses to humans could represent the first time that a new pandemic can be followed on line ( http://www . who . int/csr/disease/swineflu/en/index . html ) . The genome of the influenza A viruses comprise eight single-stranded RNA molecules of negative polarity with partially complementary ends that form a closed structure . The native ribonucleoprotein ( RNP ) particles are formed by the association of these single-stranded RNAs to multiple monomers of nucleoprotein ( NP ) and a single copy of the polymerase , a heterotrimer composed by the PB1 , PB2 and PA subunits [4] , [5] . Such RNPs are independent molecular machines responsible for transcription and replication of each virus gene . When analysed structurally by electron microscopy , virion RNPs appear as flexible , supercoiled structures [6] , [7] . The helical organization of the RNPs is determined by the structure of the NP , as complexes of NP and unrelated RNA also adopt helical structures [8] , and purified NP can form RNP-like helical particles in the absence of RNA [9] . The polymerase complex binds the vRNA promoter , that is formed by the partially complementary 5′- and 3′-terminal sequences [10]–[12] , and determines the superhelical arrangement of natural virus RNPs [13] . Although the RNPs are the essential elements for virus replication and gene expression , their structural analysis has been hampered by their heterogeneity and flexibility . However , in vivo replication of recombinant model-RNPs indicated that helical- , elliptic- or circular-shaped structures could be generated with RNA templates of diminishing lengths [14] . The clone 23 model-RNP , which represents the smallest efficient replicon , was circular in shape and showed sufficient structural rigidity to be analysed by electron microscopy and image processing after negative staining [15] . Here we report the purification of recombinant clone 23 RNPs to near homogeneity and their structural analysis by cryo-electron microscopy ( cryo-EM ) . It is important to stress that the RNPs analysed were the final products of in vivo replication , as no RNP accumulation was observed when NP or polymerase negative mutants were used for in vivo reconstitution . The final structure shows a resolution of 12 Å for the NP and 18 Å for the polymerase complex and represents the first structure of a functional influenza virus RNP and indeed of the RNP from any negative-stranded RNA virus .
Previously , we used recombinant RNPs purified by successive glycerol gradient centrifugation steps to analyse their structure by electron microscopy of negative-stained samples [15] . To improve the purity and yield of the RNP preparations , we used a PB2 subunit containing a His-tag at the C-terminus , a modification that did not alter the in vivo replication activity of the RNPs , as described previously [16] . The purification protocol involved an optimised Ni-NTA-agarose affinity step , a gel-filtration chromatography and a final concentration on a Ni-NTA-agarose resin . Such procedure allowed the routine preparation of essentially homogeneous and biologically active RNPs with a concentration appropriate for structural analysis ( Fig . 1A–E ) . Most of the cellular contaminants could be removed in the first Ni-NTA column , while active RNPs were concentrated ( Fig . 1A , B ) . The remaining contaminants were eliminated by gel filtration ( Fig . 1E ) , a step in which the signals for the polymerase and NP co-migrated with the in vitro transcriptional activity ( Fig . 1C , D ) . The purified RNPs ( Fig . 1E , frame ) were concentrated by binding to and elution from Ni-NTA-agarose ( data not shown ) and used for cryo-EM . To generate an initial model for reconstruction , a purified RNP sample was stained with uranyl-acetate and imaged at 20° tilt in a FEI Tecnai G2 field emission gun microscope . A total of 2035 particle images were employed to generate a three-dimensional reconstruction using the SPIDER algorithms [17] . To generate a three-dimensional reconstruction of frozen-hydrated RNPs , samples of purified RNPs ( Fig . 1E , frame ) were fast-frozen on holey-grids and imaged in the same microscope . A total of 9571 individual particle images were selected from the micrographs after CTF correction and used for refinement ( see Fig . S1 for a gallery of single particle images ) . Two independent refinement processes were carried out , with and without imposing 9-fold symmetry . The three-dimensional reconstruction obtained by imposing 9-fold symmetry lacked information about the polymerase complex but could achieve better resolution for the NP ring . On the contrary , refinement without imposing symmetry allowed reconstruction of the complete RNP particle but the resolution obtained was significantly lower ( Fig . S2 ) . The final structure is shown in Fig . 2 and Video S1 , and represents a composite map formed by the 7 NP monomers not contacting the polymerase , that are derived from the structure refined with symmetry , while the polymerase complex , as well as the 2 adjacent NP monomers are derived from the volume refined without symmetry . Therefore , the resolutions for either section of the model are different: 12 Å for the NP ring and 18 Å for the polymerase complex ( Fig . S3 ) . Each NP monomer consists of two domains , an upper head domain and a centred body , which contains a small mass protruding at the bottom . When represented at the calculated threshold no massive contacts among the NP monomers were apparent , suggesting that the interaction sites are flexible or random coil chains . The polymerase complex is in contact with two of the NP monomers , which lack apparent interaction with each other ( Fig . 2 ) . The structure of the polymerase complex resembles that previously obtained by negative-staining [16] , [18] , but has higher resolution . A comparison between both structures allowed the localisation of specific subunit domains , as defined earlier by binding of monoclonal antibodies or tagging ( Fig . 3A ) and suggest that the main NP-polymerase interactions are mediated by the PB1 and PB2 subunits . These interactions are quite different in intensity , the former being stronger than the latter ( Fig . 2 , Fig . S2 and Video S1 ) . Docking the recently reported atomic structure of the PA ( C ) -PB1 ( N ) dimer [19] , [20] was consistent with its predicted localisation ( Fig . 3B ) [16] and would suggests that the PB1 and PA subunits account for the upper , bulkier section of the complex while PB2 would be localised at the bottom region . We also carried out a docking of the atomic structure of the NP in the cryo-EM reconstruction . The handedness of the cryo-EM map was determined on the basis of the correlation coefficient of the NP atomic structure docked into the symmetrised volume . The fitting assays were carried out with both handednesses , using either volumetric or laplacian criteria . The maximum correlation coefficients were 0 . 854 and 0 . 341 for volumetric and laplacian tests , respectively . These values were 2 to 30% better for the selected as compared to the alternative handedness . In addition , another important consideration indicates that the selected handedness is correct . In the atomic structure of the influenza NP ( pdb accession number 2IQH ) there are some portions of the molecule that are not defined . The connections between the loop 402–428 ( which is involved in NP-NP interaction; see below ) and the body of the molecule could not be determined ( sequence A428-S438 ) . The distance between these two amino acids in the selected fitting was around 25 Å , compatible with a 10 amino acids distance , whereas in the fitting performed in the structure with the opposite handedness , these two amino acids were 41 Å apart . The result of the docking is shown in Fig . 4A and confirms the quality of the structural model obtained . A good fit is observed between the two domains described in the atomic structure and the volume of the NP monomer . However , additional masses are observed at the top and at the bottom of each NP monomer . It could be argued that such additional masses arise as a consequence of using an initial negative-stain model that was derived by conical-tilting . However , we used the same image data set to carry out a control refinement , using as initial model a 9-mer-ring structure generated with the atomic model of the NP filtered to a resolution of 30 Å , and the final model obtained was indistinguishable from that shown in Fig . 2 ( data not shown ) . Furthermore , the angular coverage of the images ( Fig . S4 ) was sufficient to exclude the missing cone as an explanation for this extra volume . Thus , we feel that the additional masses detected in the cryo-EM model of the NP monomers are bona fide . We propose that the extra mass at the top of the NP monomer corresponds to the protein sequences not solved in the crystal structure [21] while that at the bottom may contain genomic RNA . In fact , when decreasing threshold values were used to represent the RNP volume , the additional mass at the bottom of the NP was persistent , suggesting a high mass density ( data not shown ) . To test the potential RNA-dependence of the RNP structure , these were purified by affinity chromatography on Ni-NTA-agarose , extensively treated with T1 and pancreatic RNAses and analysed by gel filtration . The results are shown in Fig . 5A and clearly indicate that the interaction between the polymerase complex and the NP ring is highly RNA-dependent , as both substructures could be separated after RNAse treatment . On the other hand , the size of the template RNA before and after digestion with RNAse was analysed and a resistant band of around 18 nt was apparent ( Fig . 5B ) . Since an average content of 24 nt per NP monomer has been determined [14] , this result would suggest that most of the template RNA is uniformly distributed along the RNP structure and protected by association to the NP . Docking of the atomic structure of the NP monomers into the cryo-EM structure also allowed us to predict their interaction interfaces . It was earlier proposed that interaction of the loop 420 ( positions 402–428 ) with a neighbouring NP monomer would account for NP polymerisation [21] , but this was suggested on the basis of the formation of a crystallographic trimer and no functional data was reported . The interaction among NP monomers is conserved in the NP docking presented here , with the only need to alter the angle between NP monomers from about 120° in the crystal to 40° in the RNP volume ( Fig . 4A ) . This interaction interface would be more realistic , as no NP trimeric structure has been described in natural virus RNPs , and would imply a small change in the arrangement of the connections between the loop and the body of the NP ( positions 428–438 and 396–402 ) . These connections are in any case highly flexible and were not resolved in the atomic structure of the trimer [21] . Although such a flexible connection is not detectable in the cryo-EM map at the threshold shown in Fig . 4A ( σ = 2 . 5 ) , it is clearly visible when the volume is represented at σ = 1 . 5 ( Fig . 4B , blue arrow ) . It is not clear whether the contacts between the NP monomers observed in the atomic structure of the trimer would be strictly conserved in the functional RNP nonameric structure . Hence we mutated several of the positions in the loop , affecting either conserved or non-conserved amino acids ( Fig . S5 ) , and tested the biological activity of the RNP . The replication of a viral RNP does not lead to a naked progeny RNA but rather a progeny RNP structure and it is generally accepted that encapsidation of the newly synthesised RNA by the polymerase complex and NP monomers is coupled to RNA replication . Hence , if the mutations were to affect the NP-NP interaction , a deficiency in RNP replication would be expected . Thus we reconstituted in vivo mini-RNPs by transfection of plasmids encoding the polymerase subunits ( of which PB2 as a His-tagged protein ) , a clone 23 template RNA and either wt or mutant NP , and purified them by Ni-NTA-agarose chromatography . The accumulation of progeny RNPs was determined by Western-blot using anti-NP antibodies and represents the in vivo replication phenotype . Mutants R416A and F412A were strongly affected in replication , whereas mutants S413T , F420A , K422A and S423A behaved as wt ( Fig . 6A , B and Fig . S5 ) . These results confirm the relevance of the interaction between R416 in the loop and E339 in the connecting NP [21] and suggest that residue F420 in the loop does not play an important role in the interaction . On the other hand , the phenotype of mutant F412A indicates that it is important for viral RNA replication . To further analyse the phenotype of the NP mutants generated , the amount of purified mutant RNPs recovered by replication in vivo was determined by measuring their in vitro transcription activity . The results of a typical experiment are presented in Fig . 7A and average of two independent experiments is shown in Fig . 7B . These results are consistent with the deficiency in the replication activity observed for mutants R416A and F412A . The replication-defective phenotype observed for these mutants could be the consequence of a defect in their homopolymerisation capacity . To analyse this possibility mutant or wt NP were expressed by transfection in COS1 cells and total extracts were analysed by gel-filtration after extensive RNAse treatment . Under these conditions , wt NP formed large complexes compatible with NP polymers . On the contrary , mutant R416A , that was shown as negative in NP-NP association [21] , behaved as monomer ( Fig . 8 ) . The phenotype of the other mutants correlated with their replicative activity in vivo . Thus , mutant F420A behaved as wt while mutant F412A showed an intermediate phenotype .
In this report we have presented the three-dimensional structure of an active influenza virus RNP , as determined by cryo-EM . In fact , this represents the first structure of a biologically active RNP from any negative-stranded RNA virus . Two technical developments have allowed this breakthrough: ( i ) the generation of recombinant RNPs that are efficient replicons and have sufficient structural rigidity [14] and ( ii ) the optimisation in the purification protocols of RNPs amplified in vivo . As compared to full-length virion RNPs , the structure reported here would represent a minimal RNP in which the helical section has been deleted and only the promoter region bound to the polymerase complex and the terminal loop remains . The structure obtained for the polymerase complex present in the RNP is compatible with those reported earlier by negative-staining [16] , [18] and represents the most accurate model for a complex polymerase of a negative-stranded RNA virus thus far reported . The correlation with the sites previously mapped [16] and the docking of the atomic structure of specific domains permitted the rough localisation of the polymerase subunits ( Fig . 3 ) . Unfortunately , the other polymerase domains whose atomic structure is known [22]–[24] are not large and conspicuous enough to allow unambiguous docking in the cryo-EM structure . The interaction among NP monomers was analysed by docking of the atomic structure into the NP ring . The model obtained is compatible with the interaction mode proposed earlier [21] and further indicated that additional side-by-side interactions are now possible due to the tighter packing of the monomers ( see Fig . 4A , black arrow ) . The relevance of the 420–428 loop in the NP-NP interaction was verified functionally: The contacts of amino acid R416 and F412 are essential for replication , while amino acid K422 does not appear to be important , in spite of being conserved among type A and B viruses ( Fig . S5 ) . Previous biochemical studies had shown that residue R416 is involved in NP-NP interaction [25] and that both F412 and R416 were important for RNA binding [26] . In view of the results presented here it is possible that the RNA-binding defect detected might be secondary to the homopolymerisation failure . In addition , the residue at position 412 appears to be important for the template activity of the RNP , since mutation F412A specially affected the in vitro transcription of mutant RNPs ( compare Figs . 5 and 6 ) . Contrary to the N-RNA complexes in the Mononegavirales [27] , [28] , that contain 9 nucleotides associated to each N molecule , we have estimated an average of 24 nucleotides per NP monomer in influenza RNPs [14] . The structure of the RNP presented here is compatible with the RNA-binding site being located at the groove between the head and body domains in the NP , as previously suggested [21] , [29] . Indeed , a connecting mass is apparent in the appropriate position ( see Fig . 3A , black arrow ) that could represent the template RNA in addition to protein contacts . However , most of the RNA sequence present in the RNP is resistant to extensive RNAse treatment and the main protected fragment is around 18 nt long ( Fig . 5 ) . This would suggest that the template RNA is distributed uniformly along the RNP structure , i . e . variations of the average value of 24 nt per NP monomer are small . Furthermore , the size of the protected fragment ( 18 nt ) is similar to the average assignment of RNA per NP , suggesting that the template RNA associates to several regions of the NP and could contribute to the extra mass observed at the bottom of each NP monomer . In addition , the N-terminal region of NP , which has been implicated in binding to RNA by biochemical assays [30] and is not represented in the atomic structure of the protein [21] , [29] , could also contribute to this extra mass . In summary , we have reported the first structure of a biologically active influenza RNP . This three-dimensional structure reveals the NP-NP interaction domain and will serve as a framework to generate a quasi-atomic model of the molecular machine responsible for viral RNA synthesis .
The origin of plasmids pGPB1 , pGPB2His , pGPA , pGNP ( polyA ) and pT7ΔNSRT clone 23 , containing sequences derived from the A/Victoria/3/75 influenza virus strain , has been described [14] , [16] , [31] . The vaccinia recombinant virus expressing T7 RNA polymerase ( vTF7-3 ) [32] was provided by B . Moss . The origin of antibodies specific for PB1 , PB2 and PA has been described [14] , [33] , [34] . Antibodies specific for NP were generated by immunisation of rabbits with purified His-NP . The NP mutants were generated by site-directed mutagenesis on pGNP ( polyA ) plasmid using the Stratagene Quickchange kit and specific oligonucleotides ( sequences available upon request ) and their genotype was verified by sequencing . Recombinant RNPs containing the ΔNS clone 23 genomic RNA ( 248 nt ) were generated and amplified in vivo by transfection of plasmids pGPB1 , pGPB2His , pGPA , pGNP ( polyA ) and pT7ΔNSRT clone 23 into vaccinia vTF7-3-infected COS1 cells as described previously [16] . For RNP purification , the clarified cell extracts were incubated overnight at 4°C with Ni-NTA-agarose resin in a buffer containing 50 mM Tris-HCl-100 mM KCl-5 mM MgCl2-0 . 5% Igepal-20 mM imidazol-1 u/µl RNAsin-EDTA-free protease inhibitors cocktail , pH 8 . The resin was washed with 80 volumes of 50 mM Tris-HCl-100 mM KCl-5 mM MgCl2-0 . 5% Igepal-20 mM imidazol , pH 8 and 20 volumes of the same buffer containing 50 mM imidazol . Finally , the RNPs were eluted with 50 mM Tris-HCl-100 mM KCl-5 mM MgCl2-0 . 5% Igepal-150 mM imidazol , pH 8 . The eluted RNPs were filtered on a Sephacryl S300 column equilibrated with 50 mM Tris-HCl-100 mM KCl-5 mM MgCl2-0 . 5% Igepal-20 mM imidazol , pH 8 and the peak RNP fractions were further bound to Ni-NTA-agarose in 50 mM Tris-HCl-100 mM KCl-5 mM MgCl2-0 . 5% Igepal-20 mM imidazol-1 u/µl RNAsin-EDTA-free protease inhibitors cocktail , pH 8 , washed once with 50 mM Tris-HCl-100 mM KCl-5 mM MgCl2-0 . 5% Igepal-20 mM imidazol , pH 8 and eluted with 50 mM Tris-HCl-100 mM KCl-5 mM MgCl2-0 . 3% CHAPS-150 mM imidazol , pH 8 . Western-blotting was performed as described [14] . Protein silver-staining was carried out as indicated before [35] . To determine the transcription activity of purified RNPs , samples were incubated in a buffer containing 50 mM Tris-HCl-2 mM MgCl2-100 mM KCl-1 mM DTT-10 µg/ml actinomycin D-1 u/µl RNAsin-1 mM ATP-1 mM CTP-1 mM UTP-10 µM α-P32-GTP ( 20 µCi/µmol ) -100 µM ApG for 60 min at 30°C . The RNA synthesised was TCA precipitated , filtered through a nylon filter in a dot-blot apparatus and quantified in a phosphorimager . To test the in vivo RNP replication , cultures of COS1 cells were infected with vaccinia vTF7-3 and transfected with plasmids pGPB1 , pGPB2His , pGPA , pGNP ( polyA ) ( or mutants thereof ) and pT7ΔNSRT clone 23 . Total cell extracts were used for purification by affinity chromatography on Ni-NTA-agarose as indicated above and the accumulation of progeny RNPs was determined by Western-blot with anti-NP-specific antibodies and by measuring their in vitro transcription activity . To determine the NP aggregation state , cultures of COS1 cells were infected with vaccinia vTF7-3 and transfected with plasmid pGNP ( polyA ) ( or mutants thereof ) . Total cell extracts were prepared , treated with 50 µg/ml of RNAse A for 2 hours at room temperature and analysed by filtration over a Sephacryl S300 column calibrated with ferritin ( 440 kDa ) and BSA ( 67 kDa ) . For electron microscopy of negatively stained samples 4 µl aliquots of purified RNPs were applied to glow-discharged carbon grids for 1 min and then stained for 1 min with 2% uranyl acetate . Low-dose images were taken on a 200 kV FEI Tecnai G2 Field emission gun electron cryomicroscope operated at a nominal magnification of 50 k at 20° tilt . A total of 2035 individual RNP images were extracted and processed to generate an initial model using the SPIDER software [17] . For cryoelectron microscopy , 5 µl aliquots of purified RNPs were applied to glow-discharged Quantifoil holey grids for 2 min , blotted and frozen rapidly in liquid ethane at −180°C . Images were taken with the same conditions as in the negative stain experiments but without tilting . The selected micrographs were scanned on a Zeiss scanner ( Photoscan TD , Z/I Imaging Corporation ) with a final pixel size corresponding to 2 . 8 Å . Contrast transfer function ( CTF ) of micrographs was estimated using ctffind software [36] and corrected using Bsoft [37] . A total of 9571 images were subjected to two independent refinements with and without imposing 9-fold symmetry using SPIDER software [17] . After reaching the convergence of these refinements , the reconstructions yielded resolutions of 18 and 12 Å for non-symmetrized and symmetrized structures , respectively ( FSC 0 . 3 criterion ) . The final tilt range assigned in the refinement for the whole set of individual images was checked ( Fig . S5 ) and showed an angular distribution where the effect of missing cone in the reconstruction could be considered as negligible . The absolute handedness of the volumes was determined using the atomic structure of NP protein [21] , and turned out to be the opposite to that previously published [15] , [16] . Docking experiments were carried out using SITUS software [38] . Finally , and to verify the positions of the extra mass and the quality of the three-dimensional reconstruction , an additional refinement was carried out using as initial model the structure of the 9 NP-mer ring resulting from the docking experiments , filtered at 35 Å . This refinement yielded a reconstruction similar to the final structure presented here , showing that the additional masses detected in the cryo-EM structure protruding from the NP monomers are bona fide . Volume handling was carried out using XMIPP software [39] and general visualization was performed using Chimera [40] and Amira ( http://amira . zib . de ) . The cryo-EM map has been deposited in the Electron Microscopy Data Bank ( accession code EMD-1603 ) and the fitted atomic structure in the Protein Data Dank ( accession code 2wfs ) . | The influenza viruses cause annual epidemics of respiratory disease and occasional pandemics that constitute a major public-health issue . The recent spillover of avian H5N1 and H1N1 swine influenza viruses to humans poses a serious threat of a new pandemic . These viruses contain a segmented RNA genome , which forms independent ribonucleoprotein particles including the polymerase complex and multiple copies of the nucleoprotein . Each of these ribonucleoprotein particles are replicated and express the encoding virus genes independently in the virus-infected cells . To better understand how these processes take place we have determined the three-dimensional structure of a model ribonucleoprotein particle that only contains 248 nucleotides of virus RNA but is biologically active in vitro and in vivo . The structure shows a circular appearance and includes 9 nucleoprotein monomers , two of which are associated to the polymerase complex . Docking of the available atomic structures of the nucleoprotein and domains of the polymerase complex has permitted us to propose a quasi-atomic model for this ribonucleoprotein particle and some of the predictions of the model have been confirmed experimentally by site-directed mutagenesis and phenotype analysis in vitro and in vivo . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"molecular",
"biology/rna-protein",
"interactions",
"virology/viral",
"replication",
"and",
"gene",
"regulation",
"infectious",
"diseases/respiratory",
"infections"
]
| 2009 | The Structure of a Biologically Active Influenza Virus Ribonucleoprotein Complex |
Antibiotic-resistant infections kill approximately 23 , 000 people and cost $20 , 000 , 000 , 000 each year in the United States alone despite the widespread use of small-molecule antimicrobial combination therapy . Antibiotic combinations typically have an additive effect: the efficacy of the combination matches the sum of the efficacies of each antibiotic when used alone . Small molecules can also act synergistically when the efficacy of the combination is greater than the additive efficacy . However , synergistic combinations are rare and have been historically difficult to identify . High-throughput identification of synergistic pairs is limited by the scale of potential combinations: a modest collection of 1 , 000 small molecules involves 1 million pairwise combinations . Here , we describe a high-throughput method for rapid identification of synergistic small-molecule pairs , the overlap2 method ( O2M ) . O2M extracts patterns from chemical-genetic datasets , which are created when a collection of mutants is grown in the presence of hundreds of different small molecules , producing a precise set of phenotypes induced by each small molecule across the mutant set . The identification of mutants that show the same phenotype when treated with known synergistic molecules allows us to pinpoint additional molecule combinations that also act synergistically . As a proof of concept , we focus on combinations with the antibiotics trimethoprim and sulfamethizole , which had been standard treatment against urinary tract infections until widespread resistance decreased efficacy . Using O2M , we screened a library of 2 , 000 small molecules and identified several that synergize with the antibiotic trimethoprim and/or sulfamethizole . The most potent of these synergistic interactions is with the antiviral drug azidothymidine ( AZT ) . We then demonstrate that understanding the molecular mechanism underlying small-molecule synergistic interactions allows the rational design of additional combinations that bypass drug resistance . Trimethoprim and sulfamethizole are both folate biosynthesis inhibitors . We find that this activity disrupts nucleotide homeostasis , which blocks DNA replication in the presence of AZT . Building on these data , we show that other small molecules that disrupt nucleotide homeostasis through other mechanisms ( hydroxyurea and floxuridine ) also act synergistically with AZT . These novel combinations inhibit the growth and virulence of trimethoprim-resistant clinical Escherichia coli and Klebsiella pneumoniae isolates , suggesting that they may be able to be rapidly advanced into clinical use . In sum , we present a generalizable method to screen for novel synergistic combinations , to identify particular mechanisms resulting in synergy , and to use the mechanistic knowledge to rationally design new combinations that bypass drug resistance .
Small-molecule antimicrobial therapy facilitated one of the greatest increases in lifespan in history but is endangered by the rise of antimicrobial-resistant “superbugs” [1] . The CDC estimates that antibiotic-resistant bacteria cause more than 2 million infections and 23 , 000 deaths annually in the United States alone [2] . Combating antibiotic resistance requires a regular supply of new antimicrobial drugs , as bacteria inevitably acquire resistance to any single drug . Two main approaches are commonly used to identify additional antibiotics: new drug discovery and repurposing of drugs already approved for other conditions [3–6] . New drugs are more likely to result in breakthroughs but require a large upfront capital investment in time-consuming clinical trials . Repurposing can move drugs into the clinic without extensive trials but will not identify novel drug classes or structures [7] . This study explores a third strategy to combat antimicrobial resistance: synergistic combination therapy . Synergy occurs when 2 drugs act together with efficacy beyond the additive effect of each drug on its own [8] . Since synergistic drug pairs can kill microbes that are resistant to 1 drug in the pair [9] and are thought to slow the evolution of resistance [10 , 11] , they have generated considerable interest as a promising way to overcome antimicrobial drug resistance . Delays in the commencement of treatment of severe infections can dramatically increase mortality rates—for example , septic patients face an 8% increase in mortality for each hour’s delay [12] . Therefore , combinations of antimicrobials are commonly used prior to the identification of the causal organism . Most of the combinations currently employed are additive , but meta-analyses of clinical trials indicate better outcomes if synergistic combinations are used when the causal organism is unknown [12–15] . Molecules in additive combinations also frequently act against the same target or target pathways [16] and thus are potentially more susceptible to resistance-conferring mutation than combinations with different targets . However , few synergistic combinations have been identified [17] ( S1 Table ) , and high-throughput identification has been challenging due to the numbers involved: a collection of 1 , 000 molecules has 1 million potential pairwise combinations . We previously described a new approach to high-throughput identification of synergistic small-molecule pairs: the overlap2 method ( O2M ) [18] . O2M uses at least 1 known synergistic interaction to predict many additional interactions from large-scale chemical-genetics data . The rationale was that each small molecule in a synergistic pair produces a set of phenotypes—chemical-genetic signature—in a precise set of mutants that show reduced or enhanced growth in the presence of each molecule . When mutants exhibit the same phenotype when treated with known synergistic molecules , we predict that any molecule that induces the same phenotype from the same mutant will act synergistically with each original synergistic molecule . This was indeed the case , even when the known synergistic molecules have different mechanisms of action [18] . When we validated O2M on the pathogenic fungus Cryptococcus neoformans , we identified 36 new synergistic interactions with a low false positive rate ( 73% of the predictions were verified ) [18] . Since then , several other groups published methods using chemical-genetics datasets to identify synergistic small-molecule interactions [19–21] . Chemical-genetics datasets are widely available for a variety of pathogenic microbes , including Plasmodium falciparum [22 , 23] , Mycobacterium tuberculosis [24] , Candida albicans [25 , 26] , Candida glabrata [27] , and C . neoformans [18] . Thus , methods identifying synergistic drug interactions from chemical-genetics datasets are potentially broadly applicable . In this study , we show that O2M is also applicable to bacterial pathogens and antibiotics . Furthermore , we expand its utility with a novel high-throughput screening method for synergistic combinations and elucidate the molecular mechanism of a new drug combination . From this , we go on to rationally design synergistic combinations with different targets but the same phenotypic consequences , thus bypassing the original resistance mechanism . Indeed , our rationally designed synergistic combinations efficiently inhibit growth of clinical isolates resistant to the original antibiotic combination . In sum , we have developed an adaptable method for high-throughput screening for synergistic small-molecule pairs that facilitates rational design of synergistic small-molecule combinations , thereby addressing a key medical need in the treatment of drug-resistant infections .
We first demonstrate that our method for predicting synergistic interactions between small molecules , O2M , can be successfully applied to organisms from different kingdoms . We initially developed O2M for the fungal pathogen C . neoformans , but here we apply O2M to E . coli [28] with comparable success . In this section , we describe the initial analysis of the published E . coli chemical-genetics dataset [28] . In the next section , we show how information from analysis of this dataset allows high-throughput screening for synergistic molecule pairs . O2M requires a chemical-genetics dataset , generated when a library of knockout mutants is grown in the presence of >100 different chemicals . A quantitative growth score is calculated for each mutant/small-molecule combination . Growth scores can indicate either slower growth ( negative values ) or faster growth ( positive values ) compared to wild-type growth scores . The growth scores of all mutants when treated with each small molecule is that small molecule’s “chemical-genetic signature . ” O2M is based on the rationale that similarities between chemical-genetic signatures of a known synergistic pair contains information that is somehow indicative of synergy—and thus can be used to identify additional synergistic interactions ( Fig 1A and [18] ) . When we compare the chemical-genetic signatures of a pair of small molecules already known to act synergistically , we identified a subset of mutants with similar growth scores . We term this subset of mutants “putative synergy prediction mutants . ” We hypothesized that any molecule that elicited the same phenotypes in the same mutants as the known synergistic molecules would also act synergistically with each member of the known synergistic pair . We analyzed chemical-genetic signatures for the known synergistic antibiotic pair trimethoprim and sulfamethizole from the Nichols et al . E . coli chemical-genetics dataset [28] . This resource contains quantitative growth scores for over 4 , 000 E . coli knockout mutants grown under approximately 300 different conditions ( including different types of media and lysogeny broth [LB] medium containing small molecules ) . We looked for genes whose knockout mutants exhibit a significant ( |Z| > 2 . 5 ) growth score to both trimethoprim and sulfamethizole ( Fig 1B ) . Genes that are transcribed as a single unit ( according to EcoliWiki: http://ecoliwiki . net/colipedia/index . php/Welcome_to_EcoliWiki ) were binned together . We identified 4 elements common to both chemical-genetic signatures: ECK0963-68 , ECK1082-86 , ECK1710-13 , ECK1864-66 , and ECK3930 . Knockouts of these gene ( s ) are putative synergy prediction mutants . We then calculated if each putative synergy prediction mutant successfully identified trimethoprim synergizers . Again , using the Nichols et al . dataset [28] , we identified all small molecules that elicit a significant score ( |Z| > 2 . 5 ) from the 4 putative synergy prediction mutants . We also generated a list of negative control molecules that did not elicit a phenotype from any mutant in any response pattern gene . We performed checkerboard assays , a standard measure of synergistic interactions [29] , for each predicted synergizer or negative control combined with trimethoprim or sulfamethizole . All small molecules and their minimum inhibitory concentrations ( MICs ) are listed in Table 1 . A synergistic interaction is defined as at least a 4-fold decrease in the MIC of each drug in the pair , producing a fractional inhibitory concentration index ( FICI ) of ≤0 . 5 [30] . We found that small molecules that inhibit growth of putative synergy prediction mutant eck1864-66Δ operon are enriched for synergistic interactions with both trimethoprim ( p < 0 . 03 , Fisher’s exact test ) and sulfamethizole ( p < 0 . 05 , Fisher’s exact test ) relative to a randomly generated negative control small-molecule set ( Fig 1C & 1D ) . For example , 25% of predicted synergizers acted synergistically with trimethoprim , compared to 4% of the negative control set . None of the other synergistic response pattern operons identified synergistic interactions with trimethoprim or sulfamethizole at a higher rate than chance ( S1 Fig ) . In total , we identified 5 new synergistic interactions from analyzing the small molecules used in generation of the E . coli chemical-genetics dataset [28] . Azidothymidine ( AZT ) acted synergistically with both trimethoprim and sulfamethizole . Three molecules synergize only with trimethoprim and 1 only with sulfamethizole . Since trimethoprim interacted with more molecules , we prioritize it in subsequent experiments . The genes in the ECK1864-66 operon are involved in DNA synthesis , modification , and repair , which might explain why this operon , and not our other putative synergy prediction mutants , predicted synergy with trimethoprim . ECK1864 encodes an endonuclease that resolves Holliday junctions [31 , 32] . ECK1865 gene product does not have a known molecular function , but the mutant is sensitive to ionizing radiation [33] . ECK1866 encodes dihydroneopterin triphosphate pyrophosphohydrolase , an enzyme involved in the early stages of folate biosynthesis [34] . In contrast , the other putative synergy prediction genes , whose mutants did not enrich for trimethoprim synergizers , encoded gene products that did not function in pathways related to DNA synthesis , repair , or folate biosynthesis [35 , 36] . Notably , these experiments identified new synergistic partners for trimethoprim and sulfamethizole that are not currently used as antibiotics but are approved for human use in other indications . The antiviral drug AZT is promising because of its potent interaction with both trimethoprim and sulfamethizole . AZT has previously reported antibacterial activity but has not been shown to have any synergistic interactions with antibiotics [37–39] . AZT and several other newly identified synergizers are DNA-damaging agents , thereby suggesting a significantly different mechanism of action than the trimethoprim + sulfamethizole combination . In addition , we performed a similar analysis on vancomycin , which acts synergistically with cephalosporins [40 , 41] . When we tested the cephalosporins used in the Nichols et al . dataset with vancomycin for synergistic interactions , cefaclor acted synergistically with vancomycin in a checkerboard assay ( S2 Fig ) . We found only 1 gene/operon , ECK3247-48 , with a mutant that exhibited a significant growth score ( |Z| > 2 . 5 ) when grown in the presence of vancomycin and cefaclor . We then identified all small molecules from the Nichols et al . dataset [28] that induced a significant phenotype ( |Z| > 2 . 5 ) from eck3247Δ or eck3248Δ cells , predicting that these molecules would synergize with vancomycin . When we tested these in checkerboard assays for synergy ( S2 Fig ) , we identified 3 molecules that synergized with vancomycin: chelators EDTA and EGTA and aminocoumarin antibiotic novobiocin . Ion availability is known to effect pathogenicity [42] and vancomycin efficacy [43] , but we cannot find previous reports of an interaction between vancomycin and novobiocin , which inhibits DNA gyrase activity [44] . These data demonstrate that O2M identifies synergistic interactions for multiple antibiotics . We exploited our newly identified synergy prediction mutant ( eck1864-66Δ ) to perform one of the first high-throughput screens for synergistic pairs ( Fig 2A ) . Our rationale was that because synergy prediction mutants exhibited the same phenotypic response to molecules known to act synergistically , this limited set of knockout mutants could be used to rapidly screen additional small molecules to identify those that are likely to be synergistic with the starting molecule . Our assay is extremely simple and identifies synergistic pairs without performing multidrug assays . Instead , synergy prediction mutants functionally substitute for 1 of the antibiotics . We screened the Microsource Spectrum Collection , a small-molecule library of 2 , 000 compounds that is enriched for Food and Drug Administration ( FDA ) -approved drugs . We grew wild-type and eck1864-66Δ mutant cells in the presence of each small molecule , then identified small molecules that inhibit growth of the synergy response marker strain but not wild-type cells after 18 hours of growth ( Z score < −2 . 5 ) . We identified 28 of these putative trimethoprim-synergizing molecules ( Table 2 ) . We verified the synergistic interactions between trimethoprim and our screen hits using 2 different methods: checkerboard assays and Bliss Independence . Checkerboards are preferable but require that both small molecules inhibit microbial growth on their own . Of the 18 screen hits that met this criterion , 8 were verified to act synergistically with trimethoprim ( Fig 2B ) . This 44% enrichment rate is significantly ( p < 0 . 05 ) greater than the 4% frequency of trimethoprim synergizers in a randomly selected set of small molecules ( Fig 1 ) . Six of these small molecules ( phthalylsulfacetamide , phthalysulfathiazole , sulfabenzamide , sulfacetamide , sulfaphenazole , and sulfapyridine ) are sulfonamide antibiotics that inhibit dihydropteroate synthetase , the same target as sulfamethizole . A seventh , dapsone , also inhibits dihydropteroate synthetase but belongs to a different class of drugs [45] . The final verified screen hit , mitoxanthrone , is an antineoplastic DNA-intercalating agent [46] and not a sulfonamide antibiotic . The remaining 10 screen hits do not inhibit E . coli growth on their own , so we attempted to verify their synergistic action using the Bliss independence model [30] . Briefly , in a 96-well plate containing growth medium and bacteria , we created a gradient of trimethoprim , then added each small molecule of interest at both 10 μM and 100 μM concentrations . Synergistic small molecules enhance growth inhibition by trimethoprim at both concentrations versus trimethoprim alone . We found that 2 of 10 small molecules exhibit synergy at both concentrations ( Fig 2C ) . Sulfanitran is a sulfonamide antibiotic . Cyclosporine is a cyclic peptide that inhibits calcineurin [47] but is not known to have a bacterial target . The new synergistic molecules found from O2M analysis and the high-throughput screen fall into 2 main groups . First , the sulfonamide antibiotics almost certainly act by the same mechanism as trimethoprim + sulfamethizole , so strains resistant to the combination would likely also be resistant to these new pairs . The second group consists of several DNA damaging agents ( AZT , mitomycin C , mitaxanthrone ) . This result suggests a second molecular mechanism underlying synergy . Therefore , we focus on this second group in subsequent experiments . Our most promising new synergistic interaction is between trimethoprim or sulfamethizole and AZT . The first anti-HIV drug [48] , AZT , was investigated as a chemotherapeutic before the discovery of its antiretroviral activity [49] . AZT’s MIC against multidrug-resistant ( MDR ) E . coli and K . pneumoniae is in the nanogram per milliliter range ( Table 3 ) , suggesting that it could be a powerful antibiotic . AZT causes premature chain termination during bacterial DNA replication [38 , 39 , 50] , induces the SOS response [38 , 39] , and moderately increases mutation rates [51] . We tested whether AZT acts synergistically with trimethoprim in 12 MDR clinical isolates of E . coli and 5 MDR K . pneumoniae isolates . Seven of the E . coli isolates and 4 K . pneumoniae isolates are resistant to the trimethoprim/sulfamethizole combination . In the vast majority of trimethoprim/sulfamethizole-resistant isolates , the classic combination of trimethoprim and sulfamethizole no longer acted synergistically ( FICI ≤ 0 . 5 ) ( Fig 3A ) . One possible reason behind this is that trimethoprim and sulfamethiole targets are in the same pathway [52] , so resistance to 1 drug could confer some resistance to the other and block the synergistic interaction . In contrast , trimethoprim and AZT acted synergistically against 5 E . coli and 2 K . pneumoniae trimethoprim/sulfamethizole-resistant isolates . Our new synergistic pair thus acts against multiple species of trimethoprim/sulfamethizole-resistant bacteria . The mechanisms underlying synergistic interactions are poorly explored , with the exception of trimethoprim + sulfamethizole ( or other sulfamonamides ) . Both molecules are inhibitors of folate biosynthesis , so these drugs were historically thought to synergize due to simultaneous inhibition of 2 enzymes in the folate biosynthesis pathway [52] . The Nichols et al . chemical-genetic analysis suggests that trimethoprim and sulfonamides differentially impact the steps between tetrahydrofolate and 5 , 10-methylene tetrahydrofolate production ( Fig 1B ) [28] . Regardless , since DNA-damaging agents such as AZT do not inhibit folate biosynthesis , they likely act through a second mechanism of synergy . AZT alone induces the SOS response [38 , 39] , so we hypothesized that the synergistic pairing with trimethoprim could amplify each molecule’s individual activity . We performed checkerboard assays on K12 E . coli carrying a green fluorescent protein ( GFP ) reporter plasmid under control of the SOS-induced sulA promoter . We selected the sulA promoter because it is induced late in the SOS response , indicating a robust SOS response and cell growth arrest [53] . Neither trimethoprim nor sulfamethizole alone induces the sulA reporter compared to a no-drug control ( Fig 4A ) . AZT alone induced the sulA promoter modestly but reproducibly ( 1 . 6-fold relative to the control ) . These results predict that the combination of AZT and trimethoprim would show a 3-fold induction . Instead , we see a 9-fold induction ( p < 0 . 01; Mann-Whitney test ) . We observed the same trend for mitomycin C , a DNA crosslinking agent [54] that also synergizes with trimethoprim ( Fig 1 ) . We also tested 2 additional molecules as controls . The RNA polymerase inhibitor rifampicin , which blocks the SOS response [55] and does not synergize with trimethoprim ( Fig 2 ) , exhibited a lower SOS response in combination with trimethoprim than alone ( p < 0 . 005; Mann-Whitney test ) . Similarly , the DNA-damaging agent hydroxyurea ( HU ) [56] , which does not synergize with trimethoprim ( Fig 2 ) , also does not induce the sulA promoter alone or in combination with trimethoprim . Since the SOS response induces error-prone DNA repair , we hypothesized that the combination of trimethoprim and AZT increases mutation burden beyond that caused by each small molecule alone . To test this hypothesis , we performed a fluctuation assay to measure the mutation rate [57] . We grew cells in subinhibitory concentrations of each small molecule alone or in combination , then plated cells to LB + 15 μg/ml nalidixic acid , which selects for mutations in the topoisomerase gene [58 , 59] . We calculated mutation rate from the number of resistant colonies within the total population [60] . The trimethoprim synergizers AZT and mitomycin C both increase mutation rate by at least 3-fold in combination with trimethoprim but not alone , even at the subinhibitory concentrations tested ( 1/8 of MIC ) ( Fig 4B ) . Sulfamethizole alone or in combination with trimethoprim does not increase mutation rates . These data suggest that the amplification of DNA damage is an important step in the synergistic interaction between trimethoprim and AZT ( or mitomycin C ) , while trimethoprim and sulfamethizole interact through a different mechanism . Our data support the model that DNA damage accumulates in cells treated with trimethoprim + AZT . AZT’s connection to DNA damage is clear from its known mechanism of action [39] . However , trimethoprim’s connection is indirect . Folate is necessary for the biosynthesis of purines , thymidine [61] , and methionine [62] , and treatment with trimethoprim disrupts nucleotide homeostasis [61] . We hypothesized that such reduced nucleotide availability amplifies the phenotypic consequences of premature chain termination caused by AZT . Should the synergistic interaction between trimethoprim and AZT be due to the simultaneous inhibition of chain termination and depleted nucleotide pools , then we would expect additional , unrelated small molecules that cause similar effects to also interact synergistically . Furthermore , genetic mutations that deplete nucleotide pools would result in increased sensitivity to AZT , and other chain-terminating agents would cause increased sensitivity to trimethoprim . First , we tested the proposed nucleotide homeostasis/DNA chain termination interaction chemically . We performed a checkerboard assay with AZT and HU or floxuridine , 2 FDA-approved drugs that alter deoxynucleotide triphosphate ( dNTP ) balance [63] . Both HU and floxuridine acted synergistically with AZT but not trimethoprim ( Fig 5A ) . We then tested trimethoprim in combination with nucleoside analogs other than AZT . As expected , nucleoside analogs that inhibit E . coli growth interact synergistically with trimethoprim but not AZT ( Fig 5B ) . Second , this interaction between nucleotide pools and AZT also occurs genetically . The E . coli gene deoxycytidine deaminase ( dcd ) encodes deoxycytidine triphosphate ( dCTP ) deaminase [64] . dcd deletion mutants exhibit depleted deoxythymidine triphosphate ( dTTP ) pools and elevated dCTP pools when grown in minimal medium [64–66] . dcd deletion mutant cells exhibit a 32-fold increase in AZT sensitivity compared to wild-type cells but no change in trimethoprim or rifampicin sensitivity ( Fig 5C ) . By contrast , mutants in nucleoside diphosphate kinase ( ndk ) [67] exhibit elevated dCTP and dTTP pools but lower deoxyadenosine triphosphate ( dATP ) pools [66 , 68] . ndk deletion mutant cells exhibit a 4-fold decrease in AZT sensitivity ( increased resistance ) compared to wild-type cells . These results suggest that adequate dTTP is necessary for surviving AZT exposure . Therefore , we conclude that simultaneous disruption of nucleotide homeostasis and DNA replication increase E . coli growth inhibition . We next tested if DNA repair mutants in general are hypersensitive to AZT or if the hypersensitivity is specific to disrupted nucleotide homeostasis . We grew cells deficient in mismatch repair ( mutL , mutS , or mutH deletion mutants ) or nucleotide excision repair ( uvrA , uvrB , or uvrC deletion mutants ) in AZT , trimethoprim , and rifampicin and did not observe increased sensitivity to any of these ( Fig 5C ) . Finally , to make sure that the interaction between AZT and trimethoprim was not due to methionine depletion , we tested a deletion mutant in the methionine synthase gene metH [62] . metH mutant cells exhibited wild-type levels sensitivities to AZT , trimethoprim , and rifampicin ( Fig 5C ) . Therefore , we conclude that disruption of nucleotide homeostasis ( by multiple possible mechanisms ) hypersensitizes bacterial cells to DNA damage caused by AZT . By substituting HU or floxuridine for trimethoprim , we demonstrated that we can rationally design additional synergistic pairs once the molecular mechanism underlying an interaction is understood . Finally , we tested our new , rationally designed synergistic pairs against our collection of clinical E . coli and K . pneumoniae strains . Substituting either HU ( Fig 6A ) or floxuridine ( Fig 6B ) for trimethoprim , we found that either molecule combined with AZT inhibited growth of trimethoprim/sulfamethizole-resistant isolates . HU + AZT acted synergistically in 15 of 17 clinical isolates , including all but 1 of the trimethoprim/sulfamethizole-resistant isolates ( n = 11 ) . Floxuridine + AZT also acted synergistically in most clinical isolates ( 12 of 17 ) , including all but 2 trimethoprim/sulfamethizole-resistant isolates . The MICs of floxuridine and AZT are sub-μg/ml for most MDR clinical isolates , whereas MICs for trimethoprim or sulfamethizole were up to 5 , 000 μg/ml in vitro ( Table 3 ) . Rationally designed synergistic pairs therefore bypassed resistance in clinical isolates . We then tested the floxuridine + AZT combination in a zebrafish infection model . Efficacy of synergistic drug pairs has historically been difficult to evaluate in vertebrate systems , and many prior studies use the moth larvae Galleria mellonella or perform only in vitro tests [19 , 20 , 69 , 70] . Zebrafish offer several advantages: they have a mammalian-like innate immune system [71] , a long history as microbial infection models [71–75] , and are a good platform for assessing drug toxicity [76 , 77] . We injected approximately 2 , 500 colony-forming units ( CFU ) of trimethoprim-resistant E . coli into the pericardial cavity of zebrafish embryos [75 , 78] , incubated for 3 hours , then treated with either the original drug combination ( trimethoprim + sulfamethizole ) or our rationally designed combination ( floxuridine + AZT ) . We analyzed bacterial burden at 24 hours postinoculation ( hpi ) at dosages analogous to human dosage ( see Materials and methods and Table 4 ) [79 , 80] . The new combination was indeed successful—it resulted in a 10 , 000-fold reduction in median bacterial burden in infected zebrafish embryos treated with floxuridine + AZT compared to infected embryos treated with trimethoprim + sulfamethizole ( Fig 7A ) . When we infected embryos with a trimethoprim/sulfamethizole-sensitive E . coli strain , floxuridine + AZT treatment and trimethoprim + sulfamethizole treatment were equally effective . ( Fig 7B ) . We also analyzed MICs in the presence of human serum , as a substantial increase in MIC in the presence of serum would indicate that our small molecules of interest are binding to serum proteins and are not bioavailable [81] . We tested MICs of floxuridine , AZT , trimethoprim , and sulfamethizole with and without 20% human serum and found only minor changes in MIC ( S14 Table ) . The dosages we use in zebrafish are well under the human dosage ( Table 4 ) , suggesting that it would be possible to obtain the necessary drug concentrations in humans .
Synergistic small molecules are of considerable clinical interest , but systematic identification has been challenging . This study describes 4 significant advances to such systematic identification . First , we demonstrate that O2M is generally applicable beyond the fungal pathogen for which it was originally developed [18] . Second , we and others show that previously published chemical-genetic datasets [18 , 22–26 , 28 , 82] can be successfully used as the raw input for O2M , significantly decreasing the upfront investment required . Third , O2M identifies knockout mutants that can be used as readouts for synergy in highly scalable screening assays for additional synergistic combinations . Finally , understanding the mechanisms that underlie synergistic interactions can facilitate the rational design of further synergistic combinations that bypass antibiotic resistance . Our results show that a wide variety of synergistic combinations are available if we know how to search for them . We would suggest that these discoveries represent a small fraction of the potential synergistic combinations . In support of this idea , several groups recently published analysis methods that , like our original O2M analysis [18] , use chemical-genetics data to predict synergy between antibiotics [19] or antifungals [20 , 21 , 70] . Notably , each method identifies different , complementary synergistic pairs , suggesting that current methods are far from identifying all synergistic interactions . As described here , the particular advantage of O2M is its scalability to screen for synergistic small molecules that are not commonly used as antibiotics . This scalability is critical to keeping the initial screens as broad as possible—since substituting 1 member of a synergistic pair for a second member can change the molecular mechanism underlying the interaction , identification of diverse synergistic pairs offers the greatest potential for further rational design . We screened a well-studied collection of FDA-approved small molecules , identifying 14 novel synergistic combinations with the widely used antibiotic trimethoprim . O2M is also much faster than a pairwise screen of a small-molecule library . By identifying synergy prediction mutants , we can screen 2 , 000 molecules and verify only those predicted to be synergistic . The 28 predicted synergistic molecules identified in our trimethoprim/sulfamethizole screen took less than a week to validate . Testing of the entire 2 , 000 small-molecule collection in combination with a single molecule would take months . The molecular mechanisms underlying synergistic drug interactions are generally poorly understood [4 , 28] . There are 3 main hypotheses for why any given pair of small molecules exhibit synergistic interaction: that the pairs ( 1 ) act together to cause a third , novel inhibitory activity ( “gain-of-function” hypothesis ) , ( 2 ) act in combination by simultaneously inhibiting 2 different functions to increase potency ( “two-hit” or “parallel pathway” hypothesis ) [83] , or ( 3 ) 1 drug increases the activity and/or bioavailability of the other ( “bioavailability” hypothesis ) [84] . Our data demonstrate that the trimethoprim and AZT interaction likely represent a “two-hit” mechanism , acting through the combined induction of DNA damage and blocking DNA repair by disrupting nucleotide homeostasis ( Fig 8 ) . This molecular mechanism quite likely differs from that historically thought to underlie the trimethoprim/sulfonamides [52] . Recent data suggest that the trimethoprim + sulfamethizole interaction is not as simple as simultaneous inhibition of 2 folate biosynthesis enzymes [52] . Instead , the hypothesis is that trimethoprim and sulfonamides result in buildup of different secondary metabolites , which differentially impact enzyme activities [28] . However , since the phenotypic consequences of treatment with trimethoprim + sulfamethizole differ from treatment with trimethoprim + AZT , the mechanisms are likely also different . We surmise that both combinations represent “two-hit” synergistic interactions . “Two-hit” synergistic interactions can be predicted from network analysis: genes/pathways whose knockouts exhibit synthetic lethality could be good targets of drug combinations . These analyses also demonstrate the importance of network analysis and high-throughput studies on model organisms [85]: the chemical-genomic dataset we used for our initial O2M analysis was from a nonpathogenic K12 genetic background [28] , showing that K12 data are sufficient to elucidate synergistic drug mechanisms . One potential concern about using trimethoprim + AZT or floxuridine + AZT clinically is that these combinations increase mutation rate . While this is indeed a concern , many antibiotics target DNA replication and other processes that also increase mutation rates . Our measured mutation rate for trimethoprim + AZT , approximately 2 x 10−8 mutations per cell , is well within the range of other antibiotics [86] . These vary from 10−9 mutations per cell ( clarithromycin and amoxicillin ) [86] to 10−6 mutations per cell [87] . Mutation rates will vary with species and strain [87 , 88] , but trimethoprim + AZT induces mutation rates comparable to those induced by ciprofloxacin [86] . Once we identify biological pathways and processes whose simultaneous inhibition blocks microbial growth , we can then rationally design synergistic drug treatments that bypass antibiotic resistance . Here , we present a proof-of-principle methodology that demonstrates the power of this rational design . Once we identified the molecular mechanism underlying the trimethoprim + AZT interaction , we substituted trimethoprim for another FDA-approved small molecule . This newly designed combination , floxuridine + AZT , achieved the same synergistic interaction with AZT yet bypassed trimethoprim resistance in MDR clinical isolates ( Figs 6 and 7 ) . Indeed , floxuridine + AZT was far superior to trimethoprim + sulfamethizole in a vertebrate infection model with trimethoprim/sulfamethizole-resistant E . coli . That is , we inhibited MDR E . coli infection with lower doses of structurally unrelated but functionally similar small molecules . The optimal clinical application of interacting small molecules is currently under debate . Recent work suggested that sequential , rather than simultaneous , application of synergistic small molecules prevents the development of drug resistance [8 , 89] . Others suggest that antagonistic small-molecule interactions could be beneficial [8 , 83 , 90] . Antagonistic interactions occur when 2 molecules in combination decrease each other’s efficacy . One theory is that antagonism decreases the selective advantage of a drug-resistant mutation , and thus evolution of resistance is slower to an antagonistic pair than a synergistic pair [90] . These ideas merit further exploration in clinical and animal models of infection . In sum , O2M is an important tool for high-throughput identification of synergistic small-molecule pairs and successfully identifies new treatments to combat MDR infections . With the growing antibiotic crisis , treatments that are effective against MDR bacteria need to move rapidly into the clinic . Our method of screening FDA-approved drugs identified candidate treatments that could be deployed with fewer regulatory trials than needed for new drugs . Moreover , our rationally designed treatment , floxuridine + AZT , also uses FDA-approved agents ( and AZT is well tolerated for short-term treatments , despite toxicities associated with long-term use at high doses [91] ) . We hope to spur interest among clinicians to test such designed synergistic combinations against difficult-to-treat MDR infections , when appropriate .
Animals used in this study were handled in accordance with protocols approved by the University of Utah IACUC committee ( protocol 10–02014 ) , which follow guidelines from the Guide for the Care and Use of Laboratory Animals and zfin . org . Zebrafish older than 3 days postfertilization were euthanized by immersion in a chilled water bath followed by mechanical disruption . Zebrafish younger than 3 days , which do not have developed pain sensors , were euthanized by mechanical disruption . Infections took place under tricaine anesthesia . Unless otherwise stated , experiments were performed on E . coli K12 strain MG1655 . E . coli blood isolates #1–8 ( referred to as BEC1 , BEC2 , etc . ) are described in Barber et al . [78] . Urinary tract infection ( UTI ) isolates and K . pneumoniae isolates were obtained from ARUP labs . Strains are listed in S15 Table . All assays were performed in M9 minimal medium ( 10 . 5g/L M9 broth [Amresco] , 0 . 2% casamino acids , 0 . 1M CaCl2 , 0 . 4% glucose , 1M MgSO4 , 0 . 25% nicotinic acid , 0 . 33% thiamine in H2O ) unless otherwise stated . To determine MICs , an M9 culture of MG1655 was growth overnight at 37°C with shaking , then diluted to OD600 = 0 . 002 . We then inoculated each well with approximately 1 , 000 cells ( 2 μl of culture into 200 μl of medium per well ) . Plates were incubated at 37°C unless otherwise stated . Small-molecule gradients were diluted in 2-fold dilution series unless otherwise stated . MIC values ( Table 1 ) are calculated following 24 hours incubation at 37°C . MIC values are calculated as >90% growth inhibition unless otherwise stated . MIC values for MDR strains , which grow more rapidly than lab strains , were calculated following 24 hours incubation at 37°C . When calculating MICs in the presence of human serum , we used standard techniques but substituted up to 20% of the media volume with human serum ( Sigma ) . We performed O2M analysis as previously described [18] with the following variations . ( Step-by-step instructions are available in the S1 Text section ) . The Nichols et al . paper calculates a growth score for each mutant + small molecule combination [28] . We then use these growth scores to calculate significance in O2M analysis . We considered any growth score significant if either: ( 1 ) growth score for mutant A and small molecule B ≥ average growth score for all mutants when grown on plates containing small molecule B + 2 . 5*standard deviation of all small molecule B growth scores , or ( 2 ) growth score for mutant A and small molecule B ≤ average growth score for all mutants when grown on plates containing small molecule B − 2 . 5*standard deviation of all small molecule B growth scores . This corresponds to a Z-score cutoff value of +/−2 . 5 . We identified any gene whose knockout exhibited a significant score when exposed to trimethoprim or sulfamethizole in the Nichols et al . dataset [28] . We then identified genes whose knockouts responded significantly across the majority of concentrations of both drugs . If these genes are transcribed as part of a polycistronic RNA , then a phenotype was considered significant if any mutant in that operon met the Z score requirement ( |Z| > 2 . 5 ) . For example , if eck1864Δ was significant at 1 trimethoprim concentration and eck1865Δ at another , the entire ECK1864-66 operon was considered a significant hit at both those concentrations . For trimethoprim + sulfamethizole , this method identified 5 potential synergy prediction mutants/operons . We tested all small molecules predicted as synergistic for each of these synergy prediction mutants/operons , then calculated enrichment for successful predictions using a Fisher’s exact test . We also tested additional Z scores , |Z| > 1 . 96 and |Z| > 3 . 0 , with no change in end result . |Z| > 3 . 0 identified ECK1864-66 , ECK0964 , and ECK1710-13 as putative synergy prediction mutants . As shown in Figs 1 and S1 , only ECK1864-66 enriched for small molecules that synergize with trimethoprim . At |Z| > 1 . 96 , ECK4132-33 , ECK1189 , and ECK2901-04 were also identified as putative synergy prediction mutants . None of these mutants enriched for synergistic interactions with trimethoprim ( S5 Fig ) . We followed the same method as Hsieh et al . [92] with minor modifications . For MG1655 , starting inoculation was 2 μl of an OD600 = 0 . 02 or 0 . 002 ( 10 , 000 or 1 , 000 cells per well of 200 μl medium , respectively ) . Any synergistic interaction was verified at both inoculation levels but the initial MG1655 screen using 10 , 000 cells per well . FICI assays for the MG1655 strain were incubated for 24 hours at 37°C . FICI assays of clinical strains were incubated for 12 hours at 37°C using an inoculum of 1 , 000 cells . OD600 was read on a BioTek plate reader model Synergy H1 . Growth inhibition ≥90% compared to the no-drug control was considered significant . E . coli mutant eck1864-66::kanR was made in MG1655 by deleting the candidate gene or operon using the 1-step gene deletion method [93] . The putative knockout clones were confirmed by verification PCR with primers outside the deletion region . We amplified the eck1864-66 knockout cassette with the following primers ( gene-specific region in bold , kanamycin-specific region in regular typeface ) : eck1866KO F: GTGAAGGATAAAGTGTATAAGCGTCCCGTTTCGATCTTAGTGGTCATCTATGTGTAGGCTGGAGCTGCTTCG eck1864KO R: TTAACGCAGTCGCCCTCTCGCCAGGTTCAGCCGCGATTCGCTCATCTGCATCCATATGAATATCCTCCTTAG Colonies were selected on kanamycin . The clones were colony purified before PCR confirmation , then verified by PCR with the following primers: ECK1864Ver F- CGACTCTCTGATGAGGCCTG ECK1866Ver R-CCATTTACTATGACCTGCCA We inoculated either MG1655 wild-type or eck1864-66Δ cells at 1 , 000 cells per well ( 200 μl volume ) , then added either a vehicle control ( DMSO ) or small molecule to a final concentration of 10 μM per well . Plates were incubated for 18 hours at 37°C , then OD600 was measured on a plate reader . Small molecules that inhibited growth by more than 2 . 5-fold of the standard deviation of growth within each plate were considered significant . The plasmid with either GFP expressed under control of the sulA promoter or a promoterless plasmid containing only the GFP gene was transformed into MG1655 [94] . Strains were then grown to mid-log in MG1655 ( OD600 approximately 0 . 4 ) , then subcultured to OD600 to inoculate the experiment ( 2 ul into 200 μl M9 medium per well ) . We then performed a standard checkerboard assay , reading GFP signal ( 485 nm excitation and 528 nm emission ) and OD600 at 0 , 6 , and 24 hours . The 24-hour timepoint determined the MIC90 and the difference between the 0- and 6-hour GFP signals determined the promoter activation . If a small molecule hit from the Microsource Spectrum library screen did not inhibit E . coli growth alone , we evaluated its synergistic interaction with trimethoprim by Bliss Independence [30] instead of checkerboard analysis . Briefly , we created a gradient of trimethoprim from 4 to 62 . 5 μg/ml , then added small molecules of interest at 10 μM or 100 μM . We calculated percent growth , then determined Bliss Independence by determining whether the growth inhibition caused by each small molecule alone is equal to the inhibition caused by combination treatment . If inhibition is greater in the combination , then the molecules act synergistically . Strains ( S15 Table ) were grown in M9 overnight , then diluted to an OD66 of 0 . 002 and inoculated into 96-well plates at the same density used for MIC and FICI assays . Each 96-well plate contained a gradient of either AZT , trimethoprim , or rifampicin , serially diluted in 2-fold increments . After 24 hours at 37°C , plates were measured in a BioTek Synergy H1 . Each well was then normalized to the no-drug control for each mutant . Percent growth relative to this control is shown in the heat map . Each assay was repeated 3 times and the data averaged . All zebrafish husbandry and experimental procedures were performed in accordance with the University of Utah and IACUC-approved protocols . Wildtype AB* zebrafish were maintained as breeding colonies on a 14-hour/10-hour light/dark cycle . Embryos were collected as mixed egg clutches and raised at 28 . 5°C in E3 medium ( 5 mM NaCl , 0 . 27 mM KCl , 0 . 4 mM CaCl2 , 0 . 16 mM MgSO4; pH 7 . 4 ) containing 0 . 000016% methylene blue as an antifungal agent . Embryos were anesthetized at 2 days post fertilization ( dpf ) with tricaine ( 0 . 77 mM ethyl 3-aminobenzoate methanesulfonate salt [Sigma-Aldrich] ) , embedded in 0 . 8% low-melt agarose without tricaine , and supplemented with E3 media lacking methylene blue . A bacteria culture of BEC8 or F11 was grown at 37°C overnight in 12 ml M9 minimal media . Prior to injection , 1 mL of culture at OD600 = 2 . 5 to 3 . 5 for BEC8 or OD600 = 1 . 7 was created . Filtered green food dye was added to the culture in a 1:10 dilution . 1 nL was injected into the pericardial cavity of embryos using an Olympus SZ61 stereomicroscope together with a YOU-1 micromanipulator ( Narishige ) , a Narishige IM-200 microinjector , and a JUN-AIR model 3-compressor . Embryos were left to incubate at 28 . 5°C for 3 hours . Small molecules were mixed with green food dye , and 1 nL was injected into the yolk of embryos at 3 hpi . Dosages are listed in Table 4 . ( We estimated embryo mass based on Stehr et al . [95] ) . Additionally , drugs were supplemented to water 12 hours after yolk injections ( 15 hpi ) . Any embryos that were physically damaged during the procedures were discarded and excluded from further analysis . Embryos were unembedded and placed individually in a 96-well plate and left to incubate in 0 . 03% Instant Ocean at 28 . 5°C until we determined bacterial burden at 24 hpi for BEC8 or 19 hpi for F11 . These timepoints result in comparable bacterial burden in the vehicle-treated embryos . Bacterial inoculation levels are shown in S4 Fig . Embryos were euthanized at 24 ( BEC8 ) or 19 ( F11 ) hpi , then homogenized in 500 μL of PBS using a mechanical PRO 250 homogenizer ( PRO Scientific ) . Homogenates were serially diluted and plated on LB agar plates ( F11 ) or LB agar plates containing ampicillin ( BEC8 ) and incubated overnight at 37°C . Any embryos dead and decaying by the euthanasia timepoint were excluded from further analysis , as survival curves show that dead embryos were rare at these timepoints [78] and were evenly distributed across treatment groups . | Antibiotic resistance is a growing problem that threatens our ability to treat systemic bacterial infections . One strategy to combat antibiotic resistance is the use of synergistic antibiotic pairs that , when combined , have activity that is considerably greater than the sum of each individual drug’s activity on its own . Synergistic combinations can even inhibit the growth of bacteria that are resistant to the individual treatment drugs . However , synergistic pairs are rare and difficult to identify . High-throughput identification of synergistic pairs is challenging due to scale: 1 million different pairs are possible for a relatively small collection of 1 , 000 small molecules . Here , we describe a high-throughput method for rapid identification of synergistic small-molecule pairs , termed the overlap2 method ( O2M ) , that dramatically speeds up the screening process . First , we identify mutants that show the same phenotype when treated with each individual molecule in a synergistic pair , then use this information to guide screens for additional synergistic pairs . As a proof of concept , we studied the synergistic antibiotic pair trimethoprim and sulfamethizole , and we identified several additional synergistic molecules . Among these is the antiviral drug azidothymidine ( AZT ) , which blocks bacterial DNA replication . Trimethoprim and sulfamethizole both inhibit folate biosynthesis , which is necessary for the proper synthesis of nucleotides for DNA replication and repair . We found that reduced nucleotide levels sensitize E . coli cells to AZT . When we substitute trimethoprim with other small molecules that also reduce nucleotide levels , we find that these small molecules also act synergistically with AZT . Indeed , AZT in combination with trimethoprim substitutes inhibits the growth of trimethoprim-resistant clinical isolates more potently than trimethoprim and AZT or trimethoprim and sulfamethizole . This work demonstrates that when we resolve the pathways that underlie synergistic interactions , we can then identify additional small molecules that act by similar mechanisms , providing a means to bypass antibiotic resistance . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"biotechnology",
"antimicrobials",
"medicine",
"and",
"health",
"sciences",
"chemical",
"compounds",
"small",
"molecules",
"drugs",
"microbiology",
"vertebrates",
"organic",
"compounds",
"animals",
"animal",
"models",
"osteichthyes",
"bacterial",
"diseases",
"developmental",
"biology",
"antibiotic",
"resistance",
"model",
"organisms",
"antibiotics",
"experimental",
"organism",
"systems",
"pharmacology",
"embryos",
"research",
"and",
"analysis",
"methods",
"embryology",
"infectious",
"diseases",
"antimicrobial",
"resistance",
"fishes",
"escherichia",
"coli",
"infections",
"chemistry",
"biochemistry",
"zebrafish",
"organic",
"chemistry",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"biosynthesis",
"physical",
"sciences",
"organisms"
]
| 2017 | High-throughput identification and rational design of synergistic small-molecule pairs for combating and bypassing antibiotic resistance |
In the past two decades some fundamental properties of cortical connectivity have been discovered: small-world structure , pronounced hierarchical and modular organisation , and strong core and rich-club structures . A common assumption when interpreting results of this kind is that the observed structural properties are present to enable the brain's function . However , the brain is also embedded into the limited space of the skull and its wiring has associated developmental and metabolic costs . These basic physical and economic aspects place separate , often conflicting , constraints on the brain's connectivity , which must be characterized in order to understand the true relationship between brain structure and function . To address this challenge , here we ask which , and to what extent , aspects of the structural organisation of the brain are conserved if we preserve specific spatial and topological properties of the brain but otherwise randomise its connectivity . We perform a comparative analysis of a connectivity map of the cortical connectome both on high- and low-resolutions utilising three different types of surrogate networks: spatially unconstrained ( ‘random’ ) , connection length preserving ( ‘spatial’ ) , and connection length optimised ( ‘reduced’ ) surrogates . We find that unconstrained randomisation markedly diminishes all investigated architectural properties of cortical connectivity . By contrast , spatial and reduced surrogates largely preserve most properties and , interestingly , often more so in the reduced surrogates . Specifically , our results suggest that the cortical network is less tightly integrated than its spatial constraints would allow , but more strongly segregated than its spatial constraints would necessitate . We additionally find that hierarchical organisation and rich-club structure of the cortical connectivity are largely preserved in spatial and reduced surrogates and hence may be partially attributable to cortical wiring constraints . In contrast , the high modularity and strong s-core of the high-resolution cortical network are significantly stronger than in the surrogates , underlining their potential functional relevance in the brain .
The physical brain is a network of extraordinary complexity on multiple spatial scales . On the macroscopic scale , regions are connected by a large number of white-matter projections that form an intricate system: the connectome [1] . Understanding the principles of the large-scale architecture of the brain , how this architecture shapes brain dynamics to in turn support brain function and human behaviour , is a central challenge for contemporary neuroscience [2] , [3] . Recent advances in non-invasive anatomical [4]–[6] and functional [7] imaging techniques , along with the development of automated , high throughput post-processing methods [8] now allow the application of complex network science as a principled and systematic framework for studying the connectome [2] , [3] . Accordingly , numerous principles of organisation in the large-scale structural anatomy of the brain have been characterized , including small-world properties [9] , hierarchical architecture [10] , modular structure [11] , the existence of a strong structural core [12] and a so-called ‘rich-club’ organisation [13] . Exposing both the structural origin and functional relevance of these properties of the human connectome is an essential , but difficult step towards a deeper understanding of the large-scale organisation of the brain . A common approach to evaluating the significance of a particular network property , observed in a particular network , is via surrogate or null-hypothesis comparison [14] , [15] . In this approach , a set of surrogate networks represents a null-hypothesis for the target network property by preserving some a priori chosen properties of the network under investigation , while randomizing other network properties . Quantitative comparison of the original network with the ensemble of surrogate networks allows drawing conclusions on the significance of the target property of the network with respect to those properties preserved in the ensemble . Therefore , in its essence , surrogate network comparison allows testing if some , usually very elementary , properties of the target network induce , or at least contribute to , the expression of some of its more global and complex network properties . When choosing appropriate surrogate networks , the most widely used null-hypothesis properties are size ( number of nodes ) , connection density ( number of edges ) and degree distribution ( the number of connections of each node ) . This approach – which we term the ‘random surrogate’ approach – has illuminated the topological investigation of many abstract , spatially-unembedded networks , including the World Wide Web , semantic networks , food-webs , and gene-regulatory and metabolic networks [14] , [16] . It is also routinely applied in the analysis of brain networks in order to demonstrate that global , ‘higher order’ network property of brain maps , such as modularity or ‘small-worldness’ , cannot be attributed solely to these basic network properties [10] , [11] , [17] . Physical networks like the brain are , however , embedded into physical space and are therefore subject to additional constraints deriving from the costs of developing and maintaining connections [18] which are not conserved by random surrogates . Random surrogates therefore represent a rather loosely constrained null-hypothesis set for physical networks . Specifically , they tend to possess a large number of long-range connections because they ‘smooth’ local inhomogeneities of physical networks . They thus form highly and rather homogeneously integrated networks , while at the same time lacking the high topological segregation ( locally dense , globally sparse inter-connectivity ) associated with predominantly local connectivity , which is one of the most prominent features of brain networks [19] . When compared against random surrogates , then , certain properties of brain networks may appear to be highly distinctive even though they can be attributed to the spatial constraints of its embedding into the physical world ( wiring cost ) and/or of the economic pressure of minimising the number of the energetically expensive long-range connections ( metabolic cost ) [18] . To address this problem , so-called ‘lattice surrogates’ have been introduced [15] , [20]–[22] to preserve ( or rather increase ) the high segregation of brain networks . The motivation behind lattice surrogates , originating from the Watts–Strogatz notion of ‘small-worldness’ [23] , was to represent a lattice-like , topologically over-segregated ( and thus under-integrated ) surrogate network type , the opposite of random surrogates in a sense , and to compare the target network with these two extremes . This is reflected in the rule commonly used to generate lattice surrogates from the connectivity of a brain network ( during a ‘random’ network rewiring process , edge swaps are only made if the nonzero entries of the resulting connectivity matrix are located closer to the main diagonal [15] , [20] ) , which is only indirectly linked to physical distance through some arbitrary spatial ordering of the network nodes . For this reason , lattice surrogates are only partially appropriate as a null-hypothesis network set for physical wiring constraints of brain networks . Furthermore , lattice surrogates are designed to reduce , rather than preserve , network connection lengths thus further undermining their utility in assessing the effects of wiring constraints on cortical network properties . In this paper , we introduce two new classes of surrogates , spatial surrogates and reduced surrogates , Like random surrogates , spatial surrogates preserve network size , connection density , and degree distribution , but ( unlike random surrogates ) they also preserve the wiring length distribution of the target network . Reduced surrogates are like spatial surrogates with the difference that they do not preserve but actually reduce overall network wiring , in similar way to traditional lattice surrogates , but in a spatially well-defined and controlled manner . We reasoned that in virtue of these properties , these surrogates provide improved baselines by which to assess the extent to which a target network property can be attributed to cortical wiring constraints [18] . This approach enables us to evaluate a number of prominent findings regarding the structural properties of the connectome ( see Figure 1 ) with respect to the extent to which these properties are preserved in the novel spatial surrogates as compared to random and connection length optimised ( reduced ) surrogates . To ensure robustness we perform these analyses on both weighted and unweighted ( binary ) , and on the full resolution ( 998 regions ) as well as on a lowered resolution ( 66 regions ) version of the cortical structural connectivity data set provided by Hagmann et al . [12] . Overall , the method allows us to distinguish those significant network properties of the connectome that are derivable from its predominantly local , spatially segregated connectivity ( as indicated when both the cortical network and its spatial and reduced surrogates differ from random surrogates ) from those that are the consequences of some other , primarily not ( or not only ) spatial , but potentially more functionally relevant organisation principle of cortical connectivity ( as indicated when the cortical network differs from all of its surrogate groups ) . Specifically , during the evaluation of each specific network property , the logic of our surrogate analysis is the following ( see Table 1 ) . We measure the expression of the network property in the cortical network and every surrogate group by an appropriate complex network metric . If all surrogate groups exhibit similar metric indices to that of the cortex , then the basic network properties preserved in all surrogates ( the number of regions , number of white-matter projections and regional degree distribution of the cortical network ) appear to be sufficient for the observed expression of the investigated network property . If , however , all spatially constrained networks ( cortical network , spatial and reduced surrogates ) exhibit similar values , but differ from random surrogates , we reason that cortical wiring constraints may account for the level of expression of that network property in the cortical network . Additionally , if the cortical network is more similar to spatial than to reduced surrogates , we reason that solely the presence of long-range connections in the cortex may facilitate the network property , irrespectively of the specific arrangement of these connections in the cortex . If , however , the cortical network is more similar to reduced than to spatial surrogates , then we reason that the predominantly local ( short-range ) connectivity of the cortex can account for the expression of the network property even in the absence of long-range cortical connections ( as indicated by the similarity between the cortical network and reduced surrogates ) . In addition , this case also indicates that the particular arrangement of long-range cortical connections appears to be such that it does not interfere with ( strengthen or hinder ) the expression of the network property ( as indicated by spatial surrogates , with randomised long-range connections , being different from both the cortical network and reduced surrogates ) . Finally , if the cortical network differs from every surrogate ensemble , we reason that the network property is specific to the particular connectivity of the cortex , it cannot fully be attributed to the topological properties and wiring constraints that are conserved in the surrogates , but instead may be a more functionally relevant organisation feature of cortical connectivity .
We use the cortical connectivity network of Hagmann et al . [12] ( Figure 2 ) . This data was obtained by non-invasive tracing of white-matter projections linking pairs of cortical sites in the brains of five human subjects , combining magnetic resonance imaging ( MRI ) and diffusion spectrum imaging ( DSI ) techniques , semi-automated brain parcellation , diffusion tractography and appropriate post-processing methods . The individual connectivity networks of the five subjects were aggregated into a single network in order to reduce the impact of inter-subject variability . The resulting dataset is a compact network representation of cortical grey matter regions as network nodes , and their connecting white-matter fibre bundles as edges . For a detailed description of the acquisition procedure and validation test results of the procedure , see the original paper and [8] . By the nature of its processing pipeline , the network consists of a two-level hierarchical parcellation of the cortex: it is composed of 66 anatomical regions at the higher level , and of 998 regions of interest ( ROIs ) at the lower level . Each node on the level of ROIs represents an area of the cortical surface of approximately 1 . 5 cm2 size ( region ) , and there are a total of 17 , 865 undirected weighted connections between these regions . These figures result in a fairly sparse , 3 . 6% connection density network on the high-resolution cortical parcellation ( i . e . , on the lower hierarchical level of the segmentation ) . For the low-resolution network , similarly to [23] , we calculate the strength of the connection between every two-region pair by summing the weights of all the high-resolution connections linking the ROIs that compose the two cortical regions . This method results in 574 aggregated white-matter fibre bundles between the 66 regions on the low-resolution parcellation , which increases the connection density of the low-resolution cortical connectivity to 26 . 8% . While a few studies on high-resolution structural connectivity networks have appeared recently [e . g . ] , [ 12] , [24 , 25] , many earlier results , in particular those based on the data set used here , have relied on low-resolution data [e . g . ] , [ 26] , [27 , 28] . Although focussing on low-resolution data allows comparing to earlier low-resolution studies on other brain networks [e . g . ] , [ 11 , 20] , utilizing the information afforded by the available higher resolution connectivity may influence the outcome of complex network analysis [29] , [30] and has the benefit of maximizing usage of the available information . Here , we primarily analyse the high-resolution , 998-node anatomical connectivity network ( see Figure 2 ) , but we also compare to lower resolution results where appropriate . We employ three types of null-hypothesis networks , namely random , spatial and reduced surrogate networks . All three surrogate types preserve the size ( number of nodes ) , connection density ( number of edges ) and degree distribution ( the number of connections of each node ) of the cortical network , and differ from each other only in their physical wiring constraints: random surrogates are spatially non-constrained , spatial surrogates preserve the total wiring length of the cortical regions ( and thus that of the entire cortical network globally ) , and reduced surrogates possess reduced wiring lengths . All three types of surrogate networks were generated by the widely applied iterative rewiring algorithm [14] , [31] , the basic version of which proceeds as follows: Starting from the original cortical network , in each iteration two connections , ( r1 , r2 ) and ( r3 , r4 ) , are randomly chosen ( where ri refers to region i ) . After ensuring that no self-connections or parallel links ( multiple connections between two regions ) would be created , the two original connections are swapped to ( r1 , r3 ) and ( r2 , r4 ) . The above basic rewiring algorithm is sufficient to generate random surrogate networks . For the spatially constrained surrogate network sets , we incorporated the following additional rewiring conditions: each rewiring step is only executed if the resulting total connection length of every region ( i ) does not exceed that of the region in the original cortical network ( for spatial surrogates ) , or ( ii ) is reduced in every step ( for reduced surrogates ) . Because the complex curving trajectories of pathways cannot be preserved during rewiring , connection lengths are approximated by Euclidean distances between the positions of the region-pairs , for both cortical and surrogate networks . In the case of random and spatial surrogates , the procedure is terminated when each connection has been rewired 20 times on average ( 20 * ne/2 = 178650 connection swaps ) . For the most constrained reduced surrogates this stopping criterion is too severe because , as the algorithm progresses , progressively fewer rewiring operations with connection length reductions can be found . As a compromise , for this surrogate we chose to rewire each connection only once on average ( ne/2 = 8932 connection swaps ) , resulting in a reasonably diverse ( i . e . , not overly self-similar ) and yet well-optimised set of reduced surrogate networks ( see Results ) . On both resolutions , we generated n = 20 networks for all three surrogate types . To assess the topological similarity between the cortical connectivity network and its surrogates , we calculated the overlap between the set of connections of the cortical network and the surrogate networks , both in binary and weighted fashion . Specifically , we calculated the binary and weighted overlap between the cortical network C and each of its surrogate S using a modified version of the Sørensen similarity quotient QS [32] , which measures the similarity or relative overlap between two sets by the quotient of their intersection and union . We define the binary version of the similarity measure QSb as: ( 1 ) where N is the ( identical ) set of all nodes in networks C and S , and Cb ( Sb ) is the binarized connectivity matrix of C ( S ) with Cbij ( Sbij ) being 1 if there is a link between nodes i and j in C ( S ) and 0 otherwise . Note that the number of connections in C and S , |Cb| and |Sb| , are equal , and that the product Cbij Sbij is 1 if there is a connection between node i and j both in C and S , and 0 otherwise . Similarly , we define the weighted similarity quotient QSw as: ( 2 ) where Cij and Sij are the connection weights between regions i and j in networks C and S , respectively ( 0 if the two regions are not connected ) . QSb and QSw measure the relative similarity between the connection sets of two networks C and S that are defined on the same set of regions . Both QSb and QSw are normalised similarity quotients taking the value 0 if the two networks share no common connection ( minimal similarity ) , and 1 if the networks are equivalent , that is , they are composed of exactly the same set of binary/weighted connections ( maximal similarity ) . We use both measures because they assess network similarity of two networks in a complementary manner: the overlap in the binary layout of the two networks can only be assessed faithfully by QSb ( if the networks are different only in a small number of very high weight links , QSw is already low , despite the high binary overlap ) , while QSw accounts for the importance ( weight ) of the connections ( if the networks are different only in a number of very low weight links , QSb is lower , despite the high weighted overlap ) . We assess spatial network similarity between a network and its surrogates as the average spatial replacement of the connections of each region r , that is , the average change in the positions of all topologically adjacent ( linked ) regions of r ( its topological neighbourhood ) in the original and surrogate networks . The theoretically optimal solution for measuring such spatial displacement of the connections would require finding the ‘best matching pairing’ between the original and the rewired neighbour sets of r , i . e . , the pairing in which the sum of distances between the ( original , rewired ) region-pairs is minimal . An exhaustive search for this optimal pairing is however computationally infeasible ( given that the regions on average possess 35 connections , a lower estimate on the average number of pairings to check per region is 35 ! ≈1040 ) , therefore we developed and utilized the following algorithm to find an approximation of the optimal pairing . Given the set of the original topological neighbours of region r in the cortical connectivity , L = [l1 , l2 , …] , and the set of r's rewired neighbours in the surrogate network , M = [m1 , m2 , …] , we calculate the pair-wise distances D ( L , M ) = [d ( l1 , m1 ) , d ( l1 , m2 ) , … , d ( l2 , m1 ) , d ( l2 , m2 ) , …] between all element-pairs of the two sets . Then we sort D ( L , M ) ascending ( from the closest to the farthest original-rewired neighbour pairs ) , and , while iteratively going through the region-pairs of this sorted list , we put the current ( li , mj ) pair into pairing list P if and only if neither li nor mj is currently in P . Although the resultant pairing P provided by the ‘greedy algorithm’ above is not guaranteed to be the optimal pairing Popt between L and M , i . e . , the one having the lowest sum of ( original , rewired ) pair-wise distances , it is expected to provide a reasonable estimate on Popt given the close to homogeneous spatial distribution of the regions of the cortical network on the spheroid surface of the cortex [12] . Having obtained P for every cortical region r , we calculate the global relative spatial displacement D between the cortical connectivity C and its surrogate network S as: ( 3 ) where N is the set of all regions in the networks ( identical in C and S ) , Dr ( C , S ) is the average displacement of r's neighbours in C and S , Pr ( C , S ) contains the ( original , rewired ) neighbour-pairs of r for C and S , and d ( a , b ) is the spatial distance between regions a and b . With the above definition , D measures the distances between the original and the rewired neighbours of r ( connection displacement ) normalised by the distance of the original ( cortical ) neighbour from r , averaged over all connections and all cortical regions . D = 0 if there is no spatial displacement between the two networks , meaning that they are ( both topologically and spatially ) identical . A low D value indicates that there is only minor spatial displacement in the neighbour sets of the regions on average , while higher D values indicate a greater neighbourhood displacement , hence a larger difference in the spatial layout between the cortical connectivity and its surrogate network . Generally , the upper limit of D depends on the particular spatial distribution of the nodes and edges of the original network as well as of the wiring constraints of the rewired network in a complex manner . As a simplifying rule for the sparsely and predominantly locally connected ( high-resolution ) cortical network , however , we can regard D values on the order of 1 as indicators of substantial spatial neighbourhood displacement . A basic measure of network integration , global network efficiency [33] is the average of the inverse of the shortest path lengths dij between a node i and every other network node j , averaged over all network nodes: ( 4 ) where Ei is the efficiency of node i , n is the number of nodes , and dwij is the weighted shortest path length between nodes i and j ( the minimal of the weighted sums of constituent edges along each path between i and j , where connection weights are the reciprocal of their strength ) . High global efficiency implies that , on average , nodes require fewer intermediate steps along stronger ( higher weight ) edges to reach other nodes; therefore , networks with higher global efficiency possess greater potential for efficient internal information exchange and integration . The advantage of efficiency as a measure for integration over the more traditional measure of the mean shortest path length [15] is that efficiency can be computed for networks with multiple components , and generally is a more balanced measure due to the fact that the mean shortest path length can be strongly biased by the presence of only a few , very long paths [34] . A basic metric of network segregation , the clustering coefficient [23] is the fraction of triangles around a node ( the proportion of the node's topological neighbour pairs that are connected with each other ) , averaged over all network nodes . The weighted clustering coefficient [35] , which we use in this study on weighted networks , is defined as follows: ( 5 ) where Ci is the clustering coefficient of node i , ki is the degree of i , tiw is the ( weighted ) geometric mean of triangles around i , wij is the ( normalised ) connection weight between regions i and j ( 0 if i and j are not linked ) . The clustering coefficient of a node is high ( 1 ) if many ( all ) of its neighbours are also directly connected pair-wise ( by strength 1 connections in the weighted version of the measure ) , and it is 0 if none of its neighbour-pairs are directly connected . The clustering coefficient hence measures the ( topologically ) local density of connectivity of a network . Informally , a small-world network is a highly segregated ( i . e . , preferentially locally connected ) and yet relatively highly integrated ( i . e . , easily traversable ) network [23] . For the quantitative assessment of small-worldness , the network's high integration is usually translated to relatively short path lengths , while strong segregation is measured by a high level of clustering [36] . Among the several formulae developed to assess the degree of small-worldness of complex networks ( e . g . [33] , [37] ) , we chose an altered version of the Humphries–Gurney small-worldness index [37] , modified in the following way: ( 6 ) where C and Crnd are the clustering coefficient of the network and its random surrogates , while E and Ernd are their global efficiencies , respectively [15] . We note that Humphries and Gurney in [37] use average shortest path lengths instead of efficiency; however we prefer efficiency for the reasons stated above . A network is then said to be small-world if its clustering coefficient is larger than those of its random surrogates ( C≫Crnd ) , while their efficiencies ( shortest path lengths ) are comparable ( E≈Ernd ) , resulting in SW≫1 [37] . Using the intuition that high degree nodes should occupy a topologically central position in a hierarchical network as a starting point , Ravasz and Barabási introduced the simple but elegant hierarchy coefficient β for assessing hierarchical architecture in scale-free networks [38] . Noticing a distinctively exponential relationship between node degrees and clustering coefficients for various synthetic and real-world scale-free networks , they proposed that the exponent β of this relationship quantifies the tendency of high degree nodes to be linked to a large but sparsely intra-connected neighbour set ( hence exhibiting low clustering ) and thus effectively serving as connector nodes between segregated parts of the network [38] . Unfortunately , the human cortical network under study , and therefore also its degree-distribution preserving surrogates , exhibit an exponential , rather than scale-free degree distribution [12] , and the node degree – clustering relationship does not show a clear exponential shape , so that the β index of Ravasz and Barabási [38] cannot be applied directly . However , their basic idea remains valid irrespective of the specific shape of the functional degree to clustering relationship . Therefore , we here characterize hierarchical organisation by directly observing the degree to clustering relationship in the cortical network and in its surrogates . Specifically , in sparsely connected and locally highly clustered networks , ( of the sort studied here , see Results ) , high degree nodes of a network that possess a lower than average clustering coefficient are typically in a position to connect segregated parts of the network , suggesting a hierarchical element of the architecture with these high degree nodes in its centre ( see Figure 1B ) . In contrast , equal or higher than average clustering coefficients of high degree nodes indicate more homogeneous architectures and the lack of the hierarchical organisation pattern investigated in [38] . We note that the specific kind of topological organisation described above is of course not the only conceivable network architecture that exhibits hierarchical attributes . It is nevertheless the one that has previously been discovered in many sparsely connected , but highly clustered and modular real-world networks [38] , making it a good candidate to test for here . The modularity index Q , proposed by Newman [39] , has proved to be a highly accurate and powerful indicator of the modularity strength of a given partitioning of a complex network [16] , [40] . Given a set of node groups ( modules or communities ) M , that fully partition the network without overlaps , the modularity index Q of that partition is given by ( 7 ) where Qu is the modularity index of module u , euv is the proportion of all weighted edges wij between modules u and v in the network , lw is the sum of all weights in the network , and kiw is the sum of all connection weights of node i . Numerous algorithms have been developed to recover the modular structure of complex networks utilising Q as a ‘fitness’ measure to be optimised by some means ( for reviews , see [16] , [40] ) . In this study , we use the simple and elegant spectral algorithm developed by Newman [41] . Starting from the entire network as a single module , this algorithm iteratively splits each module into two , at each step finding the optimal bipartition by utilising a so-called ‘modularity matrix’ derived from the network's connectivity matrix . The leading eigenvector of the modularity matrix determines the node composition of the two sub-modules of each module to be split . The algorithm stops when no more increase in the global modularity index Q can be achieved by any additional split [41] . Along with its high accuracy , Newman's module detection procedure has the additional advantages of being a divisive , deterministic and generalisable method with low computational cost . See [41] for a detailed description of its implementation . We measured the consistency of the cortical module partition in surrogate networks with the scaled inclusivity index [42] . Application of this measure capitalized on the fact that the cortical network and its surrogates are defined on the same set of nodes ( cortical regions ) and differ only in their connection sets . Additionally , scaled inclusivity has the advantage of making no assumptions on the investigated partitions , and is thus generically applicable even on partition-pairs which differ in the number and sizes of modules they contain . For other pair-counting , cluster-matching , and information-theoretic techniques applied to compare module ( community ) structures of different networks , see [43]–[45] . The calculation of the scaled inclusivity index proceeded as follows . First , the individual module partitions of the cortical and surrogate networks were identified independently by Newman's spectral algorithm ( introduced above ) . Then , the cortical module partition QC , composed of m modules , was taken as a reference partition , and its match with the partition QSi of each network i of surrogate group S , composed of n modules , was assessed by calculating the n×m module-by-module similarity matrix XiC , which ( p , q ) -th element is calculated as: ( 8 ) where QSi ( p ) is the set of nodes ( regions ) belonging to the p-th module in QSi and QC ( q ) is the set of nodes belonging to the q-th module in QC . The resulting values range from 0 to 1 , where XiC ( p , q ) = 0 indicates zero overlap between the modules p and q ( i . e . , they do not share any node ) , and 1 indicates that the two modules are identical ( i . e . , they are composed of the same set of nodes ) . After calculating the matrix X for all networks in a surrogate group , the scaled inclusivity index SI of each cortical region is calculated as the mean of the similarity indices XiC ( p , q ) between all modules QSi ( p ) and QC ( q ) that contain the region , averaged over all surrogate networks i . Thus , scaled inclusivity measures how consistently a region is classified in each surrogate group , based on how well its cortical modules match with its surrogate modules , on average . We stress that SI is intended as a generically applicable metric to measure the degree of similarity between the module classification of network nodes , and it does not aim to accurately measure the actual magnitude of ‘overlap’ between the partitions ( see [42] and Eq . 8 above ) . The ‘core’ of a network is usually determined by an iterative peeling algorithm . These algorithms , at each step , remove ( ‘peel off’ ) a set of ‘shell’ or ‘crust’ nodes , in order to progressively focus on the more ‘centralised’ nodes . Centralisation in these procedures is assessed by a specific ‘coreness condition’ , as described below . To find the core structures of binary and weighted networks , we used the k-core and s-core decomposition methods , respectively . The k-core of the network [46] , for a given degree k , is the maximal set of nodes that are connected to at least k other nodes in the core . The k-coreness index of a node is then the highest degree k for which the node is still a member of the k-core . Similarly , the weighted variant of the k-core , the s-core of the network [12] is the group of nodes in which each node has a summed connection strength of at least s towards the rest of the s-core ( i . e . , the sum of the weights of its intra-core connections is not less than s ) . For increasing s ( k ) , the s-core ( k-core ) shrinks progressively and the tightest or innermost s-core ( k-core ) of the network [simply s-core ( k-core ) from here on] is the set of remaining nodes in the last non-empty s-core ( k-core ) . The so-called rich-club phenomenon is the tendency of high degree nodes to be preferentially connected to each other [47] , [48] . The degree of ‘rich-clubness’ is usually measured by the k-density function φ ( k ) of the network , which is the internal connection density among all nodes with degree larger than k . There is a basic difference between k-core/s-core and rich-club properties: while k-core and s-core nodes are selected by their connections within the subnetwork formed by the core , rich-club nodes are chosen simply and solely on the basis of their global degree in the entire network . ( Of course the ‘rich-clubness’ of this subnetwork does then depend on its internal connectivity . ) A possible weighted variant of the rich-club measure , as introduced in [49] , evaluates the tendency of the highest connection weights to be distributed among high degree ( ‘rich’ ) nodes . However , this variant , due to normalisation by the number of edges , is a connection density-independent index of weight centralisation and thus loses the ability of the unweighted rich-club index to measure edge centralisation among high degree nodes . Here we propose a novel weighted version of rich-clubness , which is sensitive to both properties , connection density and weight centralisation , and may hence be a more appropriate generalisation of the unweighted rich-club index to weighted networks . We define weighted rich-clubness as the internal weighted connection density φw ( k ) of the set of nodes with degrees larger than k , N>k , which is the ratio between the sum of connection weights W>k among the nodes in N>k and the maximum of their possible weight sum , Wmax>k: ( 9 ) where is the maximum possible number of edges among the nodes in N>k , and wrankedl is the weight of the lth strongest ( highest weight ) edges in the network . φw ( k ) defines a normalised measure of coreness , which takes a value in [0 , 1] for each degree k . φw ( k ) is 1 only in the extreme case where N>k is fully connected by exactly the strongest connections of the network . In general , φw ( k ) measures the fraction of total interconnection strength within N>k relative to this theoretical maximum ( as defined by the connection weights present in the network ) . Note that in Eq . 9 the denominator is not calculable if Emax>k is greater than the number of edges , E , in the network . This condition renders the interpretation domain of φw dependent on the connection density of the investigated network , implying that φw is meaningful for weighted rich-clubness measurements only for that fraction of the highest degree nodes N>kmin . Specifically , for undirected graphs , the number of these nodes |N>kmin| cannot be larger than the real solution of the quadratic equation ( 10 ) Eq . 10 specifies the largest number of nodes x that can still be fully interconnected by the existing number of edges E in the network . The cortical network under study has E = 17865 connections , hence we obtain |N>kmin| = 188 nodes as the largest weighted rich-club size that can be assessed by our measure . This corresponds to 18 . 8% of the nodes of the entire network , and gives φw ( k ) the domain of k∈[kmin , kmax] , where kmax = 97 is the largest node degree in the network , and kmin = 49 is the degree of the 188th node in the degree rank ordered node list . We note that , apart from of this interpretation limit of the measure , when applied to unweighted ( binary ) networks φw gives the same result as the traditional rich-club metric , underlining that φw can be interpreted as a generalisation of this traditional measure for weighted networks . Degree assortativity is a global measure of the tendency of nodes to be preferentially connected to other nodes with similar degree [50] . Degree assortativity is thus closely related to the phenomenon of rich-club formation , although while the latter only accounts for high degree nodes , the former measures preferential connectedness across nodes of all degrees . The assortativity coefficient r of a network is formally defined as: ( 11 ) where ji , ki are the degrees of the nod es at the ends of edge i , and M is the number of edges [50] . Degree assortativity is a normalised measure ( −1≤r≤1 ) , so that a network has positive r assortativity values if its edges tend to connect nodes of similar degree , while negative assortativity values indicate the tendency for nodes with different degrees to be linked . A network with r≈0 expresses neither of these trends , and is non-assortative .
The high-resolution weighted cortical connectivity matrix and averaged connectivity matrices of the three surrogate sets are illustrated in Figure 3A–D . To allow meaningful comparisons , surrogate networks need to be sufficiently randomised . The rewiring algorithms , as outlined in Methods , are constrained by several factors during the randomisation of cortical connectivity . In order to assess that sufficient randomisation has been achieved in spite of these constraints , we quantified the degree of similarity between each ensemble of surrogate networks and the cortical network , and we also examined the similarity within each surrogate ensemble . To examine topological similarity , we calculated the mean binary and weighted similarity quotients , QSb and QSw ( Eqs . 1 and 2 ) of the networks in the three surrogate sets to the cortical network . For random surrogates , QSb ( C , Srnd ) = 0 . 054±0 . 002 and QSw ( C , Srnd ) = 0 . 047±0 . 002 , indicating that their connections are almost entirely different from those of the cortical network . For spatial surrogates , we obtained intermediate similarity quotient values QSb ( C , SS ) = 0 . 494±0 . 002 and QSw ( C , SS ) = 0 . 483±0 . 002 , and for reduced surrogates higher similarity quotients QSb ( C , SR ) = 0 . 670±0 . 001 and QSw ( C , SR ) = 0 . 700±0 . 001 . These results confirm that , as expected , conserving and , even more significantly , further decreasing the already short connection lengths of the cortical connectivity network limits the achievable topological randomisation of the spatial and reduced surrogate networks . The similarity quotient values described above exhibit only very small deviations around their respective means . This could reflect the combined consequence of a sufficiently extended connection shuffling process together with the relatively large size of the networks , following the law of large numbers . But it could also indicate an undesirably low diversity in the generated surrogate sets , i . e . , each set might be composed of highly similar networks . To test for this possibility we calculated the similarity quotient between every pair of surrogate networks in each of the surrogate sets . The resulting mean intra-group values and their standard deviations are QSb ( Srnd , Srnd ) = 0 . 053±0 . 001 and QSw ( Srnd , Srnd ) = 0 . 045±0 . 001 for random surrogates; QSb ( SS , SS ) = 0 . 474±0 . 003 and QSw ( SS , SS ) = 0 . 483±0 . 002 for spatial surrogates , and QSb ( SR , SR ) = 0 . 873±0 . 002 and QSw ( SR , SR ) = 0 . 861±0 . 002 for reduced surrogates . Together , these results indicate that topological differences among surrogate ensembles , although decreasing with stricter spatial constraints , are nevertheless significantly nonzero . Interestingly , the low intra-group variance of the similarity values within every surrogate set suggests that in each such set S there is a ‘characteristic similarity’ , QS ( S , S ) , between any two members of that set . In addition , the similarity of the cortical network to its surrogate networks is comparable to these characteristic intra-group similarities in the case of random and spatial surrogates ( QS ( C , SS ) ≈QS ( SS , SS ) and QS ( C , Srnd ) ≈QS ( Srnd , Srnd ) ) . This suggests that the cortical network is a generic member of the random and spatial surrogate sets in terms of its basic region-to-region connectivity , as measured by QS . This further supports the use of random and spatial surrogates as suitable null-hypothesis networks with respect to the preserved basic properties of the cortical connectivity defined by each surrogate type . By contrast , reduced surrogates appear to form a separate class of networks that are more similar to each other than to the cortical network ( QS ( C , SR ) ≪QS ( SR , SR ) ) . This is expected given the restrictive form of spatial constraint applied during their generation ( strictly decreasing total connection length in every rewiring step ) , which is likely to make them collectively drift away from their cortical origin , converging towards the ( hypothetical ) single , minimal connection length surrogate network . The QS values illustrate well the highly optimised wiring of the cortical network in terms of connection length . While random surrogate networks share only 5 . 4% of their connections with other random surrogates and with the cortical network , this ratio increases to 49 . 4% for spatial surrogates , and each reduced surrogate is only able to substitute about one third of the long-range cortical connections with shorter ones . Furthermore , as shown in Figure 3A–C , these pair-wise overlaps translate into a ‘core’ set of connections collectively shared between the cortical network and its spatial and reduced surrogates . This ‘skeleton connectivity’ is primarily located along the main diagonal of the connectivity matrices , where most of the potential short-distance connections can be placed ( due to the spatial ordering of the brain regions in the connectivity matrices , explained in detail in the caption of Figure 3 ) . We note , however , that Figure 3B–D show the averages of the connectivity matrices of the surrogate network groups and therefore exaggerate the pair-wise overlap of the networks in each group . This is a consequence of the relatively small set of potential short-range connections in cortical space ( compared to the number of all possible connections ) , a number of which are inevitably shared by many reduced and spatial surrogates . For example , to examine the most extreme case of shared connectivity , we can determine the connections that are present in all network instances of each surrogate group . As expected , there are no such collectively shared connections among random surrogates . On the other hand , the highly optimised , and hence self-similar , reduced surrogates collectively share as many as 65 . 0% of their connections , while the ‘intermediately’ constrained spatial surrogates have only 7 . 6% of their connections shared among all of them , rendering the latter surrogate group relatively diverse . Furthermore , all shared connections of reduced and spatial surrogates are also present in the cortical network . These findings , in accordance with the ones on QS above , indicate that the cortical network is indeed a generic member of its spatial ( and random ) surrogates in terms of the basic properties of its connectivity , adding some topological credibility to our surrogate analysis . Having assessed the topological similarity of the surrogate ensembles to the cortical network , we now investigate the other relevant aspect of surrogate creation , namely to what degree the spatial layout and wiring properties of the cortical network have been changed in the surrogate ensembles . Although topological and spatial similarity are related , they do not specify each other . For example , low topological similarity between the cortical network and its surrogates in itself does not exclude that connections of the cortical network may only have been displaced by a short distance , leaving the spatial layout of the network largely unaffected by the randomisation procedure . In order to assess the impact of the randomisation procedure on the spatial layout of the cortical network , we calculated the relative spatial displacement D between the high-resolution cortical network and its surrogate groups ( see Assessing spatial similarity in Methods ) . We obtained a D ( C , Srnd ) = 4 . 04±3 . 43 mean displacement value for random surrogates , indicating that on average a neighbour l of each region r in the cortical network is replaced by a new neighbour m in random surrogates , which is about four times further away from the original cortical neighbour l than the length of the original cortical connection ( r , l ) . In spatial and reduced surrogates , we measured D ( C , SS ) = 0 . 50±0 . 62 and D ( C , SR ) = 0 . 29±0 . 43 , respectively , indicating a necessarily lower mean spatial displacement of the regions' neighbourhoods in the topologically more similar spatially constrained surrogates . However , because a significant number of connections is shared by the cortical network and its surrogates ( see Topological similarity of surrogate networks ) and hence have zero displacement , the high standard deviation in D ( C , SS ) and D ( C , SR ) indicates that those connections that have actually been rewired are displaced to a location that is substantially distant from their original target region in the cortical network . This is indeed what we see if we exclude the overlap of the connectivities and calculate the spatial displacement Dr of the replaced connections only: Dr ( C , SS ) = 0 . 97±0 . 57 and Dr ( C , SR ) = 0 . 88±0 . 30 , which indicates that the average displacement of rewired connections is almost as large as the length of the original connection . The connection length distribution and total connection length of each region ( sum of distances to all neighbours ) in the high-resolution cortical network and its surrogates are shown on Figure 3E and F . Consistent with the predominantly local connectivity of the cortical network ( mean connection length per region: CLC = 27 . 625 mm ) , random rewiring of cortical connections nearly tripled the average connection length ( mean ± standard deviation of random surrogate network means: CLrnd = 75 . 971±0 . 164 mm ) . For this reason , that is , due to the natural tendency of random connection swapping to increase the length of originally short cortical connections , the simple condition applied during spatial surrogate generation ( i . e . , ‘not to exceed the original total connection length of the cortical network’ ) was sufficient to actually achieve conservation of connection lengths ( CLS = 27 . 507±0 . 120 mm ) , and resulted in a slightly narrower connection length distribution ( standard deviation of connection lengths: cortical network: σlC = 22 . 146 mm→spatial surrogates: σlS = 18 . 589 mm ) originating from a somewhat shorter tail of the distribution ( see Figure 3E ) . Wiring length optimisation in reduced surrogates of the high-resolution weighted network successfully reduced the mean cortical connection length by 29 . 6% ( CLR = 19 . 433±0 . 013 ) , effectively substituting long-range cortico-cortical projections with shorter , local ones . This also led to a much narrower distribution of connection lengths ( standard deviation of connection length: cortical network: σlC = 22 . 146 mm→reduced surrogates: σlR = 7 . 382 mm ) . As a result of the above , the total connection lengths of individual cortical regions were preserved in spatial surrogates ( cortical network – spatial surrogates mean difference: −2 . 4±5 . 5% , Wilcoxon rank-sum test for identical distribution: p = 0 . 898 ) , while reduced and random surrogates had significantly decreased ( −24 . 6±17 . 0% ) and increased ( +227 . 6±114 . 0% ) regional connection lengths , respectively ( p<10−4 in both cases ) . Several earlier studies investigated spatially minimally wired surrogates of various neural and brain connectivity networks [51]–[54] . In order to investigate how much excess wiring length cortical connectivity has over its theoretical minimum , as well as to assess how the reduced surrogates compare to ‘bottom-up’ constructed , minimally wired models , we assembled two such models . For the first , unconstrained minimally wired network model , which we call absolute minimal ( AM ) network , we took the 998 cortical regions without their connections and simply placed links between the 17865 spatially closest region-pairs . This created a network with minimal total wiring length given the spatial arrangement of the cortical regions and the total number of connections in the cortical connectivity . The resulting AM network is composed of a single component ( no disconnected regions or groups of regions ) . The sum of its connection lengths is 62 . 9% of that of the cortical network , which , importantly , is only 10 . 6±0 . 1% less than the total connection lengths of the reduced surrogate networks . Importantly , the degree distribution of the original cortical network has been completely lost in the AM network ( mean relative deviation of regional degrees between cortical network and the AM network: 52 . 5±130 . 7% ) . This means that the reduced surrogates were able to achieve highly optimised wiring lengths while fully preserving the cortical network's degree distribution , thus providing a powerful topological baseline to assess the significance of the cortex's high level network properties . Both the cortical network and its reduced surrogates share a large number of their connections with the AM network ( binary similarity quotient: QSb ( C , AM ) = 0 . 621 , QSb ( SR , AM ) = 0 . 760±0 . 0009 ) , showing once again the remarkably conservative wiring of the cortex: 62 . 1% of the cortical connections are among the theoretically shortest possible links in the cortical network . We devised a second ‘bottom-up’ constructed minimally wired connectivity model with the additional constraint of approximating the degree distribution of the cortical network . We construct this network , which we call the degree preserving minimal ( DPM ) network , in the following way . As with the AM model we start with the 998 cortical regions without any connections , and , by going through the list of potential connections ( region-pairs ) ordered from shortest to longest , we add each connection to the DPM network only if the current degrees of both corresponding regions in the DPM network are less than their original degrees in the cortical network . By this simple strategy we are able to create a network with 17799 connections ( 66 connections [0 . 4%] less than the cortical network ) that closely approximates the degree distribution of the cortical connectivity ( mean percentage deviation in regional degrees between cortical network and the DPM network: 0 . 2±1 . 8% ) . Due to the similarity in degrees , the DPM network shares an even larger number of connections with both the cortical network and the reduced surrogates than the AM network ( binary similarity quotient: QSb ( C , DPM ) = 0 . 653 , QSb ( SR , DPM ) = 0 . 855±0 . 002 ) . The sum of connection lengths in the DPM network is 72 . 1% of that of the cortical network , which is on average 2 . 4±0 . 1% more than those of the reduced surrogates , despite the fact that the DPM network has slightly less connections than the reduced surrogates . This demonstrates that simple ‘bottom-up’ algorithms are not guaranteed to be more successful in constructing minimally wired ( surrogate ) networks than the rewiring methods used in the current study . We conclude that the spatial surrogates effectively preserved the wiring length properties of the cortex , both globally and at the level of individual regions , and that the reduced surrogates significantly decreased wiring length by substituting long-range connections with shorter ones . These properties render spatial and reduced surrogates suitable for representing a wiring-length-matching and wiring-length-optimised null-hypothesis network set of the cortical connectivity , respectively . The results so far demonstrate that , as opposed to the highly unrestricted nature of random surrogates , the presence of strict wiring constraints necessarily limits the form of potential connectivities of the cortex at the basic level of region-region connections , as shown by elevated similarity between cortical network and its spatial and reduced surrogates as compared to random surrogates . In the remainder of the paper , we go beyond these basic properties , to examine which other , network-level properties of the cortical connectivity these wiring constraints preserve . We measure the degree of expression of these properties by a series of complex network metrics , in each case applying the interpretations detailed in the Introduction ( see also Table 1 ) . The need for the simultaneous presence of functional integration and segregation imposes conflicting constraints on network architecture [55] , reflected in properties collectively known as ‘small-world’ characteristics . Small-world properties have been found in many real-world complex networks [23] , including various brain networks [10] , [56]–[58] . We measured the global integration and segregation potential of the cortical network compared to its surrogates using the quantities efficiency E and clustering coefficient C ( see Methods ) . As shown in Figure 4A , the cortical network is more similar to its reduced surrogates than to its other two surrogate sets ( high-resolution weighted cortical network: EC = 0 . 174 , CC = 0 . 271 , reduced: ER = 0 . 162±0 . 001 , CR = 0 . 289±0 . 002 , spatial: ES = 0 . 214±0 . 001 , CS = 0 . 169±0 . 002 , random: Ernd = 0 . 260±0 . 001 , Crnd = 0 . 024±0 . 001 ) . Considering that the total connection length of each region in the cortical network is the same as in its spatial surrogates , and that long-range connections are largely absent in reduced surrogates , the efficiency results indicate that the long-range cortico-cortical connections are distributed in a topologically sub-optimal way for enhancing tight functional integration ( efficiency ) in the cortical network . Furthermore , the clustering coefficient indices demonstrate a prevalence of topologically segregated neighbourhoods of groups of regions , beyond what would be expected from the wiring constraints of its individual regions ( CC is significantly larger than CS and much closer to CR than to CS ) . Therefore , not only when comparing against the necessarily more highly integrated and less segregated random surrogates , but also when taking into account the total length of the connections of each cortical region in the spatial surrogates , the cortical network appears to strongly favour topological segregation over integration ( efficiency ) . In order to assess the effect of wiring constraints on its small-world attributes , we calculated the small-world index SW of the cortical network and its spatial and reduced surrogates ( see Methods ) , using random surrogates as reference networks ( random surrogates hence have SWrnd = 1 by definition ) . First we note that all three investigated network types ( cortical network , spatial and reduced surrogates ) satisfy the basic small-worldness condition [37] , that is , their clustering coefficient is larger than those of its random surrogates ( C≫Crnd ) while their efficiencies ( average shortest path lengths ) , while being lower ( higher ) , are still comparable to those of their random surrogates ( E≈Ernd ) . In case of the cortical network , this results in the relatively high small-world index SWC = 7 . 478 ( see Figure 4A ) , indicating a well-expressed small-world organisation of the cortex . In comparison , we obtain on average SWS = 5 . 746±0 . 031 for spatial surrogates , and SWR = 7 . 419±0 . 031 for reduced surrogates , both much closer to SWC than the random surrogates ( recall SWrnd = 1 ) , indicating that the small-world architecture of the cortex can be attributed to a great extent to its wiring constraints . However , considering that SWC is significantly higher than SWS , the cortical network appears to exhibit the small-world property beyond what would be implied by its local connectivity alone . Furthermore , this excess level of cortical small-world organisation does not necessitate any particular arrangement , or even the presence , of the long-range cortical connections , as indicated by SWR not being significantly different from SWC . Therefore , the highly segregated connectivity of the cortical network , also found in reduced surrogates , but not in spatial surrogates ( see above ) , appear to contribute more to the small-world organisation of the cortex than the mere existence or particular arrangement of cortical long-range connections . In their seminal work , Ravasz and Barabási [38] detected well-expressed hierarchical structure in all investigated non-spatial ( non-geographical ) , real-world networks , but not in spatial examples ( e . g . the power grid network and the Internet ) . They reasoned that the high cost of establishing physically long connections substantially limits the type of topology spatial networks can exhibit , potentially excluding strongly hierarchical forms . However , in a study of a 104-region structural network of the human cortex Bassett et al . [10] did find hierarchical properties in the brain among multimodal cortical regions , but not within unimodal and transmodal regions . Following Ravasz and Barabási [38] ( see Methods ) , we calculated the average clustering coefficients of groups of cortical regions with similar degrees , relative to the global clustering coefficient of the cortical network ( see Figure 4B ) . We observe that the cortical network exhibits a steep decline in its mean clustering – degree relation , indicating that the cortex exhibits the type of hierarchical organisation illustrated in Figure 1B . This finding supports the general notion of a hierarchically organised brain [59] , which is quite remarkable given the tendency of spatially embedded , physical networks not to develop hierarchical features due to the basic spatial ( geographical ) constraints acting on them [38] . Furthermore , there are highly similar tendencies for spatial and reduced surrogates , but not for random surrogates , in which clustering actually increases with region degree . The remarkably high consistency of the clustering – degree relationship across the cortical network and its spatial and reduced ( but not random ) surrogates indicates that the individual wiring lengths and positioning of high degree regions in the cortex by itself entails a global hierarchical organisation . Many real world networks have a characteristic topology that allows them to be separated into relatively densely intra-connected and weakly inter-connected subgroups [16] , [60] . These subgroups are usually referred to as the modules ( or clusters , communities ) of the network . One possible functional advantage of modularity is reduced systemic risk during development and evolution [61] , [62] . Another is that modular architectures are related to potentially useful dynamical properties such as high dynamical complexity [21] and metastability [63] , as well as limited sustained network activity [64] . Recent studies have reported a highly modular architecture of the human brain in its structural [12] , [13] , [65] as well as in its resting state functional connectivity ( rsFC ) [66]–[68] . Furthermore , studying the effect of ageing on the brain's modular structure , Meunier et al . [69] found marked differences in the composition and putative topological roles between the modules in the rsFC of younger and older human subjects . These results suggest that modular ‘decomposability’ is a prominent feature of the brain , which is continuously shaped during its development , maturing and ageing . In line with these results , recent theories regard the brain's modular structure as the main facilitator of regional specialisation and segregated functional processing [18] . We investigated the modular structure of the cortical network and its surrogates by utilising Newman's module detection algorithm [41] ( see Methods ) . In order to assess the strength of modular organization , that is , the magnitude of the Q modularity index , we use the modularity of the random surrogates as a baseline value ( representing the modularity index of a non-modular network with size and connection density matching that of the cortical network ) . These random surrogates , as expected due to their quasi-zero segregation , express almost no modularity ( mean modularity index: Qrnd = 0 . 087±0 . 003 , number of modules: Nrnd = 23 . 25±1 . 95 ) . In contrast , the cortical network has a strongly modular architecture ( QC = 0 . 558 ) composed of NC = 13 , spatially compact and hemispherically symmetric modules ( Figure 5A ) . Interestingly , reduced surrogates , in spite of their lack in long-range ( thus mostly inter-module ) connections , do not exhibit a significantly higher modularity index ( QR = 0 . 567±0 . 015 , NR = 15 . 55±0 . 87 , one-tail t-test assuming normal distribution: p = 0 . 274 ) , but spatial surrogates do possess a significantly lowered level of modularity ( QS = 0 . 477±0 . 020 , NS = 11 . 55±0 . 87 , p<10−4 ) than the cortical network . These results show that while the physically constrained length of cortico-cortical white-matter connections are a fundamental factor in shaping the high strength ( QC ) and granularity ( NC ) of the global modular architecture of the cortex , the cortical network nevertheless has a stronger modular organisation than these wiring constraints by themselves would suggest , indicating the functional relevance of the cortex's modular structure . The strength of the modular organisation of the cortical network can be illustrated by its inter- versus intra-modular connection distributions ( Figure 5B ) . The NC = 13 identified modules contain 63 . 2% ( n = 11294 ) of the total number of projections internally , meaning that only 36 . 8% ( n = 6571 ) of the connections cross module boundaries . This results in an average 25 . 6% intra-module and 1 . 4% inter-module connection density , indicating that while more than every fourth intra-module region-pair is linked , this ratio falls to 1∶70 for region-pairs from different modules . For comparison , the global average connection density of the entire network is 3 . 6% . The cortical connectivity matrix ordered by the recovered module partitioning is shown in Figure 5B . To compare this partitioning with an ‘average’ partitioning for each surrogate group , we calculated the frequency with which every region-pair ( ni , nj ) can be classified into a single module ( m ( ni ) = m ( nj ) ) in each of the three surrogate network groups . The resulting matrices are shown in Figure 5C–E . The high concentration of frequent co-partitioning of region groups along the main diagonal of the matrices is apparent in the case of reduced and spatial surrogates , indicating that the corresponding cortical modules are reasonably preserved across these surrogate networks . Furthermore , there is also a tendency for the formation of off-diagonal blocks in Figure 5C and 5D which suggests that parts of some of the cortical modules are frequently merged into single surrogate modules , and therefore they are at least partially preserved in reduced and spatial surrogates . Motivated by these findings , we quantitatively tested the robustness of the modular partitioning of the cortical network against the rewiring applied to its surrogate groups by assessing the consistency of the cortical partition in the surrogate groups . To do this , we used the obtained cortical modules as a reference partition and measured the scaled inclusivity index SI of each cortical region in the surrogate groups ( see Methods ) . Among the three surrogate sets , reduced surrogates exhibited the highest mean SI index , indicating the highest overall conservation of cortical modules in reduced networks , although with high variations across the individual cortical regions ( mean ± std: reduced surrogates: SIRC = 0 . 235±0 . 182 , spatial: SISC = 0 . 202±0 . 145 , random: SIrndC = 0 . 007±0 . 002 ) . The SI values for the individual cortical regions , and for each surrogate group , are illustrated in Figure 5F–H . We found elevated robustness of the cortical modules in both reduced and spatial surrogates at specific cortical sites , including the entire pre-central and post-central cortices ( composing cortical modules M2 and M6 on Figure 5A ) , large areas of the temporal lobe ( M3 and M5 ) and some frontal ( M4 and M9 ) , and superio-parietal and limbic areas ( M10 ) . The high SI of these specific areas indicates that their modular structure exhibits greater robustness against spatially constrained rewiring , as opposed to the low SI of , and thus higher variance in , the module formations in other parts of the cortical network . The results so far , regarding the small-world , hierarchical and modular architecture of the cortex , suggest the existence of specific cortical areas that are topologically centrally positioned in the modular structure of the cortical network . This ‘core formation hypothesis’ has been the topic of several studies recently ( see below ) , and we next test its significance against the wiring constraints of the cortex by again analysing the surrogate ensembles . Intuitively , the core of a network , illustrated in Figure 1C , is a set of ‘elite’ nodes that are topologically centrally positioned , forming a highly intra- and inter-connected global centre [28] . The existence of a single , but strong core formation in the topology of a network typically suggests that the network exhibits a pronounced global core-periphery structure [70]–[72] and indicates the presence of centralisation in the network's dynamics and functional operation , which is fundamentally different from that of a homogeneous network architecture composed of distributed , identically segregated units ( e . g . , Figure 1A ) . Prior studies have identified and investigated a core structure in various brain networks , including the rich-club structure of the cat thalamo-cortical complex [17] , , the k-core of the macaque brain [75] , the s-core of the human cortex [12] , and the rich-club of the entire human brain [13] , [25] . We here compare s-core and rich-club properties of the cortical network and also assess the extent of their dependence on , and emergence given , different wiring constraints using the three surrogate types . S-core analysis assesses the extent to which a network exhibits a densely intra-connected inner core , by measuring the size of , and overall connection strength within , the most strongly intra-connected group of nodes . We identify the s-core of the cortical network through a ‘peeling’ procedure that iteratively removes less connected regions from a candidate s-core ( see Methods ) . Examining the evolution of the s-core decomposition of the high-resolution cortical network and those of its surrogates ( Figure 6A ) during the peeling procedure , we can identify two characteristic phases . A longer , rather stable early phase of ‘crust peeling’ transitions into an unstable phase for s>11 , in which the s-cores of both random and spatial surrogates diminish rapidly and then abruptly vanish . The cortical network , on the other hand , closely follows the trend of its reduced surrogates and continues to sustain a substantial s-core of n = 100 regions ( 10 . 0% ) for much longer . This s-core eventually collapses at a significantly higher strength threshold ( sC = 13 . 095 ) than its counterparts in the random ( srnd = 12 . 055±0 . 078 ) or spatial surrogates ( sS = 11 . 433±0 . 124 ) , within the range of the s-cores of reduced surrogates ( sR = 13 . 027±0 . 143 ) , but with a somewhat larger size ( s-core size of cortical network SC = 100 , reduced surrogates: SR = 74 . 500±17 . 119 , see Figure 6A inset ) . Considering that the connectivity of reduced surrogates is spatially more concentrated than that of the cortex , which is a property that favours the formation of a strong s-core , the above finding suggests that cortical connectivity may be optimised towards the formation of a global s-core , which is much stronger and larger than its connection length constraints alone would suggest . An alternative measure of core formation in a network is the assessment of its rich-club index [47] , [48] . The weighted variant of a rich-club index , φw ( k ) , measures the tendency of high degree nodes to be both densely and strongly inter-connected ( see Methods ) . Examining the evolution of φw ( k ) with increasing k in the cortical network and in its surrogates ( Figure 6B ) , the cortical network demonstrates a rich-club of significant strength ( weighted k-density at n = 100 regions: φwC ( 100 ) = 0 . 164 ) compared to its random surrogates ( φwrnd ( 100 ) = 0 . 106±0 . 002 ) . However , the cortical network does not possess a significantly stronger rich-club structure than its reduced surrogates ( φwR ( 100 ) = 0 . 164±0 . 001 , one-sample t-test: p = 0 . 23 ) or its spatial surrogates ( φwS ( 100 ) = 0 . 163±0 . 003 , one-sample t-test: p = 0 . 34 ) . Previous studies [13] , [17] , [25] used only random surrogates as null-hypothesis baselines for assessing the rich-club property of brain networks , a comparison in which the cortical networks we study here also express a highly developed rich-club ( Figure 6B , compare blue and magenta lines ) . However , we show here that this property is equally , or even more , expressed in spatial and reduced surrogates . Closer inspection reveals that the relatively low variance in the spatial locations of highly connected regions ( Figure 6D and F ) , in combination with the highly clustered , local connectivity of the cortex , naturally results in a tendency for strong rich-club formation . The wiring-constraint-dependent rich-club formation tendency of the cortex is further supported by the assortativity coefficients r of the network and its surrogates ( see Methods ) . We found significantly positive assortativity coefficients for the high-resolution cortical network ( rC = 0 . 288 ) and its spatial ( rS = 0 . 283±0 . 004 ) and reduced surrogates ( rR = 0 . 326±0 . 002 ) , indicating their tendency to connect nodes of similar degree , whereas almost no degree assortativity is found in random surrogates ( rrnd = 0 . 051±0 . 006 ) . This preferentially mutual connectedness of high degree regions suggests that the rich-club patterning of the cortical network naturally arises from the physical location of cortical hubs and the cortical wiring constraints . The s-core and rich-club regions selected by the two methods ( Figure 6C–F ) , are largely consistent with earlier findings [12] , [13] . Furthermore , the s-core ( n = 100 regions in final , non-empty core ) and rich-club regions ( n = 100 highest degree regions ) exhibit a considerable , exactly 50% ( n = 50 regions ) overlap in the cortical network . There are , however , marked differences in the anatomical composition and spatial dispersion of the two structures . The s-core of the cortical network encapsulates the caudal part of the cortical midline , formed by the precuneus , the cingulate cortex and the superior part of the occipital lobe ( cuneus , lingual gyrus and pericalcarine cortex ) . This centralisation is also present , though much less pronounced , in the cortical network rich-club , since about one third of it extends to the lateral and frontal parts of the cortex . The spread of arborisation of the two cores also exhibits this difference ( see Figure 6C–F ) : while the more numerous ( n = 5662 [31 . 7%] connections ) and rather externally projected connections ( 20 . 6% internal connection density ) of the rich-club establish direct connectivity with almost the whole remainder of the cortex ( n = 795 [88 . 5%] regions ) , the s-core possesses a smaller ( n = 3921 [21 . 9%] connections ) , as well as more internally projected connection set ( 37 . 7% internal connection density ) , which connects it directly with only one third ( n = 294 [32 . 7%] regions ) of the rest of the network . These differences , originating from the definitions of the s-core and rich-club structures , demonstrate the more distributed nature of the cortex's rich-club , as opposed to the rather encapsulated , but spatially and topologically central position of the s-core . Along with the analysis on the high-resolution weighted version of the cortical connectivity dataset presented above , we also performed our surrogate analysis on four ‘subsets’ of the full dataset , namely: on the binarized ( unweighted ) version of the high-resolution cortical network , on the weighted and the binarized versions of a lower resolution ( down-sampled ) cortical network ( see Methods ) , and on a single hemisphere extracted from the high-resolution weighted cortical network ( discussed in detail in the following section ) . Similarly to the analysis of the high-resolution weighted cortical network , we first tested the surrogates of the three cortical networks considered here with the topological similarity measure QS and the measure of mean connection lengths per region , CL . Our surrogate test results on the three cortical networks showed the same pattern that we described for the weighted high-resolution cortical network ( see Figure 7 ) , albeit with an overall lower level of randomisation ( higher topological similarity ) for the low-resolution networks , due to the higher connection density of these networks ( high-resolution: 3 . 6% connection density , low-resolution: 26 . 8% ) , as well as a slightly ( but significantly ) reduced connection length in low-resolution spatial surrogate networks , likely due to the limitations of re-wiring algorithms on smaller networks . Next we assessed the global integration and segregation potential of these cortical networks by calculating their clustering coefficient C and efficiency E , respectively . In accordance with the high-resolution weighted results , we found the same pattern of higher similarity of each cortical network to its reduced than to its spatial surrogates consistently across all analysed cortical networks , to the extent that for low-resolution networks there is no significant difference between them ( see Figure 7 ) . Therefore , as with the high-resolution weighted cortical network , these networks also demonstrate a small-world index more similar to their reduced than spatial surrogates ( see Figure 7 ) . Surrogate analysis of the modularity strength Q of these cortical networks also yield highly consistent results with those of the high-resolution weighted cortical network ( see Figure 7 ) . Taken together , these findings are consistent with our results on the high-resolution weighted cortical network; they indicate that the functional segregation potential and the small-world and modular organisation of the connectome , even when observed on lower resolutions , are significantly stronger than its wiring constraints alone can account for . We next evaluated the core formation tendencies of the three cortical networks . Results on the k-core ( unweighted s-core ) of the high-resolution binary connectivity are in agreement with the high-resolution weighted network results discussed above . On low network resolution , however , we observe different characteristics ( see Figure 7 ) . Specifically , the significantly strong k-core and s-core structures of the binary and weighted high-resolution cortical networks seem to weaken to a weighted low-resolution cortical s-core of comparable strength to its wiring-constrained surrogate ensembles , and further diminish to a binary low-resolution k-core with a strength significantly weaker than any of the surrogates . When investigating the binary low-resolution cortical k-core more closely , we discovered that it contains as much as 80 . 3% ( 53 regions ) of the entire network , and any subsequent peeling step ( see Methods ) destroys the whole structure . This is in stark contrast to the k-core of the low-resolution spatial surrogates , which are on average composed of only 52 . 3% of the network , or with the k-core ( s-core ) of the binary ( weighted ) high-resolution cortical network , which contains only 11 . 2% ( 10 . 0% ) of the 998 regions . The difference between network resolutions may largely be attributable to the high degree of spatial concentration of the high-resolution s-core ( and k-core ) regions ( Figure 6C and E ) . This concentration results in the collapse of large parts of the high-resolution core structure into single low-resolution regions of the cortex ( specifically into the precuneus , the cingulate cortex and superior areas of the occipital lobe ) , the extremely dense internal connectivity of which is not accounted for during low-resolution analysis . Consequently , even the weighted , but especially the binary , cortical network , as observed on lower resolution , appear to exhibit a more distributed , homogeneous connectivity , with highly inhomogeneous intra-region connection densities , that are only accounted for at the higher resolution analysis . More generally , these findings underline the importance of multi-resolution analysis in cortical connectivity research in order to obtain a more complete and accurate picture on the inherently multi-level organisation of the connectome . In conclusion , our surrogate analysis results extend those of Hagmann et al [12] by showing that the core structure of the high-resolution cortical network is both topologically and spatially significant , as measured by both k-core and s-core analysis . Furthermore , our findings on the low-resolution connectivity also indicate that this result depends on high-resolution analysis because the cortical connectivity becomes increasingly sparse and centralised at higher resolutions . We next evaluated the tendency of the three additional cortical networks for the formation of the other putative ‘core’ structure , the rich-club . In line with the results on the weighted high-resolution connectivity , we obtained cortical network rich-clubs in the low- and high-resolution binary connectivities with strengths comparable to those of their spatial surrogates , and even somewhat weaker than those of their reduced surrogates , assessed by the traditional ( unweighted ) rich-club measure ( see Figure 7 ) . In contrast to these results , we found a rich-club in the weighted low-resolution cortical connectivity that is statistically stronger than those of its spatial surrogates ( one-sample t-test: p = 0 . 02 , see Figure 7 ) . Originating from its agglomerative construction from the high-resolution cortical network ( see Methods ) , this finding may reflect the highly non-uniform ( exponential-like ) connection weight distribution of the weighted low-resolution cortical network . In essence , the surrogate rewiring process in the random and spatial surrogates of this cortical network , but not in its reduced surrogates , was effective in relocating the few very short-range , but extremely strong cortical connections to random positions in the network , resulting in a highly variable , but on average lowered weighted rich-club strength in these random and spatial surrogates ( Figure 7 , second row ) . ( We note that by the nature of their definition , the rest of the weighted metrics investigated in this study , including the s-core structure , are largely immune to this kind of variation in the specific location of these few , extreme strength connections in the low-resolution weighted cortical network . ) Nevertheless , the results indicate that the low-resolution weighted cortical network , in agreement with the other three connectivity types , demonstrates a significantly strong rich-club structure , by comparison with traditional random surrogates ( one-sample t-test , p<10−4 ) . Contrary to the other three connectivity types , however , the strength of the rich-club in the low-resolution weighted connectivity does not seem to be fully attributable to the spatial constraints of the cortex , as indicated by spatial surrogate comparison . An analysis approximating fibre length by the Euclidean distance of the connected regions ( see Methods ) may disproportionately underestimate the length of the longer curved inter-hemispheric fibres , particularly those connecting homotopic regions around the cortical midline [8] . This , in turn , may result in an increase in the number of inter-hemispheric connections with underestimated lengths in the wiring constrained surrogate networks . Indeed , evaluating the proportion of intra- and inter-hemispheric connections in the cortical network and in the surrogate networks shows that while only 11 . 5% of the high-resolution cortical connections run between the hemispheres , this ratio increases to 13 . 2% for reduced , 18 . 1% for spatial and 50 . 2% for random surrogates . Some , although certainly not all , of these ( surrogate ) inter-hemispheric connections are likely to cause a corresponding underestimation in the connection length of reduced and spatial surrogates compared to that of the cortical network . This concern , however , is greatly eased by noting that the regions of the cortex along its midline are already highly intra-connected ( see Figure 6 ) , leaving only few potential places where such new connections can be formed . Indeed , calculating the mean ( Euclidean ) distance between inter-hemispherically connected region pairs DIH on high network resolution , we found an increase , rather than a decrease , in the DIH of spatial surrogates compared to that of the cortical network ( DIHctx = 26 . 2 mm , DIHS = 38 . 8 mm ) . In comparison , we found , as expected , that the mean distance between the inter-hemispherically connected region pairs is somewhat lowered in reduced surrogates ( DIHR = 22 . 2 mm ) and greatly increased in random surrogates ( DIHrnd = 86 . 2 mm ) . These results indicate that the newly created inter-hemispheric connections in spatial surrogates are predominantly between relatively distant regions , therefore suffer less from the disproportionate underestimation of connection length , as approximated by Euclidean distance , between homotopic regions along the cortical midline . Nevertheless , in order to test our results against potential artefacts originating from the different degree of inter-hemispheric connectedness in the cortex and its surrogates , we repeated the analyses using a single cortical hemisphere . Specifically , we extracted the right hemisphere of the weighted high-resolution dataset , generated n = 20 surrogate networks for each of the three surrogate network types using the same method as before , and measured the complex network metrics assessed in the paper . The results of single hemispheric analysis ( Figure 7 , third row ) are largely in agreement with the previous bi-hemispheric analysis . The main differences are that the ( hemi ) -cortical network has an increased small-world index compared to reduced surrogates , and its rich-club is slightly but not significantly weaker than those of spatial surrogates ( one-sample t-test: p = 0 . 1 ) , and stronger than those of reduced surrogates . We note that if there was a significant bias in the full cortex surrogate networks to form an excess number of inter-hemispheric connections between homotopic midline regions , we would expect single hemisphere surrogate analysis to detect a consistent increase , rather than decrease , in the strength of s-core and rich-club structures , given the highly central positioning of these structures along the cortical midline in the full cortical network ( see Figure 6 ) . Due the fact that we observe such an increase in only one out of the four possible cases ( the rich-club of reduced surrogates ) , we conclude that the single-hemisphere analysis validates the Euclidean approximation on fibre lengths for our surrogate analysis , and our main conclusions on the bi-hemispheric cortical network appear to largely apply to the uni-hemispheric cortical connectivity as well .
Standard models of complex network science in conjunction with the fundamentals of neuroscience shape the techniques we use for the analysis of brain networks . For example , the abstract concept of small-worldness has traditionally been defined in relation to random and lattice networks [23] . Thus , the diffuse nervous systems of coelenterates ( such as Cnidaria ) have long been recognised to exhibit a characteristically regular , lattice-like pattern [76] . These and other findings have contributed to the wide application of random and lattice surrogate techniques in brain network analysis . In this paper we have investigated how the use of more constrained null-hypothesis models , incorporating not only basic topological but also spatial properties of the human connectome , will help us better understand the structural organisation and functional operation of the inherently spatially and economically constrained brain . We analysed a dataset representing the large-scale anatomical connectivity of the human cortex in order to confirm previously reported topological organisation patterns ( network properties ) , such as small-worldness , modularity , hierarchy and core formation ( see Figure 1 ) , at both high- and low-resolution representations of cortical connectivity , and to then analyse the relationship of these patterns to the wiring constraints of the brain . To do so we devised two novel surrogate types , ‘spatial’ and ‘reduced’ surrogates that conserve the total length of connections from each region ( spatial ) or decrease it ( reduced ) . For each network property , our analysis adopted the reasoning detailed in the Introduction ( see also Table 1 ) . First , by comparing the cortical network and the spatially constrained surrogates to random surrogates , we found that cortical wiring constraints seem to contribute strongly to its relatively low potential for functional integration ( as measured by global efficiency ) and very high potential for functional segregation ( as measured by clustering coefficient ) , and thus highly , although not fully ( see below ) , account for the known small-world cortical organisation [57] , [58] . In addition , comparison of the cortical connectivity network to the new surrogates also showed a relatively low level of global efficiency in the cortical network , closer to its reduced than to its spatial surrogates . Efficiency is a measure of functional integration potential in the network [15] and is generally most effectively increased by adding sparse long-range connections [18] . Because reduced surrogates to a great extent lack these long-range connections , our findings suggest that long-range cortico-cortical connections are in fact sub-optimally placed for maximising efficiency , and therefore , to the extent that brain structure determines function , they may not contribute to tight functional integration in the cortex as much as they could . In line with this , the cortical network was also found to be more similar to its reduced than to its spatial surrogates in its very high clustering coefficient . Functional segregation , facilitated by high structural clustering coefficient [15] , is widely acknowledged to be a fundamental characteristic of the cortex [77] . Taken together , our findings indicate that the cortical network may possess an excess level of segregation and a relatively reduced level of functional integration potential over the extent that its wiring constraints alone can account for . Furthermore , spatial surrogates exhibited significantly weaker small-worldness compared to the cortical network , while reduced surrogates exhibited comparably high small-worldness to the cortical nework . These findings suggest that high cortical segregation combined with the concentrated spatial distribution of high degree regions ( see Figure 6D ) may suffice to ensure the strong small-world organisation of the cortical connectivity , even in the absence of long-range cortical connections . Hierarchical organisation is believed to be a central architectural feature of various complex social networks and the World Wide Web [38] ( Figure 1B ) . Hierarchical aspects of network architectures can fundamentally affect their evolution , development , adaptability and efficiency on multiple scales [61] , [62] . The structural connectivity of the cortex is generally regarded to have a hierarchical organisation [59] . However , neither the degree and extent of hierarchical organisation , nor the constraints governing its expression , have yet been analysed in large-scale whole-brain networks as comprehensively as for instance the concepts of modularity or regional centrality [59] . This may be due to a lack of a consensus on the formal definition and assessment of this rather informal notion , in combination with currently available data being insufficiently detailed to evaluate and characterise the exact nature of this organisation pattern on a global scale [1] , [78] . Here , we utilised the mean clustering coefficient as a function of degree , as a simple model for detecting hierarchical features in complex networks . The results indicated the presence of hierarchical organisation in the cortical network and in both spatially constrained surrogates , but not in random surrogates . One interpretation of this finding is that the predominantly local connectivity of the cortex and the central positioning of high degree regions give rise to the observed hierarchical structure . However , we cannot exclude an alternative explanation , namely that it is in fact the strong evolutionary pressure favouring the presumably functionally beneficial hierarchical organization , that led to the observed spatial embedding of cortical network nodes . Nevertheless , as both pressures , economical to conserve wiring cost and adaptive to achieve brain function , appear to benefit from a hierarchical organisation [18] , [59] , it seems most likely that their joint , mutual presence resulted in the observed hierarchical pattern in the structural connectivity of the cortex . The brain's modular architecture is organised around spatially compact modules and their predominantly short , intra-module connections [77] . This feature of cortical connectivity is believed not only to keep global wiring costs low ( economic pressure ) , but also to improve local communication efficiency within its structurally segregated and functionally specialised modular units ( functional pressure ) [18] . Our modularity analysis revealed that all spatially constrained networks indeed exhibit a strong and spatially compact modular architecture compared to random surrogates , indicating that basic wiring constraints of cortical regions naturally result in a tendency for cortical module formation . On the other hand , the high strength and granularity of the modular organisation of the cortex is more akin to its reduced surrogates , than to its relatively less modular spatial surrogates . This suggests that the long-range cortico-cortical projections may be more optimally placed towards a highly modular cortical architecture , than wiring constraints alone would suggest , supporting the widely acknowledged notion of high functional importance of cortical modules [21] , [63] , [64] . Furthermore , while the module partitions of the cortical network and its surrogates exhibit considerable differences , we found a set of cortical areas with modules that are highly preserved both in reduced and spatial surrogates . According to our analysis , the highly robust topological encapsulation of these predominantly lateral modules against the applied spatially constrained rewiring indicates that their existence can largely be explained by cortical wiring constraints . Additionally , however , the natural emergence of these module formations may enable them to provide a consistent base or ‘backbone’ to the cortex's modular structure both across individual variation and through development and ageing processes [24] . Such a modular ‘backbone’ structure could provide the structural basis for some relatively invariant , recurring components of the continuously reconfiguring functional networks of the brain [77] . While the exponential degree distribution [12] and hierarchical organisation already suggested a centralised organisation of cortical topology , we explicitly examined which , if any , parts of the cortex are located in its topological centre . Surrogate comparison revealed that the s-core of the high-resolution cortical network is stronger and larger than those of its spatial surrogates , and similar to those of its reduced surrogates . Furthermore , confirming previous results [12] , the s-core of the cortical network was found to be spatially encapsulated at a medial-caudal location , composed by the precuneus , the cingulate cortex and the superior part of the occipital lobe . The cortical network , when observed on high-resolution ( but not on low-resolution , see below ) , therefore appears to have a spatially compact , central s-core , the strength of which is significantly higher than its wiring constraints alone would suggest . One could interpret these findings to suggest that the cortical network s-core is not a by-product of wiring constraints but may instead be relevant for the brain's function; it might even serve the purpose of a putative central , global integrator substructure among the otherwise separate , functionally more specialised areas of the brain [79] . The other candidate central structure , the rich-club of the cortical network , also exhibits a significantly denser than random intra-connectedness , which is in agreement with previous studies detecting a well-expressed cortical rich-club structure [13] , [17] , [25] . However , in contrast to our results on the cortical s-core , we found rich-club structures of similar strength in the reduced and spatial surrogates . Thus , the rich-club formation of the cortical network appears to strongly correlate with its wiring constraints and the spatial distribution of the cortical hub regions ( one of the ‘basic’ network property preserved in all surrogate ensembles ) . These findings extend earlier studies consistently discovering the brain's strong rich-club structure [13] , [17] , [25] by pointing to a plausible relationship between the remarkably dense inter-connectedness of high degree cortical regions and cortical wiring constraints . It is important to note , however , that similarly to the case of the hierarchy analysis , our method does not provide information with respect to the direction of causation between these network properties . Thus , it remains to be seen whether the economical pressure to conserve connection length is in fact the primary driving factor in the spatial arrangement of hub nodes , or the functional pressure for rich-club formation necessitates the specific spatial distribution of hub nodes in the cortex in the first place , and thus ultimately the formation of the cortical rich-club patterning . Furthermore , compared to the s-core , the rich-club of the high-resolution cortical network was found to be internally relatively loosely coupled and formed by a spatially and topologically rather dispersed set of regions . These findings render the even spatially highly significant , well-confined and more tightly intra-connected cortical s-core a more appropriate candidate for a putative central cortical core [12] , while the rich-club seems to be more suited for fulfilling the role of a ‘dynamic router’ [25] , a set of distributed cortical hub regions predominantly connecting their local neighbourhoods with distant parts and the s-core of the cortex . Nevertheless , the large ( 50% ) overlap between the s-core and rich-club regions suggests a great extent of functional cooperation between these highly intertwined , both topologically and spatially central cortical structures . In line with these results , areas in the overlap between the s-core and rich-club structures of the cortex , the precuneus , the cingulate cortex and parts of the primary visual cortex ( BA 17 , 18 ) , have also been repeatedly identified as global functional hubs of the human brain [80] , [81] , and found to functionally mediate between cortical areas that are structurally not directly connected [82] . Furthermore , some of the regions that belong to both the s-core and rich-club structures , most notably the precuneus , have also been highlighted as prominent areas of the default mode network of the brain [12] , [83] . These findings suggest that the regions shared between the cortical network's s-core and rich-club , are not only topologically central , but also play a functionally pivotal role in coordinating , integrating or routing the activity of distant cortical regions in both resting and task-evoked states of the brain [25] , [79] , [83] . Figure 7 summarises the results on the investigated properties of the cortical network with respect to the three surrogate groups , at both network resolutions ( 998 regions at high-resolution and 66 regions at lower resolution ) , both for binary and weighted networks on each resolutions , as well as for the high-resolution weighted single hemisphere analysis ( 500 regions ) . First , comparing the metric values of the cortical network with those of its surrogates , we note that the cortical network tends to exhibit more similar values to its reduced than to its spatial surrogates for several network measures . One could argue that this may simply originate from the fact that the spatial surrogates are in general more randomised , and hence less similar to the cortical network , than reduced surrogates ( see QS in Figure 7 ) . However , while similarities in the expression of higher level network properties are certainly expected to be related to the extent of similarity on the lowest level of single connections , considering solely the overlap in the connection sets does not satisfactorily explain all observed tendencies . Indeed , as we showed in Results/Topological similarity , spatial surrogates are equally different from each other in their connection sets than from the cortical network , and yet their network properties are highly similar , but significantly different from that of the cortex . The overlap QS between connection sets alone is therefore not a good predictor of the obtained results , supporting our reasoning about the observed differences being attributable to the particular connectivity of the cortex – to its predominantly local connectivity and the specific arrangement of its long-range connections ( see Table 1 ) . Secondly , Figure 7 assesses the consistency of our analyses across all investigated cortical network types ( the five main rows of Figure 7 ) . We start by noting that the results for several measures , most notably clustering coefficient , efficiency , small-worldness and modularity , are highly consistent across all investigated cortical network types . There is , however , some disagreement in the results of other complex network measures , specifically the k-core/s-core and rich-club metrics , across the various cortical networks . Generally , these disagreements indicate that the significance of the corresponding network properties ( in terms of their relationship to the corresponding surrogate ensembles ) may depend on the resolution the cortical network is observed at , or on the inclusion/exclusion of connection strengths ( estimated number of fibres constituting the fibre bundles linking the regions ) , see detailed discussion in Results . Most notably , at the s-core/k-core metric , the strength of the cortical core only becomes visible in the high-resolution network , indicating a change in the organisation of the cortical connectivity at the different observable network resolutions and underlining the importance of multi-resolution approaches in connectome research . Specifically , on low resolution we found that the relatively weak cortical k-core is composed of 80% of the entire cortex , suggesting a more ‘homogeneous’ ( non-centralised ) connectivity between larger cortical regions on low network resolution . In contrast , on high-resolution the cortical network demonstrates a relatively small ( 10% ) , highly localised and significantly strong core structure , indicating a rather centralised organisation at the finer connectivity of the cortex . These findings are largely consistent with previous results on the s-core of the low-resolution [13] and high-resolution [12] cortical connectivity , and support the notion that , as we map the brain's network on increasingly higher resolutions , observed connectivity necessarily becomes sparser , leading in turn to the observation of fundamentally different organisation features at the various resolutions [78] . In this study , we focused on two distinguishable , supposedly competing factors that shape brain structure: economic pressure and functional pressure [18] . We note , however , that there are other important factors , such as evolutionary or developmental processes , that are likely to impose certain basic constraints on brain architecture [18] . Future extensions of this study may need to incorporate certain aspects of these further constraints , for example by generating surrogate networks via some neurobiologically informed developmental model [84] . It is also important to consider the accuracy of the cortical connectivity dataset used here . It is well known ( and indeed increasingly articulated ) that diffusion magnetic resonance imaging ( dMRI ) based tractography techniques suffer from certain biases and constraints , such as limitations in the ability to track fibre crossings and wide angular changes along the trajectory of the fibre tract [85] , [86] . Therefore , in the current absence of comprehensive tract-tracing data in the human brain , it will be important that the hypotheses and computational findings of our study are tested against the increasingly complete and accurate maps dMRI techniques will be delivering in the future . Relatedly , it is likely that the spatially constrained surrogate analysis introduced in this study may give insights into the relative significance and potential origin of certain properties of the brain networks of other species , such as the cat [17] or the macaque [75] , for which tract-tracing data is available . Being a real complex network with a diverse and extraordinarily complex set of functions to carry out , it is not surprising that the cortex adopts and takes advantage of several functionally beneficial organisation patterns , even given the additional constraints imposed by wiring constraints [18] . Small-world architecture has been shown to naturally foster high dynamical complexity [9] , [87] , which is one of the hallmarks of brain activity [88] and has been associated with conscious states involving the efficient coordination of multiple sensorimotor modalities in generating flexible behaviour [89] . Modularity is widely acknowledged to promote network robustness and evolvability by minimising dependencies and isolating the effect of local mutations and disturbances [2] . It also has been shown to increase dynamical metastability [63] thus hindering the pathological cases of prolonged synchronisation and seizures [90] and again supporting functional flexibility [91] . Hierarchically modular organisation has been found to facilitate limited sustained network activity [92] , it hence may serve a crucial role in maintaining the critical functional range in which the human brain operates [93] . Furthermore , the strong central core as well as the distributed and yet densely inter- and intra-connected rich-club structure may play a central role in facilitating efficient global functional integration and information flow in the cortex [13] , [25] , [74] hence providing the structural basis of various cognitive integration processes , from sensorimotor integration through attention to higher cognition and consciousness [77] , [94] . Combining all these findings into a single description of the structural connectivity of the human cortex , our results outline a hybrid , reasonably centralised and hierarchical , but nevertheless strongly modular anatomical architecture , with a remarkably strong central network core . Consistent discovery of characteristic network properties of the human connectome in this and previous studies emphasises a fundamental question: What factors contribute to the small-world , modular , hierarchical and centralised features of the cortical connectivity ? Our results , extending those of earlier studies [51] , [95] , [96] , support the notion that the emergence of these network properties is shaped by a complex interaction involving economic pressures ( towards minimising wiring and running cost of the brain ) and functional pressures ( towards stable , reliable and adaptive operation of the brain ) [18] . In this study we characterised how much each specific network property depended on the first of these factors , economic pressures , and we found that the level of dependency differed for different properties . Our results suggest that the more independent properties , such as the small-world , modular and core structure of the cortex , may be more related to the function of the brain than the more dependent ones , such as hierarchical organisation and rich-club patterning , which may be primarily driven by economic pressures . These results motivate further computational and experimental research to uncover the specific ways in which economic and functional pressures complement , reinforce or counteract each other in shaping the large-scale architecture of the human brain . | Macroscopic regions in the grey matter of the human brain are intricately connected by white-matter pathways , forming the extremely complex network of the brain . Analysing this brain network may provide us insights on how anatomy enables brain function and , ultimately , cognition and consciousness . Various important principles of organization have indeed been consistently identified in the brain's structural connectivity , such as a small-world and modular architecture . However , it is currently unclear which of these principles are functionally relevant , and which are merely the consequence of more basic constraints of the brain , such as its three-dimensional spatial embedding into the limited volume of the skull or the high metabolic cost of long-range connections . In this paper , we model what aspects of the structural organization of the brain are affected by its wiring constraints by assessing how far these aspects are preserved in brain-like networks with varying spatial wiring constraints . We find that all investigated features of brain organization also appear in spatially constrained networks , but we also discover that several of the features are more pronounced in the brain than its wiring constraints alone would necessitate . These findings suggest the functional relevance of the ‘over-expressed’ properties of brain architecture . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"connectomics",
"neuroanatomy",
"anatomy",
"nervous",
"system",
"biology",
"and",
"life",
"sciences",
"neuroscience"
]
| 2014 | Influence of Wiring Cost on the Large-Scale Architecture of Human Cortical Connectivity |
Cognitive control , which continues to mature throughout adolescence , is supported by the ability for well-defined organized brain networks to flexibly integrate information . However , the development of intrinsic brain network organization and its relationship to observed improvements in cognitive control are not well understood . In the present study , we used resting state functional magnetic resonance imaging ( RS-fMRI ) , graph theory , the antisaccade task , and rigorous head motion control to characterize and relate developmental changes in network organization , connectivity strength , and integration to inhibitory control development . Subjects were 192 10–26-y-olds who were imaged during 5 min of rest . In contrast to initial studies , our results indicate that network organization is stable throughout adolescence . However , cross-network integration , predominantly of the cingulo-opercular/salience network , increased with age . Importantly , this increased integration of the cingulo-opercular/salience network significantly moderated the robust effect of age on the latency to initiate a correct inhibitory control response . These results provide compelling evidence that the transition to adult-level inhibitory control is dependent upon the refinement and strengthening of integration between specialized networks . Our findings support a novel , two-stage model of neural development , in which networks stabilize prior to adolescence and subsequently increase their integration to support the cross-domain incorporation of information processing critical for mature cognitive control .
Cognitive control refers to the ability to execute voluntary , goal-directed behavior [1–3] . It requires flexible and adaptive coordination of core executive systems that are supported by integration among widely distributed , specialized brain circuitries [4] . The core components of cognitive control are available early in development [5] . However , in adolescence , cognitive control abilities become significantly more reliable and flexible , as response accuracy and speed stabilize in adulthood [6] . These developmental gains in information processing occur in parallel with brain maturational events , including synaptic pruning [7] and myelination [8] , which predominantly enhance collaboration among brain systems [9] . The nature of the interaction between brain network maturation and cognitive development during adolescence is not well understood [10] , limiting our ability to understand the neural basis of psychopathologies that emerge at this time , many of which are characterized by deficits in cognitive control [11] . Characterizing functional brain network interactions during the resting state ( i . e . , while the subject is not engaged in any particular task ) has become a valuable emerging approach for investigating the brain basis of cognitive development . Studies using this approach have revealed roles for these networks in supporting cognitive control [4 , 12] . Approximately 20 functional networks have been identified in the functional connectome [13] , including sensory networks , such as the somatomotor ( SM ) and visual networks; cognitive networks , such as the fronto-parietal ( FP ) and cingulo-opercular/salience ( CO/Salience ) networks; and a task-negative default mode ( DM ) network [14] . Each functional network operates as a module within the full connectome . Networks are demarcated by dense internal connectivity [15 , 16] , defining a foundational organization for the functional brain . Thus , network organization refers to the network affiliation of each region of the connectome . Initial studies characterizing age-related changes in functional network organization suggested that the organization of these networks continued to change into adulthood [17] , such that development proceeded from short-distance anatomical networks in infancy and childhood , to long-range , widely distributed networks in adulthood [17–20] . However , age-related differences in head motion artifacts may have confounded the connectivity distance findings [21–23] . Advances in data processing methods [21–23] and recent findings suggest that foundational aspects of functional network organization are established early in development , while processes related to network integration continue to mature into adulthood [24] . Network integration refers to the level of functional coupling between networks , measured by participation coefficient ( PC ) , a graph theoretical construct [25] . PC is a particularly useful construct to measure network integration , given its sensitivity to between-network connectivity , while maintaining robustness to the total number of connections ( degree ) . Degree-based measures of integration have been shown to be dependent on the size ( number of nodes ) of a network and therefore can skew results towards a greater number of hubs within larger networks , such as the default mode network [26] . PC is normalized by the degree of the node . As a result , increases in PC are driven mainly by increases in the number of between-network connections . Properties of network organization and integration could parallel cognitive development , which is characterized by enhanced adaptive and flexible integration of mature core control components [1] . Thus , in the present study , we sought to identify whether age-related changes in functional networks are determined by changes in network organization and/or network integration and whether these changes are related to developmental improvements in cognitive control . We applied graph theory [27 , 28] to a rich developmental resting-state functional magnetic resonance imaging ( RS-fMRI ) dataset obtained in 10–26-y-olds who also performed the antisaccade task . In this inhibitory control paradigm , subjects fixate a central target on a computer screen . A stimulus is then presented at an unpredictable horizontal location . Subjects are instructed to refrain from making a saccadic eye movement towards the stimulus ( i . e . , inhibitory response ) and instead make a voluntary saccadic eye movement to the mirrored opposite location on the horizontal meridian . Given that core cognitive components are on line by childhood and that the ability to adaptively and flexibly engage these components improves into adulthood [29–33] , we hypothesized that network organization , which supports component processes , would not change with age , but that network connectivity strength and integration , which both support interaction between components , would increase with age . In turn , we hypothesized increased control network integration would predict age-related improvements in cognitive control , as measured by the antisaccade task .
We used a previously defined functional connectome parcellation of 264 functional regions of interest ( ROIs ) across cortical , subcortical , and cerebellar structures [14] in a sample of 192 individuals , aged 10–26 y old ( Table 1 ) . For each subject , we correlated the time series of each ROI with that of every other ROI . We then formed group matrices by averaging each subject’s connectivity matrix within categorical age groups ( 10–12- , 13–15- , 16–19- , and 20–26-y-olds ) ( See Materials and Methods ) ( Fig 1A ) . For each group , we partitioned the full functional connectome into modules using Newman’s Q-metric coupled with an efficient optimization approach [15 , 34 , 35] across network densities ranging from the top 1% to 25% of pair-wise correlations in terms of correlation strength . Notably , Newman’s Q-algorithm returns modules of densely interconnected nodes . We interpret these modules as being functionally connected collections of brain regions sub-serving common functions and therefore refer to them as functional brain networks . The representative network partition of the full connectome was given a threshold of a density of 10% ( Fig 1B ) to partition the network into a meaningful structure while maintaining high connectedness , which would be limited with lower thresholds . This approach identified more comprehensive networks compared with those incorporating lower thresholds [14] , such that a single network encompassed the cingulo-opercular , subcortical , and salience networks . We refer to this network , which includes regions critical to cognitive control , as the CO/Salience network . We tested changes in network organization using normalized mutual information ( NMI ) , which measures the mutual dependence of two variables ( i . e . , how much information in variable one is also contained in variable two ) . NMI values range from 0 to 1 . A value of 0 indicates no mutual dependence ( no shared information ) , while a value of 1 indicates complete dependency ( completely shared information ) . We calculated NMI for networks between consecutive age groups and between children and adults ( Fig 1B ) . We used a random permutation test to compare observed NMI values to a null distribution of 1 , 000 NMI values . For the adult versus child contrast , observed NMI = 0 . 73 ( null mean [M] = 0 . 68 , null standard deviation [SD] = 0 . 07 ) ; between children and early adolescents , NMI = 0 . 67 ( null M = 0 . 73 , null SD = 0 . 08 ) ; between early adolescents and late adolescents , NMI = 0 . 69 ( null M = 0 . 76 , null SD = 0 . 06 ) ; and between late adolescents and adults , NMI = 0 . 77 ( M = 0 . 70 , SD = 0 . 06 ) ( Fig 2 ) . Importantly , all observed NMI values fell maximally just over one standard deviation of the null mean , indicating no significant differences in network organization from late childhood into adulthood . To provide statistical evidence for findings reflecting stable network organization , we took a Bayesian approach , weighting evidence in favor of the null hypothesis ( stable network organization ) versus the evidence for the alternative hypothesis ( dynamic network organization ) [36] . First , we generated a distribution of observed NMI values by performing a leave-one-out cross validation . We removed one subject from each group for any given contrast and calculated NMI on the remaining group-averaged matrices . Then , we compared the resulting distribution to the previously generated null distribution for each contrast by calculating the Jeffreys-Zellner-Siow ( JZS ) Bayes factor [36] . Values greater than 1 provide evidence supporting the null hypothesis , while values between 0 and 1 provide support for the alternative hypothesis . With respect to the null hypothesis of stable developmental network organization , values ranging from 1 to 2 indicate anecdotal evidence and from 3 to 10 , substantial evidence . For children versus early adolescents , JZS Bayes factor = 3 . 82; for early adolescents versus late adolescents , JZS Bayes factor = 2 . 49; for late adolescents versus adults , JZS Bayes factor = 5 . 34; and for children versus adults , JZS Bayes factor = 8 . 01 . These results indicate substantial evidence in favor of stable network organization throughout late childhood , adolescence , and adulthood . Importantly , these results were robust across network densities; thus , our results were not due to our choice of representational network density ( S1 Table ) . In addition to group-averaged matrices , we also calculated NMI between modules defined on the basis of individual subject data and the group-averaged adult module assignments to provide an analysis of subject variability . No significant differences were observed between groups , as any potential between-group variability was found to be smaller than that of within-group variability ( S1 Fig ) . Given network organization is on line by childhood and remains stable throughout this developmental period , it cannot account for cognitive changes during adolescence . Hence , we investigated developmental changes in network connectivity strength within networks ( reflecting the integrity of specialized networks ) and between networks ( reflecting the integration of information processing across functional domains ) . First , we partitioned each group-averaged matrix into networks according to the adult network assignment . Consecutive age group comparisons of within- and between-network connectivity were conducted using a two-tailed t test that was Bonferroni corrected for multiple comparisons ( p < 0 . 01 ) . Age-related changes in connectivity strength were unique to developmental stages . From childhood ( 10–12 y ) to early adolescence ( 13–15 y ) , there was a global decrease in connectivity strength for both within-network and between-network connectivity ( Fig 3A and 3B ) ( p < 0 . 05 , corrected ) . From early adolescence ( 13–15 y ) to late adolescence ( 16–20 y ) within-network connectivity remained stable , while between-network connectivity increased across networks , with the exception of DM/FP network connectivity , which remained stable ( Fig 3A and 3B ) . Lastly , from late adolescence ( 16–19 y ) to adulthood ( 20–26 y ) , within-network connectivity strength again decreased , while between-network connectivity continued to increase ( Fig 3A and 3B ) . These results indicate that the transition to adult-level network connectivity is characterized by a shift from predominance of within-network connectivity to reliance on between-network connectivity . Together , these results suggest that increased collaborative brain function may underlie improvements in cognitive control . Next , we examined the presence of distance-related changes with development [17 , 19 , 20] . In the present study , age-related changes in connectivity strength between ROI pairs were assessed by subtracting each pairwise relation of the averaged child connectivity matrix from the averaged adult connectivity matrix . We also calculated Euclidean distance for each pairwise relation and regressed the change in connectivity strength against Euclidean distance ( Fig 4 ) . Results showed that Euclidean distance accounted for a non-significant amount of the variance in change in connectivity with age ( R2 = 0 . 002 , p > 0 . 05 ) , indicating distance alone does not play a significant role in connectivity strength changes from childhood to adulthood [17 , 19 , 20] . We also contrasted the distributions of the top 100 increasing and decreasing connections in terms of connectivity strength between children and adults and found no significant differences ( p = 0 . 33 ) . In addition to characterizing age-related changes in the strength of connectivity both as a function of network organization and as a function of distance , we also aimed to quantitatively characterize the distribution of these between-network interactions using graph theory . Brain regions ( nodes ) within networks may either contain connections ( links ) solely to nodes within the same network or may also contain between-network links . A node that has distributed links across multiple networks can be regarded as a highly integrated region ( Fig 5A ) . Here , we operationally define integration as the level to which a region contains distributed links from its “home” network to a foreign network . Participation coefficient ( PC ) is a graph theoretical construct that is used to calculate integration between brain networks [25] . PC refers to the level to which a node establishes links to foreign networks , with values ranging from 0 to 1 . Nodes that link solely to other nodes within their “home” network would have a PC of 0 , while nodes with many distributed between-network links would have a PC closer to 1 . Delineating the level of integration using a node’s PC extends beyond defining the degree ( i . e . , number of links ) of a node , to defining the relative importance of those links with other networks [16] . To analyze developmental trajectories of integration at the network level , we calculated PC for every node within individual subject matrices at each network density . As an important aside , to remove the arbitrary bias in thresholding , all subsequent calculations involving PC are represented as the mean value across the range of network densities . Though we chose this method , PC across all nodes is significantly positively correlated with the PC of all nodes at each network density ( S2 Fig ) . If our results were only driven by a specific threshold ( e . g . , 5% ) , but not others ( e . g . , 20% ) , a significant relationship between mean PC and the specific threshold driving the effect ( 5% in this example ) would exist , but would not exist in others ( 20% in this example ) . This provides evidence that PC is robust to any biases that could be introduced by thresholding . For each subject , nodes were grouped according to the network to which they were assigned in the adult group . Then , we calculated the mean PC value for each network and tested each network for significant age-related effects on individual subjects , fitting both linear and inverse regression models , which are known to best fit this period of development [37] . The choice of superior model fit was made quantitatively , using Akaike information criterion ( AIC ) . The PC of the CO/Salience network significantly increased over the age range studied ( R2 = 0 . 09 , t = 3 . 74 , p < 0 . 001 ) ( Fig 5B ) , optimally fit with an inverse model . No other network displayed age-related changes in PC for either linear or inverse models ( p > 0 . 05 ) ( Sheet “Fig5Fig6” in S1 Data ) . One purported role of the CO/Salience network is the maintenance of cognitive control . Thus , increased integration of the CO/Salience network with other brain networks may underlie improvements in cognitive control performance during adolescence . We tested this hypothesis by investigating associations between network integration and behavioral performance in the antisaccade task . To identify any long-term fluctuations in PC that may not be captured at the individual subject level , we sorted individual subject matrices by age and then calculated average subject correlation matrices using a moving average approach ( see Materials and Methods ) . After calculating PC for each region within each moving average group , we computed the mean PC within each network . We then fit linear , inverse , quadratic , and cubic regression models to the data , with the best fit model defined as the one with the lowest AIC ( Fig 5C ) . The best fit model for the CO/Salience network was an inverse fit ( R2 = 0 . 59 , p < 0 . 05 ) , showing an increase in PC from late childhood through approximately 14 y of age , followed by relative stability ( Fig 5C , black curve ) . The quadratic model best fit age-related changes in the DM network ( R2 = 0 . 28 , p < 0 . 05 ) , which decreased in PC throughout adolescence , but increased slightly into early adulthood ( Fig 5C , red curve ) . A quadratic model best fit the visual network ( R2 = 0 . 51 , p < 0 . 05 ) , with peak levels of integration occurring late in adolescence ( Fig 5C , blue curve ) . A cubic model best fit the FP network ( R2 = 0 . 29 , p < 0 . 05 ) , where PC increased from late childhood through approximately 14 y of age before declining from approximately 14 to 20 y , and then increasing again throughout early adulthood ( Fig 5C , yellow curve ) . Lastly , the SM network remained relatively stable throughout development ( R2 = 0 . 01 , p > 0 . 05 ) ( Fig 5C , cyan curve ) . The fact that no other network demonstrated significant age-related effects in the individual subjects analysis compared to the moving average approach suggests the lack of differences is likely due to a high amount of individual subject variability . The antisaccade task is a particularly robust test of inhibitory control that reliably shows sensitivity to cognitive development through adolescence as accuracy and reaction times ( RTs ) during successful response inhibition improves through adolescence [38–40] . First , we tested the effect of age on accuracy and RT separately , with age modeled as both a linear and an inverse function . As is typical for the adolescent age range [37] , all regression models involving age were best fit by an inverse model , as determined by lower AIC , compared to linear models . Similar to previous studies [38–44] , we found developmental increases in the accuracy of correct inhibitory response ( R2 = 0 . 14 , t = 5 . 77 , p < 0 . 0000001 ) and decreases in RT through the adolescent period ( R2 = 0 . 13 , t = -5 . 51 , p < 0 . 00001 ) ( Fig 6A and 6B ) ( Sheet “Fig5Fig6” in S1 Data ) . Next , we tested the association between PC of the CO/Salience network ( i . e . , CO/Salience network integration ) and antisaccade accuracy and RT , controlling for age . Results showed no association between CO/Salience network PC and accuracy ( p = 0 . 34 ) . However , as CO/Salience network PC increased , RT to correct inhibitory responses decreased ( t = -2 . 09 , p = 0 . 03 ) ( Fig 6C ) , suggesting that greater CO/Salience network integration supports timely successful inhibitory control . Notably , no other network displayed a significant relationship between PC and accuracy or RT ( all p > 0 . 05 ) . Given the relationship between age and both antisaccade performance and CO/Salience network PC , we assessed whether CO/Salience network PC moderates the relationship between antisaccade performance and age . To test this , we ran two moderation analyses , one including CO/Salience network PC as a moderator of age and antisaccade accuracy and a second including CO/Salience network PC as a moderator of age and antisaccade RT . In each model , both regressors were centered prior to model fitting . CO/Salience network PC did not significantly moderate the relationship between age and accuracy ( p > 0 . 05 ) . However , CO/Salience network PC did moderate the relationship between age and correct antisaccade RT ( R2 = 0 . 16 , t = -3 . 28 , p < 0 . 001 ) . To identify when in development this interaction was most prominent , we investigated effects on RT within age groups by performing a median split of CO/Salience network PC ( Fig 6D ) . We observed a significant difference in individual subjects within the child group ( 10–12 y ) between RTs of subjects with high versus low CO/Salience network PC . Lower CO/Salience network PC resulted in slower RTs , while higher CO/Salience network PC resulted in faster RTs ( t = 2 . 84 , p = 0 . 02 , Bonferroni corrected ) . When we extracted the data for each subject , the results showed that as PC increased , antisaccade RT decreased ( R2 = 0 . 18 , t = -2 . 99 , p = 0 . 005 ) ( Fig 6E ) . In order to identify the contribution of regions of interest ( ROIs ) to age-related differences in network integration , which is overlooked when averaging at the network level , we tested each ROI in the network for significant increases in PC across age groups . Specifically , we permuted the connectome 1 , 000 times between consecutive age groups to generate null distributions for each brain region . Here , we report significant regional increases in PC in a stage-like manner throughout development ( Fig 7 ) .
Within the human functional connectome , densely interconnected brain regions are organized into well-defined functional networks , subserving sensory , motor , and cognitive functions . Our findings indicate that this network organization is stable between 10 and 26 y of age , countering earlier findings that suggested developmental changes in network organization reflect a shift from localized to distributed organization , which may have been confounded by head motion artifact [17 , 21 , 22 , 45 , 46] . The current study applied a wide array of advanced preprocessing steps to limit head motion artifact , including wavelet despiking [47] , simultaneous bandpass filtering the time series data and nuisance regressors [23] , as well as scrubbing [21] . These results suggest that , after controlling for head motion , there are no changes in network organization from late childhood to adulthood . Previous studies found that many aspects of human functional network topology remain stable throughout adolescence , including small-worldness [17 , 18 , 48] , global efficiency , and hub organization [24] . Combining these findings with our results showing the stability of network organization , we see strong evidence that the large-scale organization of functional networks is present by late childhood , possibly even earlier . Despite the fact the brain undergoes continual structural maturation of both gray and white matter [8 , 49–51] , key fundamental properties of large-scale functional circuitry , including organization , are stable throughout late childhood to adulthood . While non-significant age-related changes to network organization cannot be concluded through inferential statistics , Bayesian inference via JZS Bayes factors allowed us to test the likelihood of the null versus the alternative hypothesis [36] . Using this method , we confirmed the finding that network organization does not change significantly with age . Our results show age-related changes in connectivity strength . Within-network connectivity strength decreased with age , suggesting that maturity results in network refinements akin to pruning unnecessary connections , which improves signal transmission within networks . On the other hand , we found between-network connectivity strength decreased into early adolescence and subsequently increased into adulthood , ultimately enhancing the ability for different networks to collaborate . Interestingly , adolescence demarcated the period when between-network connectivity began to increase , perhaps reflecting a qualitative shift in network interactions towards collaborative network functioning . The overall trend towards increased between-network connectivity is at odds with a previous study by Stevens and colleagues , who found causal between-network coupling decreased in strength [52] , reflecting greater segregation of specialized networks . However , this study used an independent components analysis approach to define functional networks , which only coarsely correspond to the canonical networks used in the current study . Furthermore , this study was conducted before advances concerning mitigation of head-motion–related artifacts . Changes in within- and between-network connectivity strength were sensitive to network organization , not solely by the distance between regions , as initial studies had suggested [17 , 19 , 20 , 53] . Divergences from previous results are not surprising given our implementation of recent advances in head motion control that minimized its confounds on age differences in connection strength as a function of distance [21 , 22] . Distance-related changes in connectivity strength by age have been found after controlling for head motion , albeit with a weaker effect than previously reported , in a sample that included children younger than those in the current sample ( 8 versus 10 y of age ) [53] . Decreasing short-range connectivity and increasing long-range connectivity may be specific to an earlier developmental stage , when greater changes in white matter connectivity are occurring [8] . These results suggest that the adolescent transition to maturity is a period of refinements in connectivity within stable networks and concomitant increases in connectivity across widely distributed circuitry . While between-network connectivity increased with age , the distribution of links ( i . e . , integration ) among networks remained stable for most networks studied . This suggests that the framework for network integration is available by childhood , with continued increases in the strength of these established between-network links . An exception , however , was the CO/Salience network , which displayed age-related changes in integration with other networks , as assessed by PC . The CO/Salience network is involved in maintaining a task set , saliency , and configuring sensory information , cognitive state , and motor output [12 , 54] . The continued enhancement of CO/Salience network integration follows what is known about the development of cognitive control . Core cognitive control abilities are present early in development , but the consistent successful implementation of control continues to improve into adulthood . This developmental pattern has been found for a wide range of cognitive control tasks , such as the antisaccade , go-no-go , and stroop tasks [33 , 55] . Our findings of stable network organization , coupled with increased integration , are consistent with these behavioral findings , suggesting that the underlying architecture supporting mature brain functioning is present early in development , with refinements continuing into adolescence . Age differences in integration patterns at the regional scale within the CO/Salience network corroborated the network-level findings . From childhood into early adolescence , specific regions that drove increased integration of the CO/Salience network included the right aIns , bilateral dACC , anterior and mediodorsal nuclei of the thalamus , and putamen . Both the aIns and dACC are extensively anatomically connected to many major brain networks across cortical and subcortical regions [56 , 57] . Together these regions drive a control network guiding mental activity and behavior through an interaction of cognitive , affective , and homeostatic functions [54 , 58–61] . We observed an increase in the number of links between the CO/Salience network and the SM network from every region that became more integrated within the CO/Salience network , enabling more rapid access from this control system to the motor system to guide goal-directed behavior [60] . Specifically , the right aIns has been shown to play a critical developmental role as an outflow hub in directing cognitive control processes , having greater directed causal influence on other brain regions ( dACC and posterior parietal cortex ( PPC ) ) critical for proper cognitive control execution in adults compared to children . Furthermore , these functional refinements were shown to be supported structurally via enhanced white matter fiber density with development between the right aIns and PPC [62] . Additionally , it has been shown that the right aIns increases in connectivity strength to regions within network ( e . g . , dACC ) and between networks ( e . g . , DLPFC and PCC ) , supporting its increased role in network integration over the adolescent period [63] . Due to its roles in detecting salient stimuli and acting as a switch between large-scale networks [13 , 61] , the aIns likely plays a particularly important role in normative development , supporting enhanced integration of multiple brain processes . In addition to the right aIns , the dACC also plays a critical role in cognitive control execution [40 , 64] . Using a multimodal approach , Fjell and colleagues found the surface area and white matter integrity of the dACC accounted for a significant portion of variance in performance on a flanker task [65] . In sum , much like the aIns , the dACC plays a critical role in control abilities and shows a protracted development . In support of their critical developmental role , there is evidence that abnormal engagement of the aIns and dACC may underlie neurodevelopmental disorders , such as autism [53 , 60 , 66 , 67] . Many of the regions within the CO/Salience network that significantly increased in integrative properties were subcortical , including the putamen and thalamus . These regions show larger changes than cortical areas with respect to fractional anisotropy in white matter , increasing 30% to 50% from childhood into early adulthood [68] , and also show a protracted neurophysiological development [69] . This parallels our findings of increased integration of these subcortical structures with cortical networks . Given that adolescence is a period of enhanced sensation seeking [13 , 37] , the steep increase in the integrated nature of these regions with other brain networks during early adolescence suggests a mechanism by which motivational systems are reconfigured with more cognitive , sensory , and affective systems [70] . In agreement with an extensive literature [33 , 40] , we found age-related decreases in reaction times of correct inhibitory responses . Our network analyses indicated that increased CO/Salience network integration predicted faster RTs on the antisaccade task , underscoring the importance of the CO/Salience network integrating with other networks , subserving cognitive control . Importantly , we found that CO/Salience network integration moderated decreases in antisaccade latency as a function of age . This moderation was significant in the transition from late childhood to early adolescence , when ( at both the network and the regional scale ) the CO/Salience network became significantly more integrated with other functional networks . Together , these results indicate that development brings greater integration between the CO/Salience network , supporting sustained cognitive control [12] , and regions that underlie action such as the SM network , resulting in the ability to generate quicker execution of correct cognitive control signals [64] . Although intrinsic , spontaneous coupling between regions at frequencies <0 . 1 Hz has been studied for nearly 20 y , the neural substrate and the meaning of the slow frequency signal remains unclear [71 , 72] , though functional networks observed using fMRI have also been identified using magnetoencephalography [73] . Many ROI-ROI pairs demonstrate high correlations between their time courses despite a lack of monosynaptic connections [74 , 75] . Though the functional purpose of spontaneous slow frequency BOLD oscillations is not known , a range of possibilities exist . Resting-state functional networks may be groups of regions that often coactivate in task-based settings , reflecting a history of coactivation [12 , 76 , 77] . This interpretation is supported by studies finding strong resting-state correlations , despite the lack of a direct anatomical connection . However , the existence of strong functional connectivity in the absence of direct anatomical connections allows for other alternatives , including the notion that resting-state networks are constantly sampling a possibility of configurations , constrained by anatomy , to make predictions about optimal network configurations for a given input [72] . Furthermore , over long timescales , such as in this study , resting-state functional brain networks are dependent on anatomical connectivity; however , at shorter timescales , numerous configurations are possible [78] . That said , changes in the framework of integration within the functional connectome during adolescence may reflect differences in the pattern in which information is shared across distributed neural networks . Specifically , from a graph theoretic view , the regions that significantly increased in participation coefficient are areas that integrate across multiple functional networks to a greater extent . Importantly , the role these brain regions play in integrating information may reflect a particular vulnerability for the emergence of psychopathology , which emerges during adolescence—a time when the brain is reorganizing the way it shares and processes information across these networks . This study was not without limitations . The sample was cross-sectional , undermining our ability to analyze subject-specific growth trajectories . We are also limited by some inherent drawbacks of fMRI , including residual head motion , though we took multiple processing steps towards mitigating these effects , including wavelet despiking , simultaneous bandpass filtering of the time series and nuisance regressors , and scrubbing . Additionally , 5 min of resting-state data is considered a minimum requirement for analyses of resting-state fMRI data , with recent pushes for longer acquisitions [75 , 79] . However , longer acquisitions may lead to even greater differences between age groups in head motion . Lastly , because PC was averaged over all nodes within a network , it is possible that some single brain regions could be driving this effect more than others . That said , we still found CO/Salience network increases in integration with age that moderated the relationship between cognitive control performance and age . This finding stresses the importance of network integration for adult-like cognitive control performance , rather than the maturation of any singular brain region . Future studies could aim to elucidate specific brain regions driving cognitive control maturation via integration .
One hundred and ninety-five subjects aged 10–26 y participated in this study ( Table 1 ) . Written informed consent was obtained from every subject and minors did sign assents . This research was approved by the University of Pittsburgh Institutional Review Board . A phone screen questionnaire was used to assess medical history and history of psychiatric disorders at the time of recruitment . Subjects were excluded at the time of recruitment if the subject or a first-degree relative currently or previously had a psychiatric disorder . Subjects also completed a battery of self-report measures of psychopathology . As determined through the interview process , neither subjects included in this study nor their first-degree relatives currently or previously had any neurological disease , brain injury , or diagnosed psychiatric illness . Substance use was assessed using the drug use and history questionnaire . Subjects included in this study were free from substance use or abuse . A post-scan questionnaire was used to inquire if subjects had fallen asleep . Sixteen subjects reported periods when they may have briefly drifted into sleep but none reported sleeping throughout the entire resting state scan . Data from three subjects were discarded due to excessive head motion . Therefore , we report data from 192 subjects . While age was considered as a continuous variable , some analyses considered developmental stages by binning ages after first sorting individual subjects by age , similar to methods used in the past to characterize changes in childhood ( n = 41 10–12 y olds ) , early ( n = 41 13–15 y olds ) and late adolescence ( n = 53 16–19 y olds ) , and adulthood ( n = 57 20–26 y olds ) . The antisaccade task was performed by subjects outside of the MR scanner on a separate day from the MR visit . For a full description of the antisaccade task used , see [80] . Briefly , neutral trials were extracted from an incentivized antisaccade task , consisting of reward , loss , and neutral trials . There were a total of 40 of each trial type . Each neutral trial began with a white central fixation , which then turned red for 1 . 5 s , prompting subjects to prepare a response . Next , a peripheral stimulus ( yellow dot at approximately 0 . 5 degree/visual angle ) appeared at an unpredictable location on the horizontal meridian ( ±4 and 8 degrees/visual angle ) for 1 . 5 s . Subjects were instructed to inhibit making a saccade towards the stimulus , and instead to saccade to the mirror location of the stimulus . Eye movement data were scored online using interfaced E-Prime ( Psychology Software Tools , Inc . , Pittsburgh , PA ) and ASL ( Applied Science Laboratories , Bedford , MA ) eye tracking software . A script detected if at any time during the first 1 , 000 ms a subject made a saccade to the stimulus or if no eye movement was generated . An auditory tone ( 1 , 163 Hz ) was played for 400 ms if the subject made a saccade to the stimulus . If the subject made a correct saccade a “cha-ching” sound ( 1 , 516 Hz ) was presented for 400 ms . Correct responses were defined as those in which the first eye movement in the saccade was directed toward the mirror location at a velocity greater than or equal to 30°/s [81] and extended beyond a 2 . 5°/visual angle from the central fixation . A response was considered incorrect when the first saccade was directed towards the target beyond a 2 . 5°/visual angle from central fixation , but were subsequently directed to the hemifield opposite the target , similar to previously published work [80] . In addition to the online scoring , eye data were scored offline by a technician for various saccade metrics , including correct trials and errors , as well as saccade latency , using ILAB software [81] and an in-house scoring suite written in MATLAB ( Math Works , Inc . , Natic , MA ) . A correct antisaccade response was one in which the first saccade following stimulus onset was towards the mirror location of the stimulus and extended beyond a 2 . 5 degrees/visual angle central fixation zone . Errors were defined as occurring when the first saccade following stimulus onset was directed towards the stimulus and extended beyond central fixation . Data were acquired using a 12-channel Siemens 3T Tim Trio at the University of Pittsburgh Medical Center Magnetic Resonance Research Center . The resting-state scan was acquired at the end of the scanning session and was always at the same time of acquisition for all subjects . For each subject , we collected 300 s ( 200 TRs ) of resting-state data . Structural images were acquired using a sagittal magnetization-prepared rapid gradient-echo sequence ( repetition time [TR] = 1 , 570 ms , echo time [TE] = 3 . 04 ms , flip angle = 8° , inversion time [TI] = 800 ms , voxel size = 0 . 78125 × 0 . 78125 × 1 mm ) . Functional images were acquired using an echo-planar sequence sensitive to BOLD contrast ( T2*; TR = 1 . 5 s , TE = 29 ms , flip angle = 70° , voxel size = 3 . 125 × 3 . 125 mm in-plane resolution , 29 contiguous 4-mm axial slices ) . During the resting-state scan , subjects were asked to close their eyes and relax , but not fall asleep . Functional images were preprocessed using AFNI [82] and Freesurfer [83] . Standard preprocessing steps were completed , including ( 1 ) normalization based on global mode , ( 2 ) wavelet despiking [47] , ( 3 ) simultaneous multiple regression of nuisance variables from BOLD data and bandpass filtering [23] at 0 . 009 Hz < f > 0 . 08 , and ( 4 ) spatial smoothing using a 6 mm full-width at half-maximum Gaussian blur . Freesurfer was used to segment gray matter , white matter , and ventricular voxels . Nuisance regressors included ventricular signal averaged from ventricular regions of interest ( ROIs ) , six head realignment parameters obtained by rigid body head motion correction , and the derivatives of these signals and parameters . In addition to wavelet despiking , we removed any remaining high motion volumes via a scrubbing procedure [21 , 22] . For the original 195 subjects , we calculated two quality control measures with respect to head motion , volume-to-volume framewise displacement ( FD ) and the root mean square derivative of fMRI timeseries ( DVARS ) . We censored and removed volumes in individual subjects that had an FD > 0 . 5 mm and DVARS > 5 , as well as the frame preceding the motion artifact and the two subsequent frames . FD is calculated on the original motion time series ( i . e . , before motion correction with wavelet despiking ) . On the other hand , DVARS is calculated after motion correction with wavelet despiking . Large DVARS values after wavelet despiking would indicate motion/artifact-related noise in the global signal ( i . e . , brain-wide change from one volume to the next ) still remained after despiking , which we did not observe ( Table 1: note DVARS after wavelet despiking is considerably lower in all four groups than DVARS calculated prior to wavelet despiking ) . Because we collected 300 s of data , subjects were dropped entirely if >20% of their volumes were removed , leaving the minimum amount of rest data for any subject 240 s . This procedure resulted in the removal of three subjects from further analyses . Of the remaining 192 subjects , only four did not contain a full 300 s of data . For each subject , nodes ( n = 264 ) were defined from the functional parcellation defined by Power and colleagues [14] . Coordinates were derived through fc-Mapping [84 , 85] and a meta-analytic procedure [14] , covering major brain systems involved in both tasks and rest . All ROIs were modeled as 10 mm diameter spheres around a center coordinate . For each subject , the timeseries of voxels within each ROI were averaged and then correlated to produce a 264 × 264 correlation matrix . Any comparisons made between correlations were transformed to z values using Fisher z ( r ) transformation , and then reconverted to Pearson r values for reporting and visualization . Network-level age-related changes were assessed using individual correlation matrices . For all other RS-fMRI analyses , age was treated as a categorical variable to assess stage-like developmental changes in graph metrics and changes in the distribution of connections between children ( aged 10–12 ) , early adolescents ( aged 13–15 ) , old adolescents ( aged 16–19 ) , and adults ( aged 20–26 ) . Notably , no standard for binning age groups over adolescence currently exists , though binning roughly follows Luna and colleagues [37] . Since short-distance correlations ( Euclidean distance <20mm ) can arise from artifacts [21] , these connections were not included in tests for age-dependent significant strength changes in connectivity . Since there is no ideal , biologically salient threshold that definitively defines functional networks , we explored a range of network densities from 1%–25% to avoid any thresholding bias . Results involving PC at the group level reflect values that are averaged across all network densities to remove any bias of a single threshold . For a representative network assignment , we chose a network density of 10% , since this threshold results in meaningful network organization ( i . e . , five networks ) , while maintaining full connectedness . Importantly , we did not impose network assignments according to [14] , since that would erode the ability to make conclusions concerning developmental changes in network organization . To define and examine the developmental trajectory of functional network organization , we partitioned the full connectome of 264 ROIs into modules using Newman’s Q-metric coupled with an efficient optimization approach proposed by Blondel et al . [15 , 34 , 35] . This method has been verified to be one of the best-performing community detection algorithms of undirected networks [86] . We then calculated normalized mutual information ( NMI ) to determine the level of similarity between network assignments across age groups , with values closer to 0 indicating dissimilar network assignments and values closer to 1 indicating similar assignments . NMI is a standard measure for assessing the degree of similarity between two distributions , which has been used to compare sets of network assignments in resting-state fMRI data [16 , 21] . NMI measures information shared between two probability distribution functions , specifically measuring how much knowing one distribution leads to certainty of the other . Furthermore , NMI will detect any type of relationship between two distributions , making it more robust than a simple correlation coefficient . In this way , we can empirically test the level of similarity of these distributions across subjects . To this end , we permuted the labels of individual matrices between contrasts 1 , 000 times to generate a null distribution of NMI values for each contrast . Matrices between groups were randomly shuffled and partitioned into functional networks , and NMI was calculated . Upon the finding that the observed NMI values fell around one standard deviation of the mean of the null distribution , we executed a leave one out cross validation to generate a distribution of observed NMI values for the following analysis . Because conventional significance testing does not allow stating evidence in favor of null findings , we implemented a Bayes factor alternative [36] to compare the observed NMI distribution with the null distribution . Values greater than 1 indicate the likelihood of stable functional network organization is “n” times more likely than the likelihood of developmental changes in functional network organization . A general concept in the development of functional networks is that they develop from “local to distributed” [17] . To test this hypothesis , given methodological improvements for head motion and a denser , more representative functional network [14] , we contrasted connectivity values from averaged weighted matrices in children versus adults for each ROI-ROI pair . Euclidean distance was also calculated for each pairwise relation . We then performed a simple linear regression with distance as a predictor of change in connectivity strength between the children and adult matrices . We also addressed changes in connectivity strength as a function of within- and between-network interactions . First , within each group-averaged matrix , we averaged all within-network pairwise relations and all between-network pairwise relations , separately . We then performed a two-tailed t test for each consecutive age contrast . We then wanted to test for significant increases or decreases in connectivity with respect to specific network interactions . To this end , within each group-averaged matrix , the average connectivity strength was calculated for each network . We then tested each combination of within-network ( e . g . , DM/DM network ) and between-network ( e . g . , DM/FP networks ) interactions to determine significant increases or decreases in connectivity strength between consecutive age groups . For each comparison , we ran a two-tailed t test to determine significance ( Bonferroni corrected for multiple comparisons ) . For each subject , we partitioned the full network into sub-networks imposing the module assignments from the adult group in the analysis outlined above , and subsequently calculated PC for every node within each group . PC is a graph measure quantifying the degree to which a node engages in inter-network communication [25 , 26] . Higher PC indicates more distributed between network connectivity , while a PC of 0 signifies a node’s links are completely within its home network ( within network ) . Nodal PCs were then averaged within each network and were tested for significant age-related effects using linear and inverse models . To determine any long-term fluctuations in PC that may not be captured at the individual subject data , we calculated average subject correlation matrices using a moving average approach , used previously in functional brain network data [17] and commonly used in economics research . Averaged group matrices were formed using a moving average of age-ordered subjects ( e . g . , group1: subjects 1–30 , group2: subjects 2–31 , … group163: subjects 163–192 ) , thus generating 163 groups of 30 subjects in each group . Each group matrix was then parcellated according to the adult network assignment and PC was calculated for each ROI within each group . For each group , the PC for ROIs within a network were averaged and plotted as a function of age . To test the hypothesis that the relationship between age and performance ( accuracy and RT ) on the antisaccade task is moderated by integration of the CO/Salience network with other functional networks , a hierarchical multiple regression analysis was conducted separately for accuracy and reaction time . If a significant interaction was observed , age groups were binned into the four age groups previously defined and a median split of the averaged PC within the CO/Salience network was conducted . Within each bin , we tested for significant differences in RT using a t test between high and low PC groups and corrected for multiple comparisons using the Bonferroni method . We sought to discover brain regions that significantly increased in the ability to integrate information from widespread functional networks using graph theory . PC was calculated for each node within each categorical age group . Importantly , the degree , or number of links a node has , was not considered as a metric for integration since network measures that are degree-based have recently been called into question in Pearson correlation RS-fMRI networks [26] . PC for each node was contrasted between each set of chronological age groups ( children versus early adolescents , early adolescents versus late adolescents , and late adolescents versus adults ) and between adults and children by subtracting the younger group’s PCs from the older group’s PCs resulting in four total contrasts . Permutation tests were conducted on each node to test nodes for significant changes in PC . To generate a null distribution of PCs for each node , subject labels were randomized within groups 1 , 000 times and PC was calculated for every node in each run . Contrasts between age groups were then generated by subtracting the PCs for each node for the younger group from the older group . This process was repeated for each age contrast . A significant increase or decrease in participation coefficient for a node was Bonferroni corrected for multiple comparisons . Within each group , and for each node that significantly increased in PC , we calculated the degree of the ROI to each network , including its “home” network , and then contrasted these values for consecutive age groups for comparison . The degree of a node is determined by the number of links a node has . This approach allowed us to contrast the distribution of links to each network between consecutive age groups ( i . e . , within-network versus between-network connectivity ) , giving us the ability to characterize the driving factor ( s ) behind the observed significant increases in PC . AFNI [82] and Freesurfer [83] were used to process MRI images . We used the Brain Connectivity Toolbox [28] in MATLAB ( The Mathworks , Natick , MA ) for network computations and statistical testing . For brain visualizations , we used the BrainNet Viewer [87] . These results provide evidence that the period of childhood through adulthood is characterized by increased integration of widely distributed but stable networks . As such , a critical component underlying the adolescent transition to adult-level execution of control is the refinement and strengthening of integration between existing specialized networks . In particular , the CO/Salience network continues to increase its integration with and , thus , its influence on other networks , providing a mechanism for developmental improvements in cognitive control . These findings support a novel two-stage model of adolescent brain development in which network organization stabilizes prior to adolescence , while the integration of information across widely distributed circuitry increases in the transition from adolescence to adulthood . | Adolescence is a unique period of brain development , with major changes occurring across the brain at many different levels of brain functioning . At the macroscopic level , the brain is composed of individual regions that collaborate in networks to perform diverse cognitive functions . Some networks of brain regions perform lower-level sensorimotor processing , while other networks orchestrate more complex functions , such as cognitive control . The affiliation of each region to a network is referred to as network organization . Brain regions not only can communicate with other regions belonging to their own network but also with regions in other networks . Brain regions that communicate with regions belonging to other networks display a high level of integration since they link their network with another network . We found that during adolescence , network organization does not change . However , integration continues to increase , underscoring the notion that brain function becomes more distributed and collaborative during this unique period of development . Furthermore , this increased network integration underlies improvements in cognitive control . Thus , we provide a network-based account for improvements in cognitive functioning during adolescence . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | The Contribution of Network Organization and Integration to the Development of Cognitive Control |
DNA polymerase ζ ( pol ζ ) is exceptionally important for maintaining genome stability . Inactivation of the Rev3l gene encoding the polymerase catalytic subunit causes a high frequency of chromosomal breaks , followed by lethality in mouse embryos and in primary cells . Yet it is not known whether the DNA polymerase activity of pol ζ is specifically essential , as the large REV3L protein also serves as a multiprotein scaffold for translesion DNA synthesis via multiple conserved structural domains . We report that Rev3l cDNA rescues the genomic instability and DNA damage sensitivity of Rev3l-null immortalized mouse fibroblast cell lines . A cDNA harboring mutations of conserved catalytic aspartate residues in the polymerase domain of REV3L could not rescue these phenotypes . To investigate the role of REV3L DNA polymerase activity in vivo , a Rev3l knock-in mouse was constructed with this polymerase-inactivating alteration . No homozygous mutant mice were produced , with lethality occurring during embryogenesis . Primary fibroblasts from mutant embryos showed growth defects , elevated DNA double-strand breaks and cisplatin sensitivity similar to Rev3l-null fibroblasts . We tested whether the severe Rev3l-/- phenotypes could be rescued by deletion of DNA polymerase η , as has been reported with chicken DT40 cells . However , Rev3l-/- Polh-/- mice were inviable , and derived primary fibroblasts were as sensitive to DNA damage as Rev3l-/- Polh+/+ fibroblasts . Therefore , the functions of REV3L in maintaining cell viability , embryonic viability and genomic stability are directly dependent on its polymerase activity , and cannot be ameliorated by an additional deletion of pol η . These results validate and encourage the approach of targeting the DNA polymerase activity of pol ζ to sensitize tumors to DNA damaging agents .
In eukaryotes , DNA polymerase ζ ( pol ζ ) is critical for the tolerance of many types of DNA replication blocks , by playing a central role in translesion DNA synthesis ( TLS ) . Primary replicative DNA polymerases ( pol δ or pol ε ) are stalled when they encounter many types of template DNA adducts or DNA sequences forming stable secondary structures . Such stalled replication forks are prone to formation of a dangerous DNA double-strand break . The process of TLS helps avoid catastrophes by using a lower fidelity DNA polymerase ( such as pol ζ or pol η ) , to incorporate nucleotides across from a lesion . TLS may occur either in S phase during primary DNA replication or in G2 phase during post-replication DNA synthesis . In yeast and in mammalian cells , pol ζ is important for this process , but it leads to endogenous and DNA damage-induced point mutations because of errors introduced during TLS [1–5] . Elimination of the pol ζ catalytic subunit Rev3l in mice leads to death during embryogenesis ( reviewed in [6] ) . Primary cells in culture also cannot survive in the absence of Rev3l , because chromosomal DNA breaks quickly accumulate [7 , 8] . Circumvention of damage-dependent checkpoints by SV40 large T antigen ( TAg ) immortalization of cells or by Tp53 knockout allows Rev3l-deficient cell lines to grow , but the cells continue to display gross chromosomal instability and DNA damage sensitivity [8–10] . Mice conditionally deleting Rev3l in a fraction of hematopoietic cells or in basal skin keratinocytes are viable , but exhibit enhanced tumor incidence , as a consequence of the chromosomal instability of Rev3l-null cells [7 , 11] . The hypersensitivity of REV3L-defective cells to some clinically-used DNA damaging agents indicates that REV3L is a possible target for enhancing the sensitivity of tumors to chemotherapeutic agents [12] . Although the consequences of pol ζ disruption are dramatic , it is not clear that these arise from a specific DNA polymerase defect . In mammalian cells , REV3L is a large protein ( >3000 amino acids ) , with multiple functional domains . The DNA polymerase domain occupies only the last third of the protein ( Fig 1A ) . The structural integrity of REV3L may be required in DNA processing complexes and for protein-protein interactions necessary to maintain cell viability and DNA integrity . Indeed , REV3L serves as a multi-DNA polymerase scaffold . The central region harbors two adjacent binding domains for REV7 ( gene name MAD2L2 ) . REV7 is necessary for pol ζ activity in vitro and serves an important function as a bridge protein for interaction with the REV1 protein [13–15] . REV1 in turn interacts with Y-family DNA polymerases that insert bases opposite sites of DNA damage and work in tandem with pol ζ [16–18] . REV7 also has other cellular functions in chromatin assembly and structure [19–21] . An N-terminal region of REV3 is conserved with yeast homologs [22] . At the C-terminus of REV3L [23] , an Fe-S cluster is present that binds two other subunits of the pol ζ enzyme , POLD2 and POLD3 . Both of these proteins also serve as subunits of the replicative DNA polymerase δ [23–26] . More recently , a conserved positively charged domain in the central region has been recognized as necessary for the efficient polymerase function of the recombinant protein [24] . Another domain in the central region has strong homology to the KIAA2022 gene ( S1 Fig ) . A provocative hypothesis has been put forward to explain the severe genotoxic effects of Rev3l deletion [27] . It was suggested that these are the consequence of the function of a second DNA polymerase , pol η ( gene Polh ) . As in mammalian cells , chicken DT40 cells with a disruption of pol ζ exhibit growth defects , chromosomal aberrations and DNA damage sensitivity [27] . Remarkably , it was reported that co-disruption of Polh and Rev3l corrects all of these phenotypes in DT40 . The suggested interpretation was that pol η and pol ζ always work together in bypass of DNA damage , and that a toxic intermediate is formed by pol η that cannot be resolved in the absence of pol ζ . It is clearly important to determine , in mammalian cells , whether the genome instability caused by pol ζ disruption is dependent on pol η . Here we describe experiments with knockout cells and a specific knock-in mouse model to test whether the catalytic activity of pol ζ is responsible for the phenotypes observed in pol ζ knockout mutants . We describe complementation of Rev3l-deficient mouse embryonic fibroblasts ( MEFs ) by expression of full-length human wild-type REV3L , and show that DNA polymerase-defective mutant REV3L cDNA is unable to complement cell survival or increased levels of DNA breaks . Using a Rev3l polymerase-dead knock-in mouse model , we show that specific disruption of the polymerase activity prevents the completion of embryogenesis . Finally , we tested whether pol ζ defects can be rescued by ablation of pol η function .
Rev3l deletion in mouse cell lines is associated with an elevated baseline level of DNA breaks and an increased sensitivity to DNA damaging agents such as cisplatin and UV radiation [3 , 8–10] . We wanted to test definitively whether these phenotypes are caused by the disruption of Rev3l . A pOZ expression vector harboring an IL2R selectable marker ( Fig 1A ) [28] was used to express human REV3L cDNA in Rev3l-deficient MEFs [8] . Cells were selected for IL2R expression by repeated cycles of magnetic bead sorting and clonal populations were isolated . The integrity of the expression vector was confirmed by PCR-based detection , and cells were assayed for expression of REV3L mRNA by real-time RT-PCR . Human REV3L was expressed in the Rev3l-deficient MEFs at about one-half of the normal endogenous level ( Fig 1B ) . Mouse cells expressing one or two alleles of Rev3l have indistinguishable low levels of spontaneous senescence , apoptosis , and chromosome aberrations [8] and there is no haploinsufficiency apparent regarding embryonic or adult viability in mice [7] . We expressed both wild-type REV3L ( WT ) , as well as REV3L with a dual point mutation ( ASM: D2781A; D2783A ) in residues essential for divalent metal binding in conserved DNA polymerase motif I . Equivalent changes in all other tested DNA polymerases inactivate Mg2+ coordination in the active site , and eliminate enzymatic activity [29 , 30] . We tested the growth of Rev3l-proficient and deficient cells expressing an empty vector ( EV ) , as well as deficient cells expressing WT and ASM REV3L cDNA . Rev3l-deficient cells experienced S-phase associated delay and mitotic failure , leading to a population doubling time that was longer than Rev3l-proficient populations [8] . REV3L re-expression in the deficient cell lines significantly decreased their doubling time to a level similar to Rev3l-/+ cells , whereas expression of the polymerase-inactive mutant had no effect ( Fig 1C ) . Deletion of Rev3l causes sensitivity to DNA damaging agents [8–10] . To determine whether REV3L expression could rescue this phenotype , cells were exposed to cisplatin or UVC radiation and cell survival was measured . Rev3l-deficient cells displayed the expected sensitivity to these damaging agents when compared to the Rev3l-proficient cells ( Fig 1D and 1E ) . Assays were repeated with multiple clones for each genotype . Expression of wild-type REV3L rescued the sensitivity to all three DNA damaging agents , but expression of ASM REV3L did not . Rev3l-deficient cells manifest an increased formation of DNA breaks in the absence of exogenous DNA damage . We measured a 10 to 20-fold increase in cellular micronuclei ( Fig 2A and 2C ) in Rev3l-defective cells , with 30–40% of all cells displaying micronuclei . The Rev3l defect was also accompanied by an increased frequency of DNA breaks as quantified by 53BP1 foci per cell , with a pronounced shift in distribution towards larger numbers of foci per cell ( Fig 2B and 2D ) . Expression of wild-type REV3L in Rev3l-deficient MEFs rescued both of these phenotypes , but expression of ASM REV3L did not . The frequency of sister chromatid exchange ( SCE ) was not decreased in Rev3l-deficient cells ( Table 1 ) , indicating that this mitotic recombination event is not impaired by a REV3L defect . These experiments demonstrate that sensitivity to DNA damaging agents and the presence of DNA breaks in Rev3l-deficient cells is caused by the absence of the REV3L protein , and REV3L polymerase activity is required for prevention of these phenotypes . We also investigated two reported human REV3L knockout lines designated 332 and 504 , derived from the Burkitt lymphoma cell line BL2 [31] . However , Rev3l mRNA is still transcribed in the 332 and 504 subclones , the subclones were no more sensitive to cisplatin than the parental BL2 , there was no significant increase in spontaneous double-strand break incidence in the subclones , and no complementation of the mild UV sensitivity was observed with Rev3L cDNA ( S2 Fig ) . These results and uncertainties regarding the targeting strategy ( S3 Fig ) indicate that the BL2 subclones may not be pol ζ defective . To determine the in vivo consequence of specifically inactivating the DNA polymerase function of Rev3l , a genetically engineered mouse was constructed to express an ASM knock-in allele from the endogenous promoter ( Fig 3A ) . Variant lox sites [32] were used to control knock-in of the Rev3l allele . The mice were crossed to CMV-Cre , producing a constitutive ASM allele ( abbreviated the “M” allele for the mice here ) , in a pure C57BL/6J background . All steps of genomic engineering were extensively monitored by Southern blotting analysis ( Fig 3B ) , PCR analysis and DNA sequencing . Heterozygote mutant Rev3l+/M mice were viable and fertile , demonstrating that the mutant allele does not have dominant-negative activity affecting viability . Heterozygous mutant Rev3l+/M mice were bred and pups genotyped . No homozygous mutant animals were identified at weaning ( Fig 3C ) . In addition , 48 embryos from 6 pregnancies were isolated between 8 . 5 and 10 . 5 dpc . Rev3lM/M embryos were rare at the earlier timepoints , and by 10 . 5 dpc only a few very small Rev3lM/M embryos were identifiable . The severely impaired development of homozygous Rev3l ASM embryos mirrors the lethality of the Rev3l null allele on a C57BL/6 background [33] . Due to the early embryonic lethality in Rev3lM/M embryos , we were never successful in deriving MEFs from them . To circumvent this problem we crossed Rev3l+/M mice with Rev3l-/lox mice . This mating produced embryos for derivation of viable Rev3lM/lox MEFs . The floxed ( lox ) allele of Rev3l is functional , but becomes a knockout allele ( termed the Δ allele ) after action of the Cre recombinase . We expressed Cre recombinase in the cells to yield Rev3lM/Δ MEFs . The mice also harbored the mT/mG transgene to monitor Cre activity . This mT/mG transgene constitutively expresses red fluorescent protein ( RFP ) . When Cre is active , the RFP gene is removed and green fluorescent protein ( GFP ) is expressed [34] . This allows GFP to be used for flow sorting and as a marker of cells in which Cre recombinase has been expressed . Cre was introduced via an adenovirus vector into primary MEFs [8] to compare Rev3lM/Δ MEFs with Rev3lM/+ MEFs ( retaining a wild-type allele of Rev3l ) . We measured cell growth , cisplatin sensitivity and DNA double-strand breaks in GFP-positive cells . ASM MEFs had a growth defect compared to wild-type allele-containing MEFs ( Fig 3D ) and eventually failed to thrive . ASM MEFs were hypersensitive to cisplatin , compared to control MEFs ( Fig 3E ) . Additionally , there was a two to three-fold increase in the number of ASM MEFs containing 53BP1 and γ-H2AX foci ( a measure of DNA breaks ) compared to controls at 9 days after Cre recombinase expression ( Fig 3F ) . These phenotypes are similar to those seen in Rev3l null primary MEFs [8] ( and compare Figs 3F and 4B ) . This result demonstrates that the DNA polymerase activity of REV3L is specifically required to allow for cell proliferation , to protect genome stability and to moderate cisplatin sensitivity . We wanted to determine in mammalian cells whether the DNA damage sensitivity and genome instability caused by pol ζ disruption is dependent on pol η , as has been reported for the DT40 cell line [27] . We crossed parental mice with the genotypes Rev3l-/lox Polh-/- , and investigated the genotypes of the pups . In the Polh-/- background , no Rev3l-/- mice were born ( Fig 4A ) , consistent with the complete lethality of the Rev3l-/- genotype in a Polh+/+ background [6] . We attempted to produce Rev3l-/- Polh-/- MEFs from mouse embryos , but were unable to obtain sufficient material to produce viable MEFs because of the early death during embryogenesis . Instead we derived primary MEFs from viable Rev3l-/lox Polh-/- embryos . Following introduction of Cre via an adenovirus , Rev3l-/Δ Polh-/- cells were produced . These Rev3l-defective primary MEFs had an elevated level of DNA breaks that was indistinguishable from Rev3l-/Δ Polh+/+ cells ( Fig 4B ) . Consistent with published results [35] the pol η defect in Rev3l-/lox Polh-/- MEFs conferred enhanced sensitivity to cisplatin ( by comparison with Rev3l-/lox Polh+/+ MEFs ) ( Fig 4C ) . A Rev3l defect independently enhanced cisplatin sensitivity , and the sensitivity of the Rev3l-/Δ Polh-/- and the Rev3l-/Δ Polh+/+ MEFs was similar . Therefore , deletion of pol η does not rescue the cell and organismal defects caused by loss of pol ζ , showing that the absence of pol ζ does not create a pol η-dependent toxic intermediate in mouse cells .
A major objective of this study was to determine whether the catalytic activity of pol ζ is responsible for the severe consequences observed in pol ζ mutant mouse cells . These include hypersensitivity to DNA damaging agents , a greatly increased generation of double-strand breaks in unchallenged cells , a slower growth rate , and a required role for pol ζ in embryonic viability . The impetus for this question is the existence of numerous other functional domains within the catalytic subunit of REV3L . These include a conserved N-terminal domain , two REV7 binding domains [14 , 19] , and a C-terminal Fe-S cluster that interacts with the POLD2 subunit and is necessary for in vitro activity . In addition , the central region contains a conserved positively charged domain [24] that likely promotes protein-protein and protein-DNA interactions , and a KIAA2022 homology domain , described in detail here for the first time ( S1 Fig ) . The presence of all of these domains introduces the possibility that the essential functions of REV3L could be structural , rather than directly related to the DNA polymerase activity itself . A catalytically deficient but otherwise intact REV3L may have been able to specifically interact with protein partners and DNA substrates , allowing viability of cells and mice . There is ample precedent for such a situation . One example is the mammalian ERCC2/XPD gene . Complete disruption of XPD is incompatible with viability [36] . However , an amino acid substitution that inactivates the catalytic helicase activity of XPD specifically compromises nucleotide excision repair capacity , but allows cellular viability [37] . This is because the presence of XPD as a subunit of transcription factor TFIIH is necessary for the integrity of that complex , even though XPD activity itself is unnecessary for transcription [38 , 39] . Another example is provided by the REV1 protein . REV1 has a DNA polymerase domain that can catalyze dCMP incorporation in DNA . Cells lacking REV1 are hypersensitive to UV radiation , but this DNA damage tolerance activity does not require the polymerase catalytic domain of REV1 . Instead , the damage tolerance activity is conferred by a protein-protein interaction domain at the C-terminus of REV1 that interacts with REV7 in pol ζ and with Y family DNA polymerases [40] . Recently , a non-catalytic role has been reported for human DNA pol κ in protection against oxidative stresses [41] . Here , we analyzed the consequence of a homozygous mutation of the Rev3l DNA polymerase active site . No viable homozygous mice were produced , and the corresponding embryos died early in embryogenesis , as with a complete knockout allele . To investigate cell-autonomous consequences of the specific polymerase alteration , we derived primary MEFs that carried one null Rev3l allele , and one active site mutant allele . The growth defects , DNA break formation and cisplatin sensitivity of these cells were similar to cells harboring two null alleles [8] . These results show that the DNA polymerase activity of REV3L is essential for all functions so far measured in mice and in cells . Loss of Rev3l causes chromosomal instability in cells . This complicates studies of the consequences of Rev3l deficiency , as genomic alterations may accumulate during each cell cycle and lead to new phenotypes . A rigorous way to determine which phenotypes are directly caused by Rev3l loss is to complement the cells by expression of Rev3l cDNA . Here we utilized a complementation system for REV3L in mammalian cells , allowing definitive testing of whether phenotypes seen in Rev3l-deleted cells are due to Rev3l-deletion [19 , 42] . Our results with specific mutant cDNAs establish that the polymerase activity of REV3L is specifically essential for preserving genome integrity and protecting against DNA damage . It is of course possible that other domains within REV3L also have critical functions for viability or genome integrity , and this complementation system will allow investigation of that possibility . For example , we recently demonstrated that the REV7-binding domains of REV3L are essential for pol ζ function [19] . We also attempted complementation of Rev3l-deficient phenotypes using human BL2 cell lines , reported to carry disruptions of REV3L [31] . It is notable that there were no major differences in phenotypes between the wild-type BL2 cells and the nominal 332 and 504 REV3L mutants . In contrast to the marked phenotypes found with Rev3l-deficient MEFs , the BL2 lines exhibited no statistically significant differences in cell doubling times , micronuclei formation or double-strand break formation as assessed by 53BP1 foci per cell . A modest sensitivity of 332 and 504 cells to UVC radiation and cisplatin was not rescued by complementation with REV3L . The limited sensitivity of 332 and 504 cells to a variety of DNA damaging agents has been noted [43–45] . Others have also reported no significant differences in spontaneous DNA breaks in 332 and 504 cells compared to wild-type BL2 cells [43] . In a study with wild-type BL2 cell extracts and extracts from the nominal REV3L-deficient cells [46] , it was concluded that REV3L does not contribute to acetylaminofluorine-induced frameshift mutagenesis . This should probably be re-examined with a different REV3L-defective cell system . It is possible that the modest increased sensitivity of the BL2 subclones to UV radiation and cisplatin [43] might be due to inadvertent disruption of an unrelated gene by the targeting strategy , as may have occurred with BL2 cells deleted for pol ι [47–49] . Our data indicate that the 332 and 504 cell lines may not be truly ( or only ) REV3L-deficient , and are not well-suited for studies of REV3L function . The DT40 chicken cell line has been widely used to examine the consequences of DNA repair defects , because it is amenable to genetic manipulation by homologous recombination . Some characteristics of Rev3l-deficient DT40 cells are similar to Rev3l-deficient mouse cells , including elevated levels of spontaneous DNA breaks and sensitivity to DNA damaging agents . Intriguingly , it was reported that deletion of polh ( pol η ) could rescue the severe phenotypes of Rev3l-deficient DT40 cells [27] . This led to the model that the major defects in Rev3l –deficient cells are a consequence of a polh-dependent toxic intermediate . To test this model in mammalian cells , we investigated whether Rev3l-/- Polh-/- mice could be generated . We found that embryonic lethality of this double mutant was complete and similar in timing to Rev3l-/- Polh+/+ mice . Moreover , Rev3l-/Δ Polh-/- MEFs showed levels of DNA breaks and cisplatin sensitivity analogous to that seen with Rev3l deletion in the presence of pol η . The Rev3l-/lox Polh-/- MEFs were more sensitive to cisplatin than the Rev3l-/lox Polh+/+ MEFs , consistent with the cisplatin sensitivity of human polh-defective cells [35] . Notably , the pol η defective and pol η pol ζ double mutant MEFs had similar sensitivities to cisplatin . This epistatic interaction suggests that these two proteins act in the same pathway to mediate resistance to cisplatin . In fact both polymerases can cooperate to bypass a cisplatin-DNA adduct [24] . In summary , the severe phenotypes caused by Rev3l deletion cannot be rescued in murine cells by concurrent deletion of pol η . This is consistent with results found in the yeast S . cerevisiae , where a Rev3 Rad30 ( pol ζ pol η ) mutant is more sensitive to ultraviolet radiation than a single Rev3 mutant [50 , 51] . Although the absence of pol η causes sensitivity to some DNA damaging agents , it is not specifically toxic in the absence of pol ζ . In the absence of pol ζ , it is possible that TLS does not occur at all , and that other modes of replication fork rescue are relied upon , which leads to a higher prevalence of DNA double-strand breaks [1] . The genetic interaction between pol η and pol ζ reported for chicken DT40 cells might reflect a peculiarity of that cell line . DT40 cells harbor mutations in TP53 , and no poli gene has been found in the chicken genome . but it seems unlikely that either gene is relevant in this context . Previously reported Tp53-/- Rev3l-/- MEFs are also pol i deficient ( an allele from the 129 ES cell background ) , and show major genome instability and DNA damage sensitivity [9] . Polh poli double mutant mice are apparently normal with no deficits in development . A poli defect does not exacerbate the UV radiation sensitivity of polh-defective mouse cells , indicating that pol ι does not have a significant backup function protecting against lethality in the absence of pol η [52] . Our results with the Rev3l knock-in polymerase mutant mouse are relevant to development of REV3L as a target for chemotherapy . Suppression of REV3L sensitizes cancer cells to cisplatin in mouse model systems , and can limit chemo-resistance [12 , 53] because loss of pol ζ diminishes point mutagenesis [2–5] . These studies used siRNA knockdown of REV3L to demonstrate this effect , but future use of small molecule DNA polymerase inhibitors may be more clinically feasible . Until now it has not been known whether inhibition of the catalytic activity of REV3L mimics the cytotoxic effects of a knockdown of the entire gene . Our work demonstrates that loss of REV3L catalytic activity is equivalent , in the assays used here , to gene knockout . This validates and encourages strategies to directly inhibit pol ζ DNA polymerase activity .
Rev3l-deficient TAg-immortalized MEFs were derived as in Lange et al [8] . Briefly , MEFs were made from mouse embryos with the genotypes mT/mG+/- Rev3l-/lox or mT/mG+/- Rev3l+/lox , where “lox” represents a functional allele flanked by loxP sites . The mT/mG transgene constitutively expresses RFP , until Cre recombinase activity removes the RFP and allows expression of GFP [34] . The strain background of the mice used to derive these alleles was mixed C57BL/6 and 129 . We genotyped DNA polymerase iota ( pol ι ) in these cell lines , because 129 mice carry a mutant allele of pol ι [54] . All cell lines were heterozygous for this mutation , and so can be considered pol ι proficient . These cell lines were immortalized with SV40 large T-antigen , and then treated with adenovirus Cre ( AdCre ) to delete the floxed allele of Rev3l ( and generate the knockout Δ allele ) . The cells were subcloned and selected for GFP positivity and for complete deletion of the floxed Rev3l allele . They were grown as in Lange et al [8] , in an atmosphere containing 2% O2 . The primary MEFs were also derived and cultured in 2% O2 as in Lange et al [8] . They were made from mouse embryos with the genotypes mT/mG+/- Rev3lM/lox or mT/mG+/- Rev3l+/lox , as well as from Rev3l-/lox Polh-/- or Rev3l-/lox Polh+/+ embryos . The loxP-flanked allele of the Rev3l gene was deleted using AdCre adfection , and the deletion efficiency was measured as described [8] . For all cell lines , cell number was counted at each passage , and was used to calculate population doublings and doubling time . The BL2 parental cell line and subclones 332 and 504 [31] were kindly provided by Claude-Agnés Reynaud ( Institut Gustave Roussy , Villejuif , France ) . Genomic DNA samples from the three cell lines were compared using short tandem repeat ( STR ) fingerprinting by the Cell Line Identification Core at MD Anderson . All yielded identical profiles of the 16 standard STR markers , confirming the relationship of the three cell lines . The human REV3L full-length cDNA was acquired in the pUC19M1 vector from Zhigang Wang [55] . The following modifications were made to the REV3L cDNA: a C-terminal Flag tag was added and the 5’-UTR was eliminated and replaced with an optimized mammalian Kozak sequence . This cDNA was cloned into the pTSIGN vector , which contains an EF1α promoter and an internal ribosomal entry site ( IRES ) fused to a neomycin-eGFP reporter . The active site mutation ( residues D2781A/D2783A in human REV3L ) was introduced into this pTSIGN-REV3L vector using PCR primers containing the REV3L mutations , and then the mutated PCR fragment was ligated into the REV3L-vector , replacing the wild-type sequence . The full-length human REV3L gene was PCR amplified from the pTSIGN-REV3L and pTSIGN-REV3L-ASM vectors and was cloned into the pETDuet-1 vector ( Novagen ) . The REV3L gene was removed from the REV3L-pETDuet-1 vectors using XhoI/NotI digestion , and the resulting fragments were inserted into the pOZN vector ( contains a Flag-HA tag on the N-terminal side of the inserted gene [28] ) . For the pCDH vector , the XhoI/NotI fragment from the full-length REV3L-pETDuet-1 vector was inserted into the pCDH-EF1α-Flag-HA-MCS-IRES-Puro vector ( System Biosciences ) . All vectors were completely sequenced to verify the integrity of the REV3L gene and the plasmid backbone . Full-length Flag-HA tagged REV3L can be expressed from this cDNA [19] . The pOZN-REV3L or pCDH-REV3L vectors were transfected into HEK-293T cells using lipofectamine 2000 ( Life Technologies ) , together with the retroviral packaging vectors psPAX2 ( plasmid 12260 , Addgene ) and pMD2 . G ( plasmid 12259 , Addgene ) . 48 hr later , the media ( containing pOZ or pCDH lentivirus ) was collected . It was filtered , and polybrene was added to 4 μg/mL . This media was added to plates of immortalized MEFs ( pOZ ) or flasks of BL2 cells ( pCDH ) . 48 hr later , the cells began selection for puromycin expression ( pCDH , 10 day incubation ) , or for IL2R expression ( pOZ ) . The latter required incubation of the infected cells with IL2R-antibody conjugated magnetic beads followed by washing of the beads ( as in [56 , 57]; IL2R antibody from Millipore , 05–170 ) . This was repeated 5 times . The population was then sorted for single-cells , and clones were selected and verified . The cells were confirmed to contain both the N and C-terminal portions of the REV3L expression construct using the following PCR primers: NFwd: 5’ TAC ACA GTC CTG CTG ACC AC 3’ , NRev: 5’ GAG GTA AGG AAA GAT GCC ATG TAG 3’ , CFwd: 5’ ACC TAA CTC AGC ATG GCA TCT G 3’ , CRev: 5’ CGG AAT TGA TCC GCT AGA G 3’ ( at an annealing temperature of 50°C ) . Expression of the recombinant human REV3L was confirmed using a human-specific Taqman assay ( Life Technologies ) at the exon 14–15 boundary: Ex14Fwd: 5’ CAC CTG GCC TTA GCC CAT TAT 3’ , Ex15Rev: 5’ CTC TTC TAA GAG TGT CAG TAT TAC TTC CTT TC 3’ Probe: FAM-MGB-5’ CAA CAG AAC CAA AAA CA 3’ . In order to compare the recombinant expression to that of endogenous mouse Rev3l , we designed a set of primers and a probe that would recognize both human and mouse Rev3l , and would not amplify any knockout transcript . The primers/probe were at the exon 26/27 boundary: Ex26Fwd: 5’ GTG AAT GAT ACC AAG AAA TGG GG 3’; Ex27Rev: 5’ GTG AAT GAT ACC AAG AAA TGG GG 3’; Probe: FAM-MGB-5’ TAC TGA CAG TAT GTT TGT 3’ . An additional gene expression analysis was completed on the hREV3L-expressing BL2 cells in order to distinguish the endogenous REV3L transcript ( which was expressed at approximately equal levels in the REV3L knockout and wild-type BL2 cells ) from the exogenously expressed REV3L . We used primers and a probe that crossed the FLAG tag on the exogenous gene: FlagFwd: 5’–GTCTTTGTTTCGTTTTCTGTTCTG C– 3’; FlagRev: 5’–GCTTGTCATCGTCGTCCTTG– 3’; Probe: FAM-MGB-5’–GCT GTG ACC GGC GCC TAC TCT AG– 3’ . Gene expression ( with mouse or human GAPDH as an expression control ) was measured on an Applied Biosystems 7900HT Fast Real-Time PCR System . To test sensitivity to chemical DNA damaging agents , the immortalized MEFs or BL2 cells were plated into white 96-well plates ( immortalized MEFs– 5 , 000 cells/well; BL2 cells– 10 , 000 cells/well ) . The following day , various concentrations of cisplatin ( Sigma ) or bleomycin ( Sigma ) were added to the wells , and the cells were incubated for 48 hr . Then the cells were lysed , a reagent was added that emits light in the presence of ATP ( ATPLite One Step , Perkin Elmer ) , and luminescence was measured using a plate reader ( Biotek Synergy II ) . The luminescence measurement was normalized to undamaged control . To test cisplatin sensitivity in Rev3l-deleting primary MEFs , 1 day after deleting the Rev3l floxed allele with AdCre , the cells were plated into white 96-well plates ( 10 , 000 cells/well ) . On day 3 , cisplatin at various concentrations was added , and the cells were incubated for 5 days . Then ATP content was measured by luminescence , as above . To test sensitivity of immortalized MEFs or BL2 cells to UVC radiation , 3 x 105 cells were pelleted and resuspended in 300 μL of phosphate-buffered saline . Three 100 μL drops were placed into the middle of a plastic dish and 10 μL aliquots from each were plated into 100 μL of growth media in a white 96-well plate after 0 , 2 . 5 , 5 , 7 . 5 , 10 , 15 or 20 J/m2 UVC radiation at a fluence of 0 . 4 J/m2 s-1 . 48 hr after irradiation , ATP content was measured as above . To measure the formation of DNA double-strand breaks , immortalized MEFs were plated in an 8-well chamber slide . The following day they were fixed and stained for DAPI , 53BP1 and γ-H2AX , as in Lange et al [8] . BL2 cells were applied to microscope slides using a Cytospin ( Thermo Scientific ) , and then fixed and stained as with the MEFs . Immunofluorescence images were photographed through a Leica DMI6000B microscope . Micronuclei were counted based on small , separate DAPI foci associated with DAPI-stained nuclei . 53BP1 foci per cell were counted using the CellProfiler program 1 [59] with a threshold correction factor of 1 . 7 . To measure DNA double-strand breaks in the primary MEFs , cells were plated into 8-well chamber slides 7 days after deletion of the Rev3l floxed allele using AdCre . 48 hr later they were fixed and stained for DAPI , 53BP1 and γ-H2AX as above . Photographs of the immunofluorescence were taken on the Leica microscope , and cells containing double-strand breaks were scored as those with 3 or more 53BP1 + γ-H2AX foci . Assessment of the Rev3L-/Δ and Rev3L+/Δ cell lines for sister chromatid exchanges ( SCEs ) was as described [57] . BrdU ( 10 μM ) was added to growing TAg-immortalized MEFs for a period of two cell cycles , followed by a 4 hr incubation with 0 . 02 μg/mL colcemid . The cells were then harvested and incubated with hypotonic solution ( 0 . 075 M KCl ) for 10 min at 37°C . Then 50 μL of fresh fixative solution ( 3:1 methanol:acetic acid ) was added and the cells were pelleted at 1000 rpm for 10 min . After aspiration of the supernatant , 5 mL of fresh 4°C Carnoy’s fixative ( 6:3:1 ethanol: chloroform: glacial acetic acid ) was added dropwise to the pellet and the cells were incubated at 4°C for 30 min followed by centrifugation for 10 min at 1000 rpm at 4°C . The supernatant was aspirated , and this process was repeated . 1 mL of Carnoy’s fixative was added to the final cell pellet and the cells were dropped onto clean microscope slides in a humid environment to favor chromosome spreading . The slides were stained for 45 min in 0 . 5X SSC buffer containing 2 μg/mL Hoechst 33258 for 45 min , and then were washed twice in SSC buffer for 5 min each . The slides were then immersed in 0 . 5X SSC buffer and exposed to UVA light ( 350 nm wavelength , 15 W ) at a distance of 10 cm for 1 hr . Then the slides were incubated for 1 hr in fresh 0 . 5X SSC buffer at 60°C , and were stained for 15 min with 3% Giemsa dye in Sorenson’s buffer ( Sigma , diluted 1:15 in 0 . 025 M KH2PO4 pH 6 . 8 ) . The chromosome spreads were viewed at 600X magnification under oil . Thirty to thirty-five chromosome spreads were counted for each genotype , and both total chromosome number and number of SCEs was assessed . Analysis of cells with associated micronuclei or > 2 53BP1 + γ-H2AX foci was done using one-way ANOVA with Tukey’s multiple comparisons test ( P < 0 . 05 ) . Statistical analysis of cell survival after cisplatin or UVC treatment , or of cell growth , was done using linear regression analysis , and the lines were compared based on equality of slope and intercepts . | Translesion synthesis allows DNA replication to occur in the presence of damaged DNA . This process is mediated by low-fidelity DNA polymerases ( such as pol ζ or pol η ) that maintain genomic stability . The action of these polymerases is crucial to limit cancer . In mice , complete deletion of DNA pol ζ leads to embryonic lethality , and conditional deletion enhances tumorigenesis . Pol ζ is a large protein with many domains that interact with other essential proteins and maintain the structural integrity of pol ζ . It is not known if the polymerase activity of pol ζ mediates its essential activities . Using a cell culture complementation system and in vivo knock-in mice , our work shows that pol ζ–mediated maintenance of genomic stability in the presence of DNA damage is absolutely dependent on its DNA polymerase activity . Others have demonstrated in chicken cells that co-deletion of pol ζ and pol η rescues the pol ζ-dependent phenotypes , but our work in mice and in mouse cell culture does not support that conclusion . These results demonstrate the physiological importance of pol ζ polymerase activity , and show that employing small-molecule inhibitors of the polymerase reaction is a valid strategy for sensitizing tumor cells to chemotherapeutic agents . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2016 | The Polymerase Activity of Mammalian DNA Pol ζ Is Specifically Required for Cell and Embryonic Viability |
Detailed investigations of multiparasitism are scarce in the Mekong River basin . We assessed helminth ( trematode , nematode , and cestode ) , and intestinal protozoa infections , and multiparasitism in random population samples from three different eco-epidemiological settings in Champasack province , southern Lao People's Democratic Republic ( Lao PDR ) , and determined underlying risk factors . Two stool samples were collected from 669 individuals aged ≥6 months over consecutive days and examined for helminth infections using the Kato-Katz method . Additionally , one stool sample per person was subjected to a formalin-ethyl acetate concentration technique for diagnosis of helminth and intestinal protozoa infections . Questionnaires were administered to obtain individual and household-level data pertaining to behavior , demography and socioeconomic status . Risk factors for hepato-biliary and intestinal parasitic infections and multiparasitism were determined using multiple logistic regressions analyses . Multiple species intestinal parasite infections were common: 86 . 6% of the study participants harbored at least two and up to seven different parasites concurrently . Regarding nematode infections , hookworm was the most prevalent species ( 76 . 8% ) , followed by Ascaris lumbricoides ( 31 . 7% ) and Trichuris trichiura ( 25 . 0% ) . Regarding trematodes , Opisthorchis viverrini and Schistosoma mekongi infections were found in 64 . 3% and 24 . 2% of the participants , respectively . Infections with intestinal protozoa were rare . There is a pressing need to intensify and sustain helminth control interventions in the southern part of Lao PDR . Given the high prevalence with nematode and trematode infections and the extent of multiparasitism , preventive chemotherapy is warranted . This intervention should be coupled with health education and improved access to clean water and adequate sanitation to consolidate morbidity control and enhance sustainability .
Lao People's Democratic Republic ( Lao PDR ) is a landlocked country situated in the Great Mekong sub-region of Southeast Asia , where socioeconomic and eco-epidemiological characteristics vary greatly according to location . In the northern part similar ecosystems are found as in southern People's Republic of China ( P . R . China ) with mountains and highlands dominating the landscapes . These topological features are natural barriers that might impede social and economic development , since transportation of commodities , communication and other exchanges are hampered . These issues exacerbate people's access to health care , clean water and adequate sanitation . Indeed , according to the results of the national population and housing census carried out in 2005 , less than 20% and only about half of the population living in these areas had access to clean water and sanitation , respectively [1] . Water supply and sanitation are intimately linked with intestinal parasitic infections and poverty . Schistosomiasis , opisthorchiasis and infections with the common soil-transmitted helminths ( i . e . Ascaris lumbricoides , hookworm , and Trichuris trichiura ) are of particular relevance [2]–[5] . Improving socioeconomic status , including enhanced access to quality health care , safe water , and adequate sanitation have the potential to significantly reducing the prevalence and intensity of parasitic infections , and hence reduce disease-related morbidity [5]–[7] . The central and southern parts of Lao PDR are the low land along the Mekong River basin . In these regions , the socioeconomic conditions and means of communication and transport are more advanced than in northern Lao PDR . In recent years , through the formation of the ASEAN community , the economy of the Great Mekong sub-regions countries has been bolstered . Along with these changes and ecological transformations ( e . g . , deforestation and water resources developments ) , particularly in the lowlands of the Mekong River basin , patterns of parasitic infections are changing [8] . A matter of considerable public health concern is the transmission of Schistosoma mekongi which , although several rounds of preventive chemotherapy were implemented , is still transmitted in the Mekong River in the most southern province of Lao PDR [9] , [10] . Furthermore , high prevalences of Opisthorchis viverrini are a concern , as this liver fluke is the main risk factor for the fatal cholangiocarcinoma bile duct cancer [11]–[13] . An infection with O . viverrini is acquired through the consumption of traditional dishes ( e . g . , “Lap-pa” and “Koy-pa” ) prepared with raw or insufficiently cooked fish [14] , [15] . The habit of eating raw or undercooked freshwater fish and other aquatic products is also a risk for acquiring small intestinal trematode infections , such as heterophyid and lecithodendriid flukes , which are endemic in southern Lao PDR [16]–[18] . Moreover , raw fish consumption is a precondition of capillariasis transmission , which was recently documented in Lao PDR [19] , [20] . A national survey conducted among school children found that common soil-transmitted helminth infections are particularly prevalent in the northern provinces , whereas O . viverrini infections are rampant in the central and southern provinces with significant overlaps of different parasite species in all provinces [21] . It follows that multiparasitism must occur , which has been confirmed in recent surveys [15] , [16] , [22] . However , data pertaining to multiparasitism have mostly been obtained from small studies ( e . g . , in a single community in a single village ) , often looking at a narrow age range ( e . g . , school-aged children ) [23] . The present study was carried out in different eco-epidemiological settings of Champasack province , southern Lao PDR . Using a cross-sectional design , the purpose was to assess the prevalence and intensity of hepato-biliary and intestinal parasitic infections and intestinal multiparasitism , and to identify underlying risk factors .
The study was carried out in three distinct eco-epidemiological settings of Champasack province ( Figure 1 ) , located in the southern part of Lao PDR , namely ( i ) Khong , ( ii ) Mounlapamok , and ( iii ) Paksong districts . Of note , the districts represent different settings in terms of socioeconomic conditions and eco-epidemiology and are characteristic for other parts of Lao PDR . Khong district ( estimated population: 80 , 000 ) [24] is an island district which is located in the southern part of the province ( ∼120 km from Pakse city ) , which borders Cambodia ( geographical coordinates: 13 . 57°–14 . 14°N latitude and 105 . 44°–106 . 08°E longitude ) . Khong district comprises dozens of islands in the Mekong River basin and is therefore also known as “district of four thousand islands” . The waterfall ‘Khon-Phapheng’ has put a barrier in the Mekong River and has created a natural reservoir . The ecology of the area is suitable for aquatic snails , the intermediate hosts for S . mekongi and food-borne trematodes , such as O . viverrini and minute intestinal flukes ( MIF ) . The Mounlapamok district is also located in the southern part of the province ( ∼80 km from Pakse city ) with an estimated population of 40 , 000 [24] . It is a lowland district situated along the Mekong River ( geographical coordinates: 14 . 15°–14 . 25°N and 105 . 49°–106 . 11°E ) . In this area , opisthorchiasis is highly prevalent [25] . Paksong district is located on the Bolovan plateau ( geographical coordinates: 14 . 58°–15 . 23° N and 105 . 55°–106 . 48° E ) at an elevation of ∼1 , 000 m above sea level in the northeast of the province ( ∼50 km from Pakse city ) . It is a mountainous area with an estimated population of 65 , 000 [24] . Soil-transmitted helminth infections are common in Paksong district [21] . The study was approved by the Ethics Committee in Basel , Switzerland ( EKBB; reference no . 255/06 ) and the National Ethics Committee , Ministry of Health ( MoH ) in Vientiane , Lao PDR ( reference no . 027/NECHR ) . Permission for field work was obtained from MoH , the Provincial Health Office ( PHO ) and the District Health Office ( DHO ) . Village meetings were held and village authorities and villagers were given detailed explanations about the aims , procedures , potential risks and benefit of the study . An information sheet in the local language was read aloud to all household members and their questions answered . Individual oral consent was obtained from all adult household members ( literacy is very low in this part of Lao PDR , and hence we opted for oral rather than written consent ) . However , written informed consent was obtained from all heads of households . A witness observing this procedure also signed the consent form . All individuals infected with O . viverrini , S . mekongi , soil-transmitted helminths , and intestinal protozoa were treated according to national guidelines [26] . An anti-spasmodic treatment and oral rehydration was provided in case of adverse events following drug administration . Our cross-sectional surveys were carried out between March and May 2006 . In each setting , three villages were selected from the available village list in collaboration with the DHO , and 20–25 households were randomly selected in each village . All household members aged ≥6 months were invited to participate . The number of inhabitants per household was recorded . Unique identifiers were assigned to households and study participants . In each village , a house ( usually a school or a temple ) was designated as an area of work for Kato-Katz ( KK ) thick smear preparation , microscopic examination of stool samples , etc . Two members of our research team ( one interviewer and one general physician ) went from house to house and interviewed first the head of household and then the remaining household members . Two questionnaires were administrated in each household . The household questionnaire ( after pre-testing in a neighboring area ) was administered to the heads of household . Data pertaining to household characteristics ( e . g . , building type and water supply ) , asset ownership ( e . g . , farm engine and bicycle ) and ownership of animals ( e . g . , buffalo and cow ) were collected . The geographical coordinates of each household were obtained by using a hand-held global positioning system ( GPS ) receiver ( Garmin Ltd . , Olathe , USA ) . Next , a pre-tested individual questionnaire was used and all household members were interviewed for demographic data ( e . g . , age , sex , educational attainment , and professional activity ) and behavioral risks ( e . g . , food consumption habits and personal hygiene ) . Parents or legal caregivers answered for children . Finally , stool containers were prepared for all members of each study household . Participants' names and unique identifiers were marked on the containers and distributed to the heads of household with detailed instructions of how to collect a fresh morning stool sample . All study participants were asked to provide a sufficiently large stool sample ( at least 5 g ) so that both KK and the formalin-ethyl acetate concentration technique ( FECT ) could be performed . After filled containers were collected , new empty containers were handed out with the goal to obtain three stool samples from each participant over consecutive days . Stool samples were processed in the designated area of work in the study village within a maximum of 2 hours after collection by experienced laboratory technicians . A single KK thick smear was prepared from each stool sample , using a standard plastic template holding 41 . 7 mg of stool [27] . Slides were allowed to clear for 30 min prior to examination under a microscope . The number of eggs was counted and recorded for each helminth species separately . Additionally , exactly 300 mg of stool taken from one sample was fixed in a tube containing 10 ml of sodium acetate acetic-acid formalin ( SAF ) [28] . SAF-fixed samples were forwarded to the parasitological department of the Faculty of Medicine , National University of Lao PDR . The samples were subjected to FECT [29] and diagnosed for the presence of intestinal protozoa and helminth species-specific infections and intensities with the assistance of laboratory staff from the Swiss Tropical and Public Health Institute ( Basel , Switzerland ) . Data were double-entered and cross-checked using EpiData version 3 . 1 ( Epidata Association; Odense , Denmark ) . Statistical analyses were performed with STATA version 10 ( Stata Corporation; College Station , TX , USA ) . Only those individuals who had at least two KK thick smear readings and an additional FECT result , and complete questionnaire data were included in the final analyses . People's socioeconomic status was estimated using a household-based asset approach and the population was stratified into wealth quintiles , namely ( i ) poorest , ( ii ) very poor , ( iii ) poor , ( iv ) less poor , and ( v ) least poor . Wealth quintiles were constructed using principal component analysis ( PCA ) , as proposed by the Health Nutrition and Population/World Bank in 2000 [30] . Details of this widely used approach have been presented elsewhere [31] . In brief , a PCA was calculated from the following household assets: electricity radio/recorder , television , CD/DVD player , water pump , refrigerator , car , farm engine , motorcycle , rice security , house characteristics ( construction material for floor , wall , and roof ) , and animal ownership ( buffalo , cow , goat , and pig ) . The weights obtained from the first dimension were used to calculate the household index score . The first principal component ( PC ) explained 17 . 2% of the total variability . The greatest weights were attached to families living in a wooden house ( 0 . 30 ) , a bamboo house ( 0 . 29 ) , and the presence of a television at home ( 0 . 20 ) . After standardization of these weighted asset variables , families living in a cement house had the highest scores ( 0 . 47 ) . Lowest scores were attached to families living in a bamboo house ( −0 . 55 ) . The sum of total asset scores was assigned to each study participant . Point prevalence of parasitic infections were determined and stratified by study area , sex , and age group . A chi-square ( χ2 ) test was employed to investigate associations between categorical variables ( e . g . , between infection status and sex , age group , and study area ) . Study participants were subdivided into five age groups , namely ( i ) <5 years , ( ii ) 6–15 years , ( iii ) 16–30 years , ( iv ) 31–55 years , and ( v ) >55 years . The intensity of helminth egg counts was expressed as eggs per gram of stool ( EPG ) . Intensity rate ratio ( IRR ) of EPG was calculated using negative binomial regression models and associated with sex and age groups . A predictor variable with level of significance below 0 . 15 in the bivariate logistic regression models was included in the multiple logistic regressions to investigate the associations between the parasitic infections and a particular risk factor . Random effect models were fitted into all regressions , taking into account the random effect of households .
From 1 , 213 enrolled participants , 1 , 051 were present during the cross-sectional survey and responded to our questionnaire ( Figure 2 ) . A total of 314 individuals ( 29 . 9% ) failed to submit sufficient numbers and/or quantities of stool samples for laboratory diagnoses . Fourteen individuals ( 1 . 3% ) had no SAF-fixed stool sample and 192 individuals ( 18 . 3% ) were absent during the household-based interviews , and hence their socioeconomic status could not be determined . Overall , 669 individuals ( 63 . 7% ) had complete data records ( i . e . , at least 2 KK thick smears , 1 FECT result , and complete questionnaire data ) . Among this cohort , 212 individuals ( 31 . 7% ) were from Paksong district , 232 ( 34 . 7% ) from Mounlapamok district , and 225 ( 33 . 6% ) from Khong district . Most study participants belonged to the Lao-loum ethnic groups ( 68 . 5% ) , whereas the Lao-theung minority accounted for the remaining 31 . 5% . There were slightly more females ( n = 347 , 51 . 9% ) . The median age was 15 years ( range: 6 months to 87 years ) . Age structure was as follows: ≤5 years ( 17 . 3% ) , 6–15 years ( 32 . 9% ) , 16–30 years ( 16 . 4% ) , 31–55 years ( 24 . 9% ) and >55 years ( 8 . 4% ) . Adults were primarily engaged in subsistence farming ( 52 . 1% ) , while there were only few government employees ( 1 . 4% ) . No professional activity accounted for 46 . 5% of the study participants . Of those , 17 . 3% , 20 . 4% and 8 . 8% were preschool-aged children , pupils or students and elderly persons , respectively . With regard to wealth , we observed that most study participants from Paksong district belonged to the poorest group ( 53 . 5% ) , whereas none of them were classified into the group of the least poor . In Khong and Mounlapamok districts , the combined percentage of less poor and least poor was 40 . 4% and 29 . 3% , respectively . Only a few individuals ( Mounlapamok: 3 . 0% and Khong: 2 . 2% ) belonged to the poorest group ( Figure 3 ) . Table 1 summarizes the results from the cross-sectional parasitological surveys , stratified by eco-epidemiological setting , sex , and age group . Analysis of at least two stool samples using the K-K technique , supplemented with an additional FECT result revealed overall infection prevalences of O . viverrini , S . mekongi and Echinostoma spp . of 64 . 3% , 24 . 2% and 6 . 0% , respectively . The former two trematode infections were particularly prevalent in Khong district ( O . viverrini: 92 . 0% , S . mekongi: 68 . 0% ) . While a similarly high prevalence of O . viverrini infection was observed in Mounlapamok district ( 90 . 9% ) , the observed prevalence of S . mekongi was only 3 . 9% . In Paksong district only few cases of O . viverrini infections were observed ( 5 . 7% ) , owing to a highly significant difference among study location ( likelihood ratio test ( LRT ) = 51 . 35 , P <0 . 001 ) . The prevalence of O . viverrini infections increased with age and reached the highest levels in the age group above 55 years ( LRT = 28 . 83 , P <0 . 001 ) . S . mekongi infections were also significantly associated with age , with the highest prevalence observed in the oldest age group ( 32 . 1%; LRT = 13 . 91 , P = 0 . 007 ) . Neither O . viverrini nor S . mekongi infections were significantly associated with sex . The prevalence of O . viverrini infections was significantly higher in Lao-loum ethnic group compared to Lao-theung ( 91 . 1% vs . 6 . 2%; LRT = 199 . 51 , P <0 . 001 ) . The overall infection prevalence of hookworm , A . lumbricoides and T . trichiura was 76 . 8% , 31 . 7% and 25 . 0% , respectively . There was significant variation from one district to another ( LRT = 30 . 11 , P <0 . 001 ) . The highest prevalences were found in Paksong district ( hookworm: 94 . 8% , A . lumbricoides: 85 . 9% , and T . trichiura: 55 . 7% ) and the lowest prevalences were observed in Mounlapamok district ( hookworm: 66 . 0% , T . trichiura: 8 . 2% , and A . lumbricoides: 6 . 0% ) . There were no significant differences between sex and age groups for any of the three main soil-transmitted helminth infections . Cestode infections such as Taenia spp . , Hymenolepis diminuta and Diphyllobothrium latum were found at low prevalences , ranging between 0 . 5% and 3 . 7% . Blastocystis hominis ( 13 . 6% ) was the most common intestinal protozoa diagnosed , followed by Entamoeba coli ( 7 . 2% ) , Giardia intestinalis ( 4 . 9% ) , and Endolimax nana ( 0 . 6% ) . There was a significant variation in the observed prevalence ( LRT = 42 . 32 , P<0 . 001 ) for B . hominis and E . coli according to study location . Table 2 shows the adjusted IRR of helminth egg counts expressed in EPG for the most prevalent intestinal parasites investigated , using age group <5 years as a referent group . The overall intensity ratio of EPG for O . viverrini infection increased with age and reached the highest level in the adult people aged above 55 years ( IRR = 7 . 41 , 95% confidence interval ( CI ) = 4 . 82–11 . 42 ) with no significant sex difference . Children ( 6–15 years ) showed a higher infection intensity with S . mekongi ( IRR = 1 . 79 , 95% CI = 1 . 01–3 . 18 ) and hookworm ( IRR = 1 . 49 , 95% CI = 1 . 17–1 . 90 ) than their older counterparts . With regard to A . lumbricoides and T . trichiura , there was no significant difference for infection intensity in all age groups . Only 13 ( 1 . 9% ) individuals were free of intestinal parasites . Mono-infections were observed in 77 individuals ( 11 . 5% ) . Hence , most of the study participants had a multiple species intestinal parasite infection: 32 . 9% were infected with two different parasites , 53 . 5% harbored 3–6 parasite species concurrently , and in one individual seven different parasites were observed . Over a third of the study participants living in Paksong and Khong districts were infected with three different parasite species concurrently and almost half of the surveyed Mounlapamok residents were concurrently infected with at least two parasite species ( Figure 4 ) . Table 3 summarizes significant associations between different intestinal parasites . An O . viverrini infection showed a significant positive association with S . mekongi ( odds ratio ( OR ) = 5 . 09 , 95% CI = 2 . 49–10 . 42 ) , but negative association with both A . lumbricoides ( OR = 0 . 05 , 95% CI = 0 . 03–0 . 07 ) and T . trichiura ( OR = 0 . 34 , 95% CI = 0 . 20–0 . 58 ) . Conversely , S . mekongi showed a significant positive association with an O . viverrini infection ( OR = 5 . 64 , 95% CI = 2 . 75–11 . 56 ) . Moreover , there were significant positive associations between S . mekongi and Echinostoma spp . ( OR = 3 . 19 , 95% CI = 1 . 58–6 . 45 ) and between S . mekongi and two intestinal protozoa , namely B . hominis ( OR = 2 . 19 , 95% CI = 1 . 26–3 . 79 ) and E . coli ( OR = 2 . 20 , 95% CI = 1 . 01–4 . 83 ) . An infection with hookworm was significantly associated with the other common soil-transmitted helminths ( A . lumbricoides and T . trichiura ) and S . mekongi ( OR = 1 . 70 , 95% CI = 1 . 04–2 . 79 ) . More than half of our fully compliant study participants ( n = 345 , 51 . 6% ) reported to have consumed at least once raw fish dishes within 7 days prior to the interview . The habit of raw fish consumption was particularly frequent among the Lao-loum ethnic group ( 85 . 7% ) , and significantly less common among the Lao-theung ethnic group ( 14 . 3%; LRT = 98 . 04 , P <0 . 001 ) . Consumption of raw meat dishes was reported by 12 . 3% of our study population . Of those , 80 . 7% belonged to the Lao-loum and 19 . 3% to the Lao-theung ethnic group . Table 4 shows the results from the multiple logistic regression analyses regarding associations between parasitic infections and risk factors , taking into account the random effect of households . Lao-loum ethnic groups were more likely to have an O . viverrini infection than Lao-theung ethnic groups ( OR = 303 . 5 , 95% CI = 134 . 2–686 . 6 ) . The Lao-loum were at lower risks of hookworm ( OR = 0 . 12 , 95% CI = 0 . 07–0 . 23 ) . Swimming ( bathing ) in the Mekong River was a key risk factor for acquiring a S . mekongi infection . Infections with A . lumbricoides was more common in poorer population segments ( most poor: OR = 3 . 53 , 95% CI = 1 . 47–8 . 47 ) . Consuming of raw or insufficiently cooked food was a risk factor for multiparasitism in our study population ( OR = 2 . 74 , 95% CI = 1 . 44–5 . 20 ) .
Helminth infections are widespread in Lao PDR and the Great Mekong sub-region in general . S . mekongi , O . viverrini , various MIFs and soil-transmitted helminths are prevalent and there is extensive geographical overlap of various helminth infections [9] , [21] , [22] , [32] , [33] . However , there is a paucity of high-quality data to elucidate the extent of multiparasitism and underlying risk factors [23] , [34] . We conducted a cross-sectional study in three distinct eco-epidemiological settings of Champasack province situated in the southern part of Lao PDR . We employed a rigorous diagnostic approach , i . e . , at least two stool samples were collected over consecutive days and examined by the KK method , supplemented with a FECT performed on one of these stool samples . Our data confirm that multiple species intestinal parasite infections are the norm rather than the exception; indeed more than 4 out of 5 study participants with complete data records harbored at least two different species concurrently , and several intestinal parasite species were found at high prevalence rates . Worryingly , O . viverrini infections were found in over 90% of the study subjects in the two low-land settings ( Khong and Mounlapamok districts ) . In Khong district , additionally , we found a high S . mekongi infection prevalence ( 68 . 0% ) . Soil-transmitted helminths were common in the highland of Paksong district; the overall prevalence for hookworm , A . lumbricoides and T . richiura were 94 . 8% , 85 . 9% and 55 . 7% , respectively . On the other hand , intestinal protozoa infections ( B . hominis , E . coli , G . intestinalis and E . nana ) were far less prevalent ( <14 . 0% ) . Limitations of our study are as follows . First , although we employed a rigorous diagnostic approach , the ‘true’ extent of multiparasitism is still underestimated . The diagnostic techniques used in our study only have a low sensitivity for the detection of certain parasite species ( e . g . , Strongyloides stercoralis and MIF ) or are inadequate for other endemic parasitic infections such as malaria . Second , we did not differentiate eggs of O . viverrini from those of MIF . Eggs of O . viverrini and MIF are similar in size and shape , and hence it is exceedingly difficult to differentiate them under a microscope . Therefore among those study participants declared O . viverrini-positive , some might actually be infected with MIF , since many species of MIF are also endemic in Lao PDR [16] , [22] , [35] . In our own preceding work , we found that multiple trematode species infections indeed are common . For example , among 97 individuals with heavy Opisthorchis infections who were purged , 81 . 4% of the participants were multi-parasitized . O . viverrini was the most common trematode ( 97 . 9% ) , followed by Haplorchis taichui ( 78 . 4% ) . Other small intestinal fluke species were less common [22] . In studies carried out elsewhere in Lao PDR , it was also found that O . viverrini is the predominant trematode species [16] , [17] . Third , it cannot be ruled out that some of the diagnosed hookworm eggs were actually infections with Trichostrongylus spp . The latter parasite has been found in Lao PDR with notable prevalence rates [36] . Highest infection intensities of A . lumbricoides and T . trichiura were observed in pre-schoolers ( aged ≤5 years ) , whereas the peak infection intensities of S . mekongi and hookworm were observed in school-aged children ( age: 6–15 years ) . Adults aged above 55 years showed highest O . viverrini infection intensity rate ratios . The high prevalence of S . mekongi observed in Khong district must be emphasized . This finding suggests that schistosomiasis is still a public health concern in southern Lao PDR . Once schistosomiasis had been recognized as a major public health problem in southern Lao PDR and Cambodia in the early 1980s and early 1990s , respectively [10] , [37]–[39] , community-based control programs were launched . The aim of these control programs was to reduce schistosome-related morbidity . Large-scale administration of praziquantel was endorsed as the strategy of choice [9] , [10] , [38] . Multiple rounds of praziquantel reduced the prevalence of S . mekongi in the endemic areas to very low levels in 1998 ( 2 . 1% in Khong district and 0 . 4% in Mounlapamok district ) and was considered a successful public health control program [9] , [10] . However , interruption of chemotherapy-based morbidity control in face of inadequate sanitation , lack of clean water , and continued human water contacts are at the root of rapid re-infection and re-emergence of schistosomiasis . In 2006 , chemotherapy-based control has been re-established . Failure to improve access to clean water and adequate sanitation will render truly sustainable schistosomiasis control a distant goal . In 2007 in our study villages of Khong district , only 14 . 5% of the households possessed latrines and 76 . 0% reported daily use of the Mekong River for bathing ( K . Phongluxa , personal communication ) . Hence , there is also a need for more vigorous health education to avoid risky water contacts as a means of lowering the transmission of schistosomiasis and to thoroughly cook fish and other aquatic products to break the transmission cycle of opisthorchiasis and other food-borne trematode infections . Our findings underscore that intestinal multiparasitism is common throughout Champasack province . The same observations have been made in other parts of Lao PDR [15] , [16] , [22] and neighboring countries such as Vietnam [32] , [33] , [40] and southern P . R . China [41] . Indeed , multiparasitism is the rule rather than exception in the developing world [23] , [42] , [43] , and hence it is surprising that the topic has received only token attention [44] . Our data showed that O . viverrini and hookworm co-infections were highly prevalent in the plain area of Khong and Mounlapamok districts , whilst multiple species soil-transmitted helminth infections were common among the study participants in the highlands of Paksong district . From a clinical point of view , co-infection of S . mekongi and O . viverrini is of particular concern . Indeed , an infection with O . viverrini leads to severe clinical manifestations such as hepato-biliary pathologies , including hepatomegaly , obstructive jaundice , gallbladder stones , cholecystitis and cholangitis and , most importantly , the development of a fatal bile duct cancer ( cholangiocarcinoma ) [11] , [13] , [45] . Chronic infection with S . mekongi contributes to a formation of hepatomegaly , periportal fibrosis , and portal hypertension [37] , [46]–[48] . Co-infections of these two trematodes might further aggravate the host-organ pathology , especially the liver . Another interesting finding of our study was the significant association observed between S . mekongi and hookworm . Interestingly , previous studies carried out in Côte d'Ivoire found a significant association between S . mansoni and hookworm [42] , [43] , [49] . Whilst the distribution of single species infection and co-infections have been mapped [50] and risk factors elucidated , there is a lack of epidemiologic investigations focusing on symptoms and morbidities due to co-infections . Soil-transmitted helminths were also found to be highly prevalent in the present study , particularly among those living in the highlands of Paksong district . An infection with soil-transmitted helminths can lead to nutritional deficiencies and may impair growth and cognitive development in children [50]–[53] . It is widely acknowledged that children aged below 5 years are at a high risk of mortality in developing countries [54] . Although the causes of death are multi-factorial , malnutrition is a key factor and parasites contribute to a substantial fraction of this under-nourishment [54] , [55] . In Lao PDR , a national deworming program is currently being implemented at the school level in collaboration with the MoH and the Ministry of Education [56] . However , preschool-aged children are currently not part of the project . Our findings of high prevalence and infection intensity of soil-transmitted helminths among preschool–aged children should be considered for future control activities . It is conceivable that including under-5 year-old children into the deworming program might improve their health status . Epidemiologic studies have shown that prevalence and intensity of several parasitic infections are governed by behavioral , socioeconomic , and environmental characteristics [31] , [34] . In the current study , we observed that the consumption of raw or insufficient cooked fish through traditional dishes ( i . e . , “Lap-pa” and “Koy-pa” ) was commonly practiced among the Lao-loum ethnic groups . This is the most likely explanation why the Lao-loum were at a significantly higher risk of O . viverrini infection than Lao-theung . This behavioral trait is known to be a potential risk for acquiring O . viverrini and other fish-borne trematode infections [15] , [57] , particularly in an area where several food-borne trematodes co-exist , such as Lao PDR [16]–[18] , [22] . It is also important to note that intestinal parasites varied according to people's socioeconomic status . Interestingly , S . mekongi was significantly more prevalent among better-off study participants who live along the lower Mekong islands . On the other hand , as expected , the highest prevalence of soil-transmitted helminths , particularly A . lumbricoides and T . trichiura , were observed among the poorest living in the highlands . The latter finding is in line with observations reported from southern P . R . China [31] and other parts of the developing world [58] . People belonging to the poorest wealth quintiles are at higher risk of infection with soil-transmitted helminths . Finally , we found a low prevalence of intestinal protozoa in our study cohort . These findings support the previous observations , which have shown low prevalence of pathogenic intestinal protozoa in Southeast Asia [23] , [41] , [59] . We conclude that multiparasitism is the rule in different eco-epidemiological settings of Champasack province , and most likely elsewhere in Lao PDR . The extent of multiparasitism and the high infection prevalence and intensity with a host of intestinal parasites , most importantly S . mekongi and O . viverrini are a public health problem . Consequently , a chemotherapy-based morbidity control program should be re-implemented without delay . To consolidate progress and ascertain long-term sustainability , other control measures such as health education , improving access to clean water and sanitation in an intersectoral fashion must be considered . | Multiparsitism is a general public health concern in tropical countries , and is of particular importance in the Mekong River basin of Southeast Asia . Here , we report results obtained from an in-depth study of hepato-biliary and intestinal multiparasitism and associated risk factors in three settings of the most southern province of Lao People's Democratic Republic . Multiple species intestinal parasite infections were very common: more than 80% of the study participants harbored at least two and up to seven different intestinal parasites concurrently . Of particular concerns are the high prevalence of the liver fluke Opisthorchis viverrini ( 64 . 1% ) and the moderate prevalence of the blood fluke Schistosoma mekongi ( 24 . 2% ) , as these fluke infections are responsible for severe hepato-biliary morbidity , including the bile duct cancer cholangiocarcinoma . Hookworm was the most common nematode infection ( 76 . 8% ) . We conclude that given the very high prevalence rates of parasite infections and the extent of multiparasitism , regular deworming is warranted and that this intervention should be coupled with health education and improved assess to clean water and adequate sanitation to consolidate morbidity control and ensure long-term sustainability . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"infectious",
"diseases/helminth",
"infections",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"public",
"health",
"and",
"epidemiology/screening",
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases"
]
| 2011 | Helminth and Intestinal Protozoa Infections, Multiparasitism and Risk Factors in Champasack Province, Lao People's Democratic Republic |
Myeloid derived suppressor cells ( MDSCs ) , which suppress anti-tumor or anti-viral immune responses , are expanded in the peripheral blood and tissues of patients/animals with cancer or viral infectious diseases . We here show that in chronic SIV infection of Indian rhesus macaques , the frequency of MDSCs in the bone marrow ( BM ) was paradoxically and unexpectedly decreased , but increased in peripheral blood . Reduction of BM MDSCs was found in both CD14+MDSC and Lin-CD15+MDSC subsets . The reduction of MDSCs correlated with high plasma viral loads and low CD4+ T cell counts , suggesting that depletion of BM MDSCs was associated with SIV/AIDS disease progression . Of note , in SHIVSF162P4-infected macaques , which naturally control viral replication within a few months of infection , the frequency of MDSCs in the bone marrow was unchanged . To investigate the mechanisms by which BM MDSCs were reduced during chronic SIV infection , we tested several hypotheses: depletion due to viral infection , alterations in MDSC trafficking , and/or poor MDSC replenishment . We found that the possible mobilization of MDSCs from BM to peripheral tissues and the slow self-replenishment of MDSCs in the BM , along with the viral infection-induced depletion , all contribute to the observed BM MDSC reduction . We first demonstrate MDSC SIV infection in vivo . Correlation between BM CD14+MDSC reduction and CD8+ T cell activation in tissues is consistent with decreased immune suppression by MDSCs . Thus , depletion of BM MDSCs may contribute to the pathologic immune activation during chronic SIV infection and by extension HIV infection .
Myeloid-derived Suppressor Cells ( MDSCs ) are immature cells of myeloid origin , frequently found in tumor microenvironments and in the blood of cancer patients [1–3] . Bone marrow ( BM ) is the reservoir for MDSCs . Under homeostatic conditions , there is a delicate balance between the immature MDSCs and matured myeloid cells in the BM . Normally , only matured myeloid cells are released from the BM to peripheral blood and tissues . MDSCs maintain a relatively low level in peripheral blood and tissues , and do not expand under normal physiological conditions . In healthy mice , Gr-1+CD11b+ MDSCs constitute 20–30% of the total cells in BM , 2–4% of peripheral blood cells , 2–4% of spleen cells , 2–5% of liver cells , and <1% of lymph node cells [2 , 4 , 5] . Cancer leads to MDSC expansion in tumors , and lymphoid tissues , reaching up to 50% of the total cells in lymph nodes , and bone marrow of tumor-bearing mice[6] , and patients with pancreatic cancers [7] . MDSCs were implicated in the suppression of different immune cells including T cells and NK cells[8 , 9] . Several suppressive mechanisms of MDSCs have been described including: ( 1 ) direct suppressive activities through the production of arginase1 ( ARG1 ) , indoleamine 2 , 3-dioxygenase ( IDO ) , reactive oxygen species ( ROS ) , inducible nitric oxide synthase ( iNOS ) , TGF-β , IL-10 , and PD-L1 expression; ( 2 ) the expansion of regulatory T cells ( Treg ) [10–16] . In humans , MDSCs are a heterogeneous population without specific markers . Two main subtypes , monocytic , and polymorphonuclear ( or granulocytic ) -MDSCs , have been described in humans based on their phenotype and functions . Monocytic ( M ) -MDSCs were defined as HLA-DR-/lowCD33+CD11b+CD14+ ( hereafter referred to as CD14+ MDSC ) , while polymorphonuclear ( PMN , also called granulocytic ) -MDSCs were defined as Lin-HLA-DR-/lowCD33+CD11b+CD15+ ( hereafter referred to as Lin-CD15+MDSCs ) [8 , 17–19] . Recently a new subset called early-stage MDSCs ( eMDSCs ) was added with phenotypical makers as Lin- ( including CD3 , CD14 , CD15 , CD19 , CD56 ) HLA-DR-/lowCD33+ [18] . In some of the studies , cancer leads to the expansion of both PMN and M-MDSCs , with PMN-MDSC as the predominant subset [19 , 20] . Both subsets demonstrated a suppressive effect in vitro . However , M-MDSCs have been identified as the more suppressive subset in vivo , and the loss of PMN-MDSC has not altered tumor incidence [21] . MDSC expansion also occurs in non-cancer settings [12 , 22–29] . For example , in late septic mice: Gr-1+CD11b+ MDSCs were increased dramatically with up to 40% of the total cells in spleen ( vs . 2–4% in normal spleen ) , 90% of the total cells in BM ( vs 20–30% in normal BM ) , and 3–4% of the total cells in lymph nodes ( vs . <1% in normal lymph nodes ) [30 , 31] . MDSCs are significantly increased in peripheral blood of HIV/simian immunodeficiency virus ( SIV ) -infected individuals/macaques [12 , 22 , 23] . Similar to their roles in the tumor microenvironment , MDSCs in HIV/SIV-infections effectively suppress the function of T cells , and thus play an important role in dampening protective immunity [8 , 12 , 22 , 23] . However , there are no details known regarding MDSC distribution and their roles in different tissues during HIV infection in humans . Because SIV-macaque models recapitulate the pathogenesis of HIV infections , they are considered one of the best animal models for AIDS research . In this study , we investigated MDSC tissue distribution , and possible MDSC roles during SIV infection using single suspension cells obtained from eight different anatomic compartments of SIVmac251-infected rhesus macaques . Surprisingly we found that MDSCs in the BM were paradoxically decreased in chronic SIV infection ( dropping from 22% to 8% ) . Furthermore , the reduction of BM MDSC correlated with high plasma viral loads ( VL ) , low CD4+ T cell preservation , and general high CD8+ T cell immune activation . We found that an increase in the trafficking and continuous mobilization of the MDSCs from bone marrow to peripheral blood and tissues might be one cause , along with poor replenishment of MDSCs in BM . A second cause is our discovery of viral infection of MDSCs that may further exacerbate the decrease of MDSCs . Taken together , our data suggest that the decrease in BM MDSC may contribute to immune deficiency and immune activation in macaques chronically infected with SIV .
Because information on tissue distribution of the MDSCs in normal and HIV-1 infected humans is limited , we investigated the distribution of MDSCs in eight different tissue compartments of SIVmac251-infected Indian rhesus macaques . Single cell suspensions from tissues of 12 SIVmac251 chronically infected macaques ( 14 months post-infection ) were obtained and subjected to flow cytometric analysis of MDSCs [23] . We examined the frequencies of two MDSC subsets , which were both HLA-DR-/lowCD33+CD11b+ and either Lin-/low CD15+ ( Lin-CD15+MDSC ) or CD14+ ( CD14+ MDSC ) ( Fig 1A ) [19 , 23 , 32–34] . These animals showed a typical dynamic of the plasma VL after pathogenic SIVmac251 infection with peak VL at Week 2 followed by persistent set-point VL from Week 6 to the end study at week 60 ( Fig 1B ) . Among the eight compartments from the chronically infected macaques , interestingly , high frequencies of both types of MDSCs were found in BM ( 4 . 8% and 3 . 1% for Lin-CD15+ and CD14+ MDSCs , respectively , P<0 . 05 for BM vs . all the other compartments except liver and PBMC in both types of MDSCs after Dunn’s multiple comparison tests ) , liver ( 0 . 8% and 2 . 0% for Lin-CD15+ and CD14+ MDSCs , P<0 . 05 for liver vs . axillary LN , mesenteric LN , and colon lamina propria ( LP ) in both types of MDSCs after Dunn’s multiple comparison tests ) , and PBMC ( 0 . 3% and 1 . 1% for Lin- CD15+ and CD14+ MDSCs , P<0 . 05 for liver vs . axillary LN , mesenteric LN , and colon LP in both types of MDSCs after Dunn’s multiple comparison tests ) with intermediate levels of MDSCs in spleen and gut; but very few of either subset in the LNs ( Fig 1C and 1D ) . For both subsets of MDSCs , when we analyzed the associations among different tissues , we could not find any correlations between any pairs of compartments ( S2 and S3 Tables ) , suggesting compartmentalization . For example , the animal with the highest CD14+MDSCs in the BM ( 8% ) only had 0 . 2% of CD14+MDSCs in the PBMC . In addition to MDSCs , we also measured viral-specific CD8+ T cell responses by gag CM9 peptide-Mamu A*01 Dextramer as well as the total proliferating CD8+ T cells by Ki67 in these animals , because both can be affected by MDSCs . Because all the animals were Mamu A*01+ , SIVgag-CM9 Dextramer+ cell frequencies were measured to represent viral-specific CD8+ T cell responses[35] . As shown in Fig 1E , SIVgag-CM9 Dextramer+ cell frequencies in PBMC was 1 . 7% , much lower than in the remaining tissues , which ranged from 3% to 5% of total CD8+ T cells ( P<0 . 05 for PBMC vs . axillary LN , mesenteric LN , spleen and BM after Dunn’s multiple comparison tests ) . However , the frequencies of CM9 Dextramer+ in total CD8+ T cells from different tissues were highly correlated with each other ( S4 Table ) , suggesting the equilibration of antigen-specific T cell responses among different tissue compartments . The proportion of total CD8+ T cells that expressed Ki67 ranged from 35 to 54% , without significant differences among the compartments ( Fig 1F ) . We found that the frequencies of Ki67+ within the total CD8+ T cells from different tissues were also associated with each other ( S5 Table ) . Overall , the tissue distribution patterns of viral-specific CM9 Dextramer+ and total Ki67+ CD8+ T cells were different from those of MDSCs in the same SIV-infected macaques . It is of note that the MDSCs did not accumulate in the lymphoid tissues of the SIV-infected macaques during the chronic stage , which marked a sharp contrast to cancer and some infectious diseases[36] . Due to the tissue distribution shown in Fig 1 , we then focused on the MDSCs in BM , liver and PBMC . One important question was whether the frequencies of the MDSC subsets were altered during chronic SIV infections . To make the comparisons , we measured the frequencies of MDSCs among PBMCs ( from two cohorts , n = 40 and 28 ) , liver ( from two cohorts , n = 4 and 4 ) and BM ( from two cohorts , n = 28 and 12 ) from naïve macaques . Using a similar gating strategy ( Fig 2A ) , we found that while naïve animals had a low frequency of MDSCs in PBMCs ( 0 . 09±0 . 02% and 0 . 20±0 . 04% for CD14+ and Lin-CD15+MDSC ) , PBMC MDSC frequencies were elevated after SIV infection ( 1 . 15±0 . 14% and 0 . 27±0 . 08% for CD14+ and Lin-CD15+MDSC in PBMCs ) ( Fig 2B ) . This is consistent with the results from previous studies showing an expansion of MDSCs in the PBMCs after HIV-1/SIV infection[12 , 22 , 23] . Though both subsets of MDSCs were more prevalent , we observed a more dramatic increase of the CD14+ MDSC subset than the Lin-CD15+MDSC subset: 13-fold increase of CD14+ MDSC compared with almost no change of Lin-CD15+MDSC after SIV chronic infections . Due to the small number of naïve liver samples that were available , we could not reach any conclusions as to effect of SIV infection on liver MDSCs ( Fig 2C ) . Surprisingly , we found that the total number of MDSCs ( defined as the sum of CD14+ MDSC and Lin-CD15+MDSC ) in BM was paradoxically decreased 14 months after SIVmac251 infection ( Fig 2D ) . Specifically , there was roughly a 4 . 7-fold reduction in the CD14+ MDSC subset ( from 14 . 5±1 . 3% in naïve animals vs . 3 . 1±0 . 07% in SIV-infected ones , P<0 . 0001 ) , and a 1 . 6-fold reduction in the Lin-CD15+MDSC subset ( 7 . 7±0 . 6 vs . 4 . 8±0 . 7% respectively , P = 0 . 005 ) ( Fig 2D ) . This pattern of the MDSC distribution in the SIV-infected animals was markedly different from that in tumor-bearing mice , where BM MDSCs are elevated [4 , 5] . Furthermore , we found two subpopulations of CD14+ MDSCs in the BM: CD14high and CD14intermediate MDSCs ( S1 Fig ) . Most CD14intermediate population were HLA-DR-/low cells in SIV-infected and naïve BM . This HLA-DR-/low CD14intermediate population was prominent in the naive animals , but lower in the SIV-infected animals ( S1 Fig ) . Compared to naïve BM , the CD14+ population was also decreased in the SIV-infected BM ( S2 Fig ) . Because MDSCs , especially Lin-CD15+ subset , are sensitive to cryopreservation [37] , and we used frozen cells throughout the whole study , there is a possibility that SIV infection might differentially affect the loss of SIV-infected MDSCs compared to naïve cells . To test this , we compared the frequencies of MDSC subsets in paired specimens of fresh vs frozen BM and PBMC obtained from the SIV-infected ( BM: n = 6; PBMC: n = 7 ) and naïve animals ( BM: n = 12; PBMC: n = 5 ) . The PBMC and BM samples from the SIV-infected and naïve animals were isolated , and divided into two aliquots , one was immediately stained and analyzed by flow-cytometry , the other was cryopreserved in liquid N2 and thawed for testing 4–5 months later . After cryopreservation and thawing , 70–80% of CD14+MDSCs were preserved in the naïve and SIV-infected BM samples; 65% of CD14+MDSCs were also preserved in the SIV-infected PBMC samples ( S3 Fig ) . Consistent with the previous study [37] , only 20% of Lin-CD15+ MDSCs were preserved in the naïve and SIV-infected BM samples ( S3 Fig ) , suggesting Lin-CD15+MDSC subset was more sensitive to cryopreservation . Importantly , we did not observe any significant difference of MDSC subset loss for the BM samples between the SIV-infected and naïve animals . Thus , SIV infection does not differentially affect the loss of MDSC subsets from BM compared to naïve ones . We then asked whether SIV VLs correlated with the reduction of BM MDSCs . Among the 12 SIV-infected macaques , five of them controlled SIV replication , with plasma VLs below 5 , 000 copies/ml . The remaining 7 animals did not control SIV replication , with VLs >5 , 000 copies/ml . We found that the controllers maintained high BM MDSC frequencies , whereas the non-controllers had reduced BM MDSC frequencies ( Fig 3A ) . This change was not found in the PBMC or the liver , although the liver showed the same trend ( Fig 3A ) . Indeed , the frequencies of both subsets of MDSCs in BM inversely correlated with VL ( Fig 3B ) . In the BM of chronically SIV-infected macaques , the frequencies of CD14intermediate MDSCs inversely correlated with plasma viral loads better than that of CD14high MDSCs ( S4 Fig ) . We also found that both subsets of MDSCs in the BM positively correlated with absolute blood CD4 count ratio ( necropsy end-point CD4+ T cell count/pre-challenge CD4+ T cell count ratio ) ( Fig 3B ) . The association between BM MDSC depletion and high VLs and low CD4+ T cell counts suggests that MDSC depletion may contribute to disease progression , or at least serve as marker of disease progression . The data suggested that exposure to pathogenic SIV infection decreased MDSC frequencies in the BM . We then asked if similar changes occurred in Indian rhesus macaques infected with SHIVSF162P4 . All the SHIVSF162P4-infected animals demonstrated high peak VLs during the first 6 weeks of infection , followed by viral control 3 to 4 months post-infection ( Fig 4A ) . At 11 months post-infection , when VL was near or below the detection limit , there was no change in BM MDSC frequency ( Fig 4B ) . In those samples , there was no change in either the Lin-CD15+MDSC or CD14+ MDSC ( Fig 4B ) . The same was also true for MDSCs in the 11-month post SHIV-exposed PBMC and liver ( Fig 4C and 4D ) . Thus , infection with SHIV without persistent viremia did not affect MDSC frequency in BM , but infection with pathogenic SIV depleted BM MDSC . We next investigated what caused the reduction of MDSCs especially in the CD14+ subset in the BM of the SIV-infected macaques . We proposed three hypotheses: depletion due to viral infection , increase in MDSC mobilization/trafficking out of BM , and poor MDSC replenishment from hematopoietic stem cells . We then sought to clarify the consequences of the MDSC reduction in the BM . Because the main known function of MDSCs is to limit excessive T cell proliferation , and thus control tissue damage , we proposed that BM MDSC reduction would be associated with T cell activation , especially among CD8+ T cells . We measured the frequencies of Ki67+ CD8+ T cells in the tissues of the SIV-infected macaques , and then performed correlation analyses between frequency of BM CD14+ / Lin-CD15+MDSCs and the frequencies of Ki67+ CD8+ T cells in these compartments . We found that the frequency of BM CD14+ MDSCs inversely correlated with the frequency of Ki67+CD8+ T cells in BM , PBMC , spleen , ALN , but not MLN , liver , ileum LP and colonic LP ( although MLN and colonic LP showed nearly significant trends ) ( Fig 9A–9H ) ; whereas Lin-CD15+ MDSC did not correlate with the frequency of Ki67+CD8+ T cells in any of the compartments except PBMC ( S9 Fig ) . Overall , the association of BM CD14+ MDSC frequency with CD8+ T cell immune activation , plasma vRNA levels , and CD4+ T cell loss suggests that the reduction in BM CD14+ MDSC contributes to the pathogenesis of the HIV/SIV infections .
Cancer and some infectious/inflammatory diseases lead to MDSC accumulation in peripheral blood and multiple tissue compartments , including BM[1–6] . A higher frequency of MDSCs in a cancer setting is usually associated with a poor prognosis[7 , 34] . In this study , we found unexpectedly that BM MDSCs paradoxically decreased after chronic SIV infection compared with healthy controls . This was in sharp contrast to the general increase of MDSCs observed in BM during cancer and other infectious/inflammatory diseases , and also contrary to the MDSC expansion in HIV/SIV-infected PBMCs [12 , 22 , 23] . We further demonstrated that even though the changes were among both CD14+ and Lin-CD15+MDSC subsets in BM , and both subsets correlated with high plasma vRNA level and CD4+ T cell loss , it was only the reduction of CD14+ MDSC that correlated with the high levels of CD8+ T cell activation . These changes suggest that the loss of BM MDSCs , especially the CD14+ subset , was associated with the progression to HIV/AIDS disease . We then investigated the alteration of MDSCs in the BM of a less pathogenic SHIV-macaque model , in which the animals had moderate plasma vRNA levels at the first 6 weeks of infection , but spontaneously controlled the viral replication after several months . These animals had no signs of immune activation , and did not develop AIDS-related disease . Indeed , in these animals , we found that both subsets of BM MDSCs were unchanged 11 months post-infection . This further supported the association of MDSC decrease in BM with HIV/AIDS disease progression . During HIV/SIV infection , MDSCs are thought to contribute to immune-pathogenesis by dampening protective immunity through the direct inhibition of T cell function , especially the viral-specific CD8+ T cell responses , and thus are deleterious[12 , 22 , 23] . However , our data suggest more complex and somehow paradoxical roles of MDSCs in HIV/SIV infections . MDSCs suppress anti-viral specific immune responses , which hindered the control of HIV replication [22 , 23] . However , MDSCs also have the capacity to antagonize immune activation , which plays a key role in driving viral transmission and replication and many of the disease processes associated with AIDS [46–48] . Consistent with this , the CD14+ MDSCs in the BM inversely correlated with immune activation ( Ki67+CD8+T cells ) of multiple tissue compartments , suggesting that depletion of BM MDSCs contributes to the immune activation in chronically infected macaques . During late HIV infection , the balance of immune suppression and immune activation tips in favor of the latter due to factors such as gut microbe translocations[49 , 50] . It has been proposed that MDSCs are responsible for restoring the balance through various mechanisms , including arginase-1 , inducible nitric oxide synthase , reactive oxygen species , and induction of regulatory T cells [46 , 51–53] . However , the reduction of MDSCs in the BM may impair this function , and contribute to the relentless immune activation in HIV/SIV chronically infected patients/animals . We are puzzled by the fact that the MDSC reduction in BM correlated with the CD8+ T cell proliferation in not only the MDSC-high tissues but also the MDSC-low tissues . MDSCs did not play a direct role in inhibiting CD8+ T cells proliferation in situ in the tissues with low MDSC frequencies , whereas in the MDSC-high compartments , MDSCs could inhibit the proliferation of CD8+ T cells efficiently . When the CD8+ T cells are equilibrated well among different tissue compartments in the SIV-infected animals , MDSCs suppressing T cell in high-MDSC tissues could affect the levels of T cells in the MDSC-low tissues indirectly when they re-equilibrate . MDSCs could play an important role in suppressing immune activation during HIV infection . Another suppressive mechanism of MDSCs could be the induction of CD4+ Treg cells , which could travel to other MDSC-low compartments , and suppress immune responses locally[12–14] . Furthermore , MDSCs induced a type 2 polarization of macrophages in tissues , which were functionally inhibitory [54 , 55] . It is worth mentioning that even though both subsets of MDSCs were decreased in the chronically SIV-infected BM , and associated with markers of disease progression: high plasma vRNA levels , low CD4+ T cell counts , only loss of CD14+ MDSCs was associated with high immune activation . Thus CD14+ MDSC and Lin- CD15+ MDSC were not only phenotypically distinct , but also played different roles in SIV infections . In cancer settings , CD14+ MDSCs have been shown to be the dominant suppressive populations of MDSCs , and correlated with cancer incidence , whereas G-MDSC did not correlate with clinical outcomes , even though both subsets were increased[21] . We also observed more dramatic changes of CD14+ MDSC subsets in both PBMC and BM after SIV infection . This implied that the two subsets of MDSCs were under different regulatory mechanisms , and CD14+ MDSCs might be the main subset that respond to SIV infection , and determine the outcome of SIV infections . However , as cryopreservation changes the PMN-MDSC ( Lin-CD15+ ) numbers and function[37 , 55] , and all the samples used in this study were cryopreserved ones , we should be cautious on the interpretation of data on this subset . In the present study , to investigate the mechanisms by which the MDSCs were decreased in the SIV-infected BM , we found that infection by virus , migration out of BM , and/or poor self-replenishment of MDSCs might all play important roles . With the data we have , it is impossible to dissect which were more critical . It has been reported that infectious/inactivated HIV , or gp120/gp41 , and tat proteins directly promoted MDSC expansion in an in vitro culture system using healthy PBMCs [13 , 14 , 22] . However , we doubt that virus/viral proteins were the key contributors to the MDSC accumulation in the PBMCs of the infected macaques based on 1 ) dramatic increase of M-MDSCs in the PBMCs of HIV-1-infected patients on antiretroviral therapy with undetectable viremia [14]; 2 ) no correlations of VLs with MDSC accumulation in PBMC at different time-points in the same animal[23]; and 3 ) decreased rather than increased MDSCs in BM during chronic SIV infection . Consistent with the finding of CD14+ MDSC susceptibility to HIV-1 infection in vitro [22] , we further demonstrated by three independent mutually corroborative experimental criteria ( gag expression relative to CD3 and CD4 expression , spliced vRNA expression , and p27 protein production ) that both subsets of MDSCs were able to be infected in vivo . In contrast to the suggestion that the infection of myeloid cells might cause developmental defects in myeloid cells , and result in accumulation of MDSCs in the in vitro study [22] , our data suggest that infection of MDSCs by SIV in vivo might eventually lead to MDSC depletion directly or indirectly . In the SIV-infected PBMC and BM , the SIVgag+ MDSCs were a small fraction of the total MDSCs . If infection itself led to MDSC expansion , we would expect a big pool of SIVgag+ MDSCs in the infected animals . We noticed that the relative expression level of SIVgag mRNA per cell was comparable between CD4+ T cells and MDSCs ( Fig 5C ) , which suggested that MDSCs could be productively infected by SIV , like their CD4+ T cell counterparts . Productive infection was confirmed both by detection of spliced SIV mRNA ( S7 Fig ) , and by staining for expression of viral gag protein ( Fig 6 ) , under conditions in which contamination with T cells could be ruled out as an explanation . If MDSCs are productively infected with SIV , viral-induced depletion might play an important role . However , further evidence is needed to confirm whether SIV infection leads to MDSC death . HIV-1 infected patients usually develop hematopoietic dysfunction , which is mainly mediated by the altered production of stromal cell-derived hematopoietic growth factors , infection of the BM accessory cells , and impaired stromal functionality[41–43 , 56] . This abnormality could lead to defects of MDSC replenishment . Indeed , we found decreased proliferation ( Ki67+% ) of both subsets of MDSCs in BM after SIV infection . The inverse or reciprocal relationship of MDSCs in BM and PBMC , e . g . , increased in PBMC and decreased in BM , promoted us to investigate the possibility of MDSC redistribution during SIV infection . We compared the chemotactic activities of plasma from the pre- and post-SIV infection blood using in vitro chemotaxis assays . It turned out that the chemotaxis index of MDSCs , but not of T cells or total cells , was increased during chronic SIV infection . We believe that this could directly lead to MDSC loss in BM , as a consequence of more MDSCs’ migrating out of the BM . The preferential trafficking of MDSCs towards the post-infection plasma suggested that certain chemokine ( s ) /cytokine ( s ) in the chronic SIV infected plasma could preferentially pull the MDSCs out of BM . In the attempt to dissect which chemokine ( s ) /cytokine ( s ) were responsible , we tested CCL2 , known to affect myeloid cell migration . We found that CCL2 correlated with CD14+ MDSC accumulation in the PBMC , but did not correlate with MDSC chemotaxis index measured in vitro . Since CCL2 was one of multiple cytokines/chemokines induced by SIV , other chemokines/cytokines might also play roles in attracting MDSCs to the circulation . Antibody blocking for CCL2 would be a good way to assess the role of CCL2 played in chemotaxis . Nevertheless , as MDSCs are such heterogeneous populations and no single chemokine receptor is truly specific for MDSCs , the net chemoattractant effects of multiple chemokines in the plasma might account for the results of the in vitro assays . One caveat was that this was a cross-sectional study of MDSC alteration in BM during the chronic infection stage . A longitudinal study with multiple time-points during acute and chronic infection stages will give us a better picture of the MDSC dynamics , and better understanding of the roles of MDSCs during SIV infection . We speculate that there is a delicate balance of immunity and suppression established at the onset of the SIV infection . During the early stage of the pathogenic SIV infection , it is possible that the MDSC increase/expansion also occurred in BM , as we have observed in the SHIV-infected macaques , in which MDSCs initially significantly increased 4 months post-infection ( S10 Fig ) , and then reverted to normal . In the pathogenic SIV infections , however , the presence of persistent high viral loads , and the poor self-replenishment by MDSC proliferation due to bone marrow stromal functional defects could lead to the gradual decrease of MDSCs in BM . Future studies on the kinetics of MDSCs in the BM of pathogenic SIV infections will clarify this . This depletion of MDSCs eliminates one of the body’s key mechanisms to limit tissue damage from aberrant immune activation . Thus , in the setting of acute SIV infection , MDSCs can limit antiviral T cell responses , whereas in chronic infection , BM MDSC depletion can contribute to the disease progression . Overall , MDSCs act as a double-edged sword in HIV/SIV-infection , and the decrease of MDSCs in BM after SIV infection could serve as an indicator of immune regulatory exhaustion . Further elucidating the roles of MDSCs in HIV/SIV infection will contribute to our understanding of the immune-pathogenesis of HIV infections and contribute to the development of therapeutic approaches for HIV/AIDS .
All the adult Indian rhesus macaques ( Macaca mulatta ) were used with the approval of Institutional Animal Care and Use Committees . The macaques were housed at the NCI Animal Facility , Bethesda , MD ( Protocols No . VB010 and VB011 approved by the NCI IACUC ) , which is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) International . The housing and the standard practices closely followed the recommendations of the Standards and Guide for the Care and Use of Laboratory Animals of the United States—National Institutes of Health . All efforts including provision of peri-operative and post-operative analgesia were made to minimize discomfort of the animals . Details of animal welfare , including housing , feeding , environmental enrichment , and steps to minimize suffering , were in accordance with the Guide and the recommendations of the Weatherall report , ‘‘The use of non-human primates in research” , as approved by the IACUCs . Macaques were housed in temperature controlled facilities with temperature of 21–26°C , humidity of 30%– 70% , and a 12 h light/dark cycle . A commercial primate diet and fresh fruit were provided twice daily with water freely available at all times . Macaques were monitored twice daily for overall health including activity , food and water intake . The macaques were singly housed in stainless steel wire-bottomed cages due to the nature of the experiment . Rotating toys , visual and auditory stimuli , and foraging opportunities were provided daily . The animals were anesthetized with approximately 10 mg/kg of ketamine hydrochloride injected intramuscularly for blood and bone marrow collections . When IACUC defined endpoints were reached , macaques were humanely euthanized with an overdose of barbiturate in accordance with the recommendations of the most recent American Veterinary Medical Association Panel on Euthanasia . The 12 simian immunodeficiency viruses ( SIV ) mac251-infected animals have been described in our previous studies [23 , 57] ( S1 Table , cohort 1A ) . Briefly , the animals were intrarectally challenged by three serial SIVmac251 viruses at 2-week intervals until they have been infected . The SIVmac251 viral stock was provided by Nancy Miller of the National Institute of Allergy and Infectious Diseases ( NIAID ) . Twelve simian–human immunodeficiency virus ( SHIV ) -infected macaques were also included in this study ( 6 males and 6 females; age: 3 , 3±1 . 3 years; weight: 4 . 6±1 . 7 kg ) ( S1 Table , cohort 2A ) . The animals were intrarectally exposed to 8 serial SHIVSF162P4 viral challenges with a week of interval ( 1:35 dilution ) until they were infected . After the viral challenges , SIV/SHIV RNA levels were monitored by Advanced BioScience Laboratories , Inc . The cut-off threshold for viral RNA detection was 50 copies/ml . Blood and tissue samples from naïve macaques were collected and included in the study for comparisons ( S1 Table , cohort 1 , 2 , 3 and other ) . Liver tissues from 8 naive ( 4 adenovirus-exposed , 4 non-exposed , all SIV/SIHV-negative ) macaques were used as normal controls ( S1 Table , other ) . Other naïve macaques which have been enrolled in this study were described in S1 Table . Additional SIVmac251 infected bone marrow samples were obtained from California National Primate Research Center ( S1 Table ) . Single-cell suspensions from blood , bone marrow , lymph nodes , and spleens were prepared as previously described [58] . We have used the same protocol to process bone marrow and blood samples . Briefly , the samples were first 1:1 diluted with 1XPBS or HBSS , and then the diluted samples were overlaid on lymphocyte separation medium Lympholyte-H ( Human , Cedarlane , Ontario , Canada ) . After centrifugation at 800 x g for 20 min at room temperature ( 25°C ) , with centrifuge brake on OFF position , the middle opaque fluid containing the PBMC/BM cells was collected . The cells were washed with R10 medium three times before cryopreservation in liquid nitrogen freezers . Ileum lamina propria , colon lamina propria , and liver were collected from the macaques as previously described with modification[23 , 57–59] . Briefly , the tissues were first rinsed 2–3 times with 1XHBSS containing 5mM EDTA and 1-2mM DTT to remove mucous and intestinal intraepithelial lymphocytes ( IEL ) ( only for the gut ) . The tissues were then cut into small pieces and incubated with Liberase ( Roche ) at 37°C for 30 min . After enzyme digestion , the tissue chunks were mashed through a syringe end and filtered through a 100 μm cell strainer . After washes , the cells were re-suspended in 40% Percoll , and underlaid with 80% Percoll . The cells in the interface were collected after spinning down for 20min , 800 x g at room temperature ( 25°C ) , with centrifuge brake on OFF position . The cells were cryopreserved in liquid nitrogen freezers after three washes with R10 medium . All cell samples used in this study were cryopreserved . For flow cytometric analysis , the single-cell suspensions from different compartments were first incubated with Fc Receptor blocking reagent ( Miltenyi Biotec ) , and then stained with viability dye ( Invitrogen ) and antibody mixtures including anti-CD45 ( BD Pharmingen ) to exclude the dead and CD45 negative cells . For immune activation , the following antibodies were used: CD3-PE-Cy7 , CD4-BV605 , CD8-APC-Cy7 , CD14-V450 , Ki67-APC , HLA-DR PE-Cy5 , and CCR5-PE ( BD Pharmingen ) ; CD69-Alexa Fluor 700 ( Biolegend ) ; and CD38-FITC ( STEMCELL Technologies ) . For MDSC analysis , the following antibodies were used: CD3-PE-Cy7 , CD4-BV605 , CD14-V450 , Ki67-APC , HLA-DR-APC-Cy7 , Lin-FITC ( Lin1 , BD Pharmingen ) ; CD33-PE ( Miltenyi Biotec ) , CD15-Alexa700 and CD11b-PE-Cy5 ( Biolegend ) . CM9 Dextramer was obtained from Immudex ( Denmark ) . For gag-P27 staining of MDSCs , besides the antibodies listed above , CD2-PE-Cy7 ( BD Pharmingen ) , CD68-Perp-cy5 . 5 , and P27 antibody [60] ( clone KK64 , cat#2321 , from NIH AIDS reagent program ) were also added to the panel . For gag-P27 staining , we have incubated the cells for 2 hrs with 5 μl of human TruX blocker ( Biolegend ) , and 2 μl of human IgG ( 1 mg/ml , from R&D systems ) before adding the surface antibody mixture . We also extended the P27 staining overnight on ice . An LSRII flow cytometer was used for data acquisition , and FlowJo software ( Tree Star Inc . ) was used for data analyses . The gating strategies for MDSCs were shown in Fig 1A and Fig 2A . MDSCs were sorted by using a BD FACSAria ( BD Biosciences ) with the following markers: CD3–HLA-DR–CD11b+CD33+ cells from live PBMCs ( purity greater than 95% ) or spleen . CD4+ T cells were sorted as positive controls . The sorted CD4+ T cells were either serially diluted in 10-fold dilutions or directly put into Trizon . After Trizon-lysis , the sorted cells were then subjected to RNA isolation , reverse transcription , and qPCR reactions to detect the relative expression levels of SIVgag , spliced SIV rev , tat , nef , and vif , CD3 , CD4 , and GAPDH in MDSCs and CD4+ T cells ( ABI , Bioline ) [58] . Taqman probe and primer sets for gag[61] , CD3 , CD4 , and GAPDH were used ( ABI ) . Spliced SIV rev , tat , nef , and vif primers and probes were synthesized at ABI , and multiplexed to detect the rev , tat , nef , and vif targets simultaneously [62] . Each PCR mixture contained 5 μl of cDNA , 10 μl of 2X universal PCR master mix ( ABI ) or SensiFast probe kit ( Bioline USA Inc . ) , 1 μl of primer/probe set . All PCR were run using ABI 7500 with program consisting of 2 min at 50°C , 10 min at 95°C , and then followed by 45 cycles of 15 sec at 95° C , and 1 min at 60°C . Relative mRNA expression levels were compared by the comparative threshold cycle ( Ct ) method of relative quantitation ( PerkinElmer User Bulletin no . 2 ) as described[63] , except in cases where standards were available to determine copy number , as indicated . As each cycle represents a doubling ( log2 ) , subtracting Ct for GAPDH is essentially taking a ratio to or normalizing to the GAPDH housekeeping gene . The 96-well ChemoTx chemotaxis system ( NeuroProbe Inc . , 5um pore ) was used for chemotaxis assays . The lower wells were blocked with 301 μl of 1% BSA for 30 min at room temperature , which was aspirated and replaced with 301 μl of plasma from pre- or post-infected macaques as test samples . 301 μl of RPMI-1640 /0 . 1% BSA buffer was included in the experiments as buffer controls . Single cell suspensions from bone marrow of naive macaques were collected and 2 x 105 bone marrow single cell suspensions were re-suspended in 50 μl RPMI-1640 /0 . 1% BSA and loaded above the membrane . After incubation for 2 hr at 37°C , 5% CO2 , the top wells were removed with a scraper and the migrated cells in the bottom wells were counted , and stained with the antibody mixture: CD3-PE-Cy7 , CD14-V450 , HLA-DR-APC-Cy7 , Lin-FITC , CD33-PE , and CD11b-PE-Cy5 , Cd45-Alexa 700 , and yellow viability dye . 50 μl of CountBright absolute counting beads ( Molecular Probes ) were added to each tube before data acquisition using an LSRII flow cytometer , and FlowJo software ( Tree Star Inc . ) was used for data analyses . Chemotaxis index was defined as the ratio of the absolute cell numbers in the test samples over the absolute cell numbers in the buffer control samples . We performed statistical analyses with Prism version 6 ( Graph Pad ) . One-way ANOVA with Dunn’s multiple comparison corrections , and Mann-Whitney , and Wilcoxon tests were used as shown in the figures . Spearman analysis was used for correlations . A two-sided significance level of 0 . 05 was used for all analyses . | Both cancer and infectious diseases including HIV/AIDS lead to the accumulation of myeloid-derived suppressor cells ( MDSCs ) , which can effectively suppress anti-tumor and anti-viral T cell responses to dampen protective immunity . Using a macaque model , we found unexpectedly that the MDSCs in bone marrow ( BM ) decreased after chronic simian immunodeficiency virus ( SIV ) infection compared with healthy controls . This was in sharp contrast to the general increase of MDSCs observed in BM during cancer and other infectious/inflammatory diseases , and also contrary to the MDSC expansion in HIV/SIV-infected PBMCs . We further demonstrated that the loss of MDSCs in the bone marrow was associated with the progression to AIDS disease . Investigating the mechanisms by which the MDSCs were decreased in the SIV-infected bone marrow , we found that the possible mobilization of MDSCs from bone marrow to peripheral tissues and the slow self-replenishment of MDSCs in the bone marrow , along with the viral infection-induced depletion , all contribute to the observed bone marrow MDSC reduction . Indeed , this is the first demonstration to our knowledge of SIV infection of MDSCs in vivo . Because of the suppressive nature of the MDSCs , the CD8+ T cells might not be effective in killing the virally infected MDSCs . It is tempting to speculate that MDSCs may constitute latent reservoirs . Overall , our data showed that MDSCs act as a double-edged sword in HIV/SIV-infection , and the decrease of MDSCs in bone marrow after SIV infection could serve as an indicator of immune regulatory exhaustion and also contribute to the observed immune hyperactivation seen in HIV/AIDS . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"immune",
"physiology",
"pathogens",
"immunology",
"microbiology",
"vertebrates",
"animals",
"mammals",
"retroviruses",
"primates",
"immunodeficiency",
"viruses",
"viruses",
"bone",
"marrow",
"cells",
"rna",
"viruses",
"cytotoxic",
"t",
"cells",
"old",
"world",
"monkeys",
"white",
"blood",
"cells",
"monkeys",
"animal",
"cells",
"medical",
"microbiology",
"t",
"cells",
"microbial",
"pathogens",
"blood",
"plasma",
"siv",
"macaque",
"immune",
"system",
"blood",
"cell",
"biology",
"anatomy",
"viral",
"pathogens",
"physiology",
"bone",
"marrow",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"lentivirus",
"amniotes",
"organisms"
]
| 2017 | Paradoxical myeloid-derived suppressor cell reduction in the bone marrow of SIV chronically infected macaques |
Immunogenicity is a major problem during the development of biotherapeutics since it can lead to rapid clearance of the drug and adverse reactions . The challenge for biotherapeutic design is therefore to identify mutants of the protein sequence that minimize immunogenicity in a target population whilst retaining pharmaceutical activity and protein function . Current approaches are moderately successful in designing sequences with reduced immunogenicity , but do not account for the varying frequencies of different human leucocyte antigen alleles in a specific population and in addition , since many designs are non-functional , require costly experimental post-screening . Here , we report a new method for de-immunization design using multi-objective combinatorial optimization . The method simultaneously optimizes the likelihood of a functional protein sequence at the same time as minimizing its immunogenicity tailored to a target population . We bypass the need for three-dimensional protein structure or molecular simulations to identify functional designs by automatically generating sequences using probabilistic models that have been used previously for mutation effect prediction and structure prediction . As proof-of-principle we designed sequences of the C2 domain of Factor VIII and tested them experimentally , resulting in a good correlation with the predicted immunogenicity of our model .
Protein-based drugs ( biotherapeutics ) are increasingly used to treat a wide variety of diseases[1 , 2] . Although biotherapeutics show high activity and specificity at the initiation of treatment , the gradual build-up of a patient immune response is a bottleneck for even wider usage[3] . The immunogenicity of the biotherapeutic is influenced by multiple factors that can be roughly divided into extrinsic—such as dosage , rout of administration , duration and production impurities—and intrinsic properties like the protein sequence or post-translational modifications [3] . This immune response involves the formation of anti-drug antibodies ( ADAs ) that target the biotherapeutic itself and cause loss of effect or adverse reactions[3–5] . A prominent example of this adverse effect is in the treatment of hemophilia A ( HA ) with coagulation Factor VIII , where ADAs develop in 10–15% of all HA patients and as much as 30% of those patients with the most severe form of HA[6] . Patients with the highest need for therapy are thus least likely to benefit . This correlation between severity of the disease and lack of efficacy follows from the fact that the immune system is more likely to recognize the therapeutic Factor VIII as foreign the more severe the natural mutation is , where mutations that cause a total loss of Factor VIII production are most strongly associated with ADA development[7 , 8] . The reduction of the immunogenicity has thus become a major step in a the development of a biotherapeutic[5] . The primary focus of reducing immunogenicity has been on humanized monoclonal antibodies ( mAbs ) that are comprised of foreign complementarity-determining regions in the variable regions , with the remainder of human origin , and , more recently , on fully human mAbs using bioengineering techniques[9 , 10] . However , these approaches are not generally applicable to other classes of biotherapeutics and even humanized and full human mAbs can still induce a clinically relevant anti-drug immune response , likely through the CD4+ T-cell mediated adaptive immune system[11 , 12] . The CD4+ T-cell activation is induced by the recognition of linear sequential peptides ( called epitopes ) derived from the therapeutic protein , which are presented on human leucocyte antigen ( HLA ) class II molecules of antigen presenting cells . Therefore , the systematic removal of these epitopes by sequence alteration ( termed de-immunization ) has been successfully used as an alternative approach to reduce the immunogenicity of mAbs and other therapeutic proteins [12–16] . In recent years , computational screening approaches have been developed to suggest protein sequences with reduced overall immunogenicity . The simplest approaches focused solely on introducing point mutations to reduce the amount of CD4+ T-cell epitopes by applying well-established epitope prediction methods [17–19] . However , the suggested mutations can have a significant impact on the stability and function of the protein . Naïve approaches not considering the structural impact on the protein will inevitably produce inactive designs . More advanced methods therefore try to exclude potentially harmful mutations by predicting their impact with various metrics [20] . The most recent approaches simultaneous optimize the number of deleted epitopes as well as the stability of the protein approximated either using structural or simple sequence information [21–23] . Recent advances in statistical protein modeling now allow to accurately infer the tertiary structure [24–28] and mutational effects [29–31] of proteins using evolutionary information contained in an multiple sequence alignment of a protein family . The statistical global pairwise entropy model used for protein inference accurately captures co-evolving sites within a protein which can be utilized to identify structural and functional important position using evolutionary couplings ( EC ) analysis , infer the protein structure , and predicted the effects of mutational changes . In this work , we present a novel formulation of the de-immunization problem that uses , for the first time to our knowledge , the maximum entropy model for protein design . Incorporating the maximum entropy model , as opposed to force-field based approaches such as FoldX[32] that have been previously used for protein de-immunization , has four distinct advantages: ( i ) The statistical model does not require a known structure or depend on the conditions in which the structure was measured . ( ii ) It implicitly considers constraints on residues from interactions with ligands and other proteins , and ( iii ) models interactions between mutations rather than early filtering of deleterious singles . ( iv ) A de-immunization approach using the maximum entropy model is likely to generate more viable structures as it minimizes potential damage to protein function at the same time as minimizing the immunogenicity of the biotherapeutic design . The latter can also be achieved by incorporating a force field , such as AMBER[33] , into the optimization process [21 , 22] , which however complicates the de-immunization formulation . As the frequencies of HLA alleles differ drastically between populations , the immunogenicity of the biotherapeutic differs as well . It is thus imperative to design a biotherapeutic for a specific target population considering their HLA allele frequencies , as opposed to treating each HLA allele equally important during the design process , as all previous methods have done . We therefore developed a new quantitative immunogenicity objective that builds on HLA affinity prediction methods for immunogenicity approximation , as their exists a strong correlation between immunogenicity and HLA binding affinity[34] , and considers the HLA allele distributions within different populations . The resulting de-immunization model does not require known structural information about the protein , summarizes functional and structural information that might not be captured by a structure-based model , and considers the varying HLA frequencies in different populations . We also demonstrate how the resulting bi-objective combinatorial optimization problem can be formulated in a concise manner and solved efficiently for relevant problem sizes with a newly developed distributed solving strategy . An experimental validation of the resulting designs confirms that the algorithm can indeed lead to significantly reduced immunogenicity .
The problem of protein de-immunization can be described as identifying amino acid substitutions that reduce immunogenicity by removing T-cell epitopes while at the same time keeping the structure and function of the protein intact . We therefore define the problem of protein de-immunization as a bi-objective optimization problem . The first objective characterizes the immunogenicity of the target protein with respect to a set of HLA alleles . The immunogenicity objective I ( S|H , PH ) combines the immunogenicity of each predicted epitope over a certain binding threshold weighted by the HLA allele frequencies PH of a specific target population represented by their prevalent HLA alleles H [35] . The second objective E ( S ) approximates the protein fitness via the statistical energy of the protein sequence , computed by the pairwise maximum entropy model inferred using a multiple sequence alignment ( MSA ) of the target protein family[24–28] . More formally , we define the protein de-immunization problem as follows: Given a protein sequence S of length n and a set Mi of possible alterations per position 1 ≤ i ≤ n . We seek a mutant S′ of S with k alterations for which S′[i] ∈ Mi ∀ 1 ≤ i ≤ n holds and that minimizes: argminS′ ( I ( S′|H , PH ) , −E ( S′ ) ) s . t . S′∈Χ , The model therefore optimizes the tradeoff between these two objectives and produces a set of Pareto-optimal designs of the protein sequence . The first objective of the de-immunization model is an adaptation of the immunogenicity score introduced by Toussaint et al . for epitope selection in the context of in silico vaccine design and is defined as follows [35]: I ( E|H , PH ) =∑e∈A∑h∈Hph∙ie , h , with A being a set of epitopes , ie , h the immunogenicity of epitope e ∈ A bound to HLA allele h ∈ H . It assumes that each epitope independently influences the immune response with respect to all considered HLA alleles . The contribution of an HLA allele h ∈ H is directly proportional to its probability ph of occurring within the target population H . The second objective is an evolutionary statistical energy of sequences computed by a pairwise maximum entropy model of protein families . Under these family-specific models , the probability for a protein sequence ( X1 , … , Xn ) of length n is defined as P ( X1 , …Xn ) =1Ze−E ( x ) E ( X ) =∑1≤i<j≤nJij ( Xi , Xj ) +∑1≤i≤nhi ( Xi ) where the pair coupling parameters Jij ( Xi , Xj ) describe evolutionary co-constraints on the amino acid configuration of residue pairs i and j for all amino acids and the parameters hi ( Xi ) corresponds to single-site amino acid constraints . The partition function Z is a global normalization factor summing over all possible amino acid sequences of length n [24 , 25] . The parameters Jij and hi are inferred from a protein family sequence alignment using an iterative approximate maximum likelihood inference scheme ( pseudo-likelihood maximization ) under l2-regularization to prevent overfitting . Given an inferred probability model for a family , the statistical energy −E ( X ) can be used to quantify the fitness of specific sequences . Recent work has demonstrated that changes of E ( X ) quantitatively correspond to the experimental phenotypic consequences of mutations , including effects on protein stability and organismal growth [31] . To maintain protein function while minimizing immunogenicity , the second objective function is defined as the minimization of −E ( X ) given inferred parameters hi and Jij from a multiple sequence alignment of the biotherapeutic . Evolutionary coupling ( EC ) strength between pairs of positions i and j is computed using the Frobenius norms of the matrices Jij with subsequent correction for finite sampling and phylogenetic effects ( average product correction ) [27] . The evolutionary couplings are predictive of residue proximity in many protein families , and the cumulative score of one position to all others ( EC enrichment score ) is indicative of functionally and structurally important positions [27] . We solve the stated de-immunization problem as a bi-objective mixed integer linear program ( BOMILP ) . Solving a BOMILP finds all Pareto-optimal solutions to linear objectives with affine constraints and additional integrality constraints on a subset of the variables . The model is based on Kingsford et al . ’s ILP formulation of the side-chain placement problem [36] . But instead of selecting energetically favorable rotamers , we encode each state of the model as a possible amino acid substitution at each position . A binary decision variable xi , a for each position i ∈ {1 . . n} and each possible variation a ∈ Mi is introduced with xi , a = 1 if this variant will be part of the final mutant S′ . An additional binary variable is introduced for each pair of variants and positions notated wi , j , a , b with wi , j , a , b = 1 if variant a at position i and variant b at position j have been selected as part of the solution S′ . These variables are associated with their inferred fitness terms hi , a and Ji , j , a , b to form the second objective function ( Table 1 O2 ) . The immunogenicity objective , in contrast to the problem formulation of Toussaint et al . , in which the immunogenicity of each candidate epitope e ∈ A could be pre-calculated , does not have an easy ILP representation . Prediction methods must be directly incorporate into the ILP to approximate the immunogenicity of the current mutant S′ . Therefore , we use TEPITOPEpan [37] , a linear HLA-epitope affinity prediction method , as internal prediction engine since it has been demonstrated to have good predictive power and can be easily integrated into the ILP framework . More advanced , potentially non-convex non-linear prediction models such as artificial neural networks cannot readily be integrated into the problem formulation as the resulting optimization problem would be discreet , non-convex , and non-linear . This class of optimization problems is known to be hard to solve to optimality even for small instances [38] and thus out of reach for design problems of relevant size . With linear ( matrix-based ) methods , the integration is possible by scoring each peptide generated with a sliding window of width en for each allele h ∈ H independently by summing over TEPITOPEpan’s position specific scoring matrix Φ ( h , a , j ) →R≥0 for amino acid a ∈ Mi+j at position i + j with i ∈ {1 . . ( n – en ) } and j ∈ {0 . . ( en − 1 ) } . To only consider predicted binding epitopes , the binding threshold τh of each HLA allele h ∈ H is subtracted from the sum score of an epitope and embedded into a hinge loss function . The summarized contribution of an allele h ∈ H is than weighted by its population probability ph . To make the prediction scores comparable across HLA alleles , the position specific scoring matrices of TEPITOPEpan were z-score normalized and the binding thresholds adjusted accordingly . The final immunogenicity score consists of the sum of all allele-wise weighted sums ( Table 1 , O1 ) . To construct a consistent model , three constraints have to be introduced guaranteeing that only one amino acid per position is selected ( Table 1 , C1 ) and that only pairwise interactions are considered for selected variants ( i . e . , Ji , j , a , b , = 1 ↔ xi , a = 1 ⋀ xj , b = 1 , see Table 1 C2 and C3 ) . Constraints C2 and C3 can be further relaxed by dividing the pairwise fitness values into positive and negative sets [36] , which is done in practice but disregarded here for ease of presentation . To be able to restrict the mutant to a specific number of introduced variations , constraint C4 limits the number of deviating amino acids to the wild type sequence W . A detailed formulation of the complete optimization problem can be found in S2 Material . As proof-of-principle , we tested the ability of the model to find low immunogenic constructs of the C2 domain of Factor VIII as the domain is highly immunogenic and involved in the ADA development in hemophilia A patients when used therapeutically [39 , 40] . Evolutionary couplings computed from sequence alignments have been used successfully to predict the phenotypic effects of mutations [29–31 , 41] , as well as the 3D structure shown in earlier work [24–28] . The approach assigns an evolutionary statistical energy to any protein sequence that is hypothesized to correspond to the fitness of the molecule . The computation of the statistical energy of the protein and any changes to it after mutation is automatic and does not depend on computing or knowing the 3D structure . Therefore , we reasoned that we could use this statistical model in a generative mode for design within the algorithmic de-immunization process . Previous work on predicting the effect of mutations suggested that the model accuracy depends on the diversity of the sequence alignment and the ability to predict the 3D structure accurately[31] . We used the precision of the total epistatic constraints between residue pairs as an approximation of the model validity . Overall , 70 long-range evolutionary coupled residue pairs ( ECs ) have a probability of at least 90% of being significant and 65 of these ( 93% precision ) are close in space ( less than 5Å; Fig 1A , Supplementary S1 Material ) in a 3D structure of Factor VIII’s C2 domain ( pdb: 3hny[42]; Fig 1B ) . To assess how well maximum entropy model can predict the effects of specific mutations compared to force-field methods , we used our maximum entropy model and FoldX predictions in a multinomial and logistic regression to predict hemophilia A severity ( severe , moderate , and mild ) based on patient data collected from the Factor VIII variant database ( http://www . factorviii-db . org , Supplementary S1 Table ) . Since the severity of HA is directly correlated with instability and malfunctioning of Factor VIII , the prediction of disease severity can be seen as a proxy for functional and structural effect prediction . The multinomial regression model , using the change in statistical energy between mutant and wild type as independent variable , shows a moderate ability to predict the clinical outcome ( F1-micro of 0 . 65 ± 0 . 09 , F1-macro of 0 . 47 ± 0 . 07 ) . The performance of a FoldX-based multinomial regression model however was significantly worse ( one-sided Wilcoxon signed rank test , V = 13558 , p-value < 2 . 2e-16; F1-micro of 0 . 49 ± 0 . 11 , F1-macro of 0 . 35 ± 0 . 09 ) . We combined the severe and moderate clinical classes and performed a logistic regression , which improved the prediction performance of our maximum entropy model ( weighted AUC of 0 . 72 ± 0 . 11 , weighted F1-score of 0 . 73 ± 0 . 11; Supplementary S1 Table; Fig 1C ) . The FoldX-based logistic regression model was again outperformed by our maximum entropy model ( one-sided Wilcoxon signed rank test , V = 15633 , p-value < 2 . 2e-16; Fig 1D ) , but also yielded higher predictive power compared to its multinomial model ( weighted AUC: 0 . 62 ± 0 . 11 , weighted F1 0 . 58 ± 0 . 13 ) . We first in silico identified a narrow region of high immunogenic potential for the three most prevalent HLA alleles in the European population ( DRB1*15:01 , DRB1*03:01 and DRB1*07:01; accounting for 70% of the patients in Western Europe ) to facility experimental evaluation . We screened the C2 domain of Factor VIII using TEPITOPEpan [37] for peptides binding to the three HLA alleles . Each peptide that fell into the 95% percentile of TEPITOPEpan’s score distribution of an HLA molecule was considered an epitope . We predicted 16 epitopes in 6 regions of the C2-domain of Factor VIII ( UniProt: FA8_HUMAN ) . The region with the highest scoring immunogenicity ( residues 2 , 312–2 , 340 ) had nine of the 16 predicted strong binding epitopes ( Fig 2A and 2B ) , making it a prime candidate region for de-immunization design . However , there is evidence that this very region might be of high functional importance for the protein; The region is enriched for conserved co-variation of residues and contains a known membrane-binding motif [42] . Eight of the top ten evolutionary couplings involve residues in the high immunogenic region ( Fig 2C ) . In general , the region is enriched for strong evolutionary couplings ( sign test , s = 124 , n = 130 , p-value < 2 . 2e-16 , CI95 = [0 . 90 , 0 . 98]; Fig 2A ) . Hence there is a risk that mutations designed to minimize immunogenicity could be detrimental to protein function and the method we have developed here is specifically designed to minimize the risk of both . We next solved our bi-objective mixed integer de-immunization model to design sequences of the identified highly immunogenic region resulting in 21 Pareto-optimal sequences with up to three simultaneous point mutations ( Table 2 , Fig 3 ) . Although the model was set up to constrain sequence substitutions solely to the identified immunogenic region , the resulting fitness change was optimized based on the interactions with all sites in that protein domain . Even though none of the 21 designed sequences were predicted “fitter” than the wild–type , they were all close to the wild-type fitness . The computed fitness of 20 out of 21 designs resided in 95% percentile or higher when compared to the whole distribution of single , double , and triple mutations , suggesting that the protein would remain stable and functional . The sequence with the highest difference to wild-type fitness prediction ( Design-11; V2313M , Y2324L , V2333E ) was in the 90% percentile and still close to WT fitness ( reduction of 1 . 7% ) . It exhibited also the maximal predicted reduction of immunogenicity ( immunogenicity reduction of 45% ) deleting eight out of nine epitopes of the identified region . The next-best triple mutant ( Design-12; L2321T , I2327L , V2333E ) resulted in the deletion of eight epitopes with an immunogenicity reduction of 42% and a fitness reduction of 1 . 28% . Previous work that aims to increase the likelihood of a functional protein after mutation design , has used force-field based modeling , such as FoldX [32] . As to distinguish the differences between FoldX and the employed maximum entropy model we predicted the mutation effects of the 21 designs with the EV model and FoldX ( Table 2 ) . The predictions of the maximum entropy model only moderately correlate to those using FoldX ( r = 0 . 44 , CI95 = [0 . 02 , 0 . 73] , t = 2 . 173 , df = 20 , p-value = 4 . 2e-2; Fig 4 ) . The two most deviating mutations between the two prediction methods were Design-11 ( V2313M , Y2324L , V2333E ) and Design-3 ( Y2324L , V2333E ) , both of which introduced a mutation at a membrane-binding site[42] . FoldX predicted these designs comparatively less deleterious than the predictions of the maximum entropy model . One explanation for this discrepancy is that it would be harder for force-field based methods to capture the membrane binding constraints unless they were in the structure used . To test the designs , we synthesized twenty overlapping 15-mer peptides containing the introduced mutations and their wild-type counterparts . The peptides maximally covered the predicted epitopes around the mutations of all designed constructs that contained one and two mutations ( Supplementary S2 Table ) . The affinity of these peptides to the three HLA alleles was measured at time zero and after 24 hours ( Methods , Supplementary S3 and S4 Tables ) . We linearly combined the measured relative affinity scores across HLA alleles weighted by their allele frequencies for each peptide respectively to produce a score that approximates the overall immunogenicity ( Fig 5A ) . Results are presented for measurements taken at time point zero . Measurements made after 24 hours were very similar ( r = 0 . 94 , CI95 = [0 . 86 , 0 . 98] , t = 11 . 81 , df = 17 , p-value = 1 . 3e-09 ) and thus resulted in similar correlations . Overall , the measured and predicted immunogenicity of the tested peptides correlated well with r = 0 . 76 ( CI95 = [0 . 48 , 0 . 91] , t = 4 . 885 , df = 17 , p-value = 1 . 4e-4; Fig 5A ) . Next , we compared the predicted and measured gain or loss in immunogenicity for the whole region by reconstructing the targeted region using overlapping peptides . The measured relative scores were linearly combined ( as previously ) and then normalized to the number of overlapping peptides used in the reconstruction ( Fig 5B ) . The difference in the measured scores between the wild type and mutant regions can be thought of as a proxy for gain or loss in immunogenicity and correlated well to the predicted changes ( r = 0 . 86 , CI95 = [0 . 40 , 0 . 97] , t = 4 . 136 , df = 6 , p-value = 6 . 1e-3; Fig 5B ) .
This work introduced a novel method to reduce a protein’s immunogenicity while maintaining its structural integrity requiring only sequence information of the target protein . The method uses a different immunogenicity objective compared to all previous approaches , accounting for both relative epitope strength and HLA allele frequency information of a target population . The HLA distribution can differ tremendously between populations influencing the immunogenicity of a protein and hence the design process should account for the difference by prioritizing different T-cell epitopes . We further combined these objective functions in a bi-objective mixed-integer linear program and introduced a novel solving strategy that guarantees to find the full and exact Pareto front of our de-immunization model . While guaranteeing global optimality , an integer linear program imposes constraints on the functional form the immunogenicity and protein fitness objectives can take . Only linear or simple convex functions can be integrated into an integer linear program , thus prohibiting the use of non-linear , non-convex prediction models . However , integrating such complex , non-convex methods would lead to a highly complex optimization problem that is effectively impossible to solve to optimality for design problems of relevant size . The fact that the highly immunogenic region , a priori identified during an in silico screening and independently described by others [19] , coincides with a highly evolutionary connected as well as functionally important region underlines the need for methods that are capable of incorporating functional and structural integrity prediction in the de-immunization process . The de-immunization model introduced demonstrates the power of such approaches . The predicted immunogenicity of the complete domain could be reduced by 45% without disrupting the fitness landscape extensively . Moreover , the observed highly significant correlations between measured and predicted immunogenicity both on individual peptide and ( reconstructed ) segment level affirmed that the underlying assumptions made by the model are sufficient to predict the influence of mutation in terms of immunogenicity . In the case of this Factor VIII domain , we found no advantage to structure and force-field base approaches to assessing the effect of clinical mutation classification; structure-based approaches may even be a disadvantage when structure information is incomplete ( e . g . , binding partners not present ) . This suggests that sequence information may be sufficient for de-immunization design , and is consistent with the previous observation that sequence alignments can be used to identify constrained interacting residues across biomolecules as well as the effect of mutations [31 , 43] . However , high-quality diverse sequence alignments are not always available , especially for chimeric or synthetic proteins . In summary , we proposed a novel de-immunization model that integrates quantitative immunogenicity optimization with sequence-based fitness optimization and used the approach to design novel C2 domains of Factor VIII that can be further validated for clinical application using mouse models or T-cell proliferation assays based on PBMCs of HA patients . The approach will allow bioengineers to reliably explore the design space of the target protein to select promising candidates for experimental evaluation .
To reduce the search space , a filtering approach based on position specific amino acid frequency fi ( a ) ( i . e . , conservation ) can be applied . Only amino acids at position i ∈ {1 . . n} exceeding a certain frequency threshold ζ are considered as possible substitution at a site . Hence , the set of possible substitutions per position is defined as Mi: = {a ∈ Σ|fi ( a ) ≥ ζ} . The wild type amino acid is additionally added if it does not exceed the frequency threshold . This filtering assumes that variants that are not or infrequently observed are harmful due to either destabilizing effects , reduction of function , or intervening effects with interaction partners . Special strategies must be applied to solve a BOMILP . Popular methods to solve discrete multi-objective problems include the ε-constraint [44] , perpendicular search [45] , and the augmented weighted Chebychev method [46] . All have their own limitations . The recently published rectangle splitting approach tries to overcome these [47 , 48] . For solving the de-immunization problem , we developed a parallel two-phase version of the rectangle-splitting approach that can exploit the parallel nature of the algorithm and can effectively utilize modern distributed computing resources . In the following we sketch the newly developed two-phase approach . First , we introduce necessary notations and concepts ( adopted from Boland et . al . ) . Let z1= ( z11 , z21 ) and z2= ( z12 , z22 ) be two points in solution space with z11≤z12 and z22≤z21 . Further we define R ( z1 , z2 ) to be the rectangle spanned by z1 and z2 . A nondominated point within R ( z1 , z2 ) can be found with the following sequential operation ( see proof in[47] ) : ( 1 ) z1¯=minx∈χz1 ( x ) s . t:z ( x ) ∈R ( z1 , z2 ) ( 2 ) z2¯=minx∈χz2 ( x ) s . t:z ( x ) ∈R ( z1 , z2 ) andz1≤z1¯ These operations will be denoted as z˜=lexminx∈X{z1 ( x ) , z2 ( x ) :z ( x ) ∈R ( z1 , z2 ) } . As a first step of the two-phase parallel rectangle-splitting approach the boundaries of the Pareto front are calculated by solving zT=lexminx∈X{z1 ( x ) , z2 ( x ) :z ( x ) ∈R ( ( −∞ , ∞ ) , ( −∞ , ∞ ) ) } and zB=lexminx∈X{z2 ( x ) , z1 ( x ) :z ( x ) ∈R ( ( −∞ , ∞ ) , ( −∞ , ∞ ) ) } in parallel ( Fig 6A ) . Then , the search space within R ( zT , zB ) is evenly constraint based on boundary conditions enforced w . l . o . g . on z1 ( Fig 6B ) . The boundaries are calculated for a predefined number of constraints m with: τiz1=z1T+i∙ ( z1B−z1T ) mwith1≤i≤m . Each section of the separated search space can be independently searched by solving zi=lexminx∈X{z1 ( x ) , z2 ( x ) :z ( x ) ∈R ( ( τiz1 , z2T ) , zB ) } and the resulting new nondominated points can be used as initial approximation of the Pareto front . The found nondominated points might contain duplicates and might not resemble the complete Pareto front . Therefore , it is necessary to perform a refinement of the Pareto front to find the remaining nondominated points . To this end , the nondominated points are sorted in nondecreasing order such that z11≤z12≤⋯≤z1k . Each consecutive pair of points spans a search rectangle R ( zi , zj ) with i ≤ j . These rectangles can now be searched in parallel by the rectangle-splitting algorithm ( Fig 6C ) . The search rectangles are split in half . First , the bottom half RB is searched by solving: z¯1=lexminx∈X{z1 ( x ) , z2 ( x ) :z ( x ) ∈R ( ( z1i , z2i+z2j2 ) , zj ) } If a nondominated point is found , the upper half RT is further restricted and spans now R ( zi , ( z¯1i‑ϵ , z2i+z2j2 ) ) in which z¯2=lexminx∈X{z2 ( x ) , z1 ( x ) :z ( x ) ∈R ( zi , ( z¯1i−ϵ , z2i+z2j2 ) ) } is searched . Each newly found point spans a new independent search rectangle R ( zi , z¯2 ) and R ( z¯1 , zj ) with its adjacent point . These rectangles are searched in parallel with the described procedure ( Fig 6D ) . If the search operation yielded the known point zj for R ( ( z1i , z2i+z2j2 ) , zj ) and zi for R ( zi , ( z¯1i‑ϵ , z2i+z2j2 ) ) accordingly it proofs that the area does not contain further nondominated points . The search procedure is carried out until the complete search space has been explored . Multiple sequence alignments ( MSA ) , created by JackHMMER[49] , were used for the inference of the maximum entropy models of the Factor VIII C2-domain ( residues 2 , 188–2 , 345 of FA8_HUMAN , Supplementary S1 Material ) . To optimize residue coverage and MSA diversity , the alignment was created using five search iterations at an E-value threshold of 10−20 . Sequences with 70% or more gaps and columns with over 50% gaps were excluded from subsequent statistical inference . To reduce the influence of sampling bias in the inference step , sequences were clustered at a 90% identity threshold ( theta 0 . 9 ) , and reweighted by the inverse of the number of cluster members resulting in Meff = 1656 effective sequences . Generally , at least a Meff/L ≥ 1 is considered necessary to predict the tertiary structure of a protein [31 , 50] . Here we achieve a Meff/L ≥ 10 ( with L = 157 aa ) , which should be sufficient to guarantee a high-quality model . The parameters of the pairwise maximum entropy model and evolutionary couplings were then inferred using EVfoldPLM [51] with pseudo-likelihood maximization [52] . Substitution effects were derived by calculating the difference between the wild-type and the mutant statistical energy [31] . The validity of the maximum entropy model was verified by using the precision of the inferred top evolutionary couplings ( ECs ) between residue pairs compared to an existing 3D structure . To identify the top ECs a Gaussian-lognormal mixture model was inferred based on the overall score distribution [28] and the ECs within the tail of the distribution ( ECs with a probability ≥ 0 . 90 of belonging to the lognormal ) were used for model quality assessment [28] . Single point mutation data with known patient severity status was extracted from the Factor VIII variant database ( http://www . factorviii-db . org ) . The data was filtered for mutations residing within the C2 domain , which resulted in 40 data points in total ( Supplementary S4 Table ) . The severity status of each patient was determined based on a one-stage Factor VIII:C and categorized into three classes–severe ( <1% ) , moderate ( 1–5% ) , and mild ( >5% ) . The data points were unevenly distributed across the classes with 15 severe , 8 moderate , and 17 mild cases . To train and validate the multinomial and logistic regression models , the data was randomly divided into training and test set ( 70:30%-split ) in a stratified manner . This process was repeated two hundred times and the prediction performance averaged over the runs . In order to experimentally verify our in silico predictions for Factor VIII , we utilized the commercial REVEAL HLA-Peptide binding assay of ProImmune ( www . proimmune . com ) . Peptides were synthesized using the PEPscreen custom library synthesis method , yielding high purity peptides for experimental analysis . HLA-peptide binding was assessed for the three HLA-DRB1 alleles used in this study ( DRB1*15:01 , DRB1*03:01 and DRB1*07:01 ) . In short , the method compares the affinity of the studied peptide to the affinity of a high-affinity control peptide . Each peptide is then scored for binding to a certain HLA molecule relative to the score of the control peptide and reported as the percentage of the signal generated by the control peptide . The two-phase rectangle-splitting solver was implemented in Python 2 . 7 using the CPLEX package , Numpy 1 . 4 , and Polygon 2 . 0 . 6 package . CPLEX 12 . 6 was used as backend to solve the BOMILP models . Structure-based fitness prediction for validation purposes of the de-immunized Factor VIII C2 domain constructs were performed with FoldX [32] using default settings for the obtained mutations . TEPITOPEpan 1 . 0 was used for epitope prediction . The multinomial and logistic models were fit and evaluated using Scikit-learn 0 . 18 [53] . The statistical analysis was conducted with R 3 . 0 . 2 . Statistical significance was considered at α = 0 . 05 . The specific statistical tests used are indicated in the figures or in the results section . | Therapeutic proteins have become an important area of pharmaceutical research and have been successfully applied to treat many diseases in the last decades . However , biotherapeutics suffer from the formation of anti-drug antibodies , which can reduce the efficacy of the drug or even result in severe adverse effects . A main contributor to the antibody formation is a T-cell mediated immune reaction caused by presentation of small immunogenic peptides derived from the biotherapeutic . Targeting these peptides via sequence alterations reduces the immunogenicity of the biotherapeutic but inevitably will have effects on structure and function . Experimentally determining optimal mutations is not feasible due to the sheer number of possible sequence alterations . Therefore , computational approaches are needed that can effectively cover the complete search space . Here , we present a computational method that finds provable optimal designs that simultaneously optimize immunogenicity and structural integrity of the biotherapeutic . It relies solely on sequence information by utilizing recent advances in protein ab initio prediction and incorporates immunogenicity prediction methods . Thus , the approach presents a valuable tool for bioengineers to explore the design space to find viable candidate designs that can be experimentally tested and further refined . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"engineering",
"and",
"technology",
"synthetic",
"biology",
"immunology",
"mutation",
"synthetic",
"biotherapeutics",
"mathematics",
"forecasting",
"statistics",
"(mathematics)",
"protein",
"structure",
"prediction",
"protein",
"structure",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"immune",
"system",
"proteins",
"sequence",
"alignment",
"bioinformatics",
"proteins",
"antigens",
"mathematical",
"and",
"statistical",
"techniques",
"evolutionary",
"immunology",
"molecular",
"biology",
"biochemistry",
"point",
"mutation",
"physiology",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"evolutionary",
"biology",
"statistical",
"methods",
"macromolecular",
"structure",
"analysis"
]
| 2018 | Population-specific design of de-immunized protein biotherapeutics |
Reperfusion injury results from pathologies of cardiac myocyte physiology that develop when previously ischemic myocardium experiences a restoration of normal perfusion . Events in the development of reperfusion injury begin with the restoration of a proton gradient upon reperfusion , which then allows the sodium-proton exchanger ( NHE ) to increase flux , removing protons from the intracellular space while importing sodium . The resulting sodium overload drives increased reverse-mode sodium-calcium exchanger ( NCX ) activity , creating a secondary calcium overload that has pathologic consequences . One of the attempts to reduce reperfusion-related damage , NHE inhibition , has shown little clinical benefit , and only when NHE inhibitors are given prior to reperfusion . In an effort to further understand why NHE inhibitors have been largely unsuccessful , we employed a new mathematical cardiomyocyte model that we developed for the study of ischemia and reperfusion . Using this model , we simulated 20 minutes of ischemia and 10 minutes of reperfusion , while also simulating NHE inhibition by reducing NHE flux in our model by varying amounts and at different time points . In our simulations , when NHE inhibition is applied at the onset of reperfusion , increasing the degree of inhibition increases the peak sodium and calcium concentrations , as well as reducing intracellular pH recovery . When inhibition was instituted at earlier time points , some modest improvements were seen , largely due to reduced sodium concentrations prior to reperfusion . Analysis of all sodium flux pathways suggests that the sodium-potassium pump ( NaK ) plays the largest role in exacerbated sodium overload during reperfusion , and that reduced NaK flux is largely the result of impaired pH recovery . While NHE inhibition does indeed reduce sodium influx through that exchanger , the resulting prolongation of intracellular acidosis paradoxically increases sodium overload , largely mediated by impaired NaK function .
Ischemia-reperfusion ( IR ) injury represents a constellation of pathological events that occur when previously ischemic myocardium experiences a restoration of normal tissue perfusion . IR injury , which can manifest as dangerous arrhythmias such as ventricular tachycardias and fibrillation , reduced myocardial force development , or an increased region of cell death , is likely to become even more clinically relevant in coming years owing to an aging population and the impact of aging on susceptibility to ischemia/reperfusion injury [1] . As such , it is desirable to develop an ability to effectively treat and prevent such phenomena . Because of the danger that ischemia-reperfusion related events pose , there has been great interest in this problem for several decades . A large number of studies , directed at furthering the understanding of ischemia-reperfusion injury and examining many potential therapeutic targets , have been undertaken [2]–[4] . As a result of these studies , significant insight into the mechanisms of IR injury has been obtained . Figure 1 illustrates a chain of events that are believed to play a prominent role in ischemia-reperfusion injury [3]–[6]: While much has been determined about the pathogenesis of reperfusion injury , we argue that a comprehensive understanding remains elusive . The most salient point is that despite numerous studies investigating many potential therapeutic agents to prevent reperfusion injury , efforts at translation of potential therapies to clinical practice have been largely unsuccessful [2] , [3] , [16] . One approach to treating ischemia-reperfusion injury has been inhibition of the NHE [2] , [3] , which is attractive for several reasons . The NHE is potentially responsible for a large amount of sodium permeability in cardiomyocytes [17]; the rationale is that if NHE flux is reduced during reperfusion , sodium overload can be mitigated and hence the secondary calcium overload can be reduced as well . In addition to reducing sodium and calcium-mediated reperfusion injury , NHE inhibition has been thought to play another beneficial role . It is believed that decreased ATP availability leads to activation of proteases and phospholipases that result in damage to the cell membrane , but that these enzymes have reduced activity under the relatively acidotic environment present during ischemia . Thus , when pH recovers during reperfusion , these enzymes increase in activity and cause damage . Therefore , NHE inhibition may reduce injury caused by proteases and phospholipases [17] . Despite some success in preclinical studies [4] , NHE inhibition has had little beneficial effect in the clinical setting – when any beneficial effect is observed , it is when NHE inhibitors are administered prior to the onset of reperfusion [3] , [4] . While there are several possible reasons for the failure to develop an effective treatment for reperfusion injury thus far [3] , the most fundamental may be that cellular systems , not to mention myocardial tissue consisting of a large number of coupled cells , are highly coupled systems containing a large number of components , many exhibiting nonlinear relationships between them . Such systems are frequently robust to perturbations , often behave non-intuitively , and likely require observation and/or perturbation of more than just a couple of components in order to gain an adequate understanding of processes of interest . Because of the aforementioned complexities , we aimed to develop a mathematical model that would enable us and others to gain insight into events that occur during myocardial ischemia and reperfusion , and to make predictions about possible therapeutic strategies . While models of ischemia already exist [6] , [8] , [18] , none capture all of the relevant processes discussed above . In addition , these models were designed to address ischemia only , not reperfusion . Therefore , based upon experimental data from the literature , we have developed an improved mathematical model of the cardiomyocyte , and an accompanying protocol , that allows us to more realistically simulate myocardial ischemia and reperfusion . We then used this model to examine NHE inhibition as a therapeutic strategy . In our simulations , we observed that NHE inhibition produced no significant improvement in terms of sodium and calcium overload during reperfusion relative to control , and in some cases the amount of sodium and calcium overload was worse than control . In addition , we observed that NHE inhibition inhibited the recovery from intracellular acidosis and that the effects of prolonged acidosis on some subcellular components played a role in the failure to attenuate sodium overload .
This model ( Figure 2 ) predominantly represents guinea pig physiology . Representative action potentials are shown in Figure 3 . The starting point for this work was a model previously published by Crampin and Smith ( CS model ) [8] , which is a charge-difference implementation of the Luo-Rudy dynamic ( LRd ) guinea pig ventricular myocyte model [19] where membrane voltage is calculated based upon changes in ion concentrations rather than the transmembrane currents [20] , [21] . Modifications to the LRd designed to simulate the effects of respiratory acidosis ( Figure 2 ( black and green ) ) were introduced in the CS model , including the flux of species that are important for pH regulation across the plasma membrane , as well as dynamic intracellular pH that is subject to both bicarbonate and “intrinsic” buffering . However , extracellular pH changes in the CS model depend only on prescribed extracellular carbon dioxide ( ) and static extracellular bicarbonate ( ) concentrations . In our model , both and concentrations are dynamic during ischemia and reperfusion , as discussed below . The formulation of the NHE introduced in the CS model includes allosteric regulation whereby the exchanger is stimulated by the binding of intracellular protons . There is also the possibility [22] of additional allosteric regulation whereby the binding of extracellular protons inhibits the NHE , though whether this is truly the case is unclear . We have implemented this additional regulation in our model , as it improves model behavior in terms of extracellular acidosis that develops during ischemia . Total allosteric regulation of the NHE is calculated with the following equation: ( 1 ) The Hill coefficient for allosteric regulation by extracellular protons , , was reported to be about 1 , which is the value that we use . Introducing extracellular allosteric increases the minimum extracellular pH during ischemia , resulting in a better fit to the data shown in Figure 4B , but also results in a worse fit to what is seen in terms of intracellular sodium concentration during ischemia . As depicted in Figure 4C , intracellular sodium accumulates during ischemia , and in late ischemia the rate of sodium accumulation may decrease . Introduction of extracellular allosteric NHE regulation creates a situation in which intracellular sodium concentration starts to decrease in late ischemia , approaching the pre-ischemia value . Increasing the value of raises the minimum extracellular pH that is reached after 20 minutes of simulated ischemia , but also further decreases the intracellular sodium concentration . Therefore , we have chosen a value for that is near what was reported experimentally , allows for intracellular sodium overload during ischemia , and results in a degree of extracellular acidosis that is relatively close to what has been observed experimentally . It should be noted that here we use the same value for the binding constant of protons , , for both intracellular and extracellular regulation . The value used here is the same as that used in the CS model , and near what can be extracted for intracellular allosteric regulation from the data in [23] . When we extracted the binding constant for extracellular regulation from the same data and used this value for extracellular regulation , we saw a much worse fit in terms of the intracellular sodium profile . As mentioned above , intracellular sodium accumulation falls off during late ischemia , and in a previous version of the model , sodium levels would approach or fall below pre-ischemic values . This is largely due to inhibition of the sodium-calcium exchanger , which is the primary mover of sodium into the cell . The sodium-potassium exchanger , which removes sodium , is also strongly inhibited during late ischemia , but there is still a net loss of sodium . In order to improve upon this , creating the sustained elevation of sodium above pre-ischemic levels throughout ischemia that is observed experimentally , we modified the pH dependence of the sodium-calcium exchanger . Specifically , the binding constant and Hill coefficient for proton binding to the sodium-calcium exchanger were changed to 7 . 0 and 0 . 75 , respectively , from their published values of 7 . 37 and 0 . 99 [8] . The model , as originally implemented , exhibited high concentrations of intracellular sodium at steady state under pre-ischemic conditions . In addition , the relationship between steady state sodium concentration and pacing frequency was steeper than both experimental measurements and simulated experiments using the LRd model . In order to bring sodium homeostasis in our model in line with the LRd and closer to what has been seen experimentally , we decreased the scaling factor for the sodium-calcium exchanger , c1 , by 40 percent from the value published in [8] . The current sodium-pacing frequency relationship is provided in Figure S4 in Text S1 . Decreasing baseline sodium-calcium exchange flux reduces steady state sodium load , but also increases intracellular calcium . In order to compensate for this , it was necessary to increase maximum calcium pump current , , from 1 . 15 [8] to 1 . 65 uA/uF . In order to simulate the effects and proximal causes of extracellular potassium ( ) accumulation during ischemia , we implemented components that were introduced in a model that extended the LRd model with the addition and/or modification of components to study extracellular potassium accumulation during ischemia [6] . We will henceforth refer to this model as the Terkildsen-Crampin-Smith ( TCS ) model . Note that the TCS model is not an extension of the CS model , as it does not include a pH regulation system ( i . e . equilibration with carbon dioxide , bicarbonate and intrinsic buffering ) , acid transporters , or the inhibitory effects of acidosis on various components of the calcium handling system . Among other additions , our new model incorporates all CS and TCS components , save for those relating to simulating the development of tension in the CS model . The TCS additions ( Figure 2 ( red ) ) allow the sodium-potassium exchange pump to behave in a more realistic fashion , as it is sensitive to changes in the concentration of sodium and potassium on both sides of the plasma membrane , as well as phosphometabolites and . The TCS model also incorporates and dynamic intra- and extracellular volumes coupled to water flux . These potassium-relevant processes have been incorporated into our model , with changes . First , we have modified the function for calculating the amount of water flux to include concentrations of intracellular and extracellular chloride ( [] and [] , respectively ) : ( 2 ) where Lp is the hydraulic conductivity of the membrane , R is the universal gas constant , temp is the temperature , [] , [] , [] , [] are the intracellular concentrations of sodium , potassium , calcium and chloride , respectively , [] , [] , [] , [] are the extracellular concentrations of sodium , potassium , calcium and chloride , respectively , [] and [] are the concentrations of non-permeable osmolytes in the intra- and extracellular compartments , respectively , and and are the volumes of the myoplasm and extracellular compartment , respectively . [] is used when initializing [] , and [] is used when initializing and updating [] . Second , the equation that prescribes during ischemia has been slightly modified to match a slightly different value for at the transition from simulated pre-ischemia to ischemia . During ischemia: ( 3 ) where t is the ischemic time in minutes . It should be noted that the first term of the ischemia equation may need to be modified if the model is used with different initial conditions . As the dynamic nature of extracellular acid status is believed to play a prominent role in reperfusion arrhythmogenesis , it is highly desirable to employ a mathematical model that includes a dynamic extracellular pH system that is as realistic as possible . To this end , we have made several significant improvements to the representation of extracellular pH over preexisting models ( Figure 2 ( blue ) ) . First , we have introduced intrinsic buffering in the extracellular compartment . Second , as mentioned previously , [] is dynamic during ischemia and reperfusion . Third , flux through NHE and CHE affects not only , but also . Thus , during simulated ischemia is influenced by a combination of equilibration with dynamic and concentrations , as well as an intrinsic buffering system , and flux of protons and hydroxide ions across the plasma membrane . The change in extracellular pH at each time step is: ( 4 ) where ( 5 ) is the intrinsic buffering power in the extracellular compartment and ( 6 ) represents equilibration between protons , bicarbonate and carbon dioxide in the extracellular compartment . Eqs . 4–6 are nearly the same as their intracellular counterparts , as provided in [8] save for the reversal of signs for and and the substitution of extracellular for intracellular species concentrations and volumes . Concentrations and pK values of the intrinsic buffers are the same as in the intracellular compartment . Equations for and are provided in Text S1 and [8] . Because the concentrations of protons are explicitly represented in our model , and because part of the ischemia simulation protocol is to impose a progressive intracellular acidosis on the system during ischemia , it is necessary to account for conjugate anions that are produced as the metabolic acidosis progresses in order to maintain charge conservation . This is accomplished by the introduction of a variable representing the concentration of generic monovalent anions produced in a 1∶1 stoichiometry with the surplus protons that appear as decreases . ( 7 ) where ( 8 ) and ( 9 ) are the concentrations of protons bound to the two generic intrinsic buffers , and , and ( 10 ) and ( 11 ) are the concentrations of free intracellular protons and protons bound to the intrinsic buffers , respectively , at the beginning of a simulation , such that [] is initially equal to zero . Eqs . 8 and 9 are from [8] . Experiments have shown that the availability of L-type calcium channels ( ) is modified by the concentration of ATP [24] , [25] . Accordingly , we have included a term to represent this change in channel availability ( Figure 2 ( orange ) ) [26]: ( 12 ) where [ATP] is the concentration of ATP and is the for the binding of ATP to channels . Finally , we implemented a newer version , relative to that used in the CS model , of the SERCA pump ( Figure 2 ( purple ) ) from [13] . The cycling rate of SERCA , , ( see Text S1 or [13] ) is translated to calcium flux via the following equation: ( 13 ) The conversion factor of 0 . 00820 was derived by first implementing SERCA such that it was unidirectionally coupled ( i . e . could respond to changes in , [] , [ATP] , etc . , but could not affect the remainder of the system ) and simulating normal pre-ischemic conditions . Flux through the CS model SERCA and were recorded at the peak of the intracellular calcium transient at pre-ischemic . These values were used in our model to scale the raw value to produce output that results in appropriate transients . Throughout all simulations , the cell was constantly stimulated at a rate of 3 Hz using pulses of −80 . 0 uA/uF and lasting 0 . 5 ms . Starting from the initial conditions provided in Table 1 , the cell was paced for five minutes of simulated time , well past the required time for the model to reach steady state , before the onset of simulated ischemia . During pre-ischemia , the concentrations of all extracellular species are held constant , the assumption being that the extracellular compartment has access to , and is in equilibrium with , a pool that is many orders of magnitude larger than the volume of the immediate extracellular space . , on the other hand , is dynamic and is updated per Eq . 29 below , although it fluctuates very little under these conditions . In this study we have simulated an abrupt-onset total ischemia . During ischemia , as oxygen availability falls and the cell switches to anaerobic metabolism , metabolic acidosis develops . The metabolic byproducts responsible for this acidosis are not represented in our model , and so we impose metabolic acidosis on the intracellular compartment via Eq . 3 . Also , during ischemia the extracellular compartment becomes more acidic due to flux through the acid exchangers and the accumulation of carbon dioxide . In our model the extracellular compartment is considered to be completely isolated , such that gases and ions accumulate or deplete via flux through ion channels and transporters . While we must impose pH changes on the intracellular compartment during simulated ischemia , pH changes in the extracellular compartment are allowed to respond to acid exchanger flux and carbon dioxide accumulation , and are calculated per Eq . 4 . Accumulation of carbon dioxide in the extracellular space is modeled by: ( 14 ) where is transport of across the plasma membrane and represents the concentration of produced or consumed from equilibration with in the extracellular compartment ( Eq . S . 167 in Text S1 ) . [] , [] , [] , and [] , which are no longer fixed as these species are accumulating or being depleted from the extracellular compartment while it is isolated during simulated ischemia , are updated per the following equations: ( 15 ) ( 16 ) ( 17 ) and ( 18 ) where , , and represent the total current through relevant ion channels , the NaK exchanger and NCX exchanger for , , and , respectively . Eqs . 15–18 are derived from the equations for updating intracellular species concentrations in [8] . During ischemia , sodium channel conductance , , is decreased from 16 . 0 to 14 . 4 mS/uF . In addition , late sodium current , defined by [27]: ( 19 ) is increased by increasing the value of from 0 . 0007 to 0 . 00018 . Ischemia results in decreased availability of ATP and phosphocreatine . We impose decreasing concentrations of ATP and PCr on the system during simulated ischemia , which then affect concentrations of the other phosphometabolites , per the following equations , from [6]: ( 20 ) ( 21 ) ( 22 ) ( 23 ) ( 24 ) and ( 25 ) where t is the ischemic time in minutes and [] = 1000* . Note that Eq . 20 has been modified from [6] . Finally , as described in [6] , a variable representing intracellular osmolarity is linearly increased during ischemia: ( 26 ) where t is the ischemic time in minutes . is subsequently used to update [] at each time step: ( 27 ) for use in calculating water flux ( see Text S1 ) . As discussed above , it is believed that acidic extracellular fluid is washed out during reperfusion and replaced with fluid from elsewhere that is at normal pH , carrying away accumulated protons and carbon dioxide . Also , oxygen concentration increases , reducing the cell's dependence on anaerobic metabolism and abolishing the source of metabolic acidosis . Therefore , as opposed to simulated ischemia , during which is prescribed and evolves according to Eq . 4 , during simulated reperfusion is prescribed according to Eq . 28 in order to simulate washout and is allowed to evolve according to Eq . 29 below . is prescribed according to the following formula: ( 28 ) where 7 . 40 is the pre-ischemic and t is the length of the time step . When using a time step of 0 . 005 ms , the time constant of 1 . 5× ms produces a recovery to pre-ischemic in about 10 minutes . We have attempted to choose a recovery rate that is consistent with what has been observed experimentally , but data regarding the recovery profile during reperfusion is scarce . In one porcine experiment [28] , pre-ischemic was reached after about 10 minutes , following 10 minutes of ischemia . In another experiment performed on canines , recovery took 20–30 minutes following 90 minutes of ischemia [29] . We have chosen a recovery time of approximately 10 minutes because it produces a better fit to experimental data for other parameters , such as and [] . The change in at each time step is calculated according to the following formula: ( 29 ) where ( 30 ) is the intrinsic buffering power in the intracellular compartment , and represent flux through the NHE and CHE , respectively , and represent the volumes of the myoplasm and sarcoplasmic reticulum , respectively , and ( 31 ) represents equilibration between protons , bicarbonate and carbon dioxide in the intracellular compartment . Eqs . 29–31 are from [8] . Equations for and are provided in Text S1 and [8] . It is assumed that other extracellular species ( , , , , , and ) will return to their normal concentrations during reperfusion , now that they are again in equilibrium with a vastly larger pool . We have chosen a time constant of 3 . 75× ms , such that the concentration of a particular ion or molecule is updated according to the following equation: ( 32 ) This produces a recovery to preischemic concentrations in approximately two minutes , consistent with what has been observed during reperfusion in Langendorff-perfused tissue [5] . The idea here is to simulate the wash out ( or wash in , as the case may be ) of extracellular species as the fluid in the previously isolated extracellular compartment is replaced with fluid replenished by blood arriving from outside the ischemic region . Based upon experimental evidence , it appears that ATP and PCr can recover to as much as 40 and 75 percent of their preischemic concentrations following ischemia in Langendorff-perfused guinea pig hearts [30] . It also appears that most recovery happens within the first few minutes of reperfusion . Therefore , [ATP] and [PCr] are calculated according to the following equations during simulated ischemia: ( 33 ) ( 34 ) The concentrations of all other phosphometabolites are calculated as they were during ischemia . Finally , intracellular osmolarity is returned to normal according to the following equation: ( 35 ) Flux through the NHE is computed using the following equation: ( 36 ) ( see Text S1 for preceding equations ) . In order to simulate the effects of agents that inhibit the NHE , we reduced the value of from its default value of 1 . 0 to one of two values: 0 . 50 or 0 . 0 . In addition , we introduced each of these degrees of NHE inhibition at one of three time points: the end of 20 minutes of ischemia ( denoted reperfusion in figures ) , after 10 minutes of ischemia ( denoted mid-ischemia in figures ) , or at the onset of ischemia ( denoted ischemia in figures ) . Differential equations were solved using the forward Euler method . Model code is available as a supplemental download .
Figure 3A illustrates action potentials before ( black ) and at the end ( gray ) of ischemia , as well as following 10 minutes of reperfusion ( red ) , from the control simulation . Note that after ten minutes of reperfusion , the action potential duration and amplitude are still smaller than before the onset of ischemia , consistent with what has been observed in guinea pig experiments [31] . Figure 3B illustrates responses to four stimulus events during late ischemia . Note that the resting membrane potential is depolarized to a degree similar to that which has been observed in guinea pig experiments [5] . Also , cell excitability is extremely impaired , with only every other stimulus eliciting an action potential . Figure 4 provides a comparison of experimental data to results obtained when we simulated 20 minutes of ischemia ( grey shading ) followed by 10 minutes of reperfusion . The simulation results in all panels of Figure 4 ( solid lines ) are from the control simulation , in which no NHE inhibition was instituted . Comparing simulation results to experimental data from [30] in Figure 4A , it can be seen that our model closely reproduces the behavior of intracellular pH observed in guinea pig myocytes during ischemia and reperfusion . As discussed above , during simulated ischemia evolves according to a fixed equation ( Eq . 3 ) implemented to reproduce what occurs experimentally . However , during reperfusion , which is calculated by Eq . 29 , is not pre-determined . In Figure 4B , results from the same control simulation are compared to experimentally measured during 20 minutes of guinea pig ischemia [32] . Our model and simulation protocol produce a profile that closely reproduces what is observed during early ischemia , but continues with a steep decline for a longer period of time , producing an extracellular pH after 20 minutes of ischemia that is about 0 . 2 units lower than what was seen in the guinea pig heart . Figure 4C compares results from the same control simulation to [] measured in guinea pig hearts during ischemia and reperfusion [33] . In both cases , there is an accumulation of sodium inside the cell during ischemia , although our model shows a more modest increase . This is largely due to the higher starting concentration , which is determined from a steady state value that is heavily dependent upon NaK pump density and the amount of allosteric regulation by extracellular protons at the NHE . Also , both the experimental and simulated results show a second , more profound , sodium overload upon reperfusion . It should be noted that in the experiment cited here , reperfusion was preceded by 30 minutes of ischemia , not 20 , as is the case with the other experimental comparisons . Figure 5 shows intracellular sodium , calcium , and pH profiles from a subset of simulations . The black lines representing the control simulations in this and subsequent figures are the same as shown in Figure 4 . In addition to the control simulation , results from the following three simulations are shown: either a 50 or 100 percent reduction in NHE flux beginning at the onset of reperfusion , and a 100 percent reduction at the onset of ischemia ( denoted as R50 , R0 , and I0 in tables ) . In Figure 5A , the intracellular sodium concentration throughout each of the four aforementioned simulations can be seen . Table 2 provides peak and mean intracellular sodium concentrations during reperfusion , the concentration of sodium at the end of ischemia , and ratios of peak to end-ischemic sodium concentrations for all simulations . Peak and end-ischemia sodium concentrations for all simulations are shown in Figures 6A and 6B , respectively . One obvious feature of Figures 5A and 6A is that there is minimal , if any , benefit in terms of peak sodium concentration during reperfusion , regardless of when and how much NHE inhibition is imposed . In fact , when NHE inhibition is instituted at reperfusion , there is a worsening of peak sodium concentration during reperfusion relative to control ( green and blue lines ) . In simulations where NHE inhibition was started at earlier time points , there is a reduction in [] prior to the onset of reperfusion . However , this appears to have a relatively small impact on the concentrations of sodium at the end of the observed reperfusion period . Figure 5B plots the maximum and minimum myoplasmic calcium concentrations for each beat in the same four simulations . The three insets show calcium transients , plotted on the same vertical axis as panel B , for 2000 ms in the regions marked by dashed vertical lines . Table 3 provides peak and mean concentrations during reperfusion and at the end of ischemia , as well as ratios , as in Table 2 . Peak concentrations for all simulations are graphed in Figure 6C . The largest calcium concentrations were seen during the R0 ( complete NHE block starting at reperfusion ) simulation , as was also the case for sodium ( Figure 6A and Table 2 ) . Intracellular pH profiles for the same four simulations are shown in Figure 5C . The values of intracellular pH after 10 minutes of reperfusion ( graphed in Figure 6D ) , as well as the mean during reperfusion , are provided in Table 4 . Relative to the control simulation , 10-minute and mean values are lower regardless of when and how much NHE inhibition was imposed . The lowest pH values were observed during the R0 simulation , which is also the simulation in which the most severe sodium and calcium overloads were observed . In our model , there are eight components through which sodium can enter and/or leave the cell: the sodium-hydrogen exchanger ( NHE ) and sodium-bicarbonate symporter ( NBC ) , two of the so-called acid exchangers , the sodium-calcium exchanger ( NCX ) , sodium-potassium exchanger ( NaK ) , fast-sodium current ( ) , background sodium current ( ) , late sodium current ( ) , and sodium movement through the L-type calcium channel ( ) . In order to develop an understanding of the relative importance of each of these eight sodium flux pathways to sodium overload , we analyzed current through each of these during the ten minutes of simulated reperfusion . Table 5 and Figure 7 provide the net number of moles of sodium moved through each component during the ten minutes of reperfusion . This value was calculated as per Eq . 37 for NHE and NBC and Eq . 38 for the remaining six ion channels and exchangers: ( 37 ) ( 38 ) where t is the time step , is the capacitive area , and F is Faraday's constant , as provided in Text S1 . is the flux through either the NHE or NBC at a given time step , and is the current through one of the six ion channels or exchangers at a given time step . For the NCX and NaK , the value calculated by Eq . 38 is multiplied by 3 . Note that this is not the true number of moles of ions moved . Mathematical simulations are discretized into finite time steps , and data was not written at every time step in order to keep data file sizes within manageable limits . Nonetheless , these values provide a standard with which the contributions of each of the eight sodium components can be compared across different simulations . In these simulations , data was sampled every tenth time step . The only sodium pathway which has a net negative inward sodium movement is the NaK . The NCX , which functions in both forward ( sodium in ) and reverse ( sodium out ) modes , increasingly favors the reverse mode during sodium overload , We also calculated the proportion of total moles of sodium moved for each of the eight sodium pathways , presented in Table 6 . The proportion values were calculated by taking the absolute value of the total ion movement for a given component and simulation , and dividing by the sum of the absolute values for all eight components in the same simulation . As can be seen in Table 6 , the NaK and NCX are responsible for most of the sodium movement into and out of the cell . As additional model validation , we performed additional simulations and compared the results to an analysis of sodium fluxes in rabbit myocytes [34] . The model was allowed to reach steady state either when paced at a rate of 1 Hz under pre-ischemic conditions , left at rest under pre-ischemic conditions , or left at rest in mild acidosis ( set to 6 . 9 ) . Sodium flux through all seven components that produce net inward sodium movement were analyzed for the last minute of simulated time ( please see Figure S1 in Text S1 for results ) . As expected , and consistent with data summarized in [34] , there was less sodium influx when the cell was at rest than when paced , due to reduced current through the sodium channels and NCX . However , also as expected and consistent with experimental observations , there was more sodium influx at rest when the cell was experiencing mild acidosis than when the resting cell was at normal pH . This was due , at least in part , to increased flux through NHE and NBC . During acidosis there was decreased NCX flux in both the simulated and experimental preparations , but this is not surprising as the NCX experiences inhibition at lower pH . Given that the NaK is one of the dominant sodium flux sources and that it is inhibited by low , we wanted to examine the effects of removing this acidotic inhibition during reperfusion . For this simulation , our model was modified such that a separate variable for intracellular pH was created and incorporated into the equations for calculating NaK flux . During the pre-ischemic and ischemic phases of the simulation , this variable equals the value of the general variable . However , during reperfusion , the value of this NaK-specific pH variable is returned to normal , while the rest of the model responds to the values of such as those in Figure 5C . We repeated the R0 simulation with this NaK-specific pH variable . The effects on sodium and calcium concentrations can be seen in Figures 8A and 8B , respectively . Removing acidosis-induced NaK inhibition reduced the peak [] over 2 mM to just under 18 mM and the peak [] was reduced to 0 . 00125 mM . Figure 8C compares the maximum and minimum values of during each beat from the control and R0 simulations to the modified R0 simulation with acidotic inhibition removed . The two insets capture 150 ms , plotted on the same vertical axis as the main figure , at the end of pre-ischemia ( left ) and at the end of reperfusion ( right ) . If the NaK is not impaired by acidosis during reperfusion , it experiences greater minimum current . Maximum current is still suppressed during early reperfusion , but improves during middle to late reperfusion .
The model presented here was used for the simulation of abrupt , complete stop-flow ischemia and abrupt-onset reperfusion . However , with some relatively simple modifications to the simulation protocols , it is possible to simulate ischemia of varying severity and following different time courses ( e . g . episodes of ischemia punctuated with intervals of reperfusion ) . This model is by no means an attempt to capture all of the mechanisms of ischemic pathophysiology . For example , [40] , [41] and [41] , [42] have been reported to be modified by acidosis . However , these sensitivities are not represented in the model , as they did not change model performance in a significant way and because of possible confounding factors in the experimental evidence with regard to the data . A discussion of the issues relating to the current can be found in [8] , [43] . Also , there has been work centered around the role of other components and processes in the development of reperfusion arrhythmias ( for example , see [44] and [45] ) . Explicit mitochondrial physiology would be an interesting addition to the model , particularly in the context of mitochondrial NHE function and proton and calcium homeostasis . In the current version of the model , the equations dictating phosphometabolite concentrations serve as a proxy for mitochondrial failure , a choice we made based upon computational time considerations . However , we expect to incorporate more detailed mitochondrial physiology in the future . Despite the many facets of ischemia physiology that are not included in our model , we believe that it is an important step towards a comprehensive ischemia-reperfusion model , and one that reproduces electrical and chemical changes that have been seen in vitro . Reactive oxygen species generation is also believed to play a major role , as is apoptosis . These will be interesting potential additions in the future once suitable mathematical representations are devised , especially if longer reperfusion times are to be modeled , where apoptosis likely plays a more significant role . Comparison of our simulation results with experimental data in Figure 4 reveals some discrepancies in behavior . The level of extracellular acidosis produced by our model under these simulation conditions is more profound than what is typically observed . In addition , the initial conditions and parameters that we have chosen yield higher sodium concentrations during ischemia and a slower rate of sodium accumulation during reperfusion than observed in the cited experiment . These are the result of choices we have made in tuning many parameters to fit data from many different groups , performed under different conditions and at different points in time . For example , an improved match to the cited sodium data can be achieved with the model , but at the expense of a decreased fit to extracellular pH and potassium data . While there is still room for improvement , we believe that this model provides a good fit to phenomena that occur during myocardial reperfusion and can yield useful insight into this complex problem . Many calcium-handling components are sensitive to changes in cellular pH , and some of these relationships are represented in our model . In addition , our model includes ATP-dependencies of components that directly and indirectly affect sodium and calcium balance , such as the NaK , SERCA , and ATP-inactivated potassium efflux channels . The importance of these influences have been alluded to above and are being explored in current work . The pathophysiology of ischemia-reperfusion injury is very complex , taking place within the context of a system that is highly coupled with nonlinear relationships between many of the components . Because such systems frequently exhibit nonintuitive behavior , we believe that the model developed in this study will prove useful in illuminating underlying mechanisms . For example , here we have used the model to propose a possible explanation for why NHE inhibition , a seemingly straightforward approach to limiting sodium-mediated reperfusion injury , has shown no meaningful clinical efficacy to date , and what efficacy has been observed requires the administration of NHE blockers prior to the onset of ischemia . Our investigations indicate that NHE inhibition can paradoxically exacerbate sodium and calcium overload . NHE inhibition suppresses intracellular pH recovery during reperfusion , which decreases the ability of the sodium-potassium exchanger to remove sodium from the cell and favors influx of sodium through the sodium-bicarbonate symporter . Most of any observed benefit of NHE inhibition appears to come from decreased sodium load resulting from reduced NHE activity prior to the onset of ischemia . Suppressing pH recovery during reperfusion is also likely to negatively impact other cellular systems , such as those responsible for handling calcium , that we have not examined here . | Myocardial ischemia , commonly observed when arteries supplying the heart become occluded , results when cardiac tissue receives inadequate blood perfusion . In order to minimize the amount of cardiac damage , ischemic tissue must be reperfused . However , reperfusion can result in deleterious effects that leave the heart muscle sicker than if the ischemia had been allowed to continue . Examples of these reperfusion injuries include lethal arrhythmias and an increased region of cell death . Some of the early events that result in reperfusion injury include changes in pH and an overload of sodium inside the cell . During reperfusion , the sodium-proton exchanger ( NHE ) removes protons from the cell in an effort to restore normal pH , in turn importing sodium ions . Many strategies have been attempted to prevent reperfusion injury , including inhibition of the NHE , with little clinical effect . Using a mathematical model that we developed to study ischemia and reperfusion in cardiac cells , we found that NHE inhibition produces more severe sodium overload , largely due to adverse consequences of the delayed pH recovery produced by NHE inhibition . These results suggest that NHE inhibition alone may not be a viable strategy , and that therapies which prolong intracellular acidosis may be problematic . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"cell",
"physiology",
"medicine",
"arrhythmias",
"electrophysiology",
"physiology",
"biology",
"anatomy",
"and",
"physiology",
"cardiovascular",
"system",
"biophysics",
"computational",
"biology",
"cardiovascular"
]
| 2011 | NHE Inhibition Does Not Improve Na+ or Ca2+ Overload During Reperfusion: Using Modeling to Illuminate the Mechanisms Underlying a Therapeutic Failure |
On the Zanzibar islands , United Republic of Tanzania , elimination of urogenital schistosomiasis is strived for in the coming years . This qualitative study aimed to better understand community knowledge , perceptions , and practices associated with schistosomiasis among school-aged children on Unguja and Pemba islands , in order to inform the development of behavior change interventions contributing to eliminate urogenital schistosomiasis . In 2011 , we conducted 35 children’s discussion groups , 41 in-depth interviews with parents and teachers , and 5 focus group discussions with community members in Zanzibar . Using a modified-grounded theory approach , we transcribed and coded the narrative data followed by thematic analysis of the emergent themes . Urogenital schistosomiasis is a common experience among children in Zanzibar and typically considered a boys’ disease . Children engage in multiple high-risk behaviors for acquiring schistosomiasis because of poor knowledge on disease transmission , lack of understanding on severity of disease-associated consequences , and lack of alternative options for water related activities of daily living and recreational play . Local primary school teachers had little to no training about the disease and no teaching tools or materials for students . Conducting activities in open natural freshwater contaminated by S . haematobium larvae compromises the health of school-aged children in Zanzibar . The perception of urogenital schistosomiasis as a minor illness rather than a serious threat to a child’s well-being contributes to the spread of disease . Understanding community perceptions of disease along with the barriers and facilitators to risk reduction behaviors among children can inform health promotion activities , campaigns , and programs for the prevention , control , and elimination of urogenital schistosomiasis in Zanzibar .
Schistosomiasis is a debilitating disease that affects poor and deprived population groups , especially in rural Africa [1] . The global impact is enormous with more than 200 million people infected with blood flukes of the genus Schistosoma [2] . Urogenital schistosomiasis , caused by S . haematobium , can include acute illness such as blood in urine ( hematuria ) and anemia in children , while fibrosis of the bladder and ureter , and kidney damage can occur as infections persist [3 , 4] . Bladder cancer can be a complication in adults and female genital schistosomiasis may be associated with increased risk of human immunodeficiency virus ( HIV ) infection [5–9] . Over the past decades , programs to reduce the morbidity caused by schistosomiasis have been implemented in many endemic countries and the number of people who received treatment with praziquantel has increased annually [10] . In order to eliminate schistosomiasis as a public health problem and to interrupt transmission in areas where morbidity control has been achieved , the World Health Organization ( WHO ) and other institutions and stakeholders are advocating the intensified use of integrated schistosomiasis control approaches [7 , 11–14] . Health education and interventions based on social and behavioral science triggering behavior change , in addition to regular preventive chemotherapy with praziquantel , are likely to become a key component of future elimination efforts [12 , 13 , 15] . Behavior change in humans requires close interaction with the at-risk population [16] . To change risk behaviors , enhance the interruption of disease transmission , and finally eliminate schistosomiasis particularly in children , innovative , community-tailored approaches are needed . Understanding the community is critical to creating effective behavioral interventions promoting the adoption of protective behaviors and reducing risk behaviors [16 , 17] . Community participatory processes are fundamental to understanding the community’s current knowledge , perceptions , attitudes , and behaviors as well as motivators and barriers to behavior change [18] . Community participation can also create ownership of public health initiatives , which is often viewed as fundamental for the success of population-based health outcomes [16 , 18] . People are experts about the communities they live in and have many different ways of knowing and gathering information . On the Zanzibar islands , United Republic of Tanzania , elimination of urogenital schistosomiasis is strived for in the coming years . In the frame of a randomized operational research trial implemented on Unguja and Pemba islands from 2012 till 2017 , the impact of behavior change interventions in addition to biannual praziquantel treatment on the prevalence and intensity of S . haematobium infections is assessed [19–21] . Here we present results of the qualitative formative research that was implemented in 2011 to inform , along with future participatory community co-design workshops , the development of a community-tailored behavioral change intervention that might help to eliminate urogenital schistosomiasis in Zanzibar [19] .
In 2011 , the National Centre for Emerging Zoonotic Diseases ( NCEZID ) of the Centers for Disease Control and Prevention ( CDC ) received and approved the qualitative formative research protocol for Zanzibar ( NCEZID Tracking Number: 103111BP ) to go forward to the CDC Human Research Protection Office ( HRPO ) and Institutional Review Board ( IRB ) for review . The HRPO and IRB determined that the project activities were exempt under regulation 45 CFR 46 . 101 ( b ) ( 2 ) and issued a written waiver . The full study protocol of the “Study and implementation of schistosomiasis elimination in Zanzibar ( Unguja and Pemba islands ) using an integrated approach” received additional ethical approval from the Zanzibar Medical Research Ethics Committee in Zanzibar , United republic of Tanzania ( reference no . ZAMREC 0003/Sept/011 ) , the “Ethikkommission beider Basel” ( EKBB ) in Basel , Switzerland ( reference no . 236/11 ) and the IRB of the University of Georgia , in Athens , Georgia , United States of America ( project no . 2012-10138-0 ) . The study is registered at the International Standard Randomized Controlled Trial Number Register ( ISRCTN48837681 ) . The data collection was conducted with support from the CDC in Atlanta , Georgia , United States of America . The CDC HRPO and IRB approved the informed consent process conducted with all participants , who took part in student group discussions and interviews . Due to a limited ability of participants to read and write the informed consent was available in both English and Kiswahili , the local language , and read aloud by trained bilingual research staff . Participants provided a verbal consent , with the consent acknowledged with the signature on the informed consent document of a witness present at the time [22 , 23] . Research staff reviewed the consent procedure and all consent forms to ensure compliance with the process . In case of children below the age of 18 years , their parents or legal guardians provided written informed consent for their participation . This qualitative inquiry was conducted including school-aged children , parents , teachers , and community leaders from seven small administrative areas , called shehias , on the islands of Unguja and Pemba from July till September 2011 . The islands of Unguja and Pemba have an estimated combined population of around 1 . 3 million people and the main industries are spices , raffia , and tourism [24 , 25] . More than 99 percent of Zanzibar's population is Muslim . Urogenital schistosomiasis constituted a considerable public health problem on both islands in the past century [26–28] , but regular treatment of the at risk population with praziquantel reduced S . haematobium prevalences and intensities [29–31] . In 2012 , the baseline survey of the “Study and implementation of schistosomiasis elimination in Zanzibar ( Unguja and Pemba islands ) using an integrated approach” revealed an overall prevalence of 4 . 3% and 8 . 9% in schoolchildren from Unguja and Pemba , respectively [20] . The field team consisted of a senior social scientist from the CDC and seven Kiswahili and English speaking research assistants from the Ministry of Health , Department of Neglected Tropical Diseases , and the Ministry of Education , Department of Health Education , in Unguja , and three research assistants from the Public Health Laboratory—Ivo de Carneri in Pemba . The local team had little or no previous experience with the application of qualitative research methods . Research assistants were trained in research ethics and qualitative data collection methods by the senior social scientist and served as the primary data collectors and logistic coordinators setting up focus groups and interviews within the communities . We used purposive sampling to recruit a homogeneous study sample of school-aged children , who might engage in risky behaviors . Such risk behaviors include swimming , fishing , bathing , washing clothes , or performing other domestic chores in ponds , lakes , streams , and rivers that are potentially contaminated with S . haematobium larvae . For this initial informative research study it was decided that the easiest way to reach school-aged children was through government supported public primary schools . Schools in the selected shehias were identified with the assistance of the staff from the Ministry of Health and from the Ministry of Education . Students in grades Standard 1 to Standard 7 were recruited through the headmaster of each school . Individual teachers and parents were also recruited through public primary schools and community leaders were recruited through local social networks . The qualitative inquiry was conducted through 35 children’s group discussions ( GD ) , 5 focus group discussions ( FGD ) with community leaders , and 41 in-depth interviews ( ID ) with teachers and parents . The children’s GD included 6–8 children of the same sex , facilitated by a Kiswahili speaking research assistant using a simple topic guide to lead the discussion ( S1 Topic Guide ) . Children were provided paper and crayons and were first asked to draw anything they wanted to draw , followed by a discussion of their picture . Then they were asked to , “Please draw me a picture about everything you know about kichocho ( the local term for schistosomiasis ) . ” These drawings encouraged more robust discussions [32] . Students drew pictures and then described their drawings about the disease , risk behaviors , and prevention ideas . A note taker managed an audio recorder and took written notes in support of the group facilitator . FGD with community leaders followed a similar format without the drawings . Individual IDIs were conducted with parents and teachers using the audio recorders without the note takers . Examples of qualitative open-ended questions used with the adults are shown in Fig 1 . The study was conducted in five shehias on Unguja ( Chaani , Dole , Kilombero , Mwera , and Uzini ) and in two shehias on Pemba ( Chambani and Kizimbani ) . The shehias were selected among the 15 behavioral study shehias on each island [19] , based on their location on the island and previous knowledge about urogenital schistosomiasis in the area [29 , 31 , 33 , 34] . Data were collected until saturation was obtained [35 , 36] . All data collection tools including project overview , informed consents , and topic guides were translated from English into Kiswahili , pretested , and modified to adapt to local linguistic and cultural nuances by the research team [37] . Topic guides explored i ) knowledge and perceptions of schistosomiasis transmission , ii ) specific behaviors among children in Zanzibar that put them at risk for acquiring S . haematobium infections , iii ) symptoms of schistosomiasis along with personal health-seeking behaviors and treatment strategies , along with iv ) ideas to prevent schistosomiasis in children . Additionally , we explored other cultural factors , gender-roles , influential communication channels , and decision-making processes [37] . Additional probes allowed for deeper exploration of the topics that emerged , supporting additional areas of interest . Following group discussions and interviews , participants received a small thank you gift for their time and participation . Group discussions conducted in schools and FGD conducted in community settings were approximately 1 . 5 hours in length , IDIs took approximately 45–60 minutes . Data were collected using audio recorders ( Olympus 70 , Olympus Corporation , Tokyo , Japan ) . Additional field notes were handwritten during the interviews and reviewed during debriefing sessions to verify accuracy of the interview session [38] . We did not collect personal identifiers . Due to the paucity of behavioral information on urogenital schistosomiasis in Zanzibar we chose a qualitative approach to better understand cultural practices associated with activities of daily living linked with contact with local natural open freshwater bodies . In this study , we examined barriers to schistosomiasis prevention and control related to urination practices of children along with recommendations for improving such practices and reducing disease threats . We used a modified grounded theory approach with an emergent qualitative thematic analysis allowing the hypothesis to be generated from the data [39] . Narrative data were transcribed into English with review following translation to ensure accurate translation and local meanings . Transcripts were entered into Atlas-ti ( ATLAS . ti Scientific Software Development GmbH , Berlin , Germany ) as a Word document ( Microsoft Corporation , Redmond , WA , USA ) to facilitate text searching , data coding and analysis . Data analysis began with the first discussions and interviews allowing for emerging , unexpected , and/or inconsistent issues to be explored in subsequent discussions and interviews [36 , 39] . The coding structure evolved inductively with the codes from the narrative data of earlier interviews informing subsequent coding of the following interviews supplemented with field notes from the interviewer and note taker [32 , 38–40] . Due to time constraints and ongoing data collection tasks , the primary author , a social scientist experienced in qualitative research , was the primary data coder with verification of interpretive codes by the research assistants . Open , axial , and selective coding was used to analyze the GD , FGD , and IDI narratives [36 , 38–41] . A coding frame was developed through open coding , a word-by-word analysis used to identify , name , and categorize explanations and descriptions of the day-to-day reality of participants as related to schistosomiasis . Consensus on the coding frame was obtained through discussions with the local qualitative research assistants , who were from Zanzibar and conducted the original interviews . Axial coding , the process of relating codes to each other , via a combination of inductive and deductive thinking , was used for analysis of specific emergent themes , across themes , and for the relationships between themes [40 , 41] . Over the course of data collection , emergent themes became redundant , suggesting that all major themes had been identified and saturation reached [42] . An analysis matrix served as a framework for the resulting findings . Narrative excerpts from an analysis framework matrix are shown in Table 1 . The trustworthiness of our data was derived from standardization of methods and documentation for auditability , triangulation of the data , and verification of data findings with local staff members [43] . A standardized implementation document guided the training and implementation of the qualitative methodology with all procedures , topic guides , informed consents , timelines , interview schedules , data collection strategies , data management , and analysis strategies written out [38 , 43] . Triangulation of data was derived through the multiple data collection methods ( GD , FGD , and IDI ) , multiple perspectives ( younger and older girls and boys as well as adult women and men ) , and multiple venues ( school-based , private home-based , and public venues ) . Findings were verified amongst local staff of the Ministry of Health in Unguja and the Public Health Laboratory Ivo-de-Carneri in Pemba by corroborating results with similar findings across other settings [44 , 45] .
As shown in Table 2 , narrative data were collected from 27 children’s GD , 5 FGD with community leaders , and IDI with 21 teachers and 16 parents on Unguja Island . We also conducted 8 children’s GD and 4 IDI with teachers on Pemba Island to verify similarities among children on the two islands . Group discussions included boys and girls from both islands , who were 6–17 years old and attended grades of Standard 1–7 . FGD included 16 male and 14 female community leaders , aged 23–72 years . They were teachers , farmers , leaders of women’s groups , religious leaders , school coaches , religious schoolteachers , petty traders , and small business owners , as well as housewives and traditional leaders . In-depths interviews were conducted with 2 male and 19 female teachers from Unguja and 2 female and 2 male teachers from Pemba . Teachers were 26 to 56 years of age . Eight fathers and eight mothers from Unguja were interviewed . Parents had a median age of 41years with a range of 24 to 72 years of age .
Qualitative research , alone or in mixed methods , has been used to better understand the experiences of people affected or at risk for numerous neglected tropical diseases such chagas disease , filariasis , and schistosomiasis [46–48] . Results of our informative in-depth , qualitative investigation of schistosomiasis among school-aged children suggested that despite previous initiatives related to urogenital schistosomiasis control and prevention in Zanzibar [29 , 44 , 49] , people’s knowledge about disease symptoms , transmission , and prevention were poor . Our findings identified several barriers to optimal disease prevention and control . First , we observed that school-aged children regularly exposed themselves to contaminated natural , open freshwater bodies through recreational and domestic activities of daily living with little knowledge about routes of schistosomiasis transmission , which is in line with findings from previous studies in Zanzibar , Tanzania , Zimbabwe , and Western Kenya [50–52] . Second , S . haematobium infection was often viewed as an infection with an intestinal worm of little significance , not typically associated with severe health consequences , and little to no disease stigma . This is in contrast to reports from previous research in Nigeria , where individuals with schistosomiasis disease were stigmatized by others [53] . The Health Belief Model posits that perceived seriousness along with perceived susceptibility , perceived benefits , and perceived barriers are critical constructs used to explain and influence changes in health behaviors [54 , 55] . It also specifies that if individuals perceive a negative health outcome to be severe and perceive themselves to be susceptible to those negative outcomes , they are more likely to adopt positive protective behaviors [17 , 54 , 55] . Drawing upon the constructs of this behavioral theory supports shifting the context of schistosomiasis to that of a blood fluke , with serious health consequences such as bladder cancer and infertility , rather than the current perception of a less severe “worm . ” Elaborating on the perceived seriousness of the infection , whether through medical information or increased awareness of the serious effects of the disease on a person’s life , is critical to address in a behavioral intervention [17 , 54 , 56] . There is evidence that theory-based , behavioral interventions can increase effectiveness among a variety of public health issues [57–60] . Synthesis of behavioral intervention research and non-regulatory interventions most often advocates the application of behavioral theory as an integral step in intervention design and evaluation [55 , 61] . Third , many people described abdominal pain , blood in the urine ( hematuria ) , pain or burning during urination ( dysuria ) , and commonly genital itching as symptoms of infection . However , as observed in studies conducted elsewhere in sub-Saharan Africa [62] , these symptoms were also perceived as sexually transmitted infections that indeed may appear similar to symptoms of urogenital schistosomiasis . A person with a sexually transmitted infection may be reluctant to seek treatment due to shame and stigma [53 , 62] . Therefore , correcting the misperception that schistosomiasis is a sexually transmitted disease , while at the same time supporting the need to seek treatment for any and all similar symptoms , could be an important component of a schistosomiasis educational campaign to improve treatment seeking . Fourth , first line treatment for a few people in Zanzibar , similar to mainland Tanzania [62] , was often home remedies and occasional use of locally available herbalists followed by more conventional treatments when those earlier ones had failed . Lack of decentralized , locally available drugs and cost of transportation were also identified as barriers to seeking more conventional drug treatments . The decentralization of drug treatment to the local level as well as increasing knowledge about free drug treatment accessible through mass drug administration campaigns could improve treatment seeking among infected individuals . Further research into understanding any underlying barriers to treatment seeking behaviors should be explored [63] . Fifth , little available formal education about disease transmission contributed to myths and misperceptions about routes of transmissions , causes , and severity of disease , treatment , and ultimately prevention of disease . Schoolteachers and Koran school ( Madrassa ) teachers , viewed as influential people in children’s lives lacked formal scientific training , teaching materials , and other resources to be able to educate students about schistosomiasis . Teachers reported a need for a teacher’s training with a standardized , detailed syllabus to teach children about schistosomiasis during school sessions . Trainings could be set up similar to the Lushoto Enhanced Health Education Project that introduced interactive teaching methods into mainland Tanzanian study schools and demonstrated a feasible and effective intervention capable of changing schistosomiasis knowledge and health seeking behaviors among children [64] . The inclusion of religious teachers as change agents could maximize exposure of a schistosomiasis educational program to a broader community because they often engage children who may not attend government schools . Trained school and religious teachers could instill a perception of perceived seriousness of disease as well as perceived susceptibility of disease among children engaging in risky behaviors . Teachers could also identify and address the barriers to change and promote perceived benefits of reducing risky behavior to children . Educating through schools could encourage students to act as change agents through peer education , role modeling , and shifting social norms of acceptable behavior [65 , 66] . Peer education , defined as “the teaching or sharing of health information , values and behaviors by members of similar age or status , ” is widely used in the field of health promotion and education recently , such as the prevention of HIV/acquired immune deficiency syndrome ( AIDS ) , smoking , and alcohol and drug use [67–71] . Peer education is focused on sharing information and experiences along with trust between the people in the similar context and learning from each other . Peer education , has been noted as a feasible method for transferring schistosomiasis knowledge from students to parents [65 , 66] . Sixth , most adults , and some children recognized the difficulty of extinguishing the behavior of urinating in the ponds and streams . It was seen as a private behavior and often associated with urgent need . Children and adults described educational , behavioral , and structural interventions to prevent kichocho in children . Community members often described the need for the community to work together to prevent kichocho in children suggesting the importance of a participatory approach to intervention development and implementation . Previous research has noted that top down approaches to community interventions have been perceived by some community members as not in their best interest or being a poor fit for the socio-cultural context within the community [72–74] . The lack of attention to an individual’s social , cultural , religious , environmental , and physical context often results in a poor understanding of why an intervention is valuable and ultimately in an inadequate adoption of the desired positive behaviors and practices by community members [72–74] . This may explain why despite years of community administered preventive chemotherapy , the perception of schistosomiasis in Zanzibar was that of a commonplace , minor illness , rather than a serious threat to a child’s wellbeing . Administering preventive chemotherapy without addressing the local circumstances of community members with tailored communication and educational efforts can lead to not only misunderstandings but also to potentially poor treatment compliance [75 , 76] . Understanding community perceptions along with the social , religious , economic and environmental context of schistosomiasis risk and risk reduction behaviors among children can inform behavior change interventions that are relevant and provide meaning to the vulnerable populations in Zanzibar [74 , 77 , 78] . A recent evaluation of a comic-strip medical booklet Juma na Kichocho associated with a schistosomiasis health education campaign in 16 primary schools in Zanzibar reported disappointing findings [44 , 77] . The authors recognized that changing the behaviors of children could not be done by an isolated school curriculum but needed to consider their everyday realities of daily living [77] . The information garnered from our qualitative inquiry will allow for the ideas and problem solving solutions of community members to be incorporated into a behavioral intervention that is germane to others in their communities . Increasingly , there is a commitment to bringing a community perspective into research and implementation of interventions along with a growing body of evidence that public health and health-promotion interventions based on social and behavioral science theories are more effective than those without a theoretical foundation [55 , 79 , 80] . Drawing upon the constructs of perceived seriousness , perceived susceptibility , perceived benefits , and perceived barriers from the Health Belief Model complemented by a social ecological model that addresses multiple levels of the community could provide a functional framework for designing , implementing , and evaluating a health promotion program for the prevention and control of schistosomiasis tailored to the context of community members , particularly school-aged children [17 , 55 , 74 , 79 , 81] . There were several limitations to this inquiry . Given that we used a purposive , convenience sample , the findings may not be representative of all members of the communities in which the inquiry took place , and results are not generalizable . The triangulation of data suggests that there were similarities across behaviors of school-aged children attending the government primary schools on both islands where we conducted the student discussion groups . The behaviors we assessed appeared to be generally practiced among children across the shehias on Unguja and Pemba and the lessons learned could be used to tailor messages for future schistosomiasis control programs for primary school aged children . That being said , these findings are not generalizable to children attending private schools or not attending school at all . Further investigation is needed to explore the schistosomiasis knowledge , attitudes , perceptions , and practices of students in private schools and of students who do not attend school to assess if they are similar to those from children who participated in our student discussion groups . There may have been information bias during IDI , GD , and FGD as interview subjects may have provided answers that they believed the interviewers expected or wanted to hear . Additionally , bias may have been introduced due to only having a single coder , even though data interpretations and language translations were substantiated with local research assistants . Conducting recreational and domestic activities of daily living in water contaminated with S . haematobium larvae compromises the health of school-aged children in Zanzibar . An important objective of this study was to facilitate improved design of an educational and control program . Urogenital schistosomiasis , characterized as a minor illness typically of boys , along with the lack of formal school-based and community-wide education about disease transmission , symptoms , and treatment can contribute to undiagnosed disease and a lack of treatment among both girls and boys . Understanding community perceptions of disease along with the barriers and facilitators to risk reduction behaviors among children can inform behavior change activities and health promotion programs augmented with chemotherapies for an integrated approach in support of the prevention , control , and elimination of urogenital schistosomiasis in Zanzibar and elsewhere . | On the Zanzibar islands , United Republic of Tanzania , elimination of urogenital schistosomiasis , a disease caused by infection with a blood fluke ( Schistosoma haematobium ) , locally known as kichocho , is strived for in the coming years . This study used qualitative research methods of focus groups and in-depth interviews with adults , and group discussions with school-aged children to explore ( i ) knowledge and perceptions of kichocho transmission , ( ii ) specific behaviors among children in Zanzibar that put them at risk for acquiring infections with the kichocho parasite , ( iii ) symptoms of kichocho along with personal health-seeking behaviors and treatment strategies , and finally ( iv ) ideas for preventing kichocho in children . We found that there was little available formal education about disease transmission , which contributed to myths and misperceptions about routes of transmissions , causes , and treatment of kichocho . School-aged children regularly exposed themselves to contaminated natural , open freshwater bodies through recreational and domestic activities of daily life . Kichocho was often wrongly viewed as an infection with an intestinal worm of little significance , rarely associated with severe health consequences , and with little to no disease stigma . Local primary school teachers had little to no training about the disease and no teaching tools or materials for students . The findings add valuable insights into how current knowledge , perceptions , and practices impede optimal disease prevention and control and highlight the necessity for a community tailored behavioral intervention to interrupt transmission of urogenital schistosomiasis . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"children",
"ecology",
"and",
"environmental",
"sciences",
"medicine",
"and",
"health",
"sciences",
"rivers",
"education",
"sociology",
"tropical",
"diseases",
"geographical",
"locations",
"teachers",
"social",
"sciences",
"parasitic",
"diseases",
"age",
"groups",
"physiological",
"processes",
"tanzania",
"aquatic",
"environments",
"bodies",
"of",
"water",
"neglected",
"tropical",
"diseases",
"africa",
"families",
"urination",
"behavior",
"marine",
"and",
"aquatic",
"sciences",
"schools",
"people",
"and",
"places",
"helminth",
"infections",
"professions",
"schistosomiasis",
"freshwater",
"environments",
"earth",
"sciences",
"physiology",
"population",
"groupings",
"biology",
"and",
"life",
"sciences"
]
| 2016 | Community Knowledge, Perceptions, and Practices Associated with Urogenital Schistosomiasis among School-Aged Children in Zanzibar, United Republic of Tanzania |
Daily rhythms in gene expression play a critical role in the progression of circadian clocks , and are under regulation by transcription factor binding , histone modifications , RNA polymerase II ( RNAPII ) recruitment and elongation , and post-transcriptional mechanisms . Although previous studies have shown that clock-controlled genes exhibit rhythmic chromatin modifications , less is known about the functions performed by chromatin remodelers in animal clockwork . Here we have identified the Brahma ( Brm ) complex as a regulator of the Drosophila clock . In Drosophila , CLOCK ( CLK ) is the master transcriptional activator driving cyclical gene expression by participating in an auto-inhibitory feedback loop that involves stimulating the expression of the main negative regulators , period ( per ) and timeless ( tim ) . BRM functions catalytically to increase nucleosome density at the promoters of per and tim , creating an overall restrictive chromatin landscape to limit transcriptional output during the active phase of cycling gene expression . In addition , the non-catalytic function of BRM regulates the level and binding of CLK to target promoters and maintains transient RNAPII stalling at the per promoter , likely by recruiting repressive and pausing factors . By disentangling its catalytic versus non-catalytic functions at the promoters of CLK target genes , we uncovered a multi-leveled mechanism in which BRM fine-tunes circadian transcription .
Circadian clocks are endogenous timekeeping mechanisms that drive rhythms in physiology and behavior with an approximately 24-hour period , allowing organisms from all kingdoms of life to anticipate and operate through predictable daily environmental changes . Much progress has been made in understanding the architecture of the molecular oscillators regulating these cell-autonomous clocks in different organisms , and the mechanisms in which the core oscillators communicate temporal information via circadian transcription that ultimately drives many overt physiological rhythms [1–3] . In Drosophila melanogaster , two basic helix-loop-helix PER-ARNT-SIM ( bHLH-PAS ) transcription factors CLOCK ( CLK ) and CYCLE ( CYC ) are at the core of the molecular oscillator , which operates through two interlocked transcriptional-translational feedback loops [1] . In the major loop , CLK and CYC form heterodimers and bind to E-box regulatory elements on genes encoding the main negative factors PERIOD ( PER ) and TIMELESS ( TIM ) that inhibit the transcriptional activity of CLK-CYC , and consequently their own transcription , closing off one autoregulatory feedback circuit . CLK-CYC also initiate a second loop by activating the transcription of genes encoding regulators of Clk expression [4–5] . VRILLE ( VRI ) , a basic leucine zipper ( bZIP ) transcription factor , binds to D-box ( also called V/P box ) elements on the Clk promoter to repress Clk activation by PAR Domain Protein 1ε ( PDP1ε ) . The temporal control in the expression levels of these key oscillating mRNAs and proteins over the circadian cycle is critical for the normal progression of the clock . CLK activation of per , tim , vri , and pdp1ε occurs in late day and peaks in the early evening . Post-transcriptional and post-translational regulatory mechanisms create a time-delay , causing the levels of these proteins to peak about 6 hours later in mid to late evening [6–13] . As PER and TIM proteins accumulate and enter the nucleus , they dimerize and repress the activity of CLK-CYC . This repression is relieved upon sunrise due to the degradation of light-sensitive TIM [14] and subsequently PER a few hours later through the proteasome pathway [15–16] , thus initiating another round of CLK-CYC-mediated transcription . On the other hand , the expression of Clk is antiphase , first initiating during the late evening and peaking in the early morning . The precise timing of Clk expression is the result of differential protein accumulation of VRI and PDP1ε due to yet undiscovered mechanisms [4–5] . Whereas there is little delay in VRI accumulation following vri mRNA production , there is a 3 to 6 hour delay in PDP1ε accumulation , postponing prominent accumulation of Clk mRNA until early to mid-day . Two other clock components , clockwork orange ( cwo ) [17–18] and nejire ( nej ) /CREB-binding protein ( CBP ) [19–20] , have also been shown to regulate CLK-dependent transcription , but the exact mechanisms are still controversial . Outside of the core oscillator , CLK-CYC have been found to bind more than 800 downstream target genes , leading to their rhythmic transcription [21] . Since the observations of chromatin modifications occurring as a result of light resetting [22] and rhythmic histone acetylation in circadian promoters [23] in mammalian clock systems , the importance of chromatin remodeling in modulating the activity of clock transcription factors and circadian transcription in animal clocks , as well as clocks in other organisms , is becoming increasingly evident [24–25] . Gene-specific and genome-wide studies have now established that many clock-controlled genes ( ccgs ) exhibit rhythmic chromatin modifications . For example , many hallmarks of transcriptional activation such as H3K9ac and H3K4me3 have been found to coincide with rhythmic CLK-CYC binding to the E-boxes of per and tim in Drosophila [26] as well as CLOCK-BMAL1 dependent transcription in mammals [27–28] . Many of the histone modifying proteins and their roles in regulating circadian transcription have now been explored ( reviewed in [24–25] ) . In addition to histone modifiers , ATP-dependent chromatin remodelers can regulate transcription factor accessibility to DNA through mechanisms such as nucleosome reorganization [29–31] . This aspect of chromatin remodeling in clock systems has been studied most extensively in Neurospora crassa . Key transcription factors WHITE COLLAR-1 ( WC-1 ) and WHITE COLLAR-2 ( WC-2 ) were observed to bind differentially to the frequency ( frq ) promoter as chromatin structure is altered , and a number of ATP-dependent chromatin remodelers including CLOCKSWITCH ( CSW ) , chromodomain helicase DNA-binding ( CHD-1 ) , and Clock ATPase ( CATP ) , have been identified to play a role in this process [32–34] . Whereas CSW and CHD-1 have been shown to facilitate the downregulation of frq transcription , CATP is believed to decrease nucleosome density , increase WCC ( WHITE COLLAR complex ) binding to frq , and promote frq activation . Recently , an additional ATP-dependent remodeler , SWI/SNF ( SWItch/Sucrose NonFermentable ) , has been implicated in remodeling chromatin to allow for activation of frq [35] . Despite the growing realization that nucleosome reorganization is an important aspect of regulating circadian transcription , how it contributes to animal clocks is not well understood . Kismet ( kis ) , a subunit of an ATP-dependent remodeler has been shown to be involved in light entrainment in the Drosophila circadian timing system , but does not seem to affect the core oscillator [36] . CLOCK has recently been found to act as a pioneer transcription factor to open up the chromatin in mouse liver clocks to facilitate binding of additional transcriptional factors on CLOCK target genes , but the mechanism still needs to be characterized [28] . To better understand the regulatory role of chromatin remodeling in the Drosophila clock , we used a proteomic approach and screened for CLK interactors that are involved in nucleosome organization . Among the partners we discovered was the Brahma ( Brm ) ( SWI/SNF class ) chromatin-remodeling protein complex . Similar to the SWI/SNF complex originally identified in yeast , the orthologous complex in animals also possesses chromatin remodeling activity and is implicated in a variety of cellular processes including differentiation , proliferation , and DNA repair [29] . In Drosophila , the Brahma complex is comprised of multiple protein components , in which the brm gene encodes BRM , the catalytic subunit containing the ATPase responsible for hydrolysis of ATP to mobilize nucleosomes [37–38] . In addition to an ATPase domain , BRM has a bromodomain [37] , which can recognize and bind to acetylated lysine residues on histone tails [39–40] , hence its close association with transcriptional activation . Nonetheless , BRM activity has also been associated with repression depending on target genes and cell types [41] . Here we characterize the role of BRM in regulating the Drosophila circadian clock through modifications of the chromatin landscape . We also differentiate between its catalytic function in modifying chromatin structure and its non-catalytic function , leading to downstream effects on transcription factor binding , RNA polymerase II ( RNAPII ) occupancy , clock gene expression , and overt behavioral rhythms . Our studies suggest that the SWI/SNF ( Brahma ) complex has a key regulatory role in eukaryotic clocks and provide novel insights into how animal clocks are regulated by chromatin remodeling .
To identify chromatin-remodeling proteins that regulate CLK-dependent transcription in Drosophila , we performed a label-free quantitative proteomic screen using affinity purification followed by tandem mass spectrometry ( MS/MS ) to identify CLK-interacting proteins in Drosophila Schneider 2 ( S2 ) tissue culture cells ( Lam et al . in preparation ) . We focused on the interaction between BRM and CLK , rather than BRM and CYC , as CLK has been shown to be the limiting factor in CLK-CYC dependent transcriptional activation [42] and ectopic clocks can be generated by misexpressing CLK alone , indicating its central role in circadian gene activation [43] . Prior work has shown that exogenous expression of CLK in S2 cells can interact with endogenous CYC to activate circadian promoters and is also responsive to the inhibitory effects of PER-TIM , indicating that the core CLK-dependent transcriptional machinery can operate in S2 cells [44] . Cytoplasmic and nuclear extracts were prepared from S2 cells expressing recombinant versions of CLK modified with epitope tags to facilitate purification . Our quantitative proteomic pipeline identified multiple subunits of the Brahma complex as significant interactors of CLK , specifically when probing nuclear extracts . We validated our pull-down of CLK with endogenous BRM , the catalytic subunit of the complex , by co-expressing epitope-tagged versions of both proteins in S2 cells followed by immunoprecipitation ( IP ) . Western blot analysis of reciprocal co-IPs indicated strong interaction of BRM to CLK ( Fig 1A ) . In addition , interactions of BRM to negative factors PER and TIM were also tested . Whereas TIM was found to interact with BRM to a similar extent as interactions between BRM and CLK , the interaction between BRM and PER was detectable , but much weaker ( Fig 1A ) . Co-IP experiments using proteins extracted from whole fly heads were also performed to confirm the interactions observed in S2 cells ( S1 Fig ) . To facilitate analyzing BRM in flies , we generated responder transgenic lines that express BRM carrying a 3X FLAG epitope tag ( UAS-BRMRFL ) and directed their expression using a tim ( UAS ) -gal4 driver targeting tim-expressing cells , herein termed TUG ( II ) [45] . These flies will be discussed in more details below . Whereas BRM was found to interact with CLK when CLK target gene transcription is active ( ZT8 to ZT20 ) ( S1A Fig middle panel and S1C Fig ) , its interaction with TIM appeared stronger at ZT20 , during the downswing of the transcription cycle ( S1A Fig bottom panel and S1D Fig ) . In addition , BRM appears to interact preferentially with hypophosphorylated CLK ( S1E Fig ) , which is the predominant isoform bound to target gene regions during active transcription [46–48] . Interaction of BRM and PER was not observed in fly heads , consistent with results obtained in S2 cells ( Fig 1A ) . We next determined if the Brahma complex is involved in regulating the circadian oscillator and generating ~24 hour rhythms . We used RNAi to individually knock down the expression of each of the core subunits ( Fig 1B ) in tim-expressing clock neurons ( TUG ) and performed fly locomotor activity assays to detect abnormal free-running rhythms . Following entrainment under standard conditions of 12hr light:12hr dark ( LD ) at 25°C , flies in which individual Brahma complex subunits ( Brahma , Bap55 , Bap60 , Bap111 , Bap155 or Moira , and Snf-related 1 ) were knocked down by RNAi showed period-lengthening to a similar extent ( 1 to 2 hours longer relative to parental control ) when placed into constant dark ( DD ) conditions ( Fig 1C ) . This suggests that the Brahma complex is necessary for sustaining a normal circadian period , and that knockdown of one subunit may be sufficient in impairing the function of the entire complex within the context of clock function . Quantitative PCR ( qPCR ) analysis using RNA extracted from whole fly heads revealed a roughly 40% knockdown of brm in the flies expressing brm RNAi in tim-expressing cells ( Fig 1D–1F ) . However , since brm is involved in many physiological processes and is expressed in many cell types in addition to tim-expressing cells , we expect that the actual level of brm knockdown to be greater within tim-expressing cells . To support a role for the brm complex in circadian regulation , we also used pdf-gal4 to drive brm RNAi expression in a smaller subset of clock neurons that are important for establishing circadian period and observed a similar period-lengthening effect ( Fig 1G ) . As brm has been shown to be involved in development [37 , 49] , we sought to rule out developmental defects from decreased brm expression contributing to the observed behavioral phenotypes . To this end , we utilized a temperature-sensitive GAL80 system ( UAS-tub-Gal80ts ) to repress activation of the responder brm-RNAi transgene by GAL4 during development by maintaining flies at 18°C until 3 days before the locomotor activity assay . We then transferred the flies to 29°C to relieve the GAL80 repression and initiate brm RNAi expression . The behavioral assay was performed at 29°C and brm RNAi knockdown in clock neurons still resulted in period-lengthening ( Fig 1G ) , thus ruling out contributions from developmental effects . Finally , as an independent method to knockdown BRM function , we assayed flies expressing brmK804R in tim-expressing clock cells ( UAS-brmK804R X TUG ( II ) ) . The brmK804R transgene encodes a catalytic inactive mutant of BRM that contains a lysine to arginine substitution in the ATP-binding site [50] . Flies expressing brmK804R exhibited more severe circadian rhythm abnormalities as compared to brm RNAi knockdown when both transgenes were driven with the TUG driver ( Fig 1G ) . Closer inspection of their behavioral rhythms revealed that brmK804R mutants displayed period-lengthening in the first two days into DD as in the case of the brm RNAi knockdown , which quickly deteriorated into arrhythmicity starting DD3 ( S2 Fig ) . Since we observed BRM interacting with CLK and showed that brm knockdown and mutation affect oscillator function , we hypothesized that BRM may be interacting with CLK at its target genes to regulate expression . To examine if BRM binds to CLK targets , we sought to assay BRM binding to E-box elements at the promoters of per and tim . We generated the UAS-BRMRFL responder and directed its expression using TUG ( II ) . Introduction of the FLAG-brm transgene did not alter circadian clock function as assayed by behavioral rhythms using three independent responder lines ( Fig 1H ) . This is perhaps not surprising since BRM protein is normally incorporated into a multi-subunit protein complex ( Fig 1B ) , and overexpression of brm alone might not lead to increase in functional Brm complexes . This suggests that these transgenic flies can be used as tools to examine BRM binding to CLK target genes . In addition , the FLAG epitope tag has been used successfully for ChIP analysis in studying Drosophila clocks [21] . ChIP followed by qPCR analysis of flies expressing BRMRFL revealed that over the circadian cycle , BRM is constitutively bound to the fifth E-box element at the per promoter ( Fig 2A ) , which has previously been identified as a CLK binding site for transcriptional activation [26 , 51] and is within a 69 bp clock regulatory sequence ( CRS ) , a region critical in generating the oscillatory patterns of per expression [52–53] . Additionally , BRM was found to bind to the first E-box element on the tim locus ( Fig 2B ) , which has also been identified as a CLK binding site [26] . Control ChIP experiments were also performed on flies that do not express FLAG-tagged transgene to rule out the possibility of non-specific binding ( S3 Fig ) . It is worth noting that BRM localization on circadian E-box elements appears weakly rhythmic . However , upon quantification of FLAG-BRM expression ( Fig 2C and 2D ) , it seems that its protein levels at ZT4 and ZT22 are slightly elevated relative to levels found at ZT10 and ZT16 , suggesting that the level of BRM binding to per and tim E-box elements may reflect expression levels of FLAG-BRM . Quantitative PCR analysis of endogenous brm expression did not reveal any cycling of brm steady state mRNA ( refer to Fig 1E ) , suggesting the weak rhythmicity in FLAG-BRM binding to per and tim promoters may be the result of driving FLAG-BRM expression using a tim driver . For the target genes we tested ( per and tim ) , we observed preferential binding to the promoters over gene bodies ( S4 Fig ) , consistent with binding patterns of BRM in both Drosophila and mammals genes [54–55] . The BRM complex has previously been shown to have either stimulating or repressive effects on gene expression depending on target genes by altering DNA-histone contacts in Drosophila developmental pathways [41 , 56 , 57] . Since we found that BRM interacts with hypophosphorylated CLK , our initial hypothesis was naturally that BRM cooperates with CLK to facilitate activation of CLK targets such as per and tim , perhaps by opening up the chromatin . This is an attractive hypothesis for two reasons . First , CLK has been labeled a pioneer transcription factor with the ability to bind DNA-wrapped nucleosomes and subsequently recruit additional factors to de-condense the chromatin to perpetuate transcriptional activation [28] . Moreover , the SWI/SNF complex in Neurospora has recently been shown to decrease nucleosome density , helping to activate frq transcription [35] . Using qPCR , we assayed steady state mRNA levels of various CLK-dependent transcripts ( per , tim , vri , pdp1ε ) extracted from heads of flies expressing brm RNAi as compared to control flies ( Figs 3A and 3B and S5A ) . Surprisingly , we found that there was an increase in the levels of all transcripts tested in flies expressing brm RNAi , especially during time points when these genes are normally actively transcribing ( ZT 8 to 16 ) . We also analyzed nascent pre-mRNA levels , focusing specifically on per and tim , and observed a similar extent of increase ( S6A Fig ) . This initial evidence suggests that BRM may have a repressive role in limiting CLK-dependent transcription in the Drosophila clock . Since CLK-dependent transcription is still being effectively repressed starting in the early evening in flies in which brm is knocked down by RNAi , the role of BRM may be limited to preventing excessive transcription during the active phase , and not to initiate and maintain transcriptional repression . In support of our hypothesis that BRM has a repressive role , it is noteworthy that in a recent genome-wide study , inactivation of two core subunits ( SNF5 and BRG1 ) of the Brahma complex in mammalian cells resulted in variable gene expression outcomes , with more genes being upregulated rather than downregulated [55] . In the same study [55] , inactivation of Brahma subunits and upregulation of a large subset of genes correlated with reduced nucleosome occupancy , especially in the peri-TSS ( transcription start site ) region . We therefore investigated whether the increased gene expression observed in circadian transcripts ( Figs 3A and 3B , S5A and S6A ) in flies expressing brm RNAi can be attributed to a decrease in nucleosome occupancy by measuring histone H3 density using ChIP-qPCR on per and tim promoters ( E-boxes ) . At all time points tested , we observed a decrease in H3 density in the brm RNAi flies as compared to the control ( Fig 3E and 3F ) . We performed the same experiment using flies expressing the catalytically-inactive BRMK804R protein in tim-expressing cells and observed the same decrease in H3 density at the per and tim promoters ( Fig 3G and 3H ) , indicating that the normal catalytic function of BRM at these promoters may be to increase nucleosome density . Interestingly , in the TUG control flies , histone H3 density at both the per and tim promoters is higher at ZT10 and 16 , when transcription at these promoters are active , and is lower in the mid to late night when transcription is strongly repressed ( ZT22 ) . Although not as prominent , H3 density was also higher at ZT10 and 16 compared to ZT22 in the TUG ( II ) control flies for the data set to examine H3 density in flies expressing brmK804R ( Fig 3G and 3H ) . The variance stemming from three biological replicates for the control TUG ( II ) flies at ZT4 were greater than at other time points tested ( Fig 3G and 3H ) , and might have masked the weak rhythmicity in H3 density that appeared more prominent in the TUG control flies ( Fig 3E and 3F ) . In general , the effects of brm RNAi and brmK804R on H3 density were not as prominent at ZT22 , further suggesting that BRM does not play a major role during the strong repression phase of clock gene expression . Taken together , the results suggest that BRM may function to fine-tune clock gene expression by contributing to rhythmic changes in the chromatin landscape , leading to increased nucleosome occupancy during the active phase of CLK-mediated transcription . Since flies expressing brmK804R showed decreased H3 density as compared to control flies , similar to the results for flies expressing brm RNAi , we anticipated that expression of brmK804R would also lead to an increase in mRNA expression of CLK target genes . Curiously , when we assayed clock gene expression using qPCR in flies expressing brmK804R , we observed decreased clock gene expression with the time of peak expression remaining unchanged as compared to the control ( Figs 3C and 3D and S5B ) . Analysis of nascent per and tim pre-mRNA levels using conventional qPCR did not show consistent decrease in expression , however analysis using droplet digital PCR ( ddPCR ) , which has higher resolution , revealed small but significant decreases in nascent per and tim pre-mRNA expression in the brmK804R mutant ( S6B Fig ) . The small decrease in per and tim transcripts observed in flies expressing brmK804R as compared to control when measuring pre-mRNA and the relatively larger decrease observed when measuring steady state mRNA suggests that the regulation of per and tim expression may be affected in the brmK804R mutant at both the transcriptional and posttranscriptional level . The brmK804R allele encodes a catalytically inactive BRM protein [50] , suggesting that this mutant may represent a valuable tool to disentangle the catalytic and non-catalytic functions of BRM in regulating clock gene expression . The most parsimonious explanation for why both brm RNAi and brmK804R lead to reductions in nucleosome occupancy is that they reduce the amount of endogenous catalytically active BRM that can bind chromatin and increase nucleosome density . In the case of the brmK804R mutant , the catalytically inactive BRMK804R protein may be incorporated into endogenous Brm complexes to impair their catalytic function . However , the fact that brmK804R does not lead to an increase in clock gene expression suggests that once bound to chromatin , the catalytically inactive BRMK804R protein functions in an inhibitory manner to repress transcription despite a more open chromatin landscape , perhaps by recruiting additional inhibitory factors . Thus , BRM likely modulates CLK-CYC-mediated transcription in a complex manner that is balanced between stimulation and inhibition via adjusting the accessibility of both positive and negative factors to the chromatin . This would explain the discrepancy in the observed differences in circadian gene expression in the two classes of brm knockdown . Since relaxed chromatin structure has been associated with being more permissive to transcription factor binding ( reviewed in [58] ) , and we observed changes in clock gene expression in the two different classes of brm knockdown flies , albeit in opposite direction , we next examined whether CLK ( activator ) and PER ( repressor ) binding to per and tim promoters was affected in flies expressing either brm RNAi or the brmK804R transgene using ChIP-qPCR ( S7 Fig ) , and whether differential CLK and/or PER binding contributed to the discrepancies in gene expression levels observed in flies expressing brm RNAi and brmK804R . CLK binding was not expected to be affected by the decrease in nucleosome occupancy , since in mammals , CLOCK:BMAL1 have been shown to bind to nucleosome-bound DNA [28] . In control TUG driver lines , we observed the rhythmic binding of CLK to the per and tim promoters that has previously been reported ( Fig 4A–4D ) [51] . Indeed , we did not see an increase in CLK occupancy in flies expressing brm RNAi nor brmK804R despite the decrease in nucleosome density ( Fig 4A–4D ) . This suggests that the brm RNAi-induced increase in CLK target gene expression ( Fig 3A and 3B ) is due to mechanisms downstream of CLK binding , likely because of the more open chromatin . However , we observed a significant decrease in CLK binding on the per promoter in flies expressing BRMK804R , especially during the daily upswing in gene expression ( Fig 4A ) , suggesting that the non-catalytic activity retained by BRMK804R exhibits inhibitory effects on CLK binding to its target genes that becomes more prominent with decreased nucleosome density . As PER does not generally enter the nucleus until ~ZT18 [59] , the negative effect of BRMK804R on CLK binding to the per promoter is expected to be independent of PER repression . It is interesting to note that we did not see a significant decrease in CLK binding on the tim promoter in flies expressing brmK804R as we did for the per promoter ( Fig 4C ) . This suggests that while per and tim are both activated by CLK , binding of CLK to the tim promoter may be more conducive , and therefore less sensitive to the inhibitory effects introduced by BRMK804R ( Fig 4A and 4C ) . Supporting this , there is clear evidence that the transcription rate of tim is higher than per with higher amplitude cycling [60] . This could be partially due to the presence of additional non-canonical E-boxes in the tim promoter [61] . Another possible factor contributing to the decrease in CLK binding at the per promoter is that overall CLK levels may be reduced in flies expressing brmK804R . Since Clk expression is regulated by proteins encoded by two CLK-activated targets , vri and pdp1ε , and we have already shown that expression of these genes are reduced in flies expressing brmK804R ( S5B Fig ) , it is possible that Clk transcription is affected as well , resulting in decreased CLK expression . Indeed , gene expression analysis by qPCR ( S8 Fig ) and immunoblotting ( S9A and S9B Fig ) confirmed that both Clk mRNA and CLK proteins are decreased in the brmK804R mutant . The large decrease in CLK protein compared to the relatively modest decrease in Clk mRNA in flies expressing brmK804R suggests that the non-catalytic activity of BRM could affect CLK stability . Although , if that is indeed true , we would expect to see a decrease in CLK binding at the tim promoter as well . Overall , our results suggest that the observed decrease in CLK target gene expression in flies expressing brmK804R may partly be a consequence of reduced CLK binding to their promoters due to lower CLK levels . But since tim expression decreases in flies expressing brmK804R despite similar levels of CLK binding as compared to control flies ( Fig 4C ) , it is likely that the non-catalytic function of BRM negatively impacts CLK target gene expression through additional mechanisms outside of its effects on CLK binding . We also assayed the levels of Clk mRNA and protein in flies expressing brm RNAi and found that as with per and tim , both mRNA and protein levels were elevated ( S8 Fig and S9C and S9D Fig ) . Although CLK levels are elevated in flies expressing brm RNAi , no increase in CLK binding at the per and tim promoters was observed ( Fig 4B and 4D ) , suggesting that total CLK levels alone cannot fully account for the observed changes in CLK binding to target promoters . Nevertheless , lower CLK levels might be rate-limiting with respect to transcription under certain circumstances , e . g . per transcription in flies expressing brmK804R ( Figs 3C and 4A ) . In addition to changes in CLK binding , alterations in PER binding to CLK target gene promoters in flies expressing either brm RNAi or brmK804R could also contribute to the observed changes in gene expression . We therefore assayed PER binding to per and tim promoters in both brm mutants using ChIP-qPCR , but did not detect significant changes in PER binding to either promoter ( Fig 4E–4H ) . The level of PER binding to per and tim promoters also did not appear to be sensitive to the differential amount of PER resulting from higher per expression in brm RNAi flies and decreased per expression in flies expressing the brmK804R transgene ( S9A and S9B Fig ) . We therefore conclude that the level of PER binding to per and tim promoters does not appear to be a significant factor that could explain the observed differences in per/tim mRNA levels between flies expressing brm RNAi and brmK804R . Thus , the combined results suggest that BRM directly suppresses expression at the per and tim promoters by increasing nucleosome occupancy and in some still undefined manner inhibiting transcription in a catalytically independent manner . Of course , due to the interconnected feedback loops , the effects of BRM on the levels of positively ( e . g . , PDP1ε ) and negatively ( e . g . , PER , VRI ) acting factors could in turn contribute to changes in the levels and/or activity of CLK , further modulating clock gene expression . However , the fact that clock gene expression levels do not correlate very tightly with either CLK or PER levels , suggests that the main effects of BRM on clock gene expression are direct . Chromatin remodelers , such as the Brahma complex , are known to influence transcriptional regulation at the level of chromatin structure as more open chromatin is generally believed to be associated with less physical blockage for RNAPII . In addition , the Brahma complex was shown to interact with RNAPII and regulate its activity through transient stalling downstream of the transcription start site ( TSS ) [62] . RNAPII stalling or pausing has been shown to heavily regulate the expression of many eukaryotic genes and represents an additional step in transcriptional regulation beyond RNAPII recruitment ( reviewed in [63] ) . Pausing for most genes has been observed around 30 to 100 downstream of the TSS [64] , but pausing even at 1 . 5 kb into the coding region has been observed [62] . To obtain a more comprehensive understanding of the progression of RNAPII at the per and tim loci , two additional regions along the genes were analyzed by ChIP-qPCR in addition to the E-boxes initially identified as BRM binding sites ( S10 Fig ) . The TSS as well as a region within the gene body ( coding region ) was analyzed for each gene . The RNAPII antibody used for our ChIP analysis recognizes RNAPII with phosphorylated CTD repeat ( serine 2 and serine 5 ) with preference to serine 5 , therefore it is expected to detect initiated and transiently paused RNAPII and to a lesser extent , elongation-competent RNAPII [27 , 65] . When measuring RNAPII occupancy in control flies , we observed constitutive binding at the per CRS ( Fig 5A , left panel ) , consistent with previously reported findings [26] , as well as an increased occupancy at the TSS , a signature of possible RNAPII transient stalling ( Fig 5A , middle panel ) . This increase in RNAPII occupancy at the per TSS was particularly prominent at ZT16 , possibly for fine-tuning per RNA transcript abundance during a time in the daily cycle when CLK-CYC-mediated transcription is high . The overall RNAPII occupancy throughout a daily cycle decreased as we examined the downstream region within the per gene body ( Fig 5A , right panel ) , although the RNAPII we detected may be more reflective of initiated or transiently paused RNAPII . In flies expressing brm RNAi , there was a significant drop in RNAPII occupancy at ZT16 as compared to the parental control at ZT16 ( Fig 5A , middle panel ) , suggesting that the extent of RNAPII pausing at the per TSS was diminished due to decreased nucleosome density when brm was knocked down , which could contribute to the increase in per expression observed in these flies . At the tim locus , we did not observe an increase in RNAPII occupancy at the TSS as compared to the tim E-box 1 or the gene body when comparing RNAPII occupancy at respective time points ( Fig 5B , compare left , middle , and right panels ) , indicating that transient RNAPII pausing may not play as an important role in tim transcription . However , we observed prominent rhythmic RNAPII recruitment that coincides with the gene activation phase ( higher at ZT10 and ZT16 ) in all gene regions ( Fig 5B ) , consistent with earlier findings [26] . Knocking down brm by RNAi significantly decreased RNAPII occupancy at tim E-box 1 and the TSS at ZT16 ( Fig 5B , left and middle panel ) and this difference appeared less pronounced in the gene body ( Fig 5B , right panel ) . The decrease in RNAPII occupancy at the tim E-box and the TSS is also present earlier in the gene activation phase at ZT10 , though not statistically significant . It is intriguing that at ZT16 in brm RNAi flies , both per and tim exhibit strong reductions in initiated or paused RNAPII occupancy at their respective TSS , suggesting a common mechanism contributing to their increased expression levels ( Fig 3 ) . Nonetheless , our data supports previous findings [26] that per and tim transcription , although both activated by CLK-CYC , are regulated through different mechanisms at the chromatin level: per appears to be regulated at the level of RNAPII pausing , perhaps in addition to RNAPII recruitment , whereas rhythmic RNAPII recruitment appears to play a more important role in regulating tim transcription . Future RNAPII ChIP-seq studies can be performed to provide higher resolution insight to the extent of RNAPII stalling on circadian promoters . Besides regulating nucleosome density through its catalytic activity , non-catalytic activities of BRM could also influence RNAPII activity . We therefore assayed RNAPII occupancy and activity at the per and tim loci in flies expressing brmK804R . Interestingly , in the TUG ( II ) control flies used for these experiments , we observed weak ( but not significant ) cycling of RNAPII recruitment on per ( Fig 5C ) , which was not as apparent in the TUG control flies used for comparison with flies expressing brm RNAi ( Fig 5A and 5B ) . RNAPII recruitment for tim remains rhythmic as observed earlier ( Fig 5D ) . This suggests that although it is clear RNAPII pausing regulates per more than tim transcription , temporal changes in RNAPII recruitment may still contribute to the cyclical regulation of per transcription . When comparing RNAPII occupancy in control and flies carrying the brmK804R mutation , we did not observe a significant decrease in RNAPII occupancy at the per TSS ( Fig 5C , middle panel ) , which was very prominent at ZT16 in flies expressing brm RNAi ( Fig 5A , middle panel ) . Similarly , the significant decrease in RNAPII occupancy at the tim promoter ( tim E-box 1 and TSS ) at ZT16 when brm was knocked down by RNAi ( Fig 5B , left and middle panels ) was also absent in flies expressing the brmK804R transgene ( Fig 5D , left and middle panels ) . The fact that the occupancy of initiated and/or paused RNAPII did not show significant changes in flies expressing brmK804R in a nucleosome depleted chromatin landscape , which should be more permissive to RNAPII elongation , supports our hypothesis that the non-catalytic function of BRMK804R have a repressive effect on transcription . However , we should point out that although insignificant , flies expressing brmK804R showed some decrease in RNAPII occupancy in the per TSS at ZT 10 and 16 ( Fig 5C ) and the tim promoter in ZT10 ( Fig 5D ) , similar to the trend observed in brm RNAi flies ( Fig 5A and 5B ) . We therefore suggest that the contribution of the catalytic activity in modulating RNAPII dynamics in circadian genes cannot be ruled out at this point . Curiously , we noticed a consistent significant decrease in RNAPII occupancy in the brmK804R mutant at ZT22 in all regions sampled within per and tim ( Fig 5C and 5D ) , suggesting that the non-catalytic function of BRM may promote the removal of RNAPII at ZT22 at the end of the transcription cycle . Further investigation is required to understand the significance of the decreased RNAPII occupancy at ZT22 . The Brahma complex has previously been implicated in pre-mRNA splicing regulation of developmental genes [54 , 66] . Previous studies have also identified that in Drosophila , per undergoes alternative splicing and there are two transcripts that occur naturally , which affect accumulation of per mRNA [67–68] . These two transcripts differ only by the presence or absence of an alternative intron in the 3’ untranslated region ( UTR ) and increased per splicing leads to higher levels of per mRNA . We therefore sought to investigate if BRM affects the alternative splicing of per in the 3’ UTR , and test the hypothesis that knocking down brm could lead to changes in splicing efficiency and contribute to the change in per mRNA levels observed in flies expressing brm RNAi . Using primers that flank the previously identified alternatively spliced intron in the 3’ UTR of per [68] , we measured the relative abundances of spliced and unspliced transcripts using semi-quantitative PCR and quantified the unspliced transcripts relative to total per mRNA levels ( total of spliced and unspliced transcripts ) . Both unspliced and spliced per levels were normalized using cbp20 as an internal control . While we were able to identify both alternatively spliced transcripts , we did not observe a significant difference in the splicing efficiency in flies expressing brm RNAi knockdown as compared to control flies ( S11 Fig ) , showing that BRM is unlikely to regulate per levels via this splicing event .
The comparison between flies expressing the catalytically-inactive BRMK804R protein and brm RNAi has given us the opportunity to tease apart the catalytic and non-catalytic roles of brm in circadian transcription . Since flies with two distinct genetic manipulations to reduce the endogenous influence of BRM in clock cells both exhibit a reduction in nucleosome density , it was initially surprising that they showed opposite phenotypes with respect to changes in steady state CLK target gene expression ( Fig 3A–3D ) . In particular , our results showing that flies expressing brmK804R have lower expression levels of CLK-dependent clock gene targets as compared to control flies suggests that the interactions of BRM to other proteins , which is retained in BRMK804R , negatively regulates output during the active transcription phase . Since we showed that BRM interacts with CLK , it is possible that BRMK804R is working in a dominant negative manner to limit CLK levels , binding and/or activity to negatively regulate CLK target gene expression . Consistent with this idea , the abundance of CLK was lower in flies expressing BRMK804R and the binding of CLK to the per promoter was also reduced ( Figs 4A and S9A and S9B ) . However , BRMK804R had little to no effect on the levels of CLK binding to the tim promoter ( Fig 4C ) , suggesting that effects on CLK binding to chromatin cannot fully explain the inhibitory effect of BRM on all clock target gene expression . Indeed , there is extensive evidence indicating BRM interacts with many transcriptional regulators , including repressors , as well as RNAPII and associated factors [62 , 71–75] . We postulate that BRM recruits repressive complexes to negatively regulate CLK binding and/or restrain CLK activity during the active transcription phase , and these interactions are precisely balanced by the catalytic function to increase nucleosome density to prevent over-accumulation of repressive complexes or histone marks ( Fig 6A ) . Possible candidates of repressive complexes will be discussed below . With the intact scaffolding function but defective BRM ATPase activity failing to compact chromatin in flies expressing brmK804R , the more open chromatin could augment the non-catalytic function of BRMK804R and lead to over-recruitment of repressive complexes or proteins that reduce CLK stability , resulting in decreased circadian gene transcription ( Fig 3C and 3D , Fig 6B ) . On the other hand , since both the catalytic and non-catalytic functions of BRM are reduced in brm RNAi flies , the repressive complexes normally recruited by BRM would be reduced , thereby leading to an increase in gene expression . Moreover , transcription activators or machineries independent of BRM non-catalytic activity could be more abundant due to the open chromatin , also contributing to the increase in CLK target gene transcription ( Fig 3A and 3B , Fig 6C ) . Since the timing and extent of transcriptional repression at the CLK targets that were tested remained unchanged in flies expressing either brm RNAi or brmK804R ( Fig 3A–3D ) , we suggest that BRM preferentially functions to control the rate of transcription during the daily upswing in CLK-dependent clock gene expression with minimal effects on the initiation and/or maintenance of transcriptional repression , a role likely played by a different class of chromatin remodeler . Furthermore , the fact that the interaction of CLK and BRM as assayed by co-IP in flies appeared to peak at times of active transcription lends further support to a role for BRM in modulating CLK-dependent transcription ( S1 Fig ) . Co-IP in flies also showed that BRM interacts with TIM in the late evening . It is possible that TIM might be involved in terminating BRM function at CLK target genes . Future analysis will be necessary to test this hypothesis . Our results indicate that the role of BRM on Drosophila CLK target genes is different from that of the orthologous SWI/SNF complex in Neurospora , in which it was shown to open up the chromatin at the frq locus to facilitate transcriptional activation [35] . In contrast to SWI/SNF in Neurospora , the ATP-dependent nucleosome remodeling activity of BRM on CLK targets increases nucleosome occupancy , especially during times in a daily cycle when transcriptional output from these genes are peaking . Although at a low amplitude , circadian cycles of BRM-mediated chromatin compaction and relaxation at these promoters can be observed with peak compaction coinciding with times of active transcription even in wild type control flies , and these rhythms were abolished in flies in which BRM catalytic activity was reduced ( Fig 3E–3H ) . In support of our findings indicating an overall repressive role of BRM in limiting transcription output of CLK target gene promoters and its catalytic function to increase nucleosome density , recent genome-wide studies examining BRM function in D . melanogaster larval tissues [76] and primary mouse cells [55] showed that knocking down BRM led to widespread disruption of nucleosome organization , with a bias towards a decrease in nucleosome density , especially at promoters and peri-TSS regions . Furthermore , although the relationship between loss of nucleosome density upon BRM knockdown and changes in gene transcription is highly variable , more genes appeared to be significantly upregulated than downregulated , thus corresponding with our findings in Drosophila CLK target genes . The variable outcome from loss of nucleosome density could be explained by the specificity of BRM to recruit either activators or repressors on a gene-by-gene basis . Our model proposes a complex scheme in which the catalytic activity of chromatin remodelers such as BRM can function like a rheostat to adjust the non-catalytic function to control factor recruitment and the rate of transcription in an intricate manner . Over the years , BRM has been identified to play a role in both transcriptional activation and repression [41 , 56 , 57] , and it has become increasingly evident that whether it acts as an activator or repressor highly depends on the proteins recruited through its non-catalytic activity . Based on our results , BRM appears to recruit repressive complexes to negatively regulate CLK activity and binding to target genes . In many genes including clock genes , dynamic deacetylation of histones in the promoter region has been implicated in the repression of gene expression ( reviewed in [24–25] ) . Histone deacetylation is catalyzed by a large class of histone deacetylases ( HDACs ) , and many HDACs and other epigenetic modifiers have been found to be under circadian control as well as having direct interactions with clock proteins . The transcriptionally repressive Sin3-HDAC complex , which is an evolutionarily conserved protein complex that includes HDAC1 and HDAC2 [77] , has been found to co-precipitate with PER complexes and aid in the repression of circadian transcription [78] . BRG1 ( mammalian homolog of Drosophila BRM ) and other components of the SWI/SNF complex have also been found to co-immunoprecipitate with components of Sin3-HDAC [72–74] , showing precedence in direct association of BRM ( SWI/SNF ) with proteins involved in histone deacetylation and gene repression . In addition to interacting with HDAC1 and HDAC2-containing complexes , BRG1 and other Brahma complex related proteins have also been observed to associate with a corepressor complex N-CoR-1 , which contains HDAC3 [71] , The accumulating evidence of interactions of the BRM complex with repressive factors provides support to our model in which we postulate that BRM is involved in limiting CLK target gene output . Changes in nucleosome occupancy have been known to regulate transcription factor binding [70] and differential transcription factor binding has the potential to affect transcriptional output . We therefore sought to examine CLK and PER binding at per and tim promoters in the two different types of brm knockdown ( Fig 4 ) . Although we found a decrease in CLK binding in flies expressing brmK804R that could partly explain the reduced transcriptional output at CLK target genes ( Fig 4A ) , we found no significant differences in CLK and PER binding in flies expressing brm RNAi even though they showed a large increase in transcriptional output at CLK target genes ( Fig 4B , 4D , 4F and 4H ) . We therefore explored whether changes in chromatin organization in these flies can influence RNAPII dynamics [79–80] to increase CLK target gene expression . Using RNAPII ChIP , we found that the increased nucleosome density mediated by BRM , at times when CLK-mediated transcription is peaking ( i . e . , ZT16 ) , appears to be associated with an increase in transient RNAPII stalling near the per TSS ( Fig 5A and 5C ) . This was not observed in the case of the tim TSS as RNAPII dynamics at tim appears to be more heavily regulated through RNAPII recruitment , rather than transient stalling ( Fig 5B and 5D ) . The lack of transient stalling at the per promoter observed in flies expressing brm RNAi could be due to the decrease in nucleosome density and increase in RNAPII elongation into the gene body resulting in higher overall transcription output ( Fig 5A ) . This scenario is supported by the fact that there is also a general reduction in RNAPII occupancy in the tim promoter and TSS ( Fig 5B ) in the brm RNAi flies that could also lead to increased mRNA output . Originally , we had anticipated that we might observe an increase in RNAPII occupancy in the gene body that corresponds with a decrease in RNAPII occupancy at the per/tim E-boxes and TSS in flies expressing brm RNAi , which would reflect the increased transition of initiated/paused RNAPII at the TSS into active elongation in the gene body . This was not observed , and could be due to the fact that the RNAPII antibody we used has a higher affinity for initiated/paused RNAPII , rather than elongating RNAPII . In addition to changes in nucleosome density , it is worthy to point out that there could be additional mechanisms that regulate the observed transient stalling at the per TSS , including the recruitment of pausing factors via the non-catalytic function still retained in BRMK804R . In Drosophila , it has been shown that BRM and a known interactor SAYP facilitates the formation of nucleosome-dense regions that acts as barriers to RNAPII , leading to stalling during the repressive phase of transcription [61] . Elimination of SAYP and BRM results in loss of this nucleosomal barrier , thereby releasing RNAPII and leading to an increase in transcription . These findings corroborate with our observation of the lack of RNAPII stalling at the per promoter in brm RNAi mutants . Stalling could be unaffected in flies expressing brmK804R because BRMK804R still retains the ability to interact with pausing factors such as SAYP . Altogether , our results indicate that BRM functions to fine-tune CLK-CYC-mediated expression of core clock genes by acting as a braking mechanism when transcriptional activation is in full effect . It appears that Drosophila circadian transcription during the active phase may be operating under a largely restrictive chromatin landscape , and this mechanism may be important in maintaining precise levels of cyclical gene expression . Our data suggests that this fine-tuning occurs through the coupling of the catalytic and non-catalytic functions of BRM , thereby generating a balanced chromatin landscape for transcription factor binding , transient RNAPII stalling , and possible scaffolding interactions with histone modifiers such as HDACs . Future experiments will aim at clarifying the interaction between the Brahma complex and histone modifiers , as well as the possible role of brm in regulating the expression of clock genes such as Clk that cycle in anti-phase to those that are direct targets of CLK-CYC .
Targeted RNAi knock down of Brahma ( Brm ) subunits in circadian clock neurons was achieved using the UAS/GAL4 system [81] . To knock down individual subunits , virgin female flies from UAS-RNAi responder lines were crossed to males from driver line w; UAS-dicer2; tim-UAS-GAL4 ( referred to as TUG ) [45] to achieve knockdown in tim-expressing neurons . Up to three independent UAS-RNAi responder lines targeting each gene were used . UAS-dicer2 was included to increase the efficiency of RNA interference . Male progenies of the crosses were then assayed for locomotor activity . Both male and female progenies were used for RNA and protein assays . Targeted knockdown of brm in PDF neurons ( a subset of clock neurons ) was achieved with the use of a pdf-GAL4 driver line ( obtained from P . Hardin ) . To rule out developmental effects caused by brm RNAi leading to clock defects , we utilized the temperature sensitive GAL80 mutant ( Bloomington Drosophila Stock center stock number 7108 ) to inhibit GAL4 expression during development . Progenies from crosses were placed into incubators set at 18°C to inhibit expression of dsRNA during development and transferred to 29°C to relieve the GAL4 inhibition three days before behavioral assays . To generate brm overexpressing flies , 3XFLAG-6XHis-brm was cloned into pUAST vector ( Addgene ) . This plasmid was injected into w1118 embryos by Genetic Services , Inc ( Sudbury , MA ) . Transgenic flies carrying the UAS-FLAG-brm transgene were then crossed to w; tim- ( UAS ) -GAL4 driver line ( referred to as TUG ( II ) ) to obtain expression of FLAG-brm in clock neurons . Male flies around 3 to 5 days old were subjected to locomotor activity assays using the Drosophila Activity Monitoring System ( DAMS ) ( Trikinetics , Inc . ) . Flies were entrained for four days at 12 hr light:12 hr dark ( LD ) conditions at 25°C before their free-running behavioral rhythms were assessed in total darkness ( DD ) for seven days . Fly activity monitoring using DAMS and data analysis using FaasX were as previously described [82] . Behavioral assays to rule out developmental effects of brm RNAi were performed at 29°C . For co-immunoprecipitation ( co-IP ) in Drosophila S2 cells , brm cDNA clones were obtained from Drosophila Genomics Resource Center ( DGRC ) and processed according to the stock center protocol . The brm ORF was amplified from cDNA reverse transcribed from total RNA extracted from fly heads without the stop codon using PCR and the PCR product was subcloned into pAc-3XFLAG-6XHis [10] such that the ORF of brm is in frame and located at the N terminal end of the FLAG epitope . The plasmids expressing pAc-per-V5-His and pAc-tim-3HA were previously described [83] . The pAc-clk-V5-His plasmid was generated by restriction digesting clk from pMT-HA-clk-V5 [84] and subcloned into pAc-V5-His vector . We generated α-PER ( GP5620 ) by using PCR to amplify per cDNA sequences that encode amino acids 232 to 599 and cloned the PCR fragment upstream of sequences that encode a polyhistidine stretch ( His ) in the expression vector pET28b ( Novagen ) . Fusion protein expression , purification , and antibody production were performed as previously described [85] . Antibodies to TIM ( R3 ) used in our experiments were as described [85] . Commercially available antibodies were purchased for CLK ( Santa Cruz Biotechnology H3107 ) , FLAG ( Sigma F3165 ) , histone H3 ( Abcam ab1791 ) , and RNAPII ( Abcam ab5408 ) . The RNAPII antibody used for ChIP recognizes RNAPII with phosphorylated CTD repeat ( Serine 2 and Serine 5 ) with preference to Serine 5 . Antibody dilutions are listed below . To perform co-IP assays , 3 x 106 S2 cells were transiently transfected with pAc-brm-3XFLAG-6XHis in combination with pAc-per-V5-His , pAc-tim-3HA , or pAc-clk-V5-His using the Qiagen Effectene Transfection reagent following manufacturer’s protocol . 48 hours after transfection , cells were harvested for protein extraction using M-RIPA buffer ( 20 mM Tris-HCl at pH 7 . 5 , 150 mM NaCl , 1 mM EDTA , 10% glycerol , 1% Triton X-100 , 0 . 4% DOC , 0 . 1% SDS , 0 . 5 mM PMSF , 10 μg/μl aprotinin , 10 μg/μl leupeptin , 2 μg/μl pepstatin A , 25 mM NaF , and 1x Roche protease inhibitor ) . Protein extracts were quantified and either directly analyzed by immunoblotting ( input lysate ) or aliquoted for IP . BRM IP samples were incubated with α-FLAG M2 affinity gel ( Sigma ) . In addition , reciprocal IP and negative control IP samples were incubated with either α-V5 agarose beads ( Sigma ) or α-HA agarose ( Sigma ) depending on the clock protein ( PER , TIM , or CLK ) that is coexpressed with BRM . Following 4 hours of incubation at 4°C , beads were washed a total of three times ( 10 minutes per wash ) in a M-RIPA buffer with increased salt ( 300 mM NaCl ) . Immune complexes were eluted from beads by adding 2X SDS-PAGE sample buffer and resolved using SDS-PAGE . For protein visualization , α-FLAG ( 1:7000 ) was used to detect BRM-FLAG , α-V5 ( 1:5000 ) was used to detect PER and CLK . Secondary antibody for α-FLAG and α-V5 detection was α-mouse IgG-HRP at 1:2000 ( GE Healthcare ) . α-HA ( 1:1000 ) was used to detect TIM with secondary antibody α-rat IgG at 1:1000 ( GE Healthcare ) . Flies were entrained for three full days in 12 hr light:12 hr dark ( LD ) conditions at 25°C and collected on the fourth day at the indicated time points ( ZT ) and frozen immediately on dry ice . Heads were separated using frozen metal sieves and homogenized in 3x volume of RBS buffer ( 20 mM HEPES at pH 7 . 5 , 50 mM KCl , 10% glycerol , 2 mM EDTA , 1 mM DTT , 1% Triton X-100 , 0 . 4% NP-40 , 10 μg/mL aprotinin , 5 μg/mL leupeptin , 1 μg/mL pepstatin A , 0 . 5 mM PMSF , 1X PhoStop ( Roche ) ) [46 , 51] . Homogenate was sonicated using a Fisher Scientific sonicator for five seconds and repeated five times with 10-second pauses in between . Samples were spun down at 14 , 000 rpm for 15 minutes at 4°C to remove cell debris . Supernatant was collected , transferred to new tubes , and spun down again for 10 minutes at 14 , 000 rpm at 4°C . Supernatant was collected and protein levels were quantified using a spectrophotometer ( Eppendorf ) . Proteins were resolved by SDS-PAGE ( Criterion 5% gels , Bio-Rad ) , transferred to nitrocellulose membranes ( Bio-Rad ) and incubated in 5% blocking solution ( Bio-Rad ) in 1XTBST with α-PER ( GP5620 ) ( 1:2000 ) , α-TIM R3 ( 1:2000 ) , or α-CLK ( Santa Cruz ) ( 1:1000 ) . Membranes were imaged and protein levels were quantified using the ChemiDoc MP system with Image Lab software ( Bio-Rad ) . Transgenic flies expressing BRM fused to FLAG epitope tags were entrained in 12 hr light:12 hr dark ( LD ) conditions at 25°C for three days and collected on the fourth day . Fly head collection and protein extraction with RBS buffer with sonication were performed as described above . Extracts were quantified and equal concentrations were subjected for IP . Samples were pre-cleared using sepharose beads ( Sigma ) to reduce nonspecific binding . Co-IPs were performed as described for S2 cell experiments except α-FLAG M2 ( Sigma ) , α-PER ( GP5620 ) , α-TIM ( R3 ) , and α-CLK ( Santa Cruz Biotechnology H3107 ) antibodies were used . Samples were incubated with antibodies for 4 to 6 hours at 4°C on an end-over-end rotator . 20 μl of GammaBind Plus sepharose beads ( GE ) was added and incubation was continued for 2 hours . Samples were washed with RBS buffer three times , 10 minutes each , and immune complexes were resolved by SDS-PAGE as described above . Flies were entrained in 12 hr light:12 hr dark ( LD ) conditions at 25°C for three days and collected at four or six time-points on the fourth day . At least 40 flies of each genotype were collected per time point . Heads were collected using metal sieves . Total RNA extraction and quantitative PCR gene expression analysis were performed as previously described [86] . Gene-specific primers for per , tim , vri , and brm were designed to amplify fragments of around 150 bp near the 3’ end of the coding sequence for each target gene and optimized at an annealing temperature of 60°C . Primers for clk and pdp1ε were previously described [4] . Representative results are shown in Fig 3 , S5 Fig , S6A Fig and S8 Fig . Additional biological replicates for expression analysis are presented in S12 Fig . Nascent RNA was extracted as described [87–88] with modifications . Flies were entrained in LD conditions at 25°C for three days and collected at four time points on the fourth day . Fly heads were collected on dry ice . At least 300 μl of fly heads were used for nascent RNA extraction . Fly heads were homogenized into a fine powder using a liquid nitrogen chilled ceramic mortar and pestle , and mixed in 1 . 8 ml of homogenization buffer ( 10 mM Tris-HCl at pH 7 . 6 , 10 mM KCl , 1 . 5 mM MgCl2 , 0 . 8 M sucrose , 0 . 5 mM EDTA , 1 mM DTT , 1x protease inhibitor ) . Samples were dounced 15 times on ice with the loose pestle . The resulting lysate was then filtered through a 100 μm cell strainer in a 50 ml falcon tube and centrifuged at 300 g for 2 minutes . 700 μl of the supernatant was carefully removed and the remaining supernatant and pellet were resuspended and layered over 900 μl of sucrose cushion buffer ( 10 mM Tris-HCl at pH 7 . 5 , 10 mM KCl , 1 . 5 mM MgCl2 , 1 M sucrose , 10% glycerol , 0 . 5 mM EDTA , 1 mM DTT , 1x protease inhibitor ) . Samples were spun at 11 , 000 rpm for 10 minutes . Pellets were resuspended in 1 ml of lysis buffer ( 20 mM Tris-HCl at pH 7 . 6 , 150 mM NaCL , 2 mM EDTA , 1x protease inhibitor , 0 . 5 mM PMSF , 1 mM DTT , 0 . 5 U/ml RNAseOUT/SUPERase-In ) and dounced 5 times with the tight pestle . After douncing , 1 ml of 2xNUN buffer ( 50 mM Tris-HCl at pH 7 . 6 , 2M Urea , 2% NP-40 , 600 mM NaCl , 2 mM DTT , 1x protease inhibitor , 0 . 5 mM PMSF , 0 . 5 U/ml SUPERase-In ) was added drop-by-drop while gently vortexing . Samples were incubated on ice for 20 min , then centrifuged at 14 , 000 rpm for 30 min . Supernatant was removed and 500 μl of TRI Reagent ( Sigma ) was added to the pellet . Samples were incubated at 65°C for 15 min , then the DNA pellet was resuspended by gentle pipetting . Extraction using TRI Reagent , cDNA synthesis , and qPCR analysis ( for comparison of control vs . flies expressing brm RNAi ) was performed following previously described protocol [86] . After nascent pre-mRNA isolation and cDNA synthesis , samples from TUG ( II ) and BRMK804R were diluted in nuclease-free water and ~10 ng of cDNA template was subjected to ddPCR . EvaGreen supermix reagent ( Bio-Rad ) was used following manufacturers protocol , and the QX200 Droplet Generator ( Bio-Rad ) was used to create 20 , 000 individual droplets in each reaction . Droplets were subjected to end-point PCR performed following manufacturers recommended cycling conditions . Primers for per and tim used for ddPCR were previously optimized for qPCR . Amplification of cbp20 was used for normalization . Individual droplet fluorescence was measured on a QX200 ddPCR Droplet Reader ( Bio-Rad ) and analysis was performed using QuantaSoft software ( Biorad ) . Technical triplicates from two biological replicates were performed . Data presented are unscaled expression levels normalized to cbp20 expression . Error bars = SEM for biological replicates . Chromatin immunoprecipitation ( ChIP ) was performed based on published protocols [46 , 51] with modifications . All buffers describe below , except CE buffer , contain protease inhibitors , 10 μg/μl aprotinin , 5 μg/μl leupeptin , 1 μg/μl pepstatin , and 0 . 5 mM PMSF . Flies entrained in 12 hr light:12 hr dark ( LD ) conditions at 25°C for three days were collected at four time-points ( ZT ) on the fourth day . 300 μl of fly heads were homogenized into a fine powder using a liquid nitrogen chilled ceramic mortar and pestle , mixed with 1 . 8 ml of NEB buffer ( 10 mM Tris-HCl at pH 8 . 0 , 10 mM NaCl , 0 . 1 mM EGTA at pH 8 . 0 , 0 . 5 mM EDTA at pH 8 . 0 , . 1 mM DTT , 0 . 5% Tergitol NP-10 , 0 . 5 mM spermidine , 0 . 15 mM spermine , and 1x protease inhibitor ( Sigma ) ) , and homogenized with a dounce homogenizer ( Kimble Chase ) for 20 strokes using the loose “A” pestle . Homogenate was transferred to a 70 μm cell strainer placed in a 50 ml falcon tube and centrifuged at 300 g for 1 minute . The filtered homogenate was centrifuged at 6 , 700 rpm for 10 minutes to further remove cell debris . Pellets were resuspended in 1 ml of NEB and centrifuged at 11 , 500 rpm for 20 minutes on a sucrose gradient ( 0 . 6 ml of 1 . 6 M sucrose in NEB , 0 . 35 ml of 0 . 8 M sucrose in NEB ) . The nuclei-containing pellet was resuspended in 1 ml of NEB with 1% formaldehyde ( diluted in Drosophila Schneider’s media ( Life Technologies ) ) and cross-linked for 10 minutes at room temperature with rotation . Crosslinking was quenched by adding 150 μl of 1 M glycine and samples were rotated for 5 minutes at room temperature . Nuclei were collected by centrifugation at 6 , 700 rpm for 5 minutes , washed 2x with 1 ml NEB , and resuspended in 350 μl of sonication buffer ( 10 mM Tris-HCl at pH 7 . 5 , 2 mM EDTA , 1% SDS , 0 . 2% Triton X-100 , 0 . 5 mM spermidine , 0 . 15 mM spermine , and 1x protease inhibitor cocktail ( Sigma ) ) . Samples were sonicated 3x using a Diagenode Bioruptor on high setting for 5 minutes at 30 seconds on/off and then centrifuged at 10 , 000 rpm for 10 minutes . Supernatant was collected in two 130 μl aliquots for IP and 26 μl was collected for input and frozen at -80C for analysis . Sonicated chromatin was roughly 500 bp in length ( <1000 bp ) . For each IP , 25 μl of a Protein G Dynabead slurry ( Life Technologies ) was washed twice in 75 μl of CW Buffer ( 50 mM Tris-HCl at pH 7 . 6 , 1 mM EDTA , 1% Triton X-100 , 0 . 1% DOC , 0 . 1% BSA , 0 . 5 M KCl in PBS , 150 mM NaCl , 0 . 5 M EGTA , 0 . 1% SDS , and 1x protease inhibitor ( Sigma ) ) . Beads were captured using a magnetic stand ( Millipore ) to allow for buffer removal . After the last wash , 75 μl of CW buffer was added to the beads along with the appropriate antibody and incubated with rotation for 2 hours at 4°C . Amount of antibodies used for ChIP is as follows: α-PER ( GP5620 ) ( 20 μg/ml ) ; α-CLK ( 15 μg/ml ) ; α-H3 ( 10 μg/ml ) ; α-RNAPII ( 10 μg/ml ) , α-FLAG ( 10 μg/ml ) . Following incubation , beads were collected and resuspended in 22 μl of CW buffer . 20 μl of this slurry was added to sonicated chromatin IP aliquots that were diluted 10-fold with IP buffer ( 50 mM Tris-HCl at pH 7 . 6 , 2 mM EDTA , 1% Triton X-100 , 0 . 1% DOC , 150 mM NaCl , 0 . 5 mM EGTA , and 1x protease inhibitor ) and incubated for 2 hours at 4°C . Beads were captured and washed for 30 minutes 2x in 1 ml of CW buffer , once in LW buffer ( 10 mM Tris-HCl at pH 8 . 0 , 0 . 25 M LiCl , 0 . 5% NP40 , 0 . 5% DOC , 1mM EDTA ) , and once in TE buffer for 4 minutes . Contents were then transferred to new LoBind tubes . Supernatant was removed and 150 μl of CE buffer ( 50 mM Tris-HCl at pH 8 . 0 , 10 mM EDTA , 1% SDS , 1 mM DTT , 0 . 1 mg/ml proteinase K , 50 mM NaCl , and 0 . 05 mg/ml RNase A ) was added . CE buffer ( 150 μl ) was also added to input samples . All samples were incubated for 2 hours at 37°C . Beads were then removed from IP samples and supernatant was de-crosslinked overnight at 65°C . DNA was eluted using the Qiagen PCR purification kit and subjected to qPCR . At least three technical replicates of qPCR were performed for each biological ChIP replicate and three biological replicates were performed for CLK , PER , H3 , and RNAPII assays . Background binding to a non-specific antibody ( α-V5; Life Technologies ) at 10μg/ml bound to Dynabeads was subtracted from input samples and results are presented as the percentage of the input samples . For each assay , at least three biological replicates were performed , with technical triplicates for the qPCR step for each biological replicate . The technical qPCR triplicates were averaged for each biological replicate as no significant differences were found between the technical replicates , and the error bars represent SEM calculated from variance between biological replicates . Two-tailed t-tests were used to determine statistical differences ( P < 0 . 05 ) between control and experimental treatment at each ZT . To detect splicing efficiency in period , cDNA was generated using methods described in [86] and used as DNA templates for semi-quantitative PCR as described in [67] . The PCR program was set to run for 23 cycles to ensure that amplicons were still in the log-linear phase of amplification . Primers designed to flank the 8th intron in the 3’UTR of per were used to assay splicing efficiency between WT and brm RNAi mutants . For normalization , we also included primers that amplify the non-cycling cbp20 gene . PCR products were separated and visualized by gel electrophoresis on 2% agarose gels by staining with Gelstar ( Cambrex Co . ) and DNA bands were quantified using a ChemiDoc MP with Image Lab software ( Bio-Rad ) . | The circadian clock is an endogenous timing system that enables organisms to anticipate daily changes in their external environment and temporally coordinate key biological functions that are important to their survival . Central to Drosophila clockwork is a key transcription factor CLOCK ( CLK ) . CLK activates expression of target genes only during specific parts of the day , thereby orchestrating rhythmic expression of hundreds of clock-controlled genes , which consequently manifest into daily rhythms in physiology and behavior . In this study , we demonstrated that the Brahma ( Brm ) chromatin-remodeling protein interacts with CLK and fine-tune the levels of CLK-dependent transcription to maintain the robustness of the circadian clock . Specifically , we uncovered two distinct but collaborative functions of Brm . Brm possesses a non-catalytic function that negatively regulates the binding of CLK to target genes and limits transcriptional output , likely by recruiting repressive protein complexes . Catalytically , Brm functions by condensing the chromatin at CLK target genes , specifically when transcription is active . This serves to precisely control the level of repressive factors likely recruited by Brm as well as other transcriptional regulators . By disentangling these two roles of Brm , our study uncovered a multi-layered mechanism in which a chromatin remodeler regulates the circadian clock . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | The Catalytic and Non-catalytic Functions of the Brahma Chromatin-Remodeling Protein Collaborate to Fine-Tune Circadian Transcription in Drosophila |
The identification of recurrent gene fusions in common epithelial cancers—for example , TMPRSS2/ERG in prostate cancer and EML4/ALK in nonsmall cell lung carcinomas—has raised the question of whether fusion genes are pathogenetically important also in ovarian carcinomas . The first recurrent fusion transcript in serous ovarian carcinomas was reported by Salzman et al . in 2011 , who used deep paired-end sequencing to detect the fusion gene ESRRA–C11orf20 in 10 out of 67 ( 15% ) serous ovarian carcinomas examined , a finding that holds great promise for our understanding of ovarian tumorigenesis as well as , potentially , for new treatment strategies . We wanted to test how frequent the ESRRA/C11orf20 fusion is in ovarian carcinomas of all subtypes , and therefore examined a series of 230 ovarian carcinomas of which 197 were of the serous subtype and 163 of the 197 were of stages III and IV—that is , the very same carcinoma subset where the fusion transcript had been found . We performed PCR and high-throughput sequencing analyses in search of the fusion transcript . We used the same primers described previously for the detection of the fusion and the same primer combination , but found no ESRRA/C11orf20 fusion in our series . A synthetic DNA plasmid containing the reported ESRRA/C11orf20 fusion was included as a positive control for our PCR experiments . Data from high-throughput sequencing of 23 ovarian carcinomas were screened in search of alternative partner ( s ) for the ESRRA and/or C11orf20 gene , but none was found . We conclude that the frequency of the ESRRA/C11orf20 gene fusion in serous ovarian carcinomas of stages III and IV must be considerable less than that reported previously ( 0/163 in our experience compared with 10/67 in the previous study ) . At the very least , it seems clear that the said fusion cannot be a common pathogenetic event in this tumor type .
Cancer of the ovary makes up 30% of all malignant diseases of the female genital tract . Prognosis is poor , with a mean 5-year survival rate in Europe of 32% . This unfavorable outcome is largely attributable to a lack of early warning symptoms and signs and also a lack of diagnostic tests that allow early detection . As a result , approximately 70% of patients present with advanced stage , metastatic disease [1] . A number of specific genes have been identified as playing a role in ovarian carcinogenesis; the ones that have received the most attention are BRCA1 and BRCA2 followed by TP53 . In addition , integrated genomic analysis of ovarian carcinomas has identified four ovarian cancer transcriptional subtypes , three microRNA subtypes , four promoter methylation subtypes , and a transcriptional signature associated with survival duration [2] , attesting to the genetic complexity of these tumors . The identification of recurrent gene fusions in common epithelial cancers—for example , TMPRSS2/ERG in prostate cancer [3] and EML4/ALK in nonsmall cell lung carcinomas [4] , [5]—has raised the question of whether fusion genes are pathogenetically important also in ovarian carcinomas . Salzman et al . [6] reported the first recurrent fusion transcript in serous ovarian carcinomas . They used deep paired-end sequencing to detect the fusion gene ESRRA–C11orf20 in 10 out of 67 ( 15% ) serous ovarian carcinomas examined , a finding that holds great promise for our understanding of ovarian tumorigenesis as well as , potentially , for new treatment strategies . The fusion was brought about by rearrangements in the long arm of chromosome 11 , in subband 11q13 . 1 . The gene ESRRA ( estrogen-related receptor alpha ) encodes a nuclear receptor that is closely related to the estrogen receptor , whereas its partner is but an open reading frame sequence . Because ESRRA and C11orf20 ( also known as TEX40 ) normally lie only 11 kb apart , it is possible that the rearrangement leading to their fusion is an incidental consequence of another functionally important genetic event or that it is merely a “passenger” to other structural rearrangements . To test how frequent ESRRA/C11orf20 fusion is in ovarian carcinomas of all subtypes , we performed PCR analysis of 230 ovarian carcinomas , of which 197 were of the serous subtype and 163 of the 197 were of stages III and IV—that is , the very same carcinoma subset examined by Salzman et al . [6] .
The PCR analysis of the 230 ovarian carcinomas showed no fusion transcript for the ESRRA/C11orf20 . A synthetic DNA plasmid containing the reported ESRRA/C11orf20 fusion was included as a positive control for our PCR experiments and was the only sample showing the transcript and demonstrating , at the same time , the validity of the experiments ( Figure 1 ) . We also performed high-throughput sequencing of 23 ovarian carcinomas ( already tested by PCR analysis ) , of which 10 were serous , five endometrioid , four clear cell , three mucinous , and one of a mixed endometrioid and undifferentiated subtype . Each sample was sequenced to yield about 60∼70 million reads using the Illumina HiSeq 2000 instrument . We extracted from the raw data all sequences containing the last 20 bp before the putative break of the ESRRA exon 2 gene sequence , getting 2 , 705 reads in total . We also found 58 , 59 , and 49 reads containing the first 20 bp of the C11orf20 exon 3 , exon 4 , and exon 5 gene sequences , respectively ( Table 1 ) . From the extracted ESRRA- and C11orf20-specific sequences , none contained sequences of both ESRRA and C11orf20 . The comparison was performed by investigating if the ESRRA-specific sequences contained C11orf20 exon 3 , 4 , or 5 sequences and vice versa . It is possible to argue that the fusion gene , if present , should be driven by the ESRRA promoter , and therefore that the fusion gene read counts should be more similar to the high ESRRA ones than to the low C11orf20 ones . As a result , assuming the presence of the fusion , the C11orf20 reads should have been totally dominated by the fusion , something that was not seen ( Table 1 ) . Furthermore , all 2 , 705 sequences were used in a Blast search to verify their identity . The Blast search identified specific ESRRA and C110rf20 sequences but revealed no sequences containing both ESRRA and C11orf20 gene sequences . When searching in the same series of sequenced carcinomas ( n = 23 ) for involvement of either the ESRRA or C11orf20 in alternative fusions—that is , with other partner ( s ) —none was found . We therefore conclude that the frequency of the ESRRA/C11orf20 gene fusion in serous ovarian carcinomas of stages III and IV must be considerable less than that reported by Salzman et al . ( 0/163 in our experience , compared with 10/67 in their study ) [6] . We have no explanation for the frequency differences observed . It is important to note that the difference in frequency calculated above is based only on adenocarcinomas of stages III and IV—that is , 163 tumors—as the remaining 67 tumors were of different histological subtypes or serous adenocarcinomas of grades I and II , which would not necessarily be expected to carry the same genetic fusion . Looking into the possible mechanism—that is , chromosomal rearrangement ( s ) —by which the ESRRA/C11orf20 fusion could have originated , it seems that a simple deletion or inversion could not alone have produced it . Both genes are located 5′ to 3′ from centromere to telomere on 11q , with C11orf20 proximal to ESRRA ( Figure 1 ) ; therefore , to get a fusion in which ESRRA is 5′ in a chimeric transcript would require a tandem duplication with a breakpoint in the central region . Regardless of how it may have been generated , however , it seems clear that the said fusion cannot be a common pathogenetic event in this tumor type .
The tumors were surgically removed at The Norwegian Radium Hospital from 1999 to 2010 . The RNA was extracted using Trizol reagent according to the manufacturer's instructions ( Invitrogen , Grand Island , NY ) , and its quality was checked by Experion Automated Electrophoresis System ( Bio-Rad Laboratories , Hercules , CA ) . cDNA was synthesized using the Iscript advanced cDNA synthesis kit for RT-qPCR ( Bio-Rad ) . Quality was checked using the TaqMan Gene Expression Assays for actin B ( ACTB ) and Glyceraldehyde 3-phosphate dehydrogenase ( GAPDH ) . To measure the expression of ACTB and GAPDH , the assays Hs99999903_m1 and Hs99999905_m1 , obtained from Applied Biosystems ( Life Technologies , Carlsbad , CA ) , were used and run on a CFX96 Real-Time System ( Bio-Rad ) . For the first RT-PCR reaction we used the G1P1-FWD ( 5′-GGCATTGAGCCTCTCTACATCA-3′ ) mapping between 240 and 261 bp in the ESRRA gene ( accession number NM_004451 version 4 ) and REV_pair3 ( 5′-GGGTCAGGCTTGGGTCTG -3′ ) located between 681 and 698 bp of the C11orf20 ( accession number NM_001039496 version 1 ) combination of primers—that is , the same primers as Salzman et al . [6] . The PCR cycles were as follows: initial denaturation at 94°C for 30 s , followed by 30 cycles , 15 s at 94°C , 30 s at 55°C; and 60 s at 70°C [6] . For the nested RT-PCR , the primers were G1P2-FWD ( 5′-AAAGGGTTCCTCGGAGACAGAGA-3′ ) located between 290 and 312 base pairs in the ESRRA gene ( accession number NM_004451 version 4 ) and F1-REV ( 5′-TAATTCACGTACAGCCTCTTGCTCCG-3′ ) mapping between 597 and 622 bp of the C11orf20 gene ( accession number NM_001039496 version 1 ) [6] . The cycles were as follows: 15 s at 94°C , 30 s at 55°C , and 60 s at 72°C [6] . The nested PCR was run for 30 cycles . A synthetic DNA plasmid containing the reported ESRRA/C11orf20 fusion was included as a positive control in our PCR experiments . High-throughput sequencing was performed on 23 ovarian carcinomas , of which 10 were serous , five endometrioid , four clear cell , three mucinous , and one of a mixed endometrioid and undifferentiated subtype . A total of 3 µg of RNA was sent for high-throughput pair-end RNA-sequencing to the Norwegian Sequencing Centre at Ullevål Hospital ( www . sequencing . uio . no ) . We used paired-end HTS with an average sequence read of 60–70 millions . We analyzed the sequences only with respect to the ESRRA and C11orf20 genes . The last 20 bp of the ESRRA exon 2 gene sequence before the putative break and the first 20 bp of the C11orf20 exon 3 , exon 4 , and exon 5 gene sequences have been extracted from the raw data ( fastq-files ) and further analyzed for putative gene fusions . | The identification of characteristic fusion genes in cancer helps us to understand how a particular cancer arises and also to improve classification and diagnosis , with a view to develop specific medical treatments that target exactly those aberrant molecules that trigger the disease . A fusion transcript presumed to arise from a chromosomal rearrangement involving the ESRRA and C11orf20 genes has previously been described to be present in 15% of serous ovarian carcinomas—the first fusion transcript to be associated with this common and often fatal cancer . We assessed 163 similar ovarian carcinomas for the presence of the ESRRA–C11orf20 fusion transcript , plus a further 67 ovarian carcinomas of different histologic subtypes/grades , to see if these tumors were characterized by the same fusion . Surprisingly , we found no ESRRA–C11orf20 transcripts in any of the 230 carcinomas . The question as to whether fusion genes contribute to the pathogenesis of ovarian carcinoma therefore remains open . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"oncology",
"medicine",
"cancer",
"genetics",
"ovarian",
"cancer",
"gynecological",
"tumors",
"genetics",
"biology",
"cancers",
"and",
"neoplasms"
]
| 2014 | Low Frequency of ESRRA–C11orf20 Fusion Gene in Ovarian Carcinomas |
Hair is important for thermoregulation , physical protection , sensory activity , seasonal camouflage , and social interactions . Hair is generated in hair follicles ( HFs ) and , following morphogenesis , HFs undergo cyclic phases of active growth ( anagen ) , regression ( catagen ) , and inactivity ( telogen ) throughout life . The transcriptional regulation of this process is not well understood . We show that the transcription factor Lhx2 is expressed in cells of the outer root sheath and a subpopulation of matrix cells during both morphogenesis and anagen . As the HFs enter telogen , expression becomes undetectable and reappears prior to initiation of anagen in the secondary hair germ . In contrast to previously published results , we find that Lhx2 is primarily expressed by precursor cells outside of the bulge region where the HF stem cells are located . This developmental , stage- and cell-specific expression suggests that Lhx2 regulates the generation and regeneration of hair . In support of this hypothesis , we show that Lhx2 is required for anagen progression and HF morphogenesis . Moreover , transgenic expression of Lhx2 in postnatal HFs is sufficient to induce anagen . Thus , our results reveal an alternative interpretation of Lhx2 function in HFs compared to previously published results , since Lhx2 is periodically expressed , primarily in precursor cells distinct from those in the bulge region , and is an essential positive regulator of hair formation .
Hair develops in hair follicles ( HFs ) and embryonic development of HFs ( morphogenesis ) is initiated between embryonic days 12 ( E12 ) to E15 , and is regulated by continuous epithelial-mesenchymal interactions between epidermal and dermal cells [1]–[3] . The first morphological sign of HF morphogenesis is a thickening of the epithelial cell layer leading to the formation of a placode . The placode induces a dermal condensation in the underlying mesenchyme that signals to placodal cells to proliferate and grow down into the dermis forming the primary hair germ ( HG ) , followed by the formation of the hair peg . Subsequently , at the bulbous peg stage , the dermal condensate forms the dermal papilla ( DP ) as it becomes enveloped by follicular epithelial cells [2] . At this stage the proliferating and differentiating epidermal matrix cells adjacent to the DP start to generate the different layers of the HF consisting of the medulla , cortex and cuticle of the hair shaft , and the cuticle , Huxley's layer and Henle's layer of the inner root sheath ( IRS ) . The IRS is surrounded by a distinct outer layer of outer root sheath ( ORS ) cells consisting of a layer of epidermal cells continuous with the epidermis [3] , [4] . After birth , HFs cycle through stages of active growth ( anagen ) , regression ( catagen ) and inactivity ( telogen ) [2] . The periodic growth and rest of the HF reflects the migration , proliferation and differentiation of multipotent HF stem cells suggested to be located primarily in the bulge region of the ORS [5] . Multipotent stem cells have also been detected outside of bulge region in the lower part of whisker HFs [6] , and the matrix in pelage HFs is suggested to contain restricted self-renewing stem cell for each inner structure [7] . Whether these observations reflect differences in stem cell migration and differentiation between whisker and pelage HF remains to be elucidated . The onset of anagen is characterised by the initiation of cell proliferation in the secondary HG adjacent to the DP in the proximal part of the HF [8] , leading to the invasion of the elongating HF into subcutaneous tissue . This process is accompanied by the differentiation of the matrix cells in the hair bulb leading to the formation of the hair shaft and the IRS . In pigmented animals melanins are synthesised by melanocytes located close to the forming hair shaft and this process occurs exclusively during anagen [9] . During the following catagen phase , hair shaft production stops since proliferation and differentiation is dramatically reduced leading to a regression of the HF . Catagen is followed by telogen which is characterised by a minimal signalling exchange between the DP and follicular keratinocytes [10] . Numerous signalling pathways have been implicated in the regulation of the HF and considerable overlap in the pathways promoting both HF morphogenesis and the anagen stage of postnatal HF cycling has been revealed [3] . Signalling induced by Sonic hedgehog ( Shh ) and Wnts are indispensable for HF morphogenesis and the anagen phase of HF cycle [11]–[14] . Active Wnt signalling leads to stabilization and nuclear translocation of cytoplasmic β-catenin that activate transcription of specific target genes together with the Lef1/Tcf proteins [15] . Activation of hedgehog signalling occurs upon ligand-binding to Patched ( Ptc ) that subsequently releases its inhibitory effect on Smoothened ( Smo ) leading to transcription of specific target genes by the Gli family of transcription factors [16] . Inhibition of Bone morphogenetic protein ( BMP ) signalling by Noggin ( Nog ) is important for both morphogenesis and anagen induction [17] , [18] . Mediators of paracrine and autocrine interactions trigger intracellular signals that are transmitted to the nucleus , where the activation of specific transcription factors establish and propagate proper epithelial-mesenchymal interactions during organ generation and regeneration . Therefore , it is important to identify and characterize the transcription factors that are critically involved in the development of HF as well as the regulation of HF cycling to further elucidate epithelial-mesenchymal interactions specific for HF morphogenesis and cycling . One class of transcription factors , the LIM-homeodomain family , regulates many important developmental processes such as asymmetric cell division , tissue specification and differentiation of specific cell types [19] . One member of this gene family , Lhx2 ( previously LH2 , LH2A ) , was first identified as a gene specifically expressed in pre-B cell lines and independently isolated as a transcription factor binding to the glycoprotein hormone α-subunit promoter [20] , [21] . Lhx2 has been shown to be essential in various epithelial-mesenchymal interactions as well as in the regulation of different progenitor/stem cell populations [20] , [22]–[27] , revealing its importance in regulating fundamental processes important for organ/tissue generation and regeneration . Since Lhx2 plays an important role in a variety of epithelial-mesenchymal interactions and in the regulation of various stem/progenitor cell populations , we analysed the expression pattern and function of Lhx2 in the HF . Based on mRNA expression , Lhx2 is expressed from an early stage of morphogenesis and eventually becomes restricted to the ORS and a subpopulation of the matrix cells located in the proximal part of the hair bulb . During postnatal HF cycling Lhx2 expression reveals a similar pattern and presence of Lhx2 protein in matrix cells and in cells scattered in the ORS was confirmed . Moreover , Lhx2 expression is associated with the anagen phase and was also primarily expressed by precursor cells outside of the bulge region . These results indicate that Lhx2 is involved in anagen initiation/progression and morphogenesis and contradict previously published data where Lhx2 protein was only detected in stem cells in the bulge region [27] . Rhee et al . also suggested that Lhx2 was important for maintaining quiescent stem cells and not involved in their differentiation . However , by using mouse models where Lhx2 expression could be conditionally inactivated in postnatal HFs , or significantly reduced during morphogenesis , we could confirm that Lhx2 is important for both anagen and morphogenesis progression . Furthermore , by using a mouse model where Lhx2 expression could be induced in postnatal HFs , we also showed that Lhx2 expression was sufficient to induce anagen . Thus , cyclic expression of Lhx2 in HFs regulates hair formation .
To elucidate the expression pattern of Lhx2 in detail we used in situ hybridization analysis of HFs during different stages of morphogenesis . Lhx2 starts to be expressed in patches of basal epidermal cells prior to any obvious formation of a dermal condensate in the underlying dermis ( Stage 0 of HF morphogenesis , see [28] ) ( Figure 1A ) . Lhx2 is subsequently expressed in the placode when the epidermal thickening and the adjacent dermal condensate have formed ( Stage 1 ) ( Figure 1B ) , and in the downward growing primary HG ( Stage 2–3 ) ( Figure 1C ) . This expression pattern of Lhx2 mRNA in the developing HF is in agreement with what has been reported previously [27] . At the epithelial peg stage ( Stage 4 ) Lhx2 is expressed by most cells in the epithelial portion of the developing HF ( Figure 1D ) , and we have no evidence for higher expression at the leading edge of the HF as previously reported [27] . At the stage when the concentric layers of the HF are initially formed ( bulbous peg , Stage 5 ) , Lhx2 expression becomes down-regulated in the differentiated cells in the IRS ( Figure 1E ) . During the postnatal phase of morphogenesis when the HFs are fully developed and the hair shaft has erupted through the skin surface ( Stage 8 ) , Lhx2 expression is maintained predominantly in the epidermal matrix cells located at the most proximal portion of the hair bulb and Lhx2+ cells are also scattered within the ORS ( Figure 1F ) . A similar expression pattern is also seen during whisker follicle morphogenesis ( Figure S1A ) . This expression pattern of Lhx2 has not been described previously and we did not observe an enrichment of Lhx2 expressing cells in the upper ORS at the presumptive bulge region [27] . In situ hybridization analysis of postnatal HFs in full anagen ( Sub-stage VI , [29] ) showed that Lhx2 expression was primarily detected in matrix cells in the proximal part of the bulb region and in cells scattered in the ORS ( Figure 1G ) . This expression pattern is almost identical to that during late morphogenesis ( Figure 1F ) , and is also similar to the expression pattern in adult whisker follicles which are characterised by an extended anagen ( Figure S1B ) [30] . We also detected Lhx2 protein in the nucleus of cells located in the proximal part of the hair bulb and in cells scattered in the ORS in anagen HFs ( Figure 1H and 1I ) . The Lhx2 protein appears to be more stable compared to Lhx2 mRNA since the protein can also be detected in matrix cells as they differentiate and move distally in the hair bulb , whereas the mRNA is enriched in the cells from where the matrix cells originate in the proximal part of the bulb ( Figure 1H and 1G ) . Persistence of Lhx2 protein after mRNA down-regulation during differentiation has been observed previously in the olfactory epithelium [24] . Thus , the presence of Lhx2 mRNA in the HF is a reliable indicator of functional expression of Lhx2 . However , Lhx2 expression was not detected at the time when the HFs had entered telogen at 7–8 weeks of age , ( Figure 1J ) , suggesting a cyclic expression pattern of Lhx2 in postnatal HFs . To further investigate the fluctuations in Lhx2 expression we determined the expression pattern around the time of telogen-anagen transition . Since induction of the second postnatal anagen was difficult to predict in our control animals ( Table S1 ) , we analysed HFs during the first postnatal telogen and the different Sub-stages of the subsequent and first postnatal anagen in 3–4 week old mice . During telogen in 3 week and 1 day old mice ( Figure 2A and 2D; Table S1 ) , Lhx2 expression could not be detected in any HF ( Figure 2B and 2C; Figure S3A ) , similar to what was observed in HFs in telogen in 7–8 weeks old mice ( Figure 1J ) . In slightly older animals ( ≥3 weeks and 4 days ) HFs in several individuals have initiated anagen ( Table S1 ) . However , some animals at this age that neither show any obvious morphological transition to anagen , nor express anagen-specific genes such as Shh ( Figure 2E and 2I ) , revealed distinct expression of Lhx2 in the secondary HG adjacent to the DP ( Figure 2F–2H; Figure S3B ) . HF stem cells located in the bulge area express CD34 , whereas the cells in the secondary HG are CD34− ( Figure 2B and 2F ) [31] . Few , if any , of the Lhx2 expressing cells are CD34+ since they appear as separate cell populations ( Figure 2F ) . In the early stages of anagen when Shh is expressed but prior to any pigment deposition ( e . g . anagen Sub-stages I-II . Figure 2J and 2M ) , Lhx2 is expressed both by cells in the secondary HG , where the first proliferating cells are located during anagen induction [8] , and by the future matrix cells surrounding the DP ( Figure 2K and 2L; Figure S3C ) . Although there is a sharp boundary where Lhx2 mRNA is present or absent between the secondary HG and the bulge region , we can detect Lhx2 protein in a few cells in the lower part of the bulge region as well as in the secondary HG ( Figure 2K’ ) . In the subsequent Sub-stage of anagen ( maintained Shh expression ) when melanocytes deposit pigment ( Sub-stage IIIa-c , Figure 2N and 2Q ) , the matrix cells in the proximal part of the bulb and cells in the ORS express Lhx2 whereas expression is turned off in the IRS cells ( Figure 2O and 2P; Figure S3D ) , similar to the expression pattern during early HF morphogenesis ( compare Figure 2P and Figure S3D to Figure 1D and 1E ) . At later Sub-stages of anagen ( Sub-stages IV-VI with maintained Shh expression , Figure 2R and 2U ) Lhx2 expression is maintained in the ORS ( Figure 2S and 2T; Figure 1G and 1H ) and in matrix cells in the most proximal part of the bulb ( Figure 2S’ and 2T’; Figure 1G–1I ) , similar to expression pattern during Stage 8 of HF morphogenesis ( Figure 1F ) . Also during anagen most of the Lhx2+ cells are CD34− as they represent separate cell populations ( Figure 2K , 2O , 2S , and 2S’ ) . Thus , during postnatal HF cycling , Lhx2 expression is initiated in a distinct subpopulation of CD34− epidermal progenitor cells in the secondary HG during late telogen immediately prior to anagen induction and the expression is maintained in the transient portion of the HF throughout anagen . In contrast to previously published results [27] , the anagen-associated expression during adult HF cycling , and the similarity of this expression pattern to that during HF morphogenesis , suggest that Lhx2 regulate hair generation and regeneration . We have shown that Lhx2 expression is normally associated with the anagen phase of postnatal HF cycling and we therefore wanted to elucidate whether Lhx2 is necessary for anagen initiation and/or progression . To address this issue we obtained two mouse strains in which Lhx2 exons have been flanked by loxp sites , e . g . floxed Lhx2 alleles ( Lhx2flox ) ( Figure S4B , and [32] ) , in order to conditionally inactivate the Lhx2 gene by expression of the Cre recombinase [32] . The Lhx2flox mice were bred into mice harbouring the null allele of Lhx2 ( Lhx2− ) [22] , and into transgenic mice ubiquitously expressing a fusion protein between the Cre recombinase and the ligand binding domain of the Estrogen Receptor ( CreER ) [33] , to generate CreER:Lhx2flox/flox , CreER:Lhx2flox/- , Lhx2flox/flox and Lhx2flox/- mice . Application of Tamoxifen ( Tx ) to the skin leads to nuclear translocation of the CreER fusion protein and conditional inactivation of the Lhx2 gene in HFs in the CreER:Lhx2flox/flox and the CreER:Lhx2flox/- mice . The Lhx2 gene in the Lhx2flox/flox and the Lhx2flox/- mice is unaffected by this treatment and were used as a control . The skin on the back of these mice was shaved and treated with Tx during the first postnatal telogen at approximately 3 weeks of age ( Figure S2 ) . Initiation and progression of the first postnatal anagen starting around 4 weeks of age was subsequently analysed ( see Figure S2 ) . Most of the Tx-treated CreER:Lhx2flox/- and CreER:Lhx2flox/flox mice did not re-grow their hair coat on the shaved area ( 6/8 ) whereas all the control animals did ( 9/9 ) ( Figure 3A ) , revealing that Lhx2 is required for postnatal hair regeneration . Histological analysis of the HFs where Lhx2 was inactivated showed that anagen is initiated similar to control animals and progress to anagen Sub-stage III ( Figure 3B and 3D ) , but whereas the control HFs develop normally the mutant HFs are unable to develop beyond Sub-stage III and assemble a normal hair shaft ( Figure 3C and 3E ) . HFs outside of the Tx-treated area in these mice assembled hair shafts that were indistinguishable from control HFs ( Figure 3F ) . To confirm the inactivation of the Lhx2 gene we performed in situ hybridisation using the full length Lhx2 cDNA as a probe which has been shown to hybridise to mRNA expressed from both the WT allele as well as to the inactivated Lhx2 allele [34] , [35] , and an exon 2-specific probe that only hybridises to the WT mRNA as this exon is deleted in both the conditionally inactivated Lhx2flox alleles and in the Lhx2- allele ( Figure S4B , [22] , [32] ) . As expected , cells in control HFs in anagen expressed mRNA hybridising to both probes ( Figure 3G , left panels ) , whereas cells in anagen HFs , in which Lhx2 should be conditionally inactivated , expressed a truncated mRNA hybridising exclusively to the full length probe and hence lacking exon 2 ( Figure 3G , right panels ) . Significant reduction of Lhx2 protein in the ORS and the hair bulb could also be observed in the HFs where Lhx2 had been conditionally inactivated ( Figure 3H ) , confirming that the cells expressing the truncated mRNA are unable to generate a functional protein . These results confirmed that the Lhx2 gene has been inactivated in the Tx-treated CreER:Lhx2flox/- and CreER:Lhx2flox/flox mice . Two Tx-treated CreER:Lhx2flox/flox mice started to re-grow some hair one week later compared to the control animals . Also skin from these individuals had HFs with arrested development , but there were numerous HFs containing cells expressing the control allele since the mRNA hybridised to both the full length probe as well as the exon 2-specific probe ( Figure S5 ) . None of the Tx- treated CreER:Lhx2flox/- mice re-grew their hair , revealing that conditional inactivation of the Lhx2 gene was more efficient in these mice compared to the CreER:Lhx2flox/flox mice , where a few cells retaining the control allele could rescue hair formation to some extent . Thus , Lhx2 is critically involved in hair formation by regulating anagen progression . The phenotype in Lhx2−/− mouse forebrain has been suggested to be due to lack of proliferating progenitor cells [22] . To analyse if the HF phenotype is solely due to lack of proliferating progenitor cells , we analysed the expression of the S-phase-specific histone gene Hist1h3c [36] , [37] . As expected Hist1h3c expression could not be detected in HFs in telogen whereas it is expressed in the proximal part of the hair bulb and in matrix cells in HFs in anagen ( Figure S6A and S6B; Figure 3I ) , where the proliferating progenitor cells are located . Numerous cells expressing Hist1h3c could also be observed in a similar pattern in the HFs where Lhx2 had been conditionally inactivated 1–2 weeks after the control mice re-grew their hair ( Figure 3I; Figure S6C ) . Thus , although we have not quantified the level of expression of Hist1h3c or quantified the number of proliferating cells , the developmental arrest at anagen Sub-stage III in mutant HFs cannot solely due to the fact that the progenitor cells are unable to proliferate . The presence of a significant number of proliferating progenitor cells in mutant HFs suggested that the signalling pathways critical for anagen initiation/progression were relatively unperturbed . To address this issue we analysed the expression of various genes encoding mediators of such pathways . No obvious differences in the expression of mediators of hedgehog signalling was observed since the ligand Shh , the signal transducer Smo , and the receptor and universal target of hedgehog signalling Ptc1 were equally expressed in control and mutated HFs ( Figure 3J ) . Moreover , the basal components of the canonical Wnt signalling pathway were also present , since both control and mutated HFs accumulated activated ( i . e . dephosphorylated ) β-catenin , and expressed Lef1 , the transcriptional effecter of activated β-catenin ( Figure 3J ) . Thus , inactivation of Lhx2 appears not to have a major impact on the hedgehog or Wnt signalling pathways . Since the mutated HFs contained proliferating cells and had no obvious defects in signalling pathways important for anagen progression , we wanted to elucidate if the mutated HFs are either completely blocked or delayed in their development . To address this issue we analysed mice with mutated HFs that had not re-grown any hair at 9 weeks of age . At 9 weeks of age most control animals have entered telogen ( Table S1 ) and thus do not express Lhx2 , Shh , or Hist1h3c ( Figure 1J; Figure 2B and 2C; Figure 2D and 2I; Figure S6A ) . Although many HFs have initiated hair shaft formation in the 9 week old conditional mutants , we rarely find any HFs that have developed beyond anagen Sub-stage III and some of the formed hair shafts appear distorted ( Figure 3K ) . Moreover , all HFs contain cells expressing the truncated Lhx2 allele since the mRNA hybridised only to the full length probe whereas it does not hybridise to the exon 2-specific probe ( Figure 3K ) . Also the anagen-specific marker Shh and the proliferation-specific marker Hist1h3c are expressed in the mutated HFs ( Figure 3K ) . These results suggest that the HFs where Lhx2 were efficiently inactivated are maintained in an anagen-like stage but are unable to assemble a normal full length hair shaft even after an extended period of time . Thus , Lhx2 does not primarily regulate the proliferation of HF progenitor cells but rather appears to regulate the patterning/differentiation of HF progenitor cells . Since there is significant overlap between the signalling pathways promoting HF morphogenesis and the anagen stage [3] , we wanted to elucidate if Lhx2 is also important for HF morphogenesis . A significant decrease in the number of HFs in Lhx2−/− embryos has been reported [27] , indicating that Lhx2 play a role in HF morphogenesis . However , since the lethality of Lhx2−/− embryos at E15–17 [22] coincides with and even precedes the onset of pelage HF morphogenesis , it is difficult to ascertain the precise role of Lhx2 in HF morphogenesis in Lhx2−/− embryos . To further investigate this we generated a mouse strain harbouring a hypomorphic allele of Lhx2 denoted Lhx2Neo ( Figure S4A ) . Mouse embryos homozygous for this allele ( Lhx2Neo/Neo mice ) develop an eyeless phenotype similar to the Lhx2−/− mouse embryos ( Figure S4C ) confirming that Lhx2 expression was significantly reduced in the Lhx2Neo/Neo embryos . However , Lhx2 expression is detected in the Lhx2Neo/Neo embryos revealing that it is not a null allele ( Figure 4D ) . Since the liver was less affected in these animals compared to Lhx2−/− embryos ( data not shown ) , the expected number of viable Lhx2Neo/Neo embryos were obtained at late gestation . Thus , this mouse strain allowed us to follow pelage HF morphogenesis up to a stage when it is well established [2] , [38] . Back skin of all E16 . 5 Lhx2Neo/Neo embryos contained fewer HFs compared to control animals ( Figure 4A and 4C ) , which is in agreement with the number found in Lhx2−/− embryos at a similar gestational age [27] . Moreover , the HFs that did develop in the Lhx2Neo/Neo embryos had a distinct DP but appeared to be developmentally arrested prior to the epithelial peg stage ( Stage 4 ) compared to the HFs in control animals ( Figure 4A ) . At E18 . 5 the difference in number of HFs between control and Lhx2Neo/Neo embryos was more striking since HF density did not increase between E16 . 5 and E18 . 5 in Lhx2Neo/Neo embryos ( Figure 4B and 4C ) , whereas it increased by 75% in the control embryos ( Figure 4C ) . Moreover , many HFs in E18 . 5 Lhx2Neo/Neo embryos appeared to be developmentally arrested at the same stage as those in the E16 . 5 embryos ( Figure 4A and 4B ) , similarly to the arrested anagen progression when Lhx2 is conditionally inactivated in postnatal HFs ( Figure 3E ) . To investigate whether loss of Lhx2 affected signalling pathways critically involved in HF morphogenesis , we analysed expression of various genes encoding mediators of such pathways . Similar to the postnatal HFs where Lhx2 had been conditionally inactivated , there was no obvious difference between control and Lhx2Neo/Neo HFs in expression levels in for components of the hedgehog signalling pathways ( Smo , Shh and Ptc1 ) , the canonical Wnt signalling pathway ( Lef1 ) or BMP signalling pathway ( BMP4 and Noggin ) ( Figure 4D ) . The distribution of E-cadherin ( E-cad ) in interfollicular epidermis is similar in control and Lhx2Neo/Neo skin at E16 . 5 ( Figure 4D ) . In summary , the use of a hypomorphic loss-of-function model for Lhx2 has allowed us to follow HF development to a much later stage of embryonic development compared to Lhx2−/− embryos . By using this novel loss-of-function model we have been able to demonstrate that Lhx2 is also required for HF morphogenesis , and the significant reduction in number of HFs suggests that Lhx2 also play a role in HF induction . It has been previously reported that over-expression of Lhx2 in early embryonic epidermis has no effect on HF morphogenesis [27] . This could be explained by the already high level expression of endogenous Lhx2 in the developing HF ( Figure 1A–1E ) , upon which additional Lhx2 expression would have little effect . We therefore reasoned that investigation of the function of Lhx2 in HFs would be helped by analysing the effect of transgenic expression of Lhx2 when the endogenous gene is turned off during telogen ( Figure 1J ) . To address this issue we developed a mouse model where we could induce Lhx2 expression in the epidermal portion of postnatal HFs . We utilised an expression system based on the Z/AP double reporter vector [39] , where a floxed allele of β-Geo ( encoding a β-galactosidase-Neomycin fusion protein ) is followed by an expression cassette consisting of the Lhx2 cDNA , an internal ribosomal entry site ( IRES ) and green fluorescent protein ( GFP ) cDNA ( Figure S7A ) . A Z/Lhx2-GFP founder mouse strain with high β-Galactosidase ( β-Gal ) activity in the epithelial part of HFs during telogen and in ORS cells during anagen was chosen for further breeding ( Figure S7B and S7C ) . Thus , DNA recombination by the Cre recombinase ( e . g . following CreER expression and Tx treatment ) will delete the β-Geo gene and induce expression of Lhx2-GFP primarily in the β-Gal+ cells ( Figure S7 ) . To analyse the effect of Lhx2 expression on HF cycling we applied Tx onto shaved back skin of 5 week old control mice and Z/Lhx2-GFP:CreER double transgenic mice and analysed the mice at 8–9 weeks of age when HFs should be in telogen ( Figure S2 ) . Since most animals will complete an anagen-catagen-telogen transition between the Tx treatment and the time of analysis ( Table S1; Figure S2 ) , this experimental approach would avoid confounding effects of the shaving procedure and/or of the Tx treatment . Moreover , this approach allowed us to analyse if transgenic Lhx2 expression influence the transition from anagen to telogen , and how it affects HFs in the extended telogen phase . HFs in all Tx-treated Z/Lhx2-GFP:CreER mice were in anagen at 9 weeks of age as determined by size , morphology , melanin deposition and Shh expression ( Figure 5A and 5A’; Table 1 ) , whereas HFs in almost all control mice were in telogen as expected at this age ( Figure 5B and 5B’; Table 1; Table S1 ) . Moreover , HFs that were outside of the Tx-treated area on Z/Lhx2-GFP:CreER mice remained in telogen ( Figure 5C and 5C’ ) . An extension of the growth phase of the HF by one week will generate noticeably longer hair in mice [40] . If our experimental set-up maintained anagen from Tx-treatment until the time of analysis , the length of anagen would have been extended by >2 weeks and hence lead to considerably longer hair . However , we did not observe any obvious increase in hair length in the Tx-treated Z/Lhx2-GFP:CreER mice ( data not shown ) , suggesting that our experimental approach did not prolong the anagen phase but rather prematurely induced anagen in the following extended telogen . To further address this issue we analysed HFs in Tx-treated double transgenic mice closer to the time for anagen-catagen-telogen transition to elucidate whether animals expressing transgenic Lhx2 could enter telogen . At this time ( 8 weeks and 2–4 days old ) the HFs were in telogen in one animal ( Figure 5E and 5E’ , Table 1 ) whereas the remaining mice had the HFs in anagen ( Figure 5D , 5D’ , 5F , and 5F’; Table 1 ) . The HFs in all control mice for these time points were in telogen ( Figure 5G and 5G’; Table 1 ) . Thus , HFs where Lhx2 expression had been induced were able to enter telogen at the same time as control HFs at approximately 7–8 weeks of age , showing that the former HFs prematurely initiated anagen as they all were in anagen by 9 weeks of age ( Table 1 ) . Thus , the HFs in 89% ( 8/9 ) of the Tx-treated Z/Lhx2-GFP:CreER mice were in anagen , whereas HFs in 98% of the treated control animals were in telogen as expected ( Table 1 ) . The only time we observed HFs in telogen in Tx-treated Z/Lhx2-GFP:CreER mice was when analysed at an earlier time point close to the expected time for the anagen-catagen-telogen transition , suggesting that our experimental approach lead to premature initiation of anagen . To confirm Cre-mediated recombination of the Z/Lhx2-GFP transgene and expression of Lhx2-GFP we analysed β-Gal activity and GFP expression in HFs in Tx-treated Z/Lhx2-GFP:CreER mice and control mice . Cells with robust expression of GFP were usually enriched at the proximal part of the hair bulb in 50% of the HFs in Tx-treated double transgenic mice ( Figure 5H ) whereas no GFP expression was detected in HFs in Tx-treated Z/Lhx2-GFP controls ( Figure 5J ) . Numerous cells expressing endogenous Lhx2 were also located in the proximal part of the bulb during anagen ( Figure 1G and 1H ) , suggesting that the cells expressing transgenic Lhx2 follow the same migration path as the cells expressing the endogenous gene . GFP was also expressed by cells in the ORS , which is in agreement with the distribution of the β-Gal+ cells in Z/Lhx2-GFP control mice ( Figure 5H and 5L; Figure S7C ) . Although most HFs showed GFP expression in the proximal part of the hair bulb the distribution and number of GFP+ cells in the remaining HFs varied considerably ( Figure 5I ) . Both GFP+ and Lhx2+ cells were frequently detected in the IRS ( Figure 5I and 5M ) , showing that follicular progenitor cells can migrate into the IRS despite maintained expression of Lhx2 . However , Cre-mediated excision of β-Geo was not complete in the Tx-treated Z/Lhx2-GFP:CreER mice since numerous β-Gal+ cells were present in these HFs ( Figure 5K ) . As expected , GFP expression and β-Gal activity appears to be mutually exclusive since cells lacking β-Gal activity were enriched in GFP+ regions in the proximal part of the bulb ( Figure 5K and 5H ) , whereas numerous β-Gal+ cells are located in this region in control Z/Lhx2-GFP mice ( Figure 5L and Figure S7C ) . Some HFs in Tx-treated Z/Lhx2-GFP:CreER mice contained more β-Gal+ cells which is in agreement with the decreased number of GFP+ cells in some HFs ( Figure 5K and 5I ) . Since Lhx2 expression is detected in all HFs in these mice whereas expression of GFP is not ( compare Figure 5M to Figure 5H and 5I ) , indicates that the endogenous Lhx2 gene is also expressed as in normal anagen . Thus , relatively few cells expressing the Lhx2 transgene is sufficient to induce endogenous Lhx2 expression in a cell nonautonomous manner and thereby prematurely initiate anagen .
We have analysed the function and expression of the LIM-homeobox gene Lhx2 in HFs during morphogenesis and the cyclic phases of postnatal hair growth ( summarised in Figure 6 ) . Lhx2 is expressed in basal keratinocytes prior to the formation of the hair placode at initiation of morphogenesis . Expression becomes restricted to matrix cells at the proximal part of the hair bulb and to cells in the ORS in a fully developed HF . When the HF enters telogen Lhx2 expression becomes undetectable . However , during late telogen Lhx2 expression re-appears in CD34- cells located in the secondary HG and this expression is maintained when anagen commences . The expression observed during morphogenesis is thereafter reiterated during progression of anagen . This developmental and stage-specific expression pattern suggests that Lhx2 is involved in the generation and regeneration of hair . This hypothesis is supported by mouse models where Lhx2 expression could either be conditionally inactivated , significantly reduced , or switched on in the HF . The loss-of-function mouse models revealed that Lhx2 is required for progression of both anagen and morphogenesis and hence hair formation . The gain-of-function mouse model showed that Lhx2 expression is sufficient to induce anagen . The function of Lhx2 in the HF has been addressed previously [27] , concluding that Lhx2 is important for maintaining the quiescence of the stem cells located in the bulge region . This interpretation of Lhx2 function is difficult to reconcile with the expression pattern and the phenotypes in the gain-of-function and loss-of-function mouse strains reported herein . The reasons for these discrepancies in interpretations of the function of Lhx2 in the HF are not clear . We do , however , detect Lhx2 protein in cells in the proximal part of the bulge region , which is in agreement with Rhee et al . [27] , although we are unable to detect Lhx2 mRNA in these cells . The functional relevance of this observation remains to be elucidated . The different interpretation of the function of Lhx2 could be explained by the different approaches used to study gain-of-function and loss-of-function . Rhee et al . [27] expressed Lhx2 under the control of the keratin-14 promoter leading to epidermal Lhx2 expression early in embryonic development , probably prior to HF induction [41] , and when the endogenous Lhx2 gene is widely expressed in the developing HF . We induced expression in HFs when the endogenous gene is turned off during telogen . Rhee et al . analysed loss-of-function by transplanting HFs from Lhx2−/− embryos onto skin of nude mice , whereas we analysed intact postnatal HFs where the Lhx2 gene had been conditionally inactivated . Rhee et al . also showed a significantly reduced number of HFs in the conventional Lhx2 mutant embryos , which is in agreement with our loss of function data in embryos of the same gestational age . However , since the Lhx2−/− embryos start to die in utero even prior to initiation of HF morphogenesis , we confirmed these observations in a mouse strain harbouring a hypomorphic allele of Lhx2 ( Lhx2Neo ) where HF development can be followed to a gestational age when pelage HF morphogenesis is firmly established . The Bulge Activation hypothesis has been the dominating theory of induction of anagen at the expense of the HF Predetermination hypothesis . The Bulge Activation hypothesis suggests that stem cells in the bulge region start to proliferate after responding to signal ( s ) emanating from the DP , generating transient amplifying cells that migrate to the secondary HG and initiate anagen [5] , [42] . The HF Predetermination hypothesis suggests that the secondary HG itself contains stem cells that start to proliferate after receiving signals from the DP without input of transient amplifying cells from the bulge region [43] , [44] . This hypothesis has been discarded mainly due to the failure to detect slowly dividing cells , or so-called label-retaining cells ( LRCs ) , in the secondary HG . The observation that Lhx2 expression is initiated in CD34− cells in the secondary HG , and not in CD34+ cells in the bulge region , just prior to and at initiation of anagen fits with the HF Predetermination hypothesis . Although we cannot exclude that a few CD34+Lhx2+ cells might be present in the bulge , the majority of the Lhx2+ cells are CD34− and are located in the secondary HG , the ORS , and in cells in the proximal part of the hair bulb . Thus , our data strongly suggest that Lhx2 expressing cells are involved in the expansion and patterning of the transient portion of the HF . An alternative interpretation that would fit with the Bulge Activation hypothesis is that Lhx2+ cells in the secondary HG in late telogen relay signal ( s ) from the DP to the cells in the bulge region leading to the migration of these cells into the secondary HG and hence initiation of anagen . Consequently , the stem cells start to express Lhx2 when they have left the bulge region and entered the secondary HG . Thus , irrespective of which hypothesis prevails , the Lhx2+ cells in the HG represent an epidermal progenitor/stem cell population that is distinct from those in the bulge region . The CD34+ cells in the bulge region are LRCs and many of their stem cell properties have been based on this feature [5] , [31] , whereas no LRCs are present in the secondary HG [5] , [43] . We show that Lhx2 is primarily expressed by CD34− cells and hence Lhx2 is most probably not expressed by LRCs in the bulge region . However , a recent study where the fate of cells expressing the leucine-rich G protein-coupled receptor 5 ( Lgr5 ) was analysed , question the link between stem cells and LRCs in HFs [45] . The Lgr5+ cells were suggested to define a novel stem cell population that cycles , is long lived and multipotent , contradicting the generally held view that “true” stem cells are quiescent and hence LRCs , whereas their progeny ( i . e . transient amplifying cells ) actively cycle and are short-lived . The authors suggest that Lgr5 expression defines a novel stem cell population distinct from the CD34+ LRCs in the bulge region and argue for a reinvestigation of the relationship between HF stem cells and LRCs . The expression pattern of this novel stem cell marker in the HF is very similar to that of Lhx2 since both Lgr5 and Lhx2 are expressed in the secondary HG , in the ORS and in matrix cells in the proximal part of the hair bulb . A significant overlap in Lhx2 and Lgr5 expression is also corroborated by Jaks et al . [45] who show that Lhx2 is only expressed by Lgr5+ cells in the HF . Although the exact relationship between Lhx2 and Lgr5 expression at the cellular level remains to be elucidated , the respective expression pattern suggests that they are highly overlapping and hence defines a similar , if not an identical , cell population . Our data also suggest that Lhx2 expression during late telogen might be an “activation marker” of the putative Lgr5+CD34− follicular stem/progenitor cell population in the secondary HG , and that the Lhx2+Lgr5+CD34− cells initiate the anagen phase . Moreover , recent data also suggests that cells in the secondary HG become activated during late telogen prior to anagen induction which is also accompanied by changes in gene expression pattern in the DP [46] , and the expression of Lhx2 might be a part of this process . The finding that over-expression of Lhx2 in the epidermal part of the HF is sufficient to prematurely induce anagen is in agreement with this hypothesis . However , Lhx2 is not required for anagen ( or HF morphogenesis ) induction whereas it is required for the progression of anagen and morphogenesis , revealing that redundant mechanism ( s ) can compensate for Lhx2 in the initiation process . The molecular basis of this redundancy in the initiation of anagen is unknown , but a role of other LIM homeobox genes is unlikely since the related LIM-homebox genes Lhx9 and Lmx1b , which can compensate for Lhx2 in limb development [47] , are not expressed in follicular epidermis ( unpublished data ) . It remains to be elucidated whether or not upregulation of Lhx2 expression during late telogen is an intrinsic property of the epidermal progenitor cells , or a consequence of signal ( s ) emanating from the DP , and/or linked to cyclic BMP signalling regulating stem cell activation [48] , Relatively few cells express the Lhx2 transgene in HFs of Tx-treated Z/Lhx2-GFP:CreER mice , which might explain why the HFs can enter telogen phase similar to the control HFs . A more robust expression of transgenic Lhx2 might give a different outcome , and this issue is being addressed at present . Moreover , the ability of cells expressing transgenic Lhx2 to migrate in the same manner as cells expressing endogenous Lhx2 might also explain why HFs prematurely initiate a new anagen phase . Thus , cells expressing transgenic Lhx2 would be close to the DP and convey epidermal signals to it , similar to the activation of the HG cells prior to anagen initation [46] . Consequently the DP would signal back to epidermal cells since continuous signalling between epidermal cells and DP is required for hair formation [2] , [10] . This could , in part , explain why relatively few Lhx2 expressing cells in postnatal HFs can induce endogenous Lhx2 expression in a cell nonautonomous manner and initiate anagen prematurely . Thus , in this mouse model initiation of anagen is promoted whereas the duration of anagen is not affected . The Lhx2 loss-of-function models have revealed that Lhx2 is necessary for the progression of both anagen and morphogenesis . Since the Hedgehog , Wnt and Nog/BMP signalling pathways are important for anagen and morphogenesis initiation/progression [11]–[14] , [17] , [18] , we analysed whether any of these pathways were affected in the mutant HFs . However , no difference in expression of the central mediators of Nog/BMP , Hedgehog ( Shh , Smo , Ptc1 ) or Wnt ( Lef1 ) signalling could be observed between control and mutated HFs . Furthermore , activation of the respective pathways in mutant HFs was supported by the presence of the stabilized form of β-catenin that mediates Wnt signalling [49] , expression of Ptc1 that is a downstream transcriptional target gene in Hedgehog signalling [11] , [50] , and expression of Lef1 , which is induced by Nog [51] . Signalling via NF-κB induced by the Tumour Necrosis Factor ( TNF ) family of ligands is also an important pathway promoting HF morphogenesis [52] . Although we have not directly analysed this pathway in mutant HFs , it is unlikely that Lhx2 is involved in this pathway since active NF-κB signalling induces expression of Shh [53] , which is clearly expressed in both the conditional mutant as well as the hypomorphic mouse strain . Thus , as far as we can determine , loss of Lhx2 function does not disturb any major signalling pathway important for anagen initiation/progression or HF morphogenesis . Moreover , significant numbers of proliferating cells are present in the mutated HFs showing that the developmental arrest at anagen Sub-stage III cannot solely be due to lack of proliferating cells at this stage . Co-factor of LIM domain 2 ( Clim2 ) is expressed in the ORS and the matrix cells and has been shown to physically interact with Lhx2 [54] . Transgenic expression of a dominant negative Clim in the HF using a K14 promoter leads to a progressive hair loss during postnatal life due to aberrant HF differentiation and disrupted HF structure [54] , revealing that other means to interfere with Lhx2 function also hampers hair formation . These data suggest that Lhx2 play a role in the differentiation/patterning of the HFs . Such a function of Lhx2 has previously been shown in the fetal liver where Lhx2 , both in a cell autonomous and cell nonautonomous manner , is involved in the expansion , differentiation and organisation of all cellular components of the liver into a functional 3-dimensional structure [25] . The exact mechanism ( s ) by which Lhx2 can regulate such complex processes in organ development is a central question in organogenesis . Our results show that Lhx2 can regulate two separable processes in HF cycling; firstly , Lhx2 can initiate anagen but it is not required for this process , secondly , Lhx2 is required for the progression of anagen ( and morphogenesis ) generating a fully assembled hair shaft . These observations indicate a rather complex role for Lhx2 in HF biology , suggesting that Lhx2 can initiate the signalling cascades necessary for anagen initiation , as well as regulate the process of anagen progression and hence hair formation . The significantly reduced number of HFs in the mouse strain homozygous for the hypomorphic allele of Lhx2 as well as in the Lhx2−/− mice as previously reported [27] , further supports the idea that Lhx2 has an important function during the early stages of HF development . We have not yet been able to determine the molecular basis for Lhx2 activity in the HF . However , we have previously performed a global gene expression analysis of hematopoietic stem cell-like cell lines with inducible Lhx2 expression where we could identify 141 genes that correlated to Lhx2 expression . Gene Ontology classification of these genes revealed that genes related to ‘regulation of signal transduction’ and ‘organogenesis’ were over-represented [26] . A considerable fraction of these genes ( 31% ) that we have analysed are expressed in the HF , for example Nuak1 ( NM_001004363 ) , Tmem2 ( NM_031997 ) , Etv5 ( NM_023794 ) , and Enc1 ( NM_007930 ) [26] , suggesting a putative overlap in Lhx2 function between different tissues . Further elucidation of the molecular basis of Lhx2 function in the HFs is important to fully understand the complex regulation of hair formation .
All experiments involving animals were approved by the Animal Review Board at Umeå University . Staging of hair follicle morphogenesis is according to [28] , and the staging and Sub-stages of postnatal HF cycles and the nomenclature of HFs used throughout this work are according to [29] . The initial postnatal hair cycles of back skin HFs in female C57BL/6 mice are synchronized [29] ( Figure S2 ) . Since we have analysed both female and male mice with a mixed genetic background ( C57BL/6 x 129/Sv ) , we determined the timing of the initial postnatal hair cycles in our mouse breeding stock used for these experiments to exclude any major deviations from what has been published ( summarised in Table S1 ) . The timing of the first postnatal telogen-anagen transition in 3–4 weeks old mice and the first postnatal anagen-catagen-telogen transition in 6–8 week old mice agreed fairly well with previous findings ( Table S1; Figure S2 ) [29] . Moreover , the HFs on the back skin of most control mice had entered the second and extended telogen at 7–8 weeks of age ( Table S1 ) . According to [29] ( Figure S2 ) , the second postnatal anagen is initiated at 12 weeks of age but we were unable to define a synchronized starting point of the second postnatal anagen in our control animals ( Table S1 ) . All mice were maintained at the animal facility at Umeå University under pathogen-free conditions . Generation of the Z/Lhx2-GFP transgenic mouse strain was essentially as described previously for the Z/AP double reporter mouse strain [39] , but the AP part in the Z/AP double reporter vector was replaced by Lhx2-GFP to generate the Z/Lhx2-GFP vector . E14 ES cells were electroporated ( Bio-Rad GenePulser ) with the Z/Lhx2-GFP vector and G418-resistent cells were selected in 350 µg/mL of G418 ( GIBCO , 11811-031 ) . ES cell clones were isolated and those showing the highest β-Gal activity and had a single integration of the construct was further selected and used for blastocyst injections at the Umeå Transgene Core Facility ( UTCF ) . Blastocysts were transferred to pseudo-pregnant females and chimeric offspring were crossed with C57BL/6 mice to obtain germ-line transmission of the Z/Lhx2-GFP transgene . The Z/Lhx2-GFP mouse strain with the highest β-Gal activity in HFs was chosen for further experiments ( Figure S7B and S7C ) . The mouse strain with a hypomorphic allele of Lhx2 ( Lhx2Neo ) was generated by inGenious Targeting Laboratory Inc . ( www . genetargeting . com ) . The presence of the Neo gene in the intact Lhx2 locus between exon 2 and 3 in the opposite transcriptional orientation was confirmed by Southern blot analyses on ES cells and by PCR on tail biopsies ( Figure S4A ) . Mice containing a floxed allele of Lhx2 were obtained by crossing Lhx2Neo/+ mice with mice transgenic for the Flp recombinase as this will generate offspring where Neo is deleted and leave two loxp sites flanking exon 2 ( Figure S4B ) . We have also confirmed the phenotype using another Lhx2flox mouse strain where exons 2 and 3 are flanked by loxp sites [32] , which are the exons deleted in the conventional knock-out [22] . The mutated Lhx2 gene is expressed but Lhx2 mRNA missing either exon 2 or exons 2 and 3 are unable to make Lhx2 protein containing any functional domains since both transcripts are out-of-frame down-stream of exon 1 . The Z/Lhx2-GFP and CreER [33] transgenes , and the Lhx2Neo and Lhx2flox alleles were maintained on a mixed genetic background ( C57BL/6 X 129/Sv ) . The Lhx2flox/flox and Lhx2flox/- mice are healthy and fertile . Primers used to identify the Lhx2Neo and Lhx2flox alleles were: DL1 5′-GTTCTAGAAGTGGAAGGGGAGTGG-3′ , LOX 5′-GCCAGACTAGCAGACGCTGC-3′ , SDL2 5′-CCACCGGTACTCCTCTTCAGAG-3′ , UNI 5′-AGCGCATCGCCTTCTATCGCCTTC-3′ , AT1 5′-CACTCCGAGCCTGTTTGGTG-3′ . Primers used to identify the Z/Lhx2-GFP transgene were GFPforward 5′-TTCCACCATATTGCCGTC-3′ and GFPreverse 5′-AGAACTTGCCGCTGTTCA -3′ . The morning of the vaginal plug was considered as E0 . 5 . Dorsal ( back ) skin was shaved with electric razor and 1 mg of 4-hydroxytamoxifen ( Sigma H7904 ) dissolved in acetone was applied to the shaved area once a day for 7–9 consecutive days . Skin was isolated and fixed in 4% paraformaldehyde ( PFA ) in PBS at 4°C , transferred to 30% sucrose in PBS for 24 hours at 4°C , mounted in Tissue-Tek ( Sakura ) and stored at −80°C . Sectioning ( 8 µm ) was performed in a cryostat ( Microm HM505E ) and collected on Superfrost Plus slides ( Menzel-Gläser ) . For hematoxylin-eosin staining tissue sections were incubated in Mayer’s hematoxylin solution for 2 min , water 15 min , eosin solution 2 min , 95% ethanol 2×1 min , 99% ethanol 2×1 min and in xylene for 5 min . The slides were mounted with DPX mounting media ( VWR ) . For β-Gal staining skin tissue fixed for 30 min in PFA was cryosectioned , washed for 3×20 min in wash buffer ( 0 . 1M phosphate buffer , 2 mM MgCl2 , 5 mM EGTA , 0 . 02% NP40 and 0 . 01% sodium deoxycholate ) and subsequently incubated in wash buffer supplemented with 1 mg/ml 5-bromo-4-chloro-3-indolyl-D-galactopyranoside ( X-gal , Austral ) , 5 mM potassium ferrocyanide and 5 mM potassium ferricyanide , for 4–8 hours at room temperature . The reaction was stopped with 3×5 min washes with PBS , and sections were mounted in 87% glycerol . In situ hybridization using DIG labelled probes was performed essentially as described previously [55] , [56] . Skin fixed in PFA for 24 hours was sectioned and treated with 10 µl/ml proteinase K ( Roche ) in 0 . 1 M PBS for 15 min at room temperature prior to hybridization . The DIG-signal was visualized with AP-conjugated anti-DIG Fab2 fragments and developed using NBT and BCIP ( Roche ) . The following probes were used: Lhx2 ( NM_010710 , full length cDNA nucleotides 460–1750 , exon 2 nucleotides 587–789 , exon 2–3 nucleotides 643–1172 ) , GFP ( hrGFP , complete coding region ) , Shh ( NM_017221 , nucleotides 1–1715 ) , Smo ( NM_176996 , nucleotides 2957–3813 ) , Ptc1 ( NM_008957 , nucleotides 1–841 ) , Lef1 ( BC057543 , nucleotides 1678–2999 ) , BMP4 ( NM_007554 , nucleotides 117–578 ) , and Hist1h3c ( NM_175653 , nucleotides 27–480 ) . Immunohistochemistry was performed essentially as previously described [57] . Skin fixed for 30 min or 2 hours was sectioned and slides were washed 3×5 min in TBS ( 50 mM Tris-HCl pH 7 . 4 , 150 mM NaCl ) and blocked with 10% FCS in TBST ( TBS with 0 . 1% Triton X-100 ) for 20 min . The primary antibodies rat anti-CD34 ( MEC 14 . 7 , Abcam , dilution 1:50 ) and rat anti-E-cadherin ( Zymed Laboratories Inc . , dilution 1∶300 ) were applied to slides over night at 4°C diluted in TBST with 5% FCS . After 3×5 min washing in TBST , the secondary antibody , donkey anti-rat IgG labelled with Alexa Fluor 488 ( Molecular Probes , dilution 1∶400 ) , was added together with DAPI for 1 hour at room temperature . The primary rabbit anti-mouse Lhx2 ( generous gift from Dr . Sara Wilson , dilution 1∶32 , 000 ) was applied to slides over night . After 3×5 min washing in TBST , the secondary antibody , Cy3-conjugated donkey anti-rabbit IgG ( Jackson ImmunoResearch Laboratories Inc . , dilution 1∶400 ) , was added together with DAPI for 1 hour at room temperature . Slides were subsequently washed 3×5 min in TBST before mounting with fluorescence mounting medium ( Vectashield , Vector Laboratories ) . For the immunolabelling of the active ( dephosphorylated ) form of β-catenin , skin fixed for 2 hours was used . Slides were washed 3×5 min in TBST , blocked for endogenous peroxidase activity in 2% H2O2 , 80% methanol for 20 min , washed , incubated in citrate buffer ( 10 mM sodium citrate , pH 6 . 0 ) at 95°C for 20 min for epitope retrieval and washed . After blocking with 10% FCS in TBST for 1 hour , the primary mouse anti-β-catenin antibody ( Upstate 05-665 , dilution 1∶400 , recognise non-phosphorylated Ser-37 and Thr-41 ) was applied to the slides over night at 4°C . After washing , the slides were exposed to the secondary biotinylated anti-mouse antibody ( dilution 1∶400 ) for 2 hours at room temperature , washed and incubated for 1 hour in Avidin-Biotin Complex ( ABC ) solution ( Vector; 2% buffer A , 2% buffer B and 10% FCS in TBST ) , washed 3×5 min in TBST and developed in DAB-solution ( Sigma ) with 0 , 01% H2O2 for 5 min . After thorough washing in cold PBS , the slides were mounted in 87% glycerol . Skin fixed in PFA was used for detection of AP activity . The sections were fixated in acetone at −20°C for 5 min , washed for 3×5 min in PBS , rinsed 5 min in TN buffer pH 9 . 5 ( 0 . 1 M Tris-HCl , 0 . 1 M NaCl ) and exposed to 125 . 6 µg/ml NBT and 63 µg/ml BCIP in TN buffer for 20 min in darkness . Sections were washed with PBS 3×5 min to stop the reaction and subsequently mounted in 87% glycerol . The number of HFs in embryos was calculated from two sections from two different control embryos and two sections from two different Lhx2Neo/Neo embryos at E16 . 5 , and four sections from two control animals and two sections from two Lhx2Neo/Neo embryos at E18 . 5 . Statistical analyses were carried using Student’s t-test or the Chi-square method . | Hair is generated in hair follicles , complex mini-organs in the skin that are devoted to this task . All hair follicles are generated during embryonic development . The hair follicles generate a new hair shaft by cycling through stages of regression , rest , and growth continuously throughout life . The length of the growth phase determines the length of the hair . The reason ( s ) for this complicated regulation of hair growth is not clear , but it has been suggested that it may accommodate seasonal variations in hair growth . In this study we have identified the transcription factor Lhx2 as an important regulator of hair formation . The Lhx2 gene is active during the growth phase of the hair follicle and is turned off during the resting phase . We confirm that Lhx2 is functionally involved in hair formation , since hair follicles where Lhx2 has been inactivated are unable to make hair . Moreover , activation of the Lhx2 gene in hair follicles induced the growth phase and hence hair formation . Thus , Lhx2 is an important regulator of hair growth . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"dermatology",
"developmental",
"biology",
"cell",
"biology/gene",
"expression"
]
| 2010 | Cyclic Expression of Lhx2 Regulates Hair Formation |
MicroRNAs have been shown to be important regulators of inflammatory and immune responses and are implicated in several immune disorders including systemic lupus erythematosus and rheumatoid arthritis , but their role in Lyme borreliosis remains unknown . We performed a microarray screen for expression of miRNAs in joint tissue from three mouse strains infected with Borrelia burgdorferi . This screen identified upregulation of miR-146a , a key negative regulator of NF-κB signaling , in all three strains , suggesting it plays an important role in the in vivo response to B . burgdorferi . Infection of B6 miR-146a−/− mice with B . burgdorferi revealed a critical nonredundant role of miR-146a in modulating Lyme arthritis without compromising host immune response or heart inflammation . The impact of miR-146a was specifically localized to the joint , and did not impact lesion development or inflammation in the heart . Furthermore , B6 miR-146a−/− mice had elevated levels of NF-κB-regulated products in joint tissue and serum late in infection . Flow cytometry analysis of various lineages isolated from infected joint tissue of mice showed that myeloid cell infiltration was significantly greater in B6 miR-146a−/− mice , compared to B6 , during B . burgdorferi infection . Using bone marrow-derived macrophages , we found that TRAF6 , a known target of miR-146a involved in NF-κB activation , was dysregulated in resting and B . burgdorferi-stimulated B6 miR-146a−/− macrophages , and corresponded to elevated IL-1β , IL-6 and CXCL1 production . This dysregulated protein production was also observed in macrophages treated with IL-10 prior to B . burgdorferi stimulation . Peritoneal macrophages from B6 miR-146a−/− mice also showed enhanced phagocytosis of B . burgdorferi . Together , these data show that miR-146a-mediated regulation of TRAF6 and NF-κB , and downstream targets such as IL-1β , IL-6 and CXCL1 , are critical for modulation of Lyme arthritis during chronic infection with B . burgdorferi .
Lyme Disease is caused by infection with Borrelia burgdorferi , a tick-borne spirochete [1] , and is the most common vector-borne disease in the United States with an estimated 300 , 000 cases per year [2] . Often , infection leads to acute arthritis in humans . Clinical manifestations of Lyme arthritis include inflammatory cell infiltration , edema , synovial hyperplasia and remodeling of bone and connective tissue [3] , [4] . In some cases , infection can induce autoimmunity , despite treatment with antibiotics [5] . The reason why arthritis fails to resolve remains poorly understood , but is believed to be the result of dysregulation of host immune response to infection [6] . Several inbred mouse strains exhibit varying degrees of disease severity similar to human patients [7] , [8] . Whereas the C57BL/6 ( B6 ) mouse strain develops mild arthritis , C3H and various knockout strains such as B6 IL10−/− mice develop moderate to severe arthritis [7] , [9] . Furthermore , the intensity of the inflammatory response for a given spirochete burden varies greatly among strains , implicating host immune response as driving arthritis development [9] , [10] . Our laboratory and others have used the mouse model system to elucidate key regulators of host immune response to infection . Since its discovery , nuclear factor-kappa B ( NF-κB ) has been identified as a key regulator in many cellular functions including inflammation and cancer [11] . B . burgdorferi lipoproteins are extremely potent activators of Toll-like receptor 2 ( TLR2 ) -mediated NF-κB activation and cytokine production , and are important for host defense [12]–[16] . Mice lacking TLR2 or the adapter protein myeloid differentiation primary response gene ( 88 ) ( MyD88 ) exhibit a failure to control infection [14] , [17]–[21] . Although these knockout studies clearly demonstrate an important role of NF-κB in host defense , elucidating its role in inflammation and Lyme arthritis has remained elusive . While NF-κB activation is critical in response to infection , downregulation is equally important to avoid excess inflammation , tissue damage and autoimmunity [22] . MicroRNAs ( miRNAs ) have recently been identified as being important regulators of NF-κB [23] and autoimmunity [24] . These small regulatory RNAs are posttranscriptional regulators of gene expression [25] , and one miRNA , miR-146a , has been shown to be a modulator of innate immune response to TLR ligands [26] . Targets of miR-146a include TNF receptor associated factor 6 ( TRAF6 ) and IL-1 receptor associated kinase 1 ( IRAK1 ) , adaptor molecules downstream of the MyD88-dependent TLR and cytokine signaling pathways [27] . Importantly , miR-146a itself is upregulated by IL-1β and TLRs , including TLR2 , and thus acts as a negative feedback regulator of NF-κB signaling which is required for immune homeostasis in vivo [27]–[31] . Aberrant microRNA expression , particularly miR-146a , has been associated with a variety of inflammatory disorders [32] . In systemic lupus erythematosus , a functional variant in the miR-146a promoter is associated with disease risk [33] , and abnormally low miR-146a expression has been associated with more severe symptoms [34] . In contrast , rheumatoid arthritis synovial fibroblasts express abnormally high levels of miR-146a [35] , [36] , while osteoarthritis chondrocytes express variable levels miR-146a , correlating with disease severity [37] , [38] . Despite correlative evidence linking aberrant miRNA expression to diseases such as lupus , RA and OA , determining whether miRNAs play an active role in pathogenesis has yet to be elucidated , and to our knowledge , no studies have examined the role of miRNAs in Lyme arthritis . For these reasons , we sought to determine whether changes in miRNA expression contributed to host defense and Lyme arthritis development during B . burgdorferi infection .
MicroRNA dysregulation has been associated with a number of inflammatory disorders , and we hypothesized that these may play an important role in response to B . burgdorferi infection and Lyme arthritis development . We therefore performed a genome-wide screen of changes in miRNA expression in joints of B6 , C3H and B6 IL-10−/− mice infected with B . burgdorferi at one and two weeks post-infection using an Agilent mouse microRNA microarray ( Table 1 , Table S1 ) . MicroRNAs differentially regulated included many that have been identified previously as important regulators of immune function . Interestingly , each infection model had a unique miRNA expression “signature , ” and we found that only a few dozen miRNAs showed changes in expression during infection . Most of these changes were in C3H mice , and may be due to both differences in inflammatory response and intrinsic differences in miRNA function between strains . At two weeks post-infection , two miRNAs , miR-21 and miR-146a , both induced by NF-κB and associated with TLR signaling , were the most highly upregulated in all three strains ( Table 1 ) , and were confirmed using qRT-PCR ( Figure 1 ) . Furthermore , these miRNAs maintained high expression , even at 4 weeks post-infection . Interestingly , miR-155 was significantly upregulated in B6 IL10−/− mice , but not in B6 or C3H mice . This microRNA is a proinflammatory NF-κB-induced miRNA associated with T cell-dependent inflammation and autoimmunity [39]–[41] , and expression is suppressed by IL-10 [42] . Of these , miR-146a was of particular interest , given recent reports showing a link between miR-146a and susceptibility to a variety of inflammatory disorders . Targets of miR-146a , IRAK1 and TRAF6 [27] , are involved in TLR2/NF-κB activation , which is an important pathway in controlling B . burgdorferi infection [13] , [14] , [17]–[21] . Also , the observation that miR-146a was upregulated in all three strains suggested that this miRNA likely plays a general role in regulating the immune response to B . burgdorferi . For these reasons , our focus turned to studying miR-146a . A B6 miR-146a−/− knockout mouse was recently generated [28] , which provided a powerful tool to evaluate the role of miR-146a in mildly arthritic B6 mice . While miR-146a was also upregulated in arthritis-susceptible C3H mice , we suspected that other genetic factors play a dominant role in arthritis development , including excessive Type I IFN production [43] , [44] and accumulation of undigested glycosaminoglycans in joint tissue [45] . These effects may limit the ability of miR-146a to modulate arthritis development in the C3H mouse model . It is tempting to speculate , however , that lack of miR-146a in the arthritis-susceptible C3H mouse would lead to even more severe arthritis , as has been reported in the C3H IL-10−/− mouse model [46] . Since miR-146a is an important negative regulator of NF-κB activation , we hypothesized that a B6 mouse deficient in miR-146a would develop more severe arthritis during infection with B . burgdorferi compared to WT controls . To avoid age-related pathologies associated with B6 miR-146a−/− mice [30] , we used 6–8 week-old mice , which is also the age of optimal arthritis in other mouse strains . Arthritis was assessed in B . burgdorferi-infected B6 and B6 miR-146a−/− mice . At four weeks post-infection , B6 miR-146a−/− mice developed significantly more severe arthritis . Several markers of arthritis were elevated in B6 miR-146a−/− mice , including ankle swelling ( Figure 2A ) , number and severity of lesions observed , polymorphonuclear ( PMN ) cell infiltrate , reactive/reparative score ( periosteal hyperplasia and new bone formation and remodeling ) , and tendon sheath thickness ( Table 2 ) . Cranial tibial tendon is enlarged in Figure 2B . Control ( BSK-injected ) mice showed no significant arthritis in either strain , and no significant difference between strains was seen in mononuclear cell infiltrate into inflammatory processes . Importantly , B6 miR-146a−/− mice did not display overwhelming numbers of bacteria; rather , they tended to have somewhat fewer bacteria in infected joints and similar burden in infected heart and ear tissue , as measured by B . burgdorferi-specific16S rRNA normalized to β-actin in joints and heart , and recA normalized to mouse nidogen in ear tissue ( Figure 2C ) . This difference in bacterial load in joint tissue was likely not due to differences in antibody response , since B . burgdorferi-specific IgM and IgG levels were similar between the two strains at two weeks and four weeks post-infection , respectively ( Figure 2D ) . While this does not rule out the possibility that different borrelial proteins could be opsonic targets in the two strains , these data support the notion that increased arthritis observed in B6 miR-146a−/− mice was likely due to a defect in regulation of host immune function rather than compromised host defense . In fact , the decrease in 16S rRNA in B6 miR-146a−/− mouse joints indicated that arthritis development was independent of bacterial density . This increased arthritis severity with accompanying decreases in bacterial burden is also observed in the B6 IL10−/− mouse model of arthritis [9] , [47] , [48] , and is believed to be due primarily to enhanced innate immune responses [49] . In addition to joints , the heart is another target of B . burgdorferi infection in mice . We therefore looked for evidence of miR-146a modulating inflammation in heart tissue . Mice are susceptible to Lyme carditis in an MHC-independent manner , and exhibit variation in disease severity , with C3H mice harboring a greater number of bacteria and developing more severe carditis and B6 mice being resistant and harboring fewer bacteria [7] , [50] , [51] . Lyme carditis is also observed in humans , and although rare , can be fatal [52] . To assess the role of miR-146a in modulating heart inflammation , B6 , B6 miR-146a−/− and C3H mice were infected with B . burgdorferi for 3 weeks and hearts were removed and assessed for bacterial numbers and changes in transcripts of inflammatory genes . As was seen at 4 weeks post-infection ( Figure 2C ) , bacterial burden , as measured by qRT-PCR analysis of B . burgdorferi 16S rRNA , was similar between B6 and B6 miR-146a−/− mice , and both trended lower than what was seen in heart tissue from C3H mice ( Figure 3A ) . Lesions in the heart were also scored for carditis in B6 , B6 miR-146a−/− and C3H mice at 3 weeks post-infection . Overall lesion scores were similar in B6 and B6 miR-146a−/− mice , and both trended lower than lesion severity in C3H mice ( Figure 3B ) . Lesions in hearts at 3 weeks post-infection were characterized by acute to subacute vasculitis/perivasculitis ( see Figure 3C ) of the 1 ) microvasculature ( capillaries ) at the base of the heart ( in heart muscle ) where the aorta and pulmonary arteries arise , 2 ) microvasculature ( capillaries ) within connective tissues supporting these arteries , and 3 ) microvasculature ( capillaries ) of the vasa vasorum of the aorta and pulmonary arteries . These lesions affected the vascular system , rather than being primary lesions of the heart muscle ( myocarditis ) . The character and pattern of distribution of these lesions suggested that inflammation of the microvasculature is the result of some type of localized target cell ( i . e . , endothelium ) or target substance ( i . e . , bacteria ) specificity for this location consistent with Borrelial adhesin-host ligand binding within the vascular endothelium [53] , [54] . Vascular turbulence or oxygen concentration could also be involved . Lyme carditis is associated with macrophage infiltration [50] , and invariant NKT cells have been shown to play a protective role in B6 mice [55] . We therefore used PCR analysis of macrophage ( CD11b and F4/80 ) and NKT cell ( CD4 and Vα14 ) markers to assess changes in cellularity in infected heart tissue ( Figure 3D ) , as performed previously [56] , [57] . Significant upregulation of macrophage markers CD11b and F4/80 was observed in all three strains , as expected based on previous research [50] . The magnitude of upregulation was not different between B6 and miR-146a deficient mice , indicating no role of miR-146a feedback on inflammation in this tissue . Interestingly , while CD11b and F4/80 transcripts were also significantly upregulated in C3H mice , the degree of upregulation was somewhat less than upregulation observed in B6 and B6 miR-146a−/− mice at 3 weeks post-infection . Because these data are from whole heart tissue , they reflect cumulative changes in myeloid cell numbers from the entire heart , including changes in resident cardiac macrophages [58] , which may dilute out lesion-specific changes identified by histopathology . CD4 transcript levels were not significantly different between strains , but while both B6 and B6 miR-146a−/− mice contained similar levels of Vα14 that trended higher at 3 weeks , C3H mice had very low levels of this transcript in both uninfected and infected heart tissue . This is consistent with significant variation of NKT cell numbers between different mouse strains [57] . PCR analysis was also used to determine changes in expression of various inflammatory cytokines and chemokines in heart tissue ( Figure 3D ) . As previously reported , C3H mice had elevated levels of IFNγ transcripts in infected heart tissue [59] , which was significantly higher than IFNγ mRNA in B6 and B6 miR-146a−/− hearts . This trend was also observed for IL6 transcripts , although there was significant variation in expression within C3H mice . No differences among B6 , B6 miR-146a−/− and C3H mice were observed for transcripts of IL1β , TNFα , Cxcl1 , Cxcl2 , Ccl2 , IL10 or IL12 ( data not shown ) . Together , these data suggest that the nature of host defense , macrophage and NKT cell proliferation and infiltration , as well as cytokine and chemokine expression , is very similar in infected B6 and B6 miR-146a−/− heart tissue , but is quite distinct from observations in carditis-susceptible C3H mice . Fundamental differences between strains have been reported previously in Lyme carditis studies comparing the effect of Stat1 [60] and Ccr2 [51] deficiencies on carditis-susceptible and carditis-resistant mouse strains . Together , these data show that while strain-specific variables influence differences in carditis susceptibility between B6 and C3H mice , miR-146a has no impact on carditis severity in B6 mice . Because miR-146a is known to negatively regulate NF-κB activation , we compared transcripts of genes upregulated by NF-κB in infected joint tissue from B6 and B6 miR-146a−/− mice at 2 and 4 weeks post-infection . We observed that a number of NF-κB inducible genes were significantly elevated at 4 weeks post-infection ( but not at 2 weeks ) by qRT-PCR in B6 miR-146a−/− joints , compared to WT , including cytokines IL-1β and IL-6 , as well as neutrophil chemokines Cxcl1 and Cxcl2 ( Figure 4A ) . CXCL1 has been shown to be required for full arthritis development in C3H mice [61] , [62] , and increased expression of this gene in B6 miR-146a−/− mice could be directly contributing to arthritis development through recruitment of neutrophils . This is supported by the increase in PMN infiltrate seen at 4 weeks by histopathology , shown in Table 2 . Elevated IL-1β transcript level is also particularly interesting , since the IL-1 receptor ( IL-1R ) uses the same adaptors as TLR2/1 for signal transduction , is strongly upregulated by myeloid cells during phagocytosis of B . burgdorferi [63] , and is dependent on IRAK1 and TRAF6 , two miR-146a targets [27] . Furthermore , IL-1β stimulates miR-146a upregulation in vitro [27] , suggesting that this miRNA negatively regulates IL-1β signaling . It is important to note that uninfected B6 miR-146a−/− mice did not exhibit any abnormalities in expression of these genes , indicating that this hyperactivity is due to a failure to down-regulate the NF-κB response after infection , rather than general NF-κB hyperactivity , as is observed in aging B6 miR-146a−/− mice [30] . Only a distinct subset of inflammatory cytokines and chemokines ( IL-1β , IL-6 , Cxcl1 , Cxcl2 ) appeared to be dysregulated in B6 miR-146a−/− joints . Transcript levels of other Lyme arthritis-associated genes ( IFNγ , Cxcl10 , TNFα ) were very similar between the two strains ( Figure 4A ) , and showed a peak in expression at 2 weeks post-infection , followed by resolution at 4 weeks . This is in contrast to arthritis-susceptible B6 IL10−/− mice , where previously published data show that in addition to upregulation in IL-1β , IL-6 , Cxcl1 and Cxcl2 , IFNγ and Cxcl10 are upregulated 16-fold and 141-fold at 2 weeks , and 22-fold and 189-fold at 4 weeks , respectively [47] . These data together indicate that the B6 miR-146a−/− mouse is distinct from the B6 IL10−/− model , which is associated with a dramatic IFNγ signature in joints and elevation of IFNγ in serum at 4 weeks p . i . [47] , [48] . In order to determine whether there was systemic dysregulation of NF-κB-inducible cytokines , serum was collected from B . burgdorferi-infected mice at 4 weeks post-infection and cytokine levels were measured by enzyme-linked immunosorbent assay ( Figure 4B ) . B6 miR-146a−/− mice contained higher levels of IL-6 at 4 weeks post-infection , compared to wild-type , consistent with observations in joint tissue . TNFα and IL-12 serum levels were very similar between strains , and although levels of IFNγ varied widely in B6 miR-146a−/− mice , they were not significantly greater than B6 levels . To identify the effect of miR-146a in various joint cell populations during the early phase of infection ex vivo , we digested joints with purified collagenase to release cells into a single-cell solution in order to identify and isolate cell fractions based on lineage markers , including CD45 for leukocytes , CD11b for myeloid cells , CD31 for endothelial cells , and CD29 for fibroblast-enriched cells ( Figure 5A ) . This method has been used in C3H mice to identify cellular sources of genes associated with the arthritogenic Type I IFN response early in infection , and is a sensitive assay to observe cell type-specific effects ex vivo that might be missed using whole joint tissue [43] . Using this method , we were able to determine the effect of miR-146a on specific cell types early in infection ( Figure 5B ) . Levels of IL-1β , while trending higher in myeloid cells isolated from B6 miR-146a−/− mice , were not significantly different between the two strains . In B6 mice , three genes , Cxcl2 and the IFN-inducible gene Oasl2 ( in myeloid cells ) and Cxcl1 ( in fibroblasts ) , tended to peak in expression at Day 7 post-infection . In contrast , transcripts were somewhat higher in uninfected B6 miR-146a−/− cell fractions vs . WT , and remained elevated throughout infection . This suggests that B6 miR-146a−/− mice may be poised to initiate a hyperactive immune response . There was no difference in lymphoid IFNγ expression between strains , which peaked at Day 14 post-infection , as was seen in whole joint tissue ( Figure 4A ) . It is important to note that , unlike published observations in C3H mice [43] , B6 miR-146a−/− mice did not exhibit a robust induction of IFN-responsive genes , such as Oasl2 , in fibroblasts or endothelial cells at Day 7 post-infection ( data not shown ) . Overall , these data , combined with data from Figure 4 , suggest that a number of NF-κB-inducible genes in B6 miR-146a−/− mouse joints are poised for hyper-activation prior to infection , and peak at 4 weeks post-infection , indicating that miR-146a acts to resolve the inflammatory response late in infection , rather than limiting the amplitude of inflammation during early stages of infection . During cell sorting , we observed differences in cellular infiltrate , particularly in the myeloid cell lineages , in joints during infection . Therefore , a more rigorous analysis of myeloid cells recruited to the joint by flow cytometry was performed at various times during infection . Using joint cell isolation methods described in Figure 5 , myeloid cells were characterized from infected joints at 2 and 4 weeks post-infection using fluorescently labeled antibodies against CD11b , F4/80 , Ly6C , Gr1 and CD206 . CD11b+ myeloid cell populations clustered roughly into three populations , F4/80+ Ly6Clo macrophages , Gr1hi Ly6Cint PMNs and Gr1int Ly6Chi monocytes ( Figure 6A&B ) . Furthermore , F4/80+ Ly6Clo macrophages expressed variable levels of CD206 ( MRC1 , Mannose Receptor C type 1 ) , a marker of alternatively activated ( M2-like ) macrophages [64] . There was little difference between strains in mean fluorescence intensity ( MFI ) of MRC1 and Gr1 in each myeloid subpopulation , suggesting that they were phenotypically similar populations . However , while the number of these three myeloid populations in B6 mouse joints changed only modestly in B6 mice , myeloid cell numbers in B6 miR-146a−/− joints were significantly elevated at both 2 and 4 weeks post-infection ( Figure 6C ) . An increased trend in PMN infiltration in B6 miR-146a−/− mice is also consistent with histopathology data shown in Table 2 , despite the propensity of PMNs to lyse during enzymatic digestion of joint tissue , resulting in some sample-to-sample variation . Interestingly , there was little difference between strains in infiltrating lymphoid cells at 2 or 4 weeks post-infection ( Figure S1 ) . These data , as well as the observation of similar B . burgdorferi-specific antibody levels ( Figure 2D ) , suggest that arthritis and host defense phenotypes observed in B6 miR-146a−/− mice shown in Figure 2 and Table 2 are driven primarily by myeloid cells . The data from Figure 6 implicated myeloid cells as contributors of arthritis development in B6 miR-146a−/− mice . We therefore turned to bone marrow-derived macrophages ( BMDMs ) to elucidate the molecular mechanism of miR-146a regulation of NF-κB during B . burgdorferi infection . BMDMs were cultured from bone marrow extracted from B6 or B6 miR-146a−/− mice and treated with B . burgdorferi for 6 and 24 hours . We then measured transcripts of IL1β , IL6 and TNFα ( Table 3 ) . Transcripts of IL1β were approximately 4-fold higher in B6 miR-146a−/− BMDMs , vs . WT , at both 6 and 24 hours , and IL6 levels were 7 . 5-fold higher at 6 hours and 2 . 5-fold higher at 24 hours post-stimulation . Interestingly , TNFα transcripts were only 20–30% higher in B6 miR-146a−/− BMDMs , compared to WT . Transcripts for all three cytokines were very low in uninfected cells , and were similar between the two strains ( data not shown ) . This suggests that miR-146a effect on IL1β and IL6 regulation is greater than its effect on TNFα expression . We also measured levels of several NF-κB-inducible cytokines by ELISA in cell supernatant from both B6 and B6 miR-146a−/− BMDMs at 24 hours post-stimulation , including TNFα IL-1β , IL-6 and IL-12 , CXCL1 and IL-10 ( Figure 7A ) . After 24 hours of treatment with B . burgdorferi , three cytokines , IL-1β , IL-6 and IL-12 , and the neutrophil chemokine CXCL1 , were more abundant in B6 miR-146a−/− cell supernatant than in B6 cell supernatant , consistent with hyperactive NF-κB activation and transcript analysis ( Table 3 ) . Interestingly , TNFα , an early-response NF-κB cytokine , did not share this trend , which may be due to the relatively late time point used for this analysis [65] . Production of IL-10 was robust in both strains , although somewhat greater in miR-146a-deficient BMDMs . Previous work from our laboratory showed that many macrophages are IL-10 producers in joints of B6 mice [48] . Also , macrophages produce high levels of IL-10 when treated with B . burgdorferi in vitro , which is important in regulating bacterial persistence [49] and immune response [46] , [66]–[68] . Data from Figure 6B also showed that many macrophages in joints of infected B6 and B6 miR-146a−/− mice express the alternatively activated macrophage marker MRC1 . While it is difficult to accurately determine the range of macrophage phenotypes present in joints , we used BMDMs pretreated with IL-10 as an in vitro model to study miR-146a effects on IL-10-stimulated macrophages . BMDMs were treated with 1ng/ml IL-10 for 4 hours prior to 24-hour B . burgdorferi stimulation . Surprisingly , while pretreatment with IL-10 led to an approximately 80% reduction in IL-6 production in B6 BMDMs , IL-10-mediated suppression of IL-6 in B6 miR-146a−/− BMDMs was drastically reduced , with only ∼20% decrease in IL-6 production after IL-10 pretreatment , indicating that IL-10 was unable to effectively suppress IL-6 expression in the absence of miR-146a . These data are consistent with in vivo data showing consistently elevated IL-6 protein in serum from 4 week-infected B6 miR-146a−/− mice in Figure 4 . However , IL-10 pretreatment did lead to significantly reduced IL-12 and TNFα production in both strains , as well as high production of IL-10 , after B . burgdorferi treatment , consistent with an anti-inflammatory M2-like phenotype . Both IL-1β and CXCL1 remained higher in B6 miR-146a−/− BMDMs compared to B6 BMDMs , although IL-1β levels were unaffected , and CXCL1 levels were modestly reduced by IL-10 pretreatment . Importantly , levels of IL-12 , TNF-α and IL-10 were very similar between the two strains , suggesting there was no miR-146a-mediated defect in M2 polarization in response to IL-10 pretreatment . This is consistent with in vivo observations , where TNFα , IL-12 and IFNγ serum protein levels were not significantly elevated in B6 miR-146a−/− mice at 4 weeks post-infection , relative to B6 mice ( Figure 4B ) . The role of microRNAs , including miR-146a , during inflammatory responses involves suppressing distinct mRNA targets , depending on cell type [69] . It was therefore important to determine the mRNA target most affected at the protein level by the presence or absence of miR-146a in BMDMs . Immunoblot analysis was performed on protein extracts from B6 and B6 miR-146a−/− BMDMs treated for 24 hours with B . burgdorferi to measure protein levels of three targets of miR-146a , TRAF6 , IRAK1 and STAT1 ( Figure 7B ) . TRAF6 protein expression was elevated over two-fold in both resting and stimulated B6 miR-146a−/− BMDMs compared to B6 , while protein levels of IRAK1 were similar between strains . STAT1 protein was also higher in resting B6 miR-146a−/− BMDMs compared to B6 , but this difference between strains was not observed after 24 hours stimulation . Transcript analysis of Traf6 , Irak1 and Stat1 also show this trend ( Figure 7C ) . It is interesting that in the case of TRAF6 , the difference observed at the protein level was greater than that seen at the transcript level , where transcripts were typically only 30–50% greater in B6 miR-146a−/− BMDMs vs . B6 BMDMs , suggesting that miR-146a effect on translational inhibition is more pronounced than its effect on mRNA stability . This is consistent with a growing body of evidence suggesting that microRNA-mediated translational repression is dependent on inhibition of translation initiation , rather than mRNA degradation [70] , [71] . The difference between protein and transcript levels of these three genes ( Figure 7B , C ) strongly suggests that posttranscriptional regulatory mechanisms including , but not limited to , microRNA-mediated repression , play an important role in determining cellular protein levels . STAT1 is known to be regulated by a large number of posttranslational modifications that affect function [72] . Both STAT1 and IRAK1 protein levels have been shown to be tightly regulated through ubiquitin E3 ligase-directed degradation [73] , [74] . In the case of IRAK1 and STAT1 , these data suggest that miR-146a-independent regulatory mechanisms seem to be dominant compared to miR-146a-mediated regulation . Taken together , TRAF6 protein levels appear to be the most sensitive to the presence or absence of miR-146a in myeloid cells , and imply miR-146a-mediated translational repression of TRAF6 is required to properly regulate production of NF-κB-induced cytokines in response to B . burgdorferi . The lack of difference in STAT1 protein level is also consistent with a failure to observe significant differences between B6 and B6 miR-146a−/− mice in the IFN response ( Figures 4&5 ) . One possible explanation for reduced numbers of B . burgdorferi in joints of infected B6 miR-146a−/− mice is that macrophages lacking miR-146a are more highly phagocytic . In order to measure phagocytic activity , peritoneal macrophages were collected from B6 and B6 mir-146a−/− mice and stimulated with GFP-labeled B . burgdorferi for 1 or 2 hours at 10∶1 multiplicity of infection ( MOI ) . Phagocytosis of GFP-B . burgdorferi was measured by flow cytometry ( Figure 8A&B ) . At both 1 and 2 hours post-stimulation , peritoneal macrophages lacking miR-146a had significantly higher numbers of GFP+ cells , as well as a higher mean fluorescence intensity ( MFI ) for GFP in GFP+ macrophages . These data suggest that there are more B6 miR-146a−/− peritoneal macrophages associated with higher numbers of bacteria than wild-type cells . Flow cytometry was unable to distinguish localization of the cell-associated bacteria . To determine whether GFP-B . burgdorferi were intracellular or adhering to the cell surface , confocal microscopy was used to visualize the bacteria associated with peritoneal macrophages . Peritoneal macrophages were stimulated with GFP-B . burgdorferi at 100∶1 MOI for 1 hour and stained for the lysosomal protein LAMP1 ( red ) , the macrophage-specific surface protein F4/80 ( blue ) , and nuclei were stained with DAPI ( gray , Figure 8C ) . Bacteria were visible adhering to cell surface ( white triangle ) , inside macrophage pseudopodia ( white arrow ) and inside cells associated with LAMP1 ( white chevron ) . While bacteria adhering to the cell surface and inside pseudopodia had a spirochetal shape , bacteria associated with lysosomes were amorphous , and formed bright GFP puncta , indicative of bacterial degradation . These bright GFP puncta were predominant throughout the entire sample , as represented in the image in Figure 8C for B6 mice , and in Figure S2 for B6 miR-146a−/− mice . This indicates that phagocytosis occurs very rapidly as previously reported [65] , and the flow cytometry analysis infers that miR-146a modulates the level of phagocytic activity . Although the mechanism is unknown , similar transcript levels were seen for TLR2 , CD14 , as well as the scavenger receptor MARCO ( data not shown ) , which have been recently implicated in B . burgdorferi uptake [75]–[78] . Previous reports showing phagocytosis influencing cytokine production in human mononuclear cells [78] , and B6 MyD88−/− BMDMs being defective in bacterial internalization [79] , are consistent with B6 miR-146a−/− BMDMs having elevated cytokine production and enhanced phagocytic activity ( Figure 7 , Table 3 , Figure 8 ) . While more research is necessary to elucidate this mechanism , these data suggest that B6 miR-146a−/− macrophages have enhanced phagocytosis , and may help explain why joint tissue from B6 miR-146a−/− mice contains fewer numbers of spirochetes ( Figure 2C ) .
These data have allowed us to generate a model ( Figure 9 ) where miR-146a is upregulated during B . burgdorferi infection , and acts as a nonredundant suppressor of inflammation and arthritis ( Figures 1–2 ) . Interestingly , lack of miR-146a had no effect on heart inflammation and carditis ( Figure 3 ) , indicating fundamental differences between arthritis and carditis development . Differences in carditis severity between B6 and C3H mice are believed to be closely associated with differences in bacterial dissemination and clearance between the two strains [80] . This is consistent with the positive correlation between bacterial numbers and heart lesion severity in B6 , B6 miR-146a−/− and C3H mice ( Figure 3A–B ) , and with previous reports showing no correlation between quantitative trait loci associated with arthritis severity and bacterial numbers in heart tissue [81] , [82] . Importantly , differential contribution of NF-κB regulation was not predicted from studies with mice deficient in TLR2 and MyD88 , as both hearts and joint tissues displayed increased presence of B . burgdorferi [18] , [19] , [83] . Numerous studies have revealed different mechanisms of pathogenesis in carditis development and differing contributions of innate and adaptive responses in bacterial clearance and resolution of carditis and arthritis . For example , although antibody response plays an essential role in resolution of arthritis , greater roles of CD4+ T cells and iNKT cells as sources of IFNγ are reported in protection and resolution of Lyme carditis [55] , [60] , [84]–[86] . Other gene knockout and cytokine blocking studies have shown tissue-specific effects of IL-10 [46] and chemokines [51] , [62] on arthritis and carditis severity . These results suggest future microRNA studies on carditis should focus on those miRs known to influence the balance of CD4+ T cells [87] and iNKT cell function , such as miR-150 and miR-181a/b [88]–[90] . Myeloid cells respond to a variety of Borrelia stimuli through TLRs that lead to activation of NF-κB and upregulation of hundreds of genes involved in controlling infection and initiating the adaptive response . miR-146a is also upregulated , and is an important check on the amplitude and duration of the NF-κB response . In the absence of this microRNA , this response is dysregulated , leading to increased transcription of certain NF-κB-inducible cytokines and chemokines in infected joint tissue , primarily late in infection ( Figures 4–5 ) . Myeloid cells exhibit excessive proliferation and infiltration into joint tissue of B6 miR-146a−/− mice , have increased phagocytic activity and produce excess cytokines such as IL-1β , IL-6 and CXCL1 , leading to inflammation of synovial tissue and arthritis development ( Figures 6–8 ) . Regulation of the inflammatory response via a miR-146a-mediated negative feedback loop is critical for resolution of the NF-κB response during the persistent phase of infection , and mice lacking this miRNA are poised to develop arthritis upon infection with B . burgdorferi . NF-κB activation in response to B . burgdorferi infection is a double-edged sword . On one hand , NF-κB activation is critical in mounting an effective immune response to control infection; on the other hand , dysregulated activation leads to inflammation and arthritis . Because of the dual nature of NF-κB in inflammation and host defense , decoupling these two roles has been difficult . Knockout models using B6 TLR2−/− or MyD88−/− mice have shown the important role of NF-κB in host defense , but because these mice have such a severe innate defect in bacterial defense , elucidating the role of NF-κB in arthritis using these models has remained difficult . The B6 miR-146a−/− mouse model of Lyme arthritis is unique in that it effectively decouples these two roles , leaving the bactericidal function intact while increasing the amplitude of proinflammatory NF-κB activation . This has allowed us to identify its role in arthritis development , independent of its role in host defense , and suggests that miR-146a could be a valuable therapeutic target for control of inflammation without compromising ability to clear an infection . MicroRNAs are a unique class of regulatory molecules . Unlike transcription factors , they do not act as on/off switches , rather they function as “fine tuners” of gene expression [26] . We have taken advantage of this property to decouple the roles of NF-κB in host defense and inflammation . Young B6 miR-146a−/− mice are phenotypically similar to wild-type B6 mice , and it is only upon chronic exposure to inflammatory stimuli that immunological defects are seen [29] . Consistent with this , endotoxin tolerance is highly dependent upon miR-146a expression in THP-1 cells [91] . Using Lyme arthritis as a model , we have shown that mice lacking this key miRNA fail to adequately maintain immune homeostasis , and develop inflammatory arthritis during a chronic bacterial infection ( Figure 2 , Table 2 ) . This model is also distinct from other mouse models of Lyme arthritis . For example , C3H mice exhibit a robust Type I IFN expression profile early in infection , which contributes to arthritis , and is absent in the mildly arthritic B6 mouse [43] , [44] , [47] . This IFN response was absent miR-146a−/− mice , similar to B6 . Furthermore , B6 IL10−/− mice , a model for Th1-mediated arthritis , have a very pronounced IFNγ signature beginning at 14 days post-infection that persists for several weeks [48] . This pattern was also not observed in the arthritic B6 miR-146a−/− mice ( Figures 4–5 , Figure S1 ) . Additionally , while B6 miR-146a−/− and B6 IL10−/− mice both exhibit increased bacterial clearance , likely due to an enhanced myeloid response to phagocytosis of bacteria [63] , only B6 IL10−/− mice show enhanced antibody response [49] . It was somewhat surprising that B6 miR-146a−/− mice did not exhibit a strong T-cell-mediated phenotype , based on parameters tested , since other studies have shown an important role of miR-146a in regulating Th1 responses [31] , [92] . It is possible that the elevated myeloid response could eventually lead to a dysregulated T-cell response in some cases . Indeed , several B6 miR-146a−/− mice did have elevated serum IFNγ at 4 weeks post-infection , although this was the exception rather than the rule , and average levels did not achieve statistical significance compared to wild-type mice ( Figure 4C ) . It may also be possible that robust production of IL-10 seen in B6 miR-146a−/− mice is sufficient to suppress any T-cell dysregulation due to lack of miR-146a . Nevertheless , the results of this study show that arthritis is influenced principally by hyperactive myeloid cell activation . The role of miR-146a in regulating NF-κB activation was consistent with the observed defect in downregulation of NF-κB-dependent cytokines and chemokines IL-1β , IL-6 , Cxcl1 and Cxcl2 , in B6 miR-146a−/− mice at 4 weeks post-infection ( Figure 4 ) . Dysregulation of Cxcl1 in these mice was particularly interesting because previous studies have shown that C3H mice lacking CXCL1 have reduced neutrophil infiltration and arthritis [61] , [62] . This neutrophil chemokine is tightly regulated both at the transcriptional and posttranscriptional level by both TLR dependent and cytokine dependent mechanisms [93] . Data from Figure 7 suggest that excess cytokine production by B6 miR-146a−/− macrophages may lead to enhanced CXCL1 production by resident cells in vivo . Therefore , miR-146a , expressed primarily in leukocytes [29] , likely has cell-extrinsic effects on nonhematopoietic cell function and arthritis development . Recently , IL-6 has been shown to be an important downstream target of miR-146a in regulating hematopoiesis and myeloproliferation [29] . This is consistent with increased IL-6 production shown in Figure 4 and Figure 7 , and corresponding increase in myeloid cell infiltration into joint tissue ( Figure 6 ) . Thus , miR-146a-mediated regulation of several cytokines and chemokines likely has a combined effect on inflammatory responses . Increased phagocytic activity , as well as elevated IL-1β production ( Figures 7–8 ) point to a previously unrecognized role of miR-146a in phagocytosis and caspase-1 activation . While this role remains to be elucidated , previous research has shown that B . burgdorferi induces caspase-1-dependent IL-1β production , and caspase-1 is important for inflammatory cell influx into joint tissue [94] . Additionally , phagocytosis of live B . burgdorferi is a potent activator of IL-1 β in human PBMCs [95] . Targets of miR-146a have been studied in many cell types , and it is becoming increasingly evident that the modulatory effect of miR-146a is dependent on cell type and physiological condition . For example , STAT1 appears to be an important miR-146a target in regulatory T-cells [92] , and IRAK1 and TRAF6 both appear to be important miR-146a targets in splenocytes [31] and human monocytes [30] . This study highlights the particular role of miR-146a targeting TRAF6 in myeloid cells ( Figure 7 ) , indicating that miR-146a function is , to a certain degree , cell type-specific . Importantly , several observations in the B6 miR-146a−/− mouse model are recapitulated in Lyme disease patients . Joint fluid and synovial tissue from antibiotic-refractory Lyme arthritis patients contain higher levels of IL-6 and IL-1β , as well as Th1 cytokines and chemokines , compared with patients whose arthritis is resolved after antibiotic treatment , and IL-1β remains elevated in these treatment-refractory patients long after antibiotic therapy [96] , [97] . Thus , the B6 miR-146a−/− model of Lyme arthritis could be a useful tool in further understanding how regulation of NF-κB is related to Lyme disease pathogenesis .
Mice were housed in the University of Utah Comparative Medicine Center ( Salt Lake City , UT ) , following strict adherence to the guidelines according to the National Institutes of Health for the care and use of laboratory animals , as described in the Guide for the Care and Use of Laboratory Animals , 8th Edition . Protocols conducted in this study were approved and carried out in accordance to the University of Utah Institutional Animal Care and Use Committee ( Protocol Number 12-01005 ) . Mouse experiments were performed under isofluorane anesthesia , and every effort was made to minimize suffering . C3H , C57BL/6 and B6 . 129P2-IL-10tm1Cgn/J ( B6 IL10−/− ) mice were obtained from Jackson Laboratories . B6 miR-146a−/− KO mice on a pure C57BL/6 background were generated as described [28] . Mice were infected with 2×104 B . burgdorferi strain N40 ( provided by S . Barthold , University of California , Davis , CA ) by intradermal injection into the skin of the back . Infection was confirmed in mice sacrificed before 14 d of infection by culturing bladder tissue in BSK II media containing 6% rabbit serum ( Sigma-Aldrich ) , phosphamycin and rifampicin . ELISA quantification of B . burgdorferi-specific IgM and IgG concentrations was used to confirm infection in mice sacrificed at and after 14 d of infection as described [17] . Ankle measurements were obtained using a metric caliper . Rear ankle joints were prepared for assessment of histopathology by removal of the skin and fixation of tissue in 10% neutral buffered formalin . Decalcified joints were embedded in paraffin , sectioned at 3 µm , and stained with H&E . Each slide was scored from 0 to 5 for various aspects of disease , including polymorphonuclear leukocyte ( PMN ) and mononuclear cell ( lymphocytes , monocytes , macrophages ) infiltration into inflammatory processes , tendon sheath thickening ( hypertrophy and hyperplasia of surface cells and/or underlying dense sheets of cells resembling immature fibroblasts , synoviocytes , and/or granulation tissue ) , reactive/reparative responses ( periosteal hyperplasia and new bone formation and remodeling ) , and overall lesion ( composite score based on all lesions observed in 6–8 sections per joint ) , with 5 representing the most severe lesion , and 0 representing no lesion . Ankle measurements and arthritic lesions were assessed in coded samples . Hearts of B6 , B6 miR-146a−/− and C3H mice were assessed for carditis by histopathologic evaluation at 3 weeks post-infection . Hearts were fixed in 10% neutral buffered formalin , embedded in paraffin and sectioned at 3 µm , and stained with H&E . Lesion scoring was performed in a blinded fashion based on a composite of 11 sections per sample , with a score of 5 representing the maximum lesion and 0 representing no lesion . Microarray analysis was performed with the assistance of the University of Utah Microarray and Bioinformatics core facilities . Whole joint RNA was purified from mouse joints ( 3–4 mice per sample group ) using miRNeasy kit ( Qiagen ) . RNA quality was determined using a Bioanalyzer 2100 and RNA 6000 Nano Chip ( Agilent Technologies ) . Agilent Mouse miRNA microarray v2 ( 8×15k ) was hybridized with Cyanine-3 labeled miRNA ( using 100 ng total RNA ) using the Agilent one-color GE hybridization and wash kit . Slides were scanned in a G2505C Microarray Scanner at 2 um resolution ( Agilent Technologies ) . TIF files generated from the scanned microarray image were analyzed in the Agilent Feature Extraction Software ( v . 10 . 5 ) , which was used to calculate feature intensities , background measurements and statistical analyses . Data sets for each biological sample were then filtered and log ( 2 ) transformed using an in-house java script , and were uploaded into Geospiza GeneSifter Analysis Edition ( Perkin Elmer ) . Pair-wise analysis between groups was performed using a quality cutoff for both groups of 1 , normalizing to median values , with a cutoff value of 2-fold change compared to uninfected controls . DNA was prepared from ear tissues frozen at the time of sacrifice . Tissue was incubated in 50 mM NaOH for 1 hour at 93°C and neutralized with 1M Tris ( pH 8 ) . Quantification of B . burgdorferi recA normalized to the mouse nidogen was performed using a Roche LC-480 using previously published primers [18] . Single-cell suspensions were prepared as previously described [43] . Skin was removed from rear ankle joints and digested for 1 h at 37°C in RPMI 1640 containing 0 . 2 mg/ml purified enzyme blend for tissue dissociation ( Roche ) and 100 µg/ml DNase I ( Sigma-Aldrich ) , following partial removal of tissue from bone using 20-gauge syringe needles . Single cell suspension was filtered through a 100 µm cell strainer and red blood cells were lysed using ammonium-chloride-potassium ( ACK ) lysing buffer . For all experiments examining expression in heart and joint tissue , RNA was purified from the heart or tibiotarsal joints with the skin removed . Tissue was immediately immersed in RNA stabilization solution ( Qiagen ) and stored at −80°C . Total RNA was recovered from homogenized tissue using the miRNeasy kit ( Qiagen ) . For FACS-sorted cell populations , sorted cells were collected directly in flow tubes containing 0 . 5 ml RNA stabilization solution ( Qiagen ) and RNA was recovered using the miRNeasy kit ( Qiagen ) . RNA from BMDMs was recovered using guanidium thiocyanate-phenol-chloroform extraction reagent ( Invitrogen ) . RNA recovered from tissue and cells was reverse transcribed , and transcripts were quantified using a Roche LC-480 according to our previously described protocols [47] . For mature miRNA quantification , cDNA was synthesized using the mercury Locked Nucleic Acid Universal RT microRNA PCR , Polyadenylation and cDNA synthesis kit ( Exiqon ) , and miR-146a , 5S rRNA Locked Nucleic Acid primer sets were used ( Exiqon ) to quantify miRNA using a Roche LC-480 . Other primer sequences used in this study were as follows: Itgam ( CD11b ) FWD ( 5′-CCTTCATCAACACAACCAGAGTGG-3′ ) REV ( 5′- CGAGGTGCTCCTAAAACCAAGC-3′ ) , Irak1 FWD ( 5′-TGTGCCGCTTCTACAAAGTG-3′ ) REV ( 5′-TGTGAACGAGGTCAGCTACG-3′ ) , Traf6 FWD ( 5′-AAGCCTGCATCATCAAATCC-3′ ) REV ( 5′-CTGGCACTTCTGGAAAGGAC-3′ ) . Primer sequences for B . burgdorferi 16S rRNA , β-actin , Il1β , Stat1 , Tnfa , Oasl2 [47] , Vα14 , F4/80 [48] Il10 , Ifng , Cxcl10 , Il6 [44] Cxcl1 , Cxcl2 , Pecam1 ( CD29 ) , and Ptprc ( CD45 ) [43] can be found in indicated citations . All flow cytometry data were analyzed using FlowJo ( v . 5 ) software . Sorting experiments were performed using a BD FACSAria II . All other FACS data were collected on a BD LSRII flow cytometer . 7-aminoactinomycin D ( eBioscience ) or DAPI ( Invitrogen ) was used in all experiments , and dead cells and cell doublets were excluded from analyses . All Abs used for flow cytometry were purchased from either BioLegend or eBioscience . Unconjugated Fc blocking Ab ( clone 93; BioLegend ) was included in all Ab-labeling experiments . Position of gates for sorting and analysis was based on analysis of appropriate isotype controls . Fluorochrome-conjugated Abs and isotype controls used in this study were as follows: APC/Cy7-conjugated anti-CD11b ( M1/70 ) and anti-CD45 . 2 ( 104 ) ; FITC-conjugated anti-CD8a ( 53-6 . 7 ) , anti-CD11b ( M1/70 ) and anti-Gr-1 ( RB6-8C5 ) ; PerCP/Cy5 . 5-conjugated anti-Ly6C ( HK1 . 4 ) , anti-CD4 ( RM4-4 ) and anti-CD31 ( 390 ) ; PE-conjugated anti-F4/80 ( BM8 ) anti-LAMP-1 ( 1D4B ) and anti-NK1 . 1 ( PK136 ) ; PE/Cy-7–conjugated anti-CD4 ( GK1 . 5 ) and anti-TCR β ( H57-597 ) ; APC-conjugated anti-CD206 ( MMR ) and anti-F4/80 ( BM8 ) ; and Brilliant Violet 605-conjugated anti-B220 ( RA3-6B2 ) . Confirmation of cell sorting efficiency was performed using qRT-PCR of surface markers used . Bone marrow-derived macrophages ( BMDMs ) were isolated from the femurs and tibias of mice , as previously described [98] . Macrophage cultures were plated in 12-well plate at a density of 6×105/ml in media containing the serum replacement Nutridoma ( Roche ) and stimulated with live B . burgdorferi cN40 ( 7 . 5×106/ml ) . Priming of macrophages was performed by pre-incubating cells with 1 ng/ml mouse recombinant IL-10 for 4 hours prior to addition of B . burgdorferi . After 24 hours , cell supernatants were collected and analyzed by enzyme-linked immunosorbent assay ( ELISA ) . For expression analysis , RNA was collected from cells at 6 hours and 24 hours post-stimulation , and mRNA quantification was performed by qRT-PCR using methods described above . Blood from mice was obtained by submandibular puncture at the time of euthanasia . Blood was allowed to clot , centrifuged , and serum was collected and stored at −20°C prior to analysis . Cell supernatant was used immediately or stored at −20°C prior to analysis . Cytokine concentration in serum samples and cell supernatant was detected by sandwich ELISA using capture and biotinylated antibodies against mouse IL-1β ( clones B122 and Poly5158 , Biolegend ) , IL-6 ( clones MP5-20F3 and MP5-32C11 , BD Biosciences ) , IL-10 ( clones JESS-2A5 and SXC-1 , BD Biosciences ) IL-12 ( clones C15 . 6 and C17 . 8 , BD Biosciences ) . IFNγ ( clones R46A2 and XMG1 . 2 , BD Biosciences ) , TNFα ( clones G281-2626 and MP6-XT3 , BD Biosciences ) and CXCL1 ( clone 48415 and Cat BAF453 , R&D Systems ) . Cells were washed and lysed at 4°C . with NP-40 lysis buffer ( 0 . 5% NP-40 ) for 1 hour followed by boiling for 5 minutes in SDS sample buffer . Protein concentration was measured using a BCA protein assay ( Thermo Scientific ) . Proteins were separated by polyacrylamide gel electrophoresis ( PAGE ) and transferred overnight at 4°C . onto an Immobilon-P membrane ( Millipore ) . Membrane was blocked with 5% milk in TBST and stained with the following antibodies: rabbit anti-TRAF6 ( clone H-274 , Santa Cruz ) , rabbit anti-IRAK1 ( clone D51G7 , Cell Signaling ) rabbit anti-STAT1 ( Cell Signaling #H9172S ) and rabbit anti-β-actin ( clone 13E5 , Cell Signaling ) as a loading control . Horseradish peroxidase-conjugated goat anti-rabbit IgG ( BioRad ) was used as a secondary antibody prior to incubation with enhanced chemoluminescent substrate ( Thermo Scientific ) . Membrane was exposed to autoradiography film ( GeneMate ) and developed using a medical film processor ( SRX-701 , Konica Minolta ) . Mice received an intraperitoneal ( IP ) injection of 3 ml of 3% thioglycollate 4 days prior to harvesting of peritoneal macrophages . Macrophages were removed from sacrificed mice by IP injection of 5 ml ice-cold PBS . Red blood cells were lysed using ACK lysis buffer . 5×105 cells were allowed to adhere to a 12-well plate in RPMI+10% FBS for 4 hours , after which cells were washed and unadhered cells removed . 5×106 B . burgdorferi strain N40 constitutively expressing GFP under the flaB promoter [99] ( a gift from Dr . Jay Carroll ) were added to cells in RPMI . B ( 75% RPMI+10% FBS+25% BSKII ) , as described [100] , plates were centrifuged at 500 g for 5 minutes and incubated for 1 or 2 hours . After which cells were washed gently 3× in warm PBS and gently removed from the plate using a cell scraper . Cells were washed 2× with ice-cold PBS and supernatant discarded following centrifugation . Washed cells were then resuspended in flow buffer and analyzed by flow cytometry using a BD LSRII flow cytometer . As a negative control , untreated cells and cells incubated with unlabeled B . burgdorferi N40 were used . Peritoneal macrophages were harvested as described above and allowed to adhere to the surface of etched microscope cover slides for 4 hours . Cells were incubated with GFP-B . burgdorferi for 1 hour at a 100∶1 MOI as described above , followed by 4× washes with warm PBS , fixed in 4% paraformaldehyde , and incubated with antibody blocking solution ( 3% BSA 0 . 05% milk 0 . 2% Tween20 in PBS ) for 1 hour at RT . Cells were then stained with PE-conjugated LAMP-1 , APC-conjugated F4/80 and DAPI for 1 hour in antibody solution ( 1% BSA 0 . 02% Tween20 in PBS ) , washed and mounted onto a glass slide using fluorescent mounting reagent ( Calbiochem EMD Millipore ) . Confocal imaging was performed on a FV1000 inverted confocal microscope ( Olympus ) using FV10-ASW software ( Olympus ) . Images were taken using a 60× oil lens with a 1024×1024 2× zoom , and captured at a plane dissecting the middle of cell nuclei . All imaging was performed at the University of Utah Cell Imaging Core Facility , with the assistance of Dr . Christopher Rodesh . Microarray data statistical analysis was performed using the Agilent Feature Extraction Software ( v . 10 . 5 ) and Geospiza GeneSifter Analysis Edition ( Perkin Elmer ) , as described . Raw and adjusted p values were derived by Welch's t test with Benjamini and Hochberg correction , using a raw p value cutoff of p<0 . 05 signifying statistical significance . All other graphical data represent the mean ± SEM . Statistical analysis was performed using Prism 5 . 0c software . Multiple-sample data sets were analyzed by one-way ANOVA with Dunnet's or Tukey's post hoc test for pair-wise comparisons , as appropriate and indicated in figure legends . Two-sample data sets were analyzed by Student t test . Categorical data for histopathology was assessed by Mann-Whitney U test . Statistical significance indicated in figure legends . | Lyme Disease is caused by infection with the bacteria Borrelia burgdorferi , is transmitted through infected deer ticks ( Ixodes scapularis ) , and often leads to arthritis that can persist , even after antibiotic treatment . Here , we have identified a microRNA that is critical in modulating Lyme arthritis , but not carditis . This microRNA , miR-146a , is a negative regulator of NF-κB signaling , known to be important in host defense against pathogens , and long suspected to play a role in Lyme arthritis development . Mice lacking miR-146a develop more severe arthritis and show signs of hyperactive NF-κB activation during the persistent phase of infection . Heart manifestations of disease were not altered . Furthermore , this severe arthritis is independent of host defense , since these mice are better able to clear invading bacteria in joints , and bacterial numbers are similar in heart and ear tissue . We identified TRAF6 as an important target of miR-146a-mediated NF-κB regulation of pro-inflammatory cytokines IL-6 and IL-1β , as well as chemokines CXCL1 and CXCL2 . Our data demonstrate the importance of maintaining appropriate regulation of amplitude and resolution of NF-κB activation during Borrelia burgdorferi infection , and provide a novel model for elucidating the role of NF-κB in Lyme arthritis development , independent of effect on host defense . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"blood",
"cells",
"borrelia",
"infection",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"gene",
"regulation",
"immune",
"activation",
"immunology",
"microbiology",
"animal",
"models",
"bacterial",
"diseases",
"model",
"organisms",
"emerging",
"infectious",
"diseases",
"molecular",
"genetics",
"immunoregulation",
"immunomodulation",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"white",
"blood",
"cells",
"inflammation",
"animal",
"cells",
"medical",
"microbiology",
"gene",
"expression",
"microbial",
"pathogens",
"mouse",
"models",
"pathogenesis",
"immune",
"response",
"immune",
"system",
"cell",
"biology",
"monocytes",
"clinical",
"immunology",
"immunity",
"host-pathogen",
"interactions",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types"
]
| 2014 | MicroRNA-146a Provides Feedback Regulation of Lyme Arthritis but Not Carditis during Infection with Borrelia burgdorferi |
Autophagy is a conserved degradative pathway used as a host defense mechanism against intracellular pathogens . However , several viruses can evade or subvert autophagy to insure their own replication . Nevertheless , the molecular details of viral interaction with autophagy remain largely unknown . We have determined the ability of 83 proteins of several families of RNA viruses ( Paramyxoviridae , Flaviviridae , Orthomyxoviridae , Retroviridae and Togaviridae ) , to interact with 44 human autophagy-associated proteins using yeast two-hybrid and bioinformatic analysis . We found that the autophagy network is highly targeted by RNA viruses . Although central to autophagy , targeted proteins have also a high number of connections with proteins of other cellular functions . Interestingly , immunity-associated GTPase family M ( IRGM ) , the most targeted protein , was found to interact with the autophagy-associated proteins ATG5 , ATG10 , MAP1CL3C and SH3GLB1 . Strikingly , reduction of IRGM expression using small interfering RNA impairs both Measles virus ( MeV ) , Hepatitis C virus ( HCV ) and human immunodeficiency virus-1 ( HIV-1 ) -induced autophagy and viral particle production . Moreover we found that the expression of IRGM-interacting MeV-C , HCV-NS3 or HIV-NEF proteins per se is sufficient to induce autophagy , through an IRGM dependent pathway . Our work reveals an unexpected role of IRGM in virus-induced autophagy and suggests that several different families of RNA viruses may use common strategies to manipulate autophagy to improve viral infectivity .
Macroautophagy ( thereafter referred to as autophagy ) is a highly regulated self-degradative mechanism for intracellular clearance and recycling of cytoplasmic contents [1] . During this process large portions of the cytoplasm are engulfed into autophagosomes that subsequently fuse with lysosomes to form acidic autolysosomes where degradation occurs . The autophagy process results from a cascade of reactions orchestrated by autophagy-related genes ( atg ) encoding ATG proteins mostly defined in yeast and for which numerous mammalian orthologs have been identified [2] . However , the function of most of these atg remains poorly characterized and several non atg mammalian genes were also described to regulate autophagy . During autophagy , the formation of an isolation membrane is initiated by class III phophatidylinositol 3-kinase ( PIK3C3 ) /Beclin1 containing complexes [3]–[5] . The elongation of the isolation membrane involves two ubiquitin-like conjugation systems [6] , [7] . In one of them , ATG12 associates with ATG5 for the formation of ATG12-ATG5-ATG16L1 molecular complexes that bind the outer membrane of the isolation membrane . In the second , LC3 is coupled with phosphatidylethanolamine to generate a lipidated LC3-II form that is integrated in both the outer and inner membranes of the autophagosome . Whereas required at a basal level for cellular homeostasis maintenance , autophagy is used as a universal innate cell defense mechanism to fight intracellular pathogens allowing their delivery to degradative lysosomes [8] , [9] . Studies involving overexpression or knock-down of atg have demonstrated an important role for autophagy in both innate antibacterial [10]–[12] and antiviral defense [13] , [14] . Autophagy contributes to immune surveillance via cytoplasmic sampling and delivery of intracellular pathogens or components of these pathogens to endosomes and major histocompatibility complex ( MHC ) -II molecules rich compartments , thus promoting innate recognition by endosomal Toll-like receptors ( TLR ) [15] and pathogen-adaptive immune response [16]–[18] , respectively . However , since autophagy is a conserved pathway , intracellular pathogens were submitted to an evolutionary pressure that led to the selection of pathogens with different molecular strategies to avoid or subvert this process to their own benefit [8] . RNA viruses include several viral species that are of major concerns in public health such as Hepatitis C virus ( HCV ) , human immunodeficiency virus 1 ( HIV-1 ) , influenza A , Measles virus ( MeV ) or Dengue virus . These viruses dispose of a limited number of viral proteins to control major cellular pathways such as protein production or degradation , cell survival and evasion from host cell defense . Several RNA viruses have been shown to subvert autophagy , nevertheless few viral molecular adaptations to host autophagy have been identified [19]–[25] . HIV-1 and influenza A are two viruses that block autophagosome maturation . It has been shown that both HIV-1-NEF and influenza A-M2 proteins target Beclin1 to prevent autolysosome formation [21] , [24] . The identification of new viral factors able to physically interact with autophagy-associated proteins and the characterization of their functional consequences on autophagy might provide insights on the strategies used by RNA viruses to manipulate and/or subvert this pathway . The growing knowledge of the molecular partners underlying the execution and the regulation of the autophagy process prompted us to analyze whether this machinery is particularly targeted by RNA viruses . Using a yeast two-hybrid approach and bioinformatics analysis , we have determined how proteins from 5 different RNA virus families ( Paramyxoviridae , Flaviviridae , Orthomyxoviridae , Togaviridae and Retroviridae ) physically interact with host autophagy-associated proteins . We have found that autophagy is a functional network highly targeted by RNA virus proteins . In particular , we observed that IRGM is able to interact with proteins from 5 different RNA virus families . Although IRGM was previously reported to play an autophagy-dependent anti-bacterial function [26] , [27] , the mechanisms underlying the role of IRGM in autophagy remain poorly understood [27]–[29] . We found that this protein interacts with several key proteins of the autophagy process . Furthermore , we describe a role of IRGM in both virus-dependent autophagy induction and virus production . Our results suggest that different RNA virus families have a conserved way to overcome the host autophagy pathway in order to lead to a successful infection .
To gain an insight on the molecular mechanisms by which RNA viruses modulate the autophagy process , we tested whether viral proteins belonging to various strains of 5 different viral families ( Paramyxoviridae , MeV , Mumps virus , Nipah virus; Flaviviridae , HCV , Dengue virus , West Nile virus , Tick borne encephalitis virus and Kyasanur forest disease virus; Orthomyxoviridae , Influenza A virus; Retroviridae , HIV-1; Togaviridae , Chikungunya virus ) were able to physically interact with human autophagy proteins . We established a list of 44 autophagy-associated proteins closely involved in human autophagy , based on data available from the literature showing the functional involvement of the proteins in autophagy using either short interfering ( si ) -RNA shutting down the expression of the considered protein or protein overexpression ( Figure 1A and Table S1 ) . Proteins that despite their crucial involvement in autophagy are broad signalling regulators of different cellular pathways were not included in our study . From this list 35 autophagy-associated proteins ( Figure 1A in blue and Table S2 ) were available to be tested pairwise in a yeast two-hybrid array against 80 different viral proteins ( Table S3 ) . We found 42 new protein-protein interactions ( ppi ) between viral proteins and autophagy-associated proteins to which were added the 10 ppi between RNA virus proteins and autophagy-associated proteins described in the literature ( Figure 1B and Table S4 ) . We further tested the ability of IRGM , GOPC and SQSTM1 , the most common RNA virus targets , to interact with three different HIV-1 proteins ( Table S3/S4 ) . Overall , among the 17 different autophagy-associated proteins we found to be targeted by RNA virus proteins , 9 proteins interact with one RNA virus family , 5 interact with 2 different families ( BECN1 , BCL2 , UVRAG , ATG5 , BNIP3 ) , 2 interact with 3 different RNA virus families ( SQSTM1 and GOPC ) and one autophagy-associated protein , IRGM , is a common target of 5 different families of RNA viruses ( Paramyxoviridae , Flaviviridae , Orthomyxoviridae , Togaviridae , Retroviridae ) ( Figure 1B , Figure S3 and Table S4 ) . Here , our results show that RNA virus proteins interact with more than 35% of the autophagy-associated proteins suggesting that autophagy is a widely targeted functional network . This targeting is highly significant as compared to the targeted human proteome counterpart ( exact Fisher test , p value <2 , 2×10−16 ) . Nevertheless , a defined protein may play important roles in several cellular functions; therefore a viral interaction with a particular autophagy protein does not preclude an effect in a precise pathway . To understand whether the proteins targeted by RNA viruses are particularly dedicated to the autophagy process or may be important to connect this conserved cellular function to other cellular processes , we have established a comprehensive map of ppi between the 44 autophagy-associated proteins ( Figure 1C , each node represents one autophagy-associated protein ) . We first determined this network by systematically testing pairwise 35 autophagy-associated proteins ( proteins in blue in Figure 1A ) in a yeast two-hybrid array and incremented our own set of data with those from the literature to build the autophagy network [30] ( Figure 1C , Table S5 ) . Overall , the human autophagy network is composed of a total of 150 ppi , among them we identified 23 novel intra-autophagy network interactions ( Figure 1C , blue edges ) . Interestingly , the human autophagy network appears as a highly interconnected cellular network , with a single connected component of 40 proteins , and only four isolated proteins ( ATG4D , ATG9A , WIPI1 and HDAC6 ) for which to date no protein interaction was identified within the network . This interconnectivity is significantly higher than the theoretical interconnectivity computed from resampled subnetworks ( resampling test , n = 10000 , p value <10−4 , Figure S1 ) . This high significance supports the functional consistency of the initially chosen group of 44 autophagy-associated proteins . The 44 proteins of the autophagy network do not function in isolation but interact with roughly at least 450 other cellular proteins within the whole human protein interaction network [30] . To analyze the relative functional contribution of each of the 44 autophagy-associated proteins to the autophagic process without a priori , we determined their respective connectivity and centrality within the autophagy network and within the whole human protein interaction network ( Figure 2A and Table S6 ) [31] . The connectivity or degree of a protein is the number of direct interacting partners of this protein We have defined the autophagy context-dependent connectivity as the ratio of the protein connectivity within the autophagy network over those within the whole human protein interaction network [31] ( Figure 2A and Figure S2A ) . This highlights 10 different proteins that have more than two thirds of their interactions inside the autophagy network , suggesting that they might be particularly dedicated to the autophagy network ( Figure 2A , x axis>0 . 66 ) . Examples include several different ATGs ( ATG3 , ATG4A/B , ATG7 ) but also BNIP3 and IRGM . Interestingly , half of the autophagy-associated proteins have less than one third of their interactions within the autophagy network i . e . are essentially connected outside this network ( Figure 2A , x axis<0 . 33 ) . Examples include BCL2 , ATG5 , ATG12 or BECN1 , indicating that an important fraction of autophagy-associated proteins might be involved in the crosstalk between autophagy and other cellular pathways . Although the number of connections of a protein is relevant , the flux of information passing through this protein , illustrated by its centrality or betweenness , is another critical parameter that determines the influence of a protein in a network . We have defined the autophagy context dependent-centrality as the ratio of the protein betweenness within the autophagy network over those within the whole human protein interaction network [31] ( Figure 2A and Figure S2B ) . Interestingly , most of the 44 autophagy-associated proteins appear as essential components of the autophagy system ( Figure 2A , y axis>1 ) . Among them were found most of the ATGs , ULK1 , BECN1 , AMBRA1 , PIK3C3 and PIK3R4 , BNIP3 and IRGM that exhibit increased betweenness values in the autophagy network compared to the whole human interactome . Our results show that RNA viruses mainly target proteins that although central to this functional network connect autophagy to other cellular functions ( Figure 2A , y axis>1 , targeted proteins represented in red ) . Indeed , with the exception of three proteins ( IRGM , BNIP3 and TMEM74 ) , all RNA virus-targeted autophagy-associated proteins have more than 60% of their interactions that take place out of the autophagy network ( Figure 2A , x axis<0 . 4 targeted proteins represented in red ) . Strikingly , IRGM is noteworthy by being commonly targeted by 5 different families of RNA viruses ( Figure 1B and Figure S4 ) and by making all its cellular protein interactions within the autophagy network ( Figure 2 ) . These results prompted us to investigate whether IRGM and the autophagy interactors we identified by yeast two-hybrid ( Figure 2B ) could associate in mammalian cells . In co-transfected human HeLa cells , we first found that IRGM co-localizes with ATG10 , ATG5 , SH3GLB1 and MAP1LC3C ( Figure 2C ) . Furthermore , physical interactions between IRGM and each of these autophagy-associated proteins were confirmed by GST-co-affinity experiments in human cells ( Figure 2D ) . We found that endogenous IRGM colocalizes with endogenous SH3GLB1 and ATG5 ( Figure 2E ) , and that a small fraction of endogenous ATG5 interacts with overexpressed IRGM but not with SQSTM1 ( Figure 2F and Figure S5 ) . Together , our results show that IRGM can interact with several autophagy-associated proteins . IRGM being both particularly targeted by RNA viruses and interacting with several autophagy-associated proteins , prompted us to test whether this protein was involved in virus-induced autophagy . We have found that IRGM interacts with MeV ( Paramyxoviridae family ) , HCV ( Flaviviridae family ) , influenza A ( Orthomyxoviridae family ) and HIV-1 ( Retroviridae family ) proteins ( Figure 1B , Figure S4 and Table S4 ) , four viruses described to induce autophagosome accumulation upon infection [19] , [21] , [24] , [32] . To determine whether IRGM is involved in autophagosome formation observed upon MeV , HCV and influenza A infections we have abrogated IRGM expression using specific si-RNA ( Figure S6A–G ) , prior to infection in GFP-LC3-HeLa cells , GFP-LC3-Huh 7 . 5 or GFP-LC3-A549 cells , respectively . IRGM mRNA ( Figure S6F ) and endogenous protein ( Figure S6C–E , G ) was detected in cell lines used in our studies . The level of expression of the endogenous protein is specifically decreased in cells treated with si-IRGM ( Figure S6D/E ) , as previously observed [28] . Autophagosomes were monitored by tracking the formation of GFP-LC3-labelled structures representing LC3-II-containing autophagosomes . The reduced expression of ATG5 induced by si-RNA was used as a control for autophagy extinction ( Figure S7 ) . First , we found that the reduced expression of IRGM did not significantly modulate ongoing autophagy in each tested cell line ( Figure 3A/B , D/E , G/H ) . Second , as previously reported , we found upon MeV ( Figure 3A/B ) , HCV ( Figure 3D/E ) and influenza A ( Figure 3G/H ) infections an increased number of autophagosomes . Interestingly , the inhibition of expression of either IRGM or ATG5 prevented the increase of autophagosomes induced by MeV and HCV ( Figure 3A/B/D/E ) . However , contrarily to the reduced expression of ATG5 , the inhibition of expression of IRGM did not prevent significantly autophagosome accumulation upon influenza A infection ( Figure 3G/H ) . We further tested the requirement of IRGM on MeV , HCV , influenza A and HIV-1-dependent autophagy modulation by monitoring the conversion of LC3-I ( cytosolic form ) to LC3–II ( autophagosome-bound lipidated form ) by western blot in HeLa , Huh 7 . 5 , A549 cells and human monocyte-derived macrophages ( MDM ) , respectively . Without infection , we found that si-IRGM-treated cells do not display modulation of the total amount of LC3-II when compared to si-control treated cells ( Figure 3C/F/I/J ) . We next found that MeV infection did not lead to an increased amount of LC3-II in si-control treated cells but instead a decrease , when compared with uninfected si-control-treated cells . This results might be the consequence of the increase of the autophagy flux induced upon MeV infection leading to the formation of productive autolysosomes , as we already reported [32] . Nevertheless , diminished IRGM expression reduced MeV-induced LC3-II amount , similarly to si-ATG5 treatment ( Figure 3C ) . HCV , influenza A and HIV-1 were all reported to inhibit autophagosome maturation [21] , [24] , [33] . We found that infection with these viruses lead to an increased amount of LC3-II , when compared to uninfected cells . Moreover , the diminished expression of IRGM reduced HCV and HIV-1-increased LC3-II amount , similarly to si-ATG5 treatment ( Figure 3F/J ) , whereas it has no effect on influenza A infection ( Figure 3I ) . Finally , we determined autophagy modulation during MeV , HCV , influenza A and HIV-1 infections in the presence of bafilomycin A1 ( BAF ) , which inhibits acidification of the autolysosomes ( Figure S8 ) . We found that upon BAF treatment the total number of GFP-LC3 dots in MeV-infected cells was further increased when compared with untreated MeV-infected cells , and no further increased was observed upon HCV or influenza A infections ( Figure S8A–F ) . Upon HIV-1 infection a slight increase of the total amount of LC3-II was detected in BAF-treated cells when compared with uninfected cells ( Figure S8G ) . Interestingly , the reduced expression of IRGM prevented the accumulation of autophagosomes upon MeV , HCV or HIV-1 infections , but not upon influenza A virus infection in BAF-treated cells ( Figure S8 ) . Thus , altogether these results indicate that IRGM is involved in MeV , HCV and HIV-1-mediated autophagosome accumulation . We next evaluated whether IRGM could modulate MeV , HCV , influenza A and HIV-1 infectious particle formation . To this end , we have impaired IRGM expression by si-RNA , infected HeLa cells , Huh 7 . 5 cells , A549 cells or human MDM with the appropriate viruses and evaluated its effect on viral particle formation . Looking at MeV infectivity , we first found that shutting down the expression of the autophagy essential gene ATG5 impaired the production of infectious particles by more than 65% , indicating that MeV hijacks autophagy to its own benefit ( Figure 4A ) . Importantly , MeV infectious particle production was equally compromised in absence of IRGM expression ( Figure 4A ) . Similar results were obtained concerning HCV infectivity for which the involvement of autophagy in the formation of viral particles was previously reported [20] . Interestingly we found that the absence of IRGM compromised the production of infectious HCV particles by more than 70% , similarly to the absence of ATG5 ( Figure 4B ) . In contrast , the reduced expression of IRGM does not impair influenza A particle production ( Figure 4C ) . Down regulation of ATG5 has also no influence on the viral production , as previously reported [21] . Finally , it was recently shown that autophagy modulates HIV production in human MDM [24] , [34] . We found here that absence of IRGM compromised the production of HIV-1 particles by more than 30% , similarly to the reduced expression of ATG5 ( Figure 4D ) . Overall our results show that IRGM is involved in the production of viral particles of at least three different RNA viruses , MeV , HCV and HIV-1 . To get a deeper insight on the molecular mechanisms underlining IRGM's role in both virus induced autophagy and viral particle formation we have first analysed whether IRGM and its putative viral MeV , HCV or HIV interactors could associate in human cells . In co-transfected human HeLa cells , we found that IRGM co-localizes with MeV-C , HCV-NS3 and HIV-NEF , ( Figure 5A ) . Furthermore , we confirmed the physical interaction between IRGM and MeV-C , HCV-NS3 and HIV-NEF using GST-co-affinity experiments ( Figure 5B ) . Thus IRGM interacts with proteins from several different RNA virus families . We then determined whether the viral proteins able to interact with IRGM , could modulate autophagosome formation . To this end , MeV-C , HCV-NS3 or HIV-NEF were expressed in GFP-LC3-HeLa cells and autophagy analyzed by tracking GFP+ autophagosomes . We found that each of these viral proteins induced a significant increase of the number of autophagosomes compared to the overexpression of GST , used as a control ( Figure 6A/B ) . Importantly , impairment of IRGM expression leads to a decrease of the number of autophagosomes observed upon overexpression of each of the three viral proteins , as the reduced expression of ATG5 ( Figure 6C/D/E/F ) . Thus , our results show that the MeV-C , HCV-NS3 and HIV-NEF proteins promote autophagosome accumulation via a molecular process involving IRGM .
We found that the autophagy network is highly and significantly targeted by RNA viruses suggesting that this cellular process plays an important role in virus biology . Our data suggest that RNA viruses target proteins that although central to autophagy also present a high number of physical connections with proteins involved in other cellular functions . A recent analysis of the global organization of the autophagy network revealed that this conserved cellular function is connected to other cellular functions such as proteolysis , signal transduction , phosphorylation and vesicle transport [35] . Thus , molecular bridges between autophagy and other cellular functions might be preferentially targeted by RNA viruses . The ability to manipulate multifunctional proteins might empower RNA viruses with the ability to fine-tune different complementary cellular functions which are necessary for successful virus infection and replication . A major challenge remains to determine how these viral/cellular protein interactions translate into functional changes imposed by RNA viruses on autophagy and/or other connected cellular functions . Interaction between viral proteins and autophagy-associated proteins can be explained either by a viral adaptation to this cellular function or alternatively autophagy-associated proteins might be devoted to detection/binding of pathogen proteins to promote anti-viral function . For instance , SQSTM1 ( also called p62 ) brings cargos of ubiquitinated proteins to autophagy degradation [36] . Recently , SQSTM1 was shown to protect mice against Sindbis virus infection by promoting autophagy-dependent but ubiquitination-independent capsid protein degradation [13] . We found here that SQSTM1 has a potential to bind proteins of at least 3 different families of human-infecting RNA viruses . Whether these interactions act in the host cellular defense or are viral subversive pathways has to be tested in specific virus/host cell contexts . GOPC is another highly targeted autophagy-associated protein . However , GOPC appears to be poorly dedicated to autophagy suggesting that its targeting might essentially affect non-autophagy cellular processes . Only two highly autophagy-dedicated proteins were found to interact with RNA virus proteins , BNIP3 and IRGM . Another targeted autophagy-associated protein , TMEM74 is exclusively connected to the autophagy network , since we found it interacts with BNIP3 . BNIP3 and TMEM74 are involved in stress-induced autophagy regulation [37] , [38] . Interestingly we found that BNIP3 is able to interact with a large number of proteins within the autophagy network acting at different steps of the autophagic process . This suggests that under specific conditions this protein might allow the coordination and the concerted action of different sub-networks necessary for autophagy progression/inhibition . Whether BNIP3 and TMEM74 are effective during viral infection is unknown , although they might both ( co ) -regulate cellular responses to viral infection through autophagy induction . We focused our functional work on IRGM since our data support that this protein might be an important regulator of autophagy . We would thus expect that the interaction of a viral protein with IRGM might trigger a functional and specific effect on this cellular function . Many immunity associated GTPases ( IRG ) exist in mammals and these proteins play an important role in defense against intracellular pathogens [39] . Only two IRG exist in humans , IRGC and IRGM [40] . IRGM is a genetic risk factor in Crohn's disease and tuberculosis [41]–[43] . Human IRGM is constitutively expressed , contrarily to its mouse homolog Irgm1 , and it was shown to regulate both IFN-γ- and rapamycin-induced autophagy in human macrophages [27] . Nevertheless the molecular mechanism underlying its function in autophagy remains poorly understood . A recent study reported a role of IRGM in mitochondrial fission important for autophagic control of intracellular mycobacteria [29] . Furthermore loss of tight regulation of IRGM expression compromises the control of intracellular replication of Crohn's disease-associated adherent invasive Escherichia coli by autophagy [28] . However , prior to our study no molecular protein connection of IRGM with autophagy has been reported . We found that IRGM is able to directly interact with several autophagy proteins , whereas we did not find any other IRGM interacting human proteins through a yeast two-hybrid screen against a normalized human spleen cDNA library ( data not shown ) . IRGM molecular partners in the autophagy network are involved in the initiation/elongation phases ( ATG5 , ATG10 , MAP1CL3C and SH3GLB1 ) suggesting that IRGM would modulate the initial steps of autophagy . As previously described [29] , we found that IRGM is located in the mitochondrial fraction ( Figure S9 ) . This suggests that IRGM interactions with autophagy-associated proteins might take place at the mitochondria , in initial phases of the autophagic process . At least two of the potential IRGM interactors might be partially located at the mitochondria under particular cellular contexts . ATG5 was shown to be able to physical associate with a mitochondria located protein IPS-1/MAVS [44] . Additionally , a fraction of SH3GLB1 is located in the mitochondria and is required for the maintenance of mitochondrial morphology [45] . IRGM is known to play a protective role against bacterial infection favoring IFNγ-mediated autophagy elimination of Mycobacterium bovis in macrophages [27] and anti-bacterial autophagy responses in epithelial cells against Salmonella thyphymurium [26] and adherent-invasive Escherichia coli [28] , [46] . However , we found that in the context of RNA virus infections , IRGM does not contribute to a protective role but instead promotes virus replication . Virus/host co-evolution might have lead to subversion of an initial protective mechanism initiated by IRGM/viral protein interaction . Indeed , autophagy induction involving IRGM is ultimately exploited by MeV , HCV and HIV-1 and favours viral infectious particle production . Thus for viruses able to inhibit autophagy maturation , additional molecular virus/autophagy interaction would be necessary to block this specific step . In line with this hypothesis , HIV was recently shown to inhibit autophagosome maturation in 293T and we also found here that HIV-1 may partially inhibit autophagosome maturation in human MDM ( Figure S8G ) . HIV-1-mediated inhibition of autophagy maturation was described to involve HIV-NEF via its interaction with BECN-1 [24] . HCV infection was also reported to prevent autophagosome maturation at an early time of infection ( [33] and Figure S8C/D ) . Alternatively , viral proteins may specifically target IRGM to promote autophagy . Importantly , IRGM-interacting viral proteins MeV-C , HCV-NS3 and HIV-1-NEF , induce autophagy in an IRGM-dependent pathway . Viral proteins might facilitate IRGM interaction with its autophagy partners , by facilitating its relocalization to or stabilization with autophagy-associated proteins involved in the initiation phases of autophagosome formation . Interestingly , the mouse IRGM homolog , Irgm1 was found to bind specific phosphoinositides , through a carboxy ( C ) -terminal amphipathic helical segment , allowing the recruitment of Irgm1 on nascent and early phagosomes [47] . Although we found that IRGM can interact with influenza A proteins , we found that its absence does not influence influenza A-induced autophagy . Cell infection by influenza A was reported to inhibit autophagy maturation , what we also observed using BAF-treatment ( Figure S8E/F ) [21] . It is possible that interaction might either not be engaged upon influenza A infection or be involved in other not yet identified IRGM-associated cellular process . Autophagy is a process that has the potential to degrade pathogens or pathogen-derived molecules trapped within autophagosomes . Viruses and viral proteins are not an exception . Nevertheless its ability to promote cell survival under stress conditions might be beneficial to virus since a major defense mechanism against viral infection is cell death . Overriding this mechanism can give rise to infected cell survival and further viral spread . Furthermore autophagy might provide membranous surfaces required for viral replication . Therefore the molecular analysis of the interplay between viruses and autophagy as well as of its consequences on viral and cellular biology might be of importance to control viral infection . We highlight here and unrevealed role of IRGM in autophagy subversion by RNA viruses . Our semi-global interactome approach opens many doors for a better understanding of the interplay between autophagy and RNA viruses by suggesting many possible molecular targets of RNA viruses among the autophagy-associated proteins .
The experiments in this article were performed at Biological Safety Level 2 and 3 in accordance with the regulations set forth by the by the national French committee of genetic ( commission de génie génétique ) . Venous blood from anonymous healthy human volunteers was obtained from the blood bank ( Etablissement Français du Sang ) in accordance with its guidelines , published in the French Journal Officiel , with informed written consent from each volunteer . All constructions were performed with a Gateway recombinational cloning system ( Invitrogen ) . Complete cDNA for 35 autophagy-associated proteins were purchased from several providers ( Table S2 ) . Most cDNAs were available in a pDONR vector ( Gateway technology , Invitrogen ) . For ATG4D , ATG9A , BECN1 , IRGM and ULK1 , a PCR product containing attB sites was generated . Primers used in Gateway cloning are available upon request . These attB-PCR products were used in a BP recombination reaction ( Invitrogen ) . All 80 viral ORFs used are available in ViralORFfeome ORFeotheque in a gateway pDONR vector [48] ( Table S3 ) . For the HIV-1 ORFs NEF , VPR , VIF a PCR product containing attB sites was generated . 35 autophagy-associated cDNAs were transferred by in vitro recombination from a pDONR into both pGBKT7 and pACT2 . These constructs were respectively transformed in both bait strain AH109 ( Clontech ) and prey strain Y187 ( Clontech ) . Viral ORFs ( baits ) ( Table S3 ) were transferred by in vitro recombination from a pDONR into the yeast expression vector pGBKT7 and transformed into the yeast bait strain AH109 . Autophagy-associated ORFs ( prey ) were transferred into pACT2 and transformed into the yeast strain Y187 . Yeast cells were mated and subsequently plated on a selective medium lacking histidine to test the interaction-dependent transactivation of the HIS3 reporter gene . All binary interactions between human autophagy proteins were extracted from the VirHostNet knowledge base and manually checked in each original paper . Protein-protein interaction network graphics were performed using the networks GUESS tool ( http://graphexploration . cond . org ) . The degree k of a node v in a graph G is the number of edges that are incident to this node . The betweenness b of a node v in a graph G can be defined by the number of shortest paths going through the node v and is normalized by twice the total number of protein pairs in the graph G ( n* ( n-1 ) ) . The equation used to compute betweenness centrality , b ( v ) , for a node v is:where gij is the number of shortest paths going from node i to j , i and j , V and gij ( v ) the number of shortest paths from i to j that pass through the node v: The overall statistical significance of the observed autophagy-associated proteins interconnectivity ( number of protein-protein interactions ) was assessed by a random resampling testing procedure ( n = 10 , 000 permutations ) . HeLa and GFP-LC3-HeLa cells were maintained in RPMI 1640 supplemented with 0 , 5 mg/mL gentamicin , 2 mM L-glutamine and 10% fetal calf serum . HEK293T , Huh7 . 5 , GFP-LC3-Huh7 . 5 , A549 , GFP-LC3-A549 , Vero cells and MDCK were maintained in DMEM supplemented with 0 , 5 mg/mL gentamicin , 10% fetal calf serum . Additionally Huh7 . 5 and Huh7 . 5-GFP-LC3 were supplemented with 1% of non-essential amino acids . Monocytes were purified from the blood of healthy human donors . Human monocytes were cultured in RPMI 1640 supplemented with 10% fetal calf serum and differentiated into macrophages using 10 ng/mL of rh-M-CSF during 6 days ( Immunotools , Friesoythe , Germany ) . HEK293T cells were transfected using jetPEI ( PolyPlus , Illkirch , France ) according to manufacturer's instructions . HeLa , GFP-LC3-HeLa , Huh 7 . 5 , GFP-LC3-Huh 7 . 5 were transfected using lipofectamine 2000 ( Invitrogen ) . A549 and GFP-LC3-A549 were reverse transfected using lipofectamine 2000 . Anti-Glutathione-S-Transferase ( GST ) peroxidase ( A7340 ) , anti-FLAG M2 peroxidase ( A8592 ) , anti-MAP1LC3B ( L7543 ) , anti-Actin ( A2066 ) , anti-Myc ( C3956 ) , anti-ATG5 ( A0856 ) and anti-eGFP ( G6795 ) were from Sigma ( St Louis , Mo , USA ) . Anti-IRGM ( NT ) antibody PK-AB718-4543 was purchased from PromoKine ( Heidelberg , Germany ) was used for immunofluorescence studies . Anti-IRGM ( ab93901 ) purchased from abcam ( Cambridge , UK ) was used to detect IRGM by western blot . Anti-ATG5 mouse monoclonal antibody clone 177 . 19 from Millipore was used to detect endogenous ATG5 by immunofluorescence . Anti-SH3GLB3 mouse monoclonal purchased from Sigma ( St Louis , Mo , USA ) was use to detect endogenous protein by immunofluorescence . Anti-cytochrome c mouse monoclonal was purchased from BD Biosciences ( Le Pont de Claix , France ) and the anti-GAPDH mouse monoclonal from Santa Cruz Biotechnology ( Santa Cruz , USA ) . Anti-rabbit HRP ( NA 934 ) was from Amersham Biosciences ( Uppsala , Sweden ) . Anti-mouse Alexa Fluor 568 and 488 was purchased from Invitrogen ( Molecular Probes ) . HeLa cells were co-transfected with GFP–IRGM and FLAG tagged ATG10 , ATG5 , SH3GLB1 , MAP1LC3C , MeV-C , HCV-NS3 or HIV-NEF . After 24 h cells were fixed with 2% paraformaldehyde stained using an anti-FLAG antibody followed by secondary antibody conjugated to Alexa Fluor 568 . For endogenous IRGM detection by immunofluorescence , MeV infected HeLa cells were fixed in cold acetone and IRGM was detected using an anti-IRGM polyclonal antibody from PromoKine ( Heidelberg , Germany ) followed by a secondary antibody conjugated to Alexa Fluor 488 . Endogenous ATG5 or SH3GLB1 were detected respectively using an anti-ATG5 mouse monoclonal antibody clone 177 . 19 from Millipore or an anti-SH3GLB3 mouse monoclonal from Sigma ( St Louis , Mo , USA ) . Virus infected GFP-LC3-expressing cells were fixed with 4% paraformaldehyde . Bafilomycin A1 for flux experiments was purchased from Sigma ( St Louis , Mo , USA SIGMA ) . GFP-LC3 HeLa cells and GFP-LC3 A549 were treated for 5 hrs with 100 nM of bafilomycin while GFP-LC3 Huh 7 . 5 cells were treated for 24 hrs . The number of GFP+ vesicles per cell profile was numerated from one single plan section per cell . In each case , number of GFP+ vesicules was numerated from 100 to 200 cells for each experiment . All the cells were analyzed using a Axiovert 100 M microscope ( Zeiss , Göttingen , Germany ) equipped with the LSM 510 system ( Zeiss ) and observed with a magnification of 63× ( oil immersion ) . 1 . 5 µg of each expression vector were transfected in HEK293T cells . Cell lysis was done 48 hrs post-transfection . Glutathione-sepharose 4B beads ( GE healthcare , Saint Cyr au Mont d'Or , France ) were used for the co-AP purification . For the indicated experiments HEK293T cells were transfected with both pCi-neo-3X FLAG and pDESTmyc expression vectors . Protein G sepharose 4B beads coated with 1 µg of anti-Myc antibody were used for a co-AP . 10×106 HeLa cells were transfected or not with pCi-neo-3X FLAG IRGM and infected with MeV ( MOI = 1 ) 24 hrs post-transfection . Cells were lysed 24 hrs post-infection . Protein G sepharose 4B beads coated with 1 µg of anti-FLAG mouse monoclonal antibody were used for a co-AP of the FLAG tagged IRGM . Endogenous ATG5 associated to FLAG-tagged IRGM was detected using anti-ATG5 from Sigma ( A0856 ) . Smartpool si-ATG5 , si-IRGM and control si-RNA were from Dharmacon ( Perbio , Brebières , France ) . 1 . 105 HeLa , HeLa-GFP-LC3 , Huh 7 . 5 , Huh 7 . 5-GFP LC3 cells were plated in 6-well plates 24 hrs prior to transfection with 100 ρmol si-RNA using Lipofectamine RNAiMAX ( Invitrogen ) according to manufacturer's instructions . A549 cells and GFP-LC3-A549 were reverse transfected . Human MDM were transfected with 3 µg of si-RNA using Amaxa Human Macrophage Nucleofector Kit according to manufacturer's instructions . Total RNA was extracted from 10×106 cells , isolated by using total RNeasy isolation kit Qiagen and treated with DNAse ( DNase Ambion Turbo DNA free , Ambion ) to remove genomic DNA according to manufacturers instructions . Oligotex Direct mRNA isolation kit ( Qiagen ) was used to isolate mRNA from total RNA and cDNA was synthesized using mRNA ( 0 . 5 µg ) by High Capacity RNA-to-cDNA Master Mix ( Applied BioSystems ) according to manufacturers instructions . IRGM mRNA was quantified with a quantitative real-time polymerase chain reaction ( qRT-PCR ) . qRT-PCR reactions were performed with the StepOnePlus Real-Time PCR System ( Applied ) using the FastStart Universal SYBR Green Master ( Rox ) ( Roche ) . cDNA was synthesized using the mRNA from the cells as template , and RT2 qPCR Primer Assay for Human IRGM ( Qiagen ) was used . The amount of measured transcripts was normalized to the amount of the Ribosomal protein S9 transcript . Melting curve analysis was performed after each run to analyse specificity of primers . We assessed the presence of contaminating genomic DNA using a minus-reverse transcriptase control in qRT-PCR experiments . 10×106 HeLa or Huh 7 . 5 cells were treated with Smartpool si-IRGM and control si-RNA were from Dharmacon . After mRNA extraction as previously described 4 volumes of cold ( −20°C ) acetone was added to cell lysates and incubated for 60 min at −20°C . Solution was centrifuged 20 min at 13 , 000 x g . Supernatant was discarded , pellet was dried and ressupended in NuPage LDS sample buffer ( Invitrogen ) with bond-breaker TCEP solution ( Thermo Scientific ) . HeLa cells were infected with the MeV Edmonston strain 6 h post-seeding at MOI of 3 for 24 or 48 hrs . Following 5 cycles of freezing at −80°C and defrosting at 37°C total infectious particles were quantified by limiting dilution on confluent Vero cells . Huh-7 . 5 cells were infected with HCV JFH1 strain 6 h post-seeding at a MOI of 1 . The level of virus particles present in culture supernatants was determined by end-point dilution and Core-specific immunofluorescence staining as described 24 or 48 hrs post-infection [49] . A549 cells were infected with influenza virus A/New Caledonia ( H1N1 ) at a MOI of 0 . 1 . Cell supernatants were harvested at 24 hrs post-infection and samples were titrated by plaque assay ( PFU ) in MDCK cells under agar overlay . HIV-1 infections were performed with normalized amounts of supernatants of R5 HIV-1-transfected cells . MDM were infected with 250 µL of a viral solution containing 170 ng/mL p24 for 2 hrs at 37°C . Cells were then centrifuged during 5 minutes at 1200 rpm and the supernatant was removed . The cells were cultured for 1 day in 1 mL of complete medium containing M-CSF . Infection was followed by measuring HIV-1 gag p24 in the supernatants of the infected cells using a p24 antigen capture ELISA ( Innogenetics ) . | Autophagy is a highly regulated cellular degradative pathway for recycling of long-lived proteins and damaged organelles . Autophagy is also used by host cells as a defense mechanism against intracellular pathogens . Autophagy can degrade pathogens or pathogen-derived molecules trapped within specialized vesicles named autophagosomes . Viruses and viral proteins are not an exception . However , since autophagy is a conserved pathway , viruses were submitted to an evolutionary pressure that led to the selection of molecular strategies which avoid or subvert this process to promote viral replication . Nevertheless the molecular details of viral interaction with autophagy remain largely unknown . We determined the ability of 83 proteins of several families of RNA viruses ( including Hepatitis C virus ( HCV ) , human immunodeficiency virus 1 ( HIV-1 ) , Measles virus ( MeV ) and influenza A virus ) to interact with 44 human proteins known to regulate autophagy and found that autophagy is highly targeted by RNA viruses . Strikingly , immunity-associated GTPase family M ( IRGM ) , known for its role in autophagy against bacteria , is the most targeted autophagy protein . Its absence is detrimental for HCV , HIV-1 and MeV production . Therefore , our data show that different RNA viruses families use similar strategies to fine tune autophagy to their own benefit . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"systems",
"biology",
"viral",
"immune",
"evasion",
"immunity",
"virology",
"immunity",
"to",
"infections",
"immunology",
"biology",
"microbiology",
"host-pathogen",
"interaction",
"genetics",
"and",
"genomics"
]
| 2011 | IRGM Is a Common Target of RNA Viruses that Subvert the Autophagy Network |
Pyrosequencing of PCR-amplified fragments that target variable regions within the 16S rRNA gene has quickly become a powerful method for analyzing the membership and structure of microbial communities . This approach has revealed and introduced questions that were not fully appreciated by those carrying out traditional Sanger sequencing-based methods . These include the effects of alignment quality , the best method of calculating pairwise genetic distances for 16S rRNA genes , whether it is appropriate to filter variable regions , and how the choice of variable region relates to the genetic diversity observed in full-length sequences . I used a diverse collection of 13 , 501 high-quality full-length sequences to assess each of these questions . First , alignment quality had a significant impact on distance values and downstream analyses . Specifically , the greengenes alignment , which does a poor job of aligning variable regions , predicted higher genetic diversity , richness , and phylogenetic diversity than the SILVA and RDP-based alignments . Second , the effect of different gap treatments in determining pairwise genetic distances was strongly affected by the variation in sequence length for a region; however , the effect of different calculation methods was subtle when determining the sample's richness or phylogenetic diversity for a region . Third , applying a sequence mask to remove variable positions had a profound impact on genetic distances by muting the observed richness and phylogenetic diversity . Finally , the genetic distances calculated for each of the variable regions did a poor job of correlating with the full-length gene . Thus , while it is tempting to apply traditional cutoff levels derived for full-length sequences to these shorter sequences , it is not advisable . Analysis of β-diversity metrics showed that each of these factors can have a significant impact on the comparison of community membership and structure . Taken together , these results urge caution in the design and interpretation of analyses using pyrosequencing data .
The recent advent of massively-parallelized pyrosequencing platforms has enabled a renaissance in the field of microbial ecology [1] , [2] . Pyrosequencing has engendered much enthusiasm since it is now possible to obtain nearly 100-times as many sequences by pyrosequencing for the same cost as using traditional Sanger sequencing technology . Although pyrosequencing is capable of generating 105–106 sequences per run , the sequences are between 100 and 400 bp in length . This method has become widely used among microbial ecologists to sequence PCR amplicons from variable regions within the ca . 1 , 500-bp 16S rRNA gene . These massive datasets have been analyzed through the generation of phylogenetic trees [e . g . 3] , assignment of sequences to operational taxonomic units ( OTUs ) for based on distance thresholds [e . g . 4] , and classification of sequences to phylogenentic bins based on similarity to reference sequences [e . g . 5] . Each approach has received some level of evaluation using pyrotag sequencing . Liu et al . [6] asserted that phylogenies generated using pyrotags were as good as full-length sequences based on similarity of UniFrac test statistics . Several studies have evaluated various regions and methods for assigning sequences to phylotypes [7]–[9] . Finally , a recent study emphasized differences in α-diversity metrics using different regions within the 16S rRNA gene and OTU definitions [10] . Each of these studies have focused on a limited range of phylogenetic groups found in a particular environment ( e . g . soil , mouse cecum , human feces ) and have glossed over more fundamental questions related to how alignment quality , methods of calculating pairwise genetic distances , sequence filtering , and region affects downstream analysis and their relationship to full-length sequences . Alignment quality is expected to significantly affect pairwise distances . Investigators have either used reference alignments to align sequences that implicitly incorporate the secondary structure of the 16S rRNA molecule [11]–[14] or they have used methods that do not consider the secondary structure [15] , [16] . Previous results have shown that the manually-curated SILVA reference alignment provides superior complementary base-pairing within the secondary structure compared to the greengenes alignment , which appears haphazard; the RDP alignment does not align the variable regions [11] . Considering the focus of these studies is on the variable region , there is the added complication that these areas are difficult to align accurately . To overcome limitations in alignment of variable regions , many studies have employed the use of masks to filter the troublesome regions [e . g . 3] . Yet , these filters remove a considerable amount of information from already information-sparse data ( Table 1 ) . The actual method of calculating distances is also typically taken for granted . Practically every 16S rRNA survey has made use of substitution models that assume that an alignment gap represents missing data instead of a mutation [e . g . 17] . The decision to use such a model seems motivated more by a sense of phylogenetic guilt than by biology . It is also unknown how distances calculated between partial sequences predict distances between full-length sequences . To make data analysis more tractable , some have employed heuristics based on correlations between kmer- and sequence-based pairwise distances to select which pairs of sequences to align and group within OTUs [16] . It is unclear how these correlations vary across regions within the 16S rRNA gene or what the level of risk is for falsely ignoring pairs of similar sequences . Finally , most studies make the implicit assumption that distances between partial sequences are not significantly different from those of full-length sequences; however , this is a questionable assumption as it is well-established that the 16S rRNA gene does not evolve uniformly along its length . This is apparent in the choice a 3% distance cutoff , which is used as a proxy species definition for full-length sequences , to define species using sequences from variable regions [e . g . 2] , [4] . Each of these factors is expected to have a significant effect on the analysis , interpretation , and generalizability of 16S rRNA gene surveys . Here , I used a collection of full-length 16S rRNA gene sequences representing 43 bacterial phyla to quantify how alignment quality , distance calculation methods , masking , and region within the 16S rRNA gene affect out ability to assess α- and β-diversity . The results of these analyses urge greater caution in how surveys are designed and interpreted .
For each of the 13 regions I used various alignment methods to calculate 91 , 131 , 750 pairwise distances assuming that a series of consecutive gaps represented one insertion or deletion . The SILVA , greengenes , and RDP alignments represent a gradation in the level of attention given to aligning the variable regions and are each guided by the secondary structure of the 16S rRNA gene . In contrast , the MUSCLE and pairwise alignments are attempts to optimize the alignment between sequences based on a limited number of parameters that are set a priori . To compare the pairwise distances calculated for the same pairs of sequences across alignments , I calculated the regression coefficients describing the relationship between the distances for the greengenes , RDP , MUSCLE , and pairwise alignments and the SILVA alignment for each region ( Table 2 ) . Distance calculations for this analysis assumed that consecutive gap positions were the product of a single insertion or deletion mutation ( i . e . one gap ) . With the exception of the V3 and V4 regions , the RDP alignment for each of the regions predicted greater genetic diversity than that of the SILVA alignment . Interestingly , the greengenes alignment , which does a poor job of aligning the variable regions , predicted between 9 and 33% more genetic diversity for each region than the RDP alignment , which does not attempt to align the variable regions . Visual inspection of the greengenes alignment suggests that in many instances the variable region alignments are somewhat random [11] . I observed that the MUSCLE-generated alignments described considerably greater genetic diversity than any of the other methods for the V3 , V6 , and V9 regions ( Table 2 ) ; however , the use of pairwise alignments yielded smaller distances than those calculated with the other alignment methods because pairwise alignment methods optimize the alignment without the constraint of preserving positional homology across multiple sequences ( Table 2 ) . Perhaps most worrisome is the observation that with the exception of the distances calculated from pairwise alignments , regressions of the other alignment methods to the SILVA-based alignment typically did a poor job of accounting for the variation in the distances ( Table 2 ) . These data make it clear that variation in alignment quality can have a significant impact on the genetic diversity that is calculated between the same pairs of sequences . Considering the poor correlation between the distances generated from the five alignment methods , it was necessary to determine the effect of this variation on the ability to accurately describe and compare communities . As expected based on the genetic distance analysis , the number of OTUs observed using the greengenes alignment was routinely higher than that observed using the other alignment methods and the number of OTUs observed using the pairwise alignment method was routinely the lowest ( Fig . 1 ) . Inspection of these lineage through time plots identified a stair-like appearance for many of the regions . This was due to the loss of information as sequence length decreased . The most extreme example of this phenomenon was for the V6 region that had an average sequence length of 60 bp . Each difference between a pair of V6 sequences changed the distance by approximately 0 . 0167 units , which is the step-length observed for the V6 data in Fig . 1 . When the phylogenetic diversity of the datasets was calculated , the greengenes aligned sequences had the highest phylogenetic diversity and the pairwise aligned sequences had the lowest ( Fig . 2 ) . One limitation of the phylogenetic diversity metric is that it is difficult to interpret the statistic and so it is unclear how biologically meaningful the level of variation observed is in Fig . 2 . To describe β-diversity , I used two OTU-based metrics ( Figs . 3 and 4 ) and two phylogenetic-based metrics ( Fig . 5 ) to measure the sensitivity of the metrics to alignment quality . Sequences were partitioned so that they would represent two samplings of communities whose Jaccard similarity index was 0 . 80 , but whose Morisita-Horn similarity index was 0 . 60 with a cutoff of 0 . 05 when defining OTUs with full-length sequences . Because the sampling of the two simulated communities was limited ( ca . 6 , 750 sequences per community ) , the Jaccard and unweighted UniFrac statistics did not equal the expected values . Within this simulation framework , the effect of alignment was generally highly statistically significant across metrics of β-diversity ( p≪0 . 001 ) ; however it is unclear how biologically meaningful the observed differences were . Using the same SILVA-aligned sequences that I analyzed above , I investigated the effect of different distance calculation methods on downstream analyses . Specifically , I considered the one gap calculator ( i . e . a gap of any length between two sequences represents a single mutation ) and each gap ( i . e . gaps length n , represent n mutations ) and ignore gap calculators ( i . e . gapped characters are not considered in calculating a distance; Table 3 ) . The slope of lines forced through the origin indicated that the each gap calculator calculated between 0 ( V4 ) and 9% ( V3 ) more genetic diversity than the one gap calculator . With the exception of the V3 region ( 69% ) , the regression between the each gap and one gap calculators accounted for more than 87% of the variation in the distances . The differences in the explanatory power of the regression were a function of frequency of gaps longer than 1 nucleotide . The ignore gap calculator calculated between 2 ( V4 ) and 7% ( V9 ) less genetic diversity than the one gap calculator . The regression between the ignore gap and one gap calculators accounted for more than 94% of the variation in the data . Until there is a more well-developed theoretical basis for selecting a method for treating gaps in sequence alignments , these results suggest that treating gaps of any length as a single mutation is a middle ground between ignoring them and treating each of them as a separate evolutionary event . Pairwise kmer distances were much larger than the alignment-based calculators and their regression onto the one gap calculated distances accounted for between 83 and 97% of the variation observed between the distances . In order to have no risk of falsely ignoring true one gap pairwise distances smaller than 0 . 10 , it was necessary to keep kmer distances smaller than 0 . 45 ( V19 ) to 0 . 73 ( V6 ) . This would result in needing to calculate between 3 . 3- and 9 . 1-fold more distances than would be needed by alignment-based methods . Lacking a theoretical basis for treating gaps as a single evolutionary event , I was curious how much measures of α- and β-diversity are affected by the choice of a distance calculator . I used an OTU-based approach to determine the effect of distance calculation methods on the richness of OTUs within the dataset ( Fig . 6 ) and a phylogeny-based approach using total branch length to measure phylogenetic diversity ( Fig . 7 ) . As would be predicted , the number of observed OTUs at any genetic distance was greatest with the each gap and least with the ignore gap calculators; the one gap and each gap calculators generated comparable numbers of OTUs . When I analyzed the effect of region and distance calculation method on the phylogenetic diversity of the datasets , there were qualitative trends between methods and regions that could have been predicted from the regression analysis in Table 2 ( Fig . 7 ) . These analyses suggest that the difference observed in α-diversity when using either the one gap or each gap calculator is unlikely to be biologically meaningful . I next investigated what effect each calculator method had on two OTU-based ( Figs . 8 and 9 ) and two phylogeny-based β-diversity measures ( Fig . 10 ) . For the OTU-based metrics , ignoring gaps resulted in an over-estimate of the similarity between the two communities and counting each gap resulted in an under-estimate . Increasing and decreasing the cutoff used to define the OTUs had a parallel effect on the Jaccard and Morisita-Horn indices ( Figs . 8 and 9 ) . These results occurred because ignoring gaps and increasing the threshold each dampen the differences between sequences and pull more sequences into an OTU so that more OTUs are likely to be shared; the same phenomenon was observed when sequences were filtered using the Lane mask ( see below ) . Penalizing each gap or making the OTU definition more stringent had the opposite effect . The calculated Jaccard coefficients were not significantly different between the one gap and each gap distance calculation methods when using the 0 . 03 and 0 . 05 OTU cutoffs ( Fig . 8 ) ; all four distance calculation methods yielded statistically significant differences in Morisita-Horn coefficients , regardless of the OTU cutoff . For the phylogeny-based methods , the observed differences between each of the distance calculation methods were statistically significant ( Fig 10 ) . Although the differences between distance calculation methods were highly statistically significant ( p≪0 . 001 ) , it is unclear how biologically meaningful the differences were . To circumvent alignment quality problems , the Lane mask has been used to filter variable regions from 16S rRNA genes . Results of analyses using filtered sequences aligned by any method or when distances were calculated by any method did not vary to a meaningful degree . Comparison of distances calculated using filtered sequences to those calculated using unfiltered sequences showed that filtering significantly reduced the genetic diversity observed between sequences ( Table 3 ) . With the exception of the V4 and V6 regions , masking removed between 15 and 45% of the genetic diversity . The V4 region is largely unaffected by the Lane mask and the average length of V6 sequences following the Lane mask treatment was only 27 bp , which made the resulting pairwise distances of dubious value ( Table 1 ) . As would be expected , the number of OTUs and phylogenetic diversity observed using Lane mask-filtered sequences was significantly lower than those calculated with the unfiltered sequences . For the four β-diversity measures , when the Lane mask-filtered sequences were analyzed , the communities appeared more similar than for non-filtered SILVA-aligned sequences ( Figs . 8–10 ) . One explanation for this observation is that because filtering makes sequences more similar to each other , it also makes communities appear more similar to each other . Although useful for broad-scale phylogenetic analysis at the level of a kingdom or phylum , filters remove the sequence information necessary to differentiate populations within a community . Ultimately , application of such filters is troublesome because it mutes the signals that differentiate communities . I compared the one gap distances calculated for each of the 12 regions from each alignment to the one gap distances calculated from the full-length SILVA alignments ( Table 4 ) . The regression of pairwise distances calculated from a sub-region onto distances calculated from full-length sequences was rarely near 1 . 00 . The most extreme case was the V6 region for which distances were nearly 3-fold higher than distances calculated using full-length sequences . Conversely , sequences from the V9 region were 33% less diverse than their full-length counterparts . In general , genetic diversity decreased along the length of the 16S rRNA gene . Although one could use these regression coefficients to relate data collected from one region to that from full-length sequences , the ability of the regression to explain the variation observed between sub-region and full-length sequences was quite poor . As expected , longer regions did the best job of relating the variation between sub-regions and full-length sequences . For example , when using the SILVA alignments , the regression of the V14 , V35 , and V69 distances onto the full-length distances accounted for 87 , 77 , and 77% of the variation in distances . Shorter regions such as the V3 , V6 , and V9 accounted for 26 , 36 , and 46% of the variation ( Table 4 ) . This analysis revealed that all sub-regions are limited in their capacity to serve as surrogates for full-length 16S rRNA gene sequences . The distance-based analysis clearly showed significant differences between distances calculated from sub-regions and full-length sequences . The OTU-based analysis in Fig . 1 demonstrates that there was a clear difference in the number of OTUs observed across regions for a given genetic distance as well as the level of curvature observe observed in the lineage-through-time plots ( Figs . 1 and 6 ) . In the phylogenetic-based analysis those regions that described more genetic diversity than the full-length sequences had greater phylogenetic diversity than the phylogenetic diversity calculated for the full-length sequences whereas the regions that described less genetic diversity yielded greater phylogenetic diversity ( Figs . 2 and 7 ) . Using pyrotag data introduces several complexities to β-diversity analyses . Moving across regions , but using the same OTU definition could lead one to overestimate community similarity . For example , the average Morisita-Horn similarity for full-length SILVA-aligned sequences with one gap distances was 0 . 56 . Using similarly treated sequences from the V12 , V13 , V14 , and V23 regions I calculated Morisita-Horn values between 0 . 57 and 0 . 60; however those from the other 8 regions yielded values between 0 . 64 ( V2 ) and 0 . 79 ( V9 ) . For a single region , changing the OTU cutoff also had a significant effect on the Morisita-Horn index . For instance , full-length SILVA-aligned sequences yielded 0 . 52 , 0 . 56 , and 0 . 86 for cutoffs of 0 . 03 , 0 . 05 , and 0 . 10 . This spread in Morisita-Horn values between the 0 . 03 and 0 . 10 OTU cutoffs ( 0 . 34 ) was the largest of any region . The narrowest spread was observed for the V6 region ( 0 . 06 ) . In contrast to the Morisita-Horn values , there was little variation in the unweighted or weighted UniFrac statistic when comparing sequences analyzed by the same alignment and distance calculation method . With the exception of the V6 region ( 0 . 33 ) , the average unweighted UniFrac values varied between 0 . 24 ( V13 , V14 , V19 ) and 0 . 30 ( V9 ) and with the exception of the V12 region ( 0 . 69 ) , the average weighted UniFrac values varied between 0 . 80 ( V13 ) and 0 . 87 ( V9 ) ; the value for the full-length sequence was 0 . 82 . Similar to the α-diversity measure of phylogenetic diversity , an added complication of phylogeny-based methods is the complexity of interpreting the proportion of branch length that is shared between or unique to two communities and how such proportions relate to classical β-diversity measures . Thus , it is difficult to interpret the biological significance of such variation . Regardless , the results of the OTU- and phylogeny-based analyses demonstrate that caution must be taken in extrapolating results from one region to another .
The ability to define OTUs and reconstruct phylogenies allows an investigator to approach their problem using the data as they present themselves without being confined to an a priori taxonomy . Regardless , the analysis I have presented indicates that comparing results obtained by sequencing one region of the 16S rRNA gene can not be easily compared to those obtained using full-length sequences . Ultimately , the fact that the 16S rRNA gene does not evolve uniformly across its length complicates its analysis . Technical limitations require investigators to select a region based on the availability of conserved PCR primers , fragment length , and the ability to generate high quality sequence . Analytical limitations require investigators to select a region based on the availability of database sequences for that region , the ability to accurately classify sequences , and the level of genetic diversity found in the region . Until there is a standardized approach , individual investigators will continue to select different regions for their analysis . Studies such as this are necessary to inform investigators about the strengths and weaknesses of the various regions within the 16S rRNA gene . Based on this analysis , it is clear that regardless of the region , longer reads will improve one's ability to relate their analysis to full-length sequences . As sequence lengths increase to the point that pyrosequencing full-length 16S rRNA genes is possible , this discussion will be unnecessary . Ultimately , all pyrotag regions represent a marker of a marker of genomic diversity . Even if full-length 16S rRNA gene sequencing is possible , it is still just a marker of genomic diversity . Correlations between the complete genome sequence and full-length 16S rRNA gene sequences are probably just as poor as correlations between full-length 16S rRNA gene sequences and their sub-regions [e . g . 18] . Although one may endeavor to characterize and compare the composition of multiple communities , any cutoffs that are employed are at best empirical and hopefully have some biological meaning . I have shown that alignment quality has a significant impact on downstream data analysis . Because the 16S rRNA gene sequence follows a well-determined secondary structure , it is possible to objectively state that one alignment is better than another . Furthermore , pairwise and multiple sequence alignments that ignore the secondary structure are unadvisable on theoretical grounds . Such methods are also unadvisable on technical grounds as the time and memory required to complete them typically scales in excess of the number of sequences squared; the time required to perform a profile-based alignment scales linearly with the number of sequences . A significant factor in the analysis of DNA sequences is the calculation of pairwise distances . The rich literature developed for protein-coding sequences has generated the Jukes-Cantor , Kimura , Hasegawa-Kishino-Yano and other substitution models [reviewed in 19] . Yet these models ignore gapped positions , which I have shown to have a significant impact on downstream analyses . Substitution models for structural RNA molecules such as the 16S rRNA gene are not well developed or widely used [20] , [21] . It is underappreciated that use of short sequence or filtering methods such as the Lane mask reduces the precision and information represented by a distance . For instance , if there are fewer than 200 bases being considered , then it is difficult to place much confidence in an OTU threshold of 0 . 03 ( i . e . 6 differences ) when one considers the potential impact of PCR , sequencing , and alignment artifacts . Furthermore , reducing the information content of a 1 , 500 bp molecule to a 200-bp sequence read will affect the confidence placed in the generation of phylogenetic trees and OTU assignments . Althoguh removing non-informative positions can be helpful for reconstructing broad phylogenies , the α- and β-diversity analyses described here are adversely affected by removing this fine level sequence diversity . These are clearly issues that warrant further attention . This study has ramifications on how analyses are performed . Since it is clear that the 16S rRNA gene does not evolve uniformly across its length , it is critical that sequences fully overlap before they are compared . For example , consider an analysis that includes sequences from the V2 region and those from the V12 region . The V12 sequences will have higher pairwise distances amongst each other than compared to the V2 region because the V1 region is evolving at a faster rate . Thus , the comparison of short and long sequence reads will add artifacts into the analysis , which will overstate the richness within the community . Although not explored here , it is likely that similar problems will be encountered in analyses where a taxonomy hierarchy is used to assign sequences to bins . Thus it is critical that sequences are trimmed to start and end at the same sequence-based landmarks . Because pyrosequencing does not yield a uniform length sequence read , this introduces a conundrum of whether to favor fewer long reads or many short reads . Because it is impossible to compare pyrotags to the full-length sequences accurately , it seems appropriate to increase the power of other statistical analyses by sacrificing sequence length in favor of having more sequence reads . Next generation sequence analysis of 16S rRNA genes offers the first opportunity to replicate analyses , develop more complex experimental designs , and to increase sampling depth and breadth . The results of this study encourage one to see pyrotags as markers within a metagenome and suggest a different way of considering microbial community analysis . Just as single nucleotide polymorphisms ( SNPs ) have been used as markers of disease in genome-wide association studies ( GWAS ) , which may have no direct effect on a genes phenotype , pyrotags no doubt will serve as a useful analog to SNPs for the nascent field of metagenome-wide association studies ( MWAS ) .
I obtained the SSURef 16S rRNA gene sequence database from the SILVA project ( version 98; http://www . arb-silva . de ) [22] . From this collection of sequences longer than 1 , 200 bp , I identified bacterial sequences that had an alignment quality score ( ARB database field “align_quality_slv” ) of 100 and were not chloroplasts , mitochondria , or suspected of being chimeric . The collection was further screened to remove sequences that had more than 5 ambiguous base positions and did not start by E . coli position 28 or end after position 1491 . Of the remaining sequences , 13 , 501 sequences were unique and shared between the SILVA [22] , greengenes [23] , and RDP sequence collections [14] . I then generated 12 datasets from the full-length sequences using the SILVA , greengenes , and RDP alignments by extracting sub-regions of various lengths ( Table 1 ) . These regions were selected because they had already been used in publications or are amenable to the available sequencing platforms . Lane masks were generated by mapping the original mask onto the E . coli reference sequence and then it was applied to each of the three reference alignments [24] . In addition to the three reference alignments I generated pairwise alignments between all pairs of sequences using the Needleman-Wunsch algorithm [25] and multiple sequence alignments using MUSCLE with two iterations ( maxiters = 2 ) and the diags option [15] . I implemented three sequence-based methods for calculating pairwise distances and a kmer-based distance metric . The first sequence-based method ignored any site that contained a gap; this method is implemented in the commonly used DNADIST program from the PHYLIP package [26] . The second sequence-based method counted gaps as a fifth character so that any comparison between a gap and a base was penalized as a mismatch; comparisons between two gaps were ignored . This approach asserts that every gap represents a distinct mutation . The third sequence-based method calculated distances by only penalizing a string of gaps as one mismatch [2] . This approach asserts that a gap , of any length , represents a single mutation . Distances were not corrected for multiple substitutions to simplify analysis of the data . Furthermore , some distances were so large that when they were corrected , they yielded undefined values . Distances were calculated as implemented in the mothur software package with precision to 0 . 0001 [27] . Finally , kmer-based distances were calculated between pairs of unaligned sequences based on their 7-base kmer profiles [28] . Pairwise distances were compared using a custom C++-coded program that calculated the linear regression coefficient using the origin as the intercept and the Pearson product-moment correlation coefficient [29] . Because several of the datasets did not demonstrate a linear correlation with the V19 region when the V19 pairwise distances were larger than 0 . 10 , all regression and correlation coefficients are presented for V19 distances smaller than 0 . 10 . Assessments of how much genetic diversity was either gained or lost represent the deviation from a slope of 1 . 0 . The square of the Pearson product-moment correlation coefficient ( i . e . R2 ) was used to quantify the fraction of the variation that was accounted for by the linear regression . OTU- and phylogeny-based analyses were performed to assess the intra-sample biodiversity . Sequences were assigned to OTUs using the mothur implementation of the furthest-neighbor clustering algorithm [27]; although parallel analyses using the nearest and average neighbor algorithms yielded different α- and β-diversity values , the overall relationships observed with furthest neighbor algorithm were observed . The observed richness ( i . e . the number of OTUs in a sample ) of the dataset was calculated using every possible cutoff that the data could describe . Traditional neighbor-joining trees were generated using the clearcut software program and the distance matrices that were used in the OTU-based analyses [30]; however , the relaxed neighbor-joining algorithm was not used . The phylogenetic diversity of the data was calculated by summing the branch length for the entire tree [31] . Both analyses were replicated 50 times to assess the effects of randomization on α-diversity . The OTU assignments and neighbor-joining trees created to study α-diversity were used to evaluate the effects of each variable on the ability to calculate β-diversity . Towards this end , I segregated the sequences to create two mock communities that shared 80% of their membership but had different structures . To create the mock communities full-length SILVA-aligned sequences were first assigned to OTUs using a furthest neighbor clustering of one gap distances with a cutoff of 0 . 05 . Second , OTUs were randomly ordered . Third , 10% of the OTUs were assigned exclusively to the first community , another 10% were assigned exclusively to the second community , and the remaining OTUs were shared . For half of the shared OTUs , the probability of a sequence being from the first community was 0 . 375 and for the other half of the shared OTUs , the probability was 0 . 625 . These probabilities were selected to simulate sampling two communities that had a Jaccard similarity index of 0 . 80 and Morisita-Horn Index value of 0 . 60 . This process was repeated to create 100 simulated communities . Because the mock communities were not exhaustively sampled , it was unlikely that the measures would actually equal 0 . 80 and 0 . 60 for the Jaccard and Morisita-Horn indices . All β-diversity calculations were made using the mothur software package [27] . The same 100 partitions were used to analyze all distance calculation methods , alignments , regions , and β-diversity measures . I analyzed the effects of region and the alignment or distance calculation methods using a two-way analysis of variance . Each factor was highly significant ( p≪0 . 001 ) and so I used the Tukey's honestly significant difference test for pairwise comparisons . Only those differences , which were non-significant ( p>0 . 05 ) are indicated in figures . All test were performed within an OTU cutoff or UniFrac method . | Microbial communities are notoriously difficult to analyze because of their inaccessibility via culturing and high diversity . Next generation sequencing technologies have made it possible to obtain deep sampling coverage of the 16S rRNA gene; however , interpretation of the resulting data is complicated by the inability to relate sequences from variable regions within the gene to the full-length gene and ultimately , the parent genome . Here , I present a comprehensive analysis quantifying the effects of varying sequence alignment quality , pairwise distances calculation methods , sequence filtering , and regions within the 16S rRNA gene on downstream analysis using OTU- and phylogeny-based methods . This analysis indicates that each factor can have a significant effect on descriptions of α- and β-diversity . Because it is not possible to relate pyrotags to full-length 16S rRNA gene sequences directly , I encourage scientists to view pyrotags as markers within a microbiome in an analogous fashion to how geneticists view single nucleotide polymorphisms as markers within genomes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"microbiology/environmental",
"microbiology",
"computational",
"biology/comparative",
"sequence",
"analysis",
"computational",
"biology/genomics"
]
| 2010 | The Effects of Alignment Quality, Distance Calculation Method, Sequence Filtering, and Region on the Analysis of 16S rRNA Gene-Based Studies |
Kinesins are nano-sized biological motors which walk by repeating a mechanochemical cycle . A single kinesin molecule is able to transport its cargo about 1 μm in the absence of external loads . However , kinesins perform much longer range transport in cells by working collectively . This long range of transport by a team of kinesins is surprising because the motion of the cargo in cells can be hindered by other particles . To reveal how the kinesins are able to accomplish their tasks of transport in harsh intracellular circumstances , stochastic studies on the kinesin motion are performed by considering the binding and unbinding of kinesins to microtubules and their dependence on the force acting on kinesin molecules . The unbinding probabilities corresponding to each mechanochemical state of kinesin are modeled . The statistical characterization of the instants and locations of binding are captured by computing the probability of unbound kinesin being at given locations . It is predicted that a group of kinesins has a more efficient transport than a single kinesin from the perspective of velocity and run length . Particularly , when large loads are applied , the leading kinesin remains bound to the microtubule for long time which increases the chances of the other kinesins to bind to the microtubule . To predict effects of this behavior of the leading kinesin under large loads on the collective transport , the motion of the cargo is studied when the cargo confronts obstacles . The result suggests that the behavior of kinesins under large loads prevents the early termination of the transport which can be caused by the interference with the static or moving obstacles .
Kinesins move 8 nm per step [1–4] along MT by using energy obtained from ATP hydrolysis [5–7] . Among the kinesin superfamily , this study focuses on kinesin–1 ( referred to simply as kinesin ) which has two identical heads [8 , 9] . Several experiments have measured the run length of cargoes transported by kinesins . Block et al . [10] reported that cargoes move about 1 . 4 μm when they are pulled on average by about one and half kinesin molecules . The distribution of the run length in those experiments follows an exponential probability distribution . The effects of external resisting loads exerted on the cargo and the concentration of ATP on the run length was observed by Schnitzer et al . [11] . The run length decreases with increasing resisting load and decreasing ATP concentration . In the experiment performed by Uemura et al . [12 , 13] , the magnitude of the loads causing unbinding were measured . They exerted loads toward the plus or the minus end of the MTs to discover the effects of the direction of the load . Their results show that kinesins tend to unbind more easily when subjected to loads toward the plus end of the MTs than by loads toward the opposite direction . However , the difference is not considerable . The experiment of Beeg et al . [14] focused on the transport of cargoes by groups of kinesins . To observe the relation between the number of kinesins and the run length , they varied the number of kinesins attached to the cargo . The run length increased as more kinesins participate in the transport . However , the run length was surprisingly reduced when the cargo was moved by considerably many kinesins . Mathematical models have been proposed to calculate the run length of kinesin . Schnitzer et al . [11] established an equation regarding the run length of a single kinesin molecule by using Arrhenius-Eyring kinetics . The approach produced a successful fit to experimental data for various ATP concentrations and external loads . However , the model is only applicable to the motion of single molecules . For transport by several kinesins , Klumpp et al . [15] utilized discrete Markov chains to obtain a master equation regarding the number of motors which effectively participate in the transport . Then , they obtained the stationary solution for the master equation . By substituting the transition rates of kinesins ( i . e . , binding rate to MT , and unbinding rate from MT ) , their model obtained an analytical solution for the mean value and the probability density function ( pdf ) of the run length . Their unbinding model accounts for the effects of load by assuming that the load is equally distributed over every kinesin bound on the MT . However , the distribution of loads over motors continuously changes due to the stochastic motion of kinesins [16–18] . Furthermore , their binding model is not able to capture the locations where rebinding occurs , despite the fact that those locations also affect the collective transport . The goal of this study is to develop a binding/unbinding model which is able to capture the stochastic unbinding and binding of kinesins to the MTs and their dependencies on the force acting on them . By using the model , the run length and velocity of collective transport under constant loads are obtained . The characterization on the unbinding of kinesins captures an interesting behavior of kinesins , namely that they spend a long time remaining on the MT when large resisting loads are applied . The model predicts that this behavior of kinesin is beneficial for the cargo to overcome obstacles . Also , the velocity of collective transport is affected by the stochastic rebinding process as well as by the velocity of kinesin molecules itself .
The unbinding probability during the state [K+MT] is calculated over time using the transition rate kD0 from the state [K+MT] to the unbound state [K] , as shown in Fig . 2 . If the ith cycle starts at an instant ti , the unbinding probability in state [K+MT] during this cycle is obtained by solving the following set of equations with the initial condition P[K+MT] ( t = ti ) = 1 . d d t P [ K + M T ] = - k D 0 P [ K + M T ] , P [ K + M T ] ( t ) + P D 0 ( t ) = 1 , ( 1 ) where P[K+MT] is the probability that kinesin remains attached to the MT , and PD0 ( t ) is the unbinding probability in state [K+MT] . Thus , the growth of PD0 over time can be calculated as P D 0 ( t ) = 1 - exp ( - k D 0 t ) ( 2 ) The time constant of the cargo motion transported by single kinesin was experimentally measured by Carter et al . [20] to be approximately 15 . 3 μs for a resisting load of 5 pN applied to the cargo . The dwell time of kinesin has also been measured experimentally by numerous researchers for that load , including Visscher et al . [21] . That dwell time is about 70 ms . Hence , the duration of state [K . ATP + MT]1 is much shorter than the dwell time and thus negligibly short . Due to this very short duration of state [K . ATP + MT]1 , a single unbinding probability value ( i . e . , PD1 ) is used for this state instead of capturing the changes of the probability over time . This model is able to predict the instant of transition between bound states . Thus , the effect of ATP concentration on unbinding is intrinsic to the model . The model also accounts for the effects of force by using Bell model [22] and expressions inspired by Boltzmann’s law as k D 0 = k D 0 , 0 exp ∣ F k ∣ d 0 k B T , P D 1 = P D 1 , 0 exp ∣ F k ∣ d 1 k B T , ( 3 ) where kD0 , 0 , PD1 , 0 , d0 , and d1 are parameters of the unbinding model . Fk is the force transferred from the cargo to the kinesin . The equation for this force is provided in S1 Text . The occurrence of an unbinding event is determined by comparing the calculated probability with uniformly distributed random numbers . The transport performed by a single kinesin does not include the rebinding process . Instead , the experimentally observed run lengths of single kinesins [11] are used to determine parameters of the unbinding model . The parameters of the model are obtained so that the model predicts run lengths measured experimentally . The fitting is done by using the nonlinear least-squares fit function ( lsqnonlin ) in MATLAB . Note that the results of the fitting indicate that the effects of the load on unbinding during the state [K + MT] are very weak . Hence , d0 is very small compared to other parameters . Thus , the value of d0 is set as zero . Table 1 represents the values of the parameters , and Fig . 3 shows the run length of the experiments and the model . The values of the parameters can be changed by the interactions between kinesins . When the number of kinesins is small ( e . g . , between 1 and 5 ) and not considerably large ( like hundreds ) , the effect of interference among kinesins on the unbinding of kinesins is assumed negligible . Thus , the parameters of unbinding used in this study apply to the cargoes transported by one to five kinesins . The kinesin released from the MT can possibly bind again to several binding sites , as shown in Fig . 4 . The probability of rebinding to the jth binding site is calculated using the transition rate kA , j . The value of kA , j decreases as the distance between the unbound kinesin and the jth binding site increases . The position of the neck and heads of the unbound kinesins are assumed to be the same . Thus , those positions are denoted by a single variable xu , k . The dependence on the distance is assumed to have a parabolic distribution , as shown in Fig . 4 . Since the time scale of the thermal fluctuations of an unbound kinesin is very short , the value of kA , j ( which depends on the position of the unbound kinesin ) also changes rapidly over time . Intensive computations are required to capture the value of kA , j over time . Instead of calculating the rapidly changing kA , j , the following method is used . First , the pdf of the position of the unbound kinesin is obtained by using the strain energy in the kinesin structure . Then , the time average of kA , j is obtained at every time step by spatially integrating the value of kA , j weighted by the pdf of the unbound kinesins . By using this method , the amount of computation reduces significantly because the time step can be determined by the dynamics of bound kinesins not by the fluctuating motion of the unbound kinesin . The simulation begins with the state where one kinesin is bound to the MT . The dynamics of the cargo , the chemical reactions and the unbinding are considered for bound kinesins , and rebinding to the MT are considered for unbound kinesins at every time step . The positions of the necks of bound kinesins and cargo are obtained by using the mechanistic model ( which is described in S1 Text ) . Then , the chemical states the bound kinesins are determined from the mechanistic model . At the beginning of every cycle of bound kinesin , the unbinding probability is zero . Four uniformly distributed random numbers ( e . g . , rw , rd0 , rd1 , and rb ) are also generated between 0 and 1 at the beginning of every cycle for each bound kinesin . The cycle of kinesin starts with state [K + MT] . First , if the value of PD0 becomes rd0 before the transition from [K + MT] to [K . ATP + MT]1 occurs , then the kinesin unbinds when PD0 is equal to the value of rd0 . Otherwise , the chemical state of kinesin changes to [K . ATP + MT]1 when P[K+MT] becomes rw . Then , if the value of PD1 is higher than rd1 , then the kinesin unbinds during the diffusion of its free head . Otherwise , the free head moves to the next binding site without unbinding . At this moment , the free head can move to the forward binding site or to the backward binding site . The backward motion of kinesin can be captured by considering the diffusing motion of the kinesin head which is affected by the force acted on the kinesin molecule [25] . In this study , if the probability of backward steps ( which is obtained using experimental results [20] ) is larger than the random number rb , the kinesin is assumed to move backward . Otherwise , it walks toward the plus end of the MT . When P[K+MT]+P[K . ATP+MT]2 becomes rw , the transition from [K . ATP + MT]2 to [K . ADP . Pi + MT] occurs . If two or more kinesins are attached on the cargo , the probability distribution of rebinding for every unbound kinesin is also calculated at every time step . First , uniformly distributed random numbers ra1 and ra2 are generated for each unbound kinesin . If the summation of the rebinding probabilities over binding sites ( shown in Fig . 5 ( g ) ) are larger than ra1 at a certain time step , the rebinding occurs . Otherwise , the unbound kinesin does not bind to the MT at this time step . If the kinesin is determined to bind at this time step , the rebinding probabilities are normalized with their summation . Then , the normalized rebinding probabilities are cumulated over binding sites . The kinesin binds to the binding site where this cumulative value is larger than ra2 . The run length and velocity calculated from the model were compared with the previous experimental data [14] . They mixed beads with kinesins to obtain cargoes coated with kinesins . Different concentrations of kinesins ( ck ) were used to observe the effects of the number of kinesins attached to single cargoes . Their results are shown in Fig . 8 . The results of the model for the used ck are obtained as follows . The run length distribution and mean velocity of the cargo transported by one to five kinesins were calculated from the model . Then , the weighted averages of these results were calculated by using the probability regarding the number of kinesins ( which was proposed in a previous study [14] ) for each ck . The run length distribution and mean velocity of the model are similar to the experimental results , as shown in Fig . 8 .
The unbinding probabilities for the states [K+MT] and [K . ATP + MT]1 ( i . e . , PD0 and PD1 ) are calculated for various forces applied to single kinesins and for various ATP concentrations , as shown in Fig . 9 . The values of PD0 in Fig . 9 are unbinding probabilities when the time t in Equation 2 is the average duration of state [K+MT] . When the resisting load is not significant , the duration is very short . Thus , PD0 is close to zero ( e . g . , for loads FL between 0 and 5 pN at [ATP] = 5 μm ) . For large resisting loads , the duration of state [K+MT] is long . Hence , PD0 is large and increases with the load ( e . g . , for FL loads between 5 and 12 pN at [ATP] = 5 μm ) . Also , PD0 exponentially converges to 1 by Equation 2 ( e . g . , for a load FL of approximately 12 pN at [ATP] = 5 μm ) , as shown in Fig . 9 ( b ) . Note that PD1 is much higher than PD0 for a wide range of forces when the ATP concentration is high . This difference suggests that most kinesin molecules unbind while in the state [K . ATP + MT]1 if the ATP concentration is high enough ( so that an ATP molecule binds to the kinesin head very fast ) . This rapid binding of ATP to the kinesin head decreases PD0 significantly . This characterization on the unbinding of kinesins predicts that kinesins mostly unbind when their free heads move to the next binding sites . If large resisting loads act on a kinesin , then the time required to complete one cycle of kinesin is very long . Thus , the interval between steps is also very long for the large loads . Consequently , the time until the unbinding of the kinesin occurs is very long when the load is large . This is a specific kinesin behavior which we refer to as highly loaded behavior ( HLB ) of kinesins . Because the leading kinesin undergoes the largest resisting force among kinesins , HLB is mostly likely to be observed in the leading kinesin . However , when the resisting load acting on the cargo is very large , then the next or second next leading kinesins also can exhibit HLB together with the first leading kinesin . The effects of HLB on the collective transport are described in the following . Transport by kinesins can be inhibited by other surrounding particles which act like obstacles . To consider the effect of cellular particles on the motion of the cargo , static and moving obstacles are modeled in this study . First , a static obstacle is located ahead of the cargo , as shown in Fig . 10 ( a ) . When the cargo confronts a static obstacle , the cargo is assumed stuck to the static obstacle until the sum of the forces generated by kinesins exceeds a force Fobs which is required to overcome the obstacle . Second , the obstacle is assumed to move backward , toward the minus end of the MTs . This motion pushes the cargo backward with a velocity Vobs , as shown in Fig . 10 ( b ) . The retrograde motion due to the moving obstacles also vanishes when the sum of the forces generated by kinesins exceeds a force Fobs . To check the role of HLB in overcoming obstacles , three virtual motors are created to compare their abilities of overcome obstacles with that of kinesin . The virtual motors are designed by modifying properties of kinesin which are necessary for HLB . First , the virtual motor 1 is modeled so that the unbinding probabilities for three states ( i . e . , [K+MT] , [K . ATP + MT]1 and [K . ATP + MT]2 ) are the same . To apply this modification , the unbinding probabilities corresponding to the current force on the kinesin are added together . Then , that summation is divided by three and assigned to the current state so that the motor has equal unbinding probabilities for those three states . Second , the virtual motor 2 has a stepping frequency which is invariant over forces acting on the motor . This modification is accomplished by removing the dependency of the chemical reactions on forces . Third , the virtual motor 3 has both of the changes of virtual motors 1 and 2 . Note that the unbinding probability ( of a single molecule ) per step is the same for all motors , virtual and actual . However , the responses of the teams composed of each type of motors in the presence of obstacle are considerably distinct . The team of two actual kinesins show higher probability Poc of overcoming one obstacle without dissociating from the MT compared to other teams , as shown in Fig . 10 . The difference is remarkable for the team of virtual motors 2 . This suggests that the low stepping frequency for high loads is the primary reason for the HLB of kinesin . In addition , if a team of motors confront n-obstacles , the overcoming probability decreases to P o c n . Then , the cargo transported by virtual motors is not able to reach the final destination , while actual kinesins can resist the interruption of obstacles by using their HLB . To obtain run lengths for teams of kinesins , both unbinding and rebinding are considered together with the mechanistic model . At the beginning of the transport , only one kinesin is bound to the MT , and every other kinesin is unbound . Each transport is assumed to be terminated when all kinesins detach from the MT . The run length is defined as the difference in the position of the cargo at the beginning and the position at the termination of the transport . To minimize the error from Monte Carlo simulation , a large number of data is obtained for each load and number of kinesins in the team . The average of the run length over the number of used data converges at about 100 to 150 sets of data . The results presented were obtained using 200 sets to calculate average values . When several kinesins are involved in the transport , the run length of the team increases with the number of kinesins in the team , as shown in Fig . 11 ( a ) . The run length of single kinesins monotonically decreases with the load . However , the run length of a team of kinesins increases with the load when the load is larger than 8 pN . This interesting feature can be explained by the HLB of kinesins , as shown in Fig . 11 ( b ) . HLB is attributed to properties of the unbinding and stepping frequency . Kinesins mostly unbind when the free heads move to the next binding site . Thus , the unbinding probability for a given time interval increases as the kinesin takes more steps in a given time interval . The stepping frequency of kinesins is decreased by resisting loads . As a consequence , when the high resisting force acts on a kinesin , the kinesin walks slowly and remains bound on the MT for a long time , like an anchor . During this long time , other unbound kinesins attached to the same cargo have time to bind to the MT , as shown in Fig . 11 ( b ) . It is unlikely that large resisting loads , larger than 8 pN , continuously acts on the cargo in cells . However , this HLB can be used in cells to make the transport more robust . Also , we note that the run length will decrease for loads FL close to FL = 7 × n [pN] , where n is the number of kinesins attached to the cargo . For those large loads , the cargo will move backward because the kinesins walk backward with high probability when a resisting load close to 7 pN is applied to a single kinesin . In the absence of load , a single kinesin moves with a velocity of about 800 nm/s [21 , 26–28] . The velocity of the transport performed by several kinesins was predicted in this study . For moderate resisting loads , the transport is realized with slower velocity compared to the motion of single kinesins . While the velocity of single kinesins reach a maximum around 800 nm/s at small loads , the motion of a team of kinesins is accelerated by assisting loads , as shown in Fig . 12 ( a ) . The motion of collective transport can be decelerated or accelerated by binding and unbinding of kinesins . When an assisting load acts on the cargo , the unbound kinesin is likely to bind to a binding site located in front of the other , already bound kinesin as shown in Fig . 12 ( b ) . Consequently , the cargo moves forward a distance longer than just 8 nm ( one kinesin step ) . Hence , the velocity of the cargo increases beyond that of a single kinesin , as shown in Fig . 12 ( a ) . This behavior contrasts single molecule experiments at high ATP concentrations where the velocity is not increased by assisting loads . Also , resisting loads ( less than 6 pN ) decrease the transport velocity . That is because , for resisting loads , the unbound kinesin rebinds with high probability behind the other , already bound kinesin as shown in Fig . 12 ( c ) . Thus , the transport is decelerated by resisting loads less than 6 pN , as shown in Fig . 12 ( a ) . The velocities for resisting loads higher than 7 pN are larger than the single kinesin velocity . This difference is due to the degree of cooperative work of kinesins for various loads [16] . Due to the slack behavior of kinesins , some bound kineins in a team do not generate forces for transport if the load is small . However , a team is predicted to work in a cooperative fashion for high resisting loads .
When several kinesins are involved in a transport , the characteristics of the transport are significantly affected by the unbinding and rebinding of the kinesins . Novel methods to predict the probability to rebind were presented to reduce the computational effort dramatically . Particularly , the conversion from time average to spatial integration is the key advantage . Note that this approach can be used for other systems which have properties fluctuating with high frequencies . This method revealed that the instant of rebinding depends very weakly on the load . The changes of the run length and of velocity over the load are different for a single kinesin and for a team of kinesins . The possibility of transport against resisting loads larger than 7 pN per motor implies that the capabilities of a single kinesin can be enhanced by teams of kinesins . To transport a cargo in cells , the cargo needs to navigate in highly viscoelastic cytoplasm which is filled with several particles [29–31] . A single kinesin is not reliable to perform that task for at least two reasons . First , when the resisting load is higher than 7 pN , processivity of single kinesins is not guaranteed because its backward motion occurs frequently . Second , the observed distribution of the run length [10 , 32–35] indicates that most of single kinesins detach from the MT before they reach 1 μm . However , a group of kinesins in this study shows comparable velocity toward the plus end of the MT for a broad range of loads . The velocity of collective transport has three noticeable characteristics compared to the velocity of single kinesins . First , the cargo of several kinesins moves faster than the cargo transported by a single kinesin for assisting loads and slower for resisting loads less than 6 pN . Second , the increase in transport velocity due to assisting loads is larger than the decrease due to resisting loads . Third , the magnitude of the changes in velocity is similar for two and three kinesins . Although the cargo of three kinesins experiences more frequent binding and unbinding events , the distances of the anterograde and retrograde motion of the cargo indicated by red arrows in Figs . 12 ( b ) and 12 ( c ) are shorter for three kinesins . These characteristics are also observed in previous experiments . Dujovne et al . [36] applied external forces on MTs in their inverted gliding assay by using an electric field . In their experiments , the velocity increases or decreases due to assisting or resisting loads . Also , the change in the velocity is larger for assisting loads , like our results . The low and medium density of kinesin shows a similar velocity change over the electric field . This is also consistent with the velocity predictions in this study . The team operation of kinesins also prevents the early termination of the transport by obstacles in cells . The interference between the cargo and the obstacles can result in significant loads on the cargo . Especially in axons where several MTs are aligned , the motion of the cargo can be affected by static obstacles such as MT associated proteins , or tangles of the proteins [37] . Also , the cargo could encounter other cargoes transported by different motor proteins such as dyneins which move reverse to the walking direction of kinesin . The cargoes transported toward the minus end of MTs act as moving obstacles to the anterograde transport of kinesins . The huge load resulting from the static and moving obstacles is fatal to the transport if the cargo is transported by a single kinesin . However , if several kinesins are attached to the cargo , one or more kinesins function as temporary anchors until the transient load vanishes . Thus , kinesins are able to continue and complete their task as a team . The comparison of transport by actual kinesins and virtual motors suggests that the reduction in velocity of kinesins for high resisting loads is necessary to maintain cellular transport in the presence of obstacles . When the motion of the cargo is blocked by obstacles , the leading kinesin acts like an anchor as its velocity decreases . Thus , time is available for the other kinesins in the team to cooperate and overcome the interruption due to the obstacle . Together with the velocity of the transport by several kinesins , the HLB suggests that cells utilize teams of kinesins for the ( harsh ) cellular transport . | Kinesins are molecular motors which work when they bind to tracks called microtubules ( MTs ) . In a cell , cargoes of kinesins are transported to very distant destinations compared to a short transport ability of single kinesins . To study this difference , we establish a novel quantitative model capable of capturing the transport by teams of kinesins . While an entire team does not completely dissociate from a MT easily , individual kinesins continuously bind and unbind from MTs . Because the binding occurs faster than the unbinding process , the transported distance increases significantly over the number of kinesins in the team . Another barrier on cellular transport is the disturbance to the motion of the cargo by other cytoplasmic particles . Thus , we also consider the effect of cellular particles by exerting temporary high resisting loads on the cargo . Our model predicts that kinesins remain bound on the MT for a long time when the high resisting loads act on them . This behavior increases the probability that other unbound kinesins rebind to MTs . Consequently , more kinesins can participate in the transport . Then , the cargo is able to overcome interruption by other cellular particles . Thus , a robust intracellular transport can be realized by this mechanism . | [
"Abstract",
"Introduction",
"Models",
"Results",
"Discussion"
]
| []
| 2015 | Highly Loaded Behavior of Kinesins Increases the Robustness of Transport Under High Resisting Loads |
We address the question of color-space interactions in the brain , by proposing a neural field model of color perception with spatial context for the visual area V1 of the cortex . Our framework reconciles two opposing perceptual phenomena , known as simultaneous contrast and chromatic assimilation . They have been previously shown to act synergistically , so that at some point in an image , the color seems perceptually more similar to that of adjacent neighbors , while being more dissimilar from that of remote ones . Thus , their combined effects are enhanced in the presence of a spatial pattern , and can be measured as larger shifts in color matching experiments . Our model supposes a hypercolumnar structure coding for colors in V1 , and relies on the notion of color opponency introduced by Hering . The connectivity kernel of the neural field exploits the balance between attraction and repulsion in color and physical spaces , so as to reproduce the sign reversal in the influence of neighboring points . The color sensation at a point , defined from a steady state of the neural activities , is then extracted as a nonlinear percept conveyed by an assembly of neurons . It connects the cortical and perceptual levels , because we describe the search for a color match in asymmetric matching experiments as a mathematical projection on color sensations . We validate our color neural field alongside this color matching framework , by performing a multi-parameter regression to data produced by psychophysicists and ourselves . All the results show that we are able to explain the nonlinear behavior of shifts observed along one or two dimensions in color space , which cannot be done using a simple linear model .
Color induction , which refers to a change in color appearance of a test stimulus under the influence of spatially neighboring stimuli in the field of view [1] , has been extensively studied in psychophysics [2] . This effect has been observed for uniform inducing surrounds [3–7] and geometrically more complex ones as well [8–15] . The geometry of spatial context , and especially the frequency of chromatic modulation , have been shown to play an important role in color induction . Many works on this subject have been devoted to the study of two induction effects in particular , known as chromatic assimilation and simultaneous contrast ( see Fig 1 ) . Chromatic assimilation is the fact that the chromatic appearance of a test stimulus changes towards the chromaticity of inducing stimuli . Conversely , simultaneous contrast corresponds to the test chromatic appearance changing away from the chromaticity of inducing stimuli . Contrast is interesting in that it involves the notion of color opponency: the change is often made towards an opponent or complementary color [5 , 7 , 16] . These increased perceptual similarity or dissimilarity can be viewed as the results of attractive or repulsive effects respectively in color space . While assimilation and contrast had previously been thought to occur separately , [15 , 17 , 18] suggested that they act simultaneously in a synergistic manner . The idea that effects induced by context result from a balance between assimilation and contrast was also proposed in cognition [19] . The experimental settings used by [18] to demonstrate the synergy relied on asymmetric color matching . In psychophysics , this is a classical procedure to objectively measure the amount of color induction caused by spatial context . A human observer views two still color images side by side , a test image Jtest whose pattern influences the perception of a test color ctest , and a comparison image Jcomp[ccomp] with a modifiable comparison color ccomp ( see Fig 2 ) . In most experiments , the two images have the same geometric patterns and are composed of elementary shapes , such as rectangular , round or concentric patches , uniformly filled with different colors [3 , 7 , 20] . The patches to be compared are filled with ctest and ccomp . The observer is asked to change ccomp until color appearance between the test and comparison patches are the same , leading to a perceptual match . The perceptual shift is then the difference between the final color cmatch and the test color ctest . In Fig 3 , we illustrate this experiment with simple square patterns . In [18] , color shifts were measured for patterns with concentric annuli as in Fig 1 ( Down ) , whose colors were distinguished by the stimulation of S cones only . The matching surround was a neutral gray . Stimuli were expressed in a variant of the cone-based chromaticity space proposed by [22] and specified by three coordinates s , l , Y , where s and l are defined as the ratios S L + M and L L + M respectively , and Y stands for luminance . The definition of the S , M , L signals is recalled below . Their results showed that the largest color shifts in s chromaticity were induced by patterns alternating between two distinct colors , such as purple and lime . In particular , shifts induced by uniform backgrounds or patterns alternating between white and purple or lime were smaller . Such large shifts could not be induced by optical factors ( spread light or chromatic aberration ) only , but implied some neural processing of the stimuli [23] . To explain this , [18] suggested that stest was shifted towards the adjacent ring thanks to assimilation , while it was also repelled away from the second ring by contrast , resulting in the matching value smatch . They proposed a S+/S- center-surround receptive field model to predict the shifts: at a point x in the test ring , shift at x ≔ s m a t c h - s t e s t = DOG * ( J t e s t - J c o m p [ s t e s t ] ) ( x ) , where Jtest and Jcomp , in s coordinates , are convolved with a Gaussian kernel beforehand to account for retinal blurring; the Difference of Gaussians DOG stands for the receptive field . Color appearance would then be the combined contributions of chromaticity at the point of interest and influence of the surround filtered by the S+/S- receptive field . After tuning the parameters , this simple linear model was capable of explaining data in the specific experimental setting of [18] , where the chromaticity s of the test ring was fixed while that of the surround was changed . However , it was unable to explain the dependency of shifts upon stest reported in a later paper [17] where , conversely , stest was varied while the surround was left unchanged . Indeed , in the above equality , the central ring of the difference image Jtest − Jcomp[stest] has chromaticity s = 0 , after cancellation of the test and comparison chromaticities . The model is also conceptually difficult to justify , because it treats the central ring and the surround as fundamentally different spatial components , while handling single chromaticities and spatial integrations at the same conceptual level . This prevents their computational framework from being extended to various patterns and other matching experiments . Here , we build a framework dedicated to general color matching experiments . It is able , in particular , to explain the nonlinear behavior of color shifts found by [17 , 18] . As a starting point , we reformulate their fundamental observation into a Principle of Synergy which relies on the notion of color opponency introduced by Hering [16]: This viewpoint implies an appropriate change of vocabulary: in the sequel , chromatic assimilation and simultaneous contrast refer to local interactions , which may act at the same time but at two different local scales . The global effect observed is then the integration of infinitesimal influences induced by spatially neighboring points . Perception can in particular result in attraction ( assimilation wins over contrast ) , repulsion ( contrast wins over assimilation ) , or none of them . Thus assimilation and contrast , which seem to be contradictory effects , can be described as concomitant local phenomena . Beyond giving a merely computational model , we aim at designing a framework consistent with the physiological and anatomical observations currently available . Light entering the eye stimulates L , M and S cones of the retina proportionally to quantal absorption rate [24] , in a “wavelength-blind” fashion according to the principle of univariance [25] . At a point x of the retina R , the stimulus Lx of L cones can be approximated at first order as [26 , 27] L x ≔ ∫ λ ∈ Λ C x ( λ ) S L x ( λ ) d λ = ⟨ C x , S L x ⟩ L 2 ( Λ ) , ( 1 ) where C is the spectral distribution of the light over the visible spectrum Λ , and the spectral sensitivity S L x of L cones located at x depends on their local density . L , M , S signals are then relayed by the Lateral Geniculate Nucleus ( LGN ) and transmitted to the primary visual cortex ( V1 ) through axonal projections . It has been established that chromatic input to the visual cortex from the LGN is encoded in an opponent fashion as Hering postulated [16] and as was later confirmed by neurophysiologists [28–30] . This justifies the use of a color-opponent framework here . LGN cells have been found to respond to linear combinations of L , M , S stimuli [23 , 30] . L − M and S − ( L + M ) signals are transmitted to single-opponent cells in layers ( 4Cβ ) and ( 2/3 and 4A ) of the visual cortex respectively [2 , 23] . In contrast to them , cells clustered inside or around Cytochrome Oxydase ( CO ) blobs in layer 2/3 of V1 are sensitive to a continuum of colors instead of three cardinal axes , nonlinearly with respect to cones [23 , 31 , 32] . As such , they may have a prominent function in encoding color in the cortex , as proposed in [33 , 34] . Most of them were found to be double-opponent cells [2 , 35 , 36] . Single- and double-opponent cells are very likely to play a fundamental role in color processing in the brain [2 , 23] . The former are important for analyzing color in large areas while the latter are sensitive to edges and orientation , implying that color assimilation would be due to single-opponent cells and color contrast to double-opponent cells [2] . The visual cortex has the specificity to be organised into hypercolumns , i . e . , groups of neurons sharing the same receptive field and coding for a particular physical quantity at this position , such as orientation , spatial frequency , and temporal frequency [32 , 37–39] . These signals are mapped from the retina to V1 following an approximately logarithmic retinotopy [40 , 41] . Unlike in the case of orientation , for which the existence of such hypercolumns in V1 is now well established [32] , the anatomical and physiological bases for a functional architecture encoding color are still debated . However , in light of the promising findings made by [31] , and as discussed in [23 , 32 , 42] , it is reasonable to assume in our work a hypercolumnar organisation of cells tuned to a continuum of colors , having double-opponent characteristics , and related to CO blobs in layer 2/3 of V1 . Our work also supposes the presence of long-range lateral connections between hypercolumns , in agreement with observations of [43] where horizontal connections tend to link blobs to blobs . Note that we do not use further assumptions about the anatomical organization of color-tuned cells with respect to blobs ( inside , around , independent ) , which is still unclear [2 , 23] . In this context , our model relies on a neural field [44–47] . It is worth noting that neural fields have been previously applied to simpler examples of sensory processing in visual cortex , in order to study the spontaneous formation of population tuning curves . Orientation tuning has been addressed in an important paper by Ben-Yishai and colleagues [48] . Their model of a single cortical hypercolumn did not take into account the spatial relations between these hypercolumns , which was done in the 2001 landmark paper by Bressloff and colleagues [49] . The problem was then revisited by Bressloff [50–52] , and later by two of the authors of the present paper [53] . Spatial frequency tuning has been addressed by Bressloff and Cowan [54 , 55] , and by Chossat and Faugeras [56] . The combination of orientation tuning in binocular vision giving rise to rivalry waves has been studied in [57] . These simpler models have the benefit of explicit knowledge regarding feature preference maps ( image orientations , image textures ) and connectivity in V1 ( orientation hypercolumns and their relations with CO blobs ) . This color neural field framework allows us to study color-space interactions . In recent years , the interactions between color , and orientation/form/space , have received increasing attention [2] . Double-opponent cells may strongly contribute to the relationship between color and form processing , since the shape of their receptive fields determines their orientation tuning [23] . This supports the hypothesis that functional architectures for color and orientation would be intermingled and realize color-form interactions . A framework was proposed by [42] for modeling color and orientation processing in V1 . They assume that two populations of neural masses , one color-insensitive but orientation-tuned , and the other sensitive to both , interact through an extended version of the ring model [48] . Our work is related but complementary to theirs , because rather than considering only two hypercolumns of each kind and using uniform inputs , without introducing space , we study the interactions between color and space by the means of multiple color hypercolumns and patterned inputs , without introducing orientation . The relation between orientation and color falls outside the scope of this work , and should be examined in the future . The goal of our work is therefore to provide a neural field model unifying assimilation and contrast inside a color-opponent framework , consistent with psychophysical data , and compatible with the physiology of V1 . We have successfully achieved this aim .
A . S . conceived and designed the experiments herself and consented to participate . The rigorous definition of color is thoroughly explained in S1 Appendix , where we also present the most common representations of the color space . Here , for the sake of simplicity , we only briefly state the minimal definitions and properties to be used in the model . Color is mathematically defined as an equivalence class of metameric lights [26 , 58–60] . Two physical lights of spectral distributions C 1 , C 2 ∈ L 2 ( Λ ) are metameric if they produce exactly the same visual effect under the same viewing conditions , and this identification strongly depends on the observer . In our framework , metamerism can be expressed as the equality of the triplets of scalar products characterizing C 1 and C 2 ( L , M , S ) = ( ⟨ C i , S L ⟩ L 2 ( Λ ) , ⟨ C i , S M ⟩ L 2 ( Λ ) , ⟨ C i , S S ⟩ L 2 ( Λ ) ) , with S L the spectral sensitivity of L cones ( and likewise for M and S cones ) , see ( 1 ) . We dropped the exponent x ∈ R 2 by considering for simplicity that cone density is constant across the retina and that light is spatially uniform ( although we can define metamerism with respect to any x ) . Color is hence naturally identified to a three-dimensional vector for trichromats , being specified by a triplet of cone stimuli ( L , M , S ) . For a light of spectral distribution C , its color is denoted [ C ] . Color space , denoted C , is then defined as the subset of physically realizable colors which are visible to the eye . Since the cone signals ( L , M , S ) are non-negative and cannot reach all possible positive values [60] , C can be identified to a bounded subset of ( R + ) 3 through a choice ϕ L M S : C → R 3 of coordinates . In this work , we suppose that an appropriate choice of coordinate system ϕ o p p : C → R 3 leads to an opponent representation of the color space C o p p ≔ ϕ o p p ( C ) ( 2 ) satisfying the following important properties . First , it is a bounded and convex subset of R 3 which enjoys symmetry: if c ∈ C o p p , then - c ∈ C o p p . Second , the symmetry operation c ↦ −c must pair any color to its opponent one , in the sense of Hering [16] . Hence , color regions of C o p p come into opposed pairs , for example Yellow and Blue or Red and Green regions , or likewise . Third , C o p p must contain the neutral or zero color 0 , opponent to itself , which would correspond to some neutral gray with no hue ( for a fourth condition , see details in S1 Appendix ) . Following up on this , we consider in this work an opponent representation in the style of Hering’s theory . Indeed , in view of the physiological results exposed in the Introduction , it is now accepted that L , M , S signals are recombined in area V1 into Yellow-Blue , Red-Green and Achromatic independent channels [61] , although the right choice of the opposite axes has been debated [62] . Our model does not depend in a decisive manner upon the choice of a specific color opponent space . In fact our particular choice of Hering’s coordinates is not very different from the cone opponency coordinates defined in [30] . Also , we do not require that the opponent axes point towards perceptually unique hues , unlike in the original theory of Hering . Supposing such a simple relationship between physiology of the cells and psychology related to hue pureness was indeed criticized [23 , 30 , 62] . Here , we rely on the ( l , s , Y ) and ( H , S , L ) representations , and restrict ourselves to a lower-dimensional subspace of the original color opponent space , also denoted C o p p for convenience ( in this context , the letters ‘S’ and ‘L’ of HSL stand for Saturation and Luminance respectively , not Short or Long cones , while ‘H’ stands for Hue ) . In the case of the ( l , s , Y ) representation , we apply our model to the one-dimensional subspace based on the chromaticity s and defined by c ≔ s − 1 . The ( l , s , Y ) representation used in [15 , 17 , 18] , a variant of the one proposed by [22] , is defined as { s = S L + M l = L L + M Y = L + M + S . We then define C o p p to be the one-dimensional color subspace based on the change of coordinates c ≔ s - 1 ∈ C o p p ≔ [ - 2 , 2 ] , where the number 2 is arbitrary , but covers the typical range of c values used in experiments ( purple , lime and white correspond to ( l , s , Y ) = ( 0 . 66 , 2 . 0 , 15cd/m2 ) , ( 0 . 66 , 0 . 16 , 15cd/m2 ) and ( 0 . 66 , 0 . 98 , 15cd/m2 ) respectively . ) . The Hue , Saturation and Luminance or ( H , S , L ) representation ( note that the letters ‘S’ and ‘L’ are not referring to Short or Long cones ) , often used in computer graphics , maps the sRGB unit cube or gamut of a device to a cylinder whose central axis is achromatic and perpendicular to a chromatic disk [58 , 63] . Standard formulas provide the change of coordinates T s R G B → H S L [21] . We use the two-dimensional chromatic disk C o p p , defined as the intersection of the constant luminance plane L = 1/2 with the HSL cylinder , and identified to the unit disk , so that ( c 1 , c 2 ) ≔ ( S cos ( H ) , S sin ( H ) ) ∈ C o p p . The gamut corresponds in fact to a subject- and device-dependent subspace C d e v strictly smaller than the subspace of chromatic colors visible by the observer , since they are not all reproducible by a screen . The gamut of standard devices however covers a large part of visible colors , hence justifying its use , and the HSL representation has already proven its efficiency in computer graphics . We claim that the specific details of the display , such as gamut and screen characteristics , do not play a major role in our methods and results , provided that all experiments are consistently made in the same conditions . The main thrust of our model is twofold: first , we put forward an evolution equation for the dynamics of neural activities , considered as a spatial and color neural field [44–47] . Second , we propose a theoretical framework for color matching experiments in the context of color perception , and introduce a formal definition of color sensation . The Color Neural Field Eq ( 3 ) describing ( at least , theoretically ) how the visual cortex reacts to a color image , we now link our model to psychophysical data . This subsection introduces the central notion produced by our model , i . e . , that of color sensation . Basically , it corresponds to some feeling produced in the brain when observing a still image . We define it to be a steady state of the neural field dynamics of Eq ( 3 ) , then restricted to the hypercolumn in correspondance to the test point . This concept allows us to propose a mathematical description of color matching experiments ( see Introduction ) , where matching is considered as the projection of the test color sensation onto a family of color sensations elicited by comparison images . Such a “matching as a projection” framework allows the model to predict color shifts . It could be generally applied to other dynamics than Eq ( 3 ) , as well as other definitions of sensation relatively to these dynamics .
In Fig 6 , we set W ( ⋅ , ⋅ ) ≔ ω ( r0 , c0 , ⋅ , ⋅ ) , and display it for two values of ( r0 , c0 ) . The two configurations of Table 2 ( excitatory or inhibitory connections ) can be clearly seen in the figure . In fact , most of the neural masses have negligible influence on ( r0 , c0 ) . This occurs when c′ is neither close to c0 nor to −c0 , or when r′ is too far from r0 ( not shown on the figure , because the outer variance of the DOG g has a great value compared to the extent of Ω ) . The properties of the connectivity values are analogous to double-opponent cells’ center-surround behavior . In Fig 7 , we show the cortical input H ( r , c ) = h ( c − I ( r ) ) ) , where the purple/lime patterned image I is as in Fig B in S2 Appendix and corresponds to the cortical counterpart of one typical test pattern of [18] . We also show the color sensations a∞ after convergence . The input H is obtained by “lifting” the image I inside Ω × C o p p . Altitudes of maximal values hence alternate between lime ( c ≃ −1 ) , purple ( c ≃ 1 ) , and white ( c ≃ 0 ) . The shape of H heavily determines that of the final activities a∞ , since it has the role of cortical input . It can be noticed that a∞ reaches values lower than 1/2 in the bottom part of the heatmap . In order to emulate a color matching experiment , the first building block of our algorithm is to simulate the Color Neural Field dynamics in Eq ( 3 ) , as shown in Fig 8 , and accordingly to Algo 1 in S2 Appendix ( with dt = 1 ) . Once again , the cortical image is given by Fig B in S2 Appendix . Color matching can then be emulated by applying Algo 2 in S2 Appendix . The parameters have to be regressed to the experimental data in order to reproduce the color shifts . Fig 9 shows that our model is able to explain the shifts measured for the observers called ‘MC’ and ‘AZ’ in [18] , by using the regressed values qMC = ( 0 . 60 , 0 . 69 , 0 . 30 , 0 . 40 , 4 . 42 , 1 . 82 , 0 . 58 , 8 . 35 , 0 . 47 , 0 . 30 , 1 . 80 ) and qAZ = ( 0 . 60 , 0 . 69 , 0 . 31 , 0 . 40 , 4 . 42 , 1 . 81 , 0 . 60 , 8 . 35 , 0 . 47 , 0 . 30 , 1 . 80 ) , respectively . The slight difference observed between the parameter values could partly account for subject differences . We compare in Fig 10 ( left ) the color sensations a q t e s t , a q c o m p [ c m a t c h ] , and a q c o m p [ c q p r e d ] ( refer to S2 Appendix for the notations ) , taken at the point of interest r0 = ( 0 , 0 ) , and with q = qMC . They are generated by the purple/lime test image ( as before ) , the comparison image filled with the experimental value cmatch , or with the predicted matching value c q p r e d , respectively . After regression , c q p r e d becomes close to the experimental value cmatch , so that the two corresponding curves nearly coincide . Among all curves { a q c o m p [ c ] } c , a q c o m p [ c m a t c h ] should be the nearest one to atest with respect to the L ∞ norm , and we illustrate this in Fig 11 . The qualitative difference between the test and comparison curves mainly comes from the difference of complexity of their respective inputs: the test image has a complicated pattern , while the comparison images are simpler . We confront our model to the nonlinear shifts observed by [17] in Fig 12 . We find that it is able to reproduce the data after regression , with the fitting parameter value qnonlin = ( 0 . 42 , 0 . 71 , 0 . 63 , 1 . 16 , 4 . 43 , 1 . 72 , 0 . 56 , 6 . 35 , 0 . 47 , 0 . 30 , 1 . 80 ) . This non-trivial result , alongside the results in Fig 9 , provides a strong justification to our framework . In Fig 13 , instead of qnonlin , we used the value qAZ that explained the data on Fig 9 ( right ) observed by [18] . Let us recall that in the experimental settings of [18] and Fig 9 , stest was fixed while the surround varied , unlike in the settings of [17] and Fig 12 , where conversely for some fixed test patterns stest was changed . The predictions in Fig 13 are hence not close to the ground truth data , especially at the endpoints where the orange and blue curves are crossing . Shifts are also smaller in magnitude . However , it is remarkable that the model predicts a similar trend with respect to stest . Crossings are also quite expected at stest of magnitude great enough , for which a reversing of the shift direction is plausible . For too great magnitudes , the shifts are likely to become negligible . We obtain similar results by using qMC . Finally , as a further important validation , we show in Fig 14 that our framework is also efficient in explaining the vector field of shifts in the chromatic disk ( Fig 5 ) . The convergence towards the opposite blue is made obvious , with the regressed parameter value qHSL = ( 0 . 73 , 0 . 15 , 0 . 52 , 0 . 68 , 4 . 41 , 1 . 84 , 0 . 51 , 8 . 35 , 0 . 47 , 0 . 30 , 1 . 80 ) . As a remark , the sampling resolution of the 2D color space is quite low for computational reasons , so that the meaningfulness of parameter values has to be carefully considered . We also obtain a smoother result than in the experimental data , as a SoftMin method is used to search for optimally matching colors ( see associated code ) .
The present work has to be distinguished from those dealing with color constancy problems . Color constancy is the ability of humans to guess the reflectance of objects despite very different illumination conditions [69] . Reflectance is a property characterizing matter , defined as the proportion of luminous energy reflected by its surface . In Eq ( 1 ) , it is linked to the spectral power distribution P of the incident illuminant ( daylight , lamp ) through the relation C x ( λ ) ≔ P x ( λ ) R x ( λ ) , where reflectance and illuminant power are taken at the point of the scene which sends light to x in the retina . The phenomenon of color constancy , first studied by E . Land who proposed the Retinex algorithm [70 , 71] , has since been the subject of much research . Given cone inputs ( L , M , S ) , how does the brain retrieve the spectral reflectance R of an object ? By contrast , we are not interested in how color is linked to reflectance , but rather in how identical stimulation of the cones can lead to different sensations because of spatial context . In view of the previous discussion , to the naive question “What is the color of this object ? ” , we see that at least three types of answers are theoretically possible . First , identify the colored material likely to produce such a visual effect , i . e . , guess its reflectance R; second , report the color [ C ] produced by cone stimulation; and third , describe the color sensation , as introduced in this work . In practice , none of these tasks is trivial , even regarding color naming , since thought and language related to color involve complex cognition [72] . Summarizing the previous ideas , the three punctual relationships R 1 x 1 = R 2 x 2 [ C 1 x 1 ] = [ C 2 x 2 ] S 1 x 1 = S 2 x 2 are independent , because there are situations where any of them can hold or not . The reader can be convinced by simple examples . For instance , the same object seen in daylight or under shadow has different colors because [ C d a y l i g h t ] ≠ [ C s h a d o w ] , but we can easily recognize it and guess that R = R daylight = R shadow . On the reverse , two objects having the same color can be guessed as being made of different materials , if their surroundings are not the same , that is , [ C 1 ] = [ C 2 ] but R 1 ≠ R 2 . Another example is given by color induction , as largely discussed here . Because of the surrounding context , two colors [ C 1 ] = [ C 2 ] deriving from identical L , M , S cones stimulation can be perceived as different color sensations S 1 ≠ S 2 . The reverse situation can occur , where different colors are perceived as similar with different contexts . However , if the whole retinal space is taken into account , then the sets { R x } x ∈ R ( 10 ) { [ C x ] } x ∈ R ( 11 ) { S x } x ∈ R ( 12 ) are linked to one another . While papers on color constancy are devoted to the link between Eqs ( 10 ) and ( 11 ) [69 , 73] , ours focuses on the relationship between Eqs ( 11 ) and ( 12 ) . Our framework supposes that { S x } x ∈ R depends on the sole knowledge of { [ C x ] } x ∈ R . Our model should not be considered as an image processing algorithm . First , our work involves a psychophysically and physiologically relevant neural structure , while image processing models often lie at the conceptual level of an image . In the literature , many algorithms are designed to make the image perceptually better ( contrast enhancement , histogram equalization , etc . ) and allow for image compression , in accordance with various perceptual criteria , an early work being [74] for example . Some of them are computational models simulating perception , and inspired by neural mechanisms underlying vision , as in [75] , and more recently [76] . Taking an image as input , their algorithms produce another image as output , which represents “the” perceived color image , in order to reproduce color induction effects . In fact , the simple convolutional receptive-field model of [18] is already a basic version in this family of algorithms . In the more sophisticated approach of [75] , the image evolves according to a Wilson-Cowan dynamic , which is akin to descending the gradient of some energy , leading to histogram equalization . M . Bertalmío also proposes a method to reproduce the lightness matching data of [77] , in particular some assimilation or contrast effects ( in the usual global sense , not at the local scale ) . His attempt to match psychophysical data based on neural-like dynamics can be considered similar to ours . However , shifts are estimated through a direct algebraic computation of the values found at different locations of the disks after convergence . This prevents any straightforward generalization of the method to spatially more complicated patterns , an impediment which we already mentioned in the Introduction for the model of [18] . More importantly , a major conceptual problem in this image-based approach is that , it seems inappropriate to simulate perception by producing an output image of the same nature as the original one , that represents “the” perceived color image . Indeed , the original and final images do look different: the final image hardly represents our perception of the original one , since it is also processed by the brain . This ambiguity should be clearly discussed when dealing with the output image . For instance , [74] proposes to map color images into an opponent perceptual space , where they are processed and compared through a perceptual metric , that has a role clearly different from the usual RGB color space of images . It is the fate of any model to meet its own limitations . Below , we depict non-exhaustively some questionable features or drawbacks encountered by ours . Just as for neural field models modeling orientation vision , we can study bifurcations of the solutions of Eq ( 3 ) around stationary states [49 , 53 , 56 , 82] . Under some hypotheses of symmetry and periodicity , we can predict , using equivariant bifurcation theory , the emergence of visual patterns or “planforms” . In the same fashion as [49] who explained orientation-based geometric hallucinations , a color neural field model can predict patterned color hallucinations . Future psychophysical experiments , may confirm this and support the relevance of this kind of model for color vision . Our work addresses the question of color-space interactions , by providing a color neural field model alongside a general framework to account for matching experiments . We propose to consider color matching as a mathematical projection , in agreement with the principles of psychophysics , where subjective notions are assessed by means of objective procedures . Our neural field unifies assimilation and contrast at the cortical level , and relies on the idea of color opponency . The notion of color sensation that we introduce bridges the gap between these cortical and perceptual levels , and is a nonlinear percept involving a whole distribution of neurons . This framework allows the study of psychophysical phenomena such as color induction , by taking advantage of a classical computational neuroscience tool . To our knowledge , this is the first color neural field model consistent with psychophysical data and compatible with physiological findings . The assumption that V1 is organized into a structure similar to color hypercolumns has still to be experimentally proved though . We believe that the proposed framework could possibly be adapted to other perceptual situations , such as hearing or touch . | The color perception produced by an image heavily depends on the spatial distribution of its colors . From this “color in context” phenomenon , extensively studied in psychophysics for decades , has arisen the question in neuroscience of how color and space interact in the brain . Visual signals are indeed processed in such a way that neighboring pixels make the perception at some point different from its real color , inducing a color shift . In this work , we propose to emulate perception in context by modeling the activity of color sensitive neurons with a neural field . Our framework unifies two antagonistic effects , assimilation and contrast , which have been suggested to occur simultaneously but at different scales . We use the notion of color opponency inspired by the work of Hering , so as to express these effects as a combination of attraction and repulsion in physical and color spaces . We introduce the concept of “color sensation” , and show how to rigorously link the neural field model to perceptual shifts , by considering color matching as a mathematical projection on color sensations . The results show that our model is able to reproduce some nontrivial behaviors of the color shifts observed in experiments . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
]
| [
"medicine",
"and",
"health",
"sciences",
"ocular",
"anatomy",
"brain",
"social",
"sciences",
"light",
"neuroscience",
"electromagnetic",
"radiation",
"perception",
"cognitive",
"psychology",
"luminance",
"visible",
"light",
"vision",
"animal",
"cells",
"visual",
"cortex",
"physics",
"psychology",
"cellular",
"neuroscience",
"retina",
"cell",
"biology",
"anatomy",
"color",
"vision",
"neurons",
"psychophysics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"sensory",
"perception",
"physical",
"sciences",
"ocular",
"system",
"cognitive",
"science"
]
| 2019 | A neural field model for color perception unifying assimilation and contrast |
Crossovers mediate the accurate segregation of homologous chromosomes during meiosis . The widely conserved pch2 gene of Drosophila melanogaster is required for a pachytene checkpoint that delays prophase progression when genes necessary for DSB repair and crossover formation are defective . However , the underlying process that the pachytene checkpoint is monitoring remains unclear . Here we have investigated the relationship between chromosome structure and the pachytene checkpoint and show that disruptions in chromosome axis formation , caused by mutations in axis components or chromosome rearrangements , trigger a pch2-dependent delay . Accordingly , the global increase in crossovers caused by chromosome rearrangements , known as the “interchromosomal effect of crossing over , ” is also dependent on pch2 . Checkpoint-mediated effects require the histone deacetylase Sir2 , revealing a conserved functional connection between PCH2 and Sir2 in monitoring meiotic events from Saccharomyces cerevisiae to a metazoan . These findings suggest a model in which the pachytene checkpoint monitors the structure of chromosome axes and may function to promote an optimal number of crossovers .
Meiotic recombination occurs during prophase I when homologous chromosomes are synapsed along their entire length . Synapsis is defined as the close and stable association of homologous chromosomes through a proteinaceous structure called the synaptonemal complex ( SC ) . In most organisms , this complex is composed of two main parts: lateral elements that attach along the axis of each homologous chromosome and transverse elements that span the central region of the SC and function to tether the homologs [1] , [2] . At the leptotene/zygotene stages of meiotic prophase , these structural proteins begin to load onto the chromosome axes , and are completely assembled at pachytene , when homologous chromosomes are synapsed along their entire length . Recombination between the homologous chromosomes initiates with DNA double-strand breaks ( DSBs ) that are repaired as either crossovers or noncrossovers [3]–[5] . Crossovers establish chromatin linkages called chiasmata , which , along with sister chromatid cohesion , hold homologs together after recombination has been completed and chromosomes have dissociated their SC proteins . Chiasmata help orient the homologous chromosomes on the metaphase I spindle and ensure their proper segregation at anaphase I . The failure to establish a crossover/chiasma can result in the nondisjunction of homologs and lead to aneuploid gametes . Crossover formation is a tightly regulated process . Mutational analysis has revealed evidence for several mechanisms that control the frequency and position of crossovers along the chromosome arms [6]–[9] . For example , in Drosophila melanogaster , the precondition class of mutants exhibit reduced crossover levels with an altered distribution pattern , suggesting these genes have a role in establishing the number and distribution of crossover sites [10] . Changes in chromosome structure can also affect crossover distribution . Heterozygous inversions suppress crossing over near the breakpoints , yet enhance the frequency of exchange on the remaining chromosome pairs , a phenomenon referred to as the “interchromosomal effect” [11] . Crossing over may also be regulated by surveillance mechanisms that coordinate the sequence of critical events throughout prophase . In Drosophila , the process of repairing meiotic DSBs is monitored by at least two checkpoints: the canonical DSB repair checkpoint that responds to DNA damage [12] , [13] and another that monitors DSB-independent events leading to crossover formation , hereafter referred to as the “pachytene checkpoint” [14] . The pachytene checkpoint induces a delay in response to defects in DSB repair genes required to repair all meiotic DSBs and genes encoding an endonuclease complex required for crossover formation ( exchange class ) . Pachytene checkpoint activity requires a group of MCM-related genes that promote crossover formation ( precondition class ) and the Drosophila homolog of the widely conserved AAA+ ATPase PCH2 . In Saccharomyces cerevisiae and Caenorhabditis elegans , pachytene checkpoint activity has been detected in mutants with disrupted SC formation [15] , [16]; however , it remains unclear what the underlying process is that the pachytene checkpoint is monitoring . For instance , yeast carrying a non-null zip1 allele appear to form SC normally , yet still exhibit a Pch2-dependent delay [17] . Mutations that impair SC initiation in C . elegans triggers a Pch2-dependent response [16] , although it is unclear whether the defect being monitored is synapsis per se , a prerequisite to synapsis such as homolog pairing and/or DSB repair . Some mutations that exhibit pch2-dependent delays in Drosophila have no obvious defects in SC formation and abolishing synapsis does not elicit any delay phenotypes [14] . Therefore , at least in Drosophila and possibly in these other organisms , it may not be the SC that is being monitored by the pachytene checkpoint . Instead , the pachytene checkpoint may be important to monitor synapsis-independent changes in chromosome structure required for crossover formation [14] . We have investigated the relationship between chromosome structure and the pachytene checkpoint and show that disruptions in chromosome axis components cause pch2-dependent delays . Unlike the delays observed in DSB repair mutants , these delays occur independently of MCM-related genes . Heterozygous chromosome aberrations also result in a MCM-independent pachytene delay and interchromosomal increase in crossovers that require pch2 . These findings suggest a model in which the pachytene checkpoint monitors two genetically distinct events: an early function of DSB repair proteins and the structure of chromosome axes . A checkpoint response to both events requires the histone deacetlyase Sir2 , showing that a functional connection between PCH2 and Sir2 in monitoring meiotic events is conserved in Saccharomyces cerevisiae and Drosophila . Checkpoint activity is also associated with prolonged PCH2 expression . We propose the pachytene checkpoint may function to promote an optimal number of crossovers by regulating the timing of a crossover determination phase defined by PCH2 expression .
In the Drosophila germarium , oocytes are born within cysts composed of 16 cells that are connected by ring canals . Two out of the sixteen cells , each with four ring canals , initially contain equivalent levels of SC proteins and are termed the pro-oocytes ( Figure 1A ) . As the developing cysts travel from the anterior ( region 2 ) toward the posterior ( region 3 ) of the germarium , the pro-oocytes proceed through the pachytene stage of meiosis where synapsis is completed and DSB formation and recombination occurs . By region 3 of the wild-type germarium , DSB repair is completed and one of the two pro-oocytes will exit meiosis , lose its SC and become a nurse cell while the other will continue through development and form the oocyte ( Figure 1A ) [18] . In DSB repair and exchange-defective mutants , the transition through pachytene is delayed by pachytene checkpoint activity [14] . This results in both pro-oocytes persisting into region 3 cysts , referred to as the “two-oocyte phenotype . ” Delays can be identified either by the persistence of the SC transverse filament C ( 3 ) G in both pro-oocytes [14] or by the concentration of ORB protein in the cytoplasm of two cells rather than one in region 3 cysts ( Figure 1B ) [19] . ORB staining , however , is less sensitive than C ( 3 ) G at detecting pachytene delays , resulting in a different frequency of the two-oocyte phenotype between the two assays [14] . Abolishing synapsis by mutation of c ( 3 ) G does not elicit the two-oocyte phenotype ( Figure 1D ) , suggesting the pachytene checkpoint is not monitoring SC formation [14] . We further investigated the relationship between chromosome structure components and the pachytene checkpoint by determining the effects of mutations in two other genes , ord and c ( 2 ) M , which encode structural proteins . ORD localizes to chromosome axes in oocytes independent of synapsis ( i . e . in c ( 3 ) G mutants ) and has roles in meiotic recombination and sister chromatid cohesion [20] , [21] . Although ord mutants initially display normal C ( 3 ) G and C ( 2 ) M localization , only rare structures resembling SC were observed by electron microscopy ( EM ) , suggesting that the ultrastructure of chromosome axes was disorganized [21] . Consistent with this interpretation , C ( 3 ) G and C ( 2 ) M staining precociously deteriorates in ord mutants as the oocytes progress through pachytene [21] . We found that ord mutants displayed a high frequency of the two-oocyte phenotype ( Figure 1D ) , indicative of a delay in meiotic progression . The two-oocyte phenotype was suppressed in ord; pch2 double mutants , indicating the delay was dependent on the pachytene checkpoint ( Figure 1D ) and supporting the hypothesis that the pachytene checkpoint is sensitive to defects in axis components . C ( 2 ) M is a component of the SC lateral element and localizes adjacent to the chromosome axes even in the absence of synapsis ( in c ( 3 ) G mutants ) , suggesting it may interact with axis components [22] , [23] . In c ( 2 ) M mutant oocytes , C ( 3 ) G protein fails to develop into complete strands along the lengths of each chromosome , but instead appears as small patches ( Figure 1C ) . The most likely explanation is that SC initiates in c ( 2 ) M mutants but polymerization is defective . Similar to ord mutants , c ( 2 ) M mutants exhibited a high frequency of the two-oocyte phenotype , which was suppressed in c ( 2 ) M; pch2 double mutants ( Figure 1D ) . The high frequency of the two-oocyte phenotype observed in c ( 2 ) M mutants was not suppressed by mutation of c ( 3 ) G ( Figure 1D ) , demonstrating the pachytene checkpoint can signal independently of SC initiation . Together , these results suggest the pachytene checkpoint may monitor a synapsis-independent function of ORD and C ( 2 ) M , such as the formation of chromosome axes . If the pachytene checkpoint monitors the integrity of chromosome axes we reasoned that structural rearrangements would also exhibit pachytene delays . Balancers are multiply-inverted chromosomes that fail to cross over with a normal homolog , presumably due to a disruption in the continuity of pairing and/or synapsis [24]–[26] . We characterized the integrity of SC-associated proteins in balancer heterozygotes with antibodies recognizing the SC components C ( 3 ) G and C ( 2 ) M . Single balancer heterozygotes ( TM3/+ ) had thread-like C ( 3 ) G and C ( 2 ) M staining that was indistinguishable from wild-type ( Figure 2A ) [26] . Double balancer heterozygotes ( CyO/+; TM3/+ ) also initially displayed normal C ( 3 ) G and C ( 2 ) M localization , but the staining became fragmented and sometimes undetectable in region 3 oocytes ( Figure 2A ) . This precocious deterioration of SC proteins during pachytene is similar to what is observed in ord mutant oocytes [21] , suggesting that rearrangement breakpoints might disrupt axis stability . Using C ( 3 ) G staining to detect oocytes , we found that FM7 , Bwinscy , TM2 and TM3 balancer heterozygotes each exhibited a significantly higher frequency of the two-oocyte phenotype compared to wild-type ( Figure 2B ) , suggestive of a pachytene delay . The high frequency of the two-oocyte phenotype was suppressed to wild-type levels in FM7/+; pch2 and TM3/+; pch2 females , confirming the delays were dependent on the pachytene checkpoint ( Figure 2B; P<0 . 05 compared to either balancer heterozygote alone ) . pch2 had no effect on the SC morphology in single balancer heterozygotes ( data not shown ) . Each balancer chromosome contains several inversions . For example , the TM3 chromosome includes 10 breakpoints [27] . To investigate the effects of a more subtle disruption in chromosome organization on the pachytene checkpoint , we tested whether a single aberration , or two breakpoints , would be enough to induce pachytene delays . Remarkably , heterozygotes of single translocations between the 2nd and 3rd chromosomes ( T ( 2;3 ) DP77/+ , T ( 2;3 ) dpD/+ , and T ( 2;3 ) ltX16/+ ) and a single inversion on the X chromosome ( In ( 1 ) dl49/+ ) each exhibited a high frequency of the two-oocyte phenotype , suggesting the threshold to trigger the pachytene checkpoint is low and requires as few as two breaks in axis continuity ( Figure 2B ) . Importantly , the delays were not dependent on C ( 3 ) G and were not significantly increased in In ( 1 ) dl49 homozygotes compared to wild-type ( Figure 2B and 2C ) , indicating the pachytene checkpoint responds to a break in alignment between homologs in a way that is independent of SC initiation . In addition to the delay in oocyte selection , DSB repair and exchange-defective mutants also exhibit a pch2-dependent delay in the response to DSBs [14] . To monitor DSB formation and repair in balancer heterozygotes , we stained ovaries with an antibody to γ-H2AV [28] , [29] . In wild-type oocytes , γ-H2AV foci are most abundant in region 2a nuclei ( cyst 3 in Figure 2D ) and absent by region 3 ( cyst 8 in Figure 2D ) , indicating DSBs have been repaired . Likewise , both FM7/+ and CyO/+ heterozygotes exhibited maximum γ-H2AV foci in region 2a oocytes at a similar cyst age to wild-type ( Figure 2D ) . The same result was also observed in the double heterozygote FM7/+; CyO/+ . Therefore , balancer heterozygotes do not cause a delay in the γ-H2AV response to DSBs , revealing a distinction between the effect of DSB repair mutants and chromosomal rearrangements on the pachytene checkpoint . While mutations in DSB repair genes induce two pch2-dependent delays in pachytene , delayed response to DSBs and delayed oocyte selection , chromosomal rearrangements only delay the latter . If the pachytene checkpoint can cause delays through two distinct pathways , it should be possible to define them genetically . This was tested with mutations in the MCM-related precondition genes mei-218 and rec , which are required for 90% of all crossovers and the pachytene delays caused by mutations in DSB repair and exchange genes [14] . Unexpectedly , the high frequency of the two-oocyte phenotype was still observed in mei-218; TM3/+ and FM7/+; rec ( Figure 2B , P<0 . 05 compared to mei-218 and rec single mutants ) . Consistent with this finding , the pachytene delay in c ( 2 ) M mutants was not suppressed in mei-218; c ( 2 ) M double mutants ( Figure 1D ) . These results show that , unlike the DSB repair and exchange-defective mutants , defective and/or misaligned chromosome axes interact with the pachytene checkpoint independent of precondition genes and possibly at a later step ( i . e . after the DSB response ) . PCH2 is required for some of the crossovers that occur in the exchange-defective mutant , hdm [14] . Consequently , hdm; pch2 double mutants exhibit an elevated frequency of nondisjunction compared to hdm single mutants . These results suggest a functional link between the pachytene checkpoint and the production of crossovers . To determine if this is a general property of mutants that exhibit pachytene delays , we tested additional double mutants with pch2 . Exchange class genes Ercc1 and mei-9 encode components of an endonuclease complex of proteins that includes HDM and is required for normal levels of meiotic crossing over [30] , [31] . Loss of ERCC1 function results in 14% X-chromosome nondisjunction , which is elevated to 30% in a pch2 mutant background , suggesting crossovers are further reduced ( Table S1 ) . In addition , the low level of crossovers that are generated along the 2nd chromosome in mei-9 mutants are mostly suppressed in mei-9; pch2 double mutants ( Figure 3A ) . These results suggest the residual crossovers in recombination-defective mutants depend on a mechanism facilitated by pch2 . When crossing over is suppressed along a normal chromosome heterozygous to a balancer , there is an interchromosomal effect that increases crossing over on the remaining chromosome pairs [11] . Since PCH2 is responsible for the residual level of crossovers in recombination-defective mutants , we asked if it was also responsible for the increase in crossovers observed in balancer heterozygotes . Consistent with previous work on interchromosomal effects [32] , [33] , we found that FM7/+ heterozygotes exhibit 151% of wild-type crossing over along the 2nd chromosome with an altered distribution ( Table 1 ) . Although there was little deviation from wild-type controls in the distal regions of the chromosome ( al-b ) , the genetic map distance was increased ∼4–5 times that observed in wild-type centromere-proximal intervals ( Table 1; Figure 3B ) . Remarkably , 2nd chromosome crossing over in FM7/+ heterozygotes was reduced to 106% of wild-type in a pch2 mutant background ( p<0 . 00005; Table 1; Figure 3B ) . Similarly , introduction of the CyO chromosome increased crossing over along the X chromosome to 149% of wild-type , which was reduced to 128% in pch2 mutants ( p<0 . 05; Table 2; Figure 3C ) . Interestingly , the closer the interval being tested was to the centromere , the greater the interchromosomal effect and pch2 dependence ( Table 2; Figure 3C ) . Since pch2 single mutants exhibited normal levels of crossing over on the X and 2nd chromosome ( Table 1; Table 2; Figure 3 ) , these data show that pch2 is required for most of the interchromosomal increase in crossover levels in balancer heterozygotes . The increased crossing over observed in balancer heterozygotes could be explained by pachytene checkpoint activity increasing DSB levels . However , we failed to observe any significant change in the number of γ-H2AV foci in oocytes single or doubly heterozygous for FM7 and CyO compared to wild-type ( Figure 2D ) . Since asynchrony of DSB formation can complicate measuring the total number of γ-H2AV foci , we repeated the above experiment in a spn-A ( Drosophila Rad51 ) mutant background , in which repair of DSBs is blocked [12] . The number of γ-H2AV foci in region 3 oocytes of these mutants is expected to be close to the total number of DSBs that occurred [28] , [34] . spn-A mutant region 3 oocytes displayed an average of 21 . 0 ( +/−1 . 5 ) γ-H2AV foci . Similarly , FM7/+; CyO/+; spn-A region 3 oocytes had an average of 24 . 0 ( +/−1 . 4 ) γ-H2AV foci . These results indicate that the ability of the pachytene checkpoint to increase crossing over in the genome is not mediated by a substantial increase in the total number of DSBs . Instead , pachytene checkpoint activity most likely increases the chance of DSBs becoming crossovers , particularly those that occur near centromeres . To investigate how PCH2 affects crossover frequency , we monitored the protein localization pattern during meiosis . A transgene was constructed containing a hemagluttin ( HA ) epitope at the N-terminus of the pch2 transcript under the control of an inducible UASP promoter . We expressed PCH2 using the germline specific driver P ( Gal4-nos . NGT ) 40 [35] , abbreviated as NGT , known to express in pachytene at moderate levels [36] . The NGT-driven P ( HA-pch2 ) 71 transgenic line restored the two-oocyte phenotype in FM7/+; pch2 females to similar levels found in FM7/+ heterozygotes ( Figure S1 ) , demonstrating that the NGT-driven pch2 transgene was functional . PCH2 staining formed foci that localized around the nucleus in zygotene and early pachytene ( region 2 ) cells ( Figure 4A ) . No PCH2 foci were detected in region 3 cells , suggesting the protein is rapidly degraded or no longer produced after early pachytene . Surprisingly , PCH2 foci did not overlap with the DNA stain . To determine if PCH2 foci localized within the nucleus , we counterstained with the nuclear envelope component , Lamin . We found that 73% of PCH2 foci showed a close association ( i . e . touching ) with the cytoplasmic side of the Lamin staining ( n = 368; Figure 4B ) , indicating they localized adjacent to the nuclear envelope and outside the nucleus . The remaining 27% of PCH2 foci not closely associated with Lamin were found dispersed within the cytoplasm . To determine if PCH2 localization pattern changes in mutant backgrounds that exhibit pachytene delays , we examined PCH2 expression in mutants that cause checkpoint delays: hdm , mei-9 and in TM3/+ heterozygotes . In hdm and mei-9 mutants , the number of PCH2 foci per oocyte was increased ∼2-fold compared to controls ( Figure 4C ) . In addition , the foci persisted into region 3 oocytes , which was never observed in control germaria ( Figure 4A and 4C ) . However , PCH2 localization was not detected past stage 2 of oogenesis ( data not shown ) , indicating the loss of PCH2 is only delayed in exchange-defective mutants . In TM3/+ heterozygotes , the levels of PCH2 foci in region 2 cells was unchanged compared to controls , but were present in region 3 ( Figure 4C ) , revealing a correlation between the prolonged expression of PCH2 and a delay in pachytene . The intracellular localization pattern of PCH2 did not change when pachytene was delayed since the foci remained juxtaposed to the nuclear envelope in hdm and mei-9 mutants and in TM3/+ heterozygotes at all stages ( Figure 4A and data not shown ) . Furthermore , mutation of mei-W68 , which eliminates DSB formation , showed a normal staining pattern of PCH2 , and hdm; mei-W68 double mutants showed the same PCH2 staining pattern as hdm single mutants ( Figure 4A and data not shown ) , consistent with our previous conclusion that the pachytene checkpoint functions independently of DSB formation [14] , [16] . These observations provide a connection between the nuclear envelope and pachytene checkpoint activity and suggest that PCH2′s role in nuclear events like crossing over is indirect and at a distance from the chromosomes . To test the significance of the correlation between pachytene delays and prolonged PCH2 expression , we manipulated the timing and expression levels of PCH2 in the germline . PCH2 levels were increased by driving the UASP:pch2 transgene with P ( Gal4::VP16-nos . UTR ) MVD1 [37] , abbreviated as MVD1 , known to drive high levels of expression in the germarium . MVD1-driven pch2 caused the protein to persist into region 3 oocytes , which was never observed with the NGT driver in a wild-type background ( Figure 4A and 4C ) . In addition to distinct foci , PCH2 was also distributed more evenly throughout the cytoplasm ( Figure 4A ) . Thus , MVD1-driven pch2 resulted in overproduction and prolonged expression of the protein throughout pachytene . Pachytene delays were not observed when the pch2 transgenes were expressed using the NGT driver ( Figure 5A ) . In contrast , MVD1-driven pch2 induced a pachytene delay that resulted in a high frequency of the two-oocyte phenotype ( Figure 5A ) . In fact , 100% ( n = 10 ) of the germaria with PCH2 expression in region 3 cysts also contained two oocytes , as viewed by C ( 3 ) G staining , suggesting prolonged PCH2 expression is sufficient to induce a delay in pachytene progression . Since overproducing PCH2 caused a pachytene delay , we determined if crossover frequency or distribution was also affected . We found that PCH2 expression driven by MVD1 altered the distribution of exchange in all 3 transgenic lines tested ( Table 1; Figure 5B ) . The most dramatic increase in crossover frequencies was observed in the centromere proximal interval of chromosome 2 , b-pr . Although all the transgenic lines that were tested showed the same change in crossover distribution , the magnitude was different , which probably reflects different transgenic protein levels . In support of this conclusion , the presence of two transgenic copies of P ( HA-pch2 ) 71 driven by MVD1 exacerbated the effect on both crossover levels and distribution ( Table 1; Figure 5B ) . These data show that the frequency and distribution of crossing over is sensitive to the timing and level of PCH2 expression during pachytene . We sought to identify factors that facilitate prolonged PCH2 expression and cause pachytene delays . The first candidate we tested was mei-218 since it is required for the pch2-dependent pachytene delays observed in DSB repair and exchange-defective mutants . mei-218 mutants , however , did not show an effect on the levels or distribution of MVD1-driven PCH2 ( Figure S2 ) . Also , the two-oocyte phenotype caused by PCH2 overexpression was not suppressed in mei-218 mutants ( Figure 5A ) , suggesting MEI-218 is not required for PCH2 localization . The second candidate tested was Sir2 , which encodes a histone deacetylase that is required for the nucleolus localization of Pch2 and the pachytene checkpoint during S . cerevisiae meiosis [15] . Five Drosophila genes belong to the Sir2 family . Of these , Sir2 is the closest homolog of the S . cerevisiae Sir2 [38] . Drosophila sir2 null alleles have no obvious effects on viability , but affect position effect variegation , heterochromatic silencing and fly life span [38]–[40] . sir2 mutants were fully fertile with wild-type levels of X-chromosome nondisjunction ( Table S1 ) , indicating Sir2 is dispensable for meiotic recombination . We investigated whether Sir2 is involved in the pachytene checkpoint and found that mutation of sir2 suppressed the high frequency of the two-oocyte phenotype observed when PCH2 is overexpressed with the MVD1 driver ( Figure 5A ) . The high frequency of the two-oocyte phenotypes observed in the exchange-defective mutant hdm and DSB repair mutant spn-B were also suppressed by sir2 ( Figure 6A ) . Likewise , Sir2 was required for the pachytene delay observed in TM3/+ heterozygotes ( Figure 6A ) and the delayed onset of γ-H2AV staining in spn-B mutants ( cyst 2–5 in Figure 6B ) . Thus , like pch2 , sir2 is required for the pachytene checkpoint . Strikingly , the region 3 localization of MVD1-driven PCH2 was eliminated in a sir2 mutant ( Figure 4A and 4C ) . In contrast , loss of sir2 only slightly reduced the level of PCH2 in region 2 cells and had no effect on the peri-nuclear localization of PCH2 driven by NGT ( Figure 4C and data not shown ) . In addition , expression of a c ( 2 ) M transgene driven by MVD1 was not affected , indicating the effect of sir2 on PCH2 was not due to a reduction in the transcription of UAS-driven genes ( Figure S3 ) . These results support the hypothesis that high levels of PCH2 are dependent on Sir2 and essential for the pachytene delays .
We have previously shown that removing the SC central element component C ( 3 ) G does not cause pch2-dependent delays in Drosophila meiotic prophase [14] . Therefore , the pachytene checkpoint is not monitoring the process of synapsis per se . Instead , two lines of evidence suggest the pachytene checkpoint is sensitive to defects in chromosome axes . First , mutations in genes that encode structural axis components , C ( 2 ) M and ORD , cause pch2-dependent pachytene delays . Second , heterozygous chromosomal rearrangements also cause a pch2-dependent delay . Homozygous rearrangements do not cause delays; therefore , the pachytene checkpoint is sensitive to any discontinuity in the alignment between homologous chromosomes . Since the delays do not depend on C ( 3 ) G , the defect must be prior to or independent of synapsis initiation . The misalignment of homologous sequences could destabilize the integrity of chromosome axes , such as the assembly of ORD or C ( 2 ) M , and expose substrates that trigger the pachytene checkpoint . Indeed , females doubly heterozygous for balancer chromosomes show deterioration of C ( 2 ) M staining in pachytene oocytes similar to ord mutants [21] , suggesting the axial elements are compromised by the heterozygous inversion breakpoints . The delays observed in c ( 2 ) M mutants and balancer heterozygotes do not depend on the MCM-related precondition genes such as mei-218 , which are required for the pachytene delays in DSB repair and exchange-defective mutants [14] . Balancer heterozygotes also do not cause a delayed response to DSBs or increase in the number of PCH2 foci as observed in DSB repair and exchange-defective mutants . Therefore , two pathways probably lead into a pch2-dependent checkpoint: a mei-218-dependent pathway involving the early function of DSB repair proteins and a mei-218-independent pathway involving the structure of chromosome axes . Of the two pathways in Drosophila , the pachytene checkpoint in other organisms has similarities to the mei-218-independent pathway involving chromosome structure . Heterozygous inversions and translocations induce a pachytene delay , suggesting a model in which the pachytene checkpoint can respond to breaks in axis continuity between paired homologs . The C . elegans pachytene checkpoint also monitors meiotic chromosome structure since a defect in a SC-nucleating “pairing center” triggers a Pch2-dependent response [16] . Similarly , the budding yeast pachytene checkpoint has been proposed to monitor SC-dependent events that may involve the relationship between recombination complexes and chromosome axes [41]–[43] . Therefore , a common feature of the pachytene checkpoints in these organisms is their sensitivity to the axis continuity between paired homologs with the main difference being SC-dependent defects ( yeast and nematodes ) versus SC-independent axis defects ( Drosophila ) . Interestingly , both yeast Pch2 and mouse Trip13/Pch2 have been proposed to have a checkpoint-independent role in the organization of chromosome axis proteins [43] , [44] . We do not know as of yet , however , if this is related to the sensitivity of paired axes at the Drosophila pachytene checkpoint , although it is tempting to suggest such a model . Pachytene checkpoint activity in budding yeast is associated with prolonged Pch2 expression that requires Sir2 [15] . As in budding yeast , Drosophila sir2 mutants are defective in the pachytene checkpoint and our overexpression studies suggest Sir2 is also required for the prolonged expression of PCH2 . These results provide evidence for an evolutionarily conserved role of Pch2 and Sir2 in monitoring changes in chromosome structure during meiotic prophase from yeast to a metazoan ( Figure 7 ) . Drosophila may have evolved an additional mei-218-dependent pachytene checkpoint , not shared with yeast and nematodes , which is sensitive to DSB repair complexes . The effect of inversion heterozygosity on the frequency of crossing over has been known since the work of Sturtevant in 1919 . Most often these intrachromosomal rearrangements cause an interchromosomal increase in recombination levels . Exhaustive work has been carried out investigating the interchromosomal effect and several models have been proposed in order to account for the increase in crossing over [11] . The most recent and generally accepted model was last described by Lucchesi and Suzuki in 1968 who proposed a timing model where pairing and crossover formation are coupled during the pachytene stage of prophase [11] . They suggested that when the pairing process between one set of homologs is perturbed or delayed by chromosome rearrangements , pachytene was lengthened and the opportunity to make crossovers was prolonged . We propose a modified version of the timing model where breaks in homology cause disruptions in the axis structure , resulting in a checkpoint-mediated delay ( Figure 7 ) . The timing model proposed by Lucchesi and Suzuki predicts that a factor exists which controls the timing of meiotic prophase and can affect the level of exchange [11] . The pachytene checkpoint may regulate this timing mechanism . Although pch2 is dispensable for crossing over in a wild-type background , it is required for most of the residual crossovers that occur in recombination-defective mutants . pch2 is also required for most of the interchromosomal effect and pachytene delays observed in inversion heterozygotes . To our knowledge , pch2 is the first example of a gene in Drosophila required for the interchromosomal effect that is not required for crossing over in general . Prolonged PCH2 expression may facilitate the formation of more crossovers by simply delaying a pachytene transition and extending the crossover determination phase , thereby allowing more crossover sites to be selected . An alternative explanation is that pch2 , while not required for crossover formation in wild-type , is required for a crossover mechanism active only in axis-defective situations . Since the interchromosomal effect is not mediated by an increase in DSBs , PCH2 most likely increases the chance of DSBs becoming crossovers at the expense of noncrossovers . Drosophila PCH2 localizes to peri-nuclear foci in zygotene and early pachytene cells and is rapidly degraded or no longer made at mid-pachytene . In mutants in which pachytene delays are observed , PCH2 expression persists longer than in wild-type . The observation that overexpressing PCH2 induces effects on both timing and crossover levels indicates prolonged PCH2 expression is necessary and sufficient for the pachytene checkpoint's downstream effects . Since the duration of early pachytene correlates with the domain of PCH2 expression , we suggest that degradation of PCH2 turns off checkpoint activity and allows progression through pachytene , which ends the crossover determination phase ( Figure 7 ) . We observed PCH2 localization to the outside of the nuclear envelope . These results were surprising considering the effect a pch2 mutation has on nuclear events like crossing over . While we cannot rule out the possibility that a small undetectable fraction of PCH2 protein enters the nucleus and interacts with the chromosomes , PCH2 may indirectly affect nuclear events by facilitating interactions between the chromosomes and the nuclear envelope . In budding yeast , the pachytene checkpoint requires the localization of Pch2 to the nucleolus [15] . Therefore , like budding yeast , PCH2 in Drosophila may mediate the pachytene checkpoint at a distance from the recombination sites . Intriguingly , the nuclear envelope has been linked to several cellular processes relevant to meiotic recombination , including homolog pairing and DSB repair [45]–[48] . In C . elegans , the pairing of homologous chromosomes first requires the relocation of chromosomal regions known as pairing centers to the nuclear envelope [45] . Chromosome deficiencies that remove the pairing center impair relocation , homolog pairing and synapsis as well as trigger a pch2-dependent response [16] . Therefore , it is possible that in other organisms , the nuclear envelope has a conserved role in transducing pachytene checkpoint effects .
Drosophila stocks and crosses were maintained on a standard medium at 25°C . The following mutant alleles were used unless otherwise noted- ord10 [20] , c ( 2 ) MEP , pch2EY01788a ( pch2EY ) , c ( 3 ) G68 [18] , hdmg7 , mei-2181 , rec1and rec2 [49] , Ercc1X [30] , mei-9a , spn-A1 , spn-BBu , sir217 [38] , and mei-W684572 . The deficiency Df ( 2L ) BSC245 deletes cytological bands 33F3-34A9 , which includes the sir2 locus . T ( 2;3 ) DP77 and T ( 2;3 ) dpD translocations were obtained from the Bloomington Stock Center . T ( 2;3 ) DP77 breakpoints are at 26E-27A on the 2nd and 85C on the 3rd . T ( 2;3 ) dpD breakpoints are at 25A on the 2nd and 95B–D on the 3rd . The T ( 2;3 ) ltX16 translocation has breakpoints at 40 ( heterochromatin ) on the 2nd and 95A3 on the 3rd and was obtained from B . Wakimoto [50] . X-chromosome nondisjunction was assayed by crossing females to y w/YBS males . The frequency of X-chromosome nondisjunction is calculated as 2 ( Bar females + Bar+ males ) /[2 ( Bar females + Bar+ males ) + Bar+ females + Bar males] . To estimate wild-type X chromosome crossing over frequency , y/y pn cv m f • y+ female flies were crossed to C ( 1:Y ) 1 , v f B: [+]; C ( 4 ) RM , ci ey males , and X chromosome recombinants were scored in males . Second chromosome crossing over was assayed by crossing al dp b pr cn/+ females to al dp b pr cn/CyO males and scoring for recombinants among the Cy+ progeny . P-values were calculated using the Fisher's exact test . For immunolocalization experiments , females were aged at room temperature for about 16 hours and ovaries were dissected and fixed using the “Buffer A” protocol [51] . The antibody to γ-H2AV was described by Mehrotra et al . [28] and used at a 1∶500 dilution . Additional primary antibodies included mouse anti-C ( 3 ) G antibody used at 1∶500 [18] , rabbit anti-C ( 2 ) M antibody used at 1∶400 [52] , a combination of two mouse anti-Orb antibodies ( 4H8 and 6H4 ) used at 1∶100 [53] , mouse anti-Lamin antibody developed by Fisher , P . A . used at 1∶800 , and a rat anti-HA antibody ( Roche ) used at 1∶15 . The secondary antibodies were Cy3 labeled goat anti-rabbit ( Jackson Labs ) used at 1∶250 , Cy3 labeled goat anti-rat ( Jackson Labs ) used at 1∶100 and Alexa fluor 488 goat anti-mouse ( Invitrogen ) used at 1∶100 . Chromosomes were stained with Hoechst 33342 at 1∶50 , 000 ( 10 mg/ml solution ) for seven minutes at room temperature . Images were collected using a Leica TCS SP2 confocal microscope with a 63X , N . A . 1 . 3 lens . In most cases , whole germaria were imaged by collecting optical sections through the entire tissue . These data sets are shown as maximum projections . The analysis of the images , however , was performed by examining one section at a time . The oocytes were observed using an anti-C ( 3 ) G antibody . In some cases , such as when C ( 3 ) G staining was not visible , anti-ORB staining was used to identify the oocyte ( s ) . A cell was scored as an oocyte if complete SC filaments were clear and distinct or by a concentration of ORB staining in the cytoplasm of a cell [54] . P-values were calculated using the Fisher's exact test . The P-value from the test compares the ratio of one-oocyte to two-oocyte cysts that were observed in two genotypes . The γ-H2AV foci were counted from germaria where the foci were clear and distinct . Foci numbers in wild-type were at a maximum in region 2a ( early pachytene ) and few foci were visible by region 2b ( mid pachytene ) . Therefore , to compare foci numbers in different genotypes , we used a method that calculates all cysts with γ-H2AV foci , averaging the amount in each pair of pro-oocytes . We compared the average foci in all the pro-oocytes or oocytes of each germarium , starting with the youngest cysts at the anterior end , by examining a full series of optical sections . For counting γ-H2AV foci in repair-defective backgrounds , ORB staining was used to identify oocytes in region 3 . The foci were counted from germaria where the foci were clear and distinct . The foci were counted manually by examining each section in a full series of optical sections containing complete pro-oocyte nucleus . Since the position of a cyst in the germarium is only a rough estimate of its meiotic stage , the foci were first counted in all the pro-oocytes/oocytes ( identified by C ( 3 ) G staining ) in the germarium . The meiotic stage of each pro-oocyte was then normalized according to the relative position of the entire cyst within the germarium since the relative position is more important than absolute position . The pro-oocytes from 13 wild-type germaria , 4 FM7/+ , 4 CyO/+ , 5 FM7/+; CyO/+ , 5 spn-BBu , and 4 sir217/Df; spn-BBu were arranged according to their relative age . The average number of γ-H2AV foci per pro-oocyte at each stage was then calculated and plotted as a function of relative cyst age . The annotated coding region of pch2 was obtained from Flybase and amplified off the pch2 cDNA clone LD24646 [55] by PCR . The coding region of pch2 was then cloned into the Gateway® pENTRTM4 vector ( Invitrogen ) . An LR ‘clonase’ reaction was then performed to recombine pch2 into the ppHW destination vector ( Invitrogen ) that contains 3 copies of an N-terminus HA-tag under the control of an inducible UASP promoter . The construct was injected into fly embryos by Model System Genomics at Duke University . To express the transgenic lines , they were crossed to flies expressing Gal4 using either the NGT ( P[Gal4-nos . NGT]40 ) [35] or MVD1 ( P[Gal4::VP16-nos . UTR]MVD1 ) drivers [37] . The HA-PCH2 foci were counted from germaria where the foci were clear and distinct . We counted the average foci surrounding nuclei in all the cysts at region 2 and region 3 of each germarium by examining a full series of optical sections . | Meiosis is a specialized cell division in which diploid organisms form haploid gametes for sexual reproduction . This is accomplished by a single round of replication followed by two consecutive divisions . At the first meiotic division , the segregation of homologous chromosomes in most organisms is dependent upon genetic recombination , or crossing over . Crossing over must therefore be regulated to ensure that every pair of homologous chromosomes receives at least one reciprocal exchange . Homologous chromosomes that do not receive a crossover frequently undergo missegregation , yielding gametes that do not contain the normal chromosome number , conditions frequently associated in humans with infertility and birth defects . The pch2 gene is widely conserved and in Drosophila melanogaster is required for a meiosis-specific checkpoint that delays progression when crossover formation is defective . However , the underlying process that the checkpoint is monitoring remains unclear . Here we show that defects in axis components and homolog alignment are sufficient to induce checkpoint activity and increase crossing over across the genome . Based on these observations , we hypothesize that the checkpoint may monitor the integrity of chromosome axes and function to promote an optimal number of crossovers during meiosis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"genetics",
"and",
"genomics/animal",
"genetics",
"cell",
"biology/nuclear",
"structure",
"and",
"function",
"molecular",
"biology/recombination",
"molecular",
"biology/chromosome",
"structure",
"genetics",
"and",
"genomics/chromosome",
"biology",
"genetics",
"and",
"genomics/gene",
"function",
"molecular",
"biology/dna",
"repair"
]
| 2010 | Chromosome Axis Defects Induce a Checkpoint-Mediated Delay and Interchromosomal Effect on Crossing Over during Drosophila Meiosis |
Stably suppressed viremia during ART is essential for establishing reliable simian models for HIV/AIDS . We tested the efficacy of a multidrug ART ( highly intensified ART ) in a wide range of viremic conditions ( 103–107 viral RNA copies/mL ) in SIVmac251-infected rhesus macaques , and its impact on the viral reservoir . Eleven macaques in the pre-AIDS stage of the disease were treated with a multidrug combination ( highly intensified ART ) consisting of two nucleosidic/nucleotidic reverse transcriptase inhibitors ( emtricitabine and tenofovir ) , an integrase inhibitor ( raltegravir ) , a protease inhibitor ( ritonavir-boosted darunavir ) and the CCR5 blocker maraviroc . All animals stably displayed viral loads below the limit of detection of the assay ( i . e . <40 RNA copies/mL ) after starting highly intensified ART . By increasing the sensitivity of the assay to 3 RNA copies/mL , viral load was still below the limit of detection in all subjects tested . Importantly , viral DNA resulted below the assay detection limit ( <2 copies of DNA/5*105 cells ) in PBMCs and rectal biopsies of all animals at the end of the follow-up , and in lymph node biopsies from the majority of the study subjects . Moreover , highly intensified ART decreased central/transitional memory , effector memory and activated ( HLA-DR+ ) effector memory CD4+ T-cells in vivo , in line with the role of these subsets as the main cell subpopulations harbouring the virus . Finally , treatment with highly intensified ART at viral load rebound following suspension of a previous anti-reservoir therapy eventually improved the spontaneous containment of viral load following suspension of the second therapeutic cycle , thus leading to a persistent suppression of viremia in the absence of ART . In conclusion , we show , for the first time , complete suppression of viral load by highly intensified ART and a likely associated restriction of the viral reservoir in the macaque AIDS model , making it a useful platform for testing potential cures for AIDS .
The study of persistence of viral sanctuaries during antiretroviral therapy ( ART ) and the possibility for their therapeutic targeting is crucial for eradication of HIV-1 . Animal models for lentiviral persistence during therapy are therefore needed . The creation of such animal models requires knowledge of the response of animal lentiviruses to antiretroviral drugs adopted in treatment of humans with HIV-1 . Finding cross-active drugs has been a difficult task because non-HIV-1 lentiviruses often mimic drug resistance mutations found in HIV-1 . This mimicry has been shown for the viral protease [1] and for the portion of reverse transcriptase ( RT ) that binds the non-nucleosidic reverse transcriptase inhibitors ( NNRTIs ) [2] . One of the current models is based on macaques infected with a molecularly engineered simian immunodeficiency virus ( SIVmac239 ) expressing HIV-1 RT , in order to overcome drug resistance mimicry of the primate lentiviruses to NNRTIs [3] . Another model ( SIV-based ) has been developed for neurotropic infection , a condition often occurring in late-stage AIDS [4] . In this case , in order to by-pass the different response to antiretrovirals , the authors used a drug combination which is not adopted in humans . However , in both of these animal models , low-level viremia persisted and viral RNA was consistently detectable in anatomical sanctuaries [3] , [4] . A model recently developed by our group is based on a polyclonal virus , such as SIVmac251 , mimicking , at least in part , the genetic diversity of HIV-1 naturally inoculated in human subjects [5] . It was recently shown that SIVmac251 responds to combined ART consisting of two nucleosidic/nucleotidic reverse transcriptase inhibitors ( NRTIs ) , i . e . tenofovir and emtricitabine , and the integrase inhibitor raltegravir [5] , [6] . In this treatment model , the virus persists during ART , and viral load rebounds following treatment suspension in a time frame remarkably similar to that observed in humans after treatment interruption [7] . Recent research has added more credit to the macaque AIDS model , showing that , similarly to humans [8] , [9] , rhesus macaques ( Macaca mulatta ) harbour a central memory CD4+ T-cell reservoir , which plays a pivotal role in AIDS pathogenesis [7] , [10] . Important insight has been derived from comparisons between rhesus macaques and sooty mangabays ( Cercocebus atys ) which , unlike M . mulatta , do not progress to AIDS [11] . M . mulatta , but not C . atys , shows up-regulation of the lentiviral co-receptor CCR5 in activated central memory T-cells , thus rendering this T-cell pool highly permissive to infection [10] . Conversely , the reduction of the long-lived memory T-cells ( CD95+CD28+ ) , including central memory T-cells , by the gold-based compound auranofin in intensified ART ( iART ) -treated rhesus macaques resulted in decreased levels of viral DNA and delayed progression of the infection upon therapy suspension [7] . Therefore , a model mimicking the effects of suppressive ART in humans is of fundamental importance also for the study of the dynamics of this viral reservoir . One major limitation of current models for HIV persistence during therapy is their large discrepancy from conditions observed in humans . So far , due to financial and temporal constraints , animals have been chosen from homogeneous cohorts in terms of timing , type and route of the inocula , and have been treated in the early phases of chronic infection [3]–[6] or during acute infection [12] . Instead , at therapy initiation , HIV-infected humans are usually characterized by different timings and routes of disease acquisition and different levels of progression of the infection . In order to obtain a robust animal model for HIV persistence during therapy , the drug regimens should display similar efficacies as compared to those employed for human treatment , and reproducible control of heterogeneous viral loads in wide cohorts of subjects with different characteristics and previous treatment histories . Here , we report a highly intensified ART ( H-iART ) regimen for the simian model , reproducibly capable of decreasing viral load to levels below assay detection limits in SIVmac251-infected macaques starting from a wide range of baseline viremic conditions , and overcoming previous treatment failures . We also report an unexpectedly impressive restriction of viral DNA in peripheral blood mononuclear cells , obtained by means of a pharmacological strategy entirely based on antiretroviral drugs .
CEMx174 and HTLV-I-transformed MT-4 cells were grown in RPMI-1640 medium supplemented with glutamine ( 200 µg/mL ) ( Invitrogen Life Technologies , Inc . Carlsbad , California ) , 10% heat-inactivated fetal bovine serum ( FBS; Invitrogen Life Technologies ) , penicillin ( 500 U/mL; Pharmacia Italia SPA ) and streptomycin ( 66 . 6 U/mL; Bristol-Myers , Sermoneta , LT ) . Peripheral blood from uninfected nonhuman primates was diluted 1∶2 with PBS 1x-NaCl , and peripheral blood mononuclear cells ( PBMCs ) were Ficoll-separated , resuspended at a concentration of 2×106/mL and stimulated for 3 days with 5 µg/mL phytohaemoagglutinin ( PHA ) ( Difco Laboratories , Detroit , MI , USA ) and 100 units/mL of human recombinant IL-2 ( Roche Diagnostics , Indianapolis , IN , USA ) . CEMx174 , MT-4 cells , and three-day old PBMCs were challenged with standard viral stock preparations for 2 hours in an incubator at 37°C with 5% CO2 , washed and incubated with increasing drug concentrations ( 0 . 0001–1 µM ) , according to a previously published protocol [5] . The assays on virus entry inhibitors such as maraviroc ( MRV ) , were conducted as in [13] . Briefly , the drug was first added during incubation with the virus and the same drug concentrations were then re-added upon cell washing . In MT-4 cells , through the MTT assay ( MT4-MTT ) , we measured inhibition of the cytopathic effect of the two viruses . The assay was performed when the majority of control infected cells were dead . At different intervals post-infection , the viral core antigen p27 was measured in supernatants by antigen-capture ELISA assays ( Advanced BioScience lab . , Inc . ) . The Indian rhesus macaques used in this study were housed at Bioqual , Inc . , according to standards and guidelines as set forth in the Animal Welfare Act , the Guide for the Care and Use of Laboratory Animals , and the Association for the Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) , following approval by the Institutional Animal Care and Use Committee ( IACUC ) . A total of eleven macaques have been enrolled for this study , while five previously enrolled macaques have been employed as historical controls . For the pilot study , four SIVmac251-infected non-human primates ( M . mulatta ) that had been stably viremic at least for the last 3 . 3 months were put under a regimen ( i . e . ART ) consisting of tenofovir ( PMPA ) , emtricitabine ( FTC ) and raltegravir [5] , for 1 . 5 months . To improve control of viral load , this regimen was intensified by the addition of darunavir ( DRV ) boosted with ritonavir ( /r ) [intensified ART ( iART ) ] . After 80 days , the treatment was further reinforced [highly intensified ART ( H-iART ) ] by the addition of maraviroc . For the second part of the study , eight additional SIVmac251 infected animals were used . These animals were divided into three treatment groups . One group ( n = 2 ) was treated with MRV/r alone for three weeks , followed by addition of tenofovir/emtricitabine/raltegravir/DRV . A second group ( n = 4 ) was treated with all H-iART drugs administered simultaneously . A third group ( n = 2 ) was treated with iART to serve as controls . For the combined antireservoir/antiretroviral therapeutic protocols , macaque P252 , previously treated with iART plus the anti-reservoir drug auranofin ( for detail , see Ref . 7 ) , was put under a H-iART regimen for one month when viral load rebounded after suspension of the previous treatment . Another macaque , P177 of the pilot study , was treated ( after the end of the follow-up aimed at monitoring the effects of H-iART alone ) with auranofin in addition to H-iART . This macaque was then subjected , similarly to P252 , to a further cycle of H-iART at viral load rebound . More detailed information on the macaques enrolled , their viro-immunological background and the therapeutic regimens adopted for each animal can be found in Table S1 . All animals were dosed subcutaneously with tenofovir , and emtricitabine , and orally ( with food ) with raltegravir , DRV/r , and MRV . Initial drug dosages were: tenofovir , 30 mg/kg/day; emtricitabine , 50 mg/kg/day; raltegravir , 100 mg bid; DRV , 375 mg bid ( for macaques starting from viral loads lower than 105 viral RNA copies/mL ) or 700 mg bid ( for macaques starting from viral loads higher than 105 viral RNA copies/mL ) ; ritonavir 50 mg bid; MRV 100 mg bid . Tenofovir and emtricitabine were kindly provided by Gilead Sciences ( Foster City , CA ) . Raltegravir , DRV/r and MRV were purchased from the manufacturers . For measurement of plasma SIVmac251 RNA levels , a quantitative TaqMan RNA reverse transcription-PCR ( RT-PCR ) assay ( Applied Biosystems , Foster City , Calif . ) was used , which targets a conserved region of the gag transcripts . The samples were then amplified according to a method previously validated in our hands [see ref 5 and Fig . S1] . The sensitivity of the method is two copies per run , which results in a detection limit as low as 40 RNA copies/mL in our routine analyses . Briefly , a 500-µL aliquot of plasma was spun down at 13 , 000× g for 1 h . The liquid was poured off and 1 mL of RNA-STAT 60 was added . After 5 min . , 250 µL of chloroform was added and vortexed . The samples were spun at the same speed for 1 h . The clear aqueous layer on top was removed , and added to 500 µL of isopropanol . Then , 10 µl of 10 µg/mL tRNA was added and precipitated overnight at −20°C . The samples were spun for 1 hour , washed with a cold ( −20°C ) 75% ethanol solution , and re-spun for 1 h . The RNA was resuspended in 30 µL of RNAse-free water . 10% of the resuspended RNA was added to Taqman reagents ( Applied Biosystems ) , plus primers and probe , and amplified in a 7700 Sequence Detection System by Applied Biosystems . Briefly , the sample was reverse transcribed at 48°C for 30 min . using One-Step RT-PCR Master Mix ( Applied Biosystems ) , then held at 95°C for 10 min . , and run for 40 cycles at 95°C for 15 sec . and 60°C for 1 min . The following PCR primer/probes were used: SIV2-U 5′ AGTATGGGCAGCAAATGAAT 3′ ( forward primer ) , SIV2-D 5′ GGCACTATTGGAGCTAAGAC 3′ ( reverse primer ) , SIV-P 6FAM-AGATTTGGATTAGCAGAAAGCCTGTTGGA-TAMRA ( TaqMan probe ) . The signal was finally compared to a standard curve of known concentrations from 107 down to 1 copy ( the linear range of concentration/signal relation spans eight Logs ) . All samples were done in triplicate for consistency and accuracy . In our increased sensitivity analyses , RNA was extracted from 6 mL of starting plasma , leading to a sensitivity threshold of 3 copies/mL . The inter-assay variability of the assay is 23 . 4%; The intra-assay variability is 20 . 6% . For proviral DNA detection , cells were spun down to a pellet , and the supernatant was poured off . The cell pellet was lysed with 1 mL of DNASTAT for 10 min . 250 µL of chloroform was added and the mixture was vortexed . The samples were spun at 13 , 000 for 1 h . and the aqueous layer was removed and added to another tube . To this , 500 µL of isopropanol was added , and the mixture was precipitated overnight at −20°C . The samples were then spun for 1 h and the precipitate was washed with a −20°C-cold , 75% ethanol solution , and re-spun for 1 h . The DNA pellet was resuspended in 30 µL of water and 10% of the resulting solution was added to Taqman reagents ( Applied Biosystems ) plus primers and probe ( the same as in previous paragraph ) and amplified in a 7700 Sequence Detection System by Applied Biosystems . The signal was finally compared to a standard curve of known concentrations from 106 down to 1 copy ( the linear range of concentration/signal relation spans seven Logs ) . The detection limit of this assay is two copies of proviral DNA/5×105 cells . The inter-assay variability is 28 . 3%; the intra-assay variability is 9 . 9% . The presence of PCR inhibitors in both the quantitative assays ( viral RNA and proviral DNA ) has been ruled out by spiking the samples with known amounts of viral RNA and proviral DNA respectively ( see Table 1 and Table S2 ) . Animals were bled before feeding in the morning , in order to obtain reliable measurements of trough drug levels . Plasma was obtained from supernatants of ficoll-centrifuged blood . For DRV , sample preparation involved addition of an internal standard and liquid-liquid extraction with 2 mL tert-butylmethylether ( tBME ) at basic pH , and reconstitution in 100 µL of mobile phase to concentrate the sample . Reversed phase chromatographic separation of the drugs and internal standard was performed on a YMC , C8 analytical column under isocratic conditions . A binary mobile phase was used consisting of 55% 20 mM sodium acetate buffer ( pH 4 . 88 ) and 45% acetonitrile . The UV detector set to monitor the 212 nm wavelength provided adequate sensitivity with minimal interference from endogenous matrix components . Calibration curves are linear over the range of 50 to 20 , 000 ng/mL . Inter- and intraday variability was less than 10% . For MRV , a protein precipitation method using acetonitrile ( AcN ) containing internal standard ( MVC-d6 ) was employed to extract the drug from macaques' plasma . An aliquot of the supernatant was further diluted with 0 . 5% tirfluoroacetic acid to maintain signal intensity within the linear range of the instrument . Reversed phase chromatographic separation was performed on an XBridge C18 analytical column under isocratic conditions . A binary mobile phase consisting of 0 . 1% formic acid in water and 0 . 1% formic acid in acetonitrile ( 72∶28 ) was used and provided adequate separation from other analytes . Detection and quantitation was achieved by multiple reaction monitoring ( MRM ) , and MVC and internal standard were detected using the following transitions for protonated molecular products [M+H]+: m/z MVC 514 . 2>106 . 0; m/z MVC-d6 520 . 3>115 . 0 . The assay has a dynamic range of 5 to 5 , 000 ng/mL using 20 µL plasma . For both DRV and MRV total drug concentrations were measured ( i . e free and protein bound ) . Hematological analyses were performed by IDEXX ( IDEXX Preclinical Research , North Grafton , MA ) . For calculation of absolute CD4+ and CD8+ T-cell numbers , whole blood was stained with anti-CD3-fluorescein isothiocyanate ( FITC ) /anti-CD4-phycoerythrin ( PE ) /anti-CD8-peridinin chlorophyll α protein ( PerCP ) /anti-CD28-allophycocyanin ( APC ) , and anti-CD2-FITC/anti-CD20-PE , and red blood cells were lysed using lysing reagent ( Beckman Coulter , Inc . , Fullerton , Calif . ) . Samples were run on a FACSCanto II ( BD Biosciences , San Jose , CA ) . Staining for naïve ( TN: CD28+CD95− ) , central and transitional memory ( TTCM/TM: CD28+CD95+ ) , and effector memory ( TEM: CD28−CD95+ ) T-cells was performed on PBMCs isolated from total blood of three rhesus macaques treated with H-iART . For each animal , the blood was collected monthly from 0 to 4 months from the addition of MRV to the drug regimen . The cells ( 3×105 per sample ) were surface stained by incubation with six appropriately titrated monoclonal antibodies ( mAbs ) for 20′ at 4°C , washed with PBS and resuspended in 1% paraformaldehyde in PBS . The following mAbs were used: anti-CD3 ( APC-Cy7 ) , anti-CD4 ( Per-CP ) , anti-CD8 ( Pe-Cy7 ) , anti-CD20 ( APC ) , anti-CD28 ( FITC ) and anti-CD95 ( PE ) . Six-parameter flow-cytometric analysis was performed on a FACS Canto II instrument ( BD Biosciences ) [7] . Staining for HLA DR+ T-cells was performed with the same procedure described above , but with the substitution of an anti-HLA-DR antibody ( APC , clone G46-6 ) to the aforementioned anti-CD20 antibody . The absolute numbers of naïve ( CD95−CD28+ ) , long-lived ( CD95+CD28+ ) and short-lived ( CD95+CD28− ) memory CD4+ T-cells and the numbers of HLA-DR+ cells were deduced from percentage values of parent cells . Specific immune responses were detected by measuring gamma interferon ( IFN-γ ) secretion of macaque PBMCs stimulated with a SIVmac239 Gag peptide ( 15-mer , obtained through the AIDS Research and Reference Reagent Program , National Institutes of Health [NIH] , catalogue no . 6204 , peptide 64 ) in an enzyme-linked immunospot ( ELISPOT ) assay . The assay was performed with the ELISpotPRO for monkey interferon-γ kit ( Mabtech AB , Nacka Strand , Sweden ) according to the manufacturer's instructions . Briefly , 1 . 5×105 Ficoll isolated macaque PBMCs were added to 96 well plates pre-coated with an anti-human/monkey IFN-γ antibody ( MAb GZ-4 ) . Cells were resuspended in RPMI 1640+10% FBS with 2 µg/mL of the peptide . After 48 hours incubation at 37°C with 5% CO2 , the cells were rinsed from the plates , and a biotinylated anti-human/monkey IFN-γ antibody ( MAb 7-B6-1; Mabtech ) was added to the wells . The plates were then washed with PBS and incubated with the substrate solution ( BCIP/NBT-plus ) . Spots were counted by using an automated reader ( Immunospot Reader , CTL analyzers , LLC , Cleveland , OH ) . Numbers of spot-forming cells ( SFC ) /106 cells for each set of wells were averaged . A response was considered positive if the number of SFC/106 cells was at least four times the background value . Data were analyzed using the software GraphPad Prism 5 . 00 . 288 ( GraphPad Software , Inc . , San Diego , CA ) . For calculation of the EC50 and EC90 values , data were transformed into percentage-of-inhibition values , plotted on x , y graphs , and subjected to linear or non-linear regression , depending on the best-fitting equation . Response to drugs in vivo was evaluated by repeated-measures ANOVA , followed by an appropriate post-test to analyze differences between time points . An appropriate transformation was employed to restore normality , where necessary . Logit analysis was adopted to investigate the influence of variables on binary outcomes , using an online calculator ( http://statpages . org/logistic . html ) . Trends in time were analyzed by regression analysis ( GraphPad Prism ) , using the most appropriate equations . Akaike's information criteria ( AICc ) were used to select the model that was most likely to have generated the data and to compare the differences between equation parameters . The inter-assay variability of quantitative real time PCR was estimated as an average of the coefficients of variation ( CV ) of matched measurements in two assays conducted on different occasions; the intra-assay variability was estimated as the coefficient of variation of multiple replicates ( at least five ) within the same assay . Numerical simulations were performed with the ordinary differential equations solver ODEPACK of the Scilab 5 . 3 . 3 software ( http://www . scilab . org/ ) . The solver is based on finite difference methods for non-stiff problems , but it dynamically monitors the data in order to decide whether the stiffness of the problem requires a Backward Differentiation Formula method . The values of the discrete five-dimensional vector function of the solution were computed every 0 . 01 days . Details on mathematical modeling are given in Text S1 .
The first part of this study was aimed at obtaining long-term viral suppression in a group of macaques ( n = 4 ) in order to develop a suitable platform for testing experimental eradication strategies . We first analyzed the susceptibility of SIVmac251 to the protease inhibitor darunavir ( DRV ) and the CCR5 blocker maraviroc ( MRV ) in order to expand the arsenal of antiretroviral options available for the macaque AIDS model . DRV was chosen because of its well documented ability to inhibit several drug-resistant HIV-1 isolates as well as HIV-2 , a virus closely related to SIVmac251 [1] , [14] , [15] . Moreover , the choice of this drug was supported by preliminary bioinformatic and molecular modeling analyses showing the potential interactions of DRV with the SIVmac251 protease [Text S2 and Fig . S2] . MRV , a CCR5 antagonist , was chosen on the basis of the important role of CCR5 as a SIVmac251 co-receptor [16] and due to the antilentiviral activity previously demonstrated by one experimental CCR5 blocker in macaques [17] . Moreover , retrospective analysis of one previous in-vivo experiment supported the anti-SIVmac251 effect of this drug [Text S3 and Fig . S3] . Results from tissue culture experiments showed that both DRV and MRV inhibited SIVmac251 replication in the nanomolar range , with EC50 values well below the trough concentrations detected in macaques treated with these drugs and described below in the text . ( Fig . 1 ) . A group of macaques [n = 4] displaying signs of immune deterioration ( eighteen months post-inoculation ) was treated with a regimen of tenofovir , emtricitabine and raltegravir ( Fig . 2 ) . These macaques were derived from viral titration experiments and selected among those maintaining stable plasma viral loads ( Fig . 2A ) . The selected animals displayed viral load set points between 103 and 105 viral RNA copies/mL . As our study was aimed at obtaining a model mimicking the conditions found in HIV-1-infected individuals under ART , such baseline values were chosen in order to reflect the average viral loads at which treatment is started in humans . The CD4 counts displayed by the macaques enrolled in this “pilot” study were significantly lower than values observed in uninfected controls ( Fig . S4 ) , suggesting that they were unlikely to be long-term non-progressors or élite controllers . The three-drug regimen proved insufficient to maintain control of viral load in three of the four animals treated ( Fig . 2A ) . DRV ( 375 mg bid ) , boosted with ritonavir ( 50 mg bid ) , henceforth referred to as DRV/r , was added to the treatment in an attempt to obtain a more stable control of viral load . DRV/r significantly improved control of viral load , inasmuch as viral RNA in plasma was maintained at a significantly lower level as compared to the pre-therapy values ( Fig . 2A ) . No similarly decreasing trend of viral load was observed in an untreated control group of macaques [n = 2] showing non-significant differences in baseline viral loads as compared to the treatment group ( two tailed t-test: P = 0 . 803; Fig . 2A ) . We conclude that the iART regimen adopted improves control of viral load in SIVmac251-infected macaques . To increase the chances for long-term control of SIVmac251 replication , we explored the in-vivo efficacy of the CCR5 inhibitor MRV . This drug ( 100 mg BID ) was eventually added to the drug cocktail in the aforementioned group of macaques ( Fig . 2 ) . After MRV was started , all macaques stably maintained viral loads below the limit of detection of the assay ( i . e . 40 copies RNA/mL; Fig . 2A ) . There were also significant increases in the absolute numbers of CD4+ T-lymphocytes ( Fig . 2B ) . Henceforth , this multidrug combination will be referred to as highly intensified ART ( H-iART ) . In order to further support the contribution of MRV to the antiretroviral effects observed , we treated two macaques with MRV ( ritonavir boosted , MRV/r ) in monotherapy ( Fig . 3 ) . In line with its CCR5-blocking ability , MRV decreased the viral loads in two drug-naïve macaques with dynamics similar to those previously shown by an investigational CCR5 blocker [17] . When the other H-iART drugs were added to MRV , a quick abatement of viral load to levels below the assay detection limit could be demonstrated ( Fig . 3 ) . Prior to treatment with antiretrovirals , approximately one third of the experimental infections of macaques with SIVmac251 results in viral set points comparable to those displayed by the macaques described in the previous paragraphs ( Fig . S5 ) . To check whether H-iART might reproducibly control viral replication in SIVmac251 infected macaques characterized by higher viral loads , five animals with viral set points ranging from 103 to 107 viral RNA copies/mL of plasma were treated with H-iART , and the viral decay dynamics were compared with those of macaques treated with iART . Results clearly showed that H-iART induced a significantly more rapid decay in viral load than did iART ( Fig . 4A ) . In line with the efficacy of H-iART , CD4+ T-cells increased in all study macaques ( Fig . S6 ) . We conclude that MRV-containing H-iART is superior to iART in abating viremia load in a group of macaques with a wide array of baseline viral loads . We then analyzed the viral load decay dynamics in macaques treated with H-iART ab-initio . SIVmac251-infected macaques responded to administration of H-iART with a two phase exponential decay , as described in humans treated with suppressive ART [18] ( Fig . 4 ) . Similarly to the average treatment outcomes in humans [19] , the level of viral load suppression depended on the baseline viral loads , with macaques starting from higher viral loads showing viral blips or residual , though markedly decreased ( >3 Logs ) , viral replication ( Fig . 4D–F ) . We increased the DRV and MRV dosage in macaques 4887 , BD64 and BD69 , i . e . those starting from higher baseline viral loads ( >105 ) and showing incomplete control of viral replication or major blips . Results showed that the improved drug regimen led to viral loads consistently below the assay detection limit in animals BD64 and BD69 ( Fig . 4D , E ) . The increased drug dosage was also able to decrease the amplitude of the remaining sporadic blips ( Fig . 4E ) . The resulting blips were lower than 103 copies of viral RNA/mL , thus mimicking those observed in humans under ART [20] . Nevertheless , one animal ( 4887 ) experienced a further viremic episode . Analysis of the cerebrospinal fluid ( CSF ) of this animal showed a viral load approximately one order of magnitude higher than that in plasma , while CSF samples were below the assay detection limit ( i . e . 40 copies/mL ) in the macaques showing stable control of viral replication ( data not shown ) . This evidence suggested that the central nervous system ( CNS ) was a likely major source for the rebounding virus in macaque 4887 . According to previously published studies: 1 ) virus levels in the CSF during the advanced stages of the disease are mostly due to CNS sources [21] , and 2 ) the protease inhibitors ( i . e . , the only drug class in our cocktail acting at a post-translational level , and hence on chronically infected cells ) are extruded from the CNS by P-glycoprotein ( P-gp ) molecules in the blood-brain barrier [22] . We thus intensified the P-gp blockade by increasing , from 50 to 100 mg bid , the dosage of ritonavir , which is a well-known P-gp inhibitor [23] . The viral load decreased in both plasma and CSF , with a more rapid decay kinetic in plasma , in which viral RNA eventually fell to levels below the assay detection limit ( Fig . 4F ) . This result is in good agreement with the hypothesis of the CNS as a major source for the rebounding virus . We conclude that macaques starting from high viral loads respond to H-iART similarly to HIV-infected humans and that viral loads can be abated to levels below the assay detection limit by adjusting the drug dosages and boosting procedures . To check the presence of low-level viremia in SIVmac251-infected macaques under H-iART , we lowered the detection limit to 3 copies of viral RNA/mL and re-measured viral loads in some selected pooled serum samples . We found no evidence for low-level viral replication in plasma of all of the macaques tested ( Table 1 ) . Of note , viral RNA was below the assay detection limit in the plasma samples taken from macaque 4887 before its last viremic episode , supporting the hypothesis that H-iART was able to completely control viral replication in the periphery , despite the presence of a major CNS reservoir ( Fig . 4F ) . Analyses conducted on lymph node biopsies ( inguinal ) showed that four out of six macaques analyzed had levels of cell-associated RNA below the limit of detection of the assay ( i . e . 2 copies/5*105 cells/mL ) ( Table 2 ) . The presence of cell-associated RNA in lymph nodes was independent of baseline viremia at treatment initiation ( Logit analysis P = 0 . 801 ) , thus supporting the idea that the suppressive efficacy of H-iART is not confined only to those macaques starting from moderate viral loads . In addition , cell associated RNA measured in samples taken from rectal biopsies was below the assay detection limit in all animals analyzed , supporting the idea of full suppression of peripheral viral replication ( Table 2 ) . This was rather surprising , because other antiretroviral regimens adopted in macaques proved unable to completely control viral RNA in anatomical sanctuaries [3] , [24] . In the pilot study presented above , we unexpectedly found that H-iART profoundly impacted on viral DNA . First , there was a late viral DNA decay to levels below the assay detection limit which was associated with the addition of MRV to the drug cocktail ( Fig . 5A ) . In addition , the CD4/CD8 ratio , the decrease of which is a marker of the viral reservoir and/or ongoing viral replication [7] , [8] , significantly increased during treatment ( Fig . 5B ) . Of note , viral DNA in PBMCs also fell below the assay detection limit in all macaques included in the group treated with H-iART ab-initio ( median treatment duration = 125 days , range from 45 to 174 days ) , i . e . no viral DNA copies were detectable in six out of six repeats with a threshold sensitivity of 2 copies/5*105 cells . Moreover , we could not detect viral DNA in lymph node and rectal tissue biopsies ( detection limit: 2 copies/5*105 cells , three repeats per sample ) in all the macaques of the pilot study tested ( Table 2 ) . Lymph node viral DNA was also below the assay detection limit in one of three macaques from those treated with H-iART ab-initio , while viral DNA was below the limit of detection in rectal biopsies of all the macaques of the same group ( Table 2 ) . The results were further validated by excluding the presence of PCR inhibitors using spiked DNA for selected samples ( Table S2 ) . The dynamics of the viral DNA decay during H-iART were studied in those animals to which all H-iART drugs were administered simultaneously and for which viral DNA measurements were available . The levels of viral DNA in PBMCs during time were consistent with a three-phase decay , with the first two phases paralleling the two-phase decay of viremia , and a third , slower phase occurring after viremia had fallen to levels below the assay detection limit ( Fig . 5C ) . This last phase of the viral decay has been ascribed to the latently infected T-cell numbers [18] . This result was noteworthy , because no such decreasing trends in viral DNA had been observed in animals treated with iART ( i . e . without MRV ) [7] . In line with the reportedly stimulating effect of the major CCR5 ligand RANTES on T-cell proliferation [25] some studies suggested that MRV , by acting as an antagonist of this cytokine , might alter the T-cell dynamics in vivo [26] . To study these phenomena , the CD4+ T-cell subpopulations were analyzed by six-color flow-cytometry at different time points following addition of MRV to the therapeutic regimen ( Fig . 6 ) . To avoid biasing the result with the possible effects of a detectable viral load on the T-cell subpopulations , these tests were conducted on PBMCs from macaques P157 , P185 and P188 which already displayed a viral load below the assay detection limit when MRV was added ( Fig . 2 ) . Results showed that H-iART decreased the memory CD4+ T-cell numbers over time ( Fig . 6A , B ) , while it carried out no significant effect on the naïve T-cell subpopulation ( Fig . 6C ) . This result is in accordance with the in-vitro inhibitory effect of MRV on the proliferation of sorted memory T-cell subpopulations ( Fig . S7 ) . MRV significantly decreased the numbers of activated ( HLA-DR+ ) CD4+ TEM cells ( Fig . 6D ) . This effect is in line with decreased levels of immune activation already observed in humans treated with this drug [26] , [27] . In conclusion , MRV decreased the number of memory T-cells as well as TEM cell-activation . Since these two parameters are linked to the magnitude of the viral reservoir and ongoing viral replication [9] , [28] , this effect is in good agreement with the aforementioned three-phase decay of viral DNA induced by MRV ( Fig . 5C ) . The results so far obtained were in line with a recently issued report which suggested that MRV decreased the magnitude of the viral reservoir in HIV-1-infected individuals [26] . This study , which was unable to provide conclusive evidence , did not show an impact of MRV on the viral set point following therapy suspension , a parameter stringently associated with the extent of the viral reservoir [7] , [29] , [30] . To test this hypothesis , we analyzed the difference in the pre and post-therapy viral set points in those macaques from our cohort that had received MRV and that had undergone therapy suspension ( for treatment details see Figs . 2 , 4 and Text S3 ) . Results show that treatment with MRV is associated with a reduction of the viral set point post-therapy ( Fig . 7A ) , and that the extent in the viral set point decrease depends on the total exposure to the drug ( Fig . 7B ) . These results are suggestive of an independent effect of MRV on the viral set point following therapy suspension and add credit to the hypothesis that MRV may contribute to an anti-reservoir effect of H-iART . Finally , given the aforementioned effects of H-iART , we tested whether this therapeutic regimen might be adopted to improve the effect of a previous anti-reservoir strategy based on the anti-memory drug auranofin in combination with antiretrovirals [7] . Upon interruption of this anti-reservoir treatment , SIVmac251-infected macaques experience an acute infection-like condition , i . e . an initial viral load peak followed by rapid containment of viral load [7] . The peak , which is rapidly reached upon virus re-appearance in plasma , is associated with the reconstitution of the viral reservoir , as shown by the previously published independent association between the area under the curve ( AUC ) describing the initial peak of viral load and the eventual viral load set point ( [7] see also Fig . 8A ) . From this association , it follows that decreasing the AUC at peak artificially through a cycle of H-iART should limit the reconstitution of the viral reservoir and may result in spontaneous control of viral load following H-iART suspension . The experiment was attempted in two macaques . A first macaque ( P252 ) was treated with a one-month cycle of H-iART at viral load rebound , after the suspension of the aforementioned auranofin/antiretroviral regimen . Another macaque ( P177 ) was treated with auranofin in addition to H-iART as a follow-up of the treatment presented in the pilot study . Eventually , following therapy suspension , P177 was subjected to a short H-iART cycle at viral rebound , similar to that administered to P252 . In both cases , the short H-iART cycle promptly abated viral load to levels below the assay detection limit , thus efficiently decreasing the initial AUC ( Fig . 8 A–C ) . The macaques showed exceptionally low viral set points after the short cycle of H-iART was suspended , in line with the expected values calculated on the basis of our AUC/viral set point correlation curve ( Fig . 8A ) . Both macaques periodically displayed viral load peaks that subsequently decreased to low-level viremia ( <500 copies of viral RNA/mL ) or to levels below the assay detection limits . The CD4 slope was non-significant during the follow-up period ( P = 0 . 7079 for P252 and P = 0 . 2319 for P177; Fig . 8D , E ) , in line with the previous observation that the CD4 slope following therapy suspension identifies the impact of a treatment on the viral reservoir [7] . Conversely , CD4 counts had shown significantly decreasing trends in both macaques before all treatments were started ( P<0 . 0001 for P252 and P = 0 . 0039 for P177; Fig . 8D , E ) , thus supporting the concept that the therapies adopted significantly impacted on the natural course of the disease . Consistently with its exceptional reduction of the AUC at peak , macaque P177 showed a remarkable degree of spontaneous control of viral load during six months of follow-up , which was not yet considerable as , but seemingly close to a drug-free remission of the disease ( Fig . 8C ) . In this macaque , viral load was maintained at levels below the assay detection limit during the periods between peaks ( detection limit: 40 RNA copies/mL ) and , when the RNA detection limit was further lowered to 3 copies/mL , no evidence of residual viremia was found ( see Table 1 ) . This control of viral replication could hardly be ascribed to cell-mediated responses , in that a moderate increase in the number of IFN-γ positive spots could be detected only at viral rebound but not during the viral set point ( Fig . S8 ) , thus suggesting that H-iART induced a true containment of the viral reservoir reconstitution , similarly to other experimental strategies restricting the formation of the viral reservoir during acute infection [29] , [30] , [31] . We conclude that a short course of H-iART , in line with the highly suppressive effect of this therapeutic regimen on SIVmac251 , may prevent the viral reservoir reconstitution following suspension of a previous anti-reservoir therapy and result in a drug-free spontaneous control of viral load .
Some investigators recently questioned the robustness of primate models , citing the difficulty of obtaining , with the cross-active drug options available , full viral suppression in sanctuaries and viral loads below the assay detection limits for prolonged periods [32] , [33] . The results reported in the present article do not support this argument . 1 ) Since a good animal model should mirror full viral suppression in humans , we checked viral loads in plasma for prolonged periods and analyzed the presence of viral nucleic acids in anatomical sanctuaries . The level of abatement of viral nucleic acids that we found in the present study in peripheral blood and anatomical sanctuaries of the majority of the macaques tested provide the maximum degree of viral suppression so far observed in antiretroviral treated primates . The level of reproducibility of these results is shown by the fact that they were obtained in a heterogeneous group of macaques , likely mirroring a wide number of possible disease conditions in humans . This is the first report , to our knowledge , of a therapy capable of stably controlling viral replication to levels below the assay detection limits also in macaques in the advanced stage of the disease , since the studies so far published have been able to report control of SIV replication only during acute infection [12] or in the early chronic phase of the disease [3]–[6] . Apart from mimicking the clinical conditions of a significant portion of HIV-infected individuals who are diagnosed in the chronic or pre-AIDS stages of the disease , this ‘late’ treatment allows excluding those macaques able to spontaneously control the infection , a phenomenon which usually occurs soon after the acute infection phase [34] . For the macaques enrolled in this study , the average plasma viral load at the time of therapy initiation was of 4 . 8±1 . 1 Log10 RNA copies/mL ( mean ± SD ) . This value is lower than those reported in some articles during chronic SIVmac infection of macaques [35] , [36] , but similar to those published in other articles [37] , [38] . As in this study we have not included macaques with viral loads during chronic infection higher than 6 . 8 Log10 RNA copies/mL or with the rapid progressor phenotype , the effect of our H-iART regimen on this more aggressive course of SIV infections remain to be ascertained . Of note , persistence of the virus at low level in the lymph nodes of a minority of H-iART treated macaques provides another similarity of our macaque model with clinical conditions observed in humans infected with HIV-1 , as this anatomical sanctuary has recently been shown to be a major site for ongoing viral replication in humans [39] . Studies of drug penetration in this anatomical compartment will be necessary to overcome this limitation in both macaques and humans . 2 ) As in any well respected science , the results are in good agreement with mathematical models ( Fig . 9 ) , and are mathematically predictable ( as an example , see Fig . 8A ) . In this regard , important insight into the necessity for a multidrug regimen to control viral loads in macaques can be derived from a mathematical model developed by Rong and Perelson [40] and based on experimental observations [8] . This model suggests that a superior drug efficacy is required in simian AIDS to control viral replication ( Fig . 9 ) because of the viral burst size , ( i . e . the average number of virions produced by a single productively infected cell in a day ) . The viral burst size was shown to be higher in SIV infection as compared to HIV-1 infection [41] , where a lower drug efficacy is expected to be sufficient to maintain viral control ( Fig . 9A–C ) . Also a drug acting on the proliferation rate of activated T-cells , such as MRV ( which antagonizes the proliferative effect of RANTES , see ref 25 and Fig . S7 ) , appears to be important for containment of the viral blips ( Fig . 9D ) . These simulations also show that the decreased proliferation rates may impact on the viral reservoir size ( half-life: ≈200 days , see Text S4 , S3 , S2 , S1 and Fig . 9D ) , which shows a half-life of the same order of magnitude as that calculated by analyzing the dynamics of the viral DNA decay during H-iART ( Fig . 5 ) . 3 ) According to the idea that a good animal model should represent a vanguard for future treatments to be tested in humans , our quest for increased drug efficacy in the macaque AIDS model allowed identifying unexpected benefits of H-iART on the immune system . Apart from the possible impact of H-iART on the viral reservoir ( a concept supported by recent data in humans [42] ) , reduction by MRV of the memory T-cell subpopulation may restrict one major source for viral spread and ongoing viral replication . A decrease in the memory T-cell size is a logical expectation of the anti-proliferative effect exerted by MRV through CCR5 inhibition ( Fig . S7 ) , as antigen-driven proliferation contributes to maintenance of the size of this T-cell subpopulation [8] . It is well known that memory T-cells are a preferential target of HIV-1 replication [43] , and that their decrease may affect the overall viral dynamics in vivo . In this regard , the MRV-induced decrease in the memory T-cell size is not only unlikely to be dangerous but , rather , likely to be beneficial . This hypothesis is supported by results showing that the pool of TCM cells is a correlate of anergy towards the viral antigens in Macaca mulatta but not in Cercocebus atys , which is naturally resistant to CD4+ T-cell loss and full-blown AIDS [44] . In addition , the results obtained with the present macaque model suggest that a short cycle of H-iART could be used for improving the efficacy of our previous anti-reservoir treatment based on auranofin and strengthen the idea that an arrest in disease progression may be obtained during the chronic phase of the disease . Although the data on the combined effect of the two subsequent treatment cycles are derived from a limited number of macaques , the result obtained is corroborated by the fact that no similar trend was observed in the same animals prior to starting therapy [5] , [7] or in historical controls that had not received H-iART at rebound [7] . Of note , although certain major histocompatibility complex ( MHC ) class I alleles , including Mamu-A*01 and Mamu-B17* are associated with slow disease progression in SIV infected macaques [45] , [46] , independently , the presence of these alleles is not predictive for disease outcome [47] , and none of our macaques presented the protective alleles in association ( Table S1 ) . Instead , P177 , which , following our therapies , remarkably controlled viral load , presented the HLA Mamu-B*01 allele , that is associated with aggressive simian lentivirus infection [48] . In line with this genotype , P177 showed a significant immune deterioration before our treatments were initiated ( Fig . 8C ) . Finally , recent analyses [reviewed in 49] re-evaluated the necessity of wide numbers of subjects as a support for breakthrough findings , such as , in this case , the obtainment of a condition close to a persistent suppression of viremia in the absence of ART . If the results of the present study should prove reproducible in humans , H-iART could represent a useful tool for improving the viro-immunological conditions of HIV-infected individuals and a useful addition to experimental anti-reservoir strategies . | Novel research aimed at finding a cure for AIDS requires animal models responding to human antiretroviral drugs . However , there have been few antiretrovirals cross-active against the simian viruses . In this study , we expanded the arsenal of drugs active against the simian retrovirus SIVmac251 and showed that this virus is inhibited by the protease inhibitor , darunavir , and the CCR5 blocker , maraviroc . Administration of these two drugs in combination with the reverse transcriptase inhibitors , tenofovir and emtricitabine , and the integrase inhibitor , raltegravir , resulted in prolonged plasma viral loads below assay detection limits , and , surprisingly , restricted the viral reservoir , a marker of which is viral DNA . We then decided to employ this multidrug regimen ( termed “highly intensified ART” ) in order to increase the potency of a previous strategy based on the gold drug auranofin , which recently proved able to restrict the viral reservoir in vivo . A short course of highly intensified ART following the previous treatment resulted , upon therapy suspension , in a remarkably spontaneous control of the infection , that may pave the way to a persistent suppression of viremia in the absence of ART . These results corroborate the robustness of the macaque AIDS model as a vanguard for potentially future treatments for HIV in humans . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"immunopathology",
"animal",
"models",
"of",
"infection",
"antivirals",
"viral",
"persistence",
"and",
"latency",
"virology",
"viral",
"nucleic",
"acid",
"immunology",
"biology",
"microbiology",
"viral",
"replication"
]
| 2012 | A Highly Intensified ART Regimen Induces Long-Term Viral Suppression and Restriction of the Viral Reservoir in a Simian AIDS Model |
Trypanosoma cruzi parasites are the causative agents of Chagas disease . These parasites infect cardiac and gastrointestinal tissues , leading to local inflammation and tissue damage . Digestive Chagas disease is associated with perturbations in food absorption , intestinal traffic and defecation . However , the impact of T . cruzi infection on the gut microbiota and metabolome have yet to be characterized . In this study , we applied mass spectrometry-based metabolomics and 16S rRNA sequencing to profile infection-associated alterations in fecal bacterial composition and fecal metabolome through the acute-stage and into the chronic stage of infection , in a murine model of Chagas disease . We observed joint microbial and chemical perturbations associated with T . cruzi infection . These included alterations in conjugated linoleic acid ( CLA ) derivatives and in specific members of families Ruminococcaceae and Lachnospiraceae , as well as alterations in secondary bile acids and members of order Clostridiales . These results highlight the importance of multi-‘omics’ and poly-microbial studies in understanding parasitic diseases in general , and Chagas disease in particular .
Trypanosoma cruzi are protozoan parasites endemic to Central and South America . They cause a range of cardiac and gastrointestinal manifestations collectively known as Chagas disease . With increasing travel and immigration , infected individuals are also now found worldwide . Six to seven million people are T . cruzi-positive , thirty to forty percent of which will develop symptomatic disease decades after their initial exposure to the parasite . Cardiac symptoms are the most common; these include conduction abnormalities , arrhythmias , aneurysms , and heart failure leading to death . Clinically apparent gastrointestinal Chagas disease is less prevalent; gastrointestinal Chagas disease is associated with enlargement of the esophagus and/or colon ( megaesophagus , megacolon ) , leading to pain , dysphagia , altered intestinal transit , altered nutrient intake , and constipation [1] . Research on cardiac Chagas disease progression has focused mainly on heart tissue . However , studies in murine models using luminescent T . cruzi cell lines showed recirculation of parasites from gastrointestinal tissues to the heart and propose a model in which gastrointestinal sites function as a reservoir for parasites to re-invade heart tissue and cause cardiac damage [2 , 3] . These suggest an important role for intestinal T . cruzi infection beyond megasyndrome pathogenesis . Gastrointestinal sites may also be a major source of parasites during post-treatment recrudescence [4] . Gastrointestinal Chagas disease has a strong geographic association; most cases represent infections acquired in Bolivia , Brazil , Argentina and Chile . Disease tropism has been strongly tied to T . cruzi strain [5] , but diet may also play a role [6] . T . cruzi infection is associated with parasite dose-dependent recruitment of inflammatory cells to the colon and colon damage [7] , all of which could perturb the intestinal microbiota . Conflicting results comparing infection outcomes in germ-free and conventional mice have been reported , with one study showing similar survival [8] , and another study showing differential survival [9] . The impact of T . cruzi infection on the gut microbiome and metabolome composition in immunocompetent animals has yet to be assessed . Such a system is more representative of human infection than germ-free models that show significant immunological defects [10] . This work applies 16S amplicon sequencing and mass spectrometry-based metabolomics on fecal pellets to characterize the functional bacterial changes associated with T . cruzi infection , in an immunocompetent murine model of Chagas disease . This joint approach enabled us to identify correlated microbiome and metabolome changes , and paves the way for further investigation of the T . cruzi-microbiota interaction in the context of Chagas disease pathogenesis .
All vertebrate animal studies were performed in accordance with the USDA Animal Welfare Act and the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . Euthanasia was performed by isoflurane overdose followed by cervical dislocation . The protocol was approved by the University of California San Diego Institutional Animal Care and Use Committee ( protocol S14187 ) . Male C3H/HeJ mice were purchased from Jackson laboratories and allowed to acclimatize to our vivarium for 2 weeks before the start of experimentation . At day 0 , mice were infected by intraperitoneal injection with 1 , 000 red-shifted luciferase-expressing T . cruzi strain CL Brener culture-derived trypomastigotes [2] ( 20 mice across four cages ) or left uninfected ( injected with DMEM media only , 20 mice divided in four cages ) , and initial fecal pellets collected . Parasite burden was measured bi-weekly during the acute stage of infection by bioluminescence imaging following D-luciferin injection using an In vivo Imaging System ( IVIS ) Lumina LT Series III ( Perkin Elmer ) . Total body luminescence , cardiac region luminescence , and abdominal luminescence were determined using Living Image 4 . 5 software . Fecal pellets were collected by monitoring the mice until they defecated naturally , at which point the freshly excreted pellets were immediately collected and snap-frozen in liquid nitrogen . Fecal pellets were collected bi-weekly in the acute stage of disease; imaging and fecal collection were performed every 2–3 weeks during the chronic stage of disease . Each time point was analyzed individually; no samples were pooled . No visual changes were observed at any time point for fecal pellets from infected mice compared to fecal pellets from uninfected mice . Infected mice showed no overt disease symptoms except slight decrease in weight at the last two collection timepoints ( days 64 and 90 , p<0 . 05 , Mann-Whitney , FDR-corrected ) ( S1A Fig ) , although four mice were found dead over the course of the experiment ( days 20 , 63 , 79 and 90 post-infection ) ( S1B Fig ) . Hematoxylin-eosin ( H&E ) staining of colon samples did not show any apparent tissue damage or inflammatory infiltrate in infected mice compared to uninfected mice ( S1D Fig ) . However , parasite distribution through the gastrointestinal tract is highly localized during chronic stage of infection with luminescent CL Brener [2] , and we cannot rule out the possibility that other colon regions were altered by infection . Weighed fecal pellets were homogenized in 50% methanol spiked with 2 μM sulfachloropyridazine using a Qiagen TissueLyzer at 25 Hz for 5 min [11] , at a constant concentration of 50 mg feces / 1000 μL of extraction solvent , followed by overnight incubation at 4°C . Samples were then centrifuged at 16 , 000g for 10 min . Equal volumes of centrifugation supernatant were dried in a vacuum concentrator and frozen at -80°C . For LC-MS/MS analysis , samples were resuspended in 50% methanol spiked with 2 μM sulfadimethoxine and analyzed on a Maxis Impact HD QTOF mass spectrometer ( Bruker Daltonics ) coupled to an UltiMate 3000 UHPLC system ( Thermo Scientific ) . A given infected or uninfected mouse was randomly assigned to one of eight 96 well plates , alternating infected and uninfected samples . Time-course samples were plated left to right in the 96 well plates , while run order was top to bottom . Controls included blanks ( resuspension solvent ) and pooled QC controls every 16 samples , and a standard mix of six compounds ( sulfamethazine , sulfamethizide , sulfachloropyridazine , sulfadimethoxine , amitryptiline and coumarin ) with known retention time at the beginning of the run and between each plate , to monitor for retention time shifts . Liquid chromatography separation was performed on a 1 . 7 μm C18 ( 50 × 2 . 1 mm ) UHPLC column ( Phenomenex ) heated to 40°C , with water + 0 . 1% formic acid as mobile phase A and acetonitrile + 0 . 1% formic acid as mobile phase B , at a constant flow rate of 0 . 5 mL/min . The LC gradient was: 0–1 min , 5% B; 1–9 min linear ramp up to 100% B; 9–11 min hold at 100% B; 11–11 . 5 min ramp down to 5% B; 11 . 5–12 . 5 min hold at 5% B . Ions were generated by electrospray ionization and MS spectra acquired in positive ion mode with the following instrument parameters: nebulizer gas pressure , 2 Bar; Capillary voltage , 3 , 500 V; ion source temperature , 200°C; dry gas flow , 9 . 0 L/min; spectra rate acquisition , 3 spectra/s . MS/MS data was collected by fragmentation of the five most intense ions , in mass range 50–1 , 500 m/z , with active exclusion after 2 spectra and release after 30s . Mass ranges representing common contaminants and the lock masses were also excluded ( exclusion list 144 . 49–145 . 49 , 621 . 00–624 . 10 , 643 . 80–646 . 00 , 659 . 78–662 . 00 , 921 . 00–925 . 00 , 943 . 80–946 . 00 , 959 . 80–962 . 00 ) . Ramped collision-induced dissociation energy parameters ranged from 10–50 eV . Daily calibration was performed with ESI-L Low Concentration Tuning Mix ( Agilent Technologies ) . Hexakis ( 1H , 1H , 3H-tetrafluoropropoxy ) phosphazene ( Synquest Laboratories ) , m/z 922 . 009798 , was present throughout the run and used as internal calibrant ( lock mass ) . LC-MS/MS raw data files were lock mass-corrected and converted to mzxml format using Compass Data analysis software ( Bruker Daltonics ) . MS1 feature identification was performed using an OpenMS-based [12] workflow ( Optimus version 1 . 1 . 0 https://github . com/alexandrovteam/Optimus , see S1 Table for parameters ) , restricting to features with MS2 data available . Feature abundance was normalized to the sulfachloropyridazine extraction control . Principle coordinates analysis ( PCoA ) was performed on the normalized data with our in-house tool ClusterApp using the Bray-Curtis-Faith dissimilarity metric [13 , 14] , and visualized in EMPeror [15] . Molecular networking was performed on the Global Natural Products Social Molecular Networking platform ( GNPS ) [16] , with the following parameters: parent mass tolerance 0 . 02 Da , MS/MS fragment ion tolerance 0 . 02 Da , cosine score 0 . 6 or greater , at least 4 matched peaks , maximum analog mass shift , 200 Da . Molecular networks and correlation networks were visualized with Cytoscape 3 . 4 . 0 [17] . Most metabolites were identified to levels 2/3 according to the 2007 metabolomics standards initiative ( putatively annotated compounds or compound classes [18] ) . Additional putative annotations were performed using the LIPID MAPS m/z search tool [19] . Linoleic acid/conjugated linoleic acid ( LA/CLA ) were identified with higher confidence by retention time and spectral matching to authentic standards ( Spectrum Chemical/Sigma Aldrich; level one annotation [18] ) . Random forest analysis over 5 , 000 trees was performed in R [20] . DNA extraction , 16S library preparation and sequencing were performed according to standard protocols from the Earth Microbiome project ( http://www . earthmicrobiome . org/protocols-and-standards/ [21] ) . Briefly , DNA extraction was performed using the MO BIO PowerSoil DNA Isolation Kit ( MoBio Laboratories ) . PCR amplification targeting the V4 region of the 16S rRNA bacterial gene was performed with barcoded primers 515F/806R as described in [22] . Equal amounts of amplicons from each sample were pooled in equal concentration and cleaned with the MoBio UltraClean PCR Clean-Up Kit . Library was PhiX-spiked and sequenced on the UC San Diego Institute for Genomic Medicine Illumina MiSeq2000 platform . Raw FASTQ data files were demultiplexed using Qiita ( https://qiita . ucsd . edu , study ID 10767 ) with the following parameters: maximum barcode errors: 1 . 5; sequence maximal ambiguous bases: 0; maximal bad run length: 3; Phred quality threshold: 3 . This resulted in 12 , 307 , 767 high-quality reads with a median of 24 , 578 sequences per non-blank sample . Closed-reference Operational Taxonomic Unit ( OTU ) picking was performed in Qiita with 97% sequence identity using sortmeRNA [23] as the clustering algorithm . Subsequent data analysis was performed using the QIIME1 pipeline [24] , rarefying to 8 , 500 reads per sample . PCoA plots were generated using the weighted UniFrac distance metric [25] and visualized in EMPeror [15] . Random forest analysis over 5 , 000 trees [20] was performed in R using jupyter notebooks [26] . Procrustes analyses [27 , 28] were performed using the QIIME1 [24] scripts beta_diversity . py to generate the weighted UniFrac distance matrix ( 16S data ) or Bray-Curtis-Faith distance matrix ( LC-MS data ) , followed by principal_coordinates . py to perform principal coordinates analysis . PCoA outputs were used as input for transform_coordinate_matrices . py ( Procrustes ) , with 1000 random permutations . The output of this analysis was visualized EMPeror [15] . Groups of bacteria and metabolites correlated with infection status were identified by Weighted Correlation Network Analysis ( WGCNA ) analysis . Average hierarchical clustering using the WGCNA R package in combination with soft-thresholded Pearson correlation was performed to independently cluster highly correlated microbes and metabolites into modules [29] . Data was pre-filtered using the goodSampleGenes function of the WGCNA package to remove metabolites or OTUs with >50% missing values . Remaining outlier samples were removed using the cutreeStatic function , with a minimum size of 10 . Soft thresholding power was determined using the pickSoftThreshold function and set to 4 for metabolites and for 30 microbiome data . Minimum module size was 30 for OTUs and 10 for metabolite features; threshold for merging modules was 0 . 25 . Using this approach , 49 metabolite modules and three microbial modules were obtained . Microbial and chemical modules were independently correlated with parasite burden using Pearson correlation . Since we were interested in identifying the changes in gut ecosystem due to parasite infection specifically , only the modules that were significantly correlated with parasite burden were retained for downstream analysis ( Student asymptotic p-value <0 . 01; positive correlation coefficient ) . This represented nine metabolite modules and one microbial module . We performed pairwise Pearson correlation between these modules , which yielded six positively correlated microbe-metabolite module pairs ( Student asymptotic p-value <0 . 01 ) . Finally , we performed pairwise Pearson correlations between microbial and chemical components of these strongly correlated module pairs to obtain candidate microbial-metabolite associations relevant to T . cruzi infection in mice ( positive correlation , p<0 . 05 ) . These metabolites were then compared with molecular networking results to identify common members of chemical families . Correlations between the members of these families and bacterial OTUs were plotted using Cytoscape 3 . 4 . 0 [17] .
Gut microenvironments are influenced by dietary components and by bacterial and host metabolism , all of which could affect parasite nutritional availability and antiparasitic immune responses [31] . Likewise , chemical changes in the gut microenvironment would influence bacterial growth and composition [32] . We therefore investigated the integration between the microbial and chemical changes we observed during experimental T . cruzi infection by performing Procrustes analysis [27 , 28] . Separation between infected and uninfected fecal microbiome and metabolome samples jointly was observed at days 21 and 90 post-infection but not at day 0 ( Fig 2A , S2 Table ) . To determine the nature of these joint changes , we performed weighted gene co-expression network analysis ( WGCNA ) [29 , 33] on microbial and chemical data . Metabolites and microbes were individually clustered into modules , and microbial and chemical modules correlated with abdominal parasite burden ( abdominal luminescence ) were identified ( significance cutoffs: Student asymptotic p-value <0 . 01; correlation coefficients > 0 ) . Only one module of 1954 bacterial OTUs ( out of three bacterial modules ) was correlated with parasite burden ( Pearson correlation coefficient , 0 . 19; Student asymptotic p-value , 3e-05 ) , S3 Fig ) . Nine metabolite modules ( out of 49 metabolite modules ) were correlated with parasite burden ( Student asymptotic p-value <0 . 01 , Pearson correlation coefficient 0 . 13–0 . 33 , S4 Fig ) . Pair-wise correlation was then performed between these burden-correlated microbial and chemical modules , six of which showed statistically significant correlation ( Student asymptotic p-value<0 . 01 , Pearson correlation coefficient 0 . 13–0 . 52 , S5 Fig ) . Metabolite feature to OTU pair-wise comparisons were then performed within each metabolite-microbe module pair ( cutoffs: positive correlation , p<0 . 05 ) . Within these six correlated module pairs , almost all the metabolite features positively correlated with parasite burden were from different molecular subnetworks , suggesting that they are part of different chemical families [16] . Strikingly however , eleven metabolite features from the most strongly correlated metabolite module ( Pearson correlation coefficient 0 . 33 , p-value , 3e-13 ) were from the same molecular subnetwork of linoleic acid derivatives ( Table 1 , S6A and S7 Figs ) . Dietary linoleic acid ( LA ) is modified in the gut environment by bacteria from the genera Lactobacillus , Bifidobacterium and Enterococcus into conjugated linoleic acid ( CLA ) and further derivatives [34 , 35] . Conjugated linoleic acid can also be taken up in the diet and further modified in the gastrointestinal tract [34–36] . m/z 281 . 251 RT 485s was confirmed as LA or CLA by retention time and accurate mass matching to authentic LA/CLA standards ( level one annotation according to the 2007 metabolomics standards initiative [18]; S6B Fig ) . Our chromatography conditions do not enable clear differentiation of LA and CLA . Specific members of the orders Bacteroidales and Clostridiales , including members of the families Ruminococcaceae and Lachnospiraceae can hydrogenate CLA [37 , 38] , and indeed we observed the strongest correlation ( Pearson correlation coefficient >0 . 4 ) between members of the order Clostridiales and m/z 283 . 266 RT 435s , putatively identified as vaccenic acid ( Fig 2B , S3 Table ) . Microbial hydration of linoleic acid by members of the Pediococcus and Lactobacillus genera has also been reported [35 , 37 , 39–41] . Molecular networking indicates that m/z 299 . 261 RT 336s and m/z 317 . 271 RT 336s could represent single and double hydration products of linoleic or conjugated linoleic acid; they were correlated with specific Ruminococcaceae and Lachnospiraceae family members ( Fig 2B , S3 Table ) . CLA absorption in the colon is limited; bacterial metabolites of linoleic acid therefore primarily exert their effects locally [42] . Linoleic acid metabolism products alter gut inflammatory responses , by promoting regulatory T cell recruitment [43] , decreasing TNF receptor expression [39] and TNFα production [44] , and increasing the anti-inflammatory cytokine TGFβ in the colon [44] . These metabolites could therefore promote gut microenvironments favoring T . cruzi persistence and gastrointestinal reservoir function . An additional group of 5 co-modulated features networked with cholic acid ( Table 2 , S8 Fig ) . m/z 357 . 281 RT 337s , m/z 357 . 281 RT 371s and m/z 375 . 291 RT 393s are identified as different close isomers or adducts of deoxycholic acid ( level two annotation according to the 2007 metabolomics standards initiative [18] ) . Host-produced primary bile acids such as cholic acid are conjugated to taurine or glycine in the liver . Further modifications of primary bile salts are specifically performed in the gastrointestinal environment: members of the gut microbiota deconjugate primary bile salts and remove the 7-hydroxy group to form secondary bile acids such as deoxycholic acid . Bacteroides , Bifidobacterium , Clostridium , Lactobacillus and Listeria genera deconjugate bile acids , which are then dehydroxylated by Clostridium and Eubacterium genera [45] . Indeed , one member of the Clostridium genus , Clostridium celatum ( OTU ID 4315688 ) was correlated with m/z 357 . 281 RT 337s ( Pearson correlation coefficient 0 . 21028 , p-value = 0 . 00000426 ) , and weakly correlated with m/z 357 . 281 RT 371s and m/z 358 . 285 RT 371s ( respective correlation coefficient , 0 . 09079 and 0 . 09770; respective p-values , 0 . 049161208 and 0 . 034207271 ) ( S4 Table ) . Likewise , two members of the Bifidobacterium genus were correlated with m/z 357 . 281 RT 371s and m/z 375 . 291 RT 393s , and five members of the Lactobacillus genus were correlated with m/z 357 . 281 RT 371s , m/z 358 . 285 RT 371s and m/z 375 . 291 RT 393s ( S4 Table ) . Further modifications can be performed by these genera and by Escherichia , Egghertella , Fusobacterium , Peptococcus , Peptostreptococcus , Ruminococcus genera [45] , several of which were also correlated with our infection-modulated secondary bile acids ( Fig 2C ) . The OTUs most strongly correlated with these secondary bile acids in our experiment ( correlation coefficient >0 . 4 ) were also members of the order Clostridiales , either from the genus Oscillospira or from unidentified genera ( S4 Table ) . Bile acid metabolism by the gut microbiota has been tied to local colon inflammation and general health [45] , all of which could affect Chagas disease pathogenesis . Several of these microbiome changes have been associated with other gastrointestinal diseases . Lactobacillus genus in particular is increased in obese individuals , while genus Bifidobacterium is decreased [46] . Members of the Lactobacillus genus and some Bifidobacterium species are increased in ileal Crohn’s disease , while members of order Clostridiales and family Lachnospiraceae are decreased [46] . Large-scale perturbations are also observed in these diseases , such as for example a trend for increased Firmicutes to Bacteroidetes ratio in obese individuals compared to lean individuals [46] . The observed microbial and metabolic perturbations in T . cruzi-infected animals may be a consequence of parasite-mediated modulations of local gastrointestinal microenvironments , such as nutrient depletion , or an off-target effect of anti-parasitic immune responses . Parasite control is associated with reactive oxygen and nitrogen species [47] , which are known to affect the gut microbiome composition by killing bacterial species sensitive to oxidative stress while promoting the growth of species that use nitrate as a terminal electron acceptor for respiration [48] . Significant bacterial and metabolic changes become apparent by day 14 post-infection ( Figs 1D and 1E and S2 ) , which coincides with induction of adaptive immune responses to T . cruzi [49] , suggesting an immune-mediated role in this disruption . Given the anti-inflammatory roles of the hydrated linoleic acid metabolites we found altered by infection [39 , 43 , 44] , the gut microbiome and metabolome changes we observed may be promoting long-term parasite gastrointestinal persistence and enabling the gastrointestinal tract to serve as a parasite reservoir . Microbiota perturbation may also contribute to the nutrient malabsorption and constipation observed in megasyndromes [50] . Modulating the infection-associated changes in the gut microbiome and its metabolism may prove to be an effective way to mitigate disease symptoms , nifurtimox gastrointestinal side effects or prevent parasite dissemination from the gastrointestinal tract to the heart . Modifying the levels of anti-inflammatory conjugated linoleic acid metabolites may be particularly useful in this context . Finally , although production of bile acid metabolites is performed in the gut environment by the local microbiota , these metabolites can be re-absorbed and circulate throughout the body , with far-ranging effects [51] . Bile acid metabolites may therefore also affect cardiac Chagas disease pathogenesis . Future work will directly investigate these possibilities , by testing whether the gut microbiome perturbations and the metabolites identified in this study are associated with Chagas disease severity , and assessing whether microbiome perturbation affects Chagas disease progression .
Research on Chagas disease pathogenesis has focused on the interaction between the mammalian host and the parasite . Our results indicate that infection modulates the fecal microbiome , suggesting that host-microbe interaction research in the context of Chagas disease should also include the microbiota and not just T . cruzi . By integrating microbiome with metabolome data , we show that these microbial alterations are associated with functional changes in the gut chemical environment that could be affecting host inflammatory responses . These results support additional investigation into the T . cruzi-microbiota connection and into the role of the microbiota in Chagas disease pathogenesis . Given new evidence on the role of gastrointestinal persistence in parasite recrudescence [4] , and our limited understanding of gastrointestinal Chagas disease compared to cardiac Chagas disease , such studies are essential to identify treatment strategies able to achieve sterile cure . Microbiota- and microbial metabolism-modulating therapies are now actively being developed for other cardiovascular diseases [52 , 53] . Our results demonstrate that such approaches are likely to be beneficial in cardiovascular Chagas disease . Modulation of the gut microbiota or its metabolism may also be a promising strategy for megasyndrome patient management , or to slow progression of asymptomatic individuals to symptomatic disease . | Host-parasite interactions are usually studied as a binary system , without considering the role of the host microbiota . This work integrates microbiome research into the study of gastrointestinal Chagas disease . We show that T . cruzi infection perturbs the fecal microbiome and metabolome , indicating functional changes affecting the gastrointestinal lumen . Our results support further investigation into the role of the microbiota-parasite interaction in gastrointestinal Chagas disease pathogenesis . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"and",
"discussion",
"Conclusions"
]
| [
"medicine",
"and",
"health",
"sciences",
"microbiome",
"tropical",
"diseases",
"microbiology",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"metabolomics",
"protozoans",
"metabolites",
"gastroenterology",
"and",
"hepatology",
"neglected",
"tropical",
"diseases",
"microbial",
"genomics",
"parasitic",
"intestinal",
"diseases",
"medical",
"microbiology",
"protozoan",
"infections",
"gastrointestinal",
"infections",
"biochemistry",
"trypanosoma",
"cruzi",
"trypanosoma",
"chagas",
"disease",
"eukaryota",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"metabolism",
"organisms"
]
| 2018 | Experimental Chagas disease-induced perturbations of the fecal microbiome and metabolome |
Hepatitis C virus ( HCV ) , a major cause of chronic liver disease in humans , is the focus of intense research efforts worldwide . Yet structural data on the viral envelope glycoproteins E1 and E2 are scarce , in spite of their essential role in the viral life cycle . To obtain more information , we developed an efficient production system of recombinant E2 ectodomain ( E2e ) , truncated immediately upstream its trans-membrane ( TM ) region , using Drosophila melanogaster cells . This system yields a majority of monomeric protein , which can be readily separated chromatographically from contaminating disulfide-linked aggregates . The isolated monomeric E2e reacts with a number of conformation-sensitive monoclonal antibodies , binds the soluble CD81 large external loop and efficiently inhibits infection of Huh7 . 5 cells by infectious HCV particles ( HCVcc ) in a dose-dependent manner , suggesting that it adopts a native conformation . These properties of E2e led us to experimentally determine the connectivity of its 9 disulfide bonds , which are strictly conserved across HCV genotypes . Furthermore , circular dichroism combined with infrared spectroscopy analyses revealed the secondary structure contents of E2e , indicating in particular about 28% β-sheet , in agreement with the consensus secondary structure predictions . The disulfide connectivity pattern , together with data on the CD81 binding site and reported E2 deletion mutants , enabled the threading of the E2e polypeptide chain onto the structural template of class II fusion proteins of related flavi- and alphaviruses . The resulting model of the tertiary organization of E2 gives key information on the antigenicity determinants of the virus , maps the receptor binding site to the interface of domains I and III , and provides insight into the nature of a putative fusogenic conformational change .
The hepatitis C virus ( HCV ) is a major cause of chronic liver disease worldwide , leading to cirrhosis and hepatocellular carcinoma [1] . In spite of being the focus of intense research efforts , no vaccine is available against HCV , and current therapeutic treatments have limited efficacy and significant side effects [2] . HCV belongs to the Flaviviridae family of enveloped , positive-strand RNA viruses [3] . Structural studies on this virus are difficult , in part because it propagates poorly in cell culture , and particles isolated from infected patients are heterogeneous and not amenable to a detailed structural characterization . Little structural information is available on the envelope proteins , which are heavily glycosylated , display hypervariable loops , and are stabilized by numerous disulfide bridges [4] . The folding kinetics of these proteins are slow , requiring several hours for completion of a complex process involving various ER chaperones of the infected cell [5] . These properties make their recombinant production - in a native conformation and in sufficient amounts for structural studies - a difficult endeavor . Yet structural information on the HCV envelope proteins would be extremely valuable , given that they carry the main antigenic determinants of the virus and play an essential role in cell entry by binding to specific receptors and inducing membrane fusion . HCV has indeed been shown to depend on a number of cellular molecules for entry , including CD81 [6] and the tight junction transmembrane proteins claudin 1 , 6 , 9 and occludin [7]–[9] , as well as the scavenger receptor B1 ( SR-B1 ) [10] . The LDL receptor also plays a role in HCV uptake , in line with the observation that HCV particles in infected plasma are associated with LDL species [11] . A direct interaction of the HCV envelope protein E2 with CD81 and SR-B1 has been demonstrated , and these interactions were shown to be necessary but not sufficient for cell entry . The mode of interaction of HCV with the claudins and occludin is not understood at present . The HCV genome codes for a single polyprotein precursor about 3000 amino acids long , spanning the ER membrane multiple times . It contains , sequentially , the viral proteins in the order Nter-C-E1-E2-p7-NS2-NS3-NS4A/B-NS5A/B-Cter . The N-terminal 1/4th of the precursor corresponds to the structural proteins C ( Core ) , E1 and E2 ( envelope proteins 1 and 2 ) and p7 , which functions as a proton channel . The remainder of the polyprotein contains the non-structural ( NS ) proteins , which have enzymatic and other activities that are necessary for virus replication . The mature viral proteins are generated by proteolytic processing of the precursor by cellular and viral proteases [3] . In particular , the envelope proteins are generated by host-cell signalases . E1 and E2 are type 1 trans-membrane ( TM ) proteins with a large N-terminal ectodomain and almost no cytoplasmic tail . In the best characterized HCV strain H77 , E1 and E2 are 192 and 366 amino acids long and contain 6 and 11 potential N-linked glycosylation sites , respectively . Biochemical studies have shown that E1 and E2 fold as a heterodimer , which is found at the surface of viral particles and is thought to be the functional glycoprotein form [4] . There are currently 6 identified HCV genotypes further divided into several subtypes [12] . The amino acid sequence identity between envelope proteins from different genotypes is about 68% for the most distant genotypes . E2 has been shown to contain 3 hypervariable regions that can be deleted without affecting the overall fold of the protein , as assayed by binding to conformation-sensitive mAbs and CD81 [13]–[15] . The genomic organization of HCV is characteristic of all members of the Flaviviridae family [3] . In particular , the envelope proteins are present in tandem within the polyprotein precursor . This arrangement of the structural part of the genome is characteristic of viruses encoding class II fusion proteins , reviewed in [16] . These proteins have been extensively characterized , structurally and biochemically , for viruses in the flavivirus genus within the Flaviviridae family [17] . Class II proteins have a common tertiary structure , which has also been observed in the fusion protein of Semliki Forest virus ( SFV ) , an alphavirus belonging to a separate family of enveloped , positive-strand RNA viruses , the Togaviridae [18] . Togaviridae and Flaviviridae display the same gene order in the structural part of their genomes . There is no amino acid sequence similarity in the alpha- and flavivirus fusion proteins , however , and in spite of sharing a common fold , they are stabilized by a different pattern of disulfide bonds . Viruses within the Flaviviridae families have no sequence similarity across the various genera either , and the fusion proteins from each genus also appear to have their own characteristic pattern of disulfide bonds . Yet the conservation of the class II fold across viral families in the absence of sequence conservation strongly suggests that it is also conserved across the different genera within the respective families . A further feature of class II viral fusion proteins is that they fold as a heterodimer with the upstream glycoprotein in the polyprotein precursor . This heterodimer later dissociates to drive membrane fusion upon interactions with the host cell . The first glycoprotein in the tandem thus acts as chaperone for folding the second one , which has the membrane fusion role . The chaperone function was experimentally demonstrated for the flavivirus prM [19] and the alphavirus p62 [20] glycoproteins , which precede the fusion proteins E and E1 , respectively , in the precursor polyprotein . The effect on folding appears to be reciprocal , since both p62 and prM also adopt their native conformation only in presence of the respective accompanying fusion protein ( unpublished observations ) . Importantly , heterodimerization upon folding has also been characterized for viruses belonging to other genera in the two families , and in particular for HCV [5] , [21] . Flavivirus E and alphavirus E1 change into a homotrimer upon interaction with lipids in the acidic environment of a target cell endosome , in a process that drives fusion of the viral and endosomal membrane and results in infection of the cell [22] , [23] . This process involves homodimer ( E-E , flavivirus ) or heterodimer ( E2-E1 , alphavirus ) dissociation , followed by homotrimerization of E ( flavivirus ) or E1 ( alphavirus ) upon binding to lipids . The tertiary structure of class II viral fusion proteins contains predominantly β-sheets segregated into three distinct domains arranged linearly , resulting in a rod-like molecule . The central domain 1 ( DI ) is a β-sandwich with two long insertions in loops connecting adjacent β-strands . These insertions form an elongated “fusion” domain ( DII ) , carrying the “fusion loop” in the first of the two insertions , at the distal end of the rod . The fusion loop is a segment of the polypeptide chain that inserts into the target membrane in the first step of membrane fusion . At its C-terminal end , DI is connected via a flexible linker to domain 3 ( DIII ) , which is located at the opposite side with respect to DII , giving rise to the linear organization of the molecule . DIII plays an important role in the fusogenic conformational change , during which it relocates to the side of the molecule , resulting in the characteristic “hairpin” conformation of the protein , which drives membrane fusion . This relocation involves a considerable stretching of the segment connecting DI to DIII , the region that changes most dramatically in conformation during the fusogenic transition ( reviewed in [16] ) . Although there is no direct experimental evidence demonstrating the role of E2 as the HCV fusion protein , the compelling similarities to viruses with class II fusion proteins suggest that membrane fusion is at least one of its biological roles . It is worth noting , however , that while totally unrelated viruses can have structurally homologous fusion proteins ( for example , rhabdoviruses , herpesviruses and baculoviruses , reviewed in [24] ) , related viruses can use non-homologous fusion proteins , as is the case with paramyxoviruses and rhabdoviruses , which belong to the Mononegavirales order ( reviewed in [25] ) . Yet the fact that viruses belonging to different genera in the Flaviviridae and Togaviridae families display a genomic arrangement that is the signature of class II fusion proteins , together with the additional common features outlined above , makes it very likely that they code for envelope glycoproteins that are at least distantly related to class II proteins . A model for E2 has actually been proposed based on the structure of the flavivirus E protein homodimer [26] , although no evidence is available for homodimerization of HCV E2 , which forms a heterodimer with E1 in infectious virions [4] . More importantly , this model does not take into account the location of the strictly conserved cysteine residues forming 9 disulfide bonds [27] . This model also lacks the third domain , which is important in the fusogenic transition . Moreover , it was also proposed that the membrane fusion function could be carried by HCV glycoprotein E1 ( i . e . , the first glycoprotein in the tandem ) [28] , [29] , in spite of the similarities with flavi- and alphaviruses discussed above , and in the absence of experimental support . Furthermore , a bioinformatics model for HCV E1 as a truncated class II protein was reported [30] , postulating that E1 has the fold of DII of an alpha- or flavivirus fusion protein , but neglecting the fact that in class II proteins , DII works in conjunction with the other two domains covalently linked within the polypeptide to induce membrane fusion . The corollary is that controversial hypotheses have been reported concerning the identity of the HCV fusion protein . It is therefore important to stress that the structural studies performed over the years on viral membrane fusion proteins strongly suggest that most animal enveloped viruses encode fusion proteins belonging to one of the three currently characterized structural classes [31] . It is thus highly unlikely that HCV would have acquired a totally novel fusion machinery ( for instance , one in which E1 would be the membrane fusion protein ) , especially when taking into account the similarities to class II proteins presented above . In order to bring more insight into the tertiary structure of HCV E2 , we report here the experimental identification of the connectivity of the 9 disulfide bonds present in the recombinant E2 ectodomain ( E2e ) generated by expression of the E1-E2ΔTM portion of the HCV genome in Drosophila S2 cells ( Fig . 1A ) . The absence of the transmembrane ( TM ) segment in E2 leads to secretion of its ectodomain after folding in the presence of E1 . This approach is based on previous results leading to production of recombinant dengue virus E protein in the presence of its viral chaperone prM [32] . We tested the conformation of recombinant HCV E2e biochemically and functionally , showing that it reacts with conformation-sensitive antibodies and inhibits infection of Huh7 . 5 cells by infectious HCV particles ( HCVcc ) in a dose-dependent manner . Knowledge of the disulfide bonds , along with functional data on deletion mutants [14] and CD81 binding [26] , [33] , [34] , together with secondary structure predictions , provide sufficient constraints to reconstitute the tertiary organization of the molecule . This information allowed the threading of the E2e polypeptide chain onto a class II template by matching the predicted β-strands . The resulting model reveals the distribution of the amino acids of HCV E2 among the different domains , maps the CD81 binding site to the DI/DIII interface , and highlights a strictly conserved segment of the polypeptide chain as a strong candidate for the HCV fusion loop .
We generated stable Drosophila S2 cell-lines expressing the E1-E2ΔTM segment of the precursor polyprotein ( Fig . 1A ) from 9 isolates spanning all 6 HCV genotypes and 4 subtypes ( Table 1 ) . In order to ensure that the recombinant E2 proteins were functional , we selected isolates previously tested for entry of retroviral particles pseudotyped with HCV glycoproteins ( HCVpp ) with the corresponding sequences [35] . Induction of expression at high cell density with CdCl2 resulted in accumulation of relatively high levels of secreted E2e in the cell culture medium . We purified the protein to homogeneity from the supernatant ( described in Text S1 ) , with the yields listed in Table 1 . E2e from the different isolates behaved similarly , as judged by size exclusion chromatography ( SEC ) followed by SDS-PAGE analysis under reducing and non-reducing conditions and Coomassie blue staining ( Fig . 1 ) . In a typical SEC profile , the majority of the protein elutes at a volume corresponding to a monomer , with additional minor peaks corresponding to disulfide linked dimers and higher multimers , which vary depending on the construct analyzed . The monomeric form was efficiently separated from the other species by pooling the corresponding fractions . Analytical ultracentrifugation and small angle X-ray scattering confirmed the monomeric state of the protein eluted in these fractions ( data not shown ) . Once isolated , E2e from all constructs listed in Table 1 remained monomeric and showed no tendency to associate into disulfide-linked aggregates over time . The construct corresponding to the genotype 2b isolate ( UKN2b_2 . 8 ) reproducibly yielded the highest amounts of purified monomeric protein ( Table 1 ) . The construct from genotype 4 ( UKN4_11 . 1 isolate ) yielded a significant fraction of disulfide-linked aggregates ( Fig . 1 ) , which are likely to correspond to misfolded protein . E2e from the remaining 7 constructs yielded slightly lower yields of purified , monomeric protein than did the genotype 2b construct , the lowest yields being from the gentoype 6 isolate ( Table 1 ) . The SEC profiles from the 7 other constructs were intermediate between the two chromatograms shown in Fig . 1 . Because of the higher production yields , we pursued most of the biochemical characterization using E2e from genotype 2b , to which we will refer to as E2e in the rest of the manuscript , except when explicitly stated . Yet because the best functionally characterized HCV strain is H77 ( genotype 1a ) , we use the amino acid numbering corresponding to the H77 polyprotein throughout the manuscript . Pull-down assays showed that E2e efficiently binds the CD81 large external loop ( LEL ) , as well as conformation-sensitive mAbs CBH-4B and CBH-4D [36] ( Fig . S1A ) . To further confirm that CD81 and the conformation-sensitive mAbs bind stoichiometrically to monomeric E2e , we used SEC to analyze the formation of various E2e/ligand complexes in defined ratios . The resulting chromatograms display a quantitative shift of the peak from monomeric protein to an E2e/mAb complex with a 2∶1 stoichiometry , as expected ( Figs . 2A and S1B ) . The SEC profile displayed in Fig . 2 shows the well-characterized conformation-sensitive mAb H53 that is specific for genotype 1a , whereas Fig . S1B shows the same analysis of E2e from the genotype 2b isolate and the human conformation-sensitive mAb CBH-4D . Similarly , SEC analysis using the Fab fragment of the corresponding mAbs under the same conditions , yielded a 1∶1 E2e/Fab stoichiometry , as expected ( data not shown ) . SEC analysis revealed that E2e also forms a stoichiometric complex with the CD81 LEL ( data not shown ) . The monomeric fraction from all isolates listed in Table 1 yielded similar results - except perhaps for the genotype 6 isolate , which was not tested - strongly suggesting that recombinant E2e adopts a conformation closely resembling that of authentic E2 present on virions . We further tested the ability of E2e to compete with infectious HCV particles for entry receptors , by measuring its ability to inhibit infection of Huh-7 . 5 cells by HCVcc ( Fig . 2B ) . As a control , we tested in parallel the effect of the flavivirus E protein ectodomain ( sE ) from West Nile encephalitis virus ( WNV ) , as well as the ectodomain of pestivirus E2 ( pE2e ) from the bovine viral diarrhea virus ( BVDV ) produced under identical conditions . In contrast to the control proteins , HCV E2e exerted a clear dose-dependent inhibition of the infection . At the lowest concentration tested ( 0 . 05 µM ) , 10% inhibition was observed , which increased with protein concentration to reach 90% inhibition at 2 µM of HCV E2e . This effect is in line with the observation that E2e makes a stoichiometric complex with CD81 , as described above . Computer algorithms for secondary structure prediction using amino acid alignments of E2 from all 6 HCV genotypes predict predominantly β-strands in E2e ( Fig . 3 ) , consistent with the fold of class II fusion proteins . We used recombinant E2e to experimentally analyze its secondary structure composition with two complementary methodologies , circular dichroism ( CD ) , which is sensitive to the presence of α-helices , and Fourier transform infrared ( FTIR ) spectroscopy , which readily detects β-sheets present in a protein . We carried out these tests in parallel with recombinant control class II envelope proteins of known structure available in the laboratory . For the CD measurements , the controls were WNV sE , ( [37] , PDB 2I69 ) and the ectodomain of glycoprotein E1 ( sE1 ) of Chikungunya virus ( CHIKV ) , which displays 62 . 5% amino acid sequence identity with the SFV E1 ectodomain , the crystal structure of which is known ( [18] , PDB 2Ala ) . Unexpectedly , the far-UV spectra of the three proteins exhibited considerable differences ( Fig . 4A ) , the spectrum of HCV E2e being in agreement with a previous study [38] . However , deconvolution to retrieve the percentage of the various secondary-structure elements suggests similar ratios for all three proteins , indicating , in particular , only about 5% α-helices in all three proteins . The strong minimum observed at 203 nm in the spectrum of HCV E2e suggests the presence of natively unfolded regions that are absent in the control proteins . Given that circular dichroism is not the most sensitive method to determine the amount of β-sheet in a protein , we used FTIR spectrometry in a comparative analysis of E2e with a class II protein of known 3D structure . Because the WNV sE was not available at the time of the experiment , we used instead sE from dengue virus serotype 3 ( DV3 ) , for which the crystal structure is also known ( PDB entry 1UZG [39] ) . The high-frequency region of the FTIR spectra of HCV E2e and DV3 sE is displayed in Fig . 4B . As expected , both proteins have their absorption maxima in the amide I band at 1637 and 1640 cm−1 , respectively , close to the 1630 cm−1 value typical for β-sheet containing polypeptides ( reviewed in [40] ) , in agreement with the structure of the flavivirus sE and strongly indicating that HCV E2e also contains predominantly β-sheets . In order to obtain a quantitative measure of the β-sheet content of E2e , we performed a further analysis to more precisely compare the secondary structure content of the two proteins by computing a difference spectrum after normalization to an identical area under the amide I band . The DV3/sE – HCV/E2e difference spectrum showed a positive peak at 1630 cm−1 , as well as a broad negative region ranging from 1645 to 1680 cm−1 ( Fig . 4B ) . The value of the positive peak indicated about 14% higher β-sheet content for DV3 sE , which , when using the value of 42% β-sheet estimated from the DV3 sE crystal structure , gives about 28% β-sheet for HCV E2e . The negative area of the difference FTIR spectrum indicates that the HCV E2e polypeptide displays higher relative amounts of secondary structure other than β-pleated sheet ( random coil , β-turns , 3/10 helices , etc . ) . This difference is likely to reflect the presence of the regions that give rise to the strong minimum at 203 nm in the CD spectrum ( Fig . 4A ) , i . e . , natively unfolded segments of the polypeptide chain . We determined the identity of the disulfide bridges by N-terminal sequencing together with comparative reducing/non-reducing mass spectrometry analyses of peptides obtained by trypsin digestion of E2e . For this purpose we selected E2e of three isolates , UKN2b_2 . 8 , H77 and JFH-1 ( genotypes , 2b , 1a and 2a , respectively ) , which display amino acid sequences with a different pattern of predicted trypsin cleavage sites ( Fig . S2 ) . We fully deglycosylated the protein with PNGase F under denaturing conditions , then digested it with trypsin followed by separation of the resulting peptides by HPLC under reducing or non-reducing conditions . Comparison of the HPLC elution profiles enabled the identification of peaks that were affected by reduction with TCEP ( asterisks in Fig . 5A ) . We analyzed the samples in these peaks by surface-enhanced laser desorption/ionization ( SELDI ) with a time-of-flight ( TOF ) spectrometer ( Table S1 ) , and identified their N-terminal sequence by Edman degradation ( Fig . S3 ) . This procedure allowed the unambiguous experimental identification of 8 out of the 9 disulfide bonds present in the protein , thereby also identifying the 9th by exclusion ( Table 2 ) . This table also shows that 5 disulfides were independently identified in at least two different strains , validating the procedure . A full account of the experiments made to determine the disulfide connectivity is provided as Supplementary Information ( Text S1 ) . The connectivity of the disulfide bonds provides key information on distant segments of the E2 polypeptide chain that come near each other in the folded protein . This knowledge can be used in conjunction with other available data to get a better picture of the tertiary structure of the protein , namely: i ) the observation that E2e is rich in β-sheet and that secondary structure predictions suggest regions with consensus β-strands along its amino acid sequence; ii ) the identity of residues that are far apart in primary structure and that are known to be part of the CD81 binding site; iii ) the postulate that E2 is the HCV fusion protein and therefore has a characteristic 3-domain class II fold , in agreement with the organization of its precursor polyprotein , which also implies iv ) that the third domain ( DIII ) should be connected to DI via a linker that can extend to stabilize a post-fusion trimer . Finally , DIII should be followed by a flexible “stem” region - the presence of which has already been reported for HCV E2 [41] - connecting to the TM segment . Further information comes from the identification of “hypervariable” regions in HCV E2 that can be deleted without affecting the reactivity of the resulting deletion mutant with conformation-sensitive mAbs and with CD81 [14] . In addition , numerous reports have shown that the E2 ectodomain truncated at position 661 , which is in the loop closed by disulfide 9 ( Table 2 ) , also reacts with conformation-sensitive mAbs and CD81 [38] , suggesting that the downstream segment is not part of the structured ectodomain . About one third of the E2e residues are predicted to form β-strands ( Fig . 3 ) , which is in overall agreement with the estimated 28% β-sheet content determined by FTIR spectroscopy . The pattern of predicted β-strands offers the possibility of threading the polypeptide chain along the template provided by the known fold of class II proteins , while simultaneously respecting all of the known constraints derived for HCV E2 by the functional studies discussed above . A useful guide for this analysis is the comparison between predicted and experimentally observed β-strands in the crystal structure of alpha- and flavivirus fusion proteins - for instance , in the alphavirus E1 alignment provided in Fig . 3B . The hallmark of the tertiary structure of class II proteins is the presence of an 8-stranded ( B0 through I0 ) central domain ( or DI ) folded as a β-sandwich with up-and-down topology ( Fig . 6 ) . Two insertions in this domain , in the D0E0 and H0I0 loops , constitute the fusion domain bearing the fusion loop in the distal part of the D0E0 insertion . DI is followed , after strand I0 , by a flexible segment connecting to a third domain ( DIII ) , the relocation of which is important for hairpin formation during the fusogenic conformational rearrangement of class II fusion proteins . Functional studies have shown that deletion of the HVR1 region did not induce a loss of virus infectivity in experimentally infected chimpanzees [42] , indicating that this segment cannot be part of a folded domain . We therefore began the threading process by assigning the 3 consecutive β-strands predicted immediately downstream of the HVR1 ( Fig . 3 , first three red boxes ) to the three conserved strands in the N-terminal part of DI , i . e . B0 , C0 and D0 . These β-strands are followed by a long intervening region that is compatible with the D0E0 insertion of the class II fold . For the assignment of strands E0 and F0 , the available data on the residues involved in CD81 binding ( small blue circles in Fig . 3 ) provide valuable information , since strand E0 must interact with D0 ( see diagram in Fig . 6B ) . Thus , assigning E0 and F0 to the two consecutive strands predicted after residue 525 brings together a patch of residues that are apart in primary structure to the same face of DI , forming the site of interaction with CD81 ( Fig . 6 ) . For the assignment of the remaining β-strands , there is crucial information provided by disulfide 1 . This disulfide bond connects Cys429 , at the end of strand C0 , with Cys552 further downstream , which therefore must be at the same end of the DI β-sandwich . This means that Cys552 must be located either at the G0H0 loop , or at the end of the I0 strand , if the molecule is to have a class II fold ( Fig . 6B ) . However , after strand F0 , there is a long strand predicted to span residues 549–555 ( Fig . 3 ) , which would have Cys552 in the middle . But the comparison of predicted versus experimentally determined β-strands of alphavirus E1 shows that , for several alphaviruses , the region of G0 and H0 is also predicted as a single long strand ( Fig . 3B ) . Indeed , in both alphaviruses and HCV , there is a glycine residue ( Gly 551 in E2e ) forming a tight turn that reverses the chain orientation , going from G0 into H0 ( some alphaviruses have two glycines at this β-turn ) . This shows that the prediction algorithms are not 100% reliable , suggesting that in HCV E2 , Gly551 is at the G0H0 turn , and that Cys552 is the first residue of strand H0 . Indeed , running at the edge of the DI β-sandwich , the sequence of G0 ( as well as the sequence of the alphavirus B0 strand , at the other end of the bottom β-sheet , Fig . 3B ) appears to be less typical than the sequences of internal β-strands in a β-sheet , which are easier to predict by computer algorithms . In addition , the short connections between strands F0 through H0 in both alpha- and flavivirus DI are also consistent with the assignment of H0 to a strand running between residues 552 and 555 in HCV E2 . Having assigned the G0 and H0 strands , additional considerations are necessary to assign strand I0 . In alpha- and flaviviruses , I0 is one of the two central β-strands of the bottom sheet of DI , and is directly followed by the linker connecting to DIII . Because it is the only strand missing to complete the 8-stranded β-sandwich , it can only make disulfide bonds to cysteines located upstream in primary sequence . In HCV , three β-strands are predicted directly downstream to the assigned H0 strand: one around residue 563 , one around 573 , and one around 593 ( Fig . 3 ) . The strand around residue 573 is part of a segment that can be deleted without affecting protein conformation [14] , indicating that it cannot be I0 . The strand around 593 ends at Cys597 , which forms disulfide 7 with Cys620 further downstream . Because class II proteins can have no interdomain disulfides - which would be incompatible with their function - this strand cannot be assigned to I0 either . Indeed , the interleaved nature of disulfide bonds 7 and 8 dictates that none of the strands predicted downstream can be in DI . The only option compatible with a class II fold is , therefore , to assign the strand around residue 563 to I0 . This assignment implies that the long insertion in the H0I0 loop of the alpha- and flavivirus fusion proteins is absent in HCV E2 . This is in line with E2 from HCV and pestiviruses being shorter than the alpha- and flavivirus fusion proteins by about 80–110 amino acids , i . e . , roughly the length of the insertion in the H0I0 loop of the latter . The assignment of the 8 strands in HCV DI also indicates that the linker connecting DI and DIII must be between disulfides 5 and 6 , encompassing the region called IgVR ( “intergenotypic variable region” ) , which can be deleted without affecting protein conformation , at least in the prefusion form of E2 . As discussed above , the segment containing disulfide 9 is likely not to be part of the structured ectodomain , further implying that DIII is comprised between disulfides 6 and 9 , spanning about 70 amino acids . The presence of two long-range disulfide bonds ( disulfides 7 and 8 ) suggests that this region is indeed structured into a separate domain . However , the secondary structure predictions point to only 3 β-strands in this domain , indicating that the Ig-like fold of DIII in alpha- and flaviviruses may not have been maintained in HCV . Moreover , we found no obvious way to propose an Ig-like arrangement of the polypeptide chain in this domain such that it would also satisfy the constraints imposed by disulfides 7 and 8 . The resulting model for the tertiary structure of E2 is presented in Fig . 6A , with the diagram of Fig . 6B higlighting , as a guide , the essential features of the resulting “class II” organization of the protein . The main features of the molecule are the following: This domain has an N-terminal extension in flaviviruses ( which includes β-strand A0 ) with respect to alphaviruses ( see review by [16] ) , and in HCV , the HVR1 also appears to be an N-terminal extension . DI contains disulfides 1 and 5 , both at the DII distal end of the DI β-sandwich; i . e . , at its DIII interacting end . Disulfide 5 connects two consecutive cysteines into a short loop at the end of strand I0 . The C0D0E0F0 β-sheet ( or “top” sheet ) contains most of the determinants of CD81 binding ( blue circles in Fig . 6A ) , and 5 of the 11 N-linked glycosylation sites of E2 ( numbered 1 , 2 , 3 , 6 and 7 , Figs . 3 and 6A ) . In contrast , the B0I0H0G0 β-sheet ( or “bottom” sheet ) has only site 8 ( Asn 556 , Fig . 6A ) , located in the H0I0 loop , at the site of the long insertion in alpha- and flavivirus fusion proteins ( yellow dotted line , Fig . 6B ) . The presence of an insertion in the other class II proteins suggests that there is space at this end of the barrel for a glycan chain attached to Asn556 . Importantly , glycan 8 was shown to be essential for the correct folding of E2 , in line with the key location in the H0I0 loop in the bottom sheet . Overall , the distribution of glycans on HCV DI is compatible with the experimentally determined orientation of flavivirus E and alphavirus E1 at the virion surface , with the bottom sheet facing the viral membrane . This pattern provides additional evidence validating our assignment of the DI β-strands . This domain has two predicted glycosylation sites and three disulfide bonds ( 2 , 3 and 4 ) , all connecting consecutive cysteine residues very close in primary structure . In the alpha- and flavivirus counterparts , the two insertions forming DII are quite intertwined , and the second one ( the H0I0 insertion ) acts as a scaffold supporting the D0E0 insertion bearing the fusion loop at the DI-distal end . In HCV E2 , the absence of the second insertion makes DII much smaller , and apparently also results in a more flexible , or disordered domain , as suggested by the absence of long-range disulfide bonds and the fact that the whole area between disulfides 2 and 3 can be eliminated without affecting the overall conformation of the molecule . The proposed tertiary organization of E2 provides a prediction for the location of the HCV E2 fusion loop , which in class II fusion proteins is a stretch of highly conserved residues within the D0E0 insertion . This segment is composed mainly of non-charged residues , and is rich in glycine and non-polar amino acids . The sequence alignment highlights the region spanning residues 502–520 ( red circles in Fig . 6A ) , which has similar characteristics and is strictly conserved within all HCV genotypes ( Fig . 3 ) . This region is thus a strong candidate for fulfilling the role of the HCV fusion loop . The smaller conserved block between residues 484–489 has been tested by site directed mutagenesis using retroviral particles pseudotyped with the HCV envelope proteins ( HCVpp ) , suggesting that it does not play a role in membrane fusion [29] . Because in all class II fusion proteins , the fusion loop is buried at an oligomeric interface in the prefusion form , the candidate fusion loop segment is also very likely to mark a contact region with E1 . Our model predicts that DIII ( blue ) , which in alpha- and flavivirus fusion proteins has an Ig-like fold , contains disulfides 6 , 7 and 8 , and the last two glycosylation sites , 10 and 11 . It is connected to the C-terminal end of DI via the IgVR , which contains glycan 9 . This linker region is such that it can be extended to allow the translocation of DIII to the side of the trimer during the fusogenic conformational change , as expected for a class II fusion protein . Yet this segment , located between disulfides 5 and 6 ( highlighted in brown in Fig . 6A ) , can also be replaced by a GSSG linker without affecting the overall protein conformation [14] , suggesting close apposition between DI and DIII at least in the prefusion form . This organization is also compatible with the observation that some of the residues important for CD81 binding map to DIII ( 613–618 , Figs . 3 and 6 ) , suggesting that CD81 bridges the surface of the two domains . Indeed , these two domains display an extended interaction surface in other class II fusion proteins . DIII is followed in sequence by a relatively flexible but conserved region , denoted the “stem” ( grey in Fig . 6 ) , which connects to the TM segment . The stem would contain a loop that is closed by the last disulfide ( number 9 ) . A number of reports on the characterization of a protein ending at position 661 indicate that the absence of this loop does not affect the conformation of the protein and it is therefore not part of DIII . Further support for this interpretation is provided by our identification of a trypsin-resistant fragment of E2e ending at position Arg648 , in between disulfides 8 and 9 ( data not shown ) . One important implication is that the residues interacting with CD81 are found in two domains , DI and DIII , which have to move apart during the fusogenic conformational change . This suggest that CD81 may have to dissociate away for such a conformational change to take place , or on the contrary , that its binding may help to lower the energy barrier for the conformational change to occur upon exposure to low pH in the endosomes . An additional information from this study is the positioning of the hypervariable regions of HCV E2 in the context of the class II fold . As suggested above , the presence of these regions is likely to be responsible for the difference in CD spectrum between HCV E2e and the other class II proteins examined ( Fig . 4A ) , which do not contain unstructured regions . Such regions are presumably important for evading the humoral immune response of the host , given that HCV can cause chronic infection , in contrast to alpha- and flaviviruses . Another important difference to the latter are the numerous glycosylation sites in HCV E2 . Our model indicates that these sites cluster in particular on the exposed face of DI . Indeed , several glycans appear to frame the CD81 binding surface , partially shielding it from recognition by circulating antibodies . These are glycans 1 , 2 , 6 , 7 , and possibly 10 and 11 ( Fig . 6A ) . Importantly , a number of studies have pointed to a role of some of these E2 glycosylation sites modulating entry and/or CD81 binding [43] , [44] . Another important implication is the identification of a strong candidate region for the fusion loop . This polypeptide segment , spanning residues 502–520 of E2 , has all the characteristics reported for the experimentally characterized class II fusion loops . It is strictly conserved , and is located in a region of the protein that is compatible with this function – in the D0E0 insertion of DI - in spite of being part of a fusion domain ( DII ) that is much smaller than its alpha - and flavivirus counterpart . The fact that in the latter DII is formed by two insertions into a simple , conserved DI , suggests that they may have evolved sequentially . HCV may have thus maintained some ancient intermediate form of the class II proteins , containing only the insertion that carries the fusion loop . The observed flexible and largely unstructured conformation of DII is likely to be due to the absence of the H0I0 scaffold . The presence of the candidate HCV fusion loop relatively close to DI is another important difference . In the alpha- and flavivirus counterparts , there are about 30 intervening residues present in an extended conformation between the fusion loop and strand E0 , whereas there are only a few residues in HCV . This difference is likely to be related to the intrinsic flexibility of HCV DII , since the fusion loop is unlikely to lie at the distant tip of an unstructured domain . During membrane fusion , the fusion loop could be further stabilized by interaction with the membrane proximal region of the stem , once the molecule adopts its fusogenic hairpin conformation . This organization also suggests a significantly shorter post-fusion HCV E2 trimer , compared to the other class II proteins . This would be analogous to the observed differences in the post-fusion trimers of class I viral fusion proteins , for instance from retroviruses , which are short [45] , and paramyxoviruses [46] or coronaviruses [47] , which display a very long hairpin conformation . Overall , our model for the HCV E2 tertiary organization provides a structural framework to understand the antigenicity of the virion , the organization of the regions that interact with CD81 , and the putative conformational changes that are likely to take place during the membrane fusion reaction to invade a target cell . In the absence of 3D structural data , our results constitute an important step to better understand the function of the HCV envelope proteins . This knowledge , in turn , can help devise possible antiviral strategies against this important pathogen . Our data also provide a handle to dissect and obtain structural data on the E2 domains separately , given that the intact ectodomain is very difficult to crystallize . Finally , this analysis highlights the power of conducting parallel structural studies on related viruses , which provide information that can be extrapolated to other members of the respective viral families , even in the absence of sequence similarity in the corresponding proteins .
The accompanying Supplementary Information ( Text S1 ) describes in detail the construction of the vectors used for expression of synthetic genes coding for the E1-E2ΔTM segment of the 9 HCV isolates tested ( Table 1 ) , the protocols for production , purification and conformational characterization of recombinant HCV E2e , as well as the computer and experimental analyses used for secondary structure predictions . Finally , a detailed description of the procedures used for the experimental identification of the cysteine residues involved in 8 disulfide bonds of E2 is provided . | Little is known about the structure of the envelope glycoproteins of the hepatitis C virus ( HCV ) , in spite of their essential role in the viral cycle of this major human pathogen . Here , we determined the connectivity of the 9 disulfide bonds formed by the strictly conserved 18 cysteines of the ectodomain of HCV glycoprotein E2 . We show that this information , together with important functional data available in the literature , impose important restrictions to the possible three-dimensional fold of the molecule . Indeed , these constraints allow the unambiguous threading of the predicted secondary structure elements along the polypeptide chain onto the template provided by the crystal structures of related flavi- and alphavirus class II fusion proteins . The resulting model of the tertiary organization of E2 shows the amino acid distribution among the characteristic class II domains , places the CD81 binding site at the interface of domains I and III , and highlights the location of a candidate fusion loop . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
]
| [
"biochemistry/molecular",
"evolution",
"virology/virion",
"structure,",
"assembly,",
"and",
"egress",
"biochemistry/protein",
"folding",
"virology",
"virology/host",
"invasion",
"and",
"cell",
"entry"
]
| 2010 | The Disulfide Bonds in Glycoprotein E2 of Hepatitis C Virus Reveal the Tertiary Organization of the Molecule |
The co-infection cases involving dixenous Leishmania spp . ( mostly of the L . donovani complex ) and presumably monoxenous trypanosomatids in immunocompromised mammalian hosts including humans are well documented . The main opportunistic parasite has been identified as Leptomonas seymouri of the sub-family Leishmaniinae . The molecular mechanisms allowing a parasite of insects to withstand elevated temperature and substantially different conditions of vertebrate tissues are not understood . Here we demonstrate that L . seymouri is well adapted for the environment of the warm-blooded host . We sequenced the genome and compared the whole transcriptome profiles of this species cultivated at low and high temperatures ( mimicking the vector and the vertebrate host , respectively ) and identified genes and pathways differentially expressed under these experimental conditions . Moreover , Leptomonas seymouri was found to persist for several days in two species of Phlebotomus spp . implicated in Leishmania donovani transmission . Despite of all these adaptations , L . seymouri remains a predominantly monoxenous species not capable of infecting vertebrate cells under normal conditions .
Flagellates of the family Trypanosomatidae are single-celled obligatory parasites . They can be either dixenous ( i . e . those with two hosts in their life cycle—Trypanosoma , Leishmania , and Phytomonas spp . ) or monoxenous ( i . e . those having only one host ) . For decades , monoxenous trypanosomatids of insects were effectively neglected . However , this situation is rapidly changing , as a remarkable diversity of these flagellates is being revealed within insects—a group which is known to be extraordinarily species rich [1 , 2] . In addition , the study of these parasites is expected to shed light on the origin of the dixenous life cycle ( alternation of an insect vector and a vertebrate or plant host ) . It is generally accepted that the dixenous species have evolved from their monoxenous kins and that this transition has happened independently at least three times during the evolution of Trypanosomatidae , as the dixenous genera Trypanosoma , Leishmania , and Phytomonas are interspersed by the monoxenous genera Angomonas , Blastocrithidia , Blechomonas , Crithidia , Herpetomonas , Kentomonas , Leptomonas , Paratrypanosoma , Sergeia , Strigomonas , and Wallacemonas ( S1 Fig ) [3 , 4] . This suggests that some ( presumably ) monoxenous species may occasionally try switching to dixeny . Indeed , the presence of the monoxenous trypanosomatids in vertebrates has been noted already about 100 years ago [5] . More recently , several monoxenous flagellates belonging to the genera Herpetomonas , Crithidia , Leptomonas , and Blechomonas have been identified from human clinical isolates [6–8] . Importantly , most of them involved immuno-compromised individuals , leading to a hypothesis that these usually non-infectious species may explore new ecological niches in vertebrates that have their immune system suppressed [9 , 10] . Within this paradigm , about two dozen cases of monoxenous trypanosomatids co-infecting humans along with various Leishmania spp . have been reported almost exclusively from the Indian subcontinent . Most of them implicated causative agents of visceral leishmaniasis ( kala-azar ) of the L . donovani complex [11] . It was also demonstrated that both dixenous and monoxenous flagellates may be transmitted by the same Phlebotomus vector , yet the evidence is not very strong [12 , 13] . The cytochrome b and 18S rRNA-based PCR analyses were confined to the isolates from a small geographical area and the identity of non-Leishmania parasites could not be elucidated to the species level . The species most often recovered from co-infections in leishmaniasis patients is Leptomonas seymouri Wallace , 1959 [14] . Together with all Leishmania spp . it belongs to the subfamily Leishmaniinae ( S1 Fig ) [15] and was originally isolated from a cotton stainer Dysdercus suturellus ( Hemiptera: Pyrrhocoridae ) [16] . Nonetheless , when a broad-scale survey of trypanosomatids parasitizing pyrrhocorids throughout the world was undertaken , none of the samples proved to contain L . seymouri [17] . So the question remains whether the original isolate was obtained from a specific host ( e . g . species that is evolutionary adapted for parasite's life cycle ) . L . seymouri can even multiply in plants under experimental conditions [18] proving it to be non-fastidious and able to adapt to quite different environments . Recent whole-genome analysis of kala-azar clinical isolates from splenic aspirates demonstrated heavy "contamination" with unidentified Leptomonas sp . [19] . This result is not so surprising provided that both parasites are almost indistinguishable by morphology and that Leptomonas outgrows Leishmania in culture [20] . We speculate that several species of monoxenous trypanosomatids are capable of surviving in the hostile environment of the vertebrate body . Molecular details of such adaptation are not worked out , yet it is clear that some monoxenous trypanosomatids must be able to tolerate heat shock up to the temperatures they might experience in warm-blooded vertebrates . Indeed , a number of representatives of the genera Crithidia and Herpetomonas can withstand elevated temperature reaching 37°C [21–23] . In this study we addressed the issue of Leishmania–Leptomonas co-infection from the point of view of the monoxenous partner . To understand molecular mechanisms and biochemical pathways responsible for survival within warm-blooded vertebrates , we have demonstrated that Leptomonas seymouri can withstand elevated temperatures in vitro , sequenced its genome , and assessed transcriptional profiles of cells cultivated in different conditions . Furthermore , we tested L . seymouri ability to survive in Phlebotomus argentipes and P . orientalis , two sand fly species implicated in Leishmania donovani transmission .
Whole genome sequencing of two clinical Indian kala-azar field isolates , a strain resistant to sodium antimony gluconate therapy ( Ld 39 , May 2000 , Muzaffarpur , Bihar ) and a strain sensitive to treatment ( Ld 2001 , February 2000 , Balia , Uttar Pradesh ) , revealed numerous ( over 95% ) sequences apparently derived from Leptomonas sp . in addition to those of L . donovani [19] . These isolates were cultivated from splenic aspirates in frame of a large screen aimed to understand molecular differences between confirmed kala-azar cases . For precise identification of the co-infecting species we applied an arsenal of molecular tools developed over the years [24–27] . Three genetic loci , namely 18S rRNA , glycosomal glyceraldehyde-3-phosphate dehydrogenase ( gGAPDH ) , and ITS regions were amplified , sequenced and compared with other representatives of the subfamily Leishmaniinae [15] . 18S rRNA sequences of the isolates Ld 39 and Ld 2001 ( GenBank accession numbers KP717894 and KP717895 , respectively ) were identical and indistinguishable from the corresponding sequence of L . seymouri ( GenBank accession number AF153040 ) . gGAPDH sequences ( GenBank accession numbers KP717896 and KP717897 for isolates Ld 39 and Ld 2001 , respectively ) were nearly identical with only 1 nt substitution in the coding sequence . They both were very similar ( except for the degenerative primer sequences ) to the gGAPDH sequence of L . seymouri ( GenBank accession number AF047495 ) . 18S rRNA and gGAPDH sequences are informative for higher level taxonomy , and are usually adequate for the genus ( and up ) level ranking [4 , 28] . For proper species identification we used other well-established markers , ITS1 and ITS2 [14 , 20 , 29] . Their sequences were identical with the exception of a 2 nt-long indel ( GenBank accession numbers KP717898 and KP717899 for isolates Ld 39 and Ld 2001 , respectively ) . BLAST search revealed 100% identity with the ITS1-5 . 8S rRNA region of L . seymouri ( GenBank accession number JN848802 ) . The data presented above allowed us to conclude that the monoxenous co-infectant of the clinical kala-azar isolates Ld 39 and Ld 2001 is L . seymouri . We also would like to note that the cases of co-infections of Leishmania and Leptomonas are likely underreported in the literature , as several sequences attributed to L . donovani in GenBank do in fact belong to L . seymouri . Our analysis of the ITS-containing region , SL , gGAPDH , HSP70 , HSP83 , RNA polymerase II , α-tubulin and some mitochondrial genes ( A6 , cytb , COI , COII , COIII , NADH ) revealed that 38 out of 170 ( 22% ) and 3 out of 217 ( 1 . 4% ) ITS sequences of L . seymouri were misidentified as Leishmania donovani and L . tropica , respectively ( see S1 Table for GenBank accession numbers ) . The presence of monoxenous L . seymouri in co-infections with dixenous L . donovani implies several adaptations to the environment of the human body . One of the important factors to be considered is temperature . Typical monoxenous trypanosomatids of the insect gut are temperature-sensitive and cannot withstand conditions of the warm-blooded vertebrates [6] . In order to investigate temperature resistance of several trypanosomatid species in vitro , we compared growth kinetics of two different Leptomonas species , L . seymouri ATCC 30220 ( hereafter used as a proxy of filed isolated Ld 39 and Ld 2001 , which were not available ) and L . pyrrhocoris H10 , under different experimental conditions . Parasites were incubated at temperatures 23°C , 29°C , and 35°C for up to 7 days . The highest temperature ( 35°C ) approximately corresponds to that faced by the flagellates upon transfer from a sand fly into a vertebrate . To imitate the conditions of insect gut and vertebrate blood , SDM and two-phased blood-agar were used , respectively . No considerable difference was observed in growth kinetics of two trypanosomatid species incubated at 23°C in both media . Interestingly , increasing the cultivation temperature to 29°C and 35°C inhibited growth of L . pyrrhocoris , while growth of L . seymouri was not significantly affected ( Fig 1 ) . We concluded that L . seymouri is capable of withstanding the elevated temperature reaching that of the human body . In contrast , L . pyrrhocoris is temperature-sensitive and halts its cell division in non-optimal conditions . In all cases , cultivation on blood-agar medium resulted in higher cells density . Light microscopy of Giemsa stained smears of L . seymouri cultivated under different experimental conditions revealed statistically significant morphological changes ( Fig 2 ) . The most noticeable one was shortening of the free portion of the flagellum observed in cells cultivated at high temperature . This phenomenon was observed for both media used but it was more pronounced in blood-agar . Also elevated temperature resulted in more diverse body sizes and shapes with the most conspicuous feature being elongated and tapered posterior end of some cells . The genome of L . seymouri ATCC 30220 was assembled into 1 , 222 scaffolds ( maximum length 326 , 845 bp ) with N50 of 70 , 646 bp and a total assembly length of approximately 27 . 3 Mbp . This is a substantial improvement over the previously reported assembly of the unidentified Leptomonas sp . ( 14 , 518 contigs with maximum length of 26 , 366 and N50 of 3 , 370 bp ) [19] . Both assemblies had almost the same total genome length ( 27 . 3 and 27 . 4 Mb ) . Importantly , over 85% of the reads could be cross-mapped ( length fraction = 0 . 9; similarity fraction = 0 . 9 ) confirming identity of the L . seymouri isolates . The number of annotated protein-coding genes , 8 , 488 , was also within the range of previously reported genomes ( 6 , 451 for Phytomonas sp . HART1; 8 , 309 for Leishmania major; 10 , 109 for Trypanosoma brucei ) [30–32] . Consistent with other trypanosomatids , the protein-coding genes lack conventional introns . The only exceptions reported so far in Trypanosoma spp . and Leishmania spp . are poly ( A ) polymerase and DEAD/H RNA helicase [32 , 33] . Indeed , their L . seymouri orthologs also contain introns and thus require cis-splicing for proper expression . A typical aspect of the L . seymouri genome is that it contains a relatively small number of genes that have undergone tandemly linked duplications . Using a cutoff value of 10−50 , the number of genes present in two or more homologous copies has been estimated at about 9 . 9% in L . seymouri . Same numbers for Phytomonas sp . , L . major , T . brucei , and C . fasciculata are 9 . 6% , 18 . 3% , 26 . 0% , and 40 . 2% , respectively . This is one of the major components determining differences in genome size among these species . Genomic information was used to predict the metabolic pathways in L . pyrrhocoris and L . seymouri , two phylogenetic kins with different sensitivity to temperature and ability to co-infect vertebrate hosts ( S2 Fig ) . In essence , the metabolism in these two species is very similar , with important features and differences highlighted below . A classical glycolytic pathway , partly inside glycosomes ( as inferred from the presence of peroxisome targeting signals ) , is responsible for the metabolism of various exogenous sugars ( S2 Table ) . Carbohydrate metabolism is characterized by an incomplete aerobic oxidation because one of the classical mitochondrial tricarboxylic acids ( TCA ) cycle enzymes ( NAD-linked isocitrate dehydrogenase ) is absent . However , the other TCA cycle enzymes can be used for the inter-conversion of metabolic building blocks required for gluconeogenesis and other biosynthetic purposes ( S3 Table ) . While both L . pyrrhocoris and L . seymouri are able to synthesize their own pyrimidines , they depend on a supply of external purines . They lack the capacity to oxidize aromatic amino acids and require an external supply of most of the essential amino acids , cofactors and vitamins for growth ( S4 Table ) . Both Leptomonas spp . have a fully developed mitochondrion with 9 of the 10 TCA cycle enzymes present , a complete respiratory chain with the respiratory complexes I—IV , and a fully functional mitochondrial F1-ATPase ( S5 Table ) . Although lacking the alternative oxidase found in many other trypanosomatids , L . seymouri possesses an alternative NADH dehydrogenase gene . Our analysis predicts that it is able to feed on a large variety of polysaccharides , carbohydrates , both hexoses and pentoses , with the anticipated end products of carbohydrate metabolism being acetate , succinate , carbon dioxide , ethanol , alanine , and D-lactate . L . seymouri has a complete set of β-oxidation enzymes , which are associated with the mitochondrion . A few additional lipid-metabolizing enzymes are present in the glycosomes . It appears that the analyzed flagellate does not possess a type-I system of fatty acid synthesis , but makes its fatty acids in the cytosol by the action of a series of elongases ( S6 Table ) . It is able to oxidize 16 of the 20 amino acids , but the necessary enzymes for the metabolism of lysine and the three aromatic amino acids ( phenylalanine , tyrosine and tryptophan ) are lacking . The urea cycle is not functional since two mitochondrial enzymes of the cycle are missing ( S7 Table ) . The remaining three cytosolic enzymes have all been acquired by lateral gene transfer and allow arginine to be utilized in polyamine biosynthesis . Surface proteins , previously identified in Trypanosoma , Leishmania and Crithidia spp . , have also been found in Leptomonas ( S8 Table ) . Homologues of GP63 , amastin , 3’-nucleotidase , integral membrane protein , prohibitin , membrane-bound acid phosphatases MBPA1 and MBPA2 and tartrate-sensitive acid phosphatase , but not oligosaccharyl transferase , are present . Protection against oxidative stress in monoxenous trypanosomatids differs from their dixenous kins . In addition to the trypanothione system and the presence of many homologues of tryparedoxins and peroxiredoxins , all monoxenous species analyzed thus far have a bacterial-type catalase acquired by lateral gene transfer ( S9 Table ) . Enzymes of the RNA interference pathway , namely the homologs of the Argonaute ( AGO1 ) and the two dicer proteins ( DCL1 and DCL2 ) were not detected in L . seymouri ( S10 Table ) . Importantly , they were found in the genome of L . pyrrhocoris arguing that these two closely related species differ in their ability to regulate gene expression by RNA interference . In the evolution of Trypanosomatidae many events of lateral gene transfer ( LGT ) have taken place , since genes of bacterial origin are frequently encountered in all trypanosomatid lineages [34] . This suggests that an ancestral flagellate had already acquired such genes , which include a number of enzymes of glycolysis , pentose-phosphate shunt and pyrimidine biosynthesis , as well as trypanothione reductase and pterin transporters [35–37] . Some LGT events including genes involved in sucrose and pentose sugar metabolism , haem synthesis and urea cycle seem to be more recent and specific to the Leishmaniinae clade that comprises Leishmania , Crithidia and Leptomonas spp . [38–40] . Even more recent acquisitions , shared only among Crithidia spp . and Leptomonas spp . include catalase , the diaminopimelate-metabolizing enzymes and those of β-glucosidase , nitroalkane oxidase , phenolic acid dehydrogenase and glycerol dehydrogenase families ( S11 Table ) . In total , 70 out of 586 , or 12% of all the metabolic genes analyzed , have resulted from the events of lateral transfer . For this analysis full proteomes for 23 trypanosomatid species were downloaded from TriTrypDB v . 7 . 0 and combined with newly annotated proteins from L . seymouri , L . pyrrhocoris , B . ayalai and Paratrypanosoma confusum ( S12 Table ) . Comprehensive characterization of L . seymouri gene family repertoire and its comparison to that of other trypanosomatids may help to shed light on possible adaptations of this species to the dixenous lifestyle . Recently , a comparative genomics approach was used to define a "gene kit" implicated in cell invasion and intracellular parasitism in Leishmania spp . and Trypanosoma cruzi [41] . Authors have found that despite substantial differences in mechanisms of host cell invasion and survival within the host cell , 3 , 340 orthologous gene clusters are exclusively shared between intracellular parasites when compared to extracellular T . brucei . Many proteins within these clusters were already proven to play a pivotal role in Leishmania and Trypanosoma virulence ( e . g . GP63 , amastin , ascorbate peroxidase ) , while functions of other proteins require further detailed investigation . In our study we were aiming to identify candidate proteins in L . seymouri that may define its ability to occasionally infect warm-blooded organisms . For that purpose Orthologous Groups ( OG ) presence/absence patterns in L . seymouri were analyzed and compared to those of other trypanosomatids . In the reference dataset for OrthoMCL analysis several Leishmania spp . ( medically and veterinary important dixenous species ) , along with C . fasciculata and L . pyrrhocoris ( both never encountered in vertebrates ) are of primary interest for comparison with L . seymouri . According to a widely accepted view of trypanosomatid phylogeny , Leptomonas spp . are most closely related to Crithidia spp . , and together they form a clade that clusters as a sister group to the genus Leishmania [2 , 15] ( S2 Fig ) . Firstly , OG content was compared in L . seymouri , C . fasciculata , and L . pyrrhocoris in order to exclude from the analysis OGs that are present in typical monoxenous trypanosomatids . Leptomonas pyrrhocoris has a typical promastigote morphology and dwells in insect species of the family Pyrrhocoridae [17 , 42] , while C . fasciculata uses various culicids as hosts [43–45] . Notably , some representatives of the genus Crithidia ( C . hutneri , C . luciliae thermophila ) can survive at temperatures of the mammalian and avian bodies [21 , 22] . Therefore C . fasciculata may possess genes involved in survival at elevated temperatures , and in order to exclude possible biases caused by the presence of C . fasciculata genes in our OrthoMCL analysis , OG repertoire comparisons were performed twice: with and without C . fasciculata in the datasets being compared . Out of 7 , 935 L . seymouri OGs , 79 OGs were absent in L . pyrrhocoris , and 26 OGs were absent from both L . pyrrhocoris and C . fasciculata ( S13 Table and Fig 3 ) . Our assumption is that among the genes belonging to the above-mentioned groups there are at least several that predispose L . seymouri metabolism to dixeny . Fifty five out of 79 OGs absent in L . pyrrhocoris do not have any functional annotation assigned and thus represent a broad field for further studies ( S13 Table ) . Nevertheless , several genes identified by comparative genomics approach in our study were already proven to play a pivotal role in parasite survival and virulence ( see below ) . In order to further narrow down the set of such genes we introduced one more condition into the comparison: gene family present in L . seymouri must be also present in all Leishmania species considered in the analysis ( L . braziliensis MHOM/BR/75/M2903 , L . braziliensis MHOM/BR/75/M2904 , L . donovani BPK282A1 , L . infantum MCAN/ES/98/JPCM5 , L . major MHOM/IL/80/Friedlin , and L . mexicana MHOM/GT/2001/U1103 ) . A reptile parasite L . tarentolae ParrotTarII was not included in the analysis due to its inability to infect warm-blooded organisms [46] . Additional BLASTP search ( E-value ≤ 10−10 ) for proteins belonging to OGs and meeting the criteria stated above was performed in order to determine whether these OGs have related OGs with homologous proteins clustered separately by the sensitive OrthoMCL algorithm . Cases when related OGs have a presence/absence pattern which violates the abovementioned criteria are not discussed here since unambiguous conclusion cannot be made concerning the role of such proteins in L . seymouri thermotolerance . Sixteen OGs absent from L . pyrrhocoris are shared by L . seymouri and Leishmania spp . Importantly , only 2 of them are absent from both L . pyrrhocoris and C . fasciculata ( Fig 3 ) . These two OGs represent a kinase-like protein and a ubiquinol-cytochrome c reductase-like protein . According to the results of additional BLASTP search , the latter protein OG does not have any related OGs and all of its orthologs in the TriTrypDB are annotated as ubiquinol-cytochrome c reductase-like proteins . Aiming to identify homologs of this protein in other species beyond the TriTrypDB , we conducted a BLAST search against the NCBI nr database and found a close homolog only in Strigomonas culicis ( ubiquinol-cytochrome c reductase subunit 6 , E-value ≤ 10−30 , protein accession number: EPY16273 . 1 ) . The kinase-like protein mentioned above has weak hits with E-value over 10−30 to several other OGs containing protein kinases . Due to the relatively high E-values of the BLAST hits and quite unspecific annotations of kinases within related OGs , this protein was not excluded from our analysis ( although several related OGs have absence/presence patterns that differ from the required ones ) , and its possible role in L . seymouri thermotolerance cannot be ruled out . Having excluded the requirement of OG being absent from C . fasciculata , the overlap mentioned above extends to 16 OGs ( Fig 3 ) , which include one more group with putative protein kinases as well as putative anaphase-promoting complex subunit , putative epsin and several hypothetical proteins with unknown functions . In order to obtain a global picture of corresponding OG distribution for subunits of the anaphase-promoting complex and putative epsin , we extended analysis of OG presence/absence patterns to the whole dataset of 27 trypanosomatid species . As expected , OGs containing these proteins have shown nearly omnipresent distribution ( being absent from L . pyrrhocoris as required in our analysis and additionally missing in several Trypanosoma spp . ) . Additional BLAST search ( with more relaxed parameters ) for these proteins against proteins belonging to other OGs also did not return any hits . Such results can be explained assuming considerable sequence diversity in these proteins families . For epsins , a group of eukaryotic proteins broadly implicated in clathrin-mediated endocytosis , there is evidence for substantial sequence dissimilarities and lineage-specific protein architecture [47] . Anaphase-promoting complex is a multi-subunit E3 ubiquitin ligase that is necessary for proteolytic degradation of crucial cell cycle regulators , which causes segregation of sister chromatids [48] . Taking into account a universal role of the proteins mentioned above and their phyletic patterns ( especially their presence in several monoxenous species ) we conclude that they are unlikely to be involved in L . seymouri thermotolerance . Interestingly , 3 OGs containing hypothetical proteins ( OG_09193 , OG_10013 , and OG_10042 ) within the group of 16 OGs fully satisfy the conditions applied in the study , including the absence of closely related OGs . Moreover , these groups of homologous proteins do not occur in any Trypanosoma spp . and in monoxenous trypanosomatids for which genome sequences are available ( except for C . fasciculata ) . Proteins within these groups represent primary targets for additional studies aiming to reveal mechanisms contributing to L . seymouri thermotolerance . Prompted by our observation that L . seymouri lacks RNAi machinery ( see above ) and by patterns of RNAi retention in Trypanosomatidae [49] , we also examined L . seymouri for the presence of dsRNA viruses . Two complementary methods , the nuclease digestion assay and immunofluorescence microscopy , were used [50] . Indeed , the anti-dsRNA antibodies detected small sharp dots , which are reminiscent of those found in the virus-positive isolate of Leishmania guyanensis [51] . Importantly , these putative viral particles did not co-localize with the mitochondrion ( Fig 4A ) . The nuclease digestion assay of L . seymouri RNA was performed in parallel with the virus-free Blechomonas pulexsimulantis used as a negative control [52] . It detected dsRNA bands resistant to DNase I and S1 nuclease , which were present in RNA preparations from L . seymouri ( Fig 4B ) . Interestingly , this dsRNAs differ in size from that of the previously characterized LRV1 virus of Leishmania guyanensis ( 1 . 5 + 2 . 9 kb versus 5 . 3 kb , respectively ) [53 , 54] . It remains to be investigated whether this reflects critical differences in genomic organization of viruses , such as segmented versus whole dsRNA genomes . To identify genes and/or pathways responsible for thermoresistance of L . seymouri , we profiled whole transcriptomes of the parasites cultivated at low ( 23°C ) and high ( 35°C ) temperature . We presumed that in addition to genetic factors ( e . g . chromosome ploidy ) regulation of gene expression may also be involved in adaptation to dixeny . Reads passing the filtering step ( 61 . 4; 52 . 5; 39 . 1 million reads for replicates at 23°C and 61 . 1; 58 . 9; 37 . 9 million reads for replicates at 35°C ) were used in subsequent analyses . Out of 8 , 488 genes identified in the L . seymouri genome 8 , 482 genes were recovered in our analysis . Results of the FDR test are shown in S3 Fig . In total , 340 genes ( 4% of the total number ) were shown to be differentially expressed at the elevated temperature ( S14 Table ) . Expression of 139 genes ( 1 . 6% of the total number ) was found to be down-regulated at 35°C , whilst 201 genes ( 2 . 4% of the total number ) were upregulated at least 1 . 5 fold ( p-value ≤ 0 . 05 ) . Several interesting cases are discussed in detail below . Experimental infections of the two proven vectors of Leishmania donovani , Phlebotomus orientalis and P . argentipes , were compared side-by-side . Insects were fed on either blood or sugar meals to mimic the range of conditions which may favor infection ( Fig 5A and 5B ) . On day 2 after infective sugar meal all females of P . orientalis were infected , while the infection rate of P . argentipes females was lower ( 59% ) ( Fig 5A ) . Intensity of infection was generally weak in both species tested . On day 6 p . i . percentages of infected sand flies decreased to 77% and 46% for P . orientalis and P . argentipes females , respectively . On day 9 every third female remained infected , yet most of them harbored only few flagellates . Infection via the blood meal was less efficient when compared to the sugar meal . On day 2 less than half of blood-fed females of P . orientalis and P . argentipes were infected ( 47% and 32% , respectively ) . Freely moving promastigotes were found enclosed in the ingested blood . On days 6 and 9 L . seymouri promastigotes persisted only in a few females ( Fig 5B ) . The experimental co-infection of sand fly females of P . argentipes were performed by blood meals containing either mCherry- ( L . seymouri , ATCC-30220 ) and/or GFP-expressing ( L . donovani , strain GR-374 ) flagellates . In the control dissection of five sand flies performed just a few hours p . i . , both mCherry- and GFP-labeled cells were encountered at about 100 cells of each species per sand fly gut . On day 2 p . i . the infection rate of L . seymouri was lower than that of L . donovani ( 82 . 6% versus 95 . 7% , respectively ) , and also the intensity of infection with the former species was significantly weaker ( Fig 5C ) . The differences between both parasite species became even more pronounced on day 5 p . i . , when the percentage of infected sand flies remained unaltered for L . donovani ( 86 . 7% ) , while it markedly dropped for L . seymouri ( 13 . 3% ) . Moreover , the intensity of infection with L . donovani was high , whereas the few insects still infected with L . seymouri harbored only negligible number of free swimming mCherry-expressing cells ( Fig 5C ) . Survival of parasites inside mammalian host cells J774 or BMMɸ was evaluated 3 , 4 , 5 and 6 days p . i . No viable L . seymouri cells were found in macrophages by fluorescent microscopy or in Giemsa-stained smears . In contrast , the control represented by L . donovani survived inside both J774 and BMMɸ cells . Similar results were obtained using peritoneal macrophages from BALB/c mice . The transformation assay has confirmed microscopic observations , as no L . seymouri cells were found after the lysis of macrophages . On the contrary , L . donovani propagated very well under the same conditions . Similar results were obtained when either J774 or BMMɸ macrophages were simultaneously co-infected with both parasites .
Here we performed a multifarious evaluation of the infective potential of L . seymouri , repeatedly isolated from kala-azar patients infected by L . donovani in India and neighboring countries , and have tested the capacity of this monoxenous trypanosomatid to utilize the sand fly vectors permissive for L . donovani . Moreover , we attempted to find genetic and corresponding metabolic adaptations responsible for its survival at 35°C . Firstly , we have sequenced the whole genome of L . seymouri and compared it with L . pyrrhocoris and C . fasciculata , the only monoxenous species for which high-quality assemblies are available . Twenty six OGs carried by the thermotolerant L . seymouri and absent in these closely related thermosensitive flagellates may potentially be associated with this adaptation . Including dixenous species into the comparative analysis narrowed down our search to just two OGs shared by L . seymouri and five Leishmania spp . and absent from L . pyrrhocoris and C . fasciculata , namely a kinase-like protein and a ubiquinol-cytochrome c reductase-like protein . It was shown previously that protein kinases are involved in amastigote differentiation in Leishmania spp . [58] , a process in which temperature switch plays a decisive role [59] . Moreover , our search has identified a number of proteins with specific distribution among trypanosomatid lineages ( e . g . absent in Trypanosoma spp . and/or Leishmania spp . but present in monoxenous flagellates ) that are prime targets for functional analysis . In any case , the fact that the vast majority of genes within OGs with this phyletic distribution are annotated as hypothetical proteins with unknown function indicates our scarce knowledge of trypanosomatid metabolism . A number of metabolic changes observed in L . seymouri exposed to elevated temperature are evocative of those in Leishmania amastigotes or T . brucei bloodstream forms in glucose-poor environment [60 , 61] . For example , inhibition of the de novo synthesis of sterols in L . seymouri resembles Leishmania amastigotes in which the relative abundance of C24-alkyl sterols was significantly decreased upon their differentiation from procyclics [62 , 63] . Similarly to their dixenous cousins , L . seymouri cells at high temperature reduce the uptake of glucose and shift their acetyl-CoA production in mitochondria from mainly pyruvate-based to the fatty acids-derived [64–66] . The detection of double-stranded viruses in L . seymouri is particularly relevant in the light of recent findings that their presence in Leishmania guyanensis correlates with its virulence and metastatic potential [51 , 67] . While molecular mechanisms of this phenomenon are just becoming to be understood , it is already clear that the host immune response is rewired [68 , 69] . We and others have detected dsRNA-containing viruses in several other monoxenous trypanosomatids parasitizing dipteran and heteropteran insects [27 , 51 , 70 , 71] , but their relationships to the characterized viruses of Leishmania still remain a mystery . Two analyzed Leptomonas spp . differ in their acceptability for dsRNA viruses . This indicates fundamentally different mechanisms they may utilize to regulate their gene expression . In summary , we conclude that although L . seymouri has developed several adaptations that allow it to grow well at 35°C , it remains a predominantly monoxenous species not able to infect mammalian macrophages either alone or in co-infection with Leishmania . This agrees with a recent report on selective elimination of Leptomonas from co-cultures with Leishmania [72] . Under certain circumstances it is able to infect mammals , but probably only when the host is immunocompromised by infection with another pathogen , such as L . donovani or HIV [14 , 73] . However , it is quite likely that such co-infections are much more frequent than the available literature suggests . This conclusion is further supported by our finding that L . seymouri can survive up to 9 days in the same sand fly species that is responsible for the transmission of pernicious Leishmania spp . Therefore , it will be important to analyze samples from patients suffering from visceral and other leishmaniases with primers specific for L . seymouri and related ( presumably ) monoxenous trypanosomatids to address the possibility that we see only the tip of the iceberg . In addition to the capacity to withstand elevated temperature , other factors , such as its ability to escape the host immune response , may likely play an important role in establishment of the Leptomonas infection in mammals . We cannot exclude the possibility that some isolates of L . seymouri may be exclusively transmitted by sandflies and spend part of their life cycle in vertebrates similar to their Leishmania spp . relatives .
Animals were maintained and handled in the animal facility of Charles University in Prague in accordance with institutional guidelines and Czech legislation ( Act Number 246/1992 and 359/2012 coll . on Protection of Animals against Cruelty in present statutes at large ) , which complies with all relevant European Union and international guidelines for experimental animals . The experiments were approved by the Committee on the Ethics of Animal Experiments of the Charles University in Prague ( Permit Number 24/773/08-10001 ) and were performed under the Certificate of Competency ( Registration Number CZU945/05 ext . CZ02573 ) and the Permission Number 31114/2013-MSMT-13 ext . 24115/2014-MZE-17214 of the Ministry of the Environment of the Czech Republic . Leptomonas seymouri isolate ATCC 30220 was obtained from the American Type Culture Collection ( ATCC , Manassas , USA ) . It was isolated from the cotton stainer Dysdercus suturellus in the United States in 1959 . Leptomonas pyrrhocoris isolate H10 [17] , Blechomonas ayalai isolate B08-376 [52] and Leishmania donovani isolate MHOM/ET/2010/GR374 have originated from the research collections at Charles University in Prague , Institute of Parasitology in České Budějovice , and Life Science Research Centre in Ostrava . Cultures of the monoxenous trypanosomatids were routinely maintained in the Schneider's Drosophila medium ( SDM ) ( Thermo Fisher Scientific , Waltham , USA ) supplemented with 10% Fetal Bovine Serum ( FBS ) ( Thermo Fisher Scientific ) , 50 units/ml of penicillin , 50 μg/ml of streptomycin ( both from Sigma-Aldrich , St . Louis , USA ) , and 10 μg/ml of hemin ( Jena Bioscience GmbH , Jena , Germany ) at 23°C . All isolates used in this work can also be cultivated in the Brain Heart Infusion ( BHI ) medium ( Sigma-Aldrich ) supplemented with 10% FBS and antibiotics as above , or in the two-phased blood-agar medium [74] . To estimate the dynamics of growth , 5 x 104 parasites were seeded into the SDM or the blood-agar medium . Cultures were incubated at 23°C , 29°C , and 35°C for 7 days . Cell numbers were counted using a hemocytometer and plotted in log scale . Morphology of the cells cultivated at low ( 23°C ) and high ( 35°C ) temperature , either in SDM or blood-agar media , was analyzed at day 4 ( exponential phase ) after staining cells with Giemsa as described previously [75 , 76] . One hundred cells per sample were measured and analyzed using ANOVA statistical models [77] . Leishmania donovani ( MHOM/ET/2010/GR374 ) transfected with Green Fluorescent Protein ( GFP ) was cultured in M199 medium ( Sigma ) containing 20% heat-inactivated FBS ( Thermo Fisher Scientific ) supplemented with 1% BME vitamins ( Sigma ) , 2% sterile urine , 50 units/ml penicillin , 250μg/ml amikacin ( Bristol-Myers Squibb , New York , USA ) , and 150 μg/ml of geneticin , G418 ( Sigma ) . The internal transcribed spacer , ITS region of the rRNA locus was amplified using primers IAMWE and Tc5 . 8-rev and conditions described elsewhere [78] . Total genomic DNA samples of clinical Indian kala-azar field isolates Ld_39 and Ld_2001 were used as templates [19] . The 18S rRNA and gGAPDH genes were PCR-amplified , cloned into the pCR2 . 1 vector system ( Thermo Fisher Scientific ) , sequenced and analyzed as described previously [79 , 80] . The obtained sequences were deposited to GenBank with the following accession numbers: KP717894 , KP717895 ( 18S rRNA ) ; KP717896 , KP717897 ( gGAPDH ) ; KP717898 , KP717899 ( ITS1 + ITS2 regions ) . The Leptomonas seymouri ATCC 30220 genome was sequenced with 100 nt paired-end reads using the Illumina HiSeq 2000 platform ( Macrogen , Seoul , South Korea ) . Prior to assembly , reads were subjected to trimming and filtering using CLC Genomics Workbench v . 7 . 0 ( CLC Inc , Aarhus , Denmark ) : regions with Phred quality < 20 were trimmed , no more than one N was allowed in the remaining sequence , then TruSeq adapter trimming and a minimum length threshold of 75 nt were applied . Draft genome of L . seymouri was assembled with the CLC Genomics Workbench v . 7 . 0 employing a De Bruijn graph-based algorithm with the average coverage of 180 x . Augustus v . 2 . 5 . 5 was used to annotate the draft genome of L . seymouri [81] . Prediction accuracy of Augustus was improved by retraining using a training set of L . seymouri conserved proteins . In brief , de novo assembled contigs were searched against proteins in the TriTrypDB v . 7 . 0 database [82] ( BlastX E-value ≤ 10−5 ) and best BLAST hits were chosen based on the following criteria: a ) E-value ≤ 10−30 , b ) hit length longer than 80 amino acids ( aa ) , c ) percent identity higher than 40 . Subsequently , a non-redundant training set of 727 high-confidence gene models with unambiguous start site positions was created based on best BLAST hits to annotated proteins from TriTrypDB and RNA-seq coverage data . Non-redundancy of the training set was achieved by excluding genes with more than 70% identity at the amino acid level . Further analysis of the Augustus annotation included manual curation of predicted genes based on transcriptome sequencing data , e . g . removing start sites predicted in regions with no transcriptomic coverage and adding transcribed ORFs >200 aa in length not predicted by Augustus . For tRNA gene prediction tRNAscan-SE Search Server [83] was used with default parameters . For annotating other non-coding RNAs BlastN algorithm ( E-value ≤ 10−10 ) was employed with subsequent manual inspection of BLAST results . As a result , 8 , 488 genes were annotated in the L . seymouri genome , which has been submitted to the NCBI ( BioProject accession number PRJNA285179 ) and the TriTryp database , a part of the EuPathDB [84] . Orthologous groups are the set of genes descended from a single common ancestral gene , containing both paralogs and orthologs . OGs for L . seymouri proteins were inferred using the OrthoMCL v . 2 . 0 software [85] . Full proteomes for 23 trypanosomatid species were downloaded from the TriTrypDB v . 7 . 0 and combined with newly annotated proteins from L . seymouri and 3 other trypanosomatid species ( Leptomonas pyrrhocoris , Blechomonas ayalai and Paratrypanosoma confusum ) . The reference protein dataset was subjected to removal of poor quality proteins ( based on sequence length and percent of in-frame stop codons ) , all vs . all BLAST ( E-value 10−10 ) and a clustering procedure implemented in the OrthoMCL algorithm . This resulted in 19 , 866 OGs , 7 , 935 of which contained proteins of L . seymouri . L . seymouri was cultivated at 23°C and 35°C for 75 hrs . Total RNA was isolated from 2 . 5 x 107 cells using RNeasy Mini Kit ( Qiagen GmbH , Hilden , Germany ) according to the manufacturer’s instruction . The mRNA-derived libraries were sequenced with 100 nt paired-end reads on the Illumina HiSeq 2000 platform ( Macrogen ) . Total of 3 independent biological replicates were analyzed . The whole transcriptome data from this study have been submitted to TriTrypDB database [82] . Differential gene expression analysis was done using the RNA-Seq tool in CLC Genomics Workbench . Raw reads were subjected to quality-based trimming ( regions with Phred quality < 20 were trimmed , no more than one N was allowed in the remaining sequence ) , adapter trimming , and a minimum length threshold of 30 bp . Processed reads were then mapped to the annotated L . seymouri genome with the following parameters: maximum number of mismatches , 2; minimum fraction of read length mapped , 0 . 8; minimum identity within the mapped sequence , 0 . 8; maximum number of best-scoring hits for a read , 30 . All libraries were mapped as paired-end , and expression values ( RPKM ) for each gene were calculated . To identify gene sets that are differentially expressed between the two conditions , the FDR test was employed [86] . Genes with expression fold change ≥ 1 . 5 and FDR p-value ≤ 0 . 05 were chosen for further analyses . Gene ontology ( GO ) terms for genes up- and down-regulated at high temperature were generated using the Blast2GO plugin in CLC Genomics Workbench [87] . Initially , BlastP search against the NCBI nr database was performed , GO terms associated with all the hits were retrieved , and most appropriate GO terms were selected according to the standard Blast2GO procedure . GO term enrichment was assessed using Fisher's exact test . For detection of dsRNA viruses , two complementary protocols were used . Cells were stained with 4′ , 6-diamidino-2-phenylindole ( DAPI ) , mitotracker Red CMXRos ( both from Thermo Fisher Scientific ) and mouse monoclonal anti-dsRNA antibody ( Scicons , Szirák , Hungary ) , followed by goat anti-mouse IgG–Alexa Fluor 488 ( Thermo Fisher Scientific ) antibody as described previously [50] . In addition , 50 μg of total RNA isolated using TRI reagent ( Sigma-Aldrich ) was treated with 1 unit of DNase I ( New England Biolabs , Ipswich , USA ) at 37°C for 1 hr , followed by digestion with 35 units of S1 nuclease ( Sigma-Aldrich ) for 45 min at the same temperature . Samples were analyzed on 0 . 8% native agarose in 1xTAE buffer [88] . A fragment encoding mCherry fluorescent protein was amplified with primers 5´-TTATCCATGGTTAGTAAAGGAGAA-3´ and 5´-TGTTAGCGGCCGCTTATGCGGTACCAGAACC-3´ using plasmid p2686 as a template [89] . The resulting 745 bp fragment was cloned into the pF4T7polNLS1 . 4sat vector digested with NcoI and NotI replacing the T7 polymerase ORF [90] . Log-phase L . seymouri cells ( 4 x 107 ) were transfected with 15 μg of SwaI-linearized pF4mCherry1 . 4sat as described before [91] . Recombinant clones were selected on agar—BHI growth medium supplemented with 10% FBS , 40mM HEPES , pH 7 . 4 and nourseothricin ( Jena Bioscience ) at final concentration of 250 μg/ml . Expression of mCherry was confirmed by fluorescence microscopy . Colonies of two sand fly species , Phlebotomus orientalis and P . argentipes , both representing major proven vectors for L . donovani , were maintained under standard conditions as described elsewhere [92] . Females of both colonies were fed either through a chick-skin membrane on suspension of heat-inactivated rabbit blood containing exponentially growing 1 x 107 promastigotes per ml of blood or on 20% sucrose solution containing 5 x 107 promastigotes per ml . In order to recognize sugar-fed females , the sucrose solution was stained by indigo carmin . Blood- and sugar-fed females were kept at 26°C with free access to 50% sucrose solution by day 1 post infection ( p . i . ) . Sand fly females were dissected at different intervals p . i . ( 1–2 , 5–6 and 7–9 days ) . Numbers and location of flagellates in the sand fly gut were checked microscopically . Parasite loads were graded as previously described , i . e . : light ( < 100 parasites/gut ) , moderate ( 100–1 , 000 parasites/gut ) and heavy ( > 1 , 000 parasites/gut ) [93] . Females of P . argentipes were fed through a chick-skin membrane on suspension of heat-inactivated rabbit blood containing 1 x 106 per ml promastigotes of Leptomonas seymouri mCherry ( passage 4 ) and 1 x 106 per ml promastigotes of Leishmania donovani ( MHOM/ET/2010/GR374 ) GFP ( passage 10 ) originating from exponentially growing cultures . Assorted blood-fed females were kept at 26°C with free access to 20% sucrose solution . Sand fly females were dissected on days 2 and 5 p . i . , and the presence of parasites as well as other characteristics were analyzed as described previously [93] . Macrophage cell line J774 was cultured in complete RPMI-1640 medium ( Sigma ) containing 10% FBS , 100 U/ml of penicillin , 100 μg/ml of streptomycin , 2mM of L-glutamine , and 0 . 05 mM of β-mercapto-ethanol ( all from Sigma ) at 37°C with 5% CO2 . Bone marrow was obtained by flushing of tibias and femurs of BALB/c mice and flagellates were cultured in complete RPMI-1640 medium ( Sigma ) supplemented as above along with 20% of L929 fibroblast cell culture supernatant serving as a source of macrophage colony-stimulating factor at 37°C with 5% CO2 . The differentiation from bone marrow precursor cells to bone marrow-derived macrophages proceeded for 7 to 8 days in sterile polystyrene Petri dishes . The bone marrow derived macrophages ( BMMɸ ) were washed and seeded into plates at density of 5 x 105 cells per ml . Consequently , stationary cultures of Leishmania donovani ( GFP ) , Leptomonas seymouri ( mCherry ) , alone or in combination were added in ratio of 8:1 ( parasites: BMMɸ ) . Three days p . i . BMMɸ were extensively washed with pre-warmed RPMI-1640 to remove excess of parasites and the viability of trypanosomatids was monitored by fluorescence microscope Olympus CX-31 ( Olympus , Tokyo , Japan ) up to 6 day p . i . In addition , Giemsa staining was used to analyze intracellular forms in macrophages by light microscopy . All experiments were performed in two independent biological replicates . To analyze survival of parasites , the transformation growth assay was used [94] . In brief , macrophages infected with Leishmania donovani , Leptomonas seymouri , alone or in combination for 96 hrs were extensively washed with RPMI-1640 and lysed with 0 . 016% SDS in RPMI-1640 for 7 min at room temperature to release their intracellular forms . The lysis reactions were neutralized by RPMI-1640 supplemented with 17% heat-inactivated FBS . Parasites were spun down at 3 , 200 rpm for 10 min at 4°C , washed in RPMI-1640 , and re-suspended in a relevant promastigote medium ( BHI or M199 ) supplemented with an appropriate selective antibiotic at 23°C . For macrophages co-infected with both parasites , two types of media and antibiotics were assessed . The status of viable parasites was checked for 6 consecutive days . | In this work we performed a comprehensive evaluation of the infective potential of Leptomonas seymouri , repeatedly isolated from kala-azar patients infected by Leishmania donovani in India and neighboring countries , and have tested the capacity of this monoxenous trypanosomatid to utilize the sand fly vectors permissive for Leishmania donovani . We concluded that despite several genetic adaptations it has developed , Leptomonas seymouri remains a predominantly monoxenous species not able to infect mammalian macrophages either alone or in co-infection with Leishmania . Under certain circumstances it is able to infect mammals , but probably only when the host is immunocompromised by infection with another pathogen , such as Leishmania donovani or HIV . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | Leptomonas seymouri: Adaptations to the Dixenous Life Cycle Analyzed by Genome Sequencing, Transcriptome Profiling and Co-infection with Leishmania donovani |
Hypervirulent strains of Clostridium difficile have emerged over the past decade , increasing the morbidity and mortality of patients infected by this opportunistic pathogen . Recent work suggested the major C . difficile virulence factor , TcdB , from hypervirulent strains ( TcdBHV ) was more cytotoxic in vitro than TcdB from historical strains ( TcdBHIST ) . The current study investigated the in vivo impact of altered TcdB tropism , and the underlying mechanism responsible for the differences in activity between the two forms of this toxin . A combination of protein sequence analyses , in vivo studies using a Danio rerio model system , and cell entry combined with fluorescence assays were used to define the critical differences between TcdBHV and TcdBHIST . Sequence analysis found that TcdB was the most variable protein expressed from the pathogenicity locus of C . difficile . In line with these sequence differences , the in vivo effects of TcdBHV were found to be substantially broader and more pronounced than those caused by TcdBHIST . The increased toxicity of TcdBHV was related to the toxin's ability to enter cells more rapidly and at an earlier stage in endocytosis than TcdBHIST . The underlying biochemical mechanism for more rapid cell entry was identified in experiments demonstrating that TcdBHV undergoes acid-induced conformational changes at a pH much higher than that of TcdBHIST . Such pH-related conformational changes are known to be the inciting step in membrane insertion and translocation for TcdB . These data provide insight into a critical change in TcdB activity that contributes to the emerging hypervirulence of C . difficile .
Clostridium difficile is a gram-positive , spore-forming anaerobe , first described by Hall and O'Toole over 75 years ago [1]; however , the organism was not associated with human disease until 1978 [2] , [3] . Over the past three decades C . difficile has become a major nosocomial pathogen and is the leading cause of diarrhea in hospitalized patients [4] . C . difficile associated disease ( CDAD ) is routinely treated by supportive therapy and regimens of vancomycin and metronidazole , but treatment of CDAD has become more difficult due to the emergence of hypervirulent ( NAP1/BI/027 ) strains of C . difficile [5] , [6] , [7] . Elucidating the major differences between historical strains of C . difficile and the NAP1/BI/027-related strains of C . difficile is critical to understanding how this serious human pathogen continues to emerge . The phenotypes of hypervirulent and historical strains of C . difficile are different [7] , [8] , [9] . C . difficile NAP1/BI/027 produces more toxin and sporulates with higher efficiency than historical strains [6] , [7] , [8] , [9] , [10] . NAP1/BI/027 strains also produce a binary toxin , CDT , which is thought to enhance colonization of C . difficile by triggering the formation of microtubule protrusions on cells of the gastrointestinal epithelium [11] , [12] , [13] . Finally , C . difficile NAP1/BI/027 strains are resistant to fluoroquinolones due to mutations in DNA gyrase genes [7] , [14] , [15] , [16] . The extent to which one or more of these differences between the two strains contributes to hypervirulence has not been determined . Recent work from Stabler and colleagues identified several genetic variations between epidemic and historical strains of C . difficile [17] . For example , the historical C . difficile strain , 630 , was found to contain 505 unique coding sequences compared to hypervirulent strains . This analysis also identified differences in flagellar genes , metabolic genes , phage islands , and transcriptional regulators . Of interest to our work was the finding that TcdB from C . difficile hypervirulent strains had a greater cytopathic effect on a variety of cell types than TcdB isolated from a C . difficile historical strain . The steps in cellular intoxication that account for these differences in TcdB activity , and whether in vivo tropism varies between the historical and hypervirulent TcdB have not been reported . TcdB ( ∼269 kDa ) is a 2366 residue single polypeptide toxin encoded on a C . difficile pathogenicity locus ( PaLoc ) that also includes genes for two regulators ( TcdC and TcdR ) of toxin expression , a putative holin ( TcdE ) , and TcdA [18] , [19] . TcdB has at least four functional domains that contribute to cell entry and glucosylation of small-GTPases within the cytosol of the cell [20] . TcdB's glucosyltransferase domain is included in the first 516 residues of the toxin , which also includes a conserved DXD motif ( Asp286/Asp288 ) and Trp102 , which form a complex with Mn2+ and UDP-Glucose [21] , [22] , [23] , [24] , [25] . A substrate recognition domain is located between residues 365–516 [26] . The cysteine protease domain at residues 544–955 is necessary for autoproteolytic activity and delivery of the enzymatic domain into the cytosol [27] , [28] , [29] . A putative membrane-spanning domain resides between residues 956–1128 , yet whether this domain is required for intoxication is not known . Finally , the fourth functional domain of TcdB is located within the carboxy-terminal region of the toxin , and is predicted to interact with receptors on target cells [30] , [31] , [32] , [33] . Sequence variations in one or more of the functional domains of TcdB could account for the differences in cytotoxicity between historical and hypervirulent isolates . In the current work we test this hypothesis and demonstrate that TcdB from hypervirulent strains exhibits broader tropism in vivo . We also demonstrate TcdB from hypervirulent C . difficile undergoes hydrophobic conformational changes at a higher pH than toxin from the historical strain , and this correlates with more rapid cell entry . These findings provide insight into a possible mechanism through which hypervirulent C . difficile causes more severe illness than historical strains of this organism .
The carboxy-terminal sequence of TcdB varies between isolates of C . difficile , including hypervirulent and historical strains [17] , [34] . Yet , whether sequence variations are more extensive in TcdB compared to other genes in the PaLoc or if the sequences outside of the carboxy-terminal domain of TcdB also varied among different strains of C . difficile has not been reported . We compared the sequences of proteins encoded within the PaLoc of C . difficile 630 ( a non-NAP1/BI/027 strain ) and C . difficile R20291 ( a 027 strain ) . The sequence of TcdR , a positive regulator of toxin expression was found to be 100% identical between the two strains of C . difficile . TcdE , the putative holin encoded in the middle of the PaLoc exhibited 99% identity and 100% similarity between the two strains of C . difficile . The enterotoxin , TcdA , exhibited 98% identity and 99% similarity between the two strains . The gene encoding TcdC from the hypervirulent strain encodes a stop codon and contains a deletion , which made it difficult to precisely compare this protein in the two strains . However , at the DNA level the gene was 95% homologous in the intact coding regions of tcdC . In contrast to these almost exact identities of TcdR , TcdE , and TcdA from the two strains , the amino-acid sequence of TcdB from the two strains was found to have the most variation with 92% identity and 96% similarity . We next compared the functional regions of TcdBHIST and TcdBHV ( Fig . 1 ) . The enzymatic region of TcdB ( encompassing residues 1–543 ) was found to be 96% identical and 98% similar between the two strains of C . difficile . Residues critical for catalytic activity , W102 and the DXD motif , did not vary between the two forms of TcdB ( Fig . 1A ) . The substrate specificity domain of TcdB ( residues 365 to 516 ) [26] exhibited 99% identity and 100% similarity ( Fig . 1A ) . The autoproteolytic region ( residues 544 to 955 ) was found to contain 96% identity and 98% similarity . Moreover , the reported catalytic triad ( D587 , H653 , and C698 ) was conserved between the two forms of TcdB . Interestingly however , the analysis found a rearrangement of a second cysteine residue in this region of TcdB . TcdBHIST contains a cysteine at residue 870 , but this residue is a tyrosine in TcdBHV ( Fig . 1B ) . Conversely , TcdBHV has a cysteine residue at 1477 , but this was found to be a glycine residue in TcdBHIST . The third putative functional domain of TcdB is between residues 956 and 1644 , and encodes a hydrophobic region thought to mediate membrane insertion . Comparison of this region found 91% identity and 96% similarity ( Fig . 1C ) . In line with earlier reports [17] , [34] the carboxy-terminal region , encompassing residues 1645 to 2366 , exhibited the highest degree of sequence variation in the toxin . The carboxy-terminal region showed 88% identity and 95% similarity between the two forms of TcdB . The number of CROP regions is identical , with TcdBHIST and TcdBHV containing 24 regions based on the YF consensus motif [30] , [32] , [35] , [36] . However , eight of these regions in TcdBHV were found to exhibit less than 80% sequence identity to TcdBHIST ( Fig . 1D ) . Fig . 1E shows an SDS-PAGE analysis of TcdBHIST and TcdBHV purified from wild-type strains of C . difficile as described in the materials and methods . Both forms of the toxin were obtained at greater than 95% purity based on minimal detection of contaminating proteins . We next used a zebrafish model to compare the in vivo effects of the two forms of this toxin . Our group has previously utilized the zebrafish embryo as a model to examine the effects of TcdBHIST in real time , and found that this toxin had potent cardiotoxic effects [37] . The zebrafish provides a distinct advantage for the purpose of examining tissue damage and tropism because it is possible to visualize these events directly with this model . Zebrafish embryos were arrayed in a 48-well plate in embryo water and TcdBHIST or TcdBHV across a range of concentrations was applied to the individual wells . At 24 h following treatment , a minimum of 20 zebrafish larvae per condition were examined by light microscopy for physiological changes , tissue damage , and viability ( Fig . 2 ) . Extensive necrosis was evident in all embryos exposed to TcdBHV , with broad tissue damage caused to the yolk sac , body , and head at concentrations as low as 1 nM ( Fig . 2B and 2D ) . Furthermore , all zebrafish treated with TcdBHV succumbed to the effects of the toxin within 48 h . In contrast , treatment with TcdBHIST resulted in more specific damage at the cardiac region in approximately 75% of embryos , and was not immediately lethal ( Fig . 2A ) . Incubation with higher doses of TcdBHIST or for longer periods of time increased toxicity but did not alter the physiological damage from this toxin . These findings indicate that TcdBHV impacts a broader number of cell types in vivo compared to TcdBHIST . However , corresponding to our previous report TcdBHIST preferentially targets cardiac cells in the zebrafish embryo system . Recent studies determined the relative cytotoxicity of TcdBHV and TcdBHIST on eight different cell types [17] . Because this analysis did not include cells of cardiac lineage , we compared the two toxins on HL-1 cells , which are derived from mouse cardiac tissue [38] . We also examined the effects of the two toxins on CHO cells for a relative comparison to the cardiomyocytes . As shown in Fig . 3 , similar to previous observations , TcdBHV was more cytotoxic to CHO cells ( TCD50 2 . 37×10−13 M ) than was TcdBHIST ( TCD50 2 . 53×10−11 M ) . In contrast , TcdBHV was not more cytotoxic on cardiomyocytes and displayed a very similar activity to TcdBHIST . Upon further investigation of the cardiomyocytes , the cytotoxicity of TcdBHV was found to be slightly lower than TcdBHIST ( p<0 . 05 ) with a TCD50 approximately 10-fold higher ( 3 . 37×10−10 M ) than TcdBHIST ( TCD50 2 . 80×10−11 M ) . These data indicate that while TcdBHV has a broader cell tropism and is most likely more cytotoxic overall , TcdBHIST cardiotropism is more pronounced between the two forms of this toxin . We next determined if the variation in cytotoxicity was due to differences in the cytosolic activities of the two forms of TcdB . As an approach to this problem we took advantage of a previously described system used for heterologous delivery of proteins and protein fragments into the cytosol of target cells [39] , [40] . This system is composed of the cell entry components of anthrax lethal toxin . Briefly , protective antigen ( PA ) delivers lethal factor ( LF ) into the cytosol of mammalian cells . The heterologous delivery system is derived from the amino-terminus of LF ( LFn ) , which interacts with PA and can be delivered into cells , but lacks enzymatic activity . In our experiments , the DNA fragment encoding the enzymatic domain of TcdB was genetically fused to lfn , yielding a DNA construct that expresses the cell entry portion of LF with the enzymatic component of TcdB . This heterologous delivery system allowed us to regulate the cell entry of the enzymatic component of TcdBHV and TcdBHIST so that these domains were identical in the way in which they entered the cell . We predicted that if the differences in cytotoxicity were due to factors other than intracellular activity of these forms of TcdB , then the fusions should exhibit identical cytotoxic effects . The results of the PA , LFn-TcdB fusion experiments are shown in Fig . 4 . CHO cells were treated with a fixed amount of PA ( 500 nM ) plus a range of concentrations of LFnTcdBHV ( enz ) or LFnTcdBHIST ( enz ) in order to generate a standard killing curve for this assay . As controls , CHO cells were treated with PA , LFnTcdBHV ( enz ) , or LFnTcdBHIST ( enz ) separately . Following 24 h of treatment the cells were assayed for viability using WST-8 colorimetric assay and the percent survival was plotted versus concentration of the fusion protein . Treatment with each of the components alone had no effect on cell viability in this assay ( data not shown ) . Treatments with PA plus LFnTcdBHV ( enz ) or PA plus LFnTcdBHIST ( enz ) resulted in similar ( p<0 . 05 ) cytotoxicity at each of the concentrations tested ( Fig . 4 ) . To confirm that PA was not limiting in these experiments , cytotoxicity of the fusions was tested with 10-fold higher amounts of PA , and this additional amount of PA did not change the level of cytotoxicity for either fusion ( data not shown ) . The results from this experiment suggested that the differences in the cytotoxicity of LFnTcdBHV ( enz ) and LFnTcdBHIST ( enz ) were not due to variations in intracellular activities of the enzymatic domains . The results from the experiment using an identical method of cell entry , suggested the differences in cytotoxicity might be associated with early steps in cell binding and cell entry . To address this hypothesis , we compared the interaction of TcdBHV and TcdBHIST with cultured cells . Cultured cells were incubated with Alexa-647-labeled TcdBHV or Alexa-647-labeled TcdBHIST and the extent of toxin binding was determined by flow cytometry . This analysis was performed on CHO cell and HL-1 cardiomyocytes . As shown in Fig . 5 , CHO cells and HL-1 cells exhibited a higher degree of fluorescence when incubated with labeled TcdBHIST than when incubated with labeled TcdBHV . A biphasic profile was detected in CHO cells with a smaller population of cells exhibiting a distinct , reduced , toxin-binding pattern . In contrast , binding to cardiomyocytes was uniform and revealed a profile expected for a single population of cells . Experiments were next performed to determine the apparent Kd for binding of TcdBHIST and TcdBHV . Interestingly , within the constraints of these experimental conditions we were not able to achieve saturable binding of either form of the toxin to target cells . Fig . 5C shows a nearly linear correlation between the increase in toxin concentration and the mean fluorescence intensity ( MFI ) of HL-1 cells despite reaching toxin concentrations of over 300 nM . Additionally , Fig . 5C further emphasizes the extremely low level of interaction of TcdBHV with target cells in comparison to the high MFI achieved with TcdBHIST . These data suggest that cell binding involves a higher order and more complex process than expected for a single receptor-ligand interaction . Experiments were next performed to assess the difference in the rates of cell entry between the two toxins . In previous work on historical TcdB , we found that lysosomotropic inhibitors could completely block cytopathic effects of the toxin for up to 16 h , even if added up to 20 min following exposure of the cells to the toxin [41] . These findings indicate interaction with the cell , uptake , and then translocation into the cytosol requires at least 20 min and acidification of endosomes is necessary . To determine if TcdBHV differed from TcdBHIST in rates of cell entry , cultured CHO cells were treated with the two forms of the toxin and a lysosomotropic agent was added to the cells at time-points ranging from 5 to 60 min following treatment with toxin . The lysosomotropic agent was also added prior to or at the same time cells were treated with the toxins . The effect of the lysosomotropic agent was then assessed by determining the level of cytopathic effects ( CPE ) either 2 h or 12 h after treatment with the toxin . For this experiment CPE was determined rather than cytotoxicity due to toxicity of ammonium chloride at the later time points necessary for cytotoxicity assays . As shown in Fig . 6 , based on the extent of cell rounding , there appeared to be a clear difference in the rates of translocation between TcdBHV and TcdBHIST . Unlike our earlier findings on TcdBHIST , the cytotoxic effects of TcdBHV could not be prevented when the lysosomotropic agent was added as soon as 10 min following treatment with the toxin ( Fig . 6A ) . Furthermore , addition of the lysosomotropic agent within 10 min of treatment of TcdBHV only provided a slight delay in CPE , as all inhibitor treated cells showed complete rounding by 12 h ( Fig . 6B ) . In contrast , the CPE of TcdBHIST could be prevented by adding the inhibitor up to 30 min following treatment with the toxin . These findings indicate TcdBHV translocates to the cytosol more rapidly than TcdBHIST . Previous studies from our group found that acidic pH triggers hydrophobic transitions in TcdBHIST [41] . Studies by Barth et al . found that this hydrophobic transition in TcdB correlated with membrane insertion by the toxin [42] . These conformational changes corresponded to the decrease in endosome pH that led to translocation of the toxin into the cytosol . Thus , it was reasonable to suspect that TcdBHV translocates more quickly into the cytosol because the hydrophobic transition was induced at a higher pH and thus at an earlier stage of endocytosis . To address this possibility , in the next series of experiments we identified the pH dependent conformational transitions of TcdBHV by observing changes in TNS fluorescence when the toxin was incubated at various pHs . To identify whether TcdBHV exhibits differential transitions compared to TcdBHIST , the proteins were preincubated with 150 µM TNS at pH 4 . 0 , 5 . 0 , 6 . 0 , and 7 . 0 , and then analyzed for changes in TNS fluorescence . As shown in Fig . 7 , TcdBHV exhibited a significant increase in hydrophobicity at pH 5 . 0 , while TcdBHIST did not undergo this transition until pH 4 . 0 . Further examination of a narrower pH range revealed that a significant shift occurred between pH 5 . 4 and 5 . 6 in TcdBHV ( Fig . 7D ) . In comparison , TNS fluorescence of TcdBHIST at these pHs was just above background levels . These pH transitions were also studied using the inherent fluorescence of TcdBHIST and TcdBHV from the emission of tryptophan residues . Unfolding of the hydrophobic region should expose portions of the protein to a more aqueous environment , quenching tryptophan fluorescence . Environmental changes surrounding the tryptophan residues over a broad range of pH are shown in Fig . 8A and 8B . A gradual quenching of fluorescence was detected in TcdBHIST from pH 7 to pH 4 , while the tryptophan emission spectra of TcdBHV indicated a sudden shift between pH 5 and pH 6 . Fig . 8D reveals that this shift took place between pH 5 . 4 and 5 . 2 , similar to the increase in TNS fluorescence seen at pH 5 . 4 .
In the current study we compared the sequences and activities of TcdB from hypervirulent and historical strains of C . difficile . Because TcdB has been shown to be the major virulence factor of C . difficile [43] , we reasoned that changes in the activity of this toxin could have a profound impact on the severity of disease . The findings support this notion , as TcdBHV exhibited a broader tropism and higher potency than TcdBHIST . Among the possible explanations for this increased toxicity are the observations that TcdBHV enters cells more rapidly than TcdBHIST , and TcdBHV undergoes conformational changes at a higher pH than TcdBHIST . Based on the sequence comparisons and the results of the experiments using the heterologous delivery system ( Figs . 1 and 3 ) , it appears that the differences in tropism and cytotoxicity are due to changes in regions outside of the enzymatic domain . Rapid cell entry could lead to more efficient cell killing by providing the toxin an endocytic condition in which the toxin is not subject to possible destruction by lysosomal proteases . The data from the lysosomotropic inhibitor assays ( Fig . 6 ) support the idea that TcdBHV does not reside within the endosome as long as TcdBHIST . Among the possible reasons for more rapid cell entry is a differential sensitivity to levels of IP6 that trigger autoproteolytic processing associated with translocation . We also noted a difference in the sequence of the hydrophobic region of TcdB , and if , as has been proposed [41] , [42] , this region mediates membrane insertion , such differences could allow TcdBHV to insert into the membrane at an earlier stage of cell entry . We reasoned that if this possibility were true , there should be a difference in the pH-induced transitions of the two forms of TcdB , with the hydrophobic regions of TcdBHV becoming exposed at a pH higher than the pH necessary for triggering this transition in TcdBHIST . The results from the TNS experiments ( Fig . 7 ) indicate that TcdBHV is able to undergo the hydrophobic transition at a higher pH than TcdBHIST , providing further evidence that TcdBHV has higher translocation efficiency than TcdBHIST . Studies looking at the environment surrounding tryptophan residues of TcdBHIST and TcdBHV at lower pH ( Fig . 8 ) support the idea that TcdBHV undergoes a structural change at higher pH than TcdBHIST . Additionally , these experiments revealed that the transition of TcdBHIST occurs gradually , while TcdBHV demonstrates sudden shifts upon lowering the pH . This could be indicative of a more efficient unfolding of TcdBHV , which may contribute to an enhanced ability to traverse the endosomal membrane . Our current working model is that TcdBHV is able to translocate at an earlier point in endocytosis and this contributes , at least in part , to a more efficient intoxication . We also recognize that the expanded tropism , along with more efficient cell entry could combine to enhance the in vivo toxicity of TcdBHV . The results from the zebrafish experiments ( Fig . 2 ) indicate TcdBHV targets a broader array of cells in vivo than does TcdBHIST . Defining the specific tropism in the murine model or an infection model is more difficult , but it is reasonable to consider the possibility that TcdBHV is more lethal because the toxin targets an extensive variety of cell types systemically . Unfortunately , the TcdB receptor has been difficult to identify . Several attempts by our group to identify the TcdB receptor using standard techniques that have been successful with other toxins have failed . The results from the flow analyses in the current study suggest that the interaction of TcdB with the cell surface does not fit a single ligand-receptor model; this observation may explain why it has been so difficult to identify a receptor for this toxin . We were not able to achieve saturable binding , and interestingly TcdBHV interacted less efficiently than TcdBHIST , despite the fact that TcdBHV is clearly more cytotoxic than TcdBHIST . Undoubtedly , future studies on characterizing this complex interaction with target cells will provide important insight into a novel mechanism of TcdB intoxication . Previous work by Razaq et al . found that C . difficile BI/NAP1/027 strains were more lethal than historical strains of C . difficile [44] . As mentioned in the introduction of this paper , there are several differences in the phenotypes of the hypervirulent and historical strains of C . difficile . NAP1 strains sporulate at a higher efficiency and are resistant to fluoroquinolones . Both of these characteristics may make the NAP1 strains more difficult to manage in the hospital setting and increase the frequency of disease , but are unlikely to increase virulence . Likewise , the binary toxin has been shown to enhance colonization [13] , but clinical data have revealed little correlation between the increase in disease severity and production of this toxin [45] , [46] . In addition , previous work found binary toxin to be enterotoxic , but strains producing binary toxin alone did not cause disease in hamsters [47] . Clearly , an increase in toxin production such as that reported for NAP1 strains could enhance virulence , but a recent report suggests that the tcdC mutation in epidemic strains does not always correlate with the overexpression of TcdA and TcdB [48] . Based on the findings from the current study , we suggest that variations in TcdB sequence and activity could be an important determining factor in the hypervirulence of NAP1 strains . The recent work of Lyras et al . [43] found that TcdB is critical to C . difficile virulence in a hamster model of CDAD . Thus , variations in the antigenic region ( e . g . carboxy terminus ) of TcdB could allow repeated C . difficile infections of the same host by strains with antigenic variants of this toxin . In a recent publication by He and colleagues it was estimated that C . difficile diverged into a distinct species between 1 . 1 and 85 million years ago , and has gone through remarkable genetic variation over time [49] . The authors also posited that immune selection could have influenced the genetic variation , and they examined candidate immunogenic proteins that might fit this profile and 12 such proteins were identified . TcdB was not among these candidate proteins . It is unclear whether TcdB fits the criteria established for a positively selected core gene of C . difficile in this study , but it is reasonable to suspect the gene may have varied to avoid immune responses and this hypervariability enriched for a more potent form of the toxin . It is worth noting that while the protein identity was around 92% , the DNA homology was 93% . Nearly all of the residue changes occur as a single nucleotide substitution that result in amino acid substitutions . This further suggests a possible change in the sequence of TcdB that has been selected through an enhancement in virulence and perhaps by immune evasion .
Chinese hamster ovary-K1 ( CHO ) cells were maintained in F-12K medium ( American Tissue and Culture Collection; ATCC ) along with 10% fetal bovine serum ( ATCC ) . HL-1 cardiomyocytes were obtained from the Claycomb laboratory [38] and maintained in Claycomb medium ( Sigma ) supplemented with 10% fetal bovine serum ( ATCC ) , 0 . 1 mM Norepinephrine ( Sigma ) , and 2 mM L-glutamine ( Invitrogen ) . Cultures were grown at 37°C in the presence of 6% CO2 . C . difficile VPI 10463 ( produces TcdB with identical sequence to the 630 strain ) and C . difficile BI17 6493 ( a gift from Dr . Dale Gerding ) , were used in this study for the purification of TcdBHIST and TcdBHV . The tcdB gene was sequenced from both of these strains and the sequence was confirmed as exact matches to Genbank deposited sequences of strain 630 and R20291 ( Genbank numbers AM180355 and FN545816 ) . Cultures were grown as previously described [41] , and TcdB was isolated by consecutive steps of anion-exchange ( Q-Sepharose ) and high-resolution anion-exchange ( Mono-Q ) chromatography in 20 mM Tris-HCl , 20 mM CaCl2 , pH 8 . 0 . Purification steps were followed by protein determination using the Bradford method , visualization of a single band by SDS-PAGE , and LC/MS/MS analysis ( University of Oklahoma Health Science Center ) to confirm protein identity . Cytotoxicity was determined using a WST-8 [2- ( 2-methoxy-4-nitrophenyl ) -3- ( 4-nitrophenyl ) -5- ( 2 , 4-disulfophenyl ) -2H-tetrazolium , monosodiumsalt] ( Dojindo Laboratories ) according to manufacturer's instructions . Zebrafish maintenance and experiments were performed in accordance with the PHS Principles for The Utilization and Care of Vertebrate Animals Used in Testing , Research , and Training , and followed the recommendations in the Guide for the Care and Use of Laboratory Animals under the approval of The University of Oklahoma Health Sciences Center Campus IACUC ( OUHSC #06-126 ) . Zebrafish were obtained from Aquatic Eco-System ( Apopka , FL ) . Zebrafish were maintained at 28 . 5°C on a 14 h light/10 h dark cycle in 10 gallon tanks equipped with pumps for mechanical and chemical filtration . Matings were performed in false bottom tanks , and embryos were washed briefly with 0 . 5% bleach after collection . Embryos were incubated in embryo water ( 60 mM NaCl , 1 . 2 mM NaHCO3 , 0 . 9 mM CaCl2 , 0 . 7 mM KCl ) in petri dishes at 28 . 5°C , and water was changed daily . For TcdB treatment experiments , embryos were used between 48 and 72 h post fertilization , with chorions removed . Embryos were placed ( 5 embryos per well ) into 48-well plates and treated with TcdBHIST or TcdBHV in embryo water at concentrations ranging from 50 nM to 0 . 01 nM . The embryos were observed for 72 h after treatment for morphological changes by using a SZX-7 microscope with a DP70 camera ( Olympus ) . All images were captured and processed by using DP controller and DP manager software ( Olympus ) . The region encoding the enzymatic domain of TcdBHV was amplified from C . difficile NAP1 genomic DNA by PCR using the forward primer 5′-ACGTCCCGGGATGAGTTTAGTTAATA-3′ and the reverse primer 5′-ACTGGATCCTCATTATACTGTATTTTG-3′ to generate the tcdB gene fragment encoding residues 1 to 1668 of tcdB ( tcdB1–1668 ) with a 3′ XmaI/SmaI and a 5′ BamHI site . The restricted gene fragment was fused to lfn by overnight ligation at 16°C with a Xma1/BamHI-restricted pET15b derivative containing lfn . The resulting plasmid was cloned into Escherichia coli XL-1 blue ( Novagen ) and candidate clones were screened for the correct insert and orientation by restriction analysis and DNA sequencing . LFnTcdBHIST ( enz ) which had been previously cloned and described [40] and the newly synthesized LFnTcdBHV ( enz ) were expressed using E . coli BL-21 Star ( Invitrogen ) . Both fusions were purified by Ni2+ affinity chromatography ( His-Trap , GE Life Sciences ) and the purified protein migrated within the predicted size range of ∼94 kDa on SDS-PAGE . Protective antigen was expressed and purified as previously described [50] . TcdBHIST or TcdBHV were labeled with Alexa Fluor 647 C5 maleimide ( Invitrogen ) according to manufacturer's instructions . Briefly , a 10 M excess of dye was added to TcdB in 20 mM Tris-HCl , pH 8 . 0 , and incubated overnight at 4°C . Conjugated protein was separated from unincorporated dye using Sephadex G-25 , and efficiency of labeling was confirmed to be between 80% and 100% . The activity of labeled TcdB was confirmed by cytotoxicity on CHO and HL-1 cells and was not reduced by >10% . Binding of each toxin to CHO and HL-1 cells was examined as follows . Cells were dissociated from flasks using 1 mM EDTA in PBS , centrifuged at 500× g , and washed once with PBS . One hundred thousand cells were incubated with a range from 10 nM to 320 nM of labeled toxin in 1 mL of PBS on ice for 1 h , washed twice , and the pellets were resuspended in 1 mL of PBS . The samples were analyzed using a FACSCalibur flow cytometer ( University of Oklahoma Health Sciences Center ) and FLOWJO software ( Tree Star , San Carlos , CA ) . The emission wavelength was set to 665 nm , and the excitation was set at 633 nm with a bandpass of 30 nm . CHO cells were plated at 5×104 cells/well in a 96-well plate and incubated overnight . The following day , TcdBHIST or TcdBHV was added to the cells at a final concentration of 0 . 1 µg/mL . At the indicated time points , the cells were washed to remove unbound toxin , and ammonium chloride ( Sigma ) was added to the cells at final concentration of 100 mM . Each sample was monitored for 24 h , and CPE ( cytopathic effect ) was determined by visualization . Percent CPE was calculated by counting a minimum of 100 cells in 3 different fields for each sample . Cells scored positive for CPE only when fully rounded , and the percent CPE was calculated as % rounded cellstest - % rounded cellscontrol , where control refers to cells treated with media alone . 2- ( p-Toluidinyl ) naphthalene-6-sulfonic acid , sodium salt ( TNS; Invitrogen ) solutions were prepared to a final concentration of 150 µM in pH specific buffers . For pHs ranging from 4 . 0 to 6 . 0 , 100 mM NaCl-100 mM ammonium acetate-1 mM EDTA was used . For pH 6 . 0 to 7 . 0 , 100 mM NaCl-100 mM MES-1 mM EDTA was used . For pH 7 . 0 to 8 . 0 , 100 mM NaCl- 100 mM HEPES-1mM EDTA was used . 40 pmol of TcdBHIST or TcdBHV was added to the buffer/TNS mixture in a final volume of 2 . 5 mL and allowed to incubate for 20 min and 37°C . Each sample was analyzed on a Fluorolog R928P PMT fluorometer ( HORIBA Jobin Yvon ) with an excitation of 366 nm and an emission scan of 380 to 500 nm with a slit width of 2 . 0 . Tryptophan fluorescence of TcdBHIST and TcdBHV was also compared in the same manner , using an excitation of 270 nm and an emission scan of 310 nm to 400 nm . Data are expressed as the means ± S . E . M . Statistical analyses were performed using two-tailed unpaired Student's t-test in GraphPad Prism ( La Jolla , CA ) . Statistical significance is indicated as * p<0 . 05; ** p<0 . 01; *** p<0 . 001 . | Clostridium difficile is a spore-forming bacterium that contaminates hospitals and infects patients undergoing antibiotic therapy . C . difficile is now the leading cause of hospital-acquired diarrhea in developed countries . Most concerning has been the recent increase in mortality of C . difficile patients due to the emergence of a hypervirulent strain of this pathogen . Results from the current study suggest this change in disease severity may be due to new strains producing a variant form of C . difficile's major virulence factor , TcdB . The findings indicate TcdB from hypervirulent strains targets a much broader range of cells in vivo and is able to translocate into target cells more quickly than TcdB from historical strains of C . difficile . The more rapid cell entry by TcdB from hypervirulent C . difficile appears to be due to the toxin's capacity to undergo conformational changes necessary for membrane translocation at a higher pH than TcdB from historical strains . To date , very little has been learned about the underlying reasons for the increased virulence of emerging C . difficile strains . These findings provide insight into this problem and suggest variations in TcdB activity could be an important contributing factor to the hypervirulence of emerging strains of C . difficile . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"infectious",
"diseases/bacterial",
"infections",
"microbiology/medical",
"microbiology"
]
| 2010 | Variations in TcdB Activity and the Hypervirulence of Emerging Strains of Clostridium difficile |
Lentiviruses are able to establish persistent infection in their respective hosts despite a potent type-I interferon ( IFN-I ) response following transmission . A number of IFN-I-induced host factors that are able to inhibit lentiviral replication in vitro have been identified , and these studies suggest a role for IFN-induced factors as barriers to cross-species transmission . However , the ability of these factors to inhibit viral replication in vivo has not been well characterized , nor have the viral determinants that contribute to evasion or antagonism of the host IFN-I response . In this study , we hypothesized that the host IFN-I response serves as a strong selective pressure in the context of SIV/HIV chimeric virus ( SHIV ) infection of macaques and sought to identify the viral determinants that contribute to IFN-I resistance . We assessed the ability of SHIVs encoding HIV-1 sequences adapted by serial passage in macaques versus SHIVs encoding HIV sequences isolated directly from infected individuals to replicate in the presence of IFNα in macaque lymphocytes . We demonstrate that passage in macaques selects for IFNα resistant viruses that have higher replication kinetics and increased envelope content . SHIVs that encode HIV-1 sequences derived directly from infected humans were sensitive to IFNα –induced inhibition whereas SHIVs obtained after passage in macaques were not . This evolutionary process was directly observed in viruses that were serially passaged during the first few months of infection–a time when the IFNα response is high . Differences in IFNα sensitivity mapped to HIV-1 envelope and were associated with increased envelope levels despite similar mRNA expression , suggesting a post-transcriptional mechanism . These studies highlight critical differences in IFNα sensitivity between HIV-1 sequences in infected people and those used in SHIV models .
The innate immune system presents the first host defense against viral infection . Host cells are able to sense the presence of viral infection and respond by producing type-I interferon ( IFN-I ) , which , in turn , leads to the up-regulation of hundreds of host genes that are potentially antiviral [1 , 2] . Infection with HIV-1 in people and SIV in non-human primates induces a robust IFN-I response within days of infection [3–7] . IFN-I production , including IFNα , is part of a larger systemic cytokine storm that precedes the establishment of set-point viral load suggesting that the IFN-I response may play a role in limiting viral replication during acute infection and influence disease progression [8] . In support of this hypothesis , a recent study of SIV infection in rhesus macaques demonstrated that blocking the IFN-I response resulted in higher plasma viral loads during acute infection , increased reservoir size and faster progression to AIDS [9] . Despite the presence of a robust antiviral IFN-I response to infection , lentiviruses are able to replicate to high levels during acute infection and establish persistent infection in their hosts . Some recent studies have provided evidence that the innate immune response selects for HIV-1 variants that are relatively resistant to IFN-I during transmission [10 , 11] . The biological properties that contribute to the ability of some HIV-1 variants to resist the IFN-I response remain unclear . One possible explanation for differences in IFN-I sensitivity of HIV-1 variants is that they have different abilities to evade or antagonize downstream effectors of the IFN-I response . Over the last decade , host antiviral proteins , referred to as restriction factors , have been identified that act at multiple stages of the lentiviral life cycle and directly inhibit viral replication [8 , 12] . Many of the restriction factors are induced by IFN-I [8 , 12] . Because the IFN-I-induced factors are effective at inhibiting viral replication , lentiviruses have evolved mechanisms to evade or antagonize their activity . Indeed , the human orthologs of the IFN-I-induced restriction factors that inhibit HIV-1 replication are largely inactive against HIV-1 because of the specificity of the viral antagonist for the human protein . The mechanisms of restriction factor inhibition and viral antagonism and the importance of these interactions for establishing productive infection in vitro have been carefully elucidated . However , the role of the IFN-I response in limiting viral replication and mechanisms of viral evasion/antagonism in the context of infection in vivo is less clear . Relevant to this , HIV-specific restriction factors have been largely studied for their ability to inhibit HIV-1 variants derived after passage in cell culture and less is known about the IFN-I-induced responses that inhibit viruses replicating in infected individuals . Due to the selective pressure of restriction factors , lentiviral proteins are adapted to antagonize factors in their respective hosts and often act in a species-specific manner [8 , 13] . For example , HIV-1 proteins are able to antagonize human restriction factors but are unable to effectively counteract the non-human primate orthologs . For this reason , SIV/HIV chimeric viruses ( SHIVs ) used to study HIV-1 pathogenesis in macaques encode SIV antagonists of well-characterized macaque restriction factors . The HIV-1 genes encoded in SHIVs typically include the env gene that encodes the Envelope surface glycoprotein ( Env ) . In most cases , SHIVs require multiple rounds of adaptation in lab-culture and/or by animal-animal serial passage in macaques in order to increase replication capacity and pathogenicity [14] . Often the process of animal-animal serial passage is performed within the first two weeks of infection when levels of IFN-I are highest in the animals [4 , 5 , 15–17] , providing a possible selective pressure to drive changes in the virus . Thus , SHIVs that have been adapted in macaques present a unique opportunity to study the mechanism of adaptation to IFN-I response . The goals of this study were to determine whether the process of adapting SHIVs for increased replication capacity and pathogenicity in macaques selects for variants that are resistant to the host’s IFN-I response and to identify the biological changes in the virus that contribute to IFN-I resistance . Given the fact that the majority of SHIVs studied to date encode HIV-1 variants derived from cell culture and represent the select group of SHIVs that were able to infect macaques , we also asked whether there are differences in IFN-I sensitivity of these viruses compared to SHIVs encoding HIV-1 sequences isolated directly from infected individuals . We demonstrate that envelope differences selected in vivo allow SHIVs to adapt to the IFNα response; adapted HIV-1 variants encode IFNα resistant Envs , whereas Envs obtained directly from infected individuals early in their infection are sensitive , suggesting that IFNα may have an inhibitory effect on viruses spreading in humans that has not been observed through the study of adapted viruses . These findings suggest that Env plays an important role in evading or antagonizing the IFNα response . Identification of IFNα resistant HIV-1 Env variants may facilitate the development of challenge viruses for macaque models of HIV-1 infection .
In order to test the hypothesis that adaptation of SHIVs results in IFN-I resistance , we compared a panel of nine SHIVs for their ability to replicate in macaque cells in the presence or absence of IFNα . The panel of SHIVs included four viruses that encode HIV-1 sequences isolated directly from infected individuals ( MG505 , Q23 , QF495 , and BG505 ) with the latter three from early infection [18–20] , two viruses that encode HIV-1 sequences obtained from individuals during late-stage chronic infection and adapted in lab-culture in human cells ( AD8 and SF162 ) [21 , 22] and three viruses that encode HIV-1 sequences adapted by animal-animal passage in macaques ( AD8-EO , SF162P3 and 1157ipd3N4 ) , two of which represent the animal-passaged derivatives of the lab-adapted viruses [16 , 21 , 23] ( S1 Table ) . Most HIV-1 variants circulating in people are unable to use the macaque CD4 receptor for entry into cells [24] , therefore , SHIVs that were made from HIV-1 variants isolated directly from individuals encode single amino acid changes that allow them to use the macaque CD4 receptor for entry . HIV-1 variants encoding these changes are able to use the macaque CD4 for entry at levels similar to those of adapted SHIVs [24] , and the viruses chosen for study represent those that were infectious in macaque CD4+ T cells . Otherwise , the HIV-1 sequences of these SHIVs are unmodified from the sequences found in the infected individual and will be referred to as circulating SHIVs as they are representative of HIV-1 variants circulating in human populations . We assessed the ability of the panel of SHIVs to replicate in immortalized pig-tailed macaque ( Ptm ) CD4+ lymphocytes [25] in the presence and absence of IFNα-2a at a concentration similar to that observed in natural infection ( 1000 U/ml ) [4 , 5] ( Fig 1A , S1 Fig ) . Intracellular staining for two IFNα-stimulated proteins , MX1 and IFIT1 , showed that nearly all immortalized Ptm lymphocytes responded to IFNα treatment ( S2 Fig ) . IFNα sensitivity was measured as the ratio of the area-under curve ( AUC ) of the replication curve in the IFN-treated cells to the AUC of the replication curve in the untreated cells . For example , SHIV AD8-EO , a pathogenic , macaque-passaged virus , replicated in the presence of IFNα nearly as well as the untreated cells ( Fig 1A ) and had an AUC ratio ( IFN+/IFN- ) of 0 . 96 . In contrast , SHIV Q23AE , a circulating SHIV , exhibited a pronounced IFNα-induced inhibition of viral replication corresponding to an ~100-fold reduction in SIV p27 levels at nine dpi and had an AUC ratio of 0 . 68 ( Fig 1A ) . The other seven viruses exhibited a range of inhibition between these two ends of the spectrum ( S1 Fig ) . We observed the same patterns of IFNα-induced inhibition when we pre-treated the cells with IFNα-2a at 24 hours prior to infection ( S3 Fig ) . Overall , SHIVs adapted by macaque-passage and by lab-culture exhibited higher AUC ratios than circulating SHIVs indicating resistance to IFNα treatment ( Fig 1B ) . Comparing AUC ratios , macaque-passaged SHIVs were significantly more resistant to IFNα treatment compared to circulating SHIVs ( 0 . 94 vs . 0 . 78 , p = 0 . 05 ) . The replication kinetics of the nine SHIVs were defined using the data from the replication time course studies where replication differences between viruses were evident even in the absence of IFNα . For example , SHIV AD8-EO demonstrated rapid replication kinetics in untreated cells reaching peak virus levels of >106 pg/ml of SIV p27 by six dpi ( Fig 1A ) . In contrast , SHIV Q23AE reached lower peak virus levels of >105 pg/ml of SIV p27 at nine dpi . In order to compare replication kinetics across the panel of nine SHIVs , we determined a summary measure of viral replication based on the slope of a best-fit straight line of the replication curve in untreated immortalized Ptm lymphocytes during the first six days of infection . Comparing replication slopes , macaque-passaged SHIVs replicated faster than circulating SHIVs ( 0 . 98 vs . 0 . 75 , p = 0 . 05 ) , and lab-adapted SHIVs were more similar to the animal-adapted SHIVs ( Fig 1C ) . There was a strong positive correlation between replication kinetics and IFNα resistance as measured by the AUC ratio ( Spearman r = 0 . 88 , p = 0 . 003 ) ( Fig 1D ) . Representative macaque-passaged and lab-cultured SHIVs also exhibited faster replication than circulating SHIVs in primary Ptm PBMCs ( Fig 1E ) . Thus , macaque-passaged SHIVs exhibited higher replication kinetics and greater IFNα resistance compared to the circulating SHIVs , and the replication kinetics in untreated immortalized Ptm lymphocytes correlate with the ability of the virus to overcome the IFNα response . In order to identify the viral determinant ( s ) of replication capacity and IFNα sensitivity , we generated chimeras between the viruses that exhibited the greatest difference in replication kinetics and IFNα sensitivity , SHIV AD8-EO , which is a prototype animal-adapted SHIV , and SHIV Q23AE , which represents a circulating SHIV . Because SHIV Q23AE was generated by cloning full HIV-1 vpu and env genes directly into SHIV AD8-EO [21] , the SHIV Q23AE and SHIV AD8-EO are isogenic except for the HIV-1 genes vpu , env and the second exons of tat/rev; thus , biological differences between them must be due to the HIV-1 sequences . Introduction of the entire env gene from SHIV AD8-EO to SHIV Q23AE resulted in complete recovery of replication capacity ( Fig 2A ) . Introduction of the gp120 surface subunit of Env resulted in a modest increase in replication kinetics while introduction of the gp41 trans-membrane subunit did not result in any detectable increase in replication ( Fig 2A and 2B ) . Because regions of the env gene contain overlapping reading frames with vpu and tat/rev , we also introduced the full vpu gene and the second tat exon , including a portion of rev , from SHIV AD8-EO to SHIV Q23AE . Introduction of neither vpu nor the second tat/rev exon resulted in a significant increase in replication kinetics ( Fig 2B ) . Introduction of the full env gene from SHIV AD8-EO to SHIV Q23AE also restored high-level replication capacity in primary Ptm PBMCs where differences in replication are similar to those of the immortalized Ptm lymphocytes ( Fig 2C ) . The chimeras were then examined for their ability to replicate in the presence of IFNα treatment . Introduction of the entire HIV-1 env gene from SHIV AD8-EO to SHIV Q23AE resulted in a nearly complete recovery of IFNα resistance ( Fig 2D ) . The gp120 chimera exhibited a modest increase in IFNα resistance while neither the gp41 nor vpu chimera demonstrated any detectable increase in IFNα resistance . Because of the poor replication capacity of the tat/rev chimera , we were unable to determine its IFNα sensitivity using the AUC ratio . Thus , the HIV-1 env gene is a major determinant of replication and of resistance to the IFNα response in Ptm cells . To address the basis for the effect of Env on replication capacity , we measured the amount of Env protein present in virions harvested from infected immortalized Ptm lymphocytes for the panel of nine SHIVs . For the day six viral lysates , the levels of HIV-1 Env in virus expressed from cells infected with the SHIVs adapted in lab-culture or by animal-passage were consistently higher than that of virus from cells infected with the four circulating SHIVs ( Fig 3A ) . These differences are exemplified by SHIV AD8-EO and SHIV Q23AE where there was a >10-fold difference between HIV-1 Env relative to SIV Gag p27 . These patterns of Env content were similar at nine dpi ( Fig 3B ) . In the purified virion lysates , we did not observe evidence of contamination from infected cells , for example the presence of unprocessed Gag , although we cannot definitively rule out very low levels of infected cell debris . Because the panel of SHIVs encodes HIV-1 Envs from diverse subtypes , we probed for Env using two different primary antibodies , a polyclonal rabbit sera from animals immunized with a subtype A Env protein [26] and HIVIG , antibodies pooled from HIV-1+ patients ( NIH AIDS Reagent Program ) . We observed the same patterns of Env content using either of the primary antibodies indicating that differences in HIV-1 Env detection were not due to differences in antibody recognition of the diverse proteins ( S4 Fig ) . In addition to virion-associated Env content , we determined the relative infectivity of prototype macaque-passaged ( AD8-EO ) and circulating ( Q23AE ) SHIVs by measuring TCID50 in immortalized Ptm lymphocytes ( S5 Fig ) . We found that the TCID50 values were very similar between the two viruses and in each case about 100-fold lower than the titer defined using TZM-bl cells . Thus , while the input levels of infectious virus were lower based on the TCID50 assay ( MOI of 0 . 0002 rather than 0 . 02 ) , the infecting virus titer was similar between the two viruses in our experiments . Interestingly , when we normalized TCID50 to p27 levels , we found that the infectivity of SHIV AD8-EO was approximately 30-fold higher than SHIV Q23AE ( S5 Fig ) suggesting that SHIVAD8-EO may have a higher ratio of infectious to non-infectious particles than SHIV Q23AE . Thus , the approach of using virus titer rather than p27 levels provided the best approach to normalizing infectious virus input ( S5 Fig ) . We next compared virion-associated Env content to replication kinetics across all nine viruses from Fig 3A in immortalized Ptm lymphocytes . We observed a strong positive correlation between the replication slope and Env content of virions produced at six dpi ( Spearman r = 0 . 90 , p = 0 . 002 ) ( Fig 3C ) . Overall , SHIVs adapted by macaque-passage and in lab-culture had higher virion Env content and higher replication kinetics compared to circulating SHIVs . Considering our previous finding that replication kinetics positively correlated with the ability to overcome the IFNα response , we compared virion-associated Env content and AUC ratio ( IFN+/IFN- ) . We observed a positive correlation between Env content in virions and resistance to IFNα treatment ( Spearman r = 0 . 78 , p = 0 . 02 ) suggesting that Env content is linked to the ability to overcome the IFNα response ( Fig 3D ) . The observed differences in Env content in SHIV virions could be due to variation in synthesis within the infected cells or to variation in incorporation into newly formed virions . In order to address these two possibilities , we measured the amount of HIV-1 Env expressed in SHIV-infected immortalized Ptm lymphocytes . The pattern of Env detection was similar in cells as in virions: at six dpi , there was higher steady state Env levels in cells infected with SHIVs adapted by macaque-passage or in lab-culture compared to circulating SHIVs despite detection of PrGag at comparable levels ( Fig 4A ) , and the same was true at nine dpi ( Fig 4B ) . For example , SHIV AD8-EO had >30-fold ( 1 . 0 vs . 0 . 04 ) more HIV-1 Env relative to total Gag compared to SHIV Q23AE . For the nine SHIVs tested , there was a strong positive correlation between the relative steady state intracellular Env expression levels and relative virion-associated Env content at both six dpi ( Spearman r = 0 . 87 , p = 0 . 005 ) ( Fig 4C ) and nine dpi ( Spearman r = 0 . 83 , p = 0 . 009 ) ( Fig 4D ) suggesting that differences in Env content in virions are reflective of differences in Env levels in the infected cells . To address whether differences in intracellular HIV-1 Env expression are the result of variation in transcription and splicing of env mRNA , we measured spliced env mRNA and un-spliced viral genomic RNA by reverse transcriptase quantitative PCR ( RT-qPCR ) for SHIV AD8-EO and SHIV Q23AE . At three and nine dpi , we did not observe any difference between SHIV AD8-EO and SHIV Q23AE with respect to the amount of spliced env mRNA relative to un-spliced viral genomic RNA ( Fig 4E ) , although there was a statistically significant difference at six dpi; SHIV AD8-EO had ~1 . 5-fold more spliced env mRNA compared to SHIV Q23AE ( 5 . 6 vs . 3 . 8 , p = 0 . 04 ) . Given the small magnitude of this RNA difference compared to protein differences ( >30-fold ) and that differences were not observed at other time points where protein differences were observed , these findings suggest that low levels of intracellular HIV-1 Env expression are due to post-transcriptional events in SHIV-infected Ptm lymphocytes . The finding that animal-passaged SHIVs were the most IFNα resistant of all viruses tested suggested that in vivo adaptation results in increased resistance to the IFNα response . To test the hypothesis that the process of adapting SHIVs by serial animal-animal passage in macaques increases IFNα resistance of SHIVs , we examined a collection of related SHIVs derived from a parental SHIV encoding an HIV-1 Env variant obtained without culturing , similar to the other circulating SHIVs described above [17 , 27] . We tested the IFNα sensitivity of the parental molecular clone ( SHIVC109mc ) , two isolates from the third animal passage–one from early ( SHIVC109P3 ) and one from late in infection ( SHIVC109P3N ) –and an isolate from the fourth animal passage ( SHIVC109P4 ) . For these studies , we applied an assay that allowed us to determined the amount of IFNα-2a required to inhibit 50% of viral replication ( IFNα IC50 ) [11] as a more quantitative method to assess IFN sensitivity . Because these viruses were adapted for replication by serial passage in rhesus macaques , we first tested their IFNα sensitivity in primary Rhm PBMCs ( Fig 5A ) . The parental circulating SHIVC109mc was the most sensitive to IFNα ( 12 . 8 U/ml ) , with adapted SHIVs derived from passage of this virus showing 25–80-fold increased resistance . The three passaged viruses were generally similar in their sensitivity to IFNα to each other ( 340–1 , 000 U/ml ) , suggesting that adaptation to become resistant to IFNα occurred by the time of the third passage . Very similar results were observed in immortalized Ptm lymphocytes . The parental molecular clone SHIVC109mc was highly sensitive to IFNα treatment ( IFNα IC50 1 . 6 U/ml ) , similar to SHIV Q23AE ( Fig 5B ) . Each of the macaque-passaged isolates was more resistant to the IFNα response induced in macaque lymphocytes . Both of the isolates from passage three exhibited IFNα IC50 values >5 , 000 U/ml while the passage four isolate was moderately sensitive ( IFNα IC50 330 U/ml ) . The % Vres values , which measures residual virus replication at the highest IFN level tested , demonstrated similar patterns of IFNα resistance ( Fig 5C and 5D ) . Replication of the parental virus was nearly completely inhibited at the highest concentration of IFNα while the passage three isolates demonstrated higher residual replication . This result was consistent between both Rhm PBMCs and immortalized Ptm lymphocytes although overall residual replication was higher for the passaged isolates in Ptm cells . Interestingly , several SHIVs that were resistant to IFNα exhibited increased replication in the presence of IFNα . This increase in replication could be the result of proliferation of cells that were initially protected from infection at early time points but later became infected . In order to sample a larger collection of viruses and compare these measures of IFNα sensitivity between circulating and animal adapted viruses , we defined IFNα IC50 and % Vres values for the three animal-adapted and four circulating SHIVs examined at a single IFN concentration in Fig 1 . The seven viruses exhibited a range of IFNα IC50 values ( 1 . 7–5 , 000 U/ml ) . For example , SHIV Q23AE was highly sensitive to IFNα treatment and exhibited a dose-dependent inhibition of viral replication with an IC50 value of 1 . 7 U/ml ( Fig 6A ) . In contrast , SHIV AD8-EO was completely IFNα resistant and did not exhibit inhibition of viral replication at any of the concentrations; the other animal-passaged SHIVs were similarly IFNα resistant . The data from these SHIVs allowed us to compare the IFNα sensitivity of SHIVs encoding Envs isolated directly from people and macaque-passaged SHIVs from a total of 11 viruses . The macaque-passaged SHIVs ( n = 6 ) were significantly more resistant to IFNα compared to the circulating SHIVs ( n = 5 ) with respect to both IFNα IC50 value ( 3 , 180 vs . 10 . 6 U/ml , p = 0 . 0003 ) and % Vres ( 76 . 9 vs . 3 . 3% , p = 0 . 01 ) ( Fig 6B and 6C ) .
We took advantage of the SHIV macaque model to define the role of IFNα in infection and found that infection selects for viruses that are resistant to the inhibitory effects of IFNα in macaques . Pathogenic SHIVs , which have been developed to model HIV-1 transmission and pathogenesis in macaques , are resistant to IFNα , whereas the SHIVs based on HIV-1 variants circulating in humans , including transmitted viruses , are inhibited by IFNα . Differences in sensitivity to IFNα were determined by the HIV-1 Envelope protein , which is considered a key feature of the SHIV models . Our findings underscore critical differences between SHIVs adapted for replication in macaques and HIV-1 variants isolated directly from infected individuals , including those that were recently transmitted , which represent the most biologically relevant targets of HIV-1 vaccine and prevention efforts . Resistance to IFNα inhibition was associated with higher replication capacity , which resulted from the process of adapting virus in animals . Both increased replication and resistance to IFNα were observed in three SHIVs derived from animal passage compared to their corresponding parental molecularly cloned viruses . In all three cases , the process of adapting SHIVs in macaques , including during the critical window of early infection , led to increased replication and IFNα resistance . The differences were most striking when comparing virus derived from a SHIV constructed from HIV-1 sequences derived directly from an infected individual early in their infection [27] and pathogenic SHIVs derived from it by serial passage during the first few months of infection , which includes a time when the IFNα response is high [3–7] . Interestingly , the IFNα sensitivity of this panel of viruses mirrors those of the in vivo viral replication: the parental cloned virus , SHIVC109mc , was the most sensitive to IFNα treatment and demonstrated the lowest peak viremia . The viruses from the third animal passage ( SHIVC109P3 and SHIVC109P3N ) demonstrated the greatest resistance to IFNα and the highest peak viremia in vivo , 100–1 , 000-fold higher than the parental virus . The virus from the fourth animal passage SHIVC109P4 was intermediate in terms of IFNα sensitivity and peak viremia between the parental and passage three isolates . A variety of amino acid substitutions and deletions in variable ( V1V2 , V3 and V4 ) and constant ( C3 ) regions of Env occurred during adaptation [17] , and thus , there are many potential amino acid changes that could contribute to IFNα resistance . Importantly , selection for IFNα resistance was observed in both Rhm PBMCs and immortalized Ptm lymphocytes , suggesting a similar mechanism of IFNα inhibition in both macaques . Taken together with data showing that blocking the IFNα response early in infection results in faster progression to AIDS [9] , these results suggest that the pathogenicity of adapted SHIVs may in part reflect their selection for IFNα resistance . These findings also suggest that macaque IFNα response in vivo exerts a strong selective pressure on SHIVs . Two of the parental SHIVs examined here ( ADA , SF162 ) were derived from HIV-1 sequences that were isolated during late-stage chronic infection by passaging the virus in vitro [28 , 29] . These viruses showed greater resistance to IFNα than SHIVs encoding sequences obtained directly from infected individuals early in infection . Conditions under which these viruses were derived in culture may have selected for IFNα resistance , perhaps by selecting those viruses with increased replication kinetics . It is also possible that these HIV-1 variants were already selected for features that made them less sensitive to IFNα inhibition because they were derived from later in infection . The IFNα sensitivity of the SHIVs that have not undergone either cell culture or animal adaptation to IFNα inhibition may help explain why it has been so difficult to identify SHIVs that encode HIV-1 variants isolated directly from infected individuals and are pathogenic in macaques . Indeed , our studies predict that the rare pathogenic SHIVs derived directly from HIV-1 infected people [30–32] may represent a small subset of variants that are resistant to IFNα inhibition , potentially allowing them to antagonize the early IFN storm and seed the viral reservoir to establish a persistent infection . By generating chimeras between a pathogenic , macaque-passaged SHIV and a circulating SHIV encoding HIV-1 sequences isolated directly from an infected individual , we demonstrated that HIV-1 Env is a critical determinant of the ability of SHIVs to overcome the IFNα response . We found that the underlying mechanism for role of Env in IFNα sensitivity is related to Env protein levels . Interestingly , differences in Env protein levels were not reflected in env RNA levels . There was a ≤1 . 5 fold difference in spliced env RNA levels between adapted and circulating SHIVs compared to protein differences of >30-fold . These findings suggest that low levels of intracellular HIV-1 Env expression are due to post-transcriptional events in SHIV-infected macaque cells . While Env content correlated with IFNα sensitivity , we observed differences in the IFNα sensitivity of lab-cultured and macaque-passaged SHIVs despite similar levels of Env . Thus , while our results overall suggest that Env content plays an important role in the ability to overcome the macaque IFNα response , other determinants in the envelope protein may also contribute to IFNα resistance . The finding that IFNα resistance mapped to HIV-1 env and was the result of high Env content was somewhat surprising considering that HIV-1 Env has not previously been implicated in viral antagonism of the IFN-I response , although some studies have suggested that IFN-I treatment predominantly affects very early stages of viral replication [33 , 34] . These findings raise several interesting possibilities . One is that HIV-1 Envs from adapted SHIVs act directly in evasion or antagonism of the IFNα response by protein-protein interactions with an IFN-induced host factor ( s ) . An alternative hypothesis is that high HIV-1 Env expression/content contributes to increased kinetics allowing the virus to overcome the IFNα response by saturating IFN-induced factor ( s ) . There is some precedent for this model as the ability to saturate IFN-induced restriction factors has been demonstrated in vitro [35–38] . In support of the model that an IFN-induced inhibitory factor is being saturated , we found a strong positive correlation between replication kinetics and IFNα resistance . IFN–induced , HIV-specific restriction factors are typically species-specific , presumably because HIV-1 has adapted to its human host . Thus , we expect that the IFN-induced factor ( s ) that limit replication of circulating SHIVs , but not animal adapted SHIVs , may have similar species-specificity . However , we could not directly test whether this pathway is active in human cells with the viruses studied here because SHIVs replicate poorly in human cells due to other host restrictions . A similar correlation between replication capacity and IFNα resistance was recently reported for HIV-1 in human lymphocytes [39] . Differences in virion-associated Env content between SHIVs adapted by lab-culture and/or macaque-passage and those based on circulating variants were reflected by similar differences in Env levels in infected cells . Cells infected with adapted SHIVs had high Env levels whereas those infected with circulating SHIVs had low Env levels . Low Env content could help minimize immune recognition . For example , low Env content has been suggested to reduce antibody avidity [40] . Given that HIV-1 can spread as both cell-free and cell-associated virus [41] , one intriguing possibility to explain our finding is that high Env expression in infected cells is leading to increased cell-cell transmission and saturation of IFN-α inhibition of cell-cell virus spread . Interestingly , the pattern of IFNα inhibition was the same whether cells were pre-treated with IFNα or treated just after infection ( S3 Fig ) ; even in cells pretreated with IFNα , there is a delay in the inhibition due to IFNα . This could be a result of an effect on cell-cell transmission that is only seen after the initial round of cell-free virus infection . Alternatively , it may indicate the IFN-induced factor is packaged into virions and inhibits later rounds of infection . Finally , a technical explanation for the observed delay in IFNα inhibition could be that the p27 ELISA used to determine virus levels is not sensitive enough to detect differences at very low levels of viral replication . Overall , the results of this study demonstrate differences in the IFNα sensitivity between SHIVs used to model HIV-1 infection and HIV-1 variants circulating in infected people . They also suggest that the common focus on lab-adapted viral variants may have limited the ability to identify important mechanisms underlying the IFN-I-induced inhibition of HIV-1 variants circulating in people . The results uncover a key role for envelope in the process of adaptation of lentiviruses to the IFNα response and help explain why SHIVs generally do not cause pathogenic infections in macaques without adaptation . These findings may provide insight into the development of improved SHIV challenge viruses for non-human primate models of HIV-1 infection that currently serve as gatekeepers for HIV-1 vaccine and prevention studies .
Full-length proviral SHIV clones encoding the region spanning the vpu and env open reading frames were generated using SHIV AD8-EO as a vector [21] . Expression plasmids encoding vpu and env open reading frames for Q23ENV . 17 [18] , BG505 . W6M . ENV . B1 [20] , MG505 . W0M . ENV . H3 [20] , and QF495 . 23M . ENV . A3 [19] A204E and G312V variants were amplified using primers designed to introduce an EcoRI site 5’ of the vpu start codon and a SalI site immediately 3’ of the env stop codon . The amplicons were then digested and ligated into the SHIV AD8-EO full-length proviral plasmid using EcoRI and SalI . Chimeras between SHIV AD8-EO and SHIV Q23AE were generated by overlap-extension PCR . The following full-length proviral plasmids of the parental SHIVs were also used in this study: SHIV AD8 [21] and SHIV SF162 [22] . Full-length , replication-competent virus was generated by transfecting 2x106 HEK 293T cells ( American Type Culture Collection , Manassas , VA ) with 4 μg of proviral plasmid DNA and 12 μl of Fugene 6 transfection reagent ( Roche ) . Virus was harvested 48 hours post-transfection , passed through a 0 . 2 μm sterile filter and concentrated ~10-fold using a 100 kDa molecular weight protein concentrator ( Amicon ) . Replication-competent stocks of SHIV SF162P3 [23] , SHIV 1157ipd3N4 [16] , SHIVC109mc , SHIVC109P3 , SHIV109P3N and SHIVC109P4 [17] were generated by expanding the virus in immortalized Ptm lymphocytes [25] . For each virus , 2x106 cells were infected at an initial multiplicity of infection ( MOI ) of ~0 . 02 by spinoculation at 1200 x g for 90 minutes at room temperature . After spinoculation , cells were washed 1x with 1 ml of Iscove’s modified Dulbecco’s medium ( IMDM ) supplemented with 10% heat-inactivated FCS , 2 mM L-glutamine , 100 U of penicillin/ml , 100 μg of streptomycin/ml and 100 U of interleukin-2/ml ( Chiron ) ( complete IMDM ) , re-suspended in 2 . 4 ml of media and plated in a 6-well plate . Every three days , infected cells and cell supernatant were harvested and separated by pelleting at 650 x g for five minutes at room temperature . Aliquots of replication-competent virus were stored at -80°C . The viral titer of each viral stock was determined by infecting TZM-bl cells and counting the number of blue cells at 48 hours post-infection after staining for β-galactosidase activity [20] . Replication of SHIVs was assessed using immortalized Ptm CD4+ lymphocytes [25] maintained in complete IMDM . One million Ptm lymphocytes were infected at an MOI of 0 . 02 by spinoculation as described above . In some cultures , recombinant human IFNα-2a ( PBL Assay Science , Piscataway , NJ ) was added at a final concentration of 1 , 000 U/ml five hours after the initial infection . Every three days , 400 μl of each cell supernatant was replaced , including with IFNα-2a if appropriate . SIV p27 concentrations were determined using a SIV p27 antigen ELISA ( ABL , Rockville , MD ) . For some experiments Ptm or Rhm PBMCs were used: these cells were isolated from whole macaque blood ( Washington National Primate Research Center , Seattle , WA ) from two separate donors using 95% Lymphoprep ( STEMCELL , Vancouver , BC ) . Isolated PBMCs were stimulated for three days prior to infection with IL-2 ( 20 U/ml ) and concanavalin A ( 5 μg/ml ) in RPMI 1640 medium supplemented with 20% FCS and 2 mM L-glutamine . Ptm or Rhm PBMCs from two donors were pooled immediately prior to infection , and 1x106 PBMCs were spinoculated and maintained as described for immortalized Ptm lymphocytes . The amount of HIV-1 envelope ( Env ) was determined by semi-quantitative western blot using the LICOR Odyssey system . For virion-associated Env content , supernatants from infected immortalized Ptm lymphocyte cultures were pelleted through a 25% sucrose cushion by ultracentrifugation for 90 minutes at 28 , 000 rpm . Virus pellets were lysed in 70 μl of radioimmunoprecipitation assay ( RIPA ) buffer for 10 minutes at room temperature . The concentration of SIV p27 antigen in the virus lysates was determined by ELISA , and virus lysate input was normalized to 5 ng of p27 . Western blotting was performed as described previously [42] using rabbit polyclonal anti-HIV-1 Env sera [26] and mouse anti-SIV p27 monoclonal antibody ( ABL , catalog no . 4323 ) . Both gp160 and gp120 bands were included in the quantification of Env signal . For western blotting of whole cell lysates , SHIV-infected immortalized Ptm lymphocytes were pelleted by spinning at 650 x g for 5 minutes at room temperature . Cell pellets were washed with 1x PBS and then lysed in 100 μl of RIPA buffer for 10 minutes at room temperature . Western blotting was performed as described for virus lysates . HIV-1 RNA was measured by reverse transcriptase quantitative PCR ( RT-qPCR ) . Total RNA was isolated from SHIV-infected immortalized Ptm cells using the miRNeasy Mini Kit ( Qiagen ) . For each reaction , 50 ng of total RNA , measured by Nanodrop spectrophotometer was amplified using the Superscript III Platinum SYBR Green One-Step RT-qPCR kit with ROX ( Invitrogen ) . Primers 5’- AGGGACTTGGCAAATGGATTGTAC-3’ and 5’ GTGTAATAGGCCATCTGCCTGCC-3’ were used to amplify gag from unspliced RNA . To amplify splice vpu/env mRNA , the forward primer 5’-AGGAACCAACCACGACGGAGTGCTC -3’ , which binds upstream of the splice donor site in the 5’ LTR , and reverse primer 5’-CATTGCCACTGTCTTCTGCTCTTTC-3’ , which binds downstream of the Vpu start codon , were used . To amplify macaque β-actin mRNA , primers 5’-CAACCGCGAGAAGATGACCCAGATCATG-3’ and 5’-AGGATGGCATGGGGGAGGGCATAC-3’ were used . Relative levels of HIV-1 env mRNA for each virus was determined using the following equation: 2-ΔΔ = [ ( CT env—CT beta actin ) ]—[ ( CT genomic—CT beta actin ) ] [43] . For each SHIV , 4 . 25x106 Ptm or Rhm lymphocytes were infected at an MOI of 0 . 02 in a final volume of 1 . 4 ml of complete IMDM using spinoculation . Approximately 2 . 5x105 infected cells in 200 μl of media were plated in each well of a 96-well plate containing 50 μl of media containing the indicated concentrations of IFNα-2a; experiments were performed in duplicate . Cell supernatants were harvested at 7 days post-infection ( dpi ) and used to determine the amount of virus using TZM-bl cells ( NIH AIDS Reagent Program ) . For the data analysis , all values were plotted and statistical analyses performed using Prism version 6 . 0c ( GraphPad Software ) . Percent viral replication was determined by dividing the amount of β-galactosidase activity in the IFNα treated sample by the untreated sample . The concentration of IFNα at which 50% viral inhibition was achieved was interpolated from a non-linear , best-fit curve . The amount residual viral replication at the highest concentration of IFNα ( 5000 U/ml ) was also determined . TCID50 assay was used to determine the end-point dilution of the SHIV stocks at which infection is detected in 50% of the immortalized Ptm CD4+ lymphocyte culture replicates . Seven serial 4-fold dilutions of the SHIV stocks were prepared in triplicate in 96-well flat-bottomed tissue culture plates . Briefly , the SHIV stocks were diluted 1:12 in complete IMDM and 200 μl of diluted virus was transferred to the first well of a 96-well plate . Next , the virus was serially diluted by transferring 50 μl of the virus to the subsequent well containing 150 μl of complete IMDM . 2 x 105 immortalized Ptm CD4+ lymphocytes , in 50 μl of complete IMDM , were seeded in each well . Every 4 days 125 μl of the culture supernatant was removed from every well and replaced with fresh 150 μl of complete IMDM . On day 12 , 100 μl of the culture supernatant from each well was harvested and tested for SIV p27 using a SIV p27 antigen ELISA . TCID50 was calculated using the Spearman-Kaber method . | The innate immune system is an important host defense against viral infection . Recently , there has been significant interest in characterizing the innate immune response to HIV-1 infection , in particular the role of type-I interferon ( IFN-I ) . Understanding the interaction of HIV-1 with the innate immune system is particularly important for the development of animal models of infection as innate host factors present potential species-specific barriers to the establishment of persistent infection . One of the most commonly used animal models of HIV-1 infection is chimeric SIV/HIV ( SHIV ) infection of macaques . Here , we demonstrate that the process of adapting SHIVs for replication in macaques selects for viruses that are resistant to the IFNα response , and we identity important viral determinants that contribute to this resistance . This improved understanding of virus interactions with the innate immune response may facilitate the development of improved animal models of HIV-1 infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"blood",
"cells",
"cell",
"physiology",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"microbiology",
"vertebrates",
"viral",
"structure",
"animals",
"mammals",
"retroviruses",
"primates",
"immunodeficiency",
"viruses",
"viruses",
"rna",
"viruses",
"old",
"world",
"monkeys",
"white",
"blood",
"cells",
"monkeys",
"animal",
"cells",
"proteins",
"medical",
"microbiology",
"hiv",
"microbial",
"pathogens",
"cell",
"immortalization",
"hiv-1",
"viral",
"replication",
"virions",
"lymphocytes",
"siv",
"macaque",
"biochemistry",
"cell",
"biology",
"virology",
"viral",
"pathogens",
"interferons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"lentivirus",
"amniotes",
"organisms"
]
| 2016 | Adapting SHIVs In Vivo Selects for Envelope-Mediated Interferon-α Resistance |
Japanese encephalitis virus ( JEV ) , a mosquito-borne flavivirus that causes fatal neurological disease in humans , is one of the most important emerging pathogens of public health significance . JEV represents the JE serogroup , which also includes West Nile , Murray Valley encephalitis , and St . Louis encephalitis viruses . Within this serogroup , JEV is a vaccine-preventable pathogen , but the molecular basis of its neurovirulence remains unknown . Here , we constructed an infectious cDNA of the most widely used live-attenuated JE vaccine , SA14-14-2 , and rescued from the cDNA a molecularly cloned virus , SA14-14-2MCV , which displayed in vitro growth properties and in vivo attenuation phenotypes identical to those of its parent , SA14-14-2 . To elucidate the molecular mechanism of neurovirulence , we selected three independent , highly neurovirulent variants ( LD50 , <1 . 5 PFU ) from SA14-14-2MCV ( LD50 , >1 . 5×105 PFU ) by serial intracerebral passage in mice . Complete genome sequence comparison revealed a total of eight point mutations , with a common single G1708→A substitution replacing a Gly with Glu at position 244 of the viral E glycoprotein . Using our infectious SA14-14-2 cDNA technology , we showed that this single Gly-to-Glu change at E-244 is sufficient to confer lethal neurovirulence in mice , including rapid development of viral spread and tissue inflammation in the central nervous system . Comprehensive site-directed mutagenesis of E-244 , coupled with homology-based structure modeling , demonstrated a novel essential regulatory role in JEV neurovirulence for E-244 , within the ij hairpin of the E dimerization domain . In both mouse and human neuronal cells , we further showed that the E-244 mutation altered JEV infectivity in vitro , in direct correlation with the level of neurovirulence in vivo , but had no significant impact on viral RNA replication . Our results provide a crucial step toward developing novel therapeutic and preventive strategies against JEV and possibly other encephalitic flaviviruses .
Japanese encephalitis virus ( JEV ) is the most common cause of viral encephalitis in Asia and parts of the Western Pacific , with ∼60% of the world's population at risk of infection [1] . Within the family Flaviviridae ( genus Flavivirus ) , JEV belongs to the JE serological group , which also includes medically important human pathogens found on every continent except Antarctica [2] , [3]: West Nile virus ( WNV ) , St . Louis encephalitis virus ( SLEV ) , and Murray Valley encephalitis virus ( MVEV ) . Historically , the JE serological group members have clustered in geographically distinct locations , but the recent emergence and spread of JEV in Australia [4] and WNV in North America [5] , [6] have caused growing concern that these viruses can spread into new territory , posing a significant challenge for global public health [3] , [7] . In the US , where WNV and SLEV are endemic , the situation is particularly problematic because the likelihood of JEV being introduced is considerable [8] , [9] . Worldwide , ∼50 , 000–175 , 000 clinical cases of JE are estimated to occur annually [10]; however , this incidence is undoubtedly a considerable underestimate because surveillance and reporting are inadequate in most endemic areas , and only ∼0 . 1–4% of JEV-infected people develop clinical disease [11] , [12] . On average , ∼20–30% of patients die , and ∼30–50% of survivors suffer from irreversible neurological and/or psychiatric sequelae [13] . Most clinical cases occur in children under age 15 in endemic areas , but in newly invaded areas , all age groups are affected because protective immunity is absent [14] . Thus , given the current disease burden and significant threat of the JEV emergence , resurgence , and spread among much larger groups of susceptible populations , control of JEV remains a high public health priority . JEV contains a nucleocapsid composed of an ∼11-kb plus-strand genomic RNA , complexed with multiple copies of the highly-basic α-helical C proteins [15] , [16] . The nucleocapsid is surrounded by a host-derived lipid bilayer containing the membrane-anchored M and E proteins [17]–[19] . The initial step in the flavivirus replication cycle involves attachment of the virions to the surface of susceptible cells [20]–[24] . The viral E protein is then assumed to bind with high affinity and specificity to an as-yet unidentified cellular receptor ( s ) , which triggers receptor-mediated , clathrin-dependent endocytosis [25]–[27] . The acidic conditions in the endosome lead to a conformational change in the E protein [28]–[32] , which triggers fusion of the viral membrane with host endosomal membrane [33] . Once the genome is released into the cytoplasm , the genomic RNA is translated into a single polyprotein , which is processed co- and post-translationally by host and viral proteases to yield at least 10 functional proteins [34]: three structural ( C , prM , and E ) and seven nonstructural ( NS1 , NS2A , NS2B , NS3 , NS4A , NS4B , and NS5 ) . The nonstructural proteins actively replicate the viral genomic RNA in the replication complex [35]–[38] that is associated with the virus-induced , ER-derived membranes [39]–[41] . Newly synthesized genomic RNA and C proteins are initially enveloped by the prM and E proteins to generate immature virions [42] , [43] that bud into the lumen of the ER [44] . These immature virions are then transported via the secretory pathway to the Golgi apparatus . In the low-pH environment of the trans-Golgi network , the furin-mediated cleavage of prM to M induces the maturation of the viral particles [45] , which is also accompanied by significant structural rearrangements of the M and E proteins [42] , [46] , [47] . Finally , mature virions are released into the extracellular space by exocytosis . JEV is maintained in an enzootic cycle involving multiple species of mosquito vectors ( primarily Culex species ) and vertebrate hosts/reservoirs ( mainly domestic pigs/wading birds ) . Humans become infected incidentally when bitten by an infected mosquito [48] . In the absence of antiviral therapy , active immunization is the only strategy for sustainable long-term protection . Four types of JE vaccines are used in different parts of the world [49] , [50]: ( i ) the mouse brain-derived inactivated vaccine based on the Nakayama or Beijing-1 strain , ( ii ) the cell culture-derived inactivated vaccine based on the Beijing-3 or SA14-14-2 strain , ( iii ) the cell culture-derived live-attenuated vaccine based on the SA14-14-2 strain , and ( iv ) the live chimeric vaccine developed on a yellow fever virus ( YFV ) 17D genetic background that carries two surface proteins of JEV SA14-14-2 . Of the four vaccines , the only one that is available internationally is the mouse brain-derived inactivated Nakayama [11] . Unfortunately , the production of this vaccine was discontinued in 2006 [51] because of vaccine-related adverse events , short-term immunity , and high production cost [13] , [52] . To date , the most commonly used vaccine in Asia is the live-attenuated SA14-14-2 [53] , but this vaccine is not recommended by the WHO for global immunization [13] , [54] . In addition to the dependence of the duration of immunity on the number of doses received , there is at least a theoretical risk of virus mutation and reversion of the vaccine virus to high virulence . Recently , the SA14-14-2 vaccine virus has been utilized to produce a new Vero cell-derived inactivated vaccine that has been approved in the US , Europe , Canada , and Australia since 2009 [51] , [55] , [56] . In the US , this vaccine is recommended for adults aged ≥17 years travelling to JEV-endemic countries and at risk of JEV exposure [51] , [57] , but no vaccine is currently available for children under 17 [58] . More recently , the prM and E genes of JEV SA14-14-2 have been used to replace the corresponding genes of YFV 17D [59] , creating a live chimeric vaccine [60] that is now licensed in Australia and Thailand [61] , [62] . Thus , the application of JEV SA14-14-2 to vaccine development and production is continuously expanding , but the viral factors and fundamental mechanisms responsible for its loss of virulence are still elusive . The virulence of JEV is defined by two properties: ( i ) neuroinvasiveness , the ability of the virus to enter the central nervous system ( CNS ) when inoculated by a peripheral route; and ( ii ) neurovirulence , the ability of the virus to replicate and cause damage within the CNS when inoculated directly into the brain of a host . Over the past 20 years , many investigators have sought to understand the molecular basis of JEV virulence , by using cell and animal infection model systems to compare the nucleotide sequences of the genomes of several JEV strains that differ in virological properties [63]–[74] . These studies have identified a large number of mutations scattered essentially throughout the entire viral genome . Because of the complexity of the mutations , however , the major genetic determinant ( s ) critical for either JEV neurovirulence or neuroinvasiveness remains unclear . In particular , the situation is more complicated for the live-attenuated SA14-14-2 virus , which has been reported to have a number of different mutations , i . e . , 47–64 nucleotide changes ( 17–27 amino acid substitutions ) , when compared to its virulent parental strain SA14; the exact number depends on both the passage history of the viruses and the type of cell substrate used for virus cultivation [63]–[65] . A more comprehensive sequence comparison with another SA14-derived attenuated vaccine strain , SA14-2-8 , together with two other virulent strains , has suggested seven common amino acid substitutions that may be involved in virus attenuation: 4 in E , 1 in NS2B , 1 in NS3 , and 1 in NS4B [64] . However , the genetic component directly responsible for the attenuation of SA14-14-2 is still unknown . Given that SA14-14-2 has been administered to >300 million children for >20 years in China and recently in other Asian countries [53] , it is striking that there is a fundamental gap in our knowledge at the molecular level about how SA14-14-2 is attenuated . Here we report the development of an infectious cDNA-based reverse genetics system for JEV SA14-14-2 that has enabled the analysis of molecular aspects of its attenuation in neurovirulence . By in vivo passage of a molecularly defined , cDNA-derived SA14-14-2 virus , we generated three isogenic variants , each displaying lethal neurovirulence in mice , with a common single G1708→A substitution that corresponds to a Gly→Glu change at position 244 of the viral E glycoprotein . By in vitro site-directed mutagenesis of the infectious SA14-14-2 cDNA , coupled with conventional virologic and experimental pathologic methods and homology-based structure modeling , we have demonstrated a novel regulatory role in JEV neurovirulence of a conserved single amino acid at position E-244 in the ij hairpin adjacent to the fusion loop of the E dimerization domain . These findings offer new insights into the molecular mechanism of JEV neurovirulence and will directly aid the development of new approaches to treating and preventing JEV infection .
As an initial step in investigating the molecular basis for the virulence attenuation of SA14-14-2 , we generated a full-length infectious SA14-14-2 cDNA to serve as a template for genetic manipulation of the viral genome ( Fig . 1A ) . The 10 , 977-nucleotide genome of SA14-14-2 ( GenBank accession number JN604986 ) was first cloned as four contiguous cDNAs into the bacterial artificial chromosome ( BAC ) designated pBAC/Frag-I to IV ( Fig . 1B ) . pBAC/Frag-I was modified to have an SP6 promoter immediately upstream of the viral 5′-end , and pBAC/Frag-IV was engineered to contain an artificial XbaI run-off site just downstream of the viral 3′-end , allowing in vitro run-off transcription of capped , genome-length RNAs bearing authentic 5′ and 3′ ends of the genomic RNA . Since the viral genome already had an internal XbaI site at nucleotide 9131 in the NS5 protein-coding region , this pre-existing site was eliminated in pBAC/Frag-III by introducing a silent point mutation , A9134→T ( Fig . 1B , asterisk ) , which in turn served as a genetic marker to identify the cDNA-derived virus . In the last cloning step , a panel of the four overlapping SA14-14-2 cDNAs was sequentially assembled by joining at three natural restriction sites ( BsrGI , BamHI , and AvaI ) to create the full-length SA14-14-2 cDNA , pBAC/SA14-14-2 ( Fig . 1C ) . The functionality of pBAC/SA14-14-2 was analyzed by determining the specific infectivity of the synthetic RNAs transcribed in vitro from the cDNA after RNA transfection into susceptible BHK-21 cells ( Fig . 1D ) . Two independent clones of pBAC/SA14-14-2 were linearized by XbaI , followed by mung bean nuclease treatment to remove the 5′ overhang left by the XbaI digestion . Each was then used as a template for SP6 polymerase run-off transcription in the presence of the m7G ( 5′ ) ppp ( 5′ ) A cap structure analog . Transfection of the synthetic RNAs into BHK-21 cells gave specific infectivities of 6 . 0–7 . 5×105 PFU/µg; the virus titers recovered from the RNA-transfected cells were 3 . 0–4 . 5×105 PFU/ml at 22 h post-transfection ( hpt ) and increased ∼10-fold to 2 . 9–3 . 7×106 PFU/ml at 40 hpt ( Fig . 1D ) . Unequivocally , the recovered virus contained the marker mutation ( A9134→T ) that had been introduced in pBAC/SA14-14-2 ( data not shown ) . Our results show that the synthetic RNAs generated from the full-length SA14-14-2 cDNA are highly infectious in BHK-21 cells , producing a high titer of molecularly defined , infectious virus . In cell cultures [75] , [76] , we assessed the in vitro growth properties of the molecularly cloned virus ( SA14-14-2MCV ) rescued from the infectious cDNA , as compared to those of the uncloned parental virus ( SA14-14-2 ) used for cDNA construction . In hamster kidney BHK-21 cells , which are used most frequently for JEV propagation in laboratories , SA14-14-2MCV replicated as efficiently as SA14-14-2 , with no noticeable difference in the accumulation of viral genomic RNA ( Fig . 2A ) and proteins ( Fig . 2B ) over the first 24 h after infection at a multiplicity of infection ( MOI ) of 1 plaque-forming unit ( PFU ) per cell . These observations were consistent with their growth kinetics , which were essentially identical for 4 days following infection at three different MOIs: 0 . 1 , 1 , and 10 PFU/cell ( Fig . 2C and data not shown ) . Similarly , there was no difference in focus/plaque morphology between SA14-14-2MCV and SA14-14-2 at 4 days post-infection ( dpi ) ( Fig . 2D ) ; as expected , their foci/plaques were ∼30% smaller than those produced by CNU/LP2 , a virulent JEV strain used as a reference ( Fig . S1 ) . Also , their growth properties were equivalent in two other cell lines , human neuroblastoma SH-SY5Y and mosquito C6/36 cells , which are potentially relevant to JEV pathogenesis and transmission , respectively ( Fig . S2 ) . These data suggest that the uncloned parental and molecularly cloned viruses are indistinguishable in viral replication and spread in both mammalian and insect cells . In mice [75] , [77] , we evaluated in vivo the attenuation phenotypes of SA14-14-2MCV and SA14-14-2 , with a virulent JEV CNU/LP2 [78] in parallel . Groups of 3-week-old ICR mice ( n = 20 ) were infected with various doses ( 1 . 5 to 1 . 5×105 PFU/mouse ) of each virus , via three different inoculation routes: intracerebral ( IC ) for neurovirulence , and intramuscular ( IM ) and intraperitoneal ( IP ) for neuroinvasiveness . As with SA14-14-2 , the 50% lethal doses ( LD50s ) of SA14-14-2MCV , regardless of the route of inoculation , were all >1 . 5×105 PFU ( Figs . 2E and S3 ) . Specifically , all mice infected with SA14-14-2MCV or SA14-14-2 remained healthy and displayed no clinical signs of JEV infection ( e . g . , ruffled fur , hunched posture , tremors , or hindlimb paralysis ) after IM or IP inoculation with any of the tested doses; on the other hand , a small fraction of the mice infected with SA14-14-2MCV ( 5–20% ) or SA14-14-2 ( 5–10% ) developed typical symptoms and death after the IC inoculation with a relatively high dose of ≥1 . 5×103 PFU/mouse , but not the low dose of ≤1 . 5×102 PFU/mouse ( Fig . S3 ) . In all dead or surviving mice , virus titration confirmed the presence ( 1 . 8–4 . 1×106 PFU/brain ) or absence , respectively , of viral replication in their brain tissues . As expected [75] , [77] , the LD50 values of CNU/LP2 [78] , irrespective of the inoculation route , were always <1 . 5 PFU ( Figs . 2E and S3 ) ; the control groups of mock-infected mice all survived with no signs of disease ( Fig . S3 ) . Thus , our data indicate that SA14-14-2MCV displays a variety of biological properties identical to those of SA14-14-2 , both in vitro and in vivo . As was true for SA14-14-2 , direct inoculation of a relatively high dose of SA14-14-2MCV into mouse brains initiated a productive infection in the CNS and caused lethal encephalitis , albeit at a very low frequency ( Fig . S3 ) . Intrigued by this observation , we decided to generate isogenic neurovirulent variants from SA14-14-2MCV by serial brain-to-brain passage in mice ( Fig . 3A ) . At passage 1 ( P1 ) , the cDNA-derived SA14-14-2MCV was directly inoculated into the brains of 3-week-old ICR mice at 1 . 5×105 PFU/mouse ( three groups , n = 10 per group ) ; one or two infected mice per group exhibited clinical symptoms of JEV infection . At the onset of hindlimb paralysis ( 6–10 dpi ) , virus was harvested from the brain of a moribund mouse in each group ( 3 total ) ; in each case , a brain homogenate was prepared for plaque titration and used as an inoculum for the next round of passage . Serial intracerebral passage was continued for three additional rounds , with a gradually decreasing inoculum in order to ensure the stability of selected mutations and a sufficiently pure population of viruses: 1 , 500 ( P2 ) , to 15 ( P3 ) , to 1 . 5 PFU/mouse ( P4 ) . Using this approach , we obtained three independently selected variants , SA14-14-2MCV/V1 to V3 ( Fig . 3A ) . We first compared the biological properties of the three SA14-14-2MCV variants , both in vitro and in vivo , to those of the parental SA14-14-2MCV . In three cell cultures ( BHK-21 , SH-SY5Y , and C6/36 ) , all three variants exhibited characteristics of viral replication identical to the parent , as demonstrated by ( i ) quantitative real-time RT-PCRs to measure the level of viral genomic RNA production , ( ii ) immunoblotting with a panel of JEV-specific rabbit polyclonal antisera to probe the profile and level of viral structural and nonstructural protein accumulation , and ( iii ) one-step growth analyses to assess the yield of progeny virions produced during a single round of infection ( data not shown ) . In 3-week-old ICR mice , however , there was a clear difference between the parent and the three variants in both phenotype and virulence level ( Fig . 3B ) . When peripherally inoculated ( i . e . , IM and IP ) , neither the parent nor its three variants caused any symptoms or death at a maximum dose of 1 . 5×105 PFU/mouse . In contrast , when inoculated IC , the three variants , unlike the parent ( IC LD50 , >1 . 5×105 PFU ) , were all highly neurovirulent ( IC LD50s , <1 . 5 PFU ) ( Fig . 3B and Table S1 ) . Our findings show that all three variants still lacked a detectable level of neuroinvasiveness but gained a high level of neurovirulence after serial IC passage in mice . Next , we determined the complete nucleotide sequence of the genome of the three SA14-14-2MCV variants to identify the nucleotide ( s ) and/or amino acid ( s ) in specific viral loci/genes that is ( are ) potentially responsible for the drastic increase in neurovirulence . According to our protocol [77] , the consensus genome sequence of each variant was generated by direct sequencing of three overlapping , uncloned cDNA amplicons covering the entire viral RNA genome except the 5′- and 3′-termini; the remaining consensus sequences of the 5′- and 3′-terminal regions were obtained by 5′- and 3′-RACE reactions , each followed by cDNA cloning and sequencing of 10–15 independent clones . In all three variants , when the consensus genome sequence was compared to that of the parent , a single nucleotide G-to-A transition was always found at nucleotide 1708 , changing a Gly ( GGG ) to Glu ( GAG ) codon at amino acid 244 of the viral E glycoprotein ( Fig . 3C ) . In addition , each of the three variants also contained a small number of unique silent point mutations scattered over the genome , confirming they were indeed independent variants ( Fig . 3C ) : one in SA14-14-2MCV/V1 ( U2580C ) , two in SA14-14-2MCV/V2 ( G317A and U8588C ) , and four in SA14-14-2MCV/V3 ( U419C , C3215U , C5987U , and G6551A ) . These results suggest that the G1708A substitution , the only mutation observed in all three variants , may contribute to the viral neurovirulence in mice . To identify a key point mutation ( s ) in three variants of SA14-14-2MCV that leads to the acquisition of neurovirulence , we generated eight derivatives of SA14-14-2MCV , each containing one of the eight point mutations found in our three variants , by cloning them individually into the infectious SA14-14-2 cDNA and transfecting the synthetic RNAs derived from each mutant cDNA into BHK-21 cells . In all cases , the mutant RNA was as infectious as the parent RNA , with a specific infectivity of 6 . 4–8 . 3×105 PFU/µg; the sizes of the foci/plaques produced by each mutant RNA were indistinguishable from those generated by the parent RNA , paralleling their levels of virus production , with an average yield of 2 . 1–4 . 5×105 PFU/ml at 22 hpt ( Fig . 4A ) . In agreement with these results , no difference was observed in the profile or expression level of the viral proteins , i . e . , three structural ( C , prM , and E ) and one nonstructural ( NS1 ) , as determined by immunoblotting of RNA-transfected cells at 18 hpt ( Fig . 4B ) . All the mutant viruses grew as efficiently as did the parental virus over the course of 96 h after infection at an MOI of 0 . 1 in BHK-21 cells ( Fig . 4C ) . Thus , there was no apparent effect of any of the eight introduced genetic changes on virus replication . In mice , we examined the neurovirulence of these eight mutant viruses . Groups of 3-week-old ICR mice ( n = 10 per group ) were infected by IC inoculation with various doses ( 1 . 5 to 1 . 5×105 PFU/mouse ) of the parent or each mutant virus . One of the eight mutants containing the G1708A substitution had an IC LD50 of <1 . 5 PFU , making it capable of killing all mice within ∼7 dpi with a minimum dose of 1 . 5 PFU/mouse; the other seven mutants had IC LD50 values all >1 . 5×105 PFU and behaved like the parental virus , with only <20% of infected mice developing clinical symptoms and death at a maximum dose of 1 . 5×105 PFU/mouse ( Fig . 4D and Table S2 ) . In all dead or surviving mice , virus titration confirmed the presence ( 1 . 4–3 . 5×106 PFU/brain ) or absence , respectively , of productive viral replication in the brain tissues; as expected , all mock-infected mice survived with no signs of disease ( data not shown ) . Thus , our findings showed that of the eight point mutations , a single G1708A substitution , replacing a Gly with Glu at amino acid residue 244 of the viral E glycoprotein , is sufficient to confer lethal neurovirulence in mice . To determine whether the mutant G1708A , unlike the parent SA14-14-2MCV , is able to replicate and spread in the CNS , we immunohistochemically stained for JEV NS1 antigen in mouse brains after IC inoculation ( Fig . 4E shows hippocampal slides , and Fig . S4 presents slides of other brain areas , i . e . , amygdala , cerebral cortex , thalamus , hypothalamus , and brainstem ) : ( i ) In brains infected with a virulent JEV CNU/LP2 ( control ) [75] , [77] , [78] , a large number of NS1-positive neurons were observed at 3 dpi in all areas we stained; this number was increased significantly at 5 dpi . In the hippocampus , most infected neurons were found in the CA2/3 region at 3 dpi and had spread to the CA1 region by 5 dpi . ( ii ) In brains infected with the parent SA14-14-2MCV , almost no NS1-positive cells were found in any brain region during the entire 7-day course of the experiment . In a few atypical cases , a small number of NS1-positive neurons were noted at 5–7 dpi in the hippocampal CA2/3 region , but not the CA1 region ( data not shown ) . ( iii ) In brains infected with the mutant G1708A , a considerable number of NS1-positive neurons were observed at 3 dpi , mainly in the hippocampal CA2/3 region , and only a few in other areas ( amygdala , cerebral cortex , thalamus , and brainstem ) ; overall , the number of infected neurons was much lower than in brains infected with JEV CNU/LP2 . At 5–7 dpi , the number of NS1-positive neurons was noticeably increased in the hippocampus ( now in the CA1 ) and amygdala , but not in other brain regions . Our findings show that , in mice , a single G1708A substitution changing a Gly with Glu at position E-244 promotes susceptibility to SA14-14-2MCV infection of neurons . To probe the functional importance of the amino acid side chain at position E-244 for the viral replication and neurovirulence of SA14-14-2MCV , we performed site-directed mutagenesis , replacing G244 with 14 other amino acids of six different classes: ( 1 ) aliphatic A , V , and L; ( 2 ) hydroxyl S and T; ( 3 ) cyclic P; ( 4 ) aromatic F and W; ( 5 ) basic R and K; and ( 6 ) acidic and their amides D , E , N , and Q . We first tested the viability of synthetic RNAs transcribed in vitro from the corresponding mutant cDNAs by measuring their infectivity after transfection of BHK-21 cells . In all cases , the mutant RNA was as viable as the parent RNA , with a specific infectivity of 6 . 5–8 . 2×105 PFU/µg ( Fig . 5A , RNA infectivity ) . However , three mutants ( G244K , G244F , and G244W ) were noticeably different from the parent and the other 11 mutants , as demonstrated by a ∼10-fold decrease in the yield of progeny virions released into culture medium during the first 22 hpt ( Fig . 5A , virus yield ) and a ∼2-2 . 5-fold reduction in the size of foci/plaques produced at 96 hpt ( Fig . 5A , foci/plaques ) , although no significant difference was observed in the level of viral proteins ( i . e . , C , prM , E , and NS1 ) accumulated in RNA-transfected cells at 18 hpt ( Fig . S5 ) . As compared to G244K , the mutant G244R exhibited a barely marginal decrease in focus/plaque size and no detectable change in virus production ( Fig . 5A ) . Overall , these findings were more evident when all mutant viruses were evaluated in multistep growth assays over the course of 96 h after infection at an MOI of 0 . 1 , assessing their ability to grow and establish a productive infection ( Fig . 5B ) . Our findings indicate that in BHK-21 cells , the amino acid side chain at position E-244 has no effect on the viability of the mutant RNAs , although it has a negative impact on the production and spread of infectious virions in the case of the three mutants G244K , G244F , and G244W . In mice , we determined the neurovirulence of our 14 mutant viruses by IC inoculating groups of 3-week-old ICR mice ( n = 10 per group ) with various doses ranging from 1 . 5 to 1 . 5×104 or 105 PFU/mouse of the parent or each mutant virus . According to their IC LD50 values , the 14 mutant viruses are classified into three groups ( Fig . 5C and Table S3 ) : ( i ) group 1 ( six mutants ) , neurovirulent , with an IC LD50 of ≤1 . 5 to 31 PFU , exemplified by replacing G244 with E , D , T , S , Q , and P; ( ii ) group 2 ( six mutants ) , non-neurovirulent or neuroattenuated , with an IC LD50 of >1 . 5×104 or 105 PFU , behaving like the parent SA14-14-2MCV and exemplified by exchanging G244 with R , K , F , W , N , and L; and ( iii ) group 3 ( two mutants ) , with an intermediate phenotype and an IC LD50 of 1 . 2–5 . 8×103 PFU , exemplified by substituting G244 with A and V . We confirmed the presence or absence of viral replication in the brain tissues of all dead or surviving mice , respectively; all mock-infected mice survived with no signs of disease ( data not shown ) . Also , the mutation and phenotype relationship was corroborated by sequence analysis of recovered viruses from brain tissues of moribund or dead mice following IC inoculation . We analyzed all of the 14 mutants except for four group 2 mutants ( G244R , G244F , G244W , and G244L ) , which failed to produce a lethal infection . In each case , the complete 2 , 001-nucleotide coding region of the prM and E genes was amplified from each of four randomly selected brain samples , followed by cloning and sequencing of at least seven independent clones per brain sample . In all six group 1 and two group 3 mutants , we found that the initial mutations introduced at the G244 codon were maintained with no second-site mutations , consistent with the high and intermediate levels of their neurovirulent phenotype ( Table 1 ) . In the remaining two group 2 mutants ( G244K and G244N ) , however , a majority of the sequenced clones contained a point mutation in the same codon that led to an amino acid substitution ( i . e . , K→E/T and N→D , respectively ) , converting both mutants into neurovirulent viruses and highlighting the biological importance of the amino acid at position E-244 for neurovirulence ( Table 1 ) . We next performed homology modeling to gain insight into the structural basis of E-244 function . The 3D model of the E monomer of JEV SA14-14-2 was constructed using the 3 . 0-Å crystal structure of the E monomer of WNV NY99 [79] as a template , with 75 . 5% sequence identity . The model was then fitted into the outer layer of the cryo-electron microscopy ( EM ) structure of WNV NY99 [18] , thereby visualizing three monomers placed into an icosahedral asymmetric unit on the viral membrane . In each E monomer of SA14-14-2 containing three domains ( DI , DII , and DIII ) , we noted that E-244 lies within the ij hairpin adjacent to the fusion loop at the tip of DII , with its amino acid side chain pointing toward the viral membrane ( Fig . 5D ) . We also confirmed the location of E-244 in the crystal structure of the E ectodomain of JEV SA14-14-2 [80] that has been described recently ( Fig . S6 ) . We hypothesized that E-244 , located at the ij hairpin of the viral E glycoprotein , plays an important role in JEV infection of neuronal cells . To test this hypothesis , we performed multistep growth assays in two neuronal cells , NSC-34 ( mouse motor neuron ) and SH-SY5Y ( human neuroblastoma ) , by infecting at an MOI of 0 . 1 with the non-neurovirulent parent SA14-14-2MCV and each of the four representative E-244 mutant viruses , i . e . , two neurovirulent ( G244E and G244D ) and two non-neurovirulent ( G244R and G244K ) . In parallel , the non-neuronal BHK-21 cells were also infected for comparison with the same set of five viruses . In NSC-34 cells , while the two neurovirulent viruses grew rapidly and reached their maximum titers of 1 . 8–2 . 4×105 PFU/ml at 72–96 hours post-infection ( hpi ) , the three non-neurovirulent viruses , including the parent , all replicated poorly , with peak titers only approaching 1 . 0–2 . 0×103 PFU/ml , ∼100-fold lower than those of the two neurovirulent viruses ( Fig . 6A ) . In SH-SY5Y cells , a similar defect in viral growth was also observed , with a ∼50- to 100-fold difference in maximum virus titers between the neurovirulent and the non-neurovirulent viruses . In addition , we noted a differential growth defect in the three non-neurovirulent viruses , with G244R replicating more poorly than the parent but better than G244K ( Fig . 6B ) . In contrast to the pattern of viral growth observed in NSC-34 and SH-SY5Y cells , we found that in BHK-21 cells , only G244K had a noticeable defect in viral growth , with ∼20-fold lower peak titers than those of the other four viruses that grew well to maximum titers of 0 . 8–2 . 5×106 PFU/ml at 48–72 hpi ( Fig . 6C ) . These data indicate that E-244 plays a crucial role in the productive infection of JEV in neuronal cells . Subsequently , we examined the infectivity/replicability of the parent and its four E-244 mutant viruses/RNAs in NSC-34 and SH-SY5Y cells , in parallel with BHK-21 cells for comparison . First , virus infectivity was quantified using flow cytometry by infecting the three cell types at an MOI of 1 with each of the five viruses and counting the number of cells stained with a mouse α-JEV antiserum at 12–15 hpi . In NSC-34 cells , the two non-neurovirulent mutants ( G244R and G244K ) exhibited infectivities nearly identical to that of the non-neurovirulent parent ( Fig . 6D ) , whereas the two neurovirulent mutants ( G244E and G244D ) showed infectivities ∼16- to 20-fold higher than that of the non-neurovirulent parent . Similarly , the E-244 mutation also altered virus infectivity in SH-SY5Y cells . Specifically , the two neurovirulent mutants had ∼3- to 4-fold higher infectivities than the non-neurovirulent parent; on the other hand , the two non-neurovirulent mutants displayed even lower infectivities than the parent ( ∼3-fold for G244R and ∼10-fold for G244K ) ( Fig . 6E ) . In contrast , no significant difference in virus infectivity was observed among all five viruses in BHK-21 cells ( Fig . 6F ) . Next , the replication efficiency of the viral genomic RNA was quantified by directly transfecting the three cell types with each of the five synthetic RNAs transcribed in vitro from the respective JEV cDNAs and estimating the number of infectious foci stained with the mouse α-JEV antiserum at 4 days post-transfection ( dpt ) . In each of the three cell types , there was no detectable difference in the specific infectivity of the five RNAs ( NSC-34 , Fig . 6G; SH-SY5Y , Fig . 6H; and BHK-21 , Fig . 6I ) . Also , quantitative real-time RT-PCRs indicated that the level of the viral genomic RNAs accumulated in the RNA-transfected cells over the first 15 h of transfection was indistinguishable between the parent and the four different E-244 mutants ( data not shown ) . Regardless of cell type , however , the G244K mutant was different from the parent and the other three E-244 mutants , as demonstrated by a ∼1-log decrease in the yield of infectious virions released into culture medium during the first 20 hpt ( NSC-34 , Fig . 6J; SH-SY5Y , Fig . 6K; and BHK-21 , Fig . 6L ) . Overall , these results show that the E-244 mutation alters JEV infectivity in a neuronal cell-specific manner , in agreement with the neurovirulence phenotype observed in mice , and it also affects infectious particle production in a cell type-nonspecific manner . We initially generated a multiple sequence alignment using all 154 full-length JEV genomes available from the GenBank sequence database . Of note is the fact that SA14 and SA14-14-2 have been fully sequenced by three and four independent groups , respectively; their nucleotide and deduced amino acid sequences are not identical [63]–[65] , [77] , [81] . The sequence alignment showed a Glu residue at position E-244 in the ij hairpin of all JEV strains isolated from infected mosquitoes , pigs , or humans , except for the Gln-encoding mosquito-derived K94P05 and three Gly-encoding SA14-derived attenuated strains ( i . e . , SA14-2-8 , SA14-12-1-7 , and all four different versions of SA14-14-2 ) ( Fig . S7 ) . In case of SA14 , it is intriguing to note that one version has a Glu residue at position E-244 , but the other two versions have a Gly residue at that position ( Fig . S7 ) ; this discrepancy is likely due to variations in the cultivation history of the virus [77] . We next performed the structure-based , ij-hairpin amino acid sequence alignment with six representative flaviviruses ( 14 strains total ) , including four encephalitic ( JEV , WNV , SLEV , and MVEV ) and two non-encephalitic ( YFV and DENV ) flaviviruses . In addition to the importance of the E-244 amino acid , we noted ( i ) the evolutionally conserved residues in the ij hairpin and its flanking region in all six flaviviruses , i . e . , W233 , F242 , H246 , A247 , V252 , L255 , G256 , Q258 , E259 , and G260; ( ii ) the sequence similarities in the four encephalitic flaviviruses , particularly in a ∼15-aa ij-hairpin-containing region; and ( iii ) the sequence differences between the four encephalitic and two non-encephalitic flaviviruses , e . g . , the 4-aa YFV-specific motif and the 3-aa DENV-specific motif ( Fig . 7 ) . Overall , our findings suggest that the ij hairpin of the E DII plays a key role in determining encephalitic flavivirus neurovirulence , and its function is regulated by the chemical properties of the amino acid at position E-244 in that hairpin .
In this work , we have developed a reverse genetics system for SA14-14-2 , a live-attenuated JE vaccine [53] , [82] , by constructing an infectious cDNA and rescuing molecularly cloned virus from the cDNA . This reverse genetics system offers us a unique opportunity to elucidate the genetic and molecular basis of JEV neurovirulence . Using our infectious SA14-14-2 cDNA technology , we ( i ) generated three isogenic SA14-14-2 variants that unlike its parent , displayed lethal neurovirulence in mice; ( ii ) identified a single point mutation , G1708→A , causing a Gly→Glu change at position 244 of the viral E glycoprotein that is sufficient to confer a full neurovirulence by promoting viral infection into neurons in the mouse CNS in vivo and mouse/human neuronal cells in vitro; and ( iii ) demonstrated the structure-function relationship for neurovirulence of E-244 in the ij hairpin adjacent to the fusion loop at the tip of the viral E DII . Thus , our findings reveal fundamental insights into the neurotropism and neurovirulence of JEV and other taxonomically related encephalitic flaviviruses , including WNV , SLEV , and MVEV . Intriguingly , our results also provide a new target , the ij hairpin , for the development of novel antivirals for the prevention and treatment of infection with the encephalitic flaviviruses . The flavivirus glycoprotein E mediates receptor-mediated endocytosis and low pH-triggered membrane fusion [33] , [83] , [84] . On the viral membrane , 180 E monomers are packed into 30 protein “rafts” , each composed of three E head-to-tail homodimers [17]–[19] . Each E monomer is composed of three parts: ( i ) an elongated ectodomain that directs receptor binding and membrane fusion; ( ii ) a “stem” region containing two amphipathic α-helices that lies flat on the viral membrane underneath the ectodomain; and ( iii ) a membrane “anchor” region containing two transmembrane antiparallel coiled-coils . The E ectodomain folds into three β-barrel domains [85]: ( i ) DI , a structural domain centrally located in the molecule; ( ii ) DII , an elongated dimerization domain containing the highly conserved fusion loop at its tip [86]; and ( iii ) DIII , an Ig-like domain implicated in receptor binding [20] , [87] , [88] and antibody neutralization [89]–[92] . Based on pre- and post-fusion crystal structures of the ectodomain [28] , [30] , [31] , [47] , [79] , [80] , [85] and biochemical analyses [29] , [32] , [93] , a current , detailed model for flavivirus membrane fusion has been developed . In this model , the fusion is initiated by a low pH-induced dissociation of the antiparallel E homodimers that leads to the exposure of the fusion loops and their insertion into the host membrane , followed by a large-scale structural rearrangement into a parallel E homotrimer [33] , [83] , [84] , [94] . In the parallel conformation , DIII folds back toward DII , presumably with the stem extended from the C-terminus of DIII along DII and toward the fusion loop ( “zipping” ) , driving the fusion of the viral and host membranes [95]–[98] . Despite our detailed knowledge about the fusion process , there is little available structural information about how flaviviruses bind to their cellular receptors . In encephalitic flaviviruses , the presence of an RGD motif in DIII and carbohydrate moieties on the viral surface suggests a mechanism involving interaction with the RGD motif-recognizing integrins and sugar-binding lectins on the cell surface , respectively . However , blocking/alteration of either the RGD motif or glycan does not abolish viral entry [22] , [99]–[101] . Thus , the viral factors and the interacting cellular counterparts required for viral entry are still elusive . In flaviviruses , the ij hairpin is a structural motif that is closely associated with the fusion loop at the tip of the viral E DII , but its role is thus far unknown . In JEV , we now report that a single amino acid in the ij hairpin , E-244 , serves as a key regulator to control the level of neurovirulence of SA14-14-2 in mice . This amino acid was also correlated with a differential ability to infect neurons , the primary target cells in the CNS . Consistent with this finding , we found that site-directed mutagenesis of the codon for E-244 in SA14-14-2 created a panel of 14 recombinant viruses of varying neurovirulence: ( i ) non-neurovirulent viruses , produced by substitutions of positively-charged ( R , K ) , aromatic ( F , W ) , polar ( N ) , or aliphatic ( L ) residues; ( ii ) neurovirulent viruses , produced by substitutions of negatively-charged ( E , D ) , hydroxyl ( T , S ) , polar ( Q ) , or cyclic ( P ) residues; and ( iii ) viruses intermediate in neurovirulence , produced by substitutions of aliphatic ( A , V ) residues . These results highlight the role of E-244 in neurovirulence , which was directed by a combination of three major properties of its amino acid side chain: ( i ) charge ( R/K vs . E/D ) ; ( ii ) size ( N vs . Q and L vs . A/V ) ; and ( iii ) functional group ( N vs . D ) . Our data suggest that the ij hairpin acts as a viral factor that promotes JEV infection of neurons within the CNS , likely through its role in one of three major steps involved in viral entry: binding , endocytosis , or membrane fusion [33] , [83] , [84] . Alternatively , it is possible that the late steps in the virus life cycle in neurons , such as assembly , maturation , and release , could be affected . For JEV , WNV , and tick-borne encephalitis virus , the assembly/release of infectious virions or virus-like particles has been shown to be affected by the N-glycosylation of the viral prM and/or E protein in non-neuronal cells [44] , [75] , [102]–[104] . Moreover , a conserved single N-glycosylation site in the JEV prM protein has been shown to be important for viral pathogenicity in mice [75] . Over the years , the virulence of JEV has been an active area of research . Initially , comparison of the genomic sequences of several JEV strains with a different degree of pathogenicity had predicted a number of potential loci in the viral genome that are involved in virulence [63]–[74] . Due to the complexity and variation of the mutations , however , the identity of the major viral factor that is critical for JEV virulence remains unclear . In particular , SA14-14-2 has been reported to have a total of 47–64 nucleotide changes ( 17–27 amino acid substitutions ) when compared to its virulent parental strain SA14; the number of mutations varies and depends on the cultivation history of the viruses [63]–[65] . Of the ten viral proteins , the E protein has been the primary target of genetic studies in virulence , mainly because it is involved in cell/tissue tropism and pathogenesis . Several amino acid residues in the E protein have been suggested to contribute to the neurovirulence and/or neuroinvasiveness of JEV in vivo: ( i ) E-123 , illustrated by an S123R substitution that is capable of enhancing the neuroinvasiveness of the Mie/41/2002 strain in 3-week-old ddY mice [105]; ( ii ) E-279 , exemplified by an M279K mutation that is able to increase the neurovirulence of ChimeriVax-JE ( a chimeric virus that carries the prM and E genes of JEV SA14-14-2 on a YFV 17D genetic background ) in suckling mice and rhesus monkeys [106]; and ( iii ) E-138 , indicated by two reciprocal mutations: ( 1 ) a K138E substitution , when combined with at least two other mutations , which elevates the neurovirulence of the ChimeriVax-JE virus in 4-week-old ICR mice [107] , and ( 2 ) an E138K substitution , which lowers the neurovirulence and neuroinvasiveness of three different JEVs ( the JaOArS982 strain in 2- to 5-week-old Swiss ICR mice [73] and the AT31 and NT109 strains in 3-week-old BALB/c mice [108] , [109] ) . These data indicate that multiple amino acid residues in the E protein of JEV function in a more coordinated way to achieve the maximal level of neurovirulence and/or neuroinvasiveness [53] , [107] . This notion is consistent with our finding that although a single G244E mutation in the E protein of SA14-14-2 is sufficient to confer lethal neurovirulence in 3-week-old ICR mice , the spread of the virus in the brains is still slow and limited , as compared to the highly virulent CNU/LP2 strain . These and previous findings suggest that in addition to E-244 , other amino acid residues in the E protein play a role in determining the neurovirulence of SA14-14-2 . In addition , JEV NS1' ( a product of ribosomal frameshifting [110] , [111] ) is reported to be produced in cells infected with SA14 but not with SA14-14-2 , and its lack of expression is shown to contribute to the attenuation phenotype of SA14-14-2 in mice [112] . Similarly , the expression of NS1' is also suggested to be associated with the neuroinvasiveness of WNV [110] . The neuroattenuation phenotype of SA14-14-2 has been tested in several laboratory animals , including mice and monkeys [53] . In 2- to 4-week-old ICR and ddY mice , no morbidity or mortality has been observed after subcutaneous and intracerebral inoculations with 104-106 PFU of SA14-14-2 [53] , [63] , [113]; in a rare case , however , the virus was found to be able to cause the death of the mice following IC inoculation [113] . In line with these previous results , we also found that none of the 3-week-old ICR mice inoculated IM or IP with up to ∼105 PFU of SA14-14-2 showed clinical signs or death; on the other hand , although there was some variability among the groups of mice and the doses of virus inoculum , ∼5–30% of the mice inoculated IC with a dose of 103–105 PFU developed JEV-specific symptoms and death . This low but unexpected morbidity and mortality after the IC inoculation of SA14-14-2 is likely caused by a combination of factors and conditions imposed on our infection experiments , particularly the age and strain of mice: ( i ) Age-dependent susceptibility of flaviviruses , including JEV , in the murine model has been documented previously , although its molecular mechanisms remain unclear [106] , [114] , [115] . ( ii ) A noticeable variability in mortality has also been reported when two different lineages of the age-matched outbred ICR mice are inoculated IC with a mutant of ChimeriVax-JE that contains two amino acid substitutions ( F107L and K138E ) in the SA14-14-2 E protein-coding region , suggesting that differences in the genetic background of mice may account for the variable neurovirulence [107] . More importantly , it should be pointed out that in our study , all the revertants recovered from the mice inoculated IC with SA14-14-2 appeared to have the G244E mutation , which is sufficient to confer lethal neurovirulence to the virus , corroborating that the parental SA14-14-2 virus is highly attenuated in neurovirulence . Furthermore , it is intriguing to note that the G244E substitution has been introduced into ChimeriVax-JE , in which no mortality occurs when eight 4-week-old ICR mice are injected IC with 104 PFU of the mutant virus [107]; therefore , it appears that neurovirulence may depend on the genetic background of the pathogen . Further investigation is needed to fully elucidate the neurovirulence and neuroinvasiveness of JEV . In summary , we show for the first time that E-244 in the ij hairpin of the viral E DII is a key regulator determining the neurovirulence of SA14-14-2 , and we also provide direct evidence that viral E can contribute to the neurovirulence of JEV and possibly other closely related encephalitic flaviviruses via its role in the early or late stage of viral replication in neurons . A detailed , complete understanding of the evolutionally conserved viral ij hairpin and its function in the virus life cycle will have direct application to the design of a novel and promising class of broad-spectrum antivirals ( e . g . , ligands and small molecules ) to expand the currently available preventive and therapeutic arsenal against infection with encephalitic flaviviruses .
An original stock of JEV SA14-14-2 was retrieved directly from a batch of commercial vaccine vials ( Chengdu Institute of Biological Products , China ) for viral genome sequencing and cDNA construction , to avoid any potential concern that its adaptation could occur during propagation in cell culture . This virus stock was propagated twice in BHK-21 cells to generate high-titer viral preparations for cell and mouse infection experiments . Stocks of JEV CNU/LP2 were derived from the infectious cDNA pBACSP6/JVFLx/XbaI [78] . BHK-21 cells were grown in alpha minimal essential medium ( α-MEM ) containing 10% fetal bovine serum ( FBS ) , 2 mM L-glutamine , vitamins , and penicillin-streptomycin at 37°C in 5% CO2 [78] . SH-SY5Y cells were cultivated in a 1∶1 mixture of MEM and Ham's F-12 nutrient mix supplemented with 10% FBS , 0 . 1 mM nonessential amino acids , and penicillin-streptomycin at 37°C in 5% CO2 [75] . NSC-34 cells were maintained in Dulbecco's modified Eagle's medium containing 10% FBS and penicillin-streptomycin at 37°C in 5% CO2 . As a vector , we used the BAC plasmid pBeloBAC11 [78] . First , four cDNA fragments covering the entire viral genome were cloned into the vector individually , then joined sequentially at three natural restriction sites ( BsrGI , BamHI , and AvaI ) to generate a single BAC clone that contained the full-length SA14-14-2 cDNA , named pBAC/SA14-14-2 ( Fig . 1 ) . The SP6 promoter sequence was positioned just upstream of the viral 5′-end , and an artificial XbaI run-off site was engineered just downstream of the viral 3′-end . A pre-existing XbaI site at nucleotide 9131 was removed by introducing a silent point mutation ( A9134→T ) ; this mutation also served as a rescue marker to identify the cDNA-derived SA14-14-2 . All mutations were created by overlap extension PCR . All PCR-generated fragments were sequenced . Detailed cloning procedures are described in Supporting Information . All BAC plasmids were purified by centrifugation using CsCl-ethidium bromide equilibrium density gradients . The closed circular plasmids were linearized by XbaI and mung bean nuclease digestion to produce DNA templates for in vitro run-off transcription . RNA was transcribed from a linearized plasmid with SP6 RNA polymerase as described [78] . The resulting RNA was stored at −80°C until needed . RNA yield was measured on the basis of the incorporation rate of [3H]UTP , and RNA integrity was evaluated by agarose gel electrophoresis . RNA was transfected by electroporation into cells under our optimized conditions ( 980 V , a 99-µs pulse length , and five pulses for BHK-21 cells; and 760 V , a 99-µs pulse length , and five pulses for NSC-34 and SH-SY5Y cells ) [78] . RNA infectivity was determined by infectious center assay as reported [78] . The infectious centers of foci were detected by decorating of cells with a mouse α-JEV antibody ( American Type Culture Collection [ATCC] , 1∶500 ) and a horseradish peroxidase-conjugated goat α-mouse IgG ( Jackson ImmunoResearch , 1∶1 , 000 ) , followed by staining with 3 , 3′-diaminobenzidine ( Vector ) . Total RNA was extracted with TRIzol reagent ( Invitrogen ) . Northern blot analysis was performed as described [78] . JEV genomic RNA was detected with an antisense riboprobe that binds to a 209-bp region ( nt 9143–9351 ) in the NS5 protein-coding region . The probe was synthesized with [α-32P]CTP by using the T7-MEGAscript kit ( Ambion ) . The blots were prehybridized , hybridized , and washed at 55°C . Autoradiographs were obtained by exposure to film for 24–48 h . Cells were lysed in sample buffer ( 80 mM Tri-HCl [pH 6 . 8] , 2 . 0% SDS , 10% glycerol , 0 . 1 M dithiothreitol , 0 . 2% bromophenol blue ) . Equal amounts of the lysates were run on SDS-polyacrylamide gels , transferred to polyvinylidene difluoride membranes , and subjected to immunoblotting as described [78] . The following polyclonal antisera were used as primary antibodies [75] , [76]: α-JEV ( mouse , 1∶1 , 000 ) , α-C ( rabbit , 1∶1 , 000 ) , α-pr ( rabbit , 1∶4 , 000 ) , α-E ( rabbit , 1∶500 ) , α-NS1 ( rabbit , 1∶1 , 000 ) , and α-GAPDH ( rabbit , 1∶10 , 000 ) . An alkaline phosphatase-conjugated goat α-mouse or α-rabbit IgG ( Jackson ImmunoResearch , 1∶5 , 000 ) was used for the secondary antibody , as appropriate . The specific signals were visualized by chromogenic membrane staining with a mixture of 5-bromo-4-chloro-3-indolyl-phosphate and nitroblue tetrazolium ( Sigma-Aldrich ) . Cells ( 5×105 ) were harvested by trypsinization , washed with phosphate-buffered saline ( PBS ) , and collected by centrifugation at 1 , 000×g for 5 min . The cells were resuspended in 250 µl of Cytofix/Cytoperm solution ( BD Biosciences ) and incubated at 4°C for 20 min in the dark . All subsequent wash and staining steps were performed in Perm/Wash buffer ( BD Biosciences ) . The cells were washed twice and incubated in 200 µl of mouse α-JEV antiserum ( ATCC , 1∶500 ) for 1 h at 4°C . Subsequently , the cells were washed twice and incubated in 200 µl of Alexa Fluor 488 goat α-mouse IgG ( Molecular Probes , 1∶1 , 000 ) for 1 h at 4°C . The cells were then washed twice and resuspended in 200 µl of Perm/Wash buffer . The samples were analyzed on a FACSAriaIII cell sorter with Diva 6 . 1 . 3 software ( BD Biosciences ) . For each sample , 50 , 000 events were collected within the linear range of detection . The full genome sequences of SA14-14-2 and its neurovirulent variants were determined as described [77] . Sequencing of the prM-E coding region of the E-244 mutants was done as follows: ( i ) amplification of a 2 , 069-bp cDNA by RT-PCR using a set of three primers ( prMErt , prMEfw , and prMErv; see Table S4 ) ; ( ii ) cloning of a 2 , 057-bp XhoI-SacII fragment into the pRS2 vector; and ( iii ) sequencing of ∼30 randomly picked independent clones containing the insert . Multiple sequence alignments were performed using ClustalX [116] . Female 3-week-old ICR mice ( Charles River ) were used . Groups of 10 or 20 mice were inoculated IC ( 20 µl ) , IM ( 50 µl ) , or IP ( 50 µl ) with 10-fold serial dilutions of virus stock in α-MEM . Mice were monitored for any JEV-induced clinical signs or death every 12 h for 24 days . The LD50 values were determined as described [75] , [77] . In all mice , viral replication in brain tissue was confirmed by plaque titration and/or RT-PCR [75] . All animal studies were conducted in strict accordance with the regulations in the Guide for the Care and Use of Laboratory Animals issued by the Ministry of Health and Welfare of the Republic of Korea . The protocol was approved by the Institutional Animal Care and Use Committee of the Chungbuk National University Medical School ( Permit Number: LML08-73 ) . All mice were housed in our animal facility located at the Chungbuk National University Medical School , and every effort was made to minimize suffering . Groups of 3-week-old female ICR mice ( n = 15 per group ) were infected IC with 103 PFU of virus in 20 µl of α-MEM; 10 control mice were inoculated IC with an equivalent volume of supernatant from uninfected control BHK-21 cell cultures at comparable dilution . At 3 , 5 , and 7 dpi , five randomly selected mice were transcardially perfused with ice-cold PBS , followed by 4% paraformaldehyde ( PFA ) . Brains were fixed in 4% PFA , embedded in paraffin , and cut into 6-µm sections . Brain sections were treated in microwave for antigen retrieval and incubated with 1% H2O2 in ice-cold methanol for 30 min to block endogenous peroxidase . They were then blocked with 1% normal goat serum and incubated with rabbit α-NS1 antiserum ( 1∶200 ) for 12 h at 4°C , followed by incubation with biotinylated α-rabbit IgG plus the avidin-biotin-peroxidase complex ( Vector ) . Signals were visualized by staining with 3 , 3′-diaminobenzidine solution containing 0 . 003% H2O2 and counterstaining with hematoxylin . The sequence and structure of the E ectodomain of WNV NY99 ( PDB accession code 2HG0 ) was used as template for the homology modeling . The sequence alignment was done using the online version of ClustalW2 [117] . Protein structure homology modeling was performed using the SWISS-MODEL Workspace , accessible via the ExPASy web server [118] . The generated model was visualized using UCSF Chimera 1 . 5 . 3 . The model is in agreement with a recent crystal structure of the E ectodomain of JEV SA14-14-2 ( PDB accession code 3P54 ) [80] . | A group of mosquito-borne flaviviruses that cause fatal encephalitis in humans is among the most important of all emerging human pathogens of global significance . This group includes Japanese encephalitis ( JE ) , West Nile , St . Louis encephalitis , and Murray Valley encephalitis viruses . In this work , we have developed a reverse genetics system for SA14-14-2 , a live JE vaccine that is most commonly used in JE-endemic areas , by constructing an infectious bacterial artificial chromosome that contains the full-length SA14-14-2 cDNA . Using this infectious SA14-14-2 cDNA , combined with a mouse model for JEV infection , we have identified a key viral neurovirulence factor , a conserved single amino acid in the ij hairpin adjacent to the fusion loop of the viral E glycoprotein , which regulates viral infectivity into neurons within the central nervous system in vivo and neuronal cells of mouse and human in vitro . Thus , our findings elucidate the molecular basis of the neurovirulence caused by JEV and other closely related encephalitic flaviviruses , a major step in understanding their neuropathogenesis . From a clinical perspective , the discovery of the viral neurovirulence factor and its role will have direct application to the design of a novel class of broad-spectrum antivirals to treat and prevent infection of JEV and other taxonomically related neurotropic flaviviruses . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"infectious",
"diseases",
"infectious",
"diseases",
"of",
"the",
"nervous",
"system",
"medicine",
"and",
"health",
"sciences",
"viral",
"diseases",
"encephalitis",
"japanese",
"encephalitis"
]
| 2014 | A Molecularly Cloned, Live-Attenuated Japanese Encephalitis Vaccine SA14-14-2 Virus: A Conserved Single Amino Acid in the ij Hairpin of the Viral E Glycoprotein Determines Neurovirulence in Mice |
An important task performed by a neuron is the selection of relevant inputs from among thousands of synapses impinging on the dendritic tree . Synaptic plasticity enables this by strenghtening a subset of synapses that are , presumably , functionally relevant to the neuron . A different selection mechanism exploits the resonance of the dendritic membranes to preferentially filter synaptic inputs based on their temporal rates . A widely held view is that a neuron has one resonant frequency and thus can pass through one rate . Here we demonstrate through mathematical analyses and numerical simulations that dendritic resonance is inevitably a spatially distributed property; and therefore the resonance frequency varies along the dendrites , and thus endows neurons with a powerful spatiotemporal selection mechanism that is sensitive both to the dendritic location and the temporal structure of the incoming synaptic inputs .
Neurons are constantly bombarded by thousands of synaptic inputs , so it is essential that neurons are able to listen selectively to subsets of these inputs . Throughout the sensory pathways , topographic maps ensure that neurons are able to sample a limited range of the stimulus space [1] . But the use of space is only one means by which input selectivity is achieved in the central nervous system . Another effective means is to respond selectively to particular temporal input patterns . A range of mechanisms can facilitate temporal selectivity ranging from pre-synaptic short-term plasticity [2]–[6] , learning strategies of specific temporal patterns [7]–[10] , to post-synaptic membrane resonances which enhance responses to specific input rates [11]–[13] . The focus of this study is the latter mechanism of resonance , membrane resonance , which has been traditionally considered a scalar property of a neuron: one neuron has one preferred resonance frequency [11] , [14] . This view , however , is inconsistent with the increasing awareness of the complexity of dendritic ramifications , the non-uniform spatial distribution of their ionic channels and highly localized non-linearities . Such elaborate biophysics can endow single neurons with multiple resonances occuring at a wide range of frequencies and bandwidths , and thus enable neurons to act as multi-dimensional input classifiers . Here , we explore this idea using both analytic methods and numerical simulations of neurons with both simplified and realistic dendritic structures . We show how spatial profiles of resonance frequencies emerge naturally in dendrites , facilitating selective filtering of synaptic inputs based on their location and temporal signature . Our findings thus counter the widely-held assumption that input selection is based on a single prefered frequency band regardless the location of the synaptic input .
Resonance in neuronal membranes has been described by many experimentalists and theoreticians [11] , [12] , [15]–[18]; it requires an interplay between at least two conductances with different dynamics . Figure 1A illustrates how an interaction between a membrane's passive electrical properties ( resistance and capacitance ) and one voltage-dependent current ( low voltage-activated potassium current , ) can give rise to a resonant membrane impedance ( ) comprised of two admittances: . The interplay between these admittances produces the impedance resonance in much the same way as the restorative and regenerative conductances interact to form a resonance . The first admittance , , is an effective leak ( red curve in Figure 1B ) that is mostly associated with the classic membrane passive RC-circuit ( time-constant ; see METHODS ) , and which acts as a shunt at high frequencies as schematically illustrated by the large red arrow below the plots . The second admittance ( blue curve in Figure 1B ) is due to the channels whose limited activation rate ( time-constant ) leaves them increasingly closed at frequencies higher than , as depicted by the small blue arrow at right . The sum of these two admittances often results in a minimum at a mid frequency range producing a peak impedance at a resonance frequency ( Figure 1A ) . This minimum occurs when the increase in counter-balances the drop of . Since the increase of takes place for frequencies higher than , the resonance frequency is always higher than . This is demonstrated in figure 1C where is color-coded for different values of and while is displayed as black line contours . Clearly , the resonance frequency and its sharpness ( Q ) depend on , , and , and through them on any biophysical parameters affecting the resting state of the membrane . As such , is affected by the reversal potential , membrane leak conductance , and maximal potassium conductance ( see METHODS , Figure 1C and Supplementary Figure S1 ) . As shown in Figure 1C , the resonance frequency increases monotonically both with increasing potassium channel density and with its steady state level ( set by ) . The sharpness of tuning Q depends on how much can decrease before the increase in takes place and on how close in frequency these two changes occur . Hence , the dependence of Q upon the biophysical parameters is complex . For instance , Supplementary figure S1 A2 illustrates how changes in the leak conductance produces nonmonotonic changes in Q . To conclude , even in an isopotential patch of membrane with a linearized model of channel dynamics , the resonance frequency can vary substantially ( 300% or 120<<350 Hz ) depending on a range of parameter values typically found at different locations of a dendrite ( Figure 1C and Supplementary Figure S1 ) . A key objective of our study is to explore the influence of “space” ( namely dendritic location ) on the resonance properties . To do so , we distinguish between local input impedance , , and the transfer impedance , that is the total transfer function between the input at location x and a recording electrode at the soma , as illustrated in Figure 1D . It has been shown [19] that if the membrane impedance is bandpass , then so are the transfer impedance and the cable space constant , a measure of the electrical compactness of the dendrite . Computing the transfer impedance using just a uniform membrane model already reveals a strong spatial profile of resonance frequencies as illustrated in Figure 1D ( see METHODS ) . This dependence arises mostly from an inherent mismatch between the resonance of the input impedance ( ) and that of the space constant ( ) as shown in figure 1E . By definition the space constant and the input impedance are related ( see Methods ) and the mismatch , which is influenced by , , and , is non-zero for a large set of parameters ( i . e . ; see Supplementary Figure S1 B1 ) . This implies that in most cases , a spatial profile of resonance frequencies emerges along the semi-infinite cable: When the injection and recording site are close to one another , the resonance frequency of the transfer impedance is mostly that of the input impedance . With increasing distance between both sites , the resonance frequency of the transfer impedance becomes more influenced by the resonance frequency of the frequency-dependent space constant . Figure 1F illustrates this effect and demonstrates that with plausible parameters the resonance frequency of the transfer impedance can change by as much as 11% over just the first 500 µm ( of a semi-infinite cable model ) . Thus , the mere spatial extent of a dendrite already results in a spatially distributed profile of resonant frequencies . A dendrite , however , is structurally far more elaborate than the simplified morphology and uniform membrane of the cable presented so far . Dendritic membranes , for example , often exhibit non-uniform distributions of ionic channels , as well as branching and tapering geometries . To understand such different cases , one can assume as a first approximation that a dendrite is constituted of small uniform cable segments ( piecewise constant approximation ) . The boundary conditions at each end of the uniform segment affect the spatial profile of resonance frequency of the transfer impedance . Therefore , we consider the effects of boundary conditions using a linearized cable model ( with parameters similar to Figure 1D , E , F ) . Figure 2B and C illustrates the spatial profile of resonant frequencies under two geometric configurations: the branching of daughter dendrites at the apical end ( Figure 2B ) and the attachment of a soma at the basal end ( Figure 2C ) . In both cases , the boundary conditon at the tip of the segment is given by a “lumped” impedance ( e . g . representing the impedance of the daugther dendrites lumped together ) . Moreover , this “lumped” impedance can be set to have different resonance frequencies by varying , , . In Figure 2A the “lumped” impedances are presented color coded by their resonant frequency from blue ( = 150 Hz ) to red ( = 420 Hz ) . The spatial profile produced by each resonant “lumped” impedance is compared to a control condition where the boundary impedance is that of an uniform semi-infinite cable ( shown in black in Figure 2A ) . Compared to the uniform semi-infinite boundary condition , the impedance at the recording location can shift considerably depending on the specific boundary condition and segment dimensions . For example , changes in the resonance frequency of the transfer impedance can be observed throughout the entire length of the segment in the case of a short segment ( 75 µm ) while in the case of a long segment ( 300 µm ) these changes are mainly located close to the modified tip . Interestingly , while boundary conditions modify strongly the profile of resonance frequency , the spatial profile of sharpness is not much affected ( see Supplementary Figure S2 ) . We then investigated the extent to which a spatially nonuniform conductance distribution contributes to the range of resonance frequencies expressed by a neuron . Simulations exploring the distribution of two conductances ( and ) were performed in four types of abstract morphologies: a cable , soma-and-dendrite , bipolar and y-dendrite model ( Figure 3 ) . The left panels of Figure 3 A–D provide a schematic of the optimized conductance distribution along the dendrite . Right panels provide the spatial profile of the resonance frequency ( red ) and sharpness ( blue ) . Optimizing the membrane properties to obtain a large range of resonant frequencies combined with moderate sharpness resulted in specific effects of the non-uniform distribution in each morphology . For the cable , a gradient of conductances with a constant but high produced the largest range of resonance frequencies as shown in Figure 3A . The spatial gradient of along the cable produces an increasing reversal potential toward its distal tip as well as an increasing total leak ( from 0 . 32 mS/cm2 to 1 mS/cm2 ) . Both effects tend to raise the input resonance frequency ( Supplementary figure S1 A1 ) . Moreover , because of the gradient of , each segment of this non-uniform cable will be connected at its proximal tip to a segment of lower characteristic frequency and at its distal tip , a segment of higher input resonance frequency . This configuration is similar to the configuration of a linear resonant cable producing the largest frequency range along its length ( Figure 2 B , C ) and the spatial profile of resonance frequency ranges from 292 to 325 Hz . Finally , the density of is constant and high ( 15 mS/cm2 ) and ensures a sharp tuning of input resonance ( Supplementary Figure S1 , A2 ) . Therefore , the optimization results extend the analytical insights obtained by linearization of the ionic channel dynamics . A similar gradient is observed in the case of a soma-and-dendrite morphology as depicted in Figure 3B . The density of is decreasing from 1 mS/cm2 to 0 . 83 mS/cm . The range of transfer resonant frequencies observed is both caused by the conductance-density gradient the discontinuous boundary condition introduced by the soma ( as analyzed in Figure 2 ) . Overall , the increased complexity of the ball-and-stick morphology increased both the range of frequencies expressed ( 256 to 315 Hz ) and the overall sharpness of tuning ( <Q> = 0 . 92 ) compared to the case of the finite cable shown in figure 3A . The density of is decreasing from 1 mS/cm2 to 0 . 83 mS/cm2 . The range of transfer resonant frequencies observed is both caused by the conductance-density gradient the discontinuous boundary condition introduced by the soma ( as analyzed in Figure 2 ) . Overall , the increased complexity of the ball-and-stick morphology increased both the range of frequencies expressed ( 256 to 315 Hz ) and the overall sharpness of tuning ( <Q> = 0 . 92 ) compared to the case of the finite cable shown in Figure 3A . The optimized conductance profile for the bipolar neuron morphology lead to an even larger range of resonant frequency and Q-factors ( Figure 3C ) . In the bipolar case , the range of transfer resonance frequencies differs in both dendrites mosty due to the different distributions of the leak conductance . In one branch , a low density of both and caused relatively low resonance frequencies of the transfer impedance along the branch while a high density in both conductances caused relatively high resonance frequencies in the other branch . As a result , the range of resonance exhibited in the whole neuron was large ( between 268 and 338 Hz ) and maintained good sharpness ( <Q> = 0 . 99 ) . Thus , thismorphological construct exploited both non-uniform densities and changes in boundary conditions between the soma and each of its two branches . Similarly , the optimized Y-branch produced a large range of resonance frequencies from its low resonance frequency in the parent branch to the high resonance frequency in the daugther branches ( Figure 3D ) . Thus , dendritic constructs such as branching , tapering and non-uniform channel distributions enrich the spatial distribution of resonant frequencies caused by space alone . For a more realistic experimentally reconstructed morphology ( downloaded from NeuroMorpho . org , see Methods ) , the non-uniform distribution of conductances , the complex branching and tapering of dendrites can lead to an even richer spatial distribution of resonance frequency as shown in Figure 4A . We optimized the density of and for each branch of this model . Each branch was allowed to have a linear gradient of these two channels and the optimization criteria was to find the model with largest range of resonance frequencies ( in the complete neuron ) while maintaing a reasonable sharpness ( <Q>>0 . 8 , see METHODS ) . Figure 4A illustrates the model neuron resulting from that first stage of optimization . At each location on the dendritic tree , the resonant frequency of is color-code ranging from 207 Hz ( blue ) to 247 Hz ( red ) . In this model based on a real morphology , the combination of dendritic geometry and non-uniform ion-channel distribution endow any morphologically realistic model neuron with a rich spatial profiles of resonance . Such spatially distributed and sharply tuned resonance frequencies can effectively act as spatiotemporal filters for a neuron's inputs , which leads us to consider in more detail the functional significance of these resonances . With distinct dendritic locations expressing a preference for certain frequencies , one can envision the dendrite as powerful spatio-temporal filter of synaptic inputs: viewed from the vantage point of the soma , each point on the dendritic tree has a preferred input modulation rate that it amplifies while attenuating all others input rates . This is demonstrated by the simulations in Figure 4B where the temporal and the spatial selectivity are illustrated separately ( see Methods ) . Temporal selectivity can be demonstrated when one set of synapses ( at fixed locations ) can cause a differential/preferential response at the soma of the neuron when stimulated with different temporal activation patterns , as illustrated in the scenario of Figure 4B1 . Here , the spatial distribution of the green synapses was chosen on the dendritic tree of Figure 4A so as the combined transfer function optimally responds to a 208 Hz modulated spike train while ignoring a 228 Hz input . This simulation demonstrates the dendritic temporal filtering abilities achieved with a combined spatial profile of transfer resonances . Note that in arriving at this result , we did not need to optimize the synapse properties , which are assumed to simply enhance signal transduction to ensure that the frequencies arising on the post-synaptic membrane are near the resonance frequencies shown in panel Figure 4A . Spatial selectivity is illustrated by two sets of synapses at distinct dendritic locations responding differentially to the same signal as shown in Figure 4B2 . The red synapses are located at dendritic locations corresponding to a resonance frequency of 228±4 Hz and the blue synapses at 208±4 Hz . When both groups were stimulated separately by Poisson processes modulated at 228 Hz ( see Methods ) , the input at the blue synapses generated only a few spikes at the soma ( blue trace ) . By contrast , the same input signal at the red synapses , elicited many more spikes ( red trace ) . The same signal therefore induced different somatic responses when conveyed to the neuron through distinct sets of synapses with different resonance properties to the soma . To conclude , a neuron can perform elaborate spatiotemporal filtering of its inputs utilizing the distribution of its dendritic resonances , a capability that is substantially more elaborate than is widely assumed possible of a neuron expressing only one prefered resonant frequency [12] , [13] , [20] .
In summary , building upon the work of Koch and colleagues [19] , [21] , we have shown that a model of a simple neuronal membrane with typical biophysical properties and ionic channels can readily exhibit a resonant transfer impedance . When viewed from a distance down the cable , the resonance can take a wider range of frequencies and bandwidths . This range expands greatly when considering nonuniform cable models with complex boundary conditions and changing ionic channel densities and types . Finally , the full power and versatility of this dendritic resonance idea comes into focus in a more realistic multi-compartmental model which allowed us to demonstrate its potential functional significance as it enables a neuron to serve as a spatiotemporal filter . Given the ubiquity and diversity of dendritic resonances , why has their functional significance been thus far neglected ? The answer probably lies in the commonly-held view that resonance mainly plays a role in synchrony ( and participation therein ) at lower frequencies ( e . g . , α , β , and θ-bands at <10 Hz ) . At those frequencies it is hard to distinguish experimental variability from a real range of resonance frequencies ( a range of 50% around 4 Hz is 2–6 Hz ) . At the much higher frequencies considered here ( and in only one previous report [14] ) , a 50% range translates to 225–375 Hz . Resonances in those ranges correspond to high gamma . Interestingly , in the lower auditory system , where neurons are known to express fast-activated potassium channels , these higher modulation frequencies can be transmitted by neuron to encode modulation of the sound energy . Temporal modulations at these frequencies convey periodicity cues critical in the perception of pitch [22] . Also , in more central neurons these rates can readily occur in the high-conductance state during which neurons are constantly bombarded with seemingly irregular firing rates [23] . As long as there is a temporal modulation ( envelope ) rate , dendritic transfer resonance can still filter relevant signals . It should be pointed that neurons with a rich variety of dendritic transfer resonance may rather be the rule than the exception . Indeed , as we have highlighted here both nonuniform channel conductance and boundary conditions enhance the usual range of transfer resonance expressed by a cable . There have been many studies demonstrating that channels are non-uniformly distributed on the dendrite [24]–[25] . Given that a diverse range of resonances is ubiquitous and inevitable in dendrites , we can speculate on further implications of our findings . A first important observation is the difference between resonant frequencies of the input versus transfer impedance: the input impedance dominates locally while the tranfer impedance is global insofar it spans the complete dendritic membrane along which an input signal travels to the soma . Plasticity can , in principle , differentially exploit local and global effects . At the local level , a signal that temporally matches the resonant frequency in the input impedance may trigger a large local voltage-depolarization giving rise to a calcium transient that , in turn , triggers plasticity mechanisms [26] . At the global level , a different ( but not mutually exclusive ) hypothesis is based on pre and post-synaptic spike times [27] . In this scenario , the combined synaptic input to a neuron triggers a post-synaptic spike , which then back-propagates into the dendritic tree and activates plasticity mechanisms . Since the strength of somatic depolarizaion depends on the global resonant frequency of the transfer impedance , the most likely inputs to induce spiking ( and hence plasticity ) are those with modulation rates that match this global resonance . A slight variation on the latter hypotheses is the case in which a “teacher” signal impinges onto the soma and triggers spikes . In that situation , the neuron can associate the modulation of the “teacher” signal to a specific the set of synapses that have an equal transfer resonance to the soma . Indeed , such a neuron would be responsive only when the preferred modulation rate at the synapses matches that of the teacher signal . Inputs from synapses with transfer resonance modulated at any other rate would not be carried out to the soma and would not interact constructively with the “teacher” signal . This situation is particularly interesting in the auditory system where low frequency cell could provide “teacher” signals to modulation detector neurons with dendritic branches spread across tonotopy ( such as octopus cells [28]–[30] or inferior colliculus stellate cells [31] ) . Since the output modulation rate of low frequency cells is determined by their location , while that of high frequency cell is not , cross-frequency modulation detectors could arise by such a learning of specific input location . This idea provides a neural basis to solve the central problem of linking the rate modulation of low and high frequency places in auditory pitch perception [32] . Thus , resonant frequencies in dendrites not only enable the neurons to perform elaborate spatio-temporal filtering , it can also have pivotal consequences for plasticity , and different plasticity mechanism could be activated by local or global post-synaptic potentials dependent on the temporal signature of the pre-synaptic signal .
The resonance introduced by can be described in the Fourier domain [16] , [19] , [34] after linearizing the current balanced equation around the resting membrane potential . A small variation in the potassium current is composed of three terms: an ohmic part ( i . e . the steady-state potassium conductance ) and two other terms describing the increase and decrease in subsequent changes in activation and inactivation of the channels . The membrane impedance is given by , where is the effective conductance of the membrane composed of the leak and the steady-state potassium conductance and is the effective membrane time constant . The conductance represents the extra conductance associated with opening additional activation gates following a variation of voltage around rest . Correspondingly , represents the decrease in conductance associated with the closing of some inactivation gates . The frequency dependence of and allows a further simplification . Since ms [33] while ms , any voltage changes at frequencies above 12 . 5 Hz have little effect on the inactivation and thus we can neglect effect of the inactivation . Therefore , we use the following expression for the membrane impedance in Figure 1A: . For the spatially extended models ( Figure 1D , E and 2 ) , the current-balanced equation for each compartment is similar to that of the membrane with the addition of terms describing the current between compartments which is proportional to the axial resistance . The space constant for a dendrite describes the distance between an injection and recording site for which the DC component has decayed of a factor . More generally , the membrane impedance determines the frequency dependent space constant , of the dendrite ( where denotes the real part of a complex number ) . The transfer impedance between any two points separated by a distance can be computed by solving the generalized cable equation given in the Fourier domain by with its appropriate boundary conditions , where . For the semi-infinite cable described in Figure 1 , its magnitude reads and this was used to compute the spatial profile of the resonant frequency and spatial profile of Q-factor , denoted ( see below ) . The space constant of the semi-inifinite cable is thus related to input impedance by where with . This relationship demonstrates why an inherent mismatch exists between the resonance frequency of the space constant is different than that of the the input impedance . When more specific boundary conditions are used ( Figure 2 ) , the transfer impedance does not easily relate to the concept of space constant . Different approaches [21] , [35] , [36] can be used to compute and from the boundary conditions . We have used the expression of rule I and III of Koch and Poggio [21] . Numerical simulations to determine the influence of complex dendritic morphologies on resonance were performed using the NEURON+Python [37] , [38] software . In order to explore the wide range of parameters that leads to significant spatio-temporal input filtering , we performed evolutionary optimizations [39] , [40] of abstract ( cable , bipolar , multipolar , “Y” dendrites ) model neurons ( Figure 3 ) as well as morphological detailed model neurons ( see Figure 4 ) . Optimization by evolutionary algorithms involved two critical steps: parametrization of the model neurons so they can be systematically optimized and , the quantitative assessment of the models to guide the optimization . The parameters used for the optimization are summarized in Table S1 . These parameters are based on neurons from the early auditory pathway [31] , [41]–[43] . Note that in each of these models the segment diameters as well as the conductance densities may follow a linear gradient between an initial and ending value . The diameter is additionally constrained not to increase . The length of the dendritic branches in the abstract models is adjusted so that the total length of the path between soma and termination point is 200 micron . The quantitative assessement of the models we are established by two means . First , the spatial profile of resonance frequency allows us to compare quantitatively the range of frequencies obtained on a fixed morphology . For the linear cable , this is obtained by numerically computing . For the compartmental model with nonlinear channel dynamics , an “impedance amplitude profile”- current ( ZAP-current [44] ) is injected at a specific location in the dendritic segment and the frequency at which the membrane potential is maximal ( ) is taken as the resonant frequency ( i . e . ) . The second assemement is based on the sharpness of tuning , also called the Q-factor . Rather than defining the Q-factor by , as done in various study [12] , [19] , [45] , we use a definition focusing on the bandpass properties offered by dendritic resonance , that is: how quickly the resonant response drops around the resonant frequency . The Q-factor is thus defined by where denotes the bandwidth of the resonance and are such that . The spatial profile of the Q-factor , is determined by computing Q at each point along the dendrite . We can then decide to optimize for range of resonance frequencies obtained , the overall Q factor or both Simultaneously ( as in Figure 3 ) . To demonstrate the spatio-temporal filtering in a spiking model with a realistic morphology , a neuron model with an archetypical multipolar morphology [46] ( “P2-DEV139” originally published in [44] available at the NeuroMorpho . org archive [47] ) is simulated and optimized . We optimize this model neuron in two steps . First , the membrane properties ( Table S1 ) are modified iteratively to obtain a large range of resonance frequencies ( resulting in a 207 to 247 Hz range – see Figure 3A ) and with reasonable sharpness in the dendrites ( 0 . 79<Q<0 . 89 ) . Second , while using these optimal membrane parameters , we optimize synaptic parameters and input parameters for two tasks: temporal or spatial filtering . Both tasks exemplify the single property of the optimized neuron , namely to perform spatio-temporal input classification . For both tasks , the synaptic input parameters optimization is performed as follows . Inputs spike trains onto 25 synapses are obtained from independent non-homogeneous Poisson processes ( NHPP ) with sinusoidal firing rate where and are both optimization parameters . A DC current is added to the soma segment representing the global background activity . To demonstrate the temporal selectivity , we fix the modulation frequency to a target frequency ( Hz ) or a null frequency ( Hz ) . The synapses' location and strength is optimized for a discrimination task: output spike rate is maximized for and minimized for , that is , the location and strength is kept identical for the two different inputs ( figure 3B1 , green dots ) . Because the synaptic locations are the same in both cases , the neuron can only use temporal information of the input to filter the target from the null signal . To demonstrate the spatial selectivity , we fix the input frequency at Hz and optimize synapses' location and strength for two different sets of synapses: the “target set” which should maximize the output firing rate and the “null set” which is optimized for a different frequency . Because the input signal is identical in both cases , the neuron can only use the location of the synapse to filter one signal but not the other ( Figure 3 , B2 ) | Neurons are constantly bombarded by thousands of inputs . Synaptic plasticity is generally accepted as a mechanism to select certain inputs by strengthening their synapses while reducing the effects of others by weakening them . Another biophysical mechanism to select inputs is through membrane resonance that enhances neuronal response to inputs arriving at a specific temporal rate while reducing others . In the classical view , a neuron has one such resonance frequency at which inputs can be preferentially filtered . By dissecting the biophysical mechanism underlying neuronal resonance we find that neurons in fact express a wide range of resonance frequencies spatially distributed along their dendrites . We further show that such dendritic resonance can endow a neuron with a true spatio-temporal filtering property of its inputs: neurons can preferentially filter inputs based on their dendritic location and/or temporal signature . We speculate that this new insight has pivotal consequences for learning and plasticity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"specimen",
"preparation",
"and",
"treatment",
"mechanical",
"treatment",
"of",
"specimens",
"cellular",
"neuroscience",
"specimen",
"disruption",
"neuronal",
"morphology",
"computational",
"neuroscience",
"single",
"neuron",
"function",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"electroporation",
"biophysics",
"neuroscience",
"coding",
"mechanisms",
"research",
"and",
"analysis",
"methods"
]
| 2014 | Spatially Distributed Dendritic Resonance Selectively Filters Synaptic Input |
The Genographic Project is studying the genetic signatures of ancient human migrations and creating an open-source research database . It allows members of the public to participate in a real-time anthropological genetics study by submitting personal samples for analysis and donating the genetic results to the database . We report our experience from the first 18 months of public participation in the Genographic Project , during which we have created the largest standardized human mitochondrial DNA ( mtDNA ) database ever collected , comprising 78 , 590 genotypes . Here , we detail our genotyping and quality assurance protocols including direct sequencing of the mtDNA HVS-I , genotyping of 22 coding-region SNPs , and a series of computational quality checks based on phylogenetic principles . This database is very informative with respect to mtDNA phylogeny and mutational dynamics , and its size allows us to develop a nearest neighbor–based methodology for mtDNA haplogroup prediction based on HVS-I motifs that is superior to classic rule-based approaches . We make available to the scientific community and general public two new resources: a periodically updated database comprising all data donated by participants , and the nearest neighbor haplogroup prediction tool .
The plethora of human mitochondrial DNA ( mtDNA ) studies in recent years has made this molecule one of the most extensively investigated genetic systems . Its abundance in human cells; uniparental , nonrecombining mode of inheritance; and high mutation rate compared to that of the nuclear genome , has made mtDNA attractive to scientists from many disciplines . Knowledge of mtDNA sequence variation is rapidly accumulating , and the field of anthropological genetics , which initially made use of only the first hypervariable segment ( HVS-I ) of mtDNA , is currently being transformed by complete mtDNA genome analysis [1] . While contemporary combined sources offers approximately 65 , 000 HVS-I records ( Oleg Balanovsky , unpublished data ) and over 2 , 000 complete mtDNA sequences , difficulties remain in standardizing these published data , as they report varying sequence lengths and different coding-region SNPs , and apply any number of methodologies for classifying haplotypes into informative haplogroups ( Hgs ) [2 , 3] . For example , some studies have defined the HVS-I range to comprise nucleotides 16093–16383 [4] , some 16024–16365 [5] , some adhered to the widely accepted definition of 16024–16383 [6] , while others extended the reported range to include positions such as 16390 and 16391 due to their predictive value in identifying certain specific clades [7 , 8] . Even more serious is the problem of Hg assignment , which , in the absence of complete sequence data , is best achieved by genotyping a combination of coding-region biallelic polymorphisms . Forensic studies ( which comprise a significant portion of the existing dataset ) and many population studies published before 2002 have predicted Hgs based on the HVS-I motif alone , thereby ignoring the occurrence of homoplasy and back mutations [2 , 9] . Moreover , it has been shown that many published mtDNA databases contain errors that distort phylogenetic and medical conclusions [10–15] . Therefore , it has become abundantly clear that a phylogenetically reliable and systematically quality-controlled database is needed to serve as a standard for the comparison of any newly reported data whether medical , forensic , or anthropological [7] . The Genographic Project , begun in 2005 , allows any individual to participate by purchasing a buccal swab kit . Male samples are analyzed for a combination of male specific Y chromosome ( MSY ) short tandem repeat loci and SNPs . Female samples undergo a standard mtDNA genotyping process that includes direct sequencing of the extended HVS-I ( 16024–16569 ) and the typing of a panel of 22 coding-region biallelic sites . Results are returned anonymously through the Internet ( http://www . nationalgeographic . com/genographic ) after passing a multi-layered quality check process in which phylogenetic principles are applied throughout , and which is supported by a specialized laboratory information management system . HVS-I haplotypes are reported based on the direct sequencing results . Hgs are defined by a combined use of the 22-SNP panel results and the HVS-I haplotypes . Following successful typing and reporting of the genotyping results , each participant may elect to donate his or her anonymous genotyping results to Genographic's research database . The magnitude of the project and its worldwide scale offer a unique opportunity to create a large , rapidly expanding , standardized database of HVS-I haplotypes and corresponding coding-region SNPs . Here , we report our experience from genotyping 78 , 590 public participants' mtDNAs during the first 18 months of the project . First , we describe our genotyping process and quality check measures and our considerations in designing them . Second , we report the unique insights that the standardized database supports with respect to estimation of the frequencies of transversions , transitions , heteroplasmies , indels , back mutations , and homoplasy occurring in both the HVS-I and the coding-region biallelic sites . Third , we present a new nearest neighbor ( NN ) –based methodology developed for Hg labeling , suggest it as an Hg prediction tool for validation of both new and previously reported databases , and demonstrate its superior performance over rule-based approaches , given a sufficiently large reference database . Finally , we make available to the scientific community and general public two new resources: a database ( which will be periodically updated ) containing the data donated by participants as an open source research database , and the NN analytical tool , which allows the comparison of any comparable data to the entire expanding Genographic dataset for quality control and predictive purposes .
The genotyping parameters associated with the reference database are presented in Table 1 . The overall first pass genotyping success rate for the entire process including DNA extraction , sequencing , and SNP genotyping was 98 . 5% . The average time needed to complete the first genotyping attempt was 31 days . All samples were attempted with bidirectional sequencing , but 13 . 9% contained the transition T16189C , which blocks the sequencing reaction beyond this position , and these provided data from only one strand . Of the remaining 86 . 1% reported samples , forward , backward , and bidirectional sequencing were successful in 99 . 7% , 99 . 7% , and 99 . 4% of the samples , respectively . The alternative forward sequencing primer ( Table S1 ) was used once , while the use of an alternative reverse sequencing primer was mandated in approximately 0 . 15% of the samples . A total of 83 . 2% of the samples was successfully genotyped for the complete panel of 22 SNPs . The success rate of inferring an Hg by this SNP panel was 94 . 7% , while 3 . 2% and 2 . 1% of the samples were labeled as inconsistent or uninformative , respectively . The total number of samples from project inception in which post-DNA-extraction sample mix-up was suspected due to clear nonconcordance between Hg labeling , as suggested by the HVS-I motif and the SNP genotyping , was 19 ( 0 . 00024% ) . The total number of samples from project inception in which the genotyping process could not be completed after attempting genotyping from both buccal swabs provided by the participant was 21 ( 0 . 00027% ) . Hg frequencies observed in the entire database , the reference database , and the consented dataset of 21 , 141 records are given in Table 2 . In the entire database , the most frequent Hg was Hg H ( 38 . 2% ) . When the database was collapsed into macro Hgs L ( xM , N ) M , and N the following frequencies were observed , respectively , 4 . 54% , 3 . 42% , and 92 . 04% . Table S4 provides the observed transitions , tranversions , insertions , and deletions for the entire database and further delineates their frequencies within each Hg for the reference database . Note that inferences regarding the number of times each mutation occurred within each Hg are impossible to determine from this table . The total numbers of distinct transitions and tranversions observed were 343 and 199 , respectively . The total numbers of distinct insertions and deletions observed were 35 and 15 , respectively . Table S5 describes , for the entire database , the number of distinct heteroplasmies observed and further delineates within the reference database their distribution within each Hg . The total number of distinct heteroplasmies was 152 . As it is difficult to establish the threshold of heteroplasmy detection by direct sequencing with current technologies , it is likely that the heteroplasmies found are an underestimate [16] . The results described in this section are from the reference database to provide maximum phylogenetic resolution . Homoplasy is the phenomenon in which the same mutation is found in two distinct phylogenetic branches of the mtDNA tree . Back mutation is defined herein as the phenomenon by which a position considered characteristic or diagnostic to a certain Hg has reverted to the ancestral state . It is clear that the phenomenon can affect any other position as well . The result of both phenomena can be haplotypes that are identical by state but not by descent ( Figure 1 ) , and can therefore bias interpretation of databases that make use of HVS-I haplotypes alone to infer Hg labeling or shared ancestry . In addition , these phenomena can also lead to an underestimation of population genetic distances . The extensive database presented herein contains numerous examples of these phenomena , of which many are well known while others are previously unreported . Table S6 shows the number of times that all classic HVS-I Hg-defining mutations are present as part of the haplotype motif in all reported Hgs . Table S7 shows the number of times that the same haplotype occurs in different Hgs for the portion of the reference database that overlaps with the consented database . Table S8 shows the number of times that a sample was assigned to an Hg by the SNP genotype , but did not harbor the classic HVS-I motif as defined in Table S2 . Unfortunately , the scope of this paper is too limited to describe all examples and , therefore , we focus on a few examples that emphasize the magnitude of these phenomena . The haplotype that shows no polymorphic changes when compared to the revised Cambridge Reference Sequence ( rCRS ) was well-reported under Hgs R* , U* , HV* , H , V , and their sub-branches [2 , 5 , 17 , 18] . These Hgs , which are frequent in populations of European ancestry , are expected to be frequent among the project's largely North American participants . Indeed , our database can be considered to contain an extensive sampling of European-derived populations . Since it is hard to decide whether the HVS-I haplotype of the ancestor of Hg R was rCRS or 16519C when mtDNA positions 16024–16569 are considered , the existence of the former in these Hgs can either represent identity by descent or identity by state due to homoplasy ( Figure 1 ) . Whether identical by descent or by state , the use of the 22-SNP panel allows the accurate placing of each mtDNA genome into a single Hg . Of a total of 463 mtDNA genome sequences that contain the rCRS ( 16024–16569 ) as their HVS-I haplotype , 416 , 26 , 1 , and 20 would have been assigned to Hgs H , HV* , U* , and V , respectively , when typed with our 22-SNP panel ( Table S6 ) . Likewise , any study of a European population that used HVS-I information only and labeled all rCRS samples as identical or as Hg H , is likely to incorrectly assign about 10% of these samples . The large size of the database allows us to estimate the frequencies of additional examples of homoplasy that were previously reported in the literature . Positions 16343G , 16356C , and 16270T are considered characteristic of Hgs U3 , U4 , and U5 , respectively . These positions actually occur in Hg H in 0 . 7% , 0 . 1% , and 3 . 9% of the cases , respectively . Positions 16224C and 16311C are widely considered to characterize Hg K . Our database , however , shows both a branch within Hg H that carries haplotype 16224C–16278T–16293G–16311C and branches within Hg K that lack positions 16224C or 16311C . The characteristic positions for Hg J and T are 16069T–16126C and 16126C–16294T , respectively . Several samples in our database shared the haplotype 16069T–16126C–16294T that contains both characteristic positions and proved to belong to Hg J . Haplotype 16223T–16519C occurred within Hgs H , M* , N* , U* , and W . More complex haplotypes , such as 16223T–16355T–16519C , occurred under both L3* and M* . Haplotype 16223T–16295T–16519C occurred in Hgs M* and W . The combination of positions 16189C–16217C occurred under both Hg B5 and N* . The important branching point between macroHg N and its daughter , macroHg R , is marked by two transitions , T12705C and T16223C . Our database shows that 2 . 5% of all preHg R mtDNA genomes have lost polymorphism 16223T and 1 . 1% of all R mtDNA genomes gained this mutation , mostly in the K1a1b1a lineage [19] . More specifically ( Table S8 ) , Hg I is characterized by HVS-I positions 16129A–16223T–16391A . Of the 421 Hg I mtDNA genomes defined by the relevant coding-region SNPs , 1 . 2% , 1 . 0% , and 3 . 3% have lost positions 16129A , 16223T , or 16391A , respectively . Of the 282 Hg W mtDNA genomes defined by the relevant coding-region SNPs , 1 . 4% and 15 . 2% have lost positions 16223T and 16292T , respectively . Of the 229 Hg C mtDNA genomes defined by the relevant coding-region SNPs , 2 . 2% , , 2 . 6% and 0 . 4% have lost positions 16223T , 16298C , and 16327T , respectively . These examples of positions that have experienced back mutation cannot indicate the number of times that each position has reverted during the Hg's evolution , as within-Hg resolution is not part of the presented database . The results presented below are from the reference database to provide the maximal phylogenetic resolution . Coding-region SNPs , used as reliable markers to define Hgs because they are considered stable evolutionary events , are nevertheless not entirely stable [19–21] . The dataset reported here supports this notion , and the portion of samples in which the SNP genotyping results were shown to be “inconsistent” with the expected phylogenetic hierarchy provides an important opportunity to estimate the extent of this phenomenon . Table S9 gives the number of times each of the tested SNPs occurs in different branches of the phylogeny . The overall frequency of samples in which inconsistency was observed was 3 . 2% . We note that excluding the 9-bp deletion at position 8280 would decrease the frequency to 2 . 0% . We highlight a few examples here . The most trivial is the occurrence of transition A13263G ( which we use to identify Hg C ) in Hg W . The phylogeny supported by the remaining panel of 21 SNPs correctly places the samples as belonging to Hg W . The occurrence of this transition under Hg W actually defines the samples as belonging to sub-Hg W3 [22] . Hg H , descending from R0 , is expected to harbor transitions A11719G , T14766C , and T7028C . However , 83 of the total 6232 Hg H samples lack transition A11719G , of which 73 share the HVS-I position 16316G . The phylogeny supported by the remaining panel of 21 SNPs correctly places the samples as belonging to Hg H , with this subset probably representing a monophyletic clade characterized by the loss of transition A11719G and gain of position 16316G that has not yet been named . An interesting issue concerns transition T12705C , which is the only coding-region mutation known to separate Hgs R and N [21] . Three samples ( 0 . 1% ) out of the 2 , 923 that were labeled by the remaining 21-SNP panel to be pre-R lineages did carry this transition , all of which were in Hg L sub-branches . Conversely , a total of 13 , 686 samples were labeled by the remaining panel of 21 SNPs to be lineages within Hg R , of which seven ( 0 . 05% ) did not carry transition T12705C ( but all carried SNPs typical of Hgs within R ) . These findings emphasize the importance of this position as separating Hg N from R . To quantify the effectiveness of the NN/weighted NN ( w-NN ) method combined with our reference database in mtDNA classification , we tested our ability to recover the classification revealed by the coding-region SNPs in the Genographic database . We consider classification into 23 basal Hgs based on our most extensive SNP typing protocol ( 22 coding-region SNPs ) as a “gold standard” classification ( correct with a very high probability ) , and use it for comparison of the performance of our rule-based and w-NN classification approaches , when classifying based on HVS-I information only ( without using the SNPs for classification ) . For this purpose , we adopted a leave-one-out cross-validation approach , i . e . , each of the 16 , 609 samples for which we have 22 SNPs was left out , and the 16 , 608 remaining samples were used as a “reference” database for NN/w-NN . The accuracy obtained for recovering the coding-region Hg assignment by the NN/w-NN approaches was 96 . 72% and 96 . 73% , respectively ( Table S10 , last row ) . While this difference is tiny , we see consistently throughout Table S10 that w-NN does slightly better than NN ( win-loss-tie ratio of 35-4-5 ) . We also applied the rule-based approach ( Table S2 ) based on HVS-I only , and obtained an accuracy of 85 . 3% ( Table S10 ) . Our conclusion from this experiment is that the NN-based approaches can support much higher accuracy in classification of our samples ( and samples coming from similar populations ) based on HVS-I only , when utilizing the Genographic database as reference . Table S10 details the results of repeating the same experiment with a variable number of SNP panels . We studied the level of haplotype saturation with respect to different HVS-I haplotypes and polymorphic sites present in the database by randomizing the order of the samples in the entire database and plotting the number of newly observed HVS-I haplotypes as a function of the accumulated number of samples ( Figure 2 ) . We repeated our analysis for the subsets of Hgs known to represent typically African , West Eurasian , East Asia-Americas , and South Asian mtDNA gene pools , and for Hg H haplotypes . Next , we repeated the analysis for the number of polymorphic sites obtained as a function of accumulated number of samples for the same categories . The entire database of 76 , 638 samples was included in this analysis , within which 29 , 267 belonged to Hg H . A total of 11 , 346 HVS-I haplotypes were observed in this set ( Table S3 shows the partial list of the observed haplotypes in the consented database ) . Note that homoplasy among these haplotypes is ignored , and the total number of phylogenetically independent haplotypes would have been higher if Hg information had been considered . Figure 2A shows the obtained results for all the haplotypes ( 11 , 346 ) , for the groups of Hgs grossly affiliated with Africa ( 1 , 348 ) , East Asia-Americas ( 1 , 663 ) , South Asia ( 583 ) , West Eurasia ( 7 , 684 ) , and Hg H ( 2 , 637 ) . Hgs in which geographic affiliation is uncertain ( N* , R* ) were excluded from the analysis . Figure 2B repeats the analysis for a limited number of samples to allow better comparison with the less-represented geographic groups . Figure 2C and 2D shows the application of the same analysis to the observed HVS-I polymorphic sites . We have utilized our database to search for evidence of Neanderthal origin for any of the samples , and for any discrepancies that might be attributed to recombination . On the Neanderthal question , we first extracted from GenBank all six Neanderthal HVS-I sequences of length at least 300 bp ( Table S11 ) . It is now accepted that a combination of five HVS-I mutations ( 16037G , 16139t , 16244A , 16262T , and 16263 . 1A ) , which appears in all of these samples , distinguishes these Neanderthal sequences from modern humans [23] . While all of these five mutations have in fact been observed in our full database of 78 , 590 samples , no combination of any two of them has appeared in any sample . However , since these six samples may not represent the full diversity of Neanderthal lineages , we have also investigated separately the level of divergence they show from our entire database . No sample in our database is as divergent as these Neanderthal samples , in terms of its distance from its nearest neighbor outside its own Hg , or its distance from the rCRS , which we take to represent a “random” modern human mtDNA ( Table S11 ) . We also observe that the most divergent samples in our database all carry well-known HVS-I motifs characteristic of African Hg L branches . While it is difficult to translate these findings into probabilities , it is clear that our results do not support the existence of mtDNA samples of Neanderthal ( or other archaic Homo ) origin in our database . In the search for recombination , we concentrated on our reference database . If there was a detectable level of recombination in mtDNA , it should lead to phylogenetic inconsistencies in the 22-SNP genotypes . For example , if there was recombination between an Hg H mtDNA and an Hg M mtDNA , where the M sample “donated” its region between nucleotides 9000 and 12000 into the Hg H sample , then positions 10400 ( Hg M ) , 10873 ( Hg N ) , and 11719 ( R0 ) , which are in this region and hence in their non-rCRS state , should be “inconsistent” with positions 7028 ( Hg H ) , 14766 ( Hg HV ) , and 12705 ( Hg R ) , which are in their rCRS state . Thus , we extracted all the samples in our reference database that were “inconsistent” ( a total of 538 records ) . Of these , 521 can be explained by a single inconsistency , which can be attributed to a single repeated/back mutation rather than recombination . The remaining 17 require two repeated mutations to explain them . Nine of these 17 cannot be explained by a single recombination event . The remaining eight fall under the Hg H branch , described above , which is marked by the back mutation at position 11719 and by the HVS-I transition A16316G . The second inconsistency in seven of these samples involves the 9-bp deletion at position 8280 and the eighth sample involves an inconsistency in position 13368 ( Hg T ) . As all eight occurred under a phylogenetically consistent branch , we attribute them to repeated mutation rather than a recombination event . We thus conclude that we can find no evidence of recombination in our reference database .
The Genographic Project allows members of the public to participate in a real-time anthropological genetic study . Since its inception in early 2005 , over 188 , 000 individuals have joined the project , of which over 55 , 000 have submitted their mtDNA or MSY results to the research effort , illustrating the high level of interest . Because any member of the public may participate in the study , rigorously controlled group affiliation data will not typically be obtained for the samples . Furthermore , it is clear that the accumulated database is biased towards countries in which the project is well known and where the kits are economically accessible to a significant fraction of the population . The fact that 95% of the participation kits were ordered in the US and Western Europe is consistent with the Hg frequencies observed , and suggests that the majority of the participants are of West Eurasian ( probably European ) ancestry . The importance of improving the quality of the global shared mtDNA database was recently reemphasized and summarized by Bandelt et al [24] . The strict uniform adherence to standard analytic and genotyping protocols across tens of thousands of samples makes the current study an ideal resource for the scientific community . We tried to consider all previously identified sources of errors while designing our genotyping , analysis , and reporting tools . Our database is unique for a project of this scale in using sequencing of both strands of the HVS-I as a standard procedure to assure high-quality data . The same goals led us to incorporate standard coding-region SNP genotyping on all samples . The entire analysis is “pen-less” to avoid any typographic mistakes , and a series of computational quality control measures are embedded in it . Despite the rigorous quality check procedures implemented , we still anticipate some inaccuracies in the database , but believe that these genotyping standards raise the bar on mtDNA genotyping and represent good progress towards more reliable databases . A few simple measures can be suggested to facilitate future assembly of mtDNA databases . First , as sequencing procedures have become more efficient and stretches of 600 bp can easily be obtained , we suggest standardizing the reported “HVS-I” range to include positions 16024–16569 as presented herein . Second , it would be worthwhile to create a standard list of coding-region SNPs used by the scientific community for Hg assignment and change to alternative coding-region SNPs defining the same Hg when there is a reason to suspect that the standard SNP is misleading due to homoplasy or back mutation . We make available our quality check measures as a model for any future mtDNA database submitted for publication . The database reported herein is very informative with respect to the mtDNA phylogeny , including the frequencies of the observed haplotypes , transversions , transitions , indels , and heteroplasmic positions both in the coding and control regions ( Tables S3–S10 ) . No highly divergent ( e . g . , Neanderthal ) sequences were observed , despite more than doubling the total number of sequences examined , and no evidence for recombination was found . The database did , however , provide evidence for homoplasy and back mutations affecting a low , but not insignificant , percentage of the samples both at the HVS-I and the coding-region SNPs chosen herein . For the coding-region SNPs , even these phenomena do not usually prevent the correct positioning of an mtDNA genome in the phylogeny , as the latter is based on the identification of a string of positions and not a single one . For the HVS-I , our analysis shows that while the use of Hg labeling techniques based on HVS-I variation have an overall good correlation with coding-region SNP genotyping , caution should be used in general , and , in certain specific cases , prediction is best avoided . In population-based studies of large sample size , these phenomena will likely have a small affect on the overall conclusions . However , for individual genotyping , as studied in genealogical or forensic cases , these percentages may be sufficient to preclude , for example , a firm conclusion regarding the time to most common recent ancestor of a set of samples for which only HVS-I information is available . The NN methodology presented herein , when jointly used with our reference database , has been shown to assign more mtDNA genomes to their correct Hg than prediction methods based on the classic set of HVS-I motifs . Our genotyping strategy , associating each of the HVS-I unique mutations with an Hg confirmed by a coding-region SNP , supplies the needed infrastructure for developing the NN methodology . It is clear that the high prediction score of NN/w-NN is a function of the size of the reference database collected within the population , in which the NN/w-NN methodology is implemented along with the length of the analyzed fragment in the HVS-I . For this study , and considering the large reference database , it was shown that , when no coding-region genotyping was done and Hg prediction was based solely on HVS-I classic Hg-determining rules , as many as 15% of the predictions were wrong , while the w-NN yielded an accuracy of 96 . 73% . In the sample set studied , the high rates of failure in predicting the correct Hg using HVS-I based rules alone is likely the result of high prevalence of Hgs for which no satisfactory predictive rules exist ( such as Hg H and HV* ) and to a lesser extent from phenomena like homoplasy or back mutations . To illustrate how the use of the w-NN methodology requires a joint use of a reliable relevant reference database for the studied population , we applied the w-NN methodology and our current reference database to published databases that are external to the Genographic Project and from various populations . West European and non-West Eurasian sequences , the two extremes , yielded prediction scores at a high and a low of 93 . 8% and 77 . 9% , respectively ( data not shown ) . Therefore , we make the NN prediction methodologies available on our Web site ( http://www . nationalgeographic . com/genographic ) in two forms: a ) the NN independent code to be used with any reference database and b ) in combination with an upload tool that allows the NN methods to be applied to uploaded samples using the Genographic reference database . As emphasized , we expect that the best prediction scores will currently be obtained in samples of West Eurasian ancestry for the 23 basal Hgs defined here , and that the predictions will gradually improve for other populations as the Genographic Project progresses and worldwide samples are obtained and included in the reference database , and as more coding-region SNPs are used to further resolve the basal Hgs into their sub-clades , a process actively underway in the Genographic research consortium . An interesting question that can be examined using our database relates to the effect of protocols using variable numbers of coding-region SNPs on the accuracy of Hg assignment when compared with the classification of the reference database using the full 22-SNP protocol as a gold standard ( as if 100% accurate ) . Table S10 gives the results for several coding-region SNP protocols of which the 10- , 20- , and 21-SNP protocols were previously used by the Genographic Project . These data show that a high degree of predictive accuracy was rapidly achieved as SNPs were added . When no SNPs were used , the best prediction methodology was with w-NN and yielded an accuracy of 96 . 73% . The most important single SNP in our population , 7028 ( Hg H ) , allows 98 . 18% accuracy ( w-NN ) on its own . The initial panel of ten SNPs , when combined with the HVS-I information , is responsible for 99 . 81% ( w-NN ) of the Hg assignment accuracy achieved , and the last 12 SNPs are needed to resolve the remaining small portion of the samples ( Table S10 ) . Our large database allows us to make some simple measurements of haplotype and polymorphic site saturation . Figure 2 shows that even the large number of samples collected in our study does not reach HVS-I haplotype saturation . The discrepancy between the shapes of the haplotype and polymorphic site curves probably means that the number of observed polymorphic sites is closer to saturation than the number of observed haplotypes , which in turn suggests that shuffling of the same polymorphic sites , through homoplasy and back mutations , is the dominant mechanism that increases haplotype variation . These results are not surprising in view of the strong signal of expansion observed in human mtDNA [1] . Indeed , given the huge state space for haplotype motifs , we would expect a large number of haplotypes at very low frequencies , keeping the saturation curve of haplotypes steadily rising . In contrast , the space of sites is tiny , and , therefore , presumably closer to saturation . The function of the control region is not completely understood , but is thought to be involved in mtDNA genome replication and transcription , and possibly contains the origin of heavy- and light-strand mtDNA replication and several transcription binding sites , with the HVS-I depauperate in regions of this kind [25] . One might expect that the parts of the control region in which these sites are found will be more conserved than others . The information obtained from all unique polymorphic transversions , transitions and indels was used to draw a “bar code” of the sequenced region to show all positions in which a mutation was observed ( Figure 3 ) . A total of 358 ( 65 . 5% ) of the possible 546 sequenced positions showed polymorphism . Some variability in density of polymorphic regions is evident , but no “polymorphism-free” regions can be detected . Note that the map does not distinguish between positions that mutated once or multiple times during the mtDNA evolution of Homo sapiens . It is also important to note that the database does not represent the worldwide variety of mtDNA and , therefore , mutations typical of other populations may not be represented . A few considerations unique to a public project should be discussed . Because the current dataset presented in this manuscript comprises members of the public who have joined Genographic's research effort , the samples herein represent a subset of the total global mtDNA diversity . To properly survey the genetic variation in non-Western Eurasian lineages , the Genographic Consortium is actively consulting and engaging with members of indigenous communities from around the world , and conducting anthropological and genetic analysis on those DNA samples . As such data are published they will also be made available anonymously as part of the reported Genographic reference database . The classification , saturation , and analytical techniques will need to be updated accordingly , as is the case with any expanding database . In addition , this manuscript presents a level of Hg resolution based on the current 22-SNP panel and HVS-I information . A stated goal of this research effort is to continue to refine and increase this resolution , which will be achieved by further genotyping or revised analysis incorporating the expanding dataset . Therefore , at present , the participants and scientific community are presented with a solid , but still rather simple , level of analytic resolution and are encouraged to return periodically to the project's Web site to access up-to-date data and analytical tools . In summary , we report both data and new classification methods developed using by far the largest standardized mtDNA database yet created , and detail the logistic , scientific , and public considerations unique to the Genographic Project . Most importantly , we return to the public a database made possible by their enthusiastic participation in the Genographic Project .
The Genographic Project's Web site allows members of the public to order a buccal swab kit ( containing two buccal swabs ) and undergo genotyping for either mtDNA or MSY analysis . To ensure anonymity , each participation kit is encoded with a randomly generated , nonsequential , Genographic Participant ID number . All samples are genotyped with informed consent according to procedures approved by the Institutional Review Boards of the University of Pennsylvania and the United States Department of Health and Human Services . Once results are obtained , the participants may consent to contribute their genetic data anonymously to the Genographic research database , to be used for anthropological studies and made available to the scientific community . The participants are also asked to provide genealogical information relevant to their deep ancestry . We use the term haplotype to describe HVS-I variation . The reported HVS-I is “extended” and covers 16024–16569 for all samples . Absolute numbers are used to describe nucleotide position ( 1–16569 ) in the mitochondrial genome , and refer to the position of the polymorphism compared with the rCRS [26] . It is common practice to label by letters the nucleotide change only for transversions ( e . g . , 16318t ) and to avoid labeling by letter transitions ( e . g . , 16093 ) , since the changed nucleotide can be inferred from the rCRS [27] . As this study also addresses the general public , who may not be familiar with rCRS nomenclature conventions , we note here that we deviate from the common practice , and to facilitate reading and the use of the released database we label both . Transitions are labeled by capital letters ( 16093C ) , transversions by small letters ( 16318t ) , and heteroplasmies by the letter “N” ( 16189N ) . Sequencing alignment always prefers 3′ gap placement for indels . Deletions are marked by the letter “D” ( 16166D ) and insertions by the point ( . ) sign ( 16188 . 1C ) . We use the term Hg to describe haplotype groups ( “haplogroups” ) [28] that usually coalesce tens of thousands of years ago and are best defined by a combination of coding-region SNPs . The Hgs currently reported by the Genographic Project are listed in Table 2 . We adopt a standard Hg nomenclature scheme [27] . Since we have noted that the asterisk ( * ) suffix used in this scheme leads to some confusion among public participants , we elaborate here on this point by giving an example . A label such as M* means that a sample belongs to Hg M , but not to any of the known subclades within M . It is temporary , and should mean that the Hg is one of many paraphyletic clades falling under the monophyletic Hg M but is not any of the known single-letter ( e . g . , Hg D ) or letter-number ( e . g . , Hg M1 ) coded Hg M sub-branches . It is therefore clear that even if all reported databases abided by this definition and labeled M* by excluding all known sub-branches at time of publication , it would be impossible to compare samples that fell into this cluster in different publications , because new sub-branches are continually defined . The solution suggested for Y chromosomal nomenclature [29] , which clearly specifies which sub-branches were excluded ( e . g . , M* ( xCZ , M1 , M3 , M51 ) ) , might ease database comparisons , especially , when phylogenetic knowledge enlarges and it becomes harder to exclude all known sub-branches of each given Hg in each study . Therefore , we suggest a slight modification to the use of the asterisk suffix . Herein , its use denotes that the sample was excluded from all sub-Hgs reported in this study only ( Table 2 ) , whether defined by a coding-region SNP or an HVS-I defining motif ( Table S2 ) . Therefore , in this study , the label M* means that the sample belongs to Hg M and was excluded only from the sub M branches reported in this study; namely , C and D by coding-region SNPs , and M1 and Z by HVS-I defining motifs . The sample could still belong , for example , to the well-defined M5 or M8 branches that are not part of the Hgs reported in this study . Sequences of an extended HVS-I ( 16024–16569 ) are determined from positions 16024 to 16569 , by use of the ABI Prism Dye Terminator cycle-sequencing protocols developed by Applied Biosystems ( http://www . appliedbiosystems . com ) . Sequencing is performed on a 3730xl DNA Analyzer ( Applied Biosystems ) . Mutations are scored relative to the rCRS [26] . The primary amplification is achieved by primers 15876F and 639R ( Table S1 ) . PCR products are cleaned using magnetic-particle technology ( BioSprint 96; Qiagen , http://www . qiagen . com ) . Following the primary amplifications , all samples are subject to bidirectional sequencing using primers 15946F and 132R ( Table S1 ) . In cases of template polymorphism at the annealing site ( s ) and failed sequencing due to primer/template mismatch , alternative primers are used ( Table S1 ) . High quality is assured by the following procedures: ( 1 ) All sequences are aligned by the software Sequencher ( Gene Codes Corporation ) and observed by an operator . ( 2 ) All positions with Phred score <30 are directly inspected by an operator [30 , 31] . ( 3 ) All positions that differ from the rCRS are recorded electronically . ( 4 ) Forward and backward sequences of all samples are electronically checked for consistency . ( 5 ) All scenarios noted herein are highlighted for review: failed samples , inconsistencies in forward and backward sequencing , successful sequencing in one direction only , sequences that contain indels or heteroplasmy , and sequences that are shorter than the required length . ( 6 ) All highlighted samples are observed again by a second operator . ( 7 ) All sequences containing two or more heteroplasmies are regarded as contaminated and DNA is re-extracted from the second swab of the participant . ( 8 ) The list of HVS-I haplotypes observed among the lab staff is presented as part of Dataset S1 . ( 9 ) All reported variant positions are digitally checked for consistency of the expected order of the mutations ( i . e . , 16093C followed by 16126C and not 16126C followed by 16093C ) . ( 10 ) All reported variants are verified to represent a real polymorphism by direct comparison to the rCRS . ( 11 ) All variants reported for the first time when compared to the entire database are highlighted and re-observed . ( 12 ) All data donated to the scientific world with consent are released . Any comments and remarks raised by external investigators after release will be addressed by re-observing the original sequences for accuracy . Following that , any unresolved result will be further examined by re-genotyping and , if necessary , immediately corrected by publishing an erratum . The biallelic sites are genotyped by means of KASPar assays [32] and are independent of the sequencing , thus playing an additional role in the quality check . Twenty one SNPs and the 9-bp deletion make up the total of 22 biallelic sites . For simplicity , we will refer to all biallelic sites as SNPs . The number of SNPs tested was gradually increased from ten at inception of the project to the 22 currently used . The ten initial SNPs were 3594 , 4580 , 5178 , 7028 , 10400 , 10873 , 11467 , 11719 , 12705 , and 14766 ( numbers refer to the nucleotide position in the mitochondrial genome ) . The panel was augmented to a total of 20 coding-region SNPs by including the following additional ten SNPs: 4248 , 6371 , 8994 , 10034 , 10238 , 10550 , 12612 , 13263 , 13368 , and 13928 . The panel was further augmented by the addition of SNP 2758 , to a total of 21 coding-region SNPs and finally by including the 9-bp deletion at position 8280 to a total of 22 coding-region SNPs ( Figure 4 ) . Two further changes were made: positions 8994 and 13928 used in some early work were respectively replaced with their phylogenetic equivalents 1243 and 3970 . Therefore , the current panel includes the following SNPs , with their respective gene locations shown in brackets [33]: 2758 ( 16S ) , 3594 ( ND1 ) , 4248 ( M ) , 4580 ( ND2 ) , 5178 ( ND2 ) , 6371 ( COI ) , 7028 ( COI ) , 8280 ( 9-bp deletion ) ( NC7 ) , 8994 ( ATPase6 ) , 10034 ( G ) , 10238 ( ND3 ) , 10400 ( R ) , 10550 ( NDRL ) , 10873 ( ND4 ) , 11467 ( ND4 ) , 11719 ( ND4 ) , 12612 ( ND5 ) , 12705 ( ND5 ) , 13263 ( ND5 ) , 13368 ( ND5 ) , 13928 ( ND5 ) , and 14766 ( Cytb ) . The coding-region SNPs were chosen based on the following considerations: ( 1 ) Major branching points in the mtDNA phylogeny obtained using complete mtDNA sequences [21] . ( 2 ) Hgs known to be frequent among the current populations in which the project is advertised [2 , 34 , 35] . For example , the R0 clade within macroHgs R and N is over-represented . ( 3 ) Hgs in which the HVS-I predictive value is known to be unsatisfactory [18] . ( 4 ) SNPs reported in previous publications that have been commonly used to identify a particular Hg [2] . For example , we choose polymorphism 7028 and not 2706 to identify Hg H . ( 5 ) Technical issues concerning the ability to validate any given assay . The SNP genotyping results are obtained digitally and analyzed automatically to suggest the appropriate Hg consistent with the mtDNA phylogenetic tree . Two possible scenarios can prevent the reliable assignment of an Hg by SNPs . First , when SNP genotyping in critical positions for labeling a particular Hg has failed due to technical problems , the genotyping result is rendered “uninformative . ” Note that most of the information might still exist with only the terminal SNP in the mtDNA phylogeny missing . Second , when SNP genotyping is complete but the reported mutations deviate in a particular SNP from the accepted mtDNA phylogeny , the genotyping result is labeled as “inconsistent” and can result from homoplasy , back mutation , a new unknown SNP next to the checked SNP that distorts the reaction , or a genotyping error . Hg labeling is achieved by combining the information obtained from ( 1 ) the coding-region SNPs , and ( 2 ) the HVS-I motifs . A third means of Hg labeling , based on NN methodology , is developed herein . Hg assignment by coding-region SNPs . The standard panel of 22 coding-region SNPs allows a reliable , deep-rooted analysis of the mtDNA phylogeny for each sample as presented in Figure 4 . The SNP panel contains a diagnostic SNP for each of the following major bifurcations in the mtDNA phylogeny: L2′3 ′4 ′5 ′6 ′7 , L3′4 ′7 , M , D , C , N , N1 , I , A , W , X , R , R9 , B , J , T , U , K , R0 , HV , V , and H [1 , 21] . Therefore , a total of 23 Hg clusters can be inferred from the SNP resolution: L0 or L1 , L2 or L5 or L6 , L4 or L7 or L3 ( xM , N ) , M ( xC , D ) , C , D , N ( xN1 , A , W , X , R ) , N1 ( xI ) , I , A , W , X , R ( xU , R0 , J , T , R9 , B ) , U ( xK ) , K , R0 ( xHV ) , HV ( xH , V ) , H , V , J , T , R9 , and B ( Figure 4 ) . To facilitate reading , these 23 Hg clusters are labeled more simply as follows: L0/L1 , L2 , L3* , M* , C , D , N* , N1* , I , A , W , X , R* , U* , K , R0* , HV* , H , V , J , T , R9 , and B . We emphasize that these labels are not equivalent to the final Hg definitions . It is important to note that the use of the coding-region SNPs is very accurate but still prone to errors . For example , under a theoretical scenario in which a sample that belongs to Hg H has a back mutation in position 7028 , the panel will label it as HV . We have no way of estimating the frequency with which such a scenario might occur , as we test only one coding-region SNP per branch , but we expect that this phenomenon is very rare . Final Hg labeling . Final resolution to Hgs and sub-Hgs is achieved by comparing and combining the information obtained from the SNP genotyping with the HVS-I motifs . All HVS-I haplotypes obtained following sequencing are digitally screened for possible Hg and sub-Hg definitions by use of accepted HVS-I diagnostic motifs ( Table S2 ) [2 , 3 , 36] . The list presents only the motifs used herein for prediction purposes and should not be treated as comprehensive for all Hg suggestions that might rise from HVS-I variation or as representing the Hg basal HVS-I motifs . First , a screen in the a priori defined order presented in Table S2 is run and stopped at the first Hg where the sample matches the motif . A second screen for all possible Hgs the sample can fit in is then conducted . It is clear that relying on the HVS-I variation alone to infer Hgs and sub-Hgs such as M1 , Z , U5 , U6 , and HV1 is prone to inaccuracies . In addition , HVS-I haplotypes alone cannot identify Hgs or sub-Hgs that have no defining motifs and ignores the possibility of homoplasy and back mutations . For example , it is clear that some of the mtDNA genomes appearing in our database as U* might actually belong to Hg U4 but , as they did not contain the diagnostic HVS-I position 16356C and in the absence of additional coding-region genotyping , we could not label them as such . Moreover , some of the sub-Hg definitions inferred from the HVS-I , for example within Hgs J and T , will have to be revised in the future as studies using complete mtDNA sequences prove they do not represent monophyletic clades [20 , 37] . Therefore , whenever analysis is done within an Hg , we refer only to one of the 23 Hgs directly inferred from the 22 SNPs genotyping to avoid any HVS-I based Hg labeling misinterpretations . Quality checking of Hg labeling is as follows: ( 1 ) All discrepancies between HVS-I and SNP labeling are observed by an operator . These discrepancies usually derive from well-known cases of homoplasy and are easily resolved by adhering to the SNP genotyping that correctly assigns the sample to a single Hg in the mtDNA phylogeny . In cases of inability to resolve the discrepancies , the genotyping process is repeated . ( 2 ) All samples in which the SNP information is uninformative are observed by an operator . An attempt to label the final Hg is made by direct observation from the partial list of SNPs available and the HVS-I motif . In case of any persisting doubt , the sample is re-genotyped . ( 3 ) All samples in which the SNP information is inconsistent are observed by an operator . The Hg assignment is accomplished after studying the entire string of available mutations and by applying the principle of parsimony . The final Hg can be further supported by the HVS-I information . General indices . Success rates of each of the genotyping processes , Hg frequencies , and distributions including the frequencies of transversions , transitions , heteroplasmies , indels , back mutations , and homoplasy occurring in the HVS-I after taking into account the checked coding-region SNPs are determined by direct counting . We report the heteroplasmic positions in Table S5 but excluded them from all other analyses . NN classification methodology . The common practice of classifying samples into Hgs based on HVS-I information relies on a set of rules that define certain HVS-I backbone haplotypes as characteristic of specific Hgs by using the state of the art knowledge in the literature [2 , 3 , 36] . These characteristic motifs , implemented by us here as one of the Hg labeling techniques , are best if previously proven to be associated with particular coding-region SNPs identifying the suggested Hg , and then used to classify newly obtained HVS-I data into Hgs . The weakness of this approach is its sensitivity to phenomena such as homoplasy or back mutations in the motif's HVS-I positions , which may occur between Hgs or within sub-branches of the same Hg . Since parallel evolution is rampant in HVS-I , this issue casts doubt on the ability of rule-based classification to reach high levels of accuracy in certain cases [1 , 17–19] . Given a large enough “reference” data base of correctly labeled samples ( for example , if all samples are verified by coding-region SNPs ) , we are likely to better assign Hgs for HVS-I haplotypes of new samples if we compare them to all available records in the reference database by identifying their “nearest neighbor , ” i . e . , the most similar sample we have already classified with confidence . This allows us to use all of the HVS-I information in each classification decision , rather than simply counting on the rule-defining sites . Thus , any mutations within Hgs that have appeared in the samples in the reference database will be useful for classification , and recent homoplasy in a single HVS-I locus will have a more minor effect on our classification , because other loci within the HVS-I will still support the correct classification . Given a backbone database D comprising correctly classified HVS-I samples s1 , … , sn and a new HVS-I sample t , we define the pair-wise distance as d ( si , t ) = Σj∈J wj I{tj≠sij} , where J is the set of HVS-I loci ( defined as 16024–16569 in our case ) , and wj is a locus-dependent weight . In a simple application ( unweighted NN ) we would simply take wj = 1 ∀j and get the ( unweighted ) Hamming distance , often used in neighbor-joining algorithms . A more reasonable approach would be to down-weigh the loci with a higher mutation rate ( such as 16311 ) . Denote these mutation rates ( in units of “mutations per year” ) as p1 , …pJ . Then a weighting of wj = log ( 20 , 000 × pj ) can lead to an interpretation of NN Hg classification based on d ( si , t ) as an approximate maximum likelihood estimate of the Hg , using the following logic: Assume that the “average” sample has a NN with coalescent time of about 10 , 000 years . Then the number of mutations separating the sample from its NN in site i has a Poisson ( 20 , 000 × pi ) distribution , under sufficiently simple substitution models . If we assume that 20 , 000 × pi is still very small , as would be the case for practically all sites , then we can approximate the Poisson by a Bernoulli ( 20 000 × pi ) ( which is 1 if the samples differ in site i ) . Now , if we treat the identity of the NN as the parameter to be estimated , we can see that a maximum likelihood estimate would lead us to choose the one minimizing d ( si , t ) = Σj∈J wj I{tj≠sij} . The w-NN analysis requires calculation of site-specific mutation rates , like the ones recently proposed by Bandelt et al . [38] We were limited in our ability to use these published rates , as they only apply to the region 16051–16365 , rather than our HVS-I definition . Thus , in our experiments below we use a set of probabilities we derived using a novel methodology ( Rosset et al . , in preparation ) . We verified that these estimates are consistent with Bandelt et al . [38] for the region in common , and use them here since they are the only complete set we could obtain . An improved set of probability estimates may improve the results further . ? In applying the NN methodology , we are bound to encounter many “ties , ” when there are two equally close NNs in two different Hgs . In our implementation , we assign the new sample to the Hg in which the most similar haplotypes are most prevalent in the reference database . | The Genographic Project was launched in 2005 to address anthropological questions on a global scale using genetics as a tool . Samples are collected in two ways . First , the project comprises a consortium of ten scientific teams from around the world united by a core ethical and scientific framework that is responsible for sample collection and analysis in their respective region . Second , the project promotes public participation in countries around the world and anyone can participate by purchasing a participation kit ( Video S1 ) . The mitochondrial DNA ( mtDNA ) , typed in female participants , is inherited from the mother without recombining , being particularly informative with respect to maternal ancestry . Over the first 18 months of public participation in the project we have built up the largest to date database of mtDNA variants , containing 78 , 590 entries from around the world . Here , we describe the procedures used to generate , manage , and analyze the genetic data , and the first insights from them . We can understand new aspects of the structure of the mtDNA tree and develop much better ways of classifying mtDNA . We therefore now release this dataset and the new methods we have developed , and will continue to update them as more people join the Genographic Project . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"homo",
"(human)",
"genetics",
"and",
"genomics",
"computational",
"biology"
]
| 2007 | The Genographic Project Public Participation Mitochondrial DNA Database |
Behavioural inhibition is a key anxiety-like behaviour in rodents and humans , distinct from avoidance of danger , and reduced by anxiolytic drugs . In some situations , it is not clear how behavioural inhibition minimises harm or maximises benefit for the agent , and can even appear counterproductive . Extant explanations of this phenomenon make use of descriptive models but do not provide a formal assessment of its adaptive value . This hampers a better understanding of the neural computations underlying anxiety behaviour . Here , we analyse a standard rodent anxiety model , the operant conflict test . We harvest Bayesian Decision Theory to show that behavioural inhibition normatively arises as cost-minimising strategy in temporally correlated environments . Importantly , only if behavioural inhibition is aimed at minimising cost , it depends on probability and magnitude of threat . Harnessing a virtual computer game , we test model predictions in four experiments with human participants . Humans exhibit behavioural inhibition with a strong linear dependence on threat probability and magnitude . Strikingly , inhibition occurs before motor execution and depends on the virtual environment , thus likely resulting from a neural optimisation process rather than a pre-programmed mechanism . Individual trait anxiety scores predict behavioural inhibition , underlining the validity of this anxiety model . These findings put anxiety behaviour into the context of cost-minimisation and optimal inference , and may ultimately pave the way towards a mechanistic understanding of the neural computations gone awry in human anxiety disorder .
Rodent models of human anxiety commonly involve a conflict between approach and avoidance [1–4] , as exemplified in the Elevated Plus Maze [5–7] , Open Field test [8] , operant conflict tests [9 , 10] , or novelty-suppressed feeding test [11] . Behavioural inhibition is a core anxiety-like readout in these models , defined for example as a delay to initiate approach [11] , and reduced by anxiolytic drugs [1] . Behavioural inhibition is also observed in non-human [12] and human primates [13] . Hippocampus lesions reduce anxiety-like behaviour in rodents [1 , 14 , 15] and humans [13] alike which suggests a neural implementation that is conserved across species . Extant theories assume that behavioural inhibition arises because of time requirements in the decision-making process: in one model because the animal uses that time to collect information about the situation ( risk assessment ) [1 , 16] , and in another suggestion because the decision whether to approach or to avoid is difficult [7] . Yet , a formal analysis of its adaptive value is lacking , and this impedes understanding the neural implementation of this behaviour . Here , we provide a normative explanation for behavioural inhibition in the framework of Bayesian Decision Theory ( BDT ) [17] by showing that it is the cost-minimising strategy in temporally correlated environments . Experimentally , we then demonstrate behavioural inhibition in humans , with a pattern that cannot be explained by previous accounts but is consistent with model predictions and may suggest a neural implementation based on goal-directed cost minimisation . Crucially , behavioural inhibition as measured in our task is related to individual anxiety scores , and this independently confirms the validity of the experimental anxiety model . Consider a rat in an operant conflict test in which it is trained to obtain a food pellet after a discriminative cue ( Fig 1 ) . On a proportion of trials , it will receive an electric shock together with the food pellet . Access to the food pellet is withdrawn if the animal does not respond within some time interval . In this scenario , the animal must make two decisions: whether or not to approach and collect the reward ( the action ) , and if yes , when to approach ( the approach latency , also termed response vigour [18 , 19] ) . It should take the approach action if the utility of approach is expected to be positive . In this case , the animal should choose its approach latency in order to maximise expected utility . To do so , BDT mandates that it computes the probabilities of gaining the food pellet and getting the electric shock as functions of approach latency , and combines them with loss functions , to maximise its gain . Crucially , the probability functions rely exclusively on the animal’s ( subjective ) prior probabilities because there is no current indicator of the action outcome ( i . e . no likelihood ) . If the animal knew the objective task statistics ( i . e . that probability of reward decreases over time , probability of shock is constant over time ) , its optimal decision would be to approach immediately after a reward appears ( scenario 1 , Fig 1 left ) . However , the animal has to learn these statistics . Before making a first response , it will rely on priors formed in other environments . Food availability is spatio-temporally coupled with predatory threat for many species [20 , 21] . In biological terms , small-scale temporal correlations reflect a situation in which a predator is alerted by the occurrence of a reward that his prey species is interested in , and loses interest after waiting in vain for some time . Prey can exploit this environmental dependency to predict predatory threat . If the animal’s initial prior encodes a small-scale temporal correlation of threat and food reward , we will show that behavioural inhibition arises as cost-minimising strategy . This is illustrated in Fig 1 ( scenario 2 , right ) and Fig 2 . Scenario 1 mandates immediate approach , which is biologically impossible if the food appears at unknown time points . Hence , an animal will show non-zero response latency in both scenarios 1 and 2 , i . e . regardless of its priors . In order to distinguish between these two scenarios , we can analyse the impact of threat magnitude on optimal approach latency . It turns out that the cost-minimising approach latency in scenario 2 but not in scenario 1 depends on threat magnitude . This mathematical insight affords an empirical distinction between scenarios 1 and 2 .
The model is formulated in general terms and specifies the optimal approach latency t 1 * as a local maximiser , i . e . by finding roots of the cost function derivative ( eq 5 ) . The first crucial finding is that non-zero approach latency , i . e . behavioural inhibition , is normative under very general priors about the temporal evolution of threat . Behavioural inhibition does not depend on the precise functional form of the prior and arises for all priors with a half-life of reward that is longer than the half-life of threat . This fact is illustrated in Fig 2 by using the same example prior as in Fig 1 . The value of the optimal approach latency in scenario 2 , t 1 * , can only be predicted when the temporal evolution of PL is known , which is not the case in an operant conflict task . Therefore , if we empirically measure a non-zero approach latency t1 , we would not know whether this is due to biological constraints that delay the optimal response in scenario 1 , or due to an non-zero optimal approach latency in scenario 2 . We therefore analysed the impact of small changes in the parameters on the optimal approach latency in scenario 2 . It turns out that small changes in potential loss , or overall threat probability , must increase the optimal approach latency , i . e . delay the response further ( eq 8 ) . This distinguishes scenario 2 from a model with time-varying biological constraints . For example , a time-dependent motor cost can also lead to a non-zero maximiser t 1 * , but here the optimal approach latency depends only on the gain , not on changes in threat probability or loss magnitude ( S1 Text ) . We experimentally tested predictions from this mathematical model in experiment 1 with n = 20 human participants . Similar to a previous approach in humans [13] , we modeled an operant conflict test as virtual computer game , with objective statistics according to scenario 1 ( Fig 3 ) . The player is tasked to collect tokens under threat of being caught by a “predator”which can catch the player to remove previously collected tokens . In this task , threat probability corresponds to the wake-up rate of the predator , termed threat level , and magnitude of potential loss corresponds to the number of already collected tokens , termed potential loss . The wake-up probability of the predator is constant over time , and the player has no possibility to escape once the predator is active . Tokens disappear according to an exponential distribution . Under the objective task statistics , optimal approach latency is independent from threat probability or magnitude , and should be minimised as much as biologically possible . However , if the human player uses ( subjective ) priors as in scenario 2 , approach latency should increase as threat probability or magnitude of potential loss increases . Importantly , tokens appear sequentially , such that a decision whether and when to collect the next token can be made before it appears . Hence , decision difficulty should not delay responses once the token appears . Fig 4 shows that participants were less likely to approach a token if threat level or potential loss were higher , as would be expected by rational agents . Strikingly , when they made a choice to approach the token , approach latency increased linearly both with increasing threat level ( Linear Mixed Effects Model , F ( 1 , 15485 ) = 21 . 9 , p < 1e-5 ) and with increasing potential loss ( F ( 1 , 15485 ) = 19 . 8 , p < 1e-5 ) . This is the cost-minimising strategy when using priors according to scenario 2 . It cannot be explained under scenario 1 . Because there is no information to be gained from approach delay , this behaviour cannot reflect risk assessment [16] . Also , because a decision whether or not to go can be made before the token appears , behavioural inhibition is not explained by invoking decision difficulty [7] . We were concerned that players might delay their approach to improve performance of going into the correct left/right direction , or to minimise the time they spent outside the safe place . Players made on average 97 . 2% correct left/right responses . Correctness did not depend on threat level or potential loss ( both p > . 25 ) , and also not on the variation in approach latency ( i . e . after subtracting the average approach latency for each combination of threat level/potential loss/player , p > . 10 ) . Hence , increasing approach latency did not improve performance . Further , analysis of the variation in response and return latency ( over and above impacts of threat level and potential loss , i . e . after subtracting the average approach latency for each combination of threat level/potential loss/player ) revealed that an increase in approach latency of 100 ms lead to a decrease in return latency of 0 . 9 ms ( t ( 12645 ) = -3 . 9 , p < . 0005 ) . Although significant , the impact of this relation on action outcomes ( i . e . probability of getting caught ) is negligible . Further , time-dependent motor costs cannot explain the pattern of behavioural inhibition: a model including these costs predicts a dependency of the approach latency on potential loss but not on threat level ( S1 Text ) . Approach latency distributions offered no evidence for a drift-to-threshold decision process [22] as reason for behavioural inhibition ( S1 Text ) . Finally , results were replicated in a similar experiment 2 , which balanced the colour-threat associations across participants . Again , approach latency increased linearly both with increasing threat level ( F ( 1 , 11808 ) = 38 . 9 , p < 1e-10 ) and with increasing potential loss ( F ( 1 , 11808 ) = 7 . 9 , p < . 01 ) . To summarise , human behaviour in this task is predicted under subjective priors encoding a temporal coupling of threat and reward . How is this behaviour instantiated neurally ? One possibility is that humans implement BDT in goal-directed online computations , in order to optimise their outcome on each trial . On the other hand , there may be a hard-wired ( “Pavlovian” ) inhibition mechanism invoked by predator/prey scenarios that reflexively delays actions whenever detecting approach-avoidance conflict , without any online considerations of future outcomes [23] . This mechanism would nevertheless invoke behaviour that is optimal in many natural environments in which reward/threat associations occur [20 , 21] . Pavlovian biases are often thought to depend on primary reinforcers [23] . In our task , all reinforcement is financial , but the graphical setup strongly frames this as “predation” which could invoke an association with a primary reinforcer . Hence , in experiment 3 with n = 22 human participants , we controlled whether behavioural inhibition depends on a predation scenario . The task had precisely the same statistics and required the same key presses as in experiment 1 but with an entirely different graphical setup and explanation such as to avoid any association with biological predation ( Fig 3B ) . Human players were instructed to collect tokens by moving a virtual left/right “lever” back and forth by pressing keys , but that their action might be corrupted by “static noise” , and then they would lose all previously collected tokens . As in experiment 1 , approach latency increased linearly with increasing threat level F ( 1 , 17398 ) = 21 . 9 , p < 1e-9 ) and with increasing potential loss F ( 4 , 17398 ) = 13 . 0 , p < 1e-3 ) . Comparing data from both experiments in one statistical model revealed no difference in the impact of threat level or potential loss ( interaction with experiment factor , both p > . 20 , see S1 Text ) . Strikingly , however , approach latency was considerably longer in experiment 1 ( Experiment 1: 576 ms; experiment 3: 501 ms; main effect of experiment: F ( 1 , 32902 ) = 587 . 9; p < 1e-128 ) . This demonstrates that behavioural inhibition does not depend on the prospect of predation , but that the amount of inhibition depends crucially on the particular environment , consistent with a goal-directed , online implementation of BDT . A possibly hard-wired , Pavlovian , inhibition mechanism might occur at a late motor stage . In a final experiment 4 , we tested whether a possible inhibition mechanism suppresses or delays motor responses . To this end , experiment 1 was repeated , and human players made their responses with a joystick . They would collect the token if the joystick was moved beyond a certain threshold . As in previous experiments , overt approach latency increased with threat level and potential loss ( Fig 4 , S1 Text ) , and the same was found for the latency of motor initiation ( S1 Text ) . If behavioural inhibition suppressed already initiated movements , one would expect more sub-threshold movements as threat level and potential loss increased , and as overt movements decreased , which was not the case . Instead , both overt and sub-threshold movements were inhibited as threat level/potential loss increased ( S1 Text ) . Motor execution after response initiation was not impacted by threat level , and was faster ( not slower ) for higher potential loss ( S1 Text ) . This demonstrates that behavioural inhibition in this task is not due to interference during motor execution and reflects action planning , again consistent with a goal-directed implementation of BDT . Rodent conflict tests are often regarded as anxiety tests by face validity of the observed behaviour , or because of the specific behavioural alterations elicited by anxiolytic drugs [5] . However , it is not entirely clear that this behaviour relates to subjective feelings of anxiety in humans , experimentally often elicited by procedures involving social evaluation [24] which are however unaffected by anxiolytic drugs [25] . Hence , we tested whether subjective feelings of anxiety , recorded before the experiment using a standard questionnaire [26] , predicted behavioural inhibition during our task . Because of the small baseline variation in anxiety , we increased power by combining data from experiments 1 and 2 that used the same graphical setup and response modality . Momentary ( “state” ) anxiety had no influence on approach latencies . Trait anxiety however , as a personality measure , impacted both the effect of threat level ( interaction: F ( 2 , 22928 ) = 10 . 1 , p < 5e-4 ) and potential loss ( F ( 4 , 22928 ) = 3 . 9 , p < . 005 ) on approach latency while leaving overall approach latency unaffected . This finding is consistent with an idea that individuals with higher trait anxiety scores use a different prior threat probability function than individuals with lower trait anxiety . Note that in our model there is no linear relation between the prior and the approach latency; instead the impact of changing threat level/potential loss on approach latency depends on the precise curvature of the prior threat probability at the optimal approach latency . In line with this , non-linear interaction terms of trait anxiety with threat level and approach latency contributed to the effect of trait anxiety . Approach latency patterns across 4 experiments were qualitatively consistent with model predictions . In particular , a monotonic dependency of approach latency on threat level and potential loss is in keeping with eq ( 8 ) . Additionally , we sought to quantitatively compare approach latencies to the model . This addresses whether a prior distribution PL exists that can explain the observed data . The model’s quantitative predictions crucially depend not only on PL but also on the internal representation of the gain probability , PG , and the internal utility of each loss and gain . To avoid overfitting the model to our data , we constrained our comparison by assuming that PG is perfectly learned , that all values are weighted linearly , and that an ideal observer evaluates possible loss on each trial . For each experiment , this ideal observer model took as inputs the average number of collected tokens per threat level , the average ( empirical ) rate of getting caught per threat level , and the average rate of collecting a token when making an approach movement . After calculating the possible loss for each condition , we used eq ( 5 ) together with the observed approach latencies per condition , and the true values for PG , to compute the time derivative of the prior threat distribution . Fig 5 ( lower panels ) shows that these values were well approximated with a linear fit , a non-trivial observation that does not follow from the model or from the way of computing these values . The linear approximation for the derivative of the prior threat distribution was constrained to be negative ( as per assumption 7 ) , integrated , and a constant parameter added to achieve the average catch rate at the average approach latency across all conditions . The resulting prior threat distribution is shown in Fig 5 ( lower panels ) and is qualitatively similar across experiments , with the exception of exp . 3 where it reached a baseline more quickly . Finally , this 3-parameter prior was fed into the model ( eq 5 ) to make quantitative predictions for approach latencies . Non-trivially , a local maximiser was found for all conditions in all experiments , and the predicted approach latencies approximated the monotonic trend observed in the data , while not accounting for experiment-specific kinks in the approach latency curves , as shown in Fig 5 ( upper panels ) .
Behavioural inhibition is a core readout of rodent anxiety tests involving a conflict between approach and avoidance . This phenomenon has been explained with time requirements imposed by the decision process—time to gather further information ( risk assessment [1 , 16] ) , or time to complete a difficult decision [7] . In the current study , we provide an alternative explanation by analysing the adaptive value of this behaviour . We mathematically demonstrate that during operant approach/avoidance conflict , behavioural inhibition is the cost-minimising strategy in environments with small-scale temporal correlations of threat and reward [20 , 21] . This mathematical model makes distinct predictions for an influence of threat level and potential loss on behavioural inhibition , which we experimentally confirm across 4 samples of human participants . The pattern of results cannot be explained by decision difficulty or risk assessment . Approach delay in our experiments depends on environment characteristics , but does not require the prospect of virtual predation , hence is not necessarily linked to primary reinforcers . Finally , this inhibition occurs before motor execution . It is therefore likely to arise from action planning and could possibly be instantiated by a goal-directed decision process as engendered in online cost minimisation according to BDT . Strikingly , human trait anxiety predicts behavioural inhibition in our study , confirming the validity of this operant conflict test as an anxiety model . Quantitative comparison of our data with the model demonstrated that a threat prior can be constructed to explain the observed data . This prior was similar across experiments , and its simple shape is biologically plausible . While the predictions from that prior approximated a monotonic trend in the observed approach latencies , it did not predict experiment-specific deviations from this trend . Note however that our reconstruction of the prior assumed an ideal observer and linear utility functions for specifying the expected loss , and it was solely based on financial loss and did not include an any additional loss ( a premium ) for the fact that one gets caught . Estimating such valuation parameters is likely to improve the fit of the model and could be achieved in independent tasks . We note that the model fit does not prove that participants actually used this prior . Also , we did not aim to make participant-specific predictions . The range of approach latencies from which data points for reconstructing the prior can be sampled is much smaller than the time window searched for a maximiser . The present approach should therefore only be regarded as providing face plausibility to the model , not as a strategy to reconstruct the true priors that participants used . Future work will aim at estimating this prior from independent tasks and making specific quantitative predictions for optimal approach latency . Capitalising on different reward distributions , it is then possible to test the current model in which behavioural inhibition crucially depends on the interplay of reward and threat priors . Across various tasks involving approach/avoidance conflict , anxiety-like behaviour intricately relies on the ventral hippocampus in rodents and humans , and is reduced by hippocampus lesions and anxiolytic drugs alike [1 , 13–15] . Hippocampal oscillations in the theta range are a hallmark of rodent approach/avoidance conflict and also reduced by anxiolytics in frequency and amplitude [1 , 2 , 27] . Despite a wealth of electrophysiological and neuropharmacological knowledge , however , the function of the hippocampus in these tasks is hardly understood , and this is in stark contrast to theoretical models of hippocampal theta oscillations in memory and spatial navigation [28–31] . This lacuna might be partly due to the fact that behavioural objectives in many ethological approach/avoidance conflict tasks are complex and opaque , rendering an analysis of neural computations a difficult endeavour . In the current work , we took the approach of isolating individual actions and their outcomes from one another , and avoiding incentives relying exclusively on unknown internal processes such as the innate propensity to explore open spaces , harnessed in many rodent anxiety models . This strategy removed from our analysis all elements that could transform action planning into a multi-step problem [18 , 19] , and hence simplified it to an extent where an experimental confirmation was tractable . The series of experiments conducted here allows us to speculate about neural implementation where the empirical evidence is consistent with goal-directed planning although we cannot rule out a possible influence of pre-programmed ( “Pavlovian” ) biases that have evolved because they are adaptive in many natural environments . The current approach using isolated actions affords experimental analysis of prior belief distributions on threat-reward correlation and their neural implementation . It may thus enable a more detailed understanding of the neural circuits mediating anxiety-like behaviour . It can easily be complemented by an analysis of this behaviour in continuous time [18 , 19] , and thus , in more realistic foraging scenarios . Finally , the theoretical model is applicable to rodent behaviour and can be tested by finessing operant conflict tests in animals [32] . To summarise , we find that behavioural inhibition is a cost-minimising behaviour in many natural environments , and possibly instantiated neurally by online cost minimisation . Our finding puts anxiety-like behaviour into the context of optimal inference [17 , 33 , 34] , action planning [18 , 19 , 35] , and biological cost minimisation [36 , 37] . Rather than being a somewhat irrational state of mind , anxiety is thereby rephrased as optimal behaviour under biologically plausible priors , which makes it accessible to the toolkit of computational analysis [38 , 39] . It will be interesting to investigate whether individual differences in anxiety stem from variation in prior assumptions about environmental statistics—even implausible priors— , suboptimal use of such priors , or variation in updating such priors with experience . Ultimately , this research may pave the way towards a more mechanistic understanding of anxiety disorders .
All variables are explained in Table 1 . These assumptions bear on the structure of the model . The following assumptions change the behaviour of the model but do not constrain its structure . | Behavioural inhibition is observed in situations of anxiety , both in animals and humans . In some situations , it is not clear how it minimises harm or maximises benefit for the agent , and can even appear counterproductive . This prevents an understanding of the underlying neural computations . Here , we furnish the first formal assessment of its adaptive value in a controlled anxiety model , and confirm predictions in four experiments with humans . Results may suggest a neural implementation that relies on online cost minimisation . This finding could afford a better understanding of human anxiety disorder and the underlying neural computations . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Models"
]
| []
| 2015 | Anxiety-Like Behavioural Inhibition Is Normative under Environmental Threat-Reward Correlations |
In order to metastasize , cancer cells need to acquire a motile phenotype . Previously , development of this phenotype was thought to rely on the acquisition of selected , random mutations and thus would occur late in cancer progression . However , recent studies show that cancer cells disseminate early , implying the existence of a different , faster route to the metastatic motile phenotype . Using a spontaneous murine model of melanoma , we show that a subset of bone marrow-derived immune cells ( myeloid-derived suppressor cells or MDSC ) preferentially infiltrates the primary tumor and actively promotes cancer cell dissemination by inducing epithelial-mesenchymal transition ( EMT ) . CXCL5 is the main chemokine attracting MDSC to the primary tumor . In vitro assay using purified MDSC showed that TGF-β , EGF , and HGF signaling pathways are all used by MDSC to induce EMT in cancer cells . These findings explain how cancer cells acquire a motile phenotype so early and provide a mechanistic explanation for the long recognized link between inflammation and cancer progression .
Tumor metastasis is the primary cause of death by cancer [1] . Metastasis is a multistep process in which cancer cells derived from the primary tumor migrate to regional or distant sites where they re-initiate tumor development [2] . Until recently , the first step of metastasis , ( i . e . tumor cell dissemination ) was thought to be a late event in cancer progression [3] . This time-lag was presumably needed to allow selected cancer cells to accumulate the additional mutations required for motility . However , recent work , including that of our laboratory , has challenged this paradigm . In fact , cancer cells disseminate even before diagnosis of the primary tumor [4]–[6] , and so a different , faster mechanism must be driving the development of the motile phenotype . Tumors do not consist of a homogeneous population of cells; rather they are a composite of cancer cells , mesenchymal and endothelial cells , and immune cell populations . Among these immune cells are the myeloid cells , which are generating increasing interest as having a dynamic influence on tumor growth . The link between cancer progression and infiltration with myeloid cells was recognized by R . Virchow in the late 19th century [7]–[11] . Infiltration with myeloid cells is usually associated with less favorable clinical outcomes . During the past decade several distinct subsets of tumor-infiltrating myeloid cells have been described [12] , among which CD11b+Gr1+ MDSC have drawn attention for having a role in cancer progression [13]–[16] . MDSC , which can be further divided into monocytic ( Mo- ) MDSC and granulocytic ( PMN- ) MDSC , accumulate in most malignant murine and human tumors [17]–[19] . These cells have been shown to favor cancer progression by dampening anti-tumor immune responses , promoting angiogenesis , and creating a pre-metastatic environment [20]–[22] . RETAAD mice are transgenic for the activated RET oncogene , which is specifically expressed in skin and eye melanocytes . RETAAD mice spontaneously develop uveal melanomas that are clinically detectable ( exophthalmos ) by 4 to 8 wk of age . In fact , microscopic eye tumors can be detected as early as 10 d after birth and cancer cells disseminate from the primary eye tumor throughout the body by 3 wk of age [5] , [23] . Disseminated cancer cells can be monitored in the eye draining lymph node and in the lung by measuring ectopic expression of Dct ( a specific marker of melanocytes ) or RET . The disseminated cancer cells remain dormant for months before developing into cutaneous and then visceral metastases . Previous work showed that most tumors developing in a given mouse share a common clonal origin [5] . The stepwise evolution of melanoma in RETAAD mice mimics the natural history of disease progression in cancer patients [24] , underlining the suitability of this model for dissecting the role of immune cells in the process of metastasis . Although melanocytes are not of epithelial origin , metastatic melanoma cells undergo morphological changes resulting in decreased intercellular adhesion and increased cell motility , a process that morphologically and mechanistically resembles epithelial-mesenchymal transition ( EMT ) [25] . EMT is a trans-differentiation process occurring during embryonic development and in carcinoma progression [26] . In the present study we used RETAAD mice to examine the presence and role of MDSC in melanoma metastasis . Interestingly , we found that PMN-MDSC preferentially accumulate in the primary tumor where they induce cancer cell EMT . PMN-MDSC favor the rapid acquisition of a motile phenotype by cancer cells resulting in multinodular growth of the primary tumor , dissemination throughout the body and distant metastasis .
RETAAD mice develop uveal melanoma that metastasizes to the skin and visceral organs . Tumor development at distant sites usually takes more than 6 mo . As previously described in other tumor models [18] , [19] , we observed an increase in CD11b+Gr1+ cells in both the spleen and blood during tumor progression ( Figure S1A and B , respectively ) . This increase begins as the primary tumor starts to cause the eye to bulge ( exophthalmos ) and peaks once distant metastases become palpable . As shown in Figure S1C , purified CD11b+Gr1+ cells strongly inhibited antigen-specific T cell proliferation in vitro and therefore represent bona fide MDSC . CD11b+Gr1+ MDSC are a heterogeneous population of myeloid cells . To further characterize MDSC in the RETAAD model , immune infiltrates of primary tumors and cutaneous metastases were analyzed . CD45+ immune cells represented on average 2 . 7% ( 95 CI = 1 . 9%–3 . 5% ) of the total cells within a tumor , in good agreement with published data on spontaneous melanoma models [27] and uveal melanoma patients [28] . Among immune cells infiltrating cutaneous tumors ( Figure 1A ) , lymphoid cells represented 28 . 6% ( 95 CI = 20 . 1%–37 . 1% ) . The tumor-infiltrating myeloid cell population comprised CD103+CD11chi dendritic cells ( 3 . 6% , 95 CI = 2 . 5%–4 . 7% ) , CD11b+CD11chi dendritic cells ( 5 . 8% , 95 CI = 4 . 2%–7 . 4% ) , B220+CD11c+ plasmacytoid dendritic cells ( 7 . 6% , 95 CI = 3 . 3%–11 . 0% ) , CD11b+Gr1−F4/80+ macrophages ( 51 . 4% , 95 CI = 42%–61% ) and CD11b+Gr1+ MDSC ( 6 . 7% , 95 CI = 3 . 6%–9 . 8% ) . MDSC could be further sub-divided into CD11b+Gr1hi F4/80− PMN-MDSC and CD11b+Gr1intF4/80lo Mo-MDSC ( Figure 1 and Figure S1D ) . Interestingly , while most subsets of immune cells were equally represented in primary tumors and cutaneous metastases ( Figure 1A ) , PMN-MDSC were 5 times more abundant in primary tumors ( Figure 1B and 1C , 24 . 5%±1 . 5% of immune cells versus 3 . 9%±0 . 2%; two-tailed t test p< 0 . 0001 ) . In order to understand the significance of this finding , we decided to examine the factors controlling the preferential accumulation of PMN-MDSC in the primary tumor and to investigate the role of these cells . To identify potential mediators of the preferential accumulation of PMN-MDSC in primary tumors , the expression of 148 inflammatory genes including chemokines , inflammatory cytokines , and their corresponding receptors was analyzed using low density qRT-PCR arrays . When primary tumors and cutaneous metastases from the same mice were compared , 21 genes were identified as differentially expressed ( fold change >2; p<0 . 05 ) , of which 20 were more highly expressed in the primary tumor than in the corresponding skin tumor ( Figure 1D and Table S1 ) . The most differentially expressed genes ( fold change >4 and p<0 . 05 ) were members of the IL-1 families and , interestingly , four chemokines: CCL19 , CXCL1 , CXCL2 , and CXCL5 . For example , the average expression of CXCL5 was 56-fold higher in primary tumors than in cutaneous metastases ( p = 8 . 38×10−5 ) . Transcriptome analysis of purified tumor-infiltrating immune cells indicated that only PMN-MDSC express Il8rb ( also known as CXCR2 ) , the receptor for CXCL1 , CXCL2 , and CXCL5 ( Figure S2 ) . Furthermore , PMN-MDSC express CXCL1 and CXCL2 while tumors express CXCL5 . We therefore tested the capacity of these three chemokines to attract immune cells from tumor-bearing RETAAD mice in vitro . As shown in Figure 1F–H , CXCL1 , CXCL2 , and CXCL5 are indeed potent attractants for PMN-MDSC , but not monocytes or lymphocytes . Moreover , RETAAD tumor cells attracted PMN-MDSC in vitro and this attraction was significantly reduced when PMN-MDSC were treated with either SB265610 or SB225002 , two specific inhibitors of CXCR2 ( Figure 1I ) . To establish the importance of CXCR2 ligands in the recruitment of PMN-MDSC to the primary tumors , bone marrow cells from Rosa mT/mG reporter mice expressing tdTomato ( used as fluorescently tagged wild-type [wt] cells ) and CXCR2 knockout mice expressing GFP ( IL8rb-KO crossed with mice expressing EGFP under the lysozyme M promoter [Il8rb-KO] ) were mixed at a 1∶1 ratio and adoptively transferred into tumor-bearing RETAAD mice . After 18 h , primary tumors were analyzed by flow cytometry for Gr1hitdTomato+ and Gr1hiGFP+ tumor-infiltrating PMN-MDSC . Wt PMN-MDSC were 7-fold more abundant than PMN-MDSC from Il8rb-KO mice ( Figure 1J ) . This preferential migration was specific for the primary tumor as equal proportions of wt and Il8rb-KO PMN-MDSC migrated to the spleen . This shows that expression of CXCR2 is necessary for PMN-MDSC migration into the primary tumor in vivo . Of the three CXCR2 ligands , CXCL5 is the most highly expressed in primary tumors ( Figure 1E ) and is therefore likely to be the major mediator of the preferential recruitment of PMN-MDSC to the primary tumor . In summary , primary tumors express unique chemokines that specifically attract PMN-MDSC . To gain insight into the role of PMN-MDSC in primary tumors , these cells were depleted using anti-Ly6G antibody NIMP-R14 [29] . The first injection of antibodies was given as soon as the primary tumor became visible ( 5 wk of age ) , then repeated twice a week , until the age of 20 wk when tumors were analyzed by flow cytometry and immunohistochemistry ( Figure S3A , scheme B ) . Antibody treatment ablated PMN-MDSC ( CD11b+Gr1hi ) in the primary tumor and spleen ( determined at the time of sacrifice ) and in the blood for the duration of the experiment ( Figure S3B ) . Importantly , the monocytic subset of MDSC ( Mo-MDSC , CD11b+Gr1loF4/80lo ) was not affected by the depletion , whereas there was a slight increase in macrophage number ( CD11b+Gr1−F4/80+ ) in the primary tumors ( Figure S3B ) . Interestingly , depletion of PMN-MDSC resulted in reduced growth of the primary tumors ( Figure 2A ) but not cutaneous tumors ( Figure S4 ) . At 20 wk of age , the mean diameter of primary tumors was 4 . 7 times smaller in depleted mice compared to mice treated with an isotype control antibody ( mean diameter 1 . 9±0 . 5 mm versus 0 . 4±0 . 2 mm; two-tailed t test p = 0 . 027 ) . Enhanced angiogenesis was unlikely to explain the effect of PMN-MDSC on tumor growth since no significant difference in the density of blood vessels or CD45−CD31+ endothelial cells existed between PMN-MDSC-depleted and control tumors ( Figure S5A and B ) . In fact , there was a trend towards an increased density of CD45−CD31+ cells in the PMN-MDSC depleted mice . Taken together this shows that PMN-MDSC favor the growth of the primary tumor . To elucidate the mechanism by which PMN-MDSC favor growth of the primary tumor , we focused on the early steps of tumorigenesis . The depletion experiment was repeated starting at 1 wk of age , before any eye abnormality was detectable , and the primary tumors were analyzed at 7 wk of age ( Figure S3A , scheme A ) . At this age macroscopic primary tumors are rarely detectable , and when microscopic tumors were analyzed by immunohistochemistry , no difference was observed in the size of the lesion developing in treated and control mice ( Figure S6 ) . However immunohistochemistry revealed that depletion of PMN-MDSC results in a reduced frequency of proliferating cancer cells , identified by double labeling with S100B- and Ki67-specific antibodies ( Figure 2C and D ) . As shown in Figure 2B , at 7 wk of age , primary tumors exhibited a mitotic index of 3 . 5%±0 . 6% that was reduced to 1 . 7%±0 . 1% ( p = 0 . 01 ) in PMN-MDSC depleted mice . We calculated that such a 2-fold difference in the mitotic indices would translate into a reduction in tumor diameter of just under 2-fold at 7 wk and 4-fold at 20 wk , showing good agreement with the observed 4 . 7-fold ( Figure 2A and Text S1 ) . Taken together , MDSC depletion experiments showed that these cells favor cancer cell proliferation ( detected at 7 wk ) leading to the development of bigger primary tumors ( at 20 wk ) . To determine whether the effect of PMN-MDSC on cancer cell proliferation was direct or required additional stromal components , in vitro co-culture experiments were performed . Melan-ret cells , a cell line derived from a RET tumor , were co-cultured for 48 h with PMN-MDSC purified from tumor-bearing RETAAD mice . Cancer cell proliferation was measured in vitro by [3H]-thymidine incorporation . Addition of PMN-MDSC resulted in increased proliferation of the cancer cells in a dose-dependent manner ( Figure 3A ) . This effect was not due to proliferation of PMN-MDSC since even irradiated PMN-MDSC induced cancer cell proliferation ( Figure 3B , p = 0 . 035 ) . Importantly this was specific to PMN-MDSC , as co-culture with F4/80+ cells purified from tumor-bearing RETAAD mice did not enhance cancer cell proliferation ( Figure 3B ) . Furthermore , it did not require direct contact between the PMN-MDSC and the cancer cells as increased proliferation was also observed when PMN-MDSC were separated from the cancer cells by a porous membrane ( Figure 3C , p = 0 . 025 ) . Taken together , these results show that PMN-MDSC secrete soluble mediators that directly promote cancer cell proliferation . In the RETAAD model , primary tumors of the eye display a multinodular structure . Given that in this model , tumors from the same mouse share a common clonal origin , this multinodular structure was suggestive of early acquisition of a motile phenotype [5] , [30]–[32] . We wondered whether PMN-MDSC played any role in the migration of cancer cells to the tumor periphery . Primary tumor morphology was analyzed by histology in 7-wk-old mice that had been depleted of PMN-MDSC from 1 wk of age onward . As shown in Figure 4 , primary tumors from depleted mice ( n = 11 ) had smaller numbers of nodules compared to control mice ( n = 11 ) . The average number of ocular nodules was 11±4 in the depleted mice and 19±5 in the controls ( two-tailed Wilcoxon p = 0 . 04 ) . This result shows that PMN-MDSC play a role in development of the multinodular structure of primary eye tumors by favoring the migration of melanoma cells to the tumor periphery . Given that PMN-MDSC favor the migration of melanoma cells to the tumor periphery , we wondered whether they also promoted cancer cell dissemination to more distant sites , such as the eye-draining mandibular lymph nodes or the lungs . Daupachrome tautomerase ( Dct ) is specifically expressed by melanocytic cells and ectopic expression of Dct represents a sensitive surrogate marker of cancer cell dissemination [5] . We therefore compared Dct expression in the eye-draining mandibular lymph nodes and lungs of 7-wk-old , PMN-MDSC-depleted and control mice . As shown in Figure 5A , there was an 8 . 4-fold decrease ( p = 0 . 03 ) in Dct expression in the mandibular lymph node draining the primary tumor ( left panel ) and a 2 . 6-fold decrease ( p = 0 . 02 ) in Dct expression in the lungs ( central panel ) of depleted mice compared to the isotype control treated animals . A similar difference ( 1 . 8-fold; p = 0 . 01 ) was also observed in Mitf ( a melanocyte-specific transcription factor ) expression in the mandibular lymph node ( Figure 5A , right panel ) . In addition , immunohistochemical analysis of lung sections showed that , in 7-wk-old mice , tumor cells present in lungs were mostly individual cells , implying that the increased Dct expression was not due to larger micro-metastases ( Figure 5E ) . This finding rather shows that , in addition to promoting peri-tumoral dissemination , PMN-MDSC favor metastasis to regional and distant sites . Cancer cell dissemination to the skin of RETAAD mice cannot be easily measured because skin contains melanocytes that express Dct and Mitf under physiological conditions . However , we were able to quantify the number of macroscopic cutaneous tumors in 7-wk-old mice depleted of PMN-MDSC from 1 wk of age and control mice . As shown in Figure 5B , there was a 40% reduction ( p = 0 . 01 , Wilcoxon matched-pairs test; two-tailed ) in the number of cutaneous tumors in mice depleted of PMN-MDSC between 1 and 7 wk of age . The distribution of tumor sizes was similar between treated and control animals ( Figure 5C ) , consistent with the low percentage of PMN-MDSC in cutaneous tumors . These data show that PMN-MDSC infiltrating the primary tumor promote cancer cell dissemination to cutaneous sites . This increased influx of disseminated cancer cells results in an increased number of metastases without affecting the average size of tumor . By comparing the pattern of mutations found in primary tumors and metastases , we previously showed that metastases derive from cancer cells that disseminate early [5] . We therefore predicted that if PMN-MDSC play a critical role in cancer cell dissemination , depleting these cells after cancer cells have already colonized the skin should have little effect on the onset of cutaneous metastases . Indeed , delaying the onset of PMN-MDSC depletion until 5 wk of age ( Figure S3A , Scheme B ) restored the incidence of cutaneous metastases to that of control mice ( Figure 5D ) . These data indicate that most cutaneous tumors derive from cancer cells that disseminate before 5 wk of age and that PMN-MDSC act at an early stage of melanoma development . To further investigate the mechanisms by which PMN-MDSC induce cancer cell dissemination and metastasis , we analyzed the morphology and motility of NBT-II bladder carcinoma cells co-cultured with PMN-MDSC purified from tumor-bearing RETAAD mice . NBT-II cells were chosen as reporter cells because they are known to undergo EMT in response to several different stimuli [33]–[35] . Following co-culture with PMN-MDSC for 24 h , NBT-II cells acquired a mesenchymal morphology ( Figure 6A ) . Actin microfilaments redistributed from the membrane cortex to the basal surface , indicating loss of cell-cell contact . Moreover , NBT-II cells exhibited numerous thin filopodia and cell surface expression of E-Cadherin , a classical marker of epithelial cells , was down-regulated ( Figure 6A ) . Similar changes were observed when NBT-II cells were treated with Hepatocyte Growth Factor ( HGF ) , a known inducer of EMT [36] . To assess motility , NBT-II cells were tracked for 6 h using time-lapse video microscopy ( Videos S1–S3 ) . As seen in Figure 6B–D , NBT-II cells co-cultured with PMN-MDSC travelled significantly further ( 174±9 µm versus 95±5 µm , p = 6 . 7×10−12 ) , with an increased average velocity ( 53 . 0±0 . 9 µm/h versus 32 . 1±0 . 4 µm/h , p = 2 . 8×10−106 ) resulting in a larger total displacement ( 80 . 0 ± 5 . 5 µm versus 29 . 3 ±3 . 9 µm , p = 3 . 1×10−11 ) . To assess the generality of this observation , PMN-MDSC were co-cultured with dissociated tumor cells from RETAAD mice . After 24 h , down-regulation of cell surface E-Cadherin was observed in tumor cells derived from both primary tumors ( 21% reduction; p = 0 . 023 ) and cutaneous metastases ( 22% reduction; p = 0 . 005; Figure 6E–F ) . Down-modulation of E-Cadherin gene expression was also observed in the human melanoma cell line 888mel after co-culture with PMN-MDSC purified from RETAAD mice ( Figure 6G; 42% reduction: p = 0 . 025 ) . Taken together , these data demonstrate that upon co-culture with PMN-MDSC , cancer cells undergo morphological , behavioral , and phenotypic changes typical of EMT . These in vitro findings led us to predict that in vivo depletion of PMN-MDSC would affect the frequency of cancer cells undergoing EMT in the primary tumor . S100A4 ( also known as Fsp1 ) is a small Ca2+ binding protein involved in EMT and cell motility [37]–[39] . We measured the number of S100A4+ melanoma cells in primary tumors . Interestingly , S100A4+ cells were found specifically in the periphery of tumor nodules ( Figure 6H ) . Importantly primary tumors from mice depleted of PMN-MDSC between 1 and 7 wk of age possessed a 3 . 6-fold lower density of S100A4+ cells ( Figure 6I; p = 3 . 42×10−8 in two-tailed t test ) . Similarly , a 3 . 3-fold decrease was seen in vimentin expression ( Figure 6J; p = 0 . 035 in paired two-tailed Wilcoxon-test ) . Closer characterization of S100A4+ cells in primary tumors showed that there were indeed delaminating melanoma cells expressing HMB45 and MART-1 antigens and exhibiting a fusiform morphology ( white arrows; Figure 6K ) . The number of S100A4+ cells expressing melanoma differentiation antigens was reduced in tumors depleted from PMN-MDSC ( Figure 6L ) . This observation indicates that PMN-MDSC play a crucial role in EMT induction in melanoma cells in vivo . To identify the molecular pathways involved in PMN-MDSC-induced EMT , the various cell subsets present in RETAAD tumors were sorted by flow cytometry before performing transcriptome analysis and searching for known inducers of EMT . As shown in Figure 7A , PMN-MDSC were found to express Hepatocyte Growth Factor ( HGF ) and Transforming Growth Factor-β1 ( TGF-β1 ) , while cancer cells were the main source of Epidermal Growth Factor ( EGF ) . To determine whether these factors played any role in PMN-MDSC-induced EMT , NBT-II cells were pre-treated with various inhibitors before co-culture with PMN-MDSC . After 24 h , NBT-II cells were stained with desmoplakin to visualize the desmosomes that are typically found at the junctions between epithelial cells ( Figure 7B ) . Addition of PMN-MDSC to untreated NBT-II cells resulted in the complete dissociation of the desmosomal complexes at the cell surface and their translocation into the cytoplasm . Cell scattering was strongly reduced by pre-treatment of NBT-II cells with SB525334 , a specific inhibitor of the receptor for TGF receptor 1 ( TGF-βR1 ) . Nevertheless , most desmoplakin junctions were lost . In contrast , most cells treated with JNJ38877605 , a specific inhibitor of the receptor for HGF ( HGFR ) , retain desmoplakin at cell-cell junctions . Further inhibition of EMT was observed when NBT-II cells were pretreated with SB525334 together with JNJ38877605 and/or PD153035 , a specific inhibitor of the receptor for EGF ( EGFR ) . Pre-treatment of NBT-II cells with all three inhibitors completely reversed the effect of PMN-MDSC . Similar results were obtained when a different set of inhibitors was used ( Figure S7 ) . Taken together , these results demonstrate that TGF-β1 , EGF , and HGF play a crucial role in EMT induction by PMN-MDSC . Recent data suggest that primary tumor cells acquire a metastatic phenotype during the early stages of cancer progression [5] , [40] . This hypothesis led to a conundrum , as the previously accepted theory of random accumulation of mutations leading to the metastatic phenotype was unable to explain such rapid cancer cell dissemination . Here we show that a particular subset of immune cells preferentially infiltrates primary tumors , induces EMT in primary tumor cells , and promotes cancer cell dissemination . Therefore , in our mouse model of melanoma , acquisition of a motile phenotype by cancer cells is instructed by tumor-infiltrating MDSC and occurs early in the development of the primary tumor . MDSC accumulate in the tumors , spleen , and blood of tumor-bearing mice [19] , [41] . In the RETAAD model , we uncovered a preferential recruitment of PMN-MDSC to primary tumors compared to cutaneous metastases , which correlated with a more inflammatory microenvironment . Importantly the chemokines CCL19 , CXCL1 , CXCL2 , and CXCL5 were significantly more highly expressed in primary tumors . CXCL1 , CXCL2 , and CXCL5 are known chemoattractants for neutrophils [42] , [43] and , as shown here for the first time to our knowledge , potent attractants of PMN-MDSC as well . CXCR2 , the receptor for CXCL2 , CXCL5 , and CXCL1 , is specifically expressed on PMN-MDSC purified from RETAAD tumors ( Figure S2 ) . Pharmaceutical inhibition of CXCR2 reduced PMN-MDSC attraction by melanoma cells in vitro and genetic deletion of CXCR2 impaired their recruitment to the primary tumor in vivo . Since CXCL5 is the most highly expressed of the three CXCR2 ligands , it is the prime candidate for mediating the preferential attraction of PMN-MDSC to the tumor . In addition , since CXCL1 and CXCL2 are expressed by PMN-MDSC purified from RETAAD tumors , we propose that these two chemokines provide a positive feedback loop that further amplifies the accumulation of PMN-MDSC in primary tumors . Of note , CXCL1 and IL-8 , the human ortholog of CXCL5 , are expressed by human melanoma cells [44] , [45] . Depletion of PMN-MDSC from 1 wk of age resulted in a decreased density of proliferating cells in primary tumors . Although no significant effect on tumor size was detected at 7 wk , when depletion was performed from 5 wk until 20 wk of age , primary tumors were much smaller . Importantly the size of cutaneous metastases , which contain few PMN-MDSC , was not affected by the depletion . Three sets of observations suggest that this effect is due to a direct action of PMN-MDSC on cancer cells . Firstly , highly purified PMN-MDSC stimulated the proliferation of Melan-Ret cells in vitro . This effect was mediated by soluble factors secreted by PMN-MDSC , since it did not require direct contact with the melanoma cells . Secondly , PMN-MDSC were the only tumor-infiltrating cells whose number was reduced by the antibody treatment; monocytes ( CD11b+Gr1loF4/80lo ) were unaffected and macrophage ( CD11b+Gr1−F4/80+ ) density was , if anything , slightly increased . Finally the number of endothelial cells ( CD45−CD31+ ) was slightly increased in PMN-MDSC depleted tumors , showing that PMN-MDSC stimulated tumor growth independently of angiogenesis . EMT has been suggested to be an essential step in cancer cell dissemination and metastasis [36] . Artificial induction of EMT in cancer cell lines promotes metastatic potential [46] . The aggressive behavior of melanoma has been attributed to its propensity to undergo EMT [47] . In the present study we show that purified PMN-MDSC induce typical features of EMT in vitro in melanoma and bladder human cell lines and in cells freshly isolated from primary and metastatic RETAAD tumors . In vivo , depletion of PMN-MDSC reduced the density of vimentin and S100A4 expressing cells in the primary tumor . Vimentin is a classical mesenchymal marker , while S100A4 , a target of the EMT inducer Snail [48] , is known to down-regulate E-Cadherin in mammary epithelial cell lines [49] . Collectively these observations indicate that PMN-MDSC play a crucial role in cancer cell EMT in vivo . Interestingly , expression of S100A4 correlates with shorter patient survival in several cancers , including melanoma [50] . PMN-MDSC-induced EMT results in multinodular development of the primary tumor , since mice depleted of PMN-MDSC had primary tumors composed of fewer nodules . This observation is in agreement with previous findings that multinodular growth requires cell motility [30] , [31] . At 7 wk , decreased nodularity in PMN-MDSC-depleted tumors was observed in the absence of reduction in the whole tumor size , consistent with spatial redistribution of cancer cells , rather than inhibition of their development . In contrast , cutaneous tumors typically do not display such a multinodular structure , which is consistent with their low content of PMN-MDSC . In humans , many primary tumors exhibit higher fractal dimensions than normal tissue , as a result of their multinodular morphology . This morphology is thought to facilitate exponential growth of small avascular tumors , since in silico modeling indicates that the optimal tumor morphology under strong nutrient limitation is fractal [31] , [51] . Therefore , induction of EMT in cancer cells is likely to be selected because it favors the growth of the primary tumor . PMN-MDSC-induced EMT was also associated with cancer cell dissemination and metastatic outgrowth . Cancer cell dissemination was assessed in the regional lymph nodes and in the lungs by measuring Dct and Mitf expression . As shown previously , Dct expression correlates with the presence of S100B+ melanoma cells detected by IHC [5] . PMN-MDSC depletion from 1 wk of age resulted in reduced dissemination by as early as 7 wk . Reduced primary tumor burden could not explain the decreased cancer cell dissemination , since decreased expression of melanoma markers in the mandibular LN and the lungs preceded the effect on primary tumor size: no change in primary tumor size could be detected at 7 wk . Decreased Dct and Mitf expression after PMN-MDSC depletion could result from decreased influx of and/or decreased proliferation of cancer cells in the lungs . The latter might be expected because PMN-MDSC favor cancer cell proliferation and suppress CD8+ T cell activity . We cannot exclude any of these explanations . In fact RETAAD mice possess increased numbers of PMN-MDSC in their lungs when compared with non-transgenic littermates ( unpublished data ) . Furthermore CD8+ T cells control the proliferation of disseminated cancer cells in the lungs [5] . However , in our experiments , Dct expression was measured at an early time point ( 7 wk of age ) and lung micrometastases have a low mitotic index in untreated animals ( 1 . 9% + 0 . 5%; [5] ) . We calculated that even an 80% inhibition of cancer cell proliferation in the lungs could not account for the observed decrease in Dct expression ( Text S1 ) . Moreover , immunohistochemistry analysis of lungs from control and depleted mice at 7 wk revealed mostly individual cancer cells ( CD45-S100B+ ) ( Figure 5E ) . We therefore favor the interpretation that the decrease observed in Dct expression is mainly due to reduced colonization of the lungs by cancer cells rather than reduced proliferation . Analysis of cutaneous metastases provided further support for decreased cancer cell dissemination in mice depleted of PMN-MDSC . Since PMN-MDSC poorly infiltrate cutaneous tumors , depletion was less likely to affect cancer cell proliferation in this site . If PMN-MDSC were acting by promoting metastatic growth rather than dissemination , we would expect to see smaller metastases in depleted mice . However , depletion of PMN-MDSC from 1 wk of age reduces the number of cutaneous tumors ( Figure 5B ) , but not their size ( Figure 5C ) . Interestingly , when PMN-MDSC depletion was started at 5 wk of age ( i . e . 2–3 wk after the initiation of cancer cell dissemination ) , no significant change was seen in the number of cutaneous metastases ( Figure 5D ) . This shows that metastatic potential had already been acquired at 5 wk and that cancer cells disseminating before 5 wk contain tumor initiating cells . Depletion of PMN-MDSC resulted in a very significant inhibition of tumor cell dissemination ( 85% reduction in the draining lymph node , 74% in the lung , 40% reduction in the incidence of cutaneous tumors ) and EMT ( 68% reduction in S100A4+ cells and 70% reduction in vimentin+ cells ) . Other tumor infiltrating immune cells have been shown to favor metastasis . In a mammary tumor model , CD11b+Gr1+ cells were shown to accumulate in the primary tumor and promote metastasis through increased production of MMP . Interestingly , in this model , EMT was not involved [52] . In a model of colorectal cancer , accumulation of immature myeloid cells distinct from MDSC favored invasion through the production of MMP [53] . PMN-MDSC from RETAAD tumors also produce MMP , mostly MMP9 , but when RETAAD mice were crossed to an MMP9-KO background , no significant difference was observed in cancer cell dissemination or in the incidence of cutaneous metastases ( unpublished data ) . Therefore , myeloid cells that infiltrate the primary tumor could contribute to metastasis through more than one mechanism . Transcriptome analysis of the various cell subsets purified from RETAAD tumors showed that PMN-MDSC express TGF-β1 and HGF , whereas melanoma cells express EGF ( Figure 7 ) . This observation extends previous studies in a mammary tumor model showing that MDSC represent an abundant source of intratumoral TGF-β [52] , [54] . We found that specific inhibitors of the corresponding receptors blocked PMN-MDSC mediated induction of EMT . The fact that EGF inhibitors reduced EMT induction while PMN-MDSC purified from RETAAD tumors did not produce EGF may suggest that PMN-MDSC stimulate EGF production by the cancer cells . Cancer progression has been often depicted as a linear process , during which the incipient cancer cell sequentially accumulates genetic and epigenetic changes that confer the hallmarks of cancer cells [55] . Here we show that some of these properties can be induced by cues provided by the immune stroma of the primary tumor . Once the cell migrates out of the primary tumor , it may well lapse back to its original phenotype and , for example , undergo mesenchymal-epithelial transformation . Such a transient phenotypic switch may accelerate carcinogenesis and participate in the plasticity of cancer cells [56] , [57] . Tumor infiltration by myeloid cells is associated with a poor prognosis for cancer patients . Most studies have focused on the role of macrophages [10] , [58]–[61] , whereas MDSC have been primarily characterized for their effects on immune cells [18] , [62]–[66] , stromal cells [67] , or endothelial cells [68] . The present study demonstrates for the first time that PMN-MDSC act on cancer cells to promote proliferation , EMT , and dissemination . In turn , it provides a mechanistic explanation for the long-recognized link between inflammation and cancer progression .
Animal care and experimental procedures were approved by the IACUC ( Application No . 090425 ) of the Biological Resource Center , 20 Biopolis Way , Singapore 138668 . The generation of RETAAD mice has been previously described [69] . Rosa mT/mG reporter mice expressing tdTomato and Il8rb-KO mice were obtained for JAX Laboratories ( Cat No . 007576 and 006848 , respectively ) . Il8rb-KO mice were crossed with mice expressing EGFP under the Lysozyme promoter [70] to obtain Gr1hi cells lacking CXCR2 and expressing EGFP . Anti-CD45-FITC , anti-CD19-PE , anti-CD11c-PerCP/Cy5 . 5 , anti-Gr1-PE/Cy7 , anti-F4/80-APC , anti-CD11b-APC/Cy7 , anti-CD3-PE , anti-CD4-PerCP/Cy5 . 5 , anti-CD8-PE/Cy7 , anti-NK1 . 1-APC , and anti-CD45-PE antibodies were from Biolegend , and anti-I-A/I-E-eFluor450 and anti-CD45-eFluor450 antibodies were from eBioscience . Anti-Melanoma antibody ( ab732 ) was obtained from Abcam . Anti-rat E-Cadherin antibody was from BD Pharmingen while anti-mouse E-cadherin antibody was from R&D Systems . Anti-mouse vimentin antibody was from Proteintech Group . Tumors were dissected from RETAAD mice and single cell suspensions were obtained by digestion with Collagenase A ( 1 mg/ml; Roche ) and DNase I ( 0 . 1 mg/ml , Roche ) for 20 min at 37°C followed by filtration through a 70 µm filter ( BD Biosciences ) . Flow cytometric data were acquired on the BD LSRII ( BD Bioscience ) and analyzed using FlowJo software ( Tree Star , Inc . ) . Dissociated tumor cells or recombinant CXCL1 , CXCL2 , and CXCL5 ( R&D Systems ) were placed in the bottom chamber of a 24-well plate , and 1×106 total cells from the blood and spleen of 12-wk-old RETAAD mice were placed in the upper chamber ( 3 µm , BD Falcon ) . Cells were allowed to migrate to the bottom well for 3 h at 37°C , 5% CO2 . Migrated cells were then analyzed by flow cytometry on the BD FACSCalibur and quantified using CountBright Absolute Counting Beads ( Invitrogen ) . For the inhibition of CXCR2 , PMN-MDSC were treated 1 h before and during the migration assay with inhibitors SB225002 and SB265610 from Tocris Bioscience . For the in vivo migration assay , 5×106 bone marrow cells from Rosa mT/mG reporter mice expressing tdTomato and an equal number of Il8r-KO bone marrow cells expressing GFP were injected intra-orbitally into tumor-bearing RETAAD mice . After 18 h , the ratio of tdTomato+ cells to GFP+ Gr1hi cells infiltrating the contra-lateral tumor was measured by flow cytometry . For each mouse ( n = 11 ) , RNA was extracted from one primary tumor and one cutaneous tumor and gene expression was analyzed using Mouse Common Cytokines PCR Arrays ( SABiosciences , PAMM-021 ) and Mouse Inflammatory Cytokines & Receptors PCR Arrays ( SABiosciences , PAMM-011 ) following the manufacturer’s instructions . CXCL2 was analyzed separately using the following primers: 5’-AGTGAACTGCGCTGTCAATG-3’ and 5’-GAGAGTGGCTATGACTTCTGTCTG-3’ . Gene expression was normalized to GAPDH expression , and p values were calculated using a two-tailed paired t test . Dct expression was measured as previously described [5] . Mice were injected intraperitoneally twice a week with 0 . 25 mg of anti-Ly6G NIMP-R14 antibody [71] or control rat IgG ( Sigma-Aldrich ) . Efficiency of PMN-MDSC depletion was monitored by flow cytometry . Mice were clinically assessed for palpable tumors once per fortnight . Injection schedules are as illustrated in Figure S3A . Formalin-fixed paraffin-embedded sections ( 5 µm ) were immunolabeled for S100B and Ki67 as described previously [5] or for S100A4 ( Abcam; ab27957 ) . Tumor areas were assessed using ImagePro Analyzer 6 . 2 software ( Media Cybernetics Inc . ) . Total cell number was calculated by dividing the tumor area by the average area of one cancer cell ( estimated to be 82 µm2 ) . The mitotic index was calculated as follows: PMN-MDSC and macrophages were isolated using an immune-magnetic separation kit—EasySep Mouse PE Positive Selection Kit ( STEMCELL Technologies ) —according to the manufacturer’s protocol . Briefly , PMN-MDSC and macrophages from the blood and spleen of 12-wk-old mice were labeled with anti-Ly6G-PE ( 1A8; BD Bioscience ) or anti-F4/80-PE ( BM8; eBioscience ) antibodies , respectively . PE-labeled cells were immuno-magnetically labeled and positively selected using an EasySep magnet . Cell purity was verified by flow cytometry and was always >80% . CD8+ T cells were isolated from the spleen of C . Cg-Rag2tm1Fwa Tg ( DO11 . 10 ) 10Dlo ( Taconic #4219 ) using a CD8a+ T Cell Isolation Kit ( Miltenyi Biotec ) . 1×105 isolated T cells were plated with 2×104 irradiated ( 2 , 000 rads ) cells from the CD8- fraction . PMN-MDSC isolated from tumor-bearing mice ( 12-wk-old ) were added at a ratio of 1∶1 , 1∶4 , 1∶16 , and 1∶64 to T cells . 5 µg/ml of SIINFEKL peptide ( GL Biochem ) was added . After 96 h incubation , T cell proliferation was measured by [3H]thymidine incorporation . NBT-II cells , expressing histone H2B-mCherry , were seeded at a density of 100 cells per well on Ibidi 8 well µ-slides and left in culture for 4 d . Inhibitors were added on day 3 . Purified PMN-MDSC ( 1×104 cells/ml ) were added on day 4 and left in co-culture for 24 h . EGF ( 100 ng/ml; Sigma ) was used as a positive control . Inhibitors used were TGF-β1 receptor inhibitor , SB 525334 ( Tocris; final concentration 10 µM ) , EGF receptor inhibitor , PD 153035 hydrochloride ( Tocris; final concentration 8 µM ) , and HGF receptor inhibitor , JNJ38877605 ( Selleck Chemicals; final concentration 5 µM ) . Time-lapse images were captured with an Olympus FV-1000 confocal system and analyzed with MacBiophotonics ImageJ [72] and Imaris ( Bitplane ) . NBT-II cells were then fixed and stained with Phalloidin-Alexa Fluor 488 ( Invitrogen ) and mouse anti-rat Desmoplakin ( Millipore; clone DP2 . 15 ) . A goat anti-mouse antibody conjugated to Alexa Fluor 488 ( Invitrogen ) was used as a secondary antibody . Fluorescent images were captured with an Olympus IX-81 Inverted microscope and Retiva-SRV CCD Camera ( QImaging ) and analyzed with ImagePro analysis software ( MediaCybernetics ) . Gene expression data were log-transformed before analysis using Student’s t test . Statistical tests used to analyze other data are described in the individual figure legends . Prism ( GraphPad Inc . ) and Excel ( Microsoft ) softwares were used for calculations and graphing . p values less than 0 . 05 were considered statistically significant . | Cancer progression has been depicted as a linear process , during which the incipient cancer cell sequentially accumulates mutations that confer the ability to metastasize . However , recent studies show that cancer cells disseminate early , before such mutations can accumulate , implying the existence of a different , faster route to the metastatic phenotype . Using a mouse model of melanoma , we show that the primary tumor attracts a subset of immune cells that actively promote cancer cell motility , dissemination , and metastasis . These tumor-infiltrating immune cells do so by reactivating a cellular program ( mesenchymal transition ) used by melanocytes during their development to colonize the skin , and also believed to be an essential step in cancer cell dissemination and metastasis . Once the melanoma cells migrate out of the primary tumor , they can lapse back to their original phenotype and lose their migratory potential . This transient phenotypic switch may accelerate carcinogenesis and participate in the plasticity of cancer . It explains how cancer cells might spread to other organs even before the original tumor is detected . In addition to the evidence gleaned from our mouse melanoma model , we show that these immune cells induce typical features of epithelial-mesechymal transition in both melanoma and bladder human cell lines when examined in culture dishes . These findings provide an underlying mechanism for the long-recognized link between inflammation and cancer progression . | [
"Abstract",
"Introduction",
"Results",
"Materials",
"and",
"Methods"
]
| [
"oncology",
"medicine",
"immune",
"cells",
"immunology",
"biology"
]
| 2011 | Mesenchymal Transition and Dissemination of Cancer Cells Is Driven by Myeloid-Derived Suppressor Cells Infiltrating the Primary Tumor |
Growth and maintenance of hydatid cysts produced by Echinococcus granulosus have a high requirement for host lipids for biosynthetic processes , membrane building and possibly cellular and developmental signalling . This requires a high degree of lipid trafficking facilitated by lipid transporter proteins . Members of the fatty acid binding protein ( FABP ) family have been identified in Echinococcus granulosus , one of which , EgFABP1 is expressed at the tegumental level in the protoscoleces , but it has also been described in both hydatid cyst fluid and secretions of protoscoleces . In spite of a considerable amount of structural and biophysical information on the FABPs in general , their specific functions remain mysterious . We have investigated the way in which EgFABP1 may interact with membranes using a variety of fluorescence-based techniques and artificial small unilamellar vesicles . We first found that bacterial recombinant EgFABP1 is loaded with fatty acids from the synthesising bacteria , and that fatty acid binding increases its resistance to proteinases , possibly due to subtle conformational changes induced on EgFABP1 . By manipulating the composition of lipid vesicles and the ionic environment , we found that EgFABP1 interacts with membranes in a direct contact , collisional , manner to exchange ligand , involving both ionic and hydrophobic interactions . Moreover , we observed that the protein can compete with cytochrome c for association with the surface of small unilamellar vesicles ( SUVs ) . This work constitutes a first approach to the understanding of protein-membrane interactions of EgFABP1 . The results suggest that this protein may be actively involved in the exchange and transport of fatty acids between different membranes and cellular compartments within the parasite .
Hydatidosis is a highly pathogenic infection with an almost global incidence caused by the larval stage ( metacestode ) of the cestode Echinococcus granulosus . In endemic areas it has serious health effects on humans , livestock and wildlife , representing a major public health and economic burden in many countries [1]–[3] . Echinococcus species , as do other tapeworms of mammals , require two hosts to complete their life cycle . The E . granulosus eggs containing the infective oncosphere are shed in the faeces of wild and domestic carnivores that are the definitive hosts harbouring the dwarf adult tapeworms . Once a suitable intermediate host ingests the eggs , they hatch and the oncosphere is released , escaping from the intestine to establish hydatid cysts in liver and lungs . The cyst produces thousand of protoscoleces , each of which can progress to the adult form when ingested by the definitive host [4] , but it is the hydatid cysts in intermediate hosts that cause significant pathology and death . Hydatid disease in humans is highly pathogenic and is particularly difficult to treat successfully , especially so when cysts develop and proliferate in the lungs . Fatty acid binding proteins ( FABPs ) are small proteins ( 14–15 kDa ) that bind non-covalently to hydrophobic ligands , mainly fatty acids ( FA ) and retinoids . FABPs are confined to the interior of the synthesising cells , the only known exceptions to this being in nematodes [5] , [6] . Several tissue-specific FABP types have been identified in vertebrates , each named after the tissue in which they are predominantly expressed , and have also been given numeric designations [7] . In mammals they are implicated in intracellular uptake , storage and transport of FAs in lipid metabolism and membrane building , as well as protection from the membrane-disruptive effects of free long chain FAs [8] . In addition , the non-FA-binding retinoid-binding isoforms contribute to regulation of gene expression [9] . However , the precise function of each FABP type remains poorly understood , but sub-specialization of functions is suggested by the tissue-specific and temporal expression , in addition to ligand preferences [10] . Despite very similar tertiary structures , FABPs have been found to interact with membranes in different ways that might reflect how they acquire and deliver their cargoes . The fluorescence-based biophysical approaches used for this have shown that most FABPs from mammals ( adipocyte FABP , intestinal FABP , heart FABP , keratinocyte FABP , myelin FABP , etc . ) and one from Schistosomes ( Sj-FABPc ) exhibit a collisional mechanism of ligand exchange , meaning that they interact by direct contact with a membrane in ligand transfer . In contrast , only liver FABP and cellular retinol binding protein II from mammals transfer ligands in a diffusional mechanism , meaning that transfer occurs without requiring direct contact between protein and membrane but through release of ligand into the aqueous phase followed by its intercalation into the membrane . Proteins like liver FABP may therefore be more involved in lipid storage and regulation in the cytoplasm rather than in direct transport of FAs [8] , [11] , [12] . FABPs of parasitic platyhelminths are interesting because these parasites are unable to synthesise most of their own lipids de novo , in particular long-chain FAs and cholesterol [13] , [14] . Such lipids must therefore be acquired from the host , and then delivered by carrier proteins to specific destinations within the parasite . Whether they are involved extracellularly in lipid acquisition from , or delivery to , host cells , remains to be seen . It is noteworthy that EgFABP1 has been found in hydatid cyst fluid and in protoscolex secretions [15] , [16] . A final reason for interest in FABPs is their potential role in drug delivery and the fact that they have been assayed as vaccine candidates [17]–[23] . EgFABP1 is considered to be a member of the heart FABP subfamily [24] , [25] , whose members are believed to be involved in lipid oxidation processes [8] . The ligand-binding properties of EgFABP1 have been investigated by the displacement of cis-parinaric acid by a set of hydrophobic ligands [26] , and its crystal structure reveals the 10-stranded β-barrel fold typical of the family of intracellular lipid-binding proteins [27] . The objective of this study was to investigate the lipid transport properties and protein-membrane interaction characteristics of EgFABP1 . We characterise the biophysical properties of the protein in a number of ways , and show that the protein exchanges FAs through a collisional , direct contact , mechanism with acceptor membranes , indicating that it may indeed be involved in FA dynamics within the parasite , but that it may also engage in direct , non-specific interactions with host cell membranes .
The cDNA encoding EgFABP1 ( UniProtKB/Swiss-Prot Q02970 ) was subcloned into pET11b . The expression of the protein was carried out in E . coli BL21 ( DE3 ) by induction with 0 . 4 mM isopropyl-beta-D-thiogalactoside for 3 hours at 37°C in Luria Bertani medium in presence of 100 µg/mL of ampicillin . Cells were lysed by sonication and the lysate clarified by ultracentrifugation ( 25 min , 61700× g , 4°C ) . Following clarification , the supernatant was subjected to salting out incubating the protein for 2 hours at 4°C with 0 . 5 volume of a saturated ammonium sulphate solution . After centrifugation , the obtained protein solution was applied into a size exclusion chromatographic column ( Sephadex G-50 , Pharmacia Biotech Inc . ) . The fractions containing EgFABP1 were subsequently subjected to ionic exchange chromatography employing a MonoQ column ( Pharmacia Biotech Inc . ) in order to remove nucleic acids contamination . Delipidation was carried out using a Lipidex 1000 column ( Sigma ) at 37°C in a high ionic strength buffer ( 10 mM phosphate ( K2HPO4 6 mM+KH2PO4 4 mM ) , 1 M KCl ) . As an approach for studying binding preferences of EgFABP1 , the lipid moiety of recombinant non-delipidated EgFABP1 was extracted according to Bligh & Dyer's method [28] and analysed on a TLC plate using a mobile phase for resolving neutral lipids ( hexane∶diethyl-ether∶acetic acid at 80∶20∶1 , v∶v∶v ) . The FA composition of EgFABP1 lipid fraction was analysed by GC of their methyl esters derivatives methylated with BF3-Methanol according to the method described by Morrison & Smith [29] , employing an HP 6890 device Hewlett Packard ) as described previously by Maté et al . [30] . In order to analyse possible conformational changes between apo- and holo- forms , EgFABP1 was subjected to limited proteolysis experiments . The protocol was a modification of that described by Arighi et al . [31] . Briefly , clostripain ( ArgC , Sigma ) was activated by preincubation in 10 mM phosphate ( K2HPO4 6 mM+KH2PO4 4 mM ) , 150 mM KCl , pH 7 . 4 and 1 mM DTT for 2 hours . Prior to digestion , delipidated EgFABP1 ( 0 , 5 mg/ml ) was incubated for 30 min with either myristic acid , palmitic acid , stearic acid or oleic acid in ethanol ( 4∶1 mol∶mol ligand∶protein ) to obtain holo-forms . As a control of the FA solvent used , an equal volume of ethanol was added to the apo-form . Additional 15 min incubation with 1 mM DTT was carried out previous to the addition of the protease . At fixed intervals , samples were collected and frozen for subsequent analysis by SDS-PAGE . SDS-PAGE was carried out according to Schägger and von Jagow [32] in 16 . 5% acrylamide Tris-Tricine . After Coomassie Blue staining digital images were collected employing an ImageQuant 350 device ( GE Healthcare ) . Fatty acid binding to EgFABP1 was assessed employing a fluorescent titration assay [33] . Briefly , 0 , 5 µM anthroyloxy-fatty acid ( AOFA , Molecular Probes ) was incubated at 25°C for 3 min in 40 mM Tris , 100 mM NaCl , pH 7 . 4 buffer ( TBS ) with increasing concentrations of EgFABP1 . The AOFAs employed for binding assays were 12- ( 9-anthroyloxy ) stearic acid ( 12AS ) and 16- ( 9-anthroyloxy ) palmitic acid ( 16AP ) . Fluorescence emission at 440 nm was registered after excitation at 383 nm in a Fluorolog-3 Spectrofluorometer ( Horiba-Jobin Yvon ) . An exact equilibrium n-sites binding model was fitted to fluorescence data ( using Microcal ORIGIN software ) as previously described [34] . For AOFA transfer experiments , small unilamellar vesicles ( SUVs ) were prepared by sonication and ultracentrifugation as described previously [35] . The standard vesicles were prepared to contain 90 mol % of egg phosphatidylcholine ( EPC ) and 10 mol % of N- ( 7-nitro-2 , 1 , 3-benzoxadiazol-4-yl ) -phosphatidylcholine ( NBD-PC ) , which served as the fluorescent quencher . To increase the negative charge density of the acceptor vesicles , either 25 mol % of phosphatidylserine ( PS ) or cardiolipin ( CL ) was incorporated into the SUVs in place of an equimolar amount of EPC . Vesicles were prepared in TBS except for SUVs containing CL which were prepared in TBS with 1 mM EDTA . SUVs containing 64 mol % EPC , 10 mol % egg phosphatidylethanolamine ( EPE ) , 25 mol % CL and 1 mol % dansyl-phosphatidylethanolamine ( DPE ) were prepared in 20 mM Tris , 0 . 1 mM EDTA , pH 7 . 4 for protein-membrane interaction assays . Large unilamellar vesicles ( LUVs ) of EPC were prepared ( 1 mM in phospholipids ) by extrusion through polycarbonate membranes of 100 nm pore diameter ( Avestin Inc . , Ottawa , Canada ) as described previously [35] . All lipids were purchased from Avanti Polar Lipids . Ligand partition between the protein and NBD-containing SUVs was determined by measuring AOFA fluorescence at different protein∶SUVs ratios obtained by adding SUV to a solution containing 10 µM EgFABP1 and 1 µM 12AS in buffer TBS at 25°C [36] . The relative partition coefficient ( KP ) was defined as: ( 1 ) Where [Ligand-SUV] and [Ligand-FABP] are the concentration of AOFA bound to membrane and EgFABP1 , respectively , and [FABP] and [SUV] are the concentration of protein and vesicles . The decrease in AOFA fluorescence as a function of SUV is related to KP by ( 2 ) Where Frel , [SUV] , [FABP] , Kp , a and b are the relative fluorescence , the molar concentration of SUV , the molar concentration of EgFABP1 , the partition constant and fitting parameters , respectively [37] . The partition coefficient was used to establish AOFA transfer assay conditions that ensure essentially unidirectional transfer , as detailed below . A Förster Resonance Energy Transfer assay was used to monitor the transfer of 12AS from EgFABP1 to acceptor model membranes as described in detail elsewhere [11] , [33] , [38] . Briefly , EgFABP1 with bound 12AS was mixed at 25°C with SUVs , prepared as above , using a stopped-flow RX2000 module ( Applied Photophysics Ltd . ) attached to the spectrofluorometer . The NBD moiety is an energy transfer acceptor of the anthroyloxy group donor; therefore , the fluorescence of the AOFA is quenched when the ligand is bound to SUVs that contain NBD-PC . Upon mixing , transfer of AOFA from protein to membrane is directly monitored by the time-dependent decrease in anthroyloxy group fluorescence . Different SUVs and buffer compositions were employed in order to analyse the ligand transfer mechanism . Transfer assay conditions were 15∶1 mol∶mol EgFABP1∶AOFA ratio . SUVs were added ranging from 1∶10 mol∶mol to 1∶40 mol∶mol EgFABP1∶SUVs . Controls to ensure that photobleaching was eliminated were performed prior to each experiment , as previously described [38] . Data were analysed employing SigmaPlot and all curves were well described by an exponential decay function . For each experimental condition within a single experiment , at least five replicates were done . To analyse the putative association of EgFABP1 with vesicles , an assay that exploits the well known membrane-interactive properties of cytochrome c was employed . The binding of cytochrome c to acidic membranes can be monitored by using a resonance energy transfer assay [39] in which the dansyl fluorescence of DPE-labelled SUV is quenched upon binding of cytochrome c , which contains the heme moiety quencher . Competition of EgFABP1 with cytochrome c for binding to SUVs was determined by the relief of cytochrome c-related quenching of the dansyl fluorescence . In a final volume of 200 µl , 0–48 µM EgFABP1 was added to 15 µM SUV in 20 mM Tris . HCl/0 . 1 mM EDTA , pH 7 . 4 . After a 2 min equilibration , fluorescence emission at 520 nm was measured ( λex = 335 nm ) . Cytochrome c ( Sigma ) was then added ( 1 µM final concentration ) , and the mixture equilibrated an additional 2 min period before monitoring again fluorescence emission at 520 nm . In the absence of bound FABP , the dose-dependent quenching of dansyl fluorescence is observed . An inhibition of cytochrome c-dependent quenching is interpreted as evidence for EgFABP1 interaction with SUVs , i . e . , EgFABP1 prevention of subsequent cytochrome c interaction with the bilayer .
This assay was performed in order to determine which lipid classes bind to EgFABP1 in a cellular environment . Despite E . coli's cytoplasm not being the natural environment of EgFABP1 , this approach could contribute to the assignment of the protein's natural ligands as it analyses the preference of EgFABP1 for different hydrophobic compounds present in the bacterial cytoplasm . TLC analysis showed that only FAs were bound to the recombinant protein ( data not shown ) . Among them , palmitic acid ( 16:0 ) and stearic acid ( 18:0 ) are important ligands , although myristic ( 14:0 ) , pentadecanoic ( 15:0 ) , palmitoleic ( 16:1 n-7 ) , 7-hexadecenoic ( 16:1 n-9 ) , oleic ( 18:1 n-9 ) , vaccenic ( 18:1 n-7 ) , and linoleic acid ( 18:2 ) were also detected ( Figure 1 ) . The latter may come from culture media , as E . coli is not able to synthesise polyunsaturated FAs , at least during log-phase growth [40] , [41] . The distribution of FAs bound to FABP may be related to the relative abundance of each of them in E . coli , and it correlates well with the reported FA composition of E . coli grown in equivalent conditions [41] . As in previous in vitro displacement of fluorescent ligand studies where palmitic and stearic acids are among those that produce moderate displacement percentages ( >50% ) [26] , this experiment shows that EgFABP1 is able to bind many FAs of different chain length and degree of insaturation . In addition , in agreement with these results , the crystal structure of recombinant EgFABP1 revealed an electronic density inside the cavity , which was interpreted as being palmitic acid [27] . We therefore proceeded to investigate protein:membrane transfer using fluorophore-tagged fatty acid analogues . Partial proteolysis can provide information related to conformational changes in proteins since this technique may reveal the differential exposure of proteolytic sites in apo and holo forms . We analysed the peptide pattern obtained by digestion of EgFABP1 in its apo- or different holo-forms with Clostripain ( ArgC ) . The FAs selected , following to the analysis of ligands bound to recombinant EgFABP1 ( Figure 1 ) , were myristic , palmitic , stearic and oleic acids . The enzyme hydrolyses the polypeptide chain at the C-terminal end of arginine residues . Qualitative differences were evident between apo-EgFABP1 and the different complexes ( Figure 2 ) . Results show that binding of FAs gives EgFABP1 significant relative protection against cleavage . After 5 minutes of proteolysis the apo-protein shows several bands corresponding to proteolytic fragments , while the holo-forms show mainly the band corresponding to full-length EgFABP1 and less intense bands corresponding to proteolytic fragments ( Figure 2A ) . This suggests that ligand-binding results in a different exposure of proteolytic sites . It is interesting to note that after 16 hours of proteolysis the holo-proteins do not seem to be further proteolysed while the apo-protein is almost completely degraded ( Figure 2B ) . Previous results obtained for other members of the family of FABPs have suggested that binding of ligands involves conformational changes , especially on the portal region of FABPs [10] , [31] , [42] . Furthermore , in silico simulations show that , upon ligand binding , subtle conformational changes can be detected inside the cavity , in the surface and in the portal region of EgFABP1 ( Esteves , unpublished data ) . These changes could make cleavage sites less accessible to the protease . As an additional approach to investigate conformational changes between apo- and holo-protein , we analysed the circular dichroism ( CD ) spectra of EgFABP1 in the far ( 200–250 nm ) and near ( 250–320 nm ) UV regions . Two different ligands were employed for the generation of holo-EgFABP1: palmitic and oleic acid . Results indicated that the far-UV spectra of apo- and the two holo-forms did not show appreciable differences as can be seen in Figure S1 . These data could be interpreted to show that no significant changes in overall secondary structure content are caused by ligand binding . On the other hand , the near-UV CD spectra ( Figure S1 ) showed differences upon ligand binding , especially with oleic acid , indicating a likely alteration in the environment of aromatic residues resulting from proximity to ligand and/or a change in the conformation of the protein . So , ligand binding to EgFABP1 could elicit a change in the tertiary structure of the protein that could be correlated to the relative resistance of the holo form to proteolytic attack observed in the previous experiment . In preparation for experiments on the interaction of EgFABP1 with phospholipid vesicles , binding experiments were performed using fluorescent analogues of stearic and palmitic acids , 12AS and 16AP , respectively . Anthroyloxy probes are useful indicators of binding site characteristics because their spectral properties are environment-sensitive . These probes usually have very low fluorescence intensity in buffer , which becomes dramatically enhanced upon interaction with a FABP [43] . 12AS showed a large increase in fluorescence emission accompanied by a substantial blue shift upon binding to EgFABP1 . On the other hand , 16AP's fluorescence was surprisingly decreased when bound to EgFABP1 , but also accompanied by a distinct blue shift in emission ( Figure 3 ) . This blue-shift indicates that the fluorophore had entered an apolar environment , almost certainly the hydrophobic binding pocket rather than a superficial , non-specific site of the protein . Following addition of artificial 100 mol % phosphatidylcholine LUVs to the 16AP:EgFABP1 complex , the intensity of fluorescence emission increased , indicating that the quenching of 16AP's fluorescence emission was reversed upon transfer to the different , lipidic , environment of the vesicles . In both cases ( 12AS and 16AP ) the titration described curves that reached saturation , in accordance to a ligand binding phenomenon consistent with 1∶1 binding , with a Kd of 0 . 12±0 . 02 µM for 12AS , and 0 . 013±0 . 006 µM for 16AP . 12AS was chosen as a ligand for the following analysis of transfer kinetics due to its fluorescence emission characteristics when bound to protein being more typical of that observed in other studies on protein to membrane transfer [37] , [44] , [45] . However , the quenching effect observed with 16AP will be very useful to analyse FA transfer between EgFABP1 and other proteins that show a typical increase of AOFA fluorescence upon binding . Regarding this , another lipid binding protein from E . granulosus which is very abundant in the hydatid fluid , Antigen B , has been investigated in its binding properties , showing that it binds 16AP with a 30-fold fluorescence enhancement of the probe [46] . The apparent partition coefficient that describes the relative distribution of 12AS between EgFABP1 and EPC-SUVs was determined by adding SUVs containing NBD-PC to a solution of 12AS:EgFABP1 complex . As a result of this experiment , a KP value of 0 . 48±0 . 23 was obtained employing Eq . 2 ( see Materials and Methods ) , which indicates that there is preference of the AOFA for the phospholipid membranes . In a collisional transfer , the limiting step is the effective protein-membrane interaction , and the rate increases as the acceptor membrane concentration increases . In a diffusional mechanism in which the rate limiting step is the dissociation of the protein-ligand complex , no change in rate is observed [33] , [37] , [38] , [44] , [45] , [47] . The values of Kd and KP were used to set the conditions for the transfer assay . The proportion of protein and ligand was such that less than 1% of AOFA remained free in the preincubation solution . On the other hand , KP value was used to calculate the final concentrations of protein and SUVs for which unidirectional transfer prevailed . Figure 4 shows that when constant concentrations of the EgFABP1-12AS donor complexes were mixed with increasing concentrations of EPC-SUV , the 12AS transfer rate from EgFABP1 to EPC-SUV increased proportionally to vesicle concentration in the SUV: EgFABP1 ratios ( 10∶1 to 40∶1 ) tested . In these conditions , the increase in transfer rate ranged from 0 . 04±0 . 01 sec−1 to 0 . 12±0 . 03 sec−1 . These results strongly suggest that the FA transfer from EgFABP1 occurs via a protein-membrane interaction rather than by simple aqueous diffusion of the free ligand . Considering the hypothesis that FA transfer from EgFABP1 occurs by collisional contact with an acceptor membrane , changes in membrane properties should modify the transfer rate . If the mechanism relied on aqueous diffusion alone , then the characteristics of acceptor membranes should be irrelevant to the transfer rate , since the rate-determining step in such a transfer process ( ligand dissociation into the aqueous phase ) is a physically and temporally distinct event from processes involving the membrane . Figure 5 shows that 12AS transfer rate from EgFABP1 to membranes increased when acceptor membranes contained 25% of negatively charged phospholipids ( PS or CL ) . In agreement with the behaviour we have previously observed for collisional mammalian FABPs [38] , [44] , [45] , [47] , EgFABP1 shows a large increase in FA transfer rate to CL vesicles compared with zwitterionic vesicles . To investigate further the effect of negative charge of the acceptor vesicles on the FA transfer mechanism from the protein , we analysed the modification of transfer rates with increasing concentrations of negatively charged acceptor vesicles . The rate of FA transfer from EgFABP1 always , and independently of the net charge of the vesicles , showed the classical proportional increase in transfer rate with acceptor concentration ( Figure 6 ) . Transfer of 12AS from EgFABP1 to membranes was examined as a function of increasing concentrations of NaCl . The results show that an important increase in transfer rate from EgFABP1 to neutral membranes was detected with increasing ionic strength of the aqueous phase ( Figure 7A ) . It is generally thought that electrostatic interactions at surfaces are diminished and hydrophobic interactions are stimulated as a function of increasing ionic strength . The effect of ionic strength on the rate of AOFA transfer from EgFABP1 to zwitterionic vesicles suggests that the elimination of electrostatic interactions by salt shielding is compensated by an increase in hydrophobic interactions . When negative charge was added to the acceptor lipid vesicles , a drastic decrease was observed at high salt concentrations ( Figure 7B ) . As shown in Figure 5 , EgFABP1 exhibited approximately a 60-fold increase in AOFA transfer rate to CL vesicles compared with EPC vesicles at low ionic strength . Upon increasing the ionic strength , a marked decrease from the very high values observed at low ionic strength was found ( Figure 7B ) . This suggests a masking of electrostatic interactions , which play a very important role at low ionic strength , caused by the high salt content of the buffer . FA transfer experiments suggest that the interaction of EgFABP1 with membranes is sensitive to surface charge density . As cytochrome c is known to interact as a peripherally associating protein with acidic membranes [48] , we analysed the ability of EgFABP1 to compete with cytochrome c for binding to membranes containing CL . Cytochrome c quenches dansyl fluorescence in a concentration-dependent manner ( ref . [49] and Figure 8 ) . Results show that preincubation of CL-containing vesicles with EgFABP1 was effective in preventing cytochrome c binding in a concentration-dependent manner ( Figure 8 ) . When EgFABP1 ( 48 µM ) was added to CL-containing SUVs , the dansyl fluorescence was twice that obtained in the absence of EgFABP1 and with 1 µM cytochrome c .
We show that recombinant EgFABP1 is able to bind FAs of different chain lengths from E . coli , mainly palmitic and stearic acids . This is clearly an incomplete inventory of ligands that it may transport in the parasite , but it does illustrate the propensity of the protein to bind FAs when exposed to an environment rich in a wide range of small hydrophobic compounds . The analysis of the natural ligands bound by EgFABP1 in the parasite environment is currently being undertaken in our laboratory . Our main finding in this work was that the protein engages in a collisional mechanism in ligand transfer , as do various FABP isoforms from mammals , and one from Schistosomes , that have been investigated in this way [8] , [11] , [12] . This involvement of direct contact between protein and membrane for this transfer was found by altering electrostatic and hydrophobic conditions in the transfer experiments . The results indicated that the interaction event is mediated by both charge and hydrophobic factors , and it would seem reasonable that the initial interaction is ionic , between the protein and charged phospholipid headgroups , followed by direct , transient hydrophobic interaction with the apolar layer of the membrane . The interaction of the protein with membranes has also been demonstrated by the competition with cytochrome c for membrane binding . The tertiary structure of EgFABP1 is virtually superimposable on FABPs that engage in collisional transfers [27] , in which the two alpha-helices adjacent to the portal of ligand entry in FABPs are important in engaging contact with membranes [38] . It may be no coincidence that EgFABP1 has , like these other collisional FABPs , a prominent pair of bulky hydrophobic amino acid sidechains ( Phe27 , Val28 ) extending into solvent from helix II , immediately adjacent to the portal . Such a ‘sticky finger’ could attract and orient ligand for entry into the protein , or be involved in the protein's interaction with membranes or other proteins [50] . Our results suggest that EgFABP1 is likely to be an active participant in the transport and exchange of lipids in vivo , which could involve uptake of FAs directly from , and delivery to , membranes within the parasite , potentially resourcing the developing protoscoleces within the hydatid cysts . This might also be the case for Antigen B , which belongs to a new family of hydrophobic ligand binding proteins of cestodes and has been proposed as a lipoprotein involved in lipid trafficking [46] , [51] . Furthermore , our proteolysis experiments with EgFABP1 and the analysis of CD spectra of apo- and holo-forms indicated that ligand binding would induce a conformational change in the protein . Such a change might modify the mechanism of interaction of EgFABP1 with membranes to facilitate upload or download of their cargo . A conformational change could also function as a signal to target the protein to different destinations , as has been suggested for other members of the FABP family [52] , [53] . The possibility that it also interacts with host cell membranes is more contentious , particularly since EgFABP1 does not have a secretory leader peptide , as is also the case for FABPs from any group of animals other than nematodes [5] , [6] , so should be confined to the interior of cells . However , if EgFABP1 appears in cyst fluid in vivo and in excretion/secretion products of protoscoleces [15] , [16] ( but not as a result of cell damage during fluid collection or imperfect culture conditions in the collection of excretion/secretion products ) then the possibility that it does interact with host cells beyond the cyst wall must be considered . Host proteins are known to cross hydatid cyst walls [15] , so it is conceivable that this permeability ( if a unidirectional transfer system is not in operation ) could mean that EgFABP1 leaves the cyst to interact with host membranes for return to the parasite , or to deliver lipids to host tissues for immunomodulation . These hypotheses remain to be tested . In this regard , future studies should also include protein interaction analysis with membranes that mimic parasite and host composition . This work is a first approach to understand the functional properties of EgFABP1 and constitutes the basis for further expanding our knowledge about this protein . This has been the case for other members of the FABP family , where this kind of studies has contributed to the understanding of the mechanisms of ligand transfer to membranes , protein-membrane and protein-protein interactions [8] , [54] . | Echinococcus granulosus is the causative agent of hydatidosis , a zoonotic infection that affects humans and livestock , representing a public health and economic burden in many countries . Since the parasites are unable to synthesise most of their lipids de novo , they must acquire them from the host and then deliver them by carrier proteins to specific destinations . E . granulosus produces in abundance proteins of the fatty acid binding protein ( FABP ) family , one of which , EgFABP1 has been characterised at the structural and ligand binding levels , but it has not been studied in terms of the mechanism of its interaction with membranes . We have investigated the lipid transport properties and protein-membrane interaction characteristics of EgFABP1 by applying biophysical techniques . We found that EgFABP1 interacts with membranes by a mechanism which involves direct contact with them to exchange their cargo . Given that the protein has been found in the secretions of the parasite , the implications of its direct interactions with host membranes should be considered . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"apolipoproteins",
"protein",
"interactions",
"parasitic",
"diseases",
"protein",
"structure",
"neglected",
"tropical",
"diseases",
"echinococcosis",
"infectious",
"diseases",
"lipids",
"proteins",
"lipoproteins",
"biology",
"recombinant",
"proteins",
"biophysics",
"biochemistry",
"protein",
"chemistry",
"lipoprotein",
"structure"
]
| 2012 | Direct Interaction between EgFABP1, a Fatty Acid Binding Protein from Echinococcus granulosus, and Phospholipid Membranes |
Sustaining a balanced intestinal microbial community is critical for maintaining intestinal health and preventing chronic inflammation . The gut is a highly dynamic environment , subject to periodic waves of peristaltic activity . We hypothesized that this dynamic environment is a prerequisite for a balanced microbial community and that the enteric nervous system ( ENS ) , a chief regulator of physiological processes within the gut , profoundly influences gut microbiota composition . We found that zebrafish lacking an ENS due to a mutation in the Hirschsprung disease gene , sox10 , develop microbiota-dependent inflammation that is transmissible between hosts . Profiling microbial communities across a spectrum of inflammatory phenotypes revealed that increased levels of inflammation were linked to an overabundance of pro-inflammatory bacterial lineages and a lack of anti-inflammatory bacterial lineages . Moreover , either administering a representative anti-inflammatory strain or restoring ENS function corrected the pathology . Thus , we demonstrate that the ENS modulates gut microbiota community membership to maintain intestinal health .
The intestinal tract serves to harvest nutrients and energy , protect against harmful toxins and pathogens , and clear out waste . These functions can be modulated by both the enteric nervous system ( ENS ) and the trillions of symbiotic bacteria that reside within the gut [1–3] . Importantly , the influence of microbiota on intestinal functions and health depends on the constituent microbes . Alterations in microbial composition from those observed in “healthy” subjects are often defined as “dysbiotic , ” which refers to communities that become perturbed in their composition such that they acquire pathogenic properties [4–6] . Given that the composition of the microbiota is critical for host health , it is significant that the intestinal microbial community is generally stable despite the highly dynamic internal environment of the intestinal tract [7 , 8] , which experiences disruptions such as influxes of ingested matter , host secretion and epithelial cell turnover , and coordinated outward flow of material . How microbial community stability is achieved amid these constant perturbations is unknown . Hosts with impaired intestinal motility can develop dysbiosis and intestinal pathology [9 , 10] , which suggests a profound role for the ENS in constraining microbiota composition . Here , we explore how the ENS shapes the ecology of the intestine , and we address key questions about the assembly of dysbiotic microbial communities , their functional properties , and strategies for their treatment—three aspects of dysbiosis that have been challenging to address from observational studies in humans . Our analysis reveals how , without ENS constraint , imbalances in pro- and anti-inflammatory members of the microbiota can drive intestinal pathology . The most severe example of ENS dysfunction in humans is Hirschsprung disease ( HSCR ) , an enteric neuropathy that results from a failure of neural crest–derived cells to form the distal ENS [3] . Approximately 30% of HSCR patients develop a severe form of intestinal dysbiosis , known as Hirschsprung-associated enterocolitis ( HAEC ) [9–11] , which is distinguished by diarrhea , distension , fever , and , in extreme cases , sepsis and death [12] . Studies suggest that the etiology of HAEC has a microbial component , as both pathogenic bacteria [13] and alterations in commensal communities [9 , 10] have been linked to HAEC . Interestingly , patients with a broad range of human diseases , such as inflammatory bowel disease ( IBD ) , cystic fibrosis [14] , diabetes [15] , malnutrition [16] , and myotonic muscular dystrophy [17 , 18] , also experience debilitating gastrointestinal ( GI ) symptoms . Although cause and effect are difficult to determine , these diseases are associated with both small intestinal bacterial overgrowth , a clinical syndrome often seen with impaired intestinal motility , and an altered microbiota , suggesting that impaired ENS function could be a driver of dysbiosis . To explore how the ENS may prevent dysbiosis by constraining microbial populations , we turned to a zebrafish model of HSCR . Multiple well-described zebrafish lines carry mutations in HSCR loci [19–22] . The most extreme ENS loss is seen in mutants homozygous for a null mutation in the HSCR gene sox10 [23 , 24]; these mutants entirely lack an ENS [24] . The mutant allele t3 ( sox10t3 ) homozygotes have diminished rhythmic peristaltic activity [21] , making this an ideal model for dissecting the role of the ENS in host–microbe interactions . Zebrafish are well suited for examining ENS contributions to microbiota composition because we can monitor ENS development , absolute bacterial abundance , and disease phenotypes , such as neutrophil accumulation , across the entire intestine of individual larvae . Thus , we can assess system-level functional readouts that describe properties of the associated microbiota . Furthermore , the high fecundity and ease of working with zebrafish provide us with large sample sizes to increase the power of our experiments such that we can monitor how natural microbiota variation at the species level drives phenotypic variation . In this study , we demonstrate that the ENS constrains the abundance and composition of the microbiota . We find that loss of the ENS in sox10t3 mutants results in assembly of a dysbiotic community leading to a microbe-driven intestinal inflammation that varies among individuals and resembles HAEC . Microbiota profiling across the spectrum of inflammatory states revealed that extreme intestinal inflammation is linked to an outgrowth of pro-inflammatory bacterial lineages and a reduction of anti-inflammatory bacterial lineages . Moreover , administering representative anti-inflammatory bacterial strains or transplanting wild-type ( WT ) ENS precursors to restore a WT ENS corrects the pathology in sox10t3 mutant hosts . Our analysis reveals that ENS function is a key feature of intestinal health that constrains the composition of the resident microbiota and prevents overgrowth of bacterial lineages that can drive disease .
The complete loss of ENS in sox10t3 mutants ( S1 Fig ) results in defective intestinal motility [21] . Given the connection between altered intestinal motility and small intestinal bacterial overgrowth , we hypothesized that functional consequences of these mutants would include changes to intestinal ecology and alterations in resident microbial populations . To visualize the abundance and distribution of bacteria along the length of the intestine , we used fluorescent in situ hybridization ( FISH ) . In sox10t3 mutants , we noted large populations of bacteria throughout the intestine , with marked accumulations of bacteria at the esophageal-intestinal junction ( Fig 1A and 1B ) , a location not typically heavily colonized with bacteria . We also quantified the total number of colony-forming units ( CFU ) per intestine and found that sox10t3 mutants had a significantly higher bacterial load ( Fig 1C ) . These results suggest that sox10t3 mutants experience bacterial overgrowth , which is consistent with defective intestinal transit . Defective intestinal transit has been observed in mutants in another allele , sox10m241 , which have intestinal peristalsis but do not clear ingested fluorescent beads as well as WTs [25] . To demonstrate delayed intestinal transit in sox10t3 mutants , we adapted a previous single color assay [26] into a two-color intestinal transit assay ( S1 Fig ) . The delayed transit and impaired clearance we observed in sox10 mutants likely contribute to bacterial overgrowth within their intestines . For the work described in this manuscript , we use sox10t3 mutants , hereafter referred to as sox10 mutant or sox10- . We next asked whether the bacterial overgrowth phenotype in sox10- resulted in signs of intestinal inflammation . Thus , we quantified intestinal neutrophil populations , a marker of inflammation , in cohoused WTs and sox10 mutants by staining for the neutrophil-specific enzyme myeloid peroxidase . At 6 d post fertilization ( dpf ) , intestinal neutrophil accumulation in sox10 mutants was significantly increased compared to WTs ( Fig 1D and 1E ) . Notably , sox10 mutants exhibited a much greater variation in intestinal neutrophil accumulation ( 0–18; n = 30 ) compared to WT siblings ( 0–7; n = 31 ) ; some sox10 mutants had intestinal neutrophil levels similar to WTs , whereas others had significantly elevated neutrophil populations . Intestinal neutrophil accumulation under homeostatic conditions in WT fish requires the pro-inflammatory tumor necrosis factor ( TNF ) pathway [27 , 28] . The increased neutrophil response in sox10 mutants also depends on this pathway , as inhibiting expression of the TNF receptor using an antisense morpholino [27 , 28] abolished the increased neutrophil response ( Fig 1D and 1E ) . Another indicator of intestinal pathology is epithelial cell proliferation . At 6 dpf , sox10 mutants had markedly increased intestinal cell proliferation relative to cohoused WT animals . Unlike the normal intestinal epithelial cell proliferation response to microbiota , which is TNF independent [29] , we found that elevated cell proliferation in the sox10 mutant intestine was TNF dependent ( Fig 1F ) , suggesting that this was an inflammation-dependent pathological response . To determine whether the intestinal microbiota of sox10- hosts is necessary to induce the increased intestinal neutrophil response , we derived sox10 mutants and their WT siblings germ free ( GF ) . We found that GF sox10 mutants have a low neutrophil population , indistinguishable from their WT siblings ( Fig 2A ) . To determine if the microbial community established in sox10 mutants is sufficient to induce inflammation , we performed an experiment in which we transferred microbiota from sox10 mutants into WTs . As donors , we used microbial communities from conventionally raised ( CV ) WT , sox10 mutant , or WT intestinal alkaline phosphatase morpholino ( iap MO ) -injected larvae . iap MO-injected fish are hypersensitive to lipopolysaccharide and thus develop elevated intestinal inflammation without evidence of dysbiosis [27] . These fish serve as control for the possibility that nonbacterial factors such as host pro-inflammatory cytokines rather than microbial derived factors cause transmissible intestinal inflammation ( Fig 2B ) [30] . At 6 dpf , for each separate group ( WT , sox10- , and iap MO ) , we dissected , pooled , and homogenized the donor intestines . As a negative control , we included transplantation from homogenized intestines of GF fish . The homogenate from each group was inoculated into flasks housing GF 4 dpf WT fish ( Fig 2C ) . We found that inoculation with microbes from sox10 mutants was sufficient to induce elevated intestinal inflammation in WTs as compared to inocula from GF , CV WT , or CV iap MO fish , none of which induced intestinal inflammation ( Fig 2D ) . To test whether the capacity of sox10 mutant microbiota to induce elevated neutrophils was due to increased bacterial load , we transplanted 5× CV WT microbes , which corresponded to the bacterial load of sox10 mutant transplants . This larger inoculum did not induce more intestinal inflammation ( S2 Fig ) , which indicates that the microbial community assembled in sox10- hosts is functionally distinct from WT microbiota and is sufficient to induce inflammation in fish with a normal , functional ENS . sox10 mutants exhibit a wide range of intestinal neutrophil populations ( Figs 1D and 2A ) as well as variation in bacterial load ( Fig 1B ) . Therefore , we asked whether intestinal neutrophil abundance corresponded to increased bacterial abundance . We used transgenic sox10 mutant hosts expressing green fluorescent protein ( GFP ) under control of the neutrophil-specific mpx promoter to quantify both neutrophil population and intestinal bacterial load in individual fish ( Fig 3A ) . When we compared sox10 mutants that fell in the bottom half of neutrophil response ( “sox10- low” ) or in the top half of neutrophil response ( “sox10- high” ) to WTs , we found that all sox10 mutants , regardless of neutrophil level , carried significantly higher bacterial loads than WTs ( Fig 3B ) . Thus , impaired intestinal clearance ( S1 Fig ) leads to an increased bacterial load; however , the bacterial overgrowth per se in sox10- does not drive an increased intestinal neutrophil response . We further characterized the pro-inflammatory signature of the sox10- high- and low-neutrophil subsets by monitoring expression of a panel of immune genes in the intestine ( Fig 3C ) . These results aligned with our observations of the neutrophil population , as the sox10- high-neutrophil subset had elevated levels of mpx , saa , and tnfα expression compared to WT and the sox10- low-neutrophil subset ( Fig 3C ) ; however , the increase in saa transcription was the only one to reach statistical significance . Consistent with the significantly elevated intestinal neutrophil response in these samples , saa is known to mediate intestinal neutrophil behavior stimulated by microbes [31] . Collectively , our results suggest that a pro-inflammatory compositional change occurs in the microbial community of a subset of sox10 mutants . To address the possibility of a pro-inflammatory compositional change in the sox10- microbiota , we profiled microbial communities by performing 16S rRNA gene sequencing on intestinal communities isolated from cohoused WT and sox10 mutant individuals . We collected samples across three independent experiments . To uncover differences in microbiota composition that explain the variable severity of neutrophil accumulation in sox10 mutants , we collected intestinal neutrophil response data for the same individuals from which we isolated microbial DNA and grouped samples as “WT , ” “sox10- low” ( intestinal neutrophil response 0–8 ) , or “sox10- high” ( intestinal neutrophil response of greater than or equal to 22 ) ; these groups include the top 26% and the bottom 29% , respectively ( Fig 4A ) . By standard metrics of community variability ( non-metric multidimensional scaling of Canberra distances , richness , Faith’s Phylogenetic Diversity , unweighted UniFrac ) , these three groups were not significantly different ( S3 Fig ) , which indicates that these communities are largely made up of the same microbes , and community differences driving neutrophil differences are perhaps due to changes in minor members [28] . We next asked whether the relative abundance of any bacterial operational taxonomic units ( OTUs ) correlated with intestinal neutrophil number across all individuals surveyed in the study . Of 129 OTUs present in at least 20 individuals , we found a small subset whose percent abundance was associated with neutrophil number , as measured by Spearman’s correlations . Strikingly , these neutrophil-associated OTUs were tightly clustered in only two genera found in this population of fish intestines ( Fig 4B ) . The Escherichia/Shigella genus ( hereafter referred to as Escherichia ) had ten OTUs that negatively correlated with neutrophil abundance , although only two had borderline significance after false discovery rate correction . All OTUs of the Vibrio genus had significant positive correlations with neutrophil abundance ( Fig 4B ) . Examination of OTU abundances revealed not only that the two most abundant genera were Vibrio and Escherichia ( S3 Fig ) but also that they were significantly decreased and increased , respectively , in the “sox10- high” group relative to the WT and “sox10- low” groups ( Fig 4C and 4D ) . The observation of pro-inflammatory activity associated with Vibrio is consistent with our previous analysis of a zebrafish-derived Vibrio strain ( ZWU0020 ) [32] that is phylogenetically closely related to the Vibrio OTUs in the current experiment ( Fig 4E ) . Previously , we showed that Vibrio strain ZWU0020 ( hereafter referred to as Vibrio Z20 ) promotes intestinal neutrophil accumulation in a concentration-dependent manner in gnotobiotic zebrafish [28] . Similarly , in the current study , the log10 ( relative abundance ) of Vibrio was significantly positively correlated with neutrophil number ( Fig 5A ) , and we also observed that the log10 ( relative abundance ) of Escherichia was negatively correlated with intestinal neutrophil accumulation , although the amount of variation explained was low ( Fig 5A ) . This relationship mirrors the relationship we previously observed in simple microbial communities in gnotobiotic zebrafish between the abundance of Shewanella strain ZOR0012 ( hereafter referred to as Shewanella Z12 ) and a proportional decrease in neutrophil number [28] . Of note , two OTUs from the Shewanella genus included in our analysis in this study did not have a significant correlation with neutrophil number . Combining the loss of Escherichia and the gain of Vibrio does not increase the amount of variation in neutrophil number explained by the gain of Vibrio alone ( Table 1 , Fig 5B ) . We used Akaike’s Information Criterion ( AIC ) [33] to test the relative quality of each of these models; the model that accounts only for Vibrio reports the lowest AIC value , which identifies Vibrio as the best microbial predictor of intestinal neutrophil number variability ( Table 1 ) . These analyses suggest that a balance of the Vibrio and Escherichia lineages may be important for maintaining intestinal homeostasis , with Vibrio abundance being a major determinant of intestinal inflammation . To confirm the functional contribution of Vibrio to the increased neutrophil responses in sox10- , we first added Vibrio Z20 to CV sox10 mutants at 4 dpf and assayed neutrophil numbers at 6 dpf . Exogenously added Vibrio Z20 induced a significant increase in neutrophil accumulation over the number seen in CV sox10 mutants ( Fig 5C ) . Furthermore , the absolute abundance of Vibrio Z20 colonizing these fish was positively correlated with intestinal neutrophil number , similar to observations made in the 16S rRNA data set ( Fig 5D ) . We noted that the extent of the increase in neutrophil accumulation upon addition of Vibrio depended on the level of intestinal neutrophils present in control hosts ( S4 Fig ) , which we think reflects fluctuations in bacterial community composition in CV zebrafish between experiments and a limited ability to change the neutrophil-inducing capacity of an intestinal microbiota already dominated by Vibrio strains . We furthermore observed the inflammation-inducing capacity of Vibrio Z20 in monoassociation by adding Vibrio Z20 to GF sox10 mutants . In these conditions , Vibrio Z20 was still sufficient to induce high intestinal neutrophil influx ( Fig 5C ) . In monoassociation the range of Vibrio colonization was too narrow ( S4 Fig ) to explore a correlative relationship . These experiments support our hypothesis , based on the microbiota profiling of these fish , that an overabundance of Vibrio species causes a dysbiotic and pro-inflammatory microbiota . We hypothesized that the dysbiotic state of the sox10 mutant intestine could be corrected by balancing the pro-inflammatory activity of Vibrio species with the addition of anti-inflammatory isolates , such as Escherichia species or Shewanella Z12 [28] . Consistent with this prediction , we found that addition of Escherichia coli HS , a commensal Escherichia strain isolated from a healthy human adult [34] that is closely related to Escherichia OTUs in CV fish ( S4 Fig ) , can colonize the zebrafish intestine ( S4 Fig ) , reducing neutrophil numbers in CV sox10 mutants ( Fig 6A ) and maintaining GF levels of intestinal neutrophil accumulation in monoassociation ( Fig 6A ) . Moreover , the absolute abundance of colonizing E . coli HS in CV sox10 mutants displayed a similar negative correlation with intestinal neutrophils ( Fig 6B ) , as observed with the sequenced OTUs ( Fig 5A ) . Shewanella Z12 , another species that displays a negative correlation between abundance and intestinal neutrophil accumulation [28] , also reduced intestinal neutrophils in sox10 mutants , which suggests this relationship may be a hallmark of anti-inflammatory bacterial strains ( S4 Fig ) . Thus , sox10- dysbiosis can be corrected by adding anti-inflammatory bacteria to the community . Shewanella Z12 uses an unidentified secreted factor present in cell-free supernatant ( CFS ) to mediate its anti-inflammatory activity [28] ( S4 Fig ) . However , E . coli HS CFS was insufficient to reduce sox10 mutant intestinal inflammation ( S4 Fig ) , suggesting that these species use two distinct mechanisms to control the host innate immune response . As an alternative to manipulating the microbiota directly , we postulated that correcting the underlying deficit in the ENS would also alleviate the inflammation in sox10 mutants . To test this hypothesis , we performed a rescue experiment in which vagal neural crest cells , the ENS precursors , were transplanted from WT donors into sox10 mutant hosts . Our previous studies showed a correlation between the number of ENS neurons and gut motility [21]; thus , after the transplant , we assayed for formation of a normal-appearing ENS along with intestinal neutrophil accumulation . Following transplantation , the sox10 mutants that developed a normal-appearing ENS extending along the entire length of the intestine had WT levels of intestinal neutrophils ( Fig 6C and 6D ) , demonstrating that the ENS is sufficient to prevent intestinal inflammation . Together , our results demonstrate that the ENS contributes to intestinal health by maintaining a balanced gut microbiota , revealing a previously unappreciated role for the ENS in host–microbe interactions .
Microbiota are assembled through fundamental ecological processes , including dispersal , local diversification , ecological drift , and environmental selection [35] . We have previously shown that a portion of early larval zebrafish intestinal communities follow a neutral pattern of assembly [36] . This observation suggests that features of the gut environment constrain which microbes colonize and persist in the gut environment . We hypothesized that the ENS , which controls motility and aspects of intestinal homeostasis [3] , may also directly or indirectly serve as a significant constraint on intestinal microbial community assembly , such that loss of the ENS constitutes a major ecological shift . Consistent with this hypothesis , we show that zebrafish lacking an ENS have an altered intestinal microbiota and deficits in clearing food from the gut , suggesting gut motility is a mechanism by which the ENS influences microbiota composition . This is further supported by the recent finding that GI transit time is one of the largest predictors of microbiota composition [37] . Moreover , intestinal motility profoundly influences the spatial organization of bacterial populations and has been found to promote competitive exclusion within resident communities [38] . This suggests that abnormal GI transit patterns can significantly reshape ecological interactions within the gut . The ENS also contributes to epithelial barrier function and secretion; however , whether and how these functions are altered in the sox10 mutant has not yet been described . Therefore , observed alterations to the microbial community may be the result of changes to any ( or all ) of these functions ( Fig 7 ) . Of all ENS mutants , the sox10 mutant has the most overlapping characteristics with the human disease HSCR; however , given that the ENS is interconnected with many other organ systems , our work reveals the need to investigate other model systems of ENS dysfunction . Currently , no other available mutant both entirely eliminates the ENS as seen in sox10 mutants and retains normal craniofacial structures [21] . For example , another severe mutant , ret , has a few residual ENS neurons and also exhibits severe craniofacial defects that may impair bacterial colonization [39] . A zebrafish sox10 cell ablation model exists [40] but requires treatment with the antibiotic metronidazole , which would alter the microbiota and confound our experiments . For future experiments , developing a new line in which it is possible to specifically ablate enteric neurons at specified developmental stages will be essential . Mounting evidence suggests that ENS defects of HSCR patients , as well as those of HSCR animal models , are not restricted to the aganglionic region of the intestine but rather extend to more proximal intestinal regions; thus , these defects are poised to precipitate dysbiosis associated with HAEC [9 , 41 , 42] . Patients with chronic IBD can also have functional and structural abnormalities in the ENS that disrupt motility [43 , 44] . Although these pathologies are generally thought to be secondary to inflammation [45] , our data raise the possibility that , regardless of the origin of ENS defects , they have the potential to disrupt the microbial community and thus contribute to a feedback loop that prevents a healthy microbiota from establishing after an inflammatory episode . Such a cycle could explain the variable outcomes of treating IBD patients with probiotics [46] . However , in healthy hosts , this feedback loop may instead enforce stability and homeostasis within the system . The intimate connection between human health and microbiota suggests that health is an effect of services provided by the microbial ecosystem [35] , and thus to manage health through the microbiota , we need to identify the taxa that provide specific ecosystem services . One strategy to identify bacterial species that specifically influence host health or disease phenotypes is to define a dose–response relationship , or correlation , between a phenotype of interest and a microbial isolate . For example , we used gnotobiotic methods to identify Vibrio Z20 as pro-inflammatory in zebrafish by its positive correlation between abundance and intestinal neutrophil number [28] , defining a dose–response relationship for this isolate . In this study , we have now expanded this approach to complex communities and discovered that Vibrio species fulfill this pro-inflammatory role in the highly inflamed sox10 mutant gut—finding a positive relationship between relative abundance of Vibrio and number of intestinal neutrophils . Correlations between OTUs and host phenotypes have been important in the identification of “indicator” species of interest in chronic obstructive pulmonary disease [47] , ulcerative colitis [48] , and asthma [49] . Perez-Losada and colleagues [49] expanded this concept by comparing the host and bacterial transcriptomes of asthmatics and healthy controls . They revealed positive correlations between both bacterial phyla ( Proteobacteria ) and functions ( adhesion ) with the pro-inflammatory cytokine IL1A [49] . Similarly , a recent large-scale study used correlations to identify microbial drivers of cytokine expression in healthy humans [50] . These studies highlight the potential for using correlation to identify bacterial species , or properties of bacterial species , that have functional consequences for the host in health and disease . The zebrafish is an especially good model for this type of analysis because we can manipulate host genetics and the environment to control microbial variability across samples . For human studies , the heterogeneity in microbial communities among subjects may be a limiting factor in performing this type of analysis . Furthermore , zebrafish microbial communities are less complex than those of humans , which allows us to probe the data at a higher resolution , with less data reduction [48] , and to analyze host–microbe interactions at the OTU level . Our analysis at this resolution revealed conserved functions at the genera level . The phylogenetic conservation of certain bacterial traits suggests that interactions between zebrafish and their resident microbiota serve as a model for identifying bacterial lineages that influence phenotypes across many host species . For example , like the pro-inflammatory Vibrio identified in sox10 mutants , some Vibrios , such as Vibrio parahaemolyticus , induce inflammatory gastroenteritis in humans [51] . We also identified the Escherichia genus as anti-inflammatory in the zebrafish , and some Escherichia are used as probiotics in the treatment of inflammatory intestinal disorders like ulcerative colitis [52] . These data suggest that characterizing species correlated with host phenotypes in model organisms may help to identify individual members of complex communities that contribute to disease phenotypes . Viewing host–microbiota interactions as an ecological system allowed us to identify two system components , the ENS and key bacterial species , which greatly influence ecosystem function , as measured by host intestinal inflammation . With this information , we can ask whether manipulation of these components provides us with control over ecosystem function . For example , an expanded population of Vibrio lineages combined with a decreased population of Escherichia lineages in sox10 mutants induces increased neutrophil influx . We manipulated this component by introducing the anti-inflammatory Escherichia or Shewanella Z12 [28] and thus ameliorated the disease phenotype . Notably , the most consistent microbial signature of IBD patients is the loss of an anti-inflammatory species , Faecalibacterium prausnitzii [53] , the colonization level of which decreases in a step-wise manner from healthy subjects , to patients in remission , to patients with active colitis , to patients with infective colitis [54] . Furthermore , administration of F . prausnitzii reduces disease severity in mice with chemically induced colitis [55] . These results highlight the important immunomodulatory role played by specific bacterial species within the intestinal microbiota and the need to identify these species to devise therapies for reestablishing control of the intestinal environment and ameliorating dysbiosis . Treatment of dysbiosis-associated diseases with probiotics is likely to require continual probiotic administration if there is an underlying disease mechanism leading to its depletion . A more fruitful approach would include a treatment for the underlying ecological perturbation along with the introduction of probiotic strains . For example , restoring ENS function via transplantation or drug administration are possible ways to treat ENS dysfunction . We demonstrated that transplantation of WT ENS precursors into sox10 mutant hosts restored a normal-appearing ENS and rescued the inflammatory gut phenotype . We think that the normal inflammatory response indicates restored ENS function . However , in future experiments , determining the functional capacity of the transplanted ENS to restore motility , secretion , and epithelial barrier function will help elucidate which specific ENS functions contribute to the constraint on the intestinal microbiota . Recently , there has been significant success in establishing a functional ENS in mouse by transplantation of induced pluripotent stem cells [56] , which , together with our results , suggests that this strategy could contribute to a successful cure of disease in cases of HAEC or IBD .
Here , we have utilized the zebrafish as a powerful model to examine the complex relationship between the ENS , the immune system , and the microbiota . We demonstrated the critical role played by the ENS in shaping the ecology of the intestine by constraining the functional properties of the resident microbiota . Our analysis reveals how , without this constraint , imbalances in pro- and anti-inflammatory members of the microbiota can drive intestinal pathology . The imbalances we discovered could not be described by large changes in phylum level abundances or the acquisition of a single pathogenic lineage but rather by subtle differences in the abundances of key commensal species that have the potential to either protect against or promote inflammation . We note that this discovery reveals the reciprocal relationship between the microbes and the ENS , as ENS activity and development can be altered by microbiota; in fact , individual bacterial species can have distinct effects on ENS function [57] . Furthermore , immune cell responses influence ENS function both under healthy conditions [58] and in inflammatory states [59] . Therefore , intestinal homeostasis depends on a complex tri-directional conversation that occurs between the microbiota , the ENS , and the immune system , with proper functioning of each branch depending on signals from the other two branches . Uncovering new therapeutic strategies for chronic intestinal diseases will require a profound understanding not only of each branch of this system but the multifaceted interactions that connect them and how alterations made to one system ripple out to affect the function of the other two branches . Developing scalable and tractable model systems , such as the zebrafish , in which we can monitor all three branches of this system will be critical for addressing these complex questions .
All zebrafish experiments were done in accordance with protocols approved by the University of Oregon Institutional Animal Care and Use Committee ( protocol numbers 15–15 , 14-14RR , and 15-83A8 ) and conducted following standard protocols as described in [60] . CV-raised WT ( AB x Tu strain ) , heterozygote sox10t3- ( referred to as sox10- ) [24] , and Tg ( BACmpx:GFP ) i114 ( referred to as mpx:GFP ) [61] fish were maintained as described [60] . Homozygous sox10 mutants were obtained by mating heterozygotes and identified by lack of pigmentation [24] . The sox10t3 line was used for neutrophil experiments unless otherwise indicated . The mpx:GFP line [61] was crossed with sox10+/- adults to create a line that when in-crossed resulted in offspring that were sox10-/- and Tg ( BACmpx:GFP ) i114 ( referred to as sox10 , mpx:GFP ) . No defects were observed in heterozygous siblings , which have pigment , develop normally , and survive to adulthood , and thus they are grouped with homozygous WTs [21 , 23 , 24 , 38 , 62] . For all experiments , WT siblings and homozygous sox10 mutants were cohoused . See S1 Text . Splice-blocking MOs ( Gene Tools , Corvallis , OR ) were injected into embryos at the one cell stage . For knockdown of TNF , the tr1v1/tr1v2 MOs ( 1 . 2 moles and 6 moles , respectively ) were used as previously described [27 , 28] . For knockdown of intestinal alkaline phosphatase , the iape212 MO ( 3 pmoles ) was used as previously described [27] . Zebrafish larvae were fixed in 4% paraformaldehyde ( PFA ) overnight . Whole larvae were stained with Myeloperoxidase kit ( Sigma ) following the manufacturer’s protocol and processed and analyzed as previously described [27] . For analysis of neutrophils in mpx:GFP fish , GFP+ cells in the intestine were quantified as previously described [28] . For proliferation , larvae were immersed in 100 μg/ml EdU ( A10044 , Invitrogen ) for 16 h prior to PFA fixation . Subsequent processing and analysis were done as previously described [29] . See also Histology and neutrophil analysis in S1 Text . At 6 dpf , larvae were humanely killed with Tricaine ( Western Chemical , Inc . , Ferndale , WA ) , mounted in 4% methylcellulose ( Fisher , Fair Lawn , NJ ) , and their intestines were dissected using sterile technique . Dissected zebrafish intestines were placed in 100-μl sterile EM , homogenized , diluted , and cultured on tryptic soy agar plates ( TSA; BD , Sparks MD ) . After incubation at 32°C for 48 h for conventionally colonized fish or for 24 h for inoculated fish , colonies were counted . Zebrafish embryos were derived GF as previously described [63] . All manipulations to the GF flasks were performed under a class II A/B3 biological safety cabinet . Zebrafish inoculated with donor microbial populations were generated by inoculating flasks with 4 dpf GF zebrafish with 104 CFU/mL of donor microbes ( 1× ) . Donor microbes were collected by dissecting CV zebrafish and based on colonization data ( Fig 1B ) each fish was assumed to carry 105 total CFU/gut; a total of 25 dissected guts were pooled and homogenized to create the donor microbes . We inoculated CV fish with live Vibrio ( 106 bacterial cells/ml ) , E . coli HS ( 107 bacterial cells/ml ) , and Shewanella Z12 ( 106 bacterial cells/ml ) as previously described [28] . For monoassociations , each strain was inoculated at 106 bacterial cells/ml . We isolated and concentrated CFS as previously described [28] . Flasks were kept at 28°C until analysis of myeloperoxidase positive cells on 6 dpf . RNA isolation and cDNA preparation were performed as previously reported [27] except either five ( for saa , mpx , and il1b primers ) or 18 ( for mmp9 and tnfα primers ) dissected intestines were pooled . RNA was harvested by homogenizing and extracting with Trizol reagent ( Invitrogen ) . Contaminating genomic DNA was eliminated using the Turbo DNA-free kit ( Ambion ) per manufacturer’s instructions . The RNA ( 100 ng for saa , mpx , and il1b primers; 320 ng for mmp9 and tnfα primers ) was used as templates for generating cDNA with Superscript III Reverse Transcriptase and random primers ( Invitrogen ) following manufacturer’s instructions . The cDNA was measured in a qPCR reaction with SYBR Fast qPCR master mix ( Kapa Biosystems ) . Assays were performed in triplicate using ABI StepOne Plue RealTime . Data were normalized to elfa and analyzed using ΔΔCt analysis . Sequences and annealing temperatures are presented in S1 Table . Dissected intestines were placed in 2-mL screw cap tubes with 0 . 1 mm zirconia silica beads and 200-μL sterile lysis buffer ( 20 mM Tris-Cl; 2 mM EDTA; 2 . 5-mL 20% Tx-100 ) and frozen in liquid nitrogen . DNA was extracted using Qiamp DNA micro Kit ( Qiagen ) as detailed in S1 Text . The microbial communities of each sample were characterized by an Illumina HiSeq 2500 Rapid Run ( San Diego , CA ) sequencing the 16S rRNA gene amplicon by the University of Oregon Genomics and Cell Characterization Facility . The read length was paired-end 150 nucleotide , targeting the V4 region ( primers listed in S2 Table ) . The 16S rRNA gene Illumina reads were clustered using USEARCH 8 . 1 . 1803 [64] . The final OTU table was rarefied to a depth of 100 , 000 ( see S2 Data for metadata , S3 Data for OTU taxonomy , and S4 Data for OTU table ) . Measures of community diversity and similarity ( OTU richness , phylogenetic distances , unweighted UniFrac ) were calculated in R using vegan , picante , and GUniFrac ( See S2 Code ) . Correlations were calculated in R , and false discovery rate was adjusted using the Benjamini & Hochberg correction in p . adjust ( See S1 Code ) . WT donor embryos were labeled by injection of 5% tetramethylrhodamine dextran ( 3000 MW ) at the 1–2 cell stage and reared until the next manipulation in filter-sterilized EM . Embryos at the 12–14 somite stage were mounted in agar , a small hole dissected in the skin , and cells transplanted as previously described [65] and detailed in the S1 Text . Statistical analysis was performed using Prism ( Graphpad software ) . Statistical significance was defined as p < 0 . 05 . Data whose distributions were bounded by 0 were log transformed + 1 prior to statistical analysis . For correlations in Figs 5 and 6 , log transformations of neutrophil number and percent OTU were performed so the data met the assumptions of normality and homoscedasticity for linear regression . We note that the relationships and result of multiple linear regression were the same if the data were not log transformed . Throughout , box plots represent the median and interquartile range; whiskers represent the 5–95 percentile . Data for all figures are available in S1 Data . | Intestinal health depends on maintaining a balanced microbial community within the highly dynamic environment of the intestine . Every few minutes , this environment is rocked by peristaltic waves of muscular contraction and relaxation through a process regulated by the enteric nervous system ( ENS ) . We hypothesized that normal , healthy intestinal microbial communities are adapted to this dynamic environment , and that their composition would become perturbed without a functional ENS . To test this idea , we used a model organism , the zebrafish , with a genetic mutation that prevents formation of the ENS . We found that some mutant individuals without an ENS develop high levels of inflammation , whereas other mutants have normal intestines . We profiled the intestinal bacteria of inflamed and healthy mutants and found that the intestines of inflamed individuals have an overabundance of pro-inflammatory bacterial lineages , lack anti-inflammatory bacterial lineages , and are able to transmit inflammation to individuals with a normally functioning ENS . Conversely , we were able to prevent inflammation in the ENS mutants by either administering a representative anti-inflammatory bacterial strain or restoring ENS function . From these experiments , we conclude that the ENS modulates intestinal microbiota community membership to maintain intestinal health . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Conclusion",
"Materials",
"and",
"methods"
]
| [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"microbiome",
"gut",
"bacteria",
"pathology",
"and",
"laboratory",
"medicine",
"vibrio",
"immunology",
"microbiology",
"vertebrates",
"animals",
"animal",
"models",
"osteichthyes",
"model",
"organisms",
"signs",
"and",
"symptoms",
"enterobacteriaceae",
"experimental",
"organism",
"systems",
"escherichia",
"bacteria",
"neutrophils",
"microbial",
"genomics",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"white",
"blood",
"cells",
"inflammation",
"fishes",
"animal",
"cells",
"medical",
"microbiology",
"immune",
"response",
"gastrointestinal",
"tract",
"zebrafish",
"diagnostic",
"medicine",
"anatomy",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"genomics",
"organisms"
]
| 2017 | The enteric nervous system promotes intestinal health by constraining microbiota composition |
Change detection is a classic paradigm that has been used for decades to argue that working memory can hold no more than a fixed number of items ( “item-limit models” ) . Recent findings force us to consider the alternative view that working memory is limited by the precision in stimulus encoding , with mean precision decreasing with increasing set size ( “continuous-resource models” ) . Most previous studies that used the change detection paradigm have ignored effects of limited encoding precision by using highly discriminable stimuli and only large changes . We conducted two change detection experiments ( orientation and color ) in which change magnitudes were drawn from a wide range , including small changes . In a rigorous comparison of five models , we found no evidence of an item limit . Instead , human change detection performance was best explained by a continuous-resource model in which encoding precision is variable across items and trials even at a given set size . This model accounts for comparison errors in a principled , probabilistic manner . Our findings sharply challenge the theoretical basis for most neural studies of working memory capacity .
Visual working memory , the ability to buffer visual information over time intervals of the order of seconds , is a fundamental aspect of cognition . It is essential for detecting changes [1]–[3] , integrating information across eye fixations [4]–[5] , and planning goal-directed reaching movements [6] . Numerous studies have found that visual working memory is limited , but the precise nature of its limitations is subject of intense debate [7]–[14] . The standard view is that visual working memory cannot hold more than about four items , with any excess items being discarded [7]–[9] , [15]–[18] . According to an alternative hypothesis , working memory limitations take the form of a gradual decrease in the encoding precision of stimuli with increasing set size [10]–[11] , [13] , [19]–[23] . In this view , encoding precision is a continuous quantity , and this hypothesis has therefore also been referred to as the continuous-resource hypothesis . Historically , the leading paradigm for studying visual working memory has been change detection , a task in which observers report whether a change occurred between two scenes separated in time [2]–[3] , [24] . Not only humans , but also non-human primates can perform multiple-item change detection [25]–[28] , and physiological studies have begun to investigate the neural mechanisms involved in this task [27] . Findings from change detection studies have been used widely to argue in favor of the item-limit hypothesis [2] , [8] , [15]–[18] . The majority of these studies , however , used stimuli that differed categorically from each other , such as line drawings of everyday objects or highly distinct and easily named colors . The logic is that for such stimuli , changes are large relative to the noise , avoiding the problem of “comparison errors” [1] , [18] , [29]–[30] that would be associated with low encoding precision ( high noise ) . When encoding precision is limited , an observer's stimulus measurements are noisy and will differ between displays for each item , even if the item did not change . The observer then has to decide whether a difference in measurements is due to noise only or to a change plus noise , which is especially problematic when changes are small . This signal detection problem results in comparison errors . Attempts to avoid such errors by using categorical stimuli run into two objections: first , using such stimuli does not guarantee that comparison errors are absent and can be ignored in modeling; second , there is no good reason to avoid comparison errors , since the pattern of such errors can help to distinguish models . Ideally , change detection performance should be measured across a wide range of change magnitudes , including small values , as we do here . Comparison errors can , in fact , be modeled rather easily within the context of a Bayesian-observer model . Bayesian inference is the decision strategy that maximizes an observer's accuracy given noisy measurements [31]–[32] , and was recently found to describe human decision-making in change detection well [33] . We conducted two change detection experiments , in the orientation and color domains , in which we varied both set size and the magnitude of change . We rigorously tested five models of working memory limitations , each consisting of an encoding stage and a decision stage . The encoding stage differed between the five models: the original item-limit model [2] , [15]–[16] , two recent variants [9] , and two continuous-resource models , one with equal precision for all items [20] , [23] , and one with item-to-item and trial-to-trial variability in precision [13] , [33] . The decision stage was Bayesian for every model . To anticipate our results , we find that variable precision coupled with Bayesian inference provides a highly accurate account of human working memory performance across change magnitudes , set sizes , and feature dimensions , and far outperforms models that postulate an item limit .
We model a task in which the observer is presented with two displays , each containing N oriented stimuli and separated in time by a delay period . On each trial , there is a 50% probability that one stimulus changes orientation between the first and the second display . The change can be of any magnitude . Observers report whether or not a change occurred . We tested five models of this task , which differ in the way they conceptualize what memory resource consists of and how it is distributed across items ( Fig . 1a ) . We conducted an orientation change detection task in which we manipulated both set size and change magnitude ( Fig . 2a ) . Consistent with earlier studies ( e . g . [10] , [15] , [17] ) , we found that the ability of observers to detect a change decreased with set size , with hit rate H monotonically decreasing and false-alarm rate F monotonically increasing ( Fig . 2b ) . Effects of set size were significant ( repeated-measures ANOVA; hit rate: F ( 3 , 27 ) = 52 . 8 , p<0 . 001; false alarm rate: F ( 3 , 27 ) = 82 . 0 , p<0 . 001 ) . The increase in F is inconsistent with the IP model , as this model would predict no dependence . For a more detailed representation of the data , we binned magnitude of change on change trials into 10 bins ( Fig . 2c ) . All no-change trials had magnitude 0 and sat in a separate bin . These psychometric curves clearly show that the probability of reporting a change increases with change magnitude at every set size ( p<0 . 001 ) . From Fig . 2c we could , in principle , compute a naïve estimate of memory capacity using the well-known formula from the IP model , K = N ( H−F ) / ( 1−F ) [16] . However , since H depends on the magnitude of change , the estimated K would depend on the magnitude of change as well , contradicting the basic premise of a fixed capacity . For example , at set size 6 , for change magnitudes between 0° and 9° , Cowan's formula would estimate K at exactly zero ( no items retained at all ) , while for magnitudes between 81° and 90° , it would estimate K at 3 . 8 , with a nearly linear increase in between . This serves as a first indication that the IP model in general and this formula in particular are wrong .
Five models of visual working memory limitations have been proposed in the literature . Here , we tested all five using a change detection paradigm . Although change detection has been investigated extensively , several of the models had never been applied to this task and no previous study had compared all models . Compared to previous studies , our use of a continuous stimulus variable and changes drawn from a wide range of magnitudes enhanced our ability to tell apart the model predictions . Our results suggest that working memory resource is continuous and variable and do not support the notion of an item limit . The variable-precision model of change detection connects a continuous-resource encoding model of working memory [13] with a Bayesian model for decision-making in change detection [33] . This improves on two related change detection studies that advocated for continuous resources . Wilken and Ma [10] introduced the concept of continuous resources , but only compared an EP model with a suboptimal decision rule to the IP model . Although the EP model won in this comparison , the more recent item-limit models ( SA and SR ) had not yet been proposed at that time . Our present results show that the SA and SR models are improvements over both the EP and IP models , but lose to the VP model . In a more recent study , we compared different variants of the Bayesian model of the decision process and found that the optimal decision rule outperformed suboptimal ones [33] , but we did not vary set size or compare different models of working memory . Other tasks , such as change localization [13] , visual search [21] , [23] , and multiple-object tracking [19] , [46] , can also be conceptualized using a resource-limited front end conjoined with a Bayesian-observer back end . Whether such a conceptualization will survive a deeper understanding of resource limitations remains to be seen . It is instructive to consider each model in terms of the distribution over precision that it postulates for a given set size . In the IP model , this distribution has mass at infinity and , depending on set size , also at zero . In the SA and SR models , probability mass resides , depending on set size , at one or two nonzero values , or at zero and one nonzero value . The EP model has probability mass only at one nonzero value . The VP model is the only model considered that assigns probability to a broad , continuous range of precision values . Roughly speaking , the more values of precision a model allows , the better it seems to fit . Although we assumed in the VP model that precision follows a gamma distribution , it is possible that a different continuous distribution can describe variability in precision better . However , the amount of data needed to distinguish different continuous precision distributions using psychophysics only might be prohibitive . Work by Rouder et al . used a change detection task to compare a continuous-resource model based on signal detection theory to a variant of the IP model [8] . Manipulating bias , they measured receiver-operating characteristics ( ROCs ) . The IP variant predicted straight-line ROCs , whereas the continuous-resource model predicted regular ROCs ( i . e . , passing through the origin ) . Unfortunately , each of the ROCs they measured contained only three points , and therefore the models were very difficult to distinguish . We ourselves , in an earlier study , had collected five-point ROCs using confidence ratings , allowing for an easier distinction between different ROC types; there , we found that the ROCs were regular [10] , in support of a continuous-resource model . A difference between the Rouder study and our current study is that Rouder et al . used ten distinct colors instead of a one-dimensional continuum; this again has the disadvantage of missing the stimulus regime in which the signal-to-noise ratio is low . Moreover , the decision process in their continuous-resource model was not optimal; an optimal observer would utilize knowledge of the distribution of the stimuli and change magnitudes used in the experiment . It is likely that the optimal decision rule would have described human behavior in Rouder et al . 's experiment better than an ad-hoc suboptimal rule [33] . Finally , Rouder et al . did not consider variability in precision . In short , our current study does not contradict the results of Rouder et al . , but offers a more plausible continuous-resource model and tests all models over a broader range of experimental conditions . The notion of an item limit on the one hand and continuous or variable resources on the other hand are not mutually exclusive . In the SR model , for example , a continuous resource is split among a limited number of items . Although this model was not the best in the present study , many other “hybrid” models can be conceived – such as a VP model augmented with an item limit , or an IP or SA model with variable capacity [47]–[48] – and testing them is an important direction for future work . Our results , however , establish the VP model as the standard against which any new model of change detection should be compared . The neural basis of working memory limitations is unknown . In the variable-precision model , encoding precision is the central concept , raising the question which neural quantity corresponds to encoding precision . We hypothesize that precision relates to neural gain , according to the reasoning laid out in previous work [13] , [19] , [33] . To summarize , gain translates directly to precision in sensory population codes [49] , increased gain correlates with increased attention [50] , and high gain is energetically costly [51] , potentially bringing encoding precision down as set size increases . The variable-precision model predicts that the gain associated with the encoding of each item exhibits large fluctuations across items and trials . There is initial neurophysiological support for this prediction [52]–[53] . Furthermore , if gain is variable , then spiking activity originates from a doubly stochastic process: spiking is stochastic for a given of value of gain , while gain is stochastic itself . Recent evidence points in this direction [54]–[55] , although formal model comparison remains to be done . The variable-precision model also predicts that gain on average decreases with increasing set size . We proposed in earlier work that this could be realized mechanistically by divisive normalization [19] . Divisive normalization could act on the gains of the input populations by approximately dividing each gain by the sum of the gains across all locations raised to some power [56] . When set size is larger , the division would be by a larger number , resulting in a post-normalization gain that decreases with set size . A spiking neural network implementation of aspects of continuous-resource models was proposed recently [57] . Taken together , the variable-precision model has plausible neural underpinnings . Our results have far-reaching implications for neural studies of working memory limitations . Throughout the field , taking a fixed item limit for granted has been the norm , and many studies have focused on finding its neural correlates [12] , [58] . Even if we restrict ourselves to change detection only , a fixed item limit has been assumed by studies that used fMRI [59]–[65] , EEG [66]–[72] , MEG [67] , [72]–[73] , voxel-based morphometry [74] , TMS [68] , [75] , lesion patients [76] , and computational models [77]–[78] . Our present results undermine the theoretical basis of all these studies . Neural studies that questioned the item-limit model or attempted to correlate neural measures with parameters in a continuous-resource model have been rare [27] , [57] . Perhaps , this is because no continuous-resource model has so far been perceived as compelling . The variable-precision model remedies this situation and might inspire a new generation of neural studies .
Stimuli were displayed on a 21″ LCD monitor at a viewing distance of approximately 60 cm . Stimuli were oriented ellipses with minor and major axes of 0 . 41 and 0 . 94 degrees of visual angle ( deg ) , respectively . On each trial , ellipse centers were chosen by placing one at a random location on an imaginary circle of radius 7 deg around the screen center , placing the next one 45° counterclockwise from the first along the circle , etc . , until all ellipses had been placed . Set size was 2 , 4 , 6 , or 8 . Each ellipse position was jittered by a random amount between −0 . 3 and 0 . 3 deg in both x- and y-directions to reduce the probability of orientation alignments between items . Stimulus and background luminances were 95 . 7 and 33 . 1 cd/m2 , respectively . Ten observers participated ( 4 female , 6 male; 3 authors ) . All were between 20 and 35 years old , had normal or corrected-to-normal vision , and gave informed consent . On each trial , the first stimulus display was presented for 117 ms , followed by a delay period ( 1000 ms ) and a second stimulus display ( 117 ms ) . In the first display , set size was chosen randomly and the orientation of each item was drawn independently from a uniform distribution over all possible orientations . The second display was identical to the first , except that there was a 50% chance that one of the ellipses had changed its orientation by an angle drawn from a uniform distribution over all possible orientations . The ellipse centers in the second screen were jittered independently from those in the first . Following the second display , the observer pressed a key to indicate whether there was a change between the first and second displays . A correct response caused the fixation cross to turn green and an incorrect response caused it to turn red . During the instruction phase , observers were informed in lay terms about the distributions from which the stimuli were drawn ( e . g . , “The change is equally likely to be of any magnitude . ” ) . Each observer completed three sessions of 600 trials each , with each session on a separate day , for a total of 1800 trials . There were timed breaks after every 100 trials . During each break , the screen displayed the observer's cumulative percentage correct . Methods for model fitting and model comparison are described in the Text S1 . | Working memory is a fundamental aspect of human cognition . It allows us to remember bits of information over short periods of time and make split-second decisions about what to do next . Working memory is often tested using a change detection task: subjects report whether a change occurred between two subsequent visual images that both contain multiple objects ( items ) . The more items are present in the images , the worse they do . The precise origin of this phenomenon is not agreed on . The classic theory asserts that working memory consists of a small number of slots , each of which can store one item; when there are more items than slots , the extra items are discarded . A modern model postulates that working memory is fundamentally limited in the quality rather than the quantity of memories . In a metaphor: instead of watering only a few plants in our garden , we water all of them , but the more plants we have , the less water each will receive on average . We show that this new model does much better in accounting for human change detection responses . This has consequences for the entire field of working memory research . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"working",
"memory",
"cognitive",
"neuroscience",
"computational",
"neuroscience",
"psychophysics",
"biology",
"sensory",
"perception",
"neuroscience"
]
| 2013 | No Evidence for an Item Limit in Change Detection |
Rabies is a fatal encephalitis caused by viruses belonging to the genus Lyssavirus of the family Rhabdoviridae . It is a viral disease primarily affecting mammals , though all warm blooded animals are susceptible . Experimental rabies virus infection in birds has been reported , but naturally occurring infection of birds has been documented very rarely . The carcass of a domestic fowl ( Gallus domesticus ) , which had been bitten by a stray dog one month back , was brought to the rabies diagnostic laboratory . A necropsy was performed and the brain tissue obtained was subjected to laboratory tests for rabies . The brain tissue was positive for rabies viral antigens by fluorescent antibody test ( FAT ) confirming a diagnosis of rabies . Phylogenetic analysis based on nucleoprotein gene sequencing revealed that the rabies virus strain from the domestic fowl belonged to a distinct and relatively rare Indian subcontinent lineage . This case of naturally acquired rabies infection in a bird species , Gallus domesticus , being reported for the first time in India , was identified from an area which has a significant stray dog population and is highly endemic for canine rabies . It indicates that spill over of infection even to an unusual host is possible in highly endemic areas . Lack of any clinical signs , and fewer opportunities for diagnostic laboratory testing of suspected rabies in birds , may be the reason for disease in these species being undiagnosed and probably under-reported . Butchering and handling of rabies virus- infected poultry may pose a potential exposure risk .
Rabies is a fatal encephalitis caused by the viruses belonging to the genus Lyssavirus of the family Rhabdoviridae . It is a viral disease primarily affecting mammals , though all warm blooded animals are susceptible . In India and other Asian countries more than 90% of human infections occur due to exposure to rabid dogs , while cats , monkeys and other wild animals are reported to transmit the infection in the rest of the cases [1] . Perusal of available literature revealed that natural infection of birds with rabies virus has been documented uncommonly . The occurrence of rabies in a chicken under natural conditions is considered extremely rare [2] . The present case of clinical rabies in a domestic fowl was identified from an area which has a significant stray dog population and is highly endemic for canine rabies .
This study included samples received for diagnostic confirmation from a domestic fowl that died naturally . No tissue or clinical sample was obtained from the dead fowl specifically for the purpose of this study . No human subject or human clinical samples were included in this study . Hence , ethical clearance from the institutional review boards ( IRB , NIMHANS and IRB , CDIO ) was not required . The carcass of a domestic fowl ( Gallus domesticus ) was brought to the rabies diagnostic laboratory at the Chief Disease Investigation Office , Department of Animal Husbandry , Kerala , India for diagnosis of rabies . The bird had been bitten by a stray dog one month back in its breast muscle . The wound was treated locally . After one month , the bird appeared droopy and off-feed for a day and succumbed . As rabies was frequently reported in the locality among dogs and other domestic animals , the owner brought the carcass for ruling out rabies . The stray dog that bit the fowl could not be traced and laboratory confirmation of its rabid status could not be done . A detailed necropsy was conducted . The whole brain was collected from the bird with all biosafety precautions . Two independent laboratories with facilities for rabies diagnosis tested the sample . Touch impressions of the brain were stained by Seller’s stain and examined for the presence of Negri bodies . Impressions , fixed in high grade chilled acetone at -20°C , were subjected to the fluorescent antibody test ( FAT ) for detection of rabies virus nucleoprotein antigens [3] . A one step TaqMan real time PCR targeting the nucleoprotein gene was carried out as described earlier [4] on the brain tissue for confirmation of rabies virus infection . Partial gene sequencing was carried out by amplifying a 446 bp region in the nucleoprotein gene using nested PCR [5] . PCR products were purified using a commercial kit ( QIAquick Gel purification kit , Qiagen , UK ) and custom sequenced from Amnion Biosciences Pvt . Ltd , Bangalore , India using gene specific primers . The sequence was deposited in the GenBank database under accession number KP316199 . Partial nucleoprotein gene sequences of rabies virus isolates from GenBank representing various geographical regions in India and two other countries from the Indian subcontinent i . e Sri Lanka and Nepal were used to investigate the phylogenetic relationship with the present rabies virus strain . The sequences were aligned using ClustalW , and a maximum likelihood phylogeny tree was constructed using MEGA5 software [6] with bootstrap replication values of 1000 .
No significant lesions could be observed in any of the visceral organs on post-mortem examination . The dog bite wound on the breast muscle was fully healed . The brain impressions tested positive for rabies virus antigens by FAT , however Negri bodies could not be demonstrated . The brain tissue sample was positive for rabies viral RNA by TaqMan real time PCR . The phylogenetic tree comprising of various nucleoprotein gene sequences of rabies virus isolates from India can be divided into 2 divergent clusters ( Fig 1 ) . The lower cluster comprises of most of the isolates from the northern and southern part of India belonging to the Arctic/Arctic-like lineage , along with the Arctic Fox isolate from Canada ( included for comparison ) . The upper cluster includes the present strain ( NNV-RAB-FOWL ) and other rabies virus strains from Southern India , Sri Lanka and Nepal which belong to the distinct Indian subcontinent lineage .
Rabies has been conventionally considered as a disease of mammals and clinical cases of naturally acquired rabies has been reported in birds infrequently . A few anecdotal reports were published in the late 1950’s , indicating the rare occurrence of rabies in birds [7 , 8]; however they remain uncorroborated by lack of additional reports with robust laboratory evidence of naturally acquired rabies in birds . Gough and Jorgenson ( 1976 ) examined 343 serum samples of birds for antibodies against rabies virus and found that 23 of them had low passive haemagglutination titers [9] . However , in another serological survey in captured birds , no significant titers were detected [10] . A variety of birds have been infected experimentally with development of clinical signs , often without development of neurological features , or with recovery from clinical signs [2 , 11] . The present report indicates that rabies is a disease that can affect birds . Lack of obvious clinical signs and fewer opportunities for diagnostic laboratory testing of suspected rabies in a bird , may be the reason for the disease in these species being undiagnosed and probably underestimated . The locality of the bird in the present case is highly endemic for rabies [12] and the presence of an easily accessible diagnostic facility may be the reason the disease could be identified in poultry . Most often birds succumb due to shock or complication of animal bite injury and may not survive until the development of clinical rabies infection . Phylogenetic analysis based on partial nucleoprotein gene sequencing revealed 98% homology of the present rabies virus strain ( NNV-RAB-FOWL ) to a canine rabies virus isolate ( AF374721 ) from Chennai , in Tamil Nadu state in south India . Interestingly , it clustered with another Indian isolate from Goa , a southern state and also with distinct isolates from Sri Lanka , rather than with other Indian isolates in the Arctic/Arctic-like virus cluster , the extensive circulation of which has been reported from India . Earlier reports had speculated the presence of this distinct Sri Lankan variant in India and suggested that movement of humans and their animals between Sri Lanka and India , particularly within the southeastern coastal area of the mainland , may have resulted in the movement of this variant between these geographically separate regions [13–15] . Other studies confirmed the circulation of this distinct variant found only in Sri Lanka and mostly the southern part of India [16] . However , recently isolates from Nepal have also been reported to be phylogenetically related to these distinct variants and this lineage has been designated as the Indian subcontinent rabies virus clade [17] . This clade with phylogenetically related strains from South India ( including the present strain ) , Sri Lanka and Nepal is also evident in Fig 1 . Sequencing and phylogenetic analysis of additional isolates representative of various geographical areas in India and other countries in the Indian sub-continent may aid in further elucidation of the epidemiological significance of these variants . In conclusion , a case of naturally acquired rabies infection in a bird species , Gallus domesticus is reported in India for the first time . It indicates that spill over of infection even to unnatural hosts is possible in highly endemic areas . The risk of exposure through consumption of infected meat , though unlikely , and butchering/handling of rabies virus-infected poultry can potentially pose a risk of transmission of rabies to humans , although never reported to date . | Rabies is a fatal viral disease affecting humans and other animals . Though all warm blooded animals are susceptible to this disease , rabies is commonly observed in mammals . Birds can be experimentally infected with this virus; however , naturally occurring rabies infection in birds has been reported very rarely . We report an unusual case of natural rabies infection in a domestic fowl from India . The bird was bitten by a stray dog and succumbed after a month . The brain tissue from the carcass was tested at a laboratory and was found to be positive for rabies virus antigens . This report indicates that rabies is a disease that can affect birds . Most often birds succumb due to shock or complication of animal bite injury and may not survive until the development of clinical signs of rabies infection . Moreover , fewer opportunities for diagnostic laboratory testing of suspected rabies in a bird may be a reason for the disease in these species being underestimated . Butchering and handling of virus-infected poultry may pose a potential biohazard . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| []
| 2015 | Natural Rabies Infection in a Domestic Fowl (Gallus domesticus): A Report from India |
Evolution depends on the manner in which genetic variation is translated into new phenotypes . There has been much debate about whether organisms might have specific mechanisms for “evolvability , ” which would generate heritable phenotypic variation with adaptive value and could act to enhance the rate of evolution . Capacitor systems , which allow the accumulation of cryptic genetic variation and release it under stressful conditions , might provide such a mechanism . In yeast , the prion [PSI+] exposes a large array of previously hidden genetic variation , and the phenotypes it thereby produces are advantageous roughly 25% of the time . The notion that [PSI+] is a mechanism for evolvability would be strengthened if the frequency of its appearance increased with stress . That is , a system that mediates even the haphazard appearance of new phenotypes , which have a reasonable chance of adaptive value would be beneficial if it were deployed at times when the organism is not well adapted to its environment . In an unbiased , high-throughput , genome-wide screen for factors that modify the frequency of [PSI+] induction , signal transducers and stress response genes were particularly prominent . Furthermore , prion induction increased by as much as 60-fold when cells were exposed to various stressful conditions , such as oxidative stress ( H2O2 ) or high salt concentrations . The severity of stress and the frequency of [PSI+] induction were highly correlated . These findings support the hypothesis that [PSI+] is a mechanism to increase survival in fluctuating environments and might function as a capacitor to promote evolvability .
[PSI+] is an epigenetic modifier of translation termination in yeast [1 , 2] . It is formed by the protein Sup35 , a subunit of the translation termination complex , which carries an intrinsically disordered prion-determining region at its N terminus . When this domain switches to an aggregating amyloid conformation ( the prion conformation ) , much of the protein becomes unavailable for translation termination . Ribosomes therefore begin to read through stop codons an appreciable fraction of the time [3–6] . This creates a host of new phenotypes [7 , 8] , because ribosomal read-through can cause changes in mRNA stability and protein function [9] . [PSI+] generates different phenotypes in different genetic backgrounds as a result of the high levels of sequence variation downstream of stop codons . These phenotypes are heritable because the prion protein is passed through the cytoplasm to progeny , where self-templating conformational change perpetuates it [4 , 5 , 7 , 8] . Remarkably , under growth conditions in which [PSI+] produces new phenotypes , these phenotypes are advantageous as much as ∼25% of the time [7] . Reducing the fidelity of translation termination should generally be deleterious . Not surprisingly , extensive sampling of Saccharomyces cerevisiae from the wild failed to recover any that were in the [PSI+] state [10 , 11] . However , the unusual amino-acid composition of the prion domain of Sup35 ( PD ) has been conserved for over 800 million years of fungal evolution [4 , 12 , 13] . Although the sequence itself is poorly conserved , its ability to switch into a prion conformation and perpetuate that conformation in a stable heritable way [11 , 12 , 14] , as well as the precise mechanism of conformational conversion [15 , 16] and the regulation of its conformations by the protein remodeling factor Hsp104 [17 , 18] , have been conserved over the same period . We have proposed that the transient appearance of [PSI+] provides a mechanism for cells to acquire complex traits in a single step by sampling hidden genetic variation on a genome-wide scale . These complex traits can also be lost in a single step if the environments that favored them disappear , by simple loss of the prion conformation . Alternatively , when the selective pressure is maintained , [PSI+] cells could propagate , allowing time for the traits to be fixed and further enhanced by genetic change [7 , 19] . In addition to increasing survival in fluctuating environments on a short time scale , this remarkable feature of the prion provides a framework for a mechanism for evolvability , which is the capacity of an organism to generate heritable phenotypic variation that has adaptive value [20] . However , the question of whether mechanisms for evolvability may themselves be the result of natural selection remains a hotly contested issue [21–23] . In considering whether [PSI+] may be maintained over such long evolutionary distances because it was occasionally selected for by the evolutionary novelties it produced , it has been noted that there should be a correlation between the appearance of the prion and exposure to the stressful environments in which it is potentially advantageous [24–26] . Under typical laboratory conditions , [PSI+] appears and is lost apparently spontaneously at low frequency ( 10−6–10−7 cells ) [3 , 27 , 28] . Deliberate experimental manipulation of protein chaperones and degradation systems [29–32] can alter the frequency with which [PSI+] appears . However , little is known about the relationship between de novo formation of [PSI+] and stressful environments , except that prolonged storage at 4 °C increases the number of [PSI+] cells [31 , 33] . Here we show that an increased frequency of [PSI+] formation is linked to the cellular stress response , both by conducting an unbiased genome-wide screen for genetic modifiers of prion induction and by testing a large array of stressful growth conditions . We propose that the appearance and loss of [PSI+] is intrinsically linked to perturbations in homeostatic mechanisms , particularly protein folding , simply because prion gain and loss are caused by rearrangements of an aggregation-prone , natively unfolded protein domain and is regulated by the activities of several different protein chaperones [5 , 32 , 34 , 35] .
To conduct an unbiased genome-wide search for factors that influence prion formation , we took advantage of two long-standing observations: First , overexpressing the prion domain increases the frequency of [PSI+] induction [36–38] . This is because the process of prion nucleation involves protein–protein interactions [35 , 39 , 40] . Second , continued overexpression , after cells have switched to the [PSI+] state , is toxic . This is because overexpression drives too much of the essential Sup35 protein into the inactive prion state [36 , 38 , 41] . When the prion domain is overexpressed in [psi–] cells , they will die upon switching to the [PSI+] state ( Figure 1A ) . This creates a toxicity that is proportional to prion induction frequency and therefore provides a means of screening for modifiers of induction frequency . We transformed each of the ∼4 , 700 strains in the S . cerevisiae single gene deletion library [42] ( YGDS ) with a galactose-inducible , plasmid-based expression construct for a prion-domain–yellow fluorescent protein fusion ( PD-YFP ) . When transformants were switched from glucose to galactose , overexpression of PD-YFP caused an intermediate level of toxicity in most of the strains ( Figure 1B , left ) . This allowed us to identify strains that enhanced or reduced toxicity ( Figures 1B and 2A ) . As expected , the deletion of hsp104 or rnq1 , which was previously known to abolish [PSI+] induction [18 , 43 , 44] , eliminated the toxicity of PD-YFP overexpression ( Figures 1B and 2B ) . To eliminate candidate deletion strains that might have enhanced toxicity independently of [PSI+] induction , we transformed them with a plasmid coding for YFP alone under control of the Gal promoter and compared the level of growth inhibition upon shift to galactose medium with that in strains carrying the PD-YFP fusion ( Figure 2B ) . This left 108 strains ( Table S1 ) that exhibited increased toxicity with PD-YFP but not with YFP ( Figure 2B ) . Next we eliminated candidate deletion strains in which the toxicity was reduced independently of changes in the [PSI+] induction frequency , for example , by reducing the copy number of the 2μ plasmid used to induce the prion or by random loss of the prion-induction factor [RNQ+] [38 , 43 , 44] . To do so , we integrated two copies of the galactose-inducible PD-YFP construct into the genome of a [RNQ+] donor strain . These were introduced into the deletion strains by mating , sporulation and selection for the deletion [45] ( Figure 2C ) . After retesting each strain ( Figure 2C ) , we confirmed that 143 gene deletions reduced the toxicity of PD-YFP overexpression ( Table S1 ) . The unexpectedly large number of genes that modify toxicity of PD-YFP overexpression indicates that prion induction frequency is highly responsive , directly or indirectly , to diverse cellular perturbations . To classify the candidate gene deletions , we used the Gene Ontology ( GO ) Slim Mapper program available at the S . cerevisiae genome database ( SGD; http://db . yeastgenome . org/cgi-bin/GO/goSlimMapper . pl ) for functional clustering analysis of the hits from our two groups: strains that enhanced toxicity and those that reduced toxicity of overexpression of PD-YFP . For the reduced toxicity group , the most prominent categories were “cytoskeleton organization and biogenesis” ( 8% of hits versus 3% of library ) and “cell budding” ( 6% of hits versus 1% of library ) ( Figure S1A ) . Several genes in these categories were found to alter the initial formation of prion foci , suggesting that they are involved in the mechanisms of prion formation and inheritance ( JT and SL , unpublished data ) . For the enhanced toxicity group the most prominent categories were “signal transduction” ( 11% of hits versus 4% of library ) , “response to stress” ( 17% of hits versus 8% of library ) and “response to chemical stimulus” ( 18% of hits versus 7% of library ) ( Figure S1B ) . This suggested a relationship between environmental stress and prion induction . Next , we directly tested the deletions that changed toxicity of PD-YFP overexpression , together with some other closely related gene deletions , for effects on the spontaneous induction of [PSI+]; that is , without overexpression of PD-YFP . These deletions were recreated by a site-directed knockout strategy [46] in a different strain background ( 74D-694 [psi–] ) to eliminate strain-specific effects and effects of genetic changes that may have accumulated in the original deletion library since it was created several years ago [47] . Because the rate of spontaneous induction of [PSI+] is so low ( 10−6–10−7 ) , we used a variant strain , R2E2 , carrying a small insertion in the prion domain of Sup35 that increases [PSI+] induction by facilitating conformational conversion in a protein-autonomous manner [28] . Because of the labor involved , we focused on a subset of 40 genes from the best represented categories that appeared to have the most reproducible effects and did not have other confounding phenotypes ( Table S2 ) . Cells carrying a stop codon mutation in the ADE1 gene were selected for the read-through phenotype characteristic of [PSI+] by growth on medium deficient in adenine . An example of the data obtained is shown in Figure 3 , which illustrates the effects of deleting the gene encoding the general stress-response transcription factor Msn2 [48 , 49] . Each genetic knockout was tested repeatedly to obtain induction rates with statistical significance . For each , we tested multiple independent transformants and all deletions were confirmed by colony PCR . Finally , we tested hundreds of colonies that appeared on the [PSI+] selection plates for the presence of bona fide [PSI+] elements by curing their read-through phenotypes with guanidine HCl [50] . Of the 40 deletions tested in this manner , 16 deletions increased [PSI+] induction frequency and 12 decreased it , both in a statistically significant manner ( Student's t-test , p < 0 . 001–0 . 05 , Figure 4 ) . It was previously shown that members of the ubiquitin–proteasome system ( UPS ) , such as Ubp6 , Ubc4 , and Doa4 , influence [PSI+] prion formation; for example , ubc4Δ facilitates spontaneous [PSI+] formation [29 , 31] . Consistent with this , we detected additional members of the UPS that significantly altered [PSI+] induction: the transcription factor Rpn4 , which regulates expression of proteasome genes [51]; the ubiquitin-specific protease Ubp7; the only nonessential 20S subunit Pre9; and Doa1 , which is involved in controlling cellular ubiquitin levels [52] . [PSI+] induction frequency increased after deleting either the kinase Ire1 [53] or the transcription factor Hac1 [53] , which are both regulators of the unfolded protein response ( UPR ) , or Der1 , a protein involved in endoplasmic-reticulum-associated degradation . Our findings also implicate proteins in other stress response pathways , such as phosphate starvation ( Pho5 ) , osmotic shock ( Ssk2 ) , and the general stress response ( Msn2 ) . Consistent with this , Whi2 regulates Msn2 and is required for the full activation of the general stress response [54]—deleting Whi2 also increased [PSI+] induction frequency significantly ( p < 0 . 05 , Figure 4 ) . We note that genes that affect [PSI+] induction frequency were originally identified in the deletion strain library , which carries a wild-type SUP35 gene . However , to confirm the effects of some of the deletions on [PSI+] induction frequency in a strain with neither Sup35 overexpression nor the R2E2 allele , we used a sensitive marker for [PSI+]-mediated translational read-through: a fusion protein with a stop codon in front of a flow-cytometry-optimized green fluorescent protein ( GFP ) marker [55 , 56] . To validate this method , cells emerging from a wild-type [psi–] culture with GFP fluorescence ( suggestive of a switch to the [PSI+] state ) were sorted by fluorescence-activated cell sorting ( FACS ) and plated on nonselective medium , on medium that selects for [PSI+] cells , and on medium that cures cells of [PSI+] ( Figure S2 ) . Surprisingly , we observed a much higher frequency of [PSI+] emergence by this method than expected from previous unsorted plating-based quantifications in our and other laboratories [3 , 27 , 28 , 57] and , indeed , than that obtained by direct plating of these cells . The higher rate of [PSI+] appearance scored by the fluorescent method was confirmed as being due to the bona fide appearance of cells that could give rise to [PSI+] colonies when plated after sorting ( Figure S2 ) . The apparent difference in induction frequency obtained by the two methods may be explained by an unstable , transient [PSI+] transition state that we have recently observed during de novo [PSI+] induction ( JT , J . Dong , M . McCaffery , H . Saibil , B . Bevis , and SL , unpublished data ) . In any case , the effects of the deletion mutations on the induction frequency of [PSI+] were confirmed by this method ( Figure S3 ) . To directly test the relationship between prion emergence and environmental stress responses , we monitored the frequency of [PSI+] formation in wild-type cells under conditions that cause stress: high salt concentrations , oxidative stress , endoplasmic reticulum stress , and high temperatures ( Table S3 ) . Identical aliquots of overnight cultures were diluted to 0 . 25 optical density units and incubated for various times under different test conditions ( or in standard synthetic medium for control ) and then quantified for [PSI+] induction frequency ( Figure 3 ) . For each condition , three to five parallel samples were tested and each experiment was repeated several times . We tested both short intense stresses ( shock ) and longer extended stresses ( 12–24 h ) . Some stresses severely affected cell number , by killing cells or by inhibiting growth . Others did not . None of the short-term exposures showed a significant effect on the frequency of [PSI+] induction ( unpublished data ) . However , longer exposure to each of the different types of stress changed the [PSI+] induction frequency significantly ( p < 0 . 001–0 . 05 ) . When conditions were sorted according to the severity of their effects on cell number ( Figure 5 , orange bars ) , a striking correlation ( p < 0 . 0001 ) was observed between the severity of the stress and [PSI+] induction frequency ( Figure 5 , blue bars ) . For example , 1 M NaCl did not influence growth or [PSI+] induction . Increasing the concentration of NaCl to 1 . 5 M decreased cell number ∼2-fold and [PSI+] induction frequency increased 4-fold . At 2 M NaCl , cell number decreased 10-fold and [PSI+] induction frequency increased ∼10-fold . This trend was independent of the type of stress ( Figure 5 ) . Importantly , [PSI+] induction was not just a consequence of reduced growth , but was associated with stress . Cultures maintained for 24 h in water or medium lacking sugar , which induce a quiescent state , also had low cell densities but had no increased induction of [PSI+] ( Figure 5 , boxed ) . If stress itself is the mechanism that increases switching to the [PSI+] state , it should increase [PSI+] induction irrespective of whether [PSI+] would prove to be advantageous or disadvantageous in that condition . We determined the effect of each condition on the relative growth of [PSI+] and [psi–] cells . At the same time that initial cultures were prepared and tested for [PSI+] induction frequency , we also pre-mixed a known ratio of wild-type [PSI+] and [psi–] cells , exposed them to the same test condition , and measured the ratio of the two cell types afterwards . As expected , in the majority of conditions tested , [PSI+] was either neutral ( Figure 5 , unlabeled ) or disadvantageous ( Figure 5 , single dots ) for growth . In the latter case , in which [PSI+] was at a selective disadvantage , the frequency of its induction was likely even higher . In two conditions , 12 mM DTT and 2 M KCl ( Fig 5 , double dots ) , [PSI+] cells had a significant growth advantage over [psi–] cells . In these cases [PSI+] induction frequencies are likely somewhat inflated .
[PSI+] formation is associated with the acquisition of new phenotypes by the exposure of previously hidden genetic variation [7] . Rather than forming stochastically with a steady low frequency , as previously supposed , we find that the frequency of [PSI+] induction increases with stress . This was first suggested by the results of an unbiased screen of 4 , 700 viable gene deletions for changes in the frequency of [PSI+] induction in a strain carrying a wild-type copy of the SUP35 gene , in which [PSI+] was induced by overexpression of the prion domain . The effects of the deletions were confirmed in a different strain background , which carried a mutation that allowed a direct quantitative measure of spontaneous induction . Finally , [PSI+] induction frequencies were found to be increased by exposure to stressful environments , in a manner that correlated with the severity of the stress . This stunning correlation fulfills a theoretical prediction: if [PSI+] represents a mechanism for evolvability [7 , 19] , its appearance should correlate with stress [23–26] . The inherent logic of the system derives from the fact that [PSI+] is generally detrimental . First , it reduces the fidelity of protein synthesis . Given the many mechanisms that have evolved to ensure the fidelity of protein synthesis , this would be expected to have at least some detrimental effect on fitness , even if that effect is too small to be noticed in most laboratory situations . Second , in the majority of cases the hidden genetic variation that is revealed by [PSI+] is either neutral or deleterious . Thus , under optimal growth conditions , the frequency of switching to the prion state is so low as to have a negligible effect on fitness ( one in ∼106–107 ) . Under conditions of extreme stress , when the expressed phenotype is poorly suited to the environment , an increase in the acquisition of [PSI+] provides an opportunity for cells to alter the phenotypic read-out of their genome , providing an immediate access to complex traits [19] that have an appreciable chance of reaching a new optimum . However , in keeping with the fact that only a fraction of [PSI+]-induced traits are beneficial , most cells in the culture still do not switch . Notably , in earlier studies , some of the same conditions that we found to increase the frequency of [PSI+] ( 2 M KCl or high concentrations of MgAc2 , as well as other types of stress ) were reported to facilitate the loss of [PSI+] [50 , 58] . The same logic applies here . If the expressed phenotype is well suited to the environment , [PSI+] is relatively stable , but if it is not , the frequency of switching increases . Mechanistically , how might this be accomplished ? It was recently reported that challenging the capacity of protein quality control in Caenorhabditis elegans , by overexpressing folding-defective mutant proteins , enhances the aggregation of other glutamine-rich amyloid-forming proteins , such as a polyglutamine-expanded fragment of huntingtin [59] . By analogy , challenges to protein homeostasis by severe stress may stabilize partially folded or misfolded Sup35 molecules , as well as other glutamine-rich proteins [43 , 44 , 60] , and facilitate nucleated conformational conversion to the prion state [35] . Furthermore , a complex web of interactions between Sup35 and the chaperone network of the cell controls the assembly and disassembly of the prion amyloid in vitro ( [4 , 32 , 61] , ) . The complex and varied nature of these interactions can plausibly explain increased switching to [psi–] in cells that are [PSI+] and vice versa . Consider , for example , the stress-induced heat shock protein Hsp104 , a protein remodeling factor that conformationally remodels the [PSI+] determinant Sup35 via several mechanistically distinct mechanisms . High concentrations of Hsp104 , which occur with stress , promote disassembly of Sup35 prions , whereas ADP , whose concentration rises during stress , reduces the inhibition of prion nucleation by Hsp104 [4 , 18] . These interactions represent an intrinsic mechanism by which the state of protein homeostasis increases the likelihood that cells will switch from [psi–] to [PSI+] , or from [PSI+] to [psi–] , when the cell is not well adapted to its environment . But how might the prion have appeared in the first place ? The extreme enrichment of the prion domain in glutamine residues suggests that it may simply have arisen as a polyglutamine expansion , an event that occurs frequently in eukaryotic genomes [62] . Simple stretches of polyglutamine can provide new functions [63] but can be toxic in their own right if they expand too far [64] , and this would have created selective pressure for the sequence to diverge away from simple polyglutamine while preserving other possible functions , as suggested by evidence of purifying selection operating on the prion-forming domain of SUP35 [12] . Together , these observations provide a credible conceptual framework for the evolution of a system for enhancing evolvability . Other systems for evolvability may include mechanisms that increase sex and recombination , and thereby variation , in response to stress [65] . The ability of the molecular chaperone Hsp90 to buffer hidden genetic variation and release it in response to stressful environments is another example [22 , 23 , 66] . These , too , may not initially have appeared because of their effects on evolvability per se , but may simply have arisen as an inherent consequence of effects of stress on genome stability and protein homeostasis .
We used the BY4741 deletion set provided by EUROSCARF [42] , 74D-694 [18] , and a 74D-694-R2E2 variant [28] . The media were complete standard synthetic medium or medium lacking particular amino acids and containing either d-glucose ( SD ) or d-galactose ( SGal ) as sole carbon source . Sporulation was performed in liquid rich sporulation medium ( 1% potassium acetate , 0 . 05% dextrose , 0 . 1% yeast extract , and 0 . 01% complete amino acid mix , Bio101 ) Primer sequences to create knockouts were from the yeast deletion project ( http://www-sequence . stanford . edu/group/yeast_deletion_project/deletions3 . html ) . Knockout cassettes were generated by PCR using the kanMX4 plasmid pFA6a [46] and individual knockout primers . Yeast were transformed with purified PCR products using lithium acetate . The knockouts were confirmed using the KanB primer [46] , which primes in the kanMX4 cassette , together with individual 20–22mers ∼700 bp upstream of the start codons of the candidate genes . Cultures were grown as described in each figure . An aliquot was plated onto SD- plates lacking adenine ( SD-Ade ) and incubated for 7–9 d at 30 °C , whereas a 1:500 dilution of the culture was plated onto complete SD plates for total cell number determination and incubated for 3 d . Non-curable colonies were excluded before calculating [PSI+] induction frequency ( ratio of [PSI+] colonies to total cell number ) as described in Figure 3 . High-throughput transformations were done in a 96-well format [67] . Transformants were selected for 3 d at 30°C on SD plates lacking uracil ( SD-Ura ) . Growth effects of PD-YFP overexpression were determined by replica plating on SGal-Ura and incubation for 3 d at 30 °C . Effects from the deletion itself were excluded by plating on SD-Ura in parallel . Candidate deletion strains were re-arrayed in two different sets in 96-well plates: increased and decreased toxicity of PD-YFP overexpression . To retest deletion strains with decreased toxicity ( using systematic gene analysis , SGA [45] ) , we mated a haploid Mat-alpha [RNQ+] donor strain containing two genomic copies of galactose inducible PD-YFP with the deletion library strains of interest in 96-well format . After pinning the two strains together on YPD plates and incubating them overnight at 30 °C , we spotted the cells four consecutive times on SD medium containing 200 μg/ml geneticin ( for selection of kanMX cassette of the deletion strain ) to select for diploids . Diploids were subjected to sporulation in liquid medium and incubated for 6 d at 23 °C . Haploid spores of interest were selected for by spotting onto selective SD medium three consecutive times [45] and incubation for 3 d at 30 °C each time . PD-YFP expression was induced by spotting onto selective SGal plates and on selective SD plates as control and incubated at 30 °C for 3–4 d . To retest deletion strains with increased toxicity , cells were transformed with a 2μ plasmid coding for PD-YFP or just YFP under control of the galactose promoter . To induce expression , cells were pinned onto synthetic selective SGal medium and selective SD medium as a control and incubated at 30 °C for 3–4 d , respectively . Calculated [PSI+] induction frequencies were log transformed to improve normality . Multiple linear regression analyses were performed with experiment and genotype ( Figure 4 ) or experiment , condition , and time ( Figure 5 ) as fixed effects . Significance ( α < 0 . 05 ) was determined by a two-tailed Student t-test . Exact p-values were calculated from the t-ratio of each effect . Effects were normalized to control ( wild-type strain or untreated sample ) to demonstrate fold increase or decrease . | One controversy in evolutionary biology concerns whether there might be plausible explanations for the rapid evolution of complex traits . An extreme and fascinating example of protein conformational change , the prion , offers a framework for this concept . Prion proteins are responsible for neurodegenerative diseases , instruct us in important aspects of amyloid formation , and furthermore , serve as ancient protein-based units of inheritance , a domain previously reserved for nucleic acids . In yeast , the [PSI+] prion causes read-through of nonsense codons . This has the capacity to rapidly unveil hidden genetic variation that may have adaptive value . The suggestion that [PSI+] might serve as a mechanism for evolvability would be strengthened if the frequency of the prion's appearance increased when the organism was under stress and therefore not ideally adapted to its environment . We investigated genetic and environmental factors that could modify the frequency with which the prion appears . Our high-throughput , genome-wide screen identified genes involved in stress response and signal transduction , whereas our cell-based assays found severe conditions that increased prion formation . Thus [PSI+] provides a possible mechanism for the organism to rapidly acquire new phenotypes in times of stress and potentially increases evolvability . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"cell",
"biology",
"neurological",
"disorders",
"evolutionary",
"biology"
]
| 2008 | Prion Switching in Response to Environmental Stress |
An object in the peripheral visual field is more difficult to recognize when surrounded by other objects . This phenomenon is called “crowding” . Crowding places a fundamental constraint on human vision that limits performance on numerous tasks . It has been suggested that crowding results from spatial feature integration necessary for object recognition . However , in the absence of convincing models , this theory has remained controversial . Here , we present a quantitative and physiologically plausible model for spatial integration of orientation signals , based on the principles of population coding . Using simulations , we demonstrate that this model coherently accounts for fundamental properties of crowding , including critical spacing , “compulsory averaging” , and a foveal-peripheral anisotropy . Moreover , we show that the model predicts increased responses to correlated visual stimuli . Altogether , these results suggest that crowding has little immediate bearing on object recognition but is a by-product of a general , elementary integration mechanism in early vision aimed at improving signal quality .
Since Korte [1] originally described perceptual phenomena of reading in peripheral vision , a substantial number of studies have shown the important role of spacing for object recognition . The phenomenon that an object becomes more difficult to recognize when surrounded by other objects is now popularly known as ‘crowding’ [2] ( see [3] , [4] for two recent reviews ) . The strength of the crowding effect depends on the spacing between objects ( Figure 1 ) . The largest spacing at which there is a measurable effect is commonly referred to as the ‘critical spacing’ . An important and often replicated finding is that the critical spacing for object recognition is proportional to the viewing eccentricity [5] . Moreover , critical spacing is found to be highly invariant to a great variety of stimulus manipulations , such as contrast and size [6]–[8] . Critical spacing is the most extensively studied crowding property and , because of its robustness , now sometimes considered the defining property of crowding [3] . Crowding is a general phenomenon in vision . It is not confined to letter and shape recognition , but affects a broad range of stimuli and tasks , including the identification of orientation [9]–[11] , object size , hue and saturation of colors [12] , recognition of faces [13] , [14] , reading [15] , and visual search [16]–[18] . Altogether , crowding emerges as a fundamental limiting factor in vision , making the question about its neural basis and functional origin rather pressing . Several theories have been proposed to explain the crowding effect [4] , [19] . Currently , there is a growing consensus that crowding results from feature integration over an area that is larger than the target object [4] . However , there is a marked controversy about both the underlying mechanism and the functional origin of the effect . Some authors assert the existence of bottom-up hardwired integration fields ( e . g . , [3] ) , while others claim that feature integration arises from limitations related to the spatial resolution of attention ( e . g . [20] , [21] ) . Postulated functions of feature integration include texture perception [10] , contour integration [22] , and object recognition [3] , [23] . In the absence of quantitative , biologically motivated models , however , it is not clear whether these theories can also quantitatively account for the ‘mysteries of crowding’ [4] , and how plausible they are from a biological perspective . Here , we present a quantitative model for spatial integration of orientation signals . Our model is based on the principles of population coding [24] , which is an approach that mathematically formalizes the idea that information is encoded in the brain by populations of cells , rather than by single cells . Motivated by findings from physiological [25] , [26] and theoretical [27] studies , we model feature integration as a ( weighted ) summation of population codes . Using simulations , we demonstrate that this approach allows to explain several fundamental crowding properties in a single , unified model , including aspects of critical spacing [6] , [15] , compulsory averaging of crowded orientation signals [10] , and an asymmetry between the effects of foveally and peripherally placed flankers [28] , [29] . Moreover , we show that the model predicts enhancement of signals that encode visual contours , which could facilitate subsequent contour detection and segmentation and adds support to earlier findings about a link between crowding and contour integration . Altogether , our main finding is that feature integration , implemented in a neurophysiologically plausible way , produces crowding as a by-product . Furthermore , our results add support to an earlier suggested link between crowding and contour integration , and they point at V4 as a likely locus for feature integration cells ( at least for the orientation domain ) .
Several different population coding schemes have been proposed in the literature [30] . Although they differ in their details , the general idea behind all of them is that variables are encoded in the brain by entire populations of cells . Our model is based on the ‘distributional population coding’ ( DPC ) scheme that was proposed by Zemel et al . [31] . In this scheme , a population code explicitly encodes a probability distribution over the stimulus domain . In this section we will only provide a general overview of our model . Mathematical details can be found in the Methods section . The input to the model consists of a set of stimuli , each one defined by a location , orientation , contrast , and size ( Figure 2a ) . The first layer of the model represents full probability distributions over the input stimuli . These distributions are assumed to be Gaussian , with a width that depends on the eccentricity , contrast , and size of the stimuli ( Figure 2b ) . Subsequently , these probability distributions are used as inputs to the DPC encoder that computes a population code representation for each of the stimuli ( Figure 2c ) . The properties of the cells ( e . g . , tuning width ) in the first layer are chosen such that they closely resemble V1 simple cells ( see Methods for parameter values ) . In the second layer , stimulus representations from the first layer are spatially integrated , in the form of weighted summations of cell responses ( Figure 2d ) . The integration weights depend on the cortical distance in primary visual cortex between the locations of the ‘integration cell’ and the cells encoding the input stimuli ( for details about the weight function and mapping of visual field to cortical locations , see Methods ) . This function can be interpreted as defining a cortical ‘integration field’ . The size and shape of these integration fields can be thought of as representing the arborization of the dendritic tree , i . e . , the distribution of lateral connections of a physiological integration cell . The weight function is a 2D Gaussian , thus reflecting that there are many short-range connections and fewer long-range connections . Unlike the first layer , which is a simulation of V1 simple cells , it is currently difficult to link the cells from the second layer to a very specific cortical area . Nonetheless , if we compare the predictions that follow from optimization of our model parameters to the current physiological literature , then we find V4 to be a likely candidate . We come back to this in the discussion section . Several of the simulation experiments that we conducted required that a response be generated ( e . g . , when simulating psychophysical experiments involving target tilt estimation ) . In those simulations , a maximum-likelihood decoder was used to decode the post-integration population code associated with the target position back to a stimulus distribution ( Figure 2e ) . The number of components of the returned mixture model was interpreted as the number of distinct orientations perceived at the location associated with the decoded population code , the mixing proportions as the amounts of evidence for the presence of an orientation , the means as estimates of these orientations , and the standard deviations as the amounts of uncertainty about these estimates . A well-established behavioral finding in human observers is that identification thresholds for a crowded target decrease as a function of target-flanker spacing until a certain critical spacing is reached . Beyond this critical spacing flankers no longer have an effect ( see , for example , the results shown in Figure 1 ) . In our model , the integration fields are implemented as weight functions of stimulus spacing in cortex . Consequently , flanker stimuli affect the identification of a target only when positioned within a certain distance from the target , yielding a critical region for target identification . To examine whether our model can quantitatively account for critical regions found for human subjects , we performed a simulation that mimicked the psychophysical experiment by Pelli et al . [15] , who estimated critical regions for letter identification at several positions in the visual field . Critical regions predicted by our model were estimated as follows . For each target position , identification thresholds were determined for a range of target-flanker spacings ( see Figures 3a and 3b; we refer to Methods for details about the procedure that was used to estimate identification thresholds ) . A ‘clipped line’ was fit to the resulting data , providing an estimate of the critical spacing ( Figure 3c ) . By varying the positions of the flankers , we estimated critical spacing in several directions around the target . Combining these spacings gives an estimate of the critical region around a given target location ( Figure 3d ) . We estimated model parameter values that result in a good model fit to one of the critical regions measured by Pelli et al . Subsequently , we repeated the experiment for the other target locations using the same parameter values , and found that the model accurately predicts all reported human critical regions ( Figure 3d ) . These results thus provide quantitative evidence for the suggestion that the behavioral crowding regions found in humans can be explained as the result of fixed-sized , hard-wired integration fields in visual cortex . The critical spacing for crowding is known to scale with eccentricity and is consistently found to be in the range 0 . 3–0 . 6 times the target eccentricity [6] . Moreover , it is found to be largely invariant under changes to the physical properties of the stimulus , such as the size , contrast , and number of flankers [6] and the ‘scaling’ of stimuli ( i . e . , changing the size of both the target and flankers ) [6]–[8] . To further verify our model , we conducted another series of simulation experiments , in which we manipulated several stimulus properties . We found that the results are compatible with findings in human subjects: critical spacing predicted by our model scales linearly with target eccentricity and is hardly affected by stimulus manipulations ( Figure 4 ) . Human observers are able to report the mean orientation of a set of crowded stimuli , but not the orientations of the individual stimuli [10] . This peculiar crowding property is generally referred to as ‘compulsory averaging’ . In the experiment of Parkes et al . , observers reported the tilt direction of a variable number of equally tilted targets positioned among horizontal flankers . Parkes et al . found that a relatively simple pooling model could account for human data when the total number of stimuli is kept constant . However , when targets are presented without flankers , identification thresholds dropped significantly slower as a function of the number of targets than predicted by their model ( Figure 5b ) . They postulated a ‘late noise’ factor to explain the discrepancy between data and model . Our model suggests the following explanation for the compulsory averaging phenomenon . When two features are highly similar , their population code representations have a high degree of overlap and will merge when summed . Consequently , the resulting post-integration code will be interpreted as representing a single feature with a value somewhere in between the values of the input stimuli ( Figure 5a ) . To examine whether our model can also quantitatively account for compulsory averaging , we conducted a simulation experiment with conditions and stimuli similar to those used in the psychophysical experiment performed by Parkes et al . [10] . The results show that our model produces accurate fits to the psychophysical data for both the condition with and without flankers ( Figure 5b ) . An important difference between our model and the pooling model proposed by Parkes et al . is that the latter integrates all stimuli with equal weight , while integration in our model is weighted by object spacing . To verify the relevance of this aspect in explaining why the models make different predictions , we reran the simulations with varying stimulus spacing ( see Text S1 and Figure S3 for results ) . We found that when we set all integration weights in our model to one ( implying an object spacing of zero ) , the identification thresholds predicted by our model are similar to those predicted by the pooling model of Parkes et al . Additionally , the predictions of the models increasingly diverge when object spacing is increased . These results confirm that object-spacing related weighting of integration is an essential difference between the models . Moreover , they challenge the need for the ‘late noise’ factor proposed by Parkes et al . to explain their results . Several studies [5] , [29] have found that , with equal target-flanker spacing , flankers positioned at the peripheral side of a target cause stronger crowding effects than flankers positioned at the foveal side . As has been noted previously [16] , this asymmetry follows directly from the way that the visual field is mapped onto the cortex . With increasing eccentricity , the representation of visual space becomes more and more compressed . Consequently , for equal target-flanker spacing in visual space , the cortical distance between the representation of a target and a foveal flanker is larger than that between a target and a peripheral flanker . Assuming that cortical integration fields are isotropic , peripheral flankers will , therefore , contribute more to the integrated target signal than foveal flankers . We conducted a simulation experiment to verify whether our model replicates the foveal-peripheral anisotropy and to investigate how its predictions depend on target-flanker spacing . For several target-flanker spacings , we estimated 75%-correct target contrast thresholds for identifying the tilt of a target without a flanker , a target with a foveal flanker , and a target with a peripheral flanker ( Figure 6a ) . The results show that while both the foveal and peripheral flanker produce crowding ( Figure 6b ) , the effect caused by a peripheral flanker is substantially larger than that caused by a foveal flanker ( Figure 6c ) . Hence , our model exhibits a foveal-peripheral flanker anisotropy . Furthermore , the model predicts the anisotropy to be strongest at intermediate spacings while it predicts no anisotropy when target-flanker spacing is very small or approaches the critical spacing ( Figure 6d ) . In these simulation data , the strongest anisotropy is found when target-flanker spacing is about 2 degrees ( i . e . , about 0 . 3 times the target eccentricity ) . At this spacing , threshold elevation caused by the peripheral flanker is predicted to be approximately 2 . 5 times that caused by the foveal flanker . This is comparable to the effect size measured for human observers [29] . The results so far suggest that crowding is what happens when signals from closely-spaced , unrelated stimuli are integrated with each other . However , in normal viewing conditions , signals from closely-spaced stimuli are often correlated ( e . g . , neighboring line segments of an edge or smooth contour ) . It has been suggested that integration of such correlated ( orientation ) signals may underlie phenomena such as contour integration [25] , [32] , [33] . To see how our model responds to signals from correlated stimuli , we ran a simulation with an input stimulus consisting of a set of line segments comprising various contours within a noisy background ( see Methods for details ) . The results are shown in Figure 7 . Line segments that are part of a contour clearly stand out in the post-integration representation . This is because both stimulus density and orientation correlation are higher for contours than for the random background . This result supports an earlier suggested link between contour integration and crowding [22] , but firm conclusions would require further extensive evaluation . Note that in areas away from fixation , in the periphery of the visual field , the decoder often returned stimulus distributions that represent more than one orientation value . This indicates that the post-integration codes at those locations are ambiguous in terms of the encoded orientation . In other words , when stimulus spacing is small relative to eccentricity , stimuli become jumbled with their neighbors , just as observed in crowding .
The cells in the first layer are modeled after V1 simple cells ( see Methods for accompanying parameter values ) . However , there is currently no agreement about the cortical locus of the ‘integration cells’ that are supposed to underlie crowding . Therefore , we decided to make minimum assumptions about their physiological origin . Consequently , the size and shape of their receptive fields , determined by σrad and σtan ( see Methods ) , were taken to be free parameters , such that the parameter values that provide a good fit to experimental data can be considered a prediction for the receptive field properties of the integration cells underlying crowding . We found that the best model fit to the data is obtained with integration cells that are strikingly similar to a type of cell that has recently been identified in V4 ( of cat and monkey ) [34] , [35] . The function of these cells is currently unknown [36] . Hence , we speculate that these V4 cells spatially integrate information from V1 ( either directly or mediated by V2 ) . Their possible function may be contour integration ( e . g . , as a precursor for shape coding ) , with crowding as a by-product . Interestingly , other , independent , lines of evidence also have suggested that crowding occurs beyond V1 [21] , [37] with V4 as a likely candidate area [38] . The parameter settings ( see Methods ) in our model were fixed over the entire range of simulations that we performed , with one minor exception ( see Figure 3 ) . We reran a number of simulations with different parameter values and found that this hardly affected our results ( see Text S1 and Figure S4 for details ) . This suggests that crowding is an inherent property of a mechanism that integrates signals by summing population codes . These results shed new light on earlier proposed crowding theories . Some authors have proposed that crowding is , at least in part , the result of ‘source confusion’ due to positional uncertainty [39] , [40] . We would like to note , however , that integrating signals over space necessarily increases positional uncertainty . Hence , we consider location uncertainty and , consequently , ‘source confusion’ a result of feature integration , rather than an additional factor in the explanation of crowding . Indeed , our results show clear evidence for ‘source confusion’ , even though we did not explicitly incorporate positional uncertainty into our model ( for an example , see Figure 5a ) . When spatially averaging signals in a retinotopically arranged ‘feature map’ ( such as V1 ) , activation patterns that are caused by closely spaced stimuli may slightly shift towards each other ( or even completely merge together , if spacing is very small ) . As a result , an averaging of stimulus positions may be perceived in such situations . In a recent paper it was shown that judgments of the position of a crowded target object are systematically biased towards the positions of flanking objects [41] . The authors of that paper explained their results by a model that averages stimulus positions . Based on the foregoing argument , their results can presumably just as well be explained as a result of averaging feature signals over space . A recent theory suggests that crowding is the ‘breakdown of object recognition’ [3] . The reasoning is that spatial integration of object features ( in the notion of ‘binding’ ) is required for object recognition , whereas crowding occurs when multiple objects fall within the same integration field . Our results indicate that the spatial signal integration underlying crowding may enhance responses for correlated signals , such as contours . This corroborates an earlier suggestion that the ‘association fields’ that have been proposed to underlie contour integration [42] may also cause crowding [22] . While such enhancement of responses to correlated signals will no doubt facilitate higher-order functions such as object recognition , integration appears to have a more elementary and general function . Other authors argue that crowding is the result of attentional limitations [20] , [21] , although evidence for these theories is considered very slim [4] . While we deem it possible that attentional factors have modulatory effects on crowding , our present results show that the general properties of crowding can very well be accounted for without invoking attentional mechanisms . It has also been suggested that crowding is ‘texture perception when we do not wish it to occur’ [10] . The motivation behind this proposal is the finding that observers cannot identify individual stimulus properties in a crowded display , but still have access to its average statistics ( i . e . , its texture properties ) . Our model is able to explain this finding ( see Figure 5 ) , and we agree that what occurs after pooling can be described as ‘texture perception’ . However , in view of the plausible connection between spatial integration and contour integration , we hesitate to conclude that texture perception is the primary function of spatial integration . Moreover , if a functional link exists between spatial integration and texture perception , then we deem it just as likely that integration serves to compress visual information , in order to reduce energy requirements at higher levels of processing . Two crowding properties that our current model does not account for are the effects of ‘target-flanker similarity’ and ‘flanker configuration’ . The former refers to the finding that crowding is stronger for target-like flankers compared to dissimilar flankers [9] , [43] , [44] . The ‘flanker configuration’ effect refers to the finding that crowding is partially ‘released’ when surrounding flankers form a contour [45] , [46] . A rather natural extension to our model may allow it to account for these two effects as well . At present , the integration fields in our model represent exclusively excitatory horizontal connections between cells . Alongside these excitatory connections , however , many of the cells in primary visual cortex are known to have inhibitory connections as well as feedback connections from higher-order brain areas [47] . Inhibition could reduce the integration of dissimilar pieces of information and thus be responsible for target-flanker similarity effects in crowding . Likewise , the feedback connections might inhibit the integration of signals that are likely to represent different objects or ‘perceptual groups’ and , therefore , be responsible for configuration influences on crowding . The model and simulations that were presented in this paper are limited to the orientation domain . However , crowding is a rather general phenomenon that affects a large number of tasks , including discrimination of letters and objects sizes , colors , and shapes . Since population coding is considered the general way by which variables are encoded in the brain [24] , crowding of other basic features such as size and color [12] can presumably be explained by a model that is largely analogous to the one presented here . Moreover , if population coding is also used to encode more complex information , and spatial integration takes place at many different levels of processing , then our model predicts that crowding should also be found at many different levels . Hence , crowding of more complex structures ( such as letters , object shapes , bodies , and faces ) could follow both from crowding in their constituent features and from crowding within higher-order population codes that represent the structures themselves [48] . Our model licenses a number of predictions that can be tested experimentally . For example , the simulations related to the ‘compulsory averaging’ effect predicts the effects of stimulus spacing and contrast on identification thresholds . Additionally , the model makes quantitative predictions regarding the effect of spacing on the foveal-peripheral flanker anisotropy of crowding . Finally , the model makes predictions about the receptive field properties of the integration cells responsible for crowding , The results that we presented here lend strong quantitative support to the theory that the mechanism behind crowding is spatial feature integration , and our model provides a computationally motivated physiological basis to this theory .
Input stimuli are specified as 4-tuples , where is the orientation , the size , the location , and the ( relative ) contrast of the stimulus . For each of these inputs we first define a corresponding probability distribution , which is subsequently used as input to the distributional population coding scheme of Zemel et al . [31] . The width of an input distribution represents the perceptual uncertainty about a stimulus and is related to stimulus eccentricity , size , and contrast c , in the following way ( see Text S1 for motivation ) : ( 1 ) In order to account for the circularity of the orientation domain , we define these distributions to be circular normal ( von Mises ) distributions . More specifically , the distribution over orientation s for a stimulus is defined as: ( 2 ) where is the modified Bessel function of order 0 , is an inverse measure of statistical dispersion , and is a value drawn from the normal distribution over s . In the simulation experiments we map the stimulus domain [−90 , 90 ) deg to [−π , π ) . The tuning curves fi ( s ) of the cells are defined as circular normal functions over s: ( 3 ) where si is the preferred orientation of cell i , the width of the tuning curves , and an S-shaped function that defines how cell gain relates to the contrast c and size α of a stimulus ( see Text S1 and Figure S1 ) . Following the DPC scheme , we compute the average response of cell i to a stimulus as follows: ( 4 ) where is the level of spontaneous activity and drawn from a normal distribution with mean and a standard deviation . In order to evaluate this integral numerically , we approximate the input distributions by histograms and the tuning functions by histograms , both with bin centres linearly spaced in the range . Hence , we can rewrite equation ( 4 ) to ( 5 ) A population code representing a stimulus is constructed by drawing responses rhi from Poisson distributions ( 6 ) The second layer of the model spatially integrates the stimulus representations in the first layer . The layer-2 population code that is associated with position is computed as a weighted sum over the population code representations of all N input stimuli: ( 7 ) where is a 2D Gaussian weight function that represents the cortical integration fields ( see Text S1 and Figure S2 for details ) . Several of our simulation experiments require that a task response is generated . In those experiments , a Bayesian decoder is used to estimate the stimulus probability distribution that is encoded in the post-integration population code associated with the target position . Subsequently , the orientation with the highest probability is interpreted as representing the most likely orientation of the target , and chosen for response . We use the Bayesian Information Criterion to choose the most likely mixture model among a set of models with 1 , 2 , and 3 mixture components . We refer to the Text S1for all mathematical details of the decoder . The parameter settings of the model were as follows . In all simulations , the width of the tuning curves was set to , the number of neurons J comprising one population code was set to 90 , the spontaneous firing rate rbase was set to 5 spikes/s , and the maximum firing rate was set to 90 spikes/s . The only parameters that varied between simulations were σrad and σtan , which determine the integration field width in the ‘radial’ and ‘tangential’ direction , respectively ( see Text S1 ) . These were set to 2 . 5 and 1 . 0 mm , respectively , in all simulations , except the one in which we estimated critical regions ( Figure 3 ) , where the values were set to 1 . 6 and 1 . 1mm , respectively . This difference is motivated by the observation that the human data in Figure 3 are from a subject with an unusually small critical spacing ( approximately 0 . 3 times the target eccentricity ) . Several simulation experiments involved estimation of target contrast thresholds for a tilt identification task . In those experiments , the procedure on a single trial was as follows . The target and flanker stimuli were encoded and their representations integrated , as described above . Subsequently , the post-integration population code associated with the target position was decoded to a mixture of normal distributions . The sign of the orientation associated with the peak location in the returned probability distribution was compared with the sign of the input target . Performance was considered ‘correct’ if the signs were the same , and ‘incorrect’ otherwise . Performance estimates were made for several target contrasts , by simulating 50 trials for each contrast . Finally , a sigmoid function with a mean a and a width b: ( 8 ) was fit to these data , in order to obtain an estimate of the target contrast that yields 75%-correct performance ( see Figure 3b for an example ) . In the simulation experiments that estimated critical spacing , the above procedure was repeated to obtain 75%-correct thresholds for several target-flanker spacings . A ‘clipped line’ was fit to these thresholds in order to estimate critical spacing ( see Figure 3c for an example ) . Input stimuli consisted of a ±10° tilted target and two 30° tilted flankers , positioned at opposite sides of the target . Flanker contrast and the size of both the target and flankers were set to 1 . Using the procedure described above , critical spacing was estimated for the same target and flanker positions as in the psychophysical experiment by Pelli et al . [15] . The input stimuli consisted of a ±10° tilted target , one −30° tilted flanker , and one +30° tilted flanker . Flanker contrast and the size of both the target and flankers were set to 1 . Critical spacing was determined for flankers positioned along the radial axis , on opposite sides of the target . In the first simulation ( Figure 5a ) , input stimuli consisted of a 0° tilted target and two flankers with 10° tilt in the first condition and 50° tilt in the second condition . The target was positioned at 2 . 5 deg of eccentricity . The flankers were positioned on opposite sides of the target , with a spacing of 0 . 5 deg of eccentricity . The contrast and size of all stimuli were set to 1 . Stimuli used in the second simulation ( Figure 5b ) were similar to those used in the psychophysical experiment by Parkes et al . [10]: N tilted targets and 9-N vertical flankers ( first condition ) or no vertical flankers ( second condition ) , with a central position of 2 . 5 deg of eccentricity and a spacing of 0 . 5 deg between the central stimulus and surrounding stimuli . The contrast and size of the stimuli were set to 0 . 5 . On a single trial , the post-integration population code associated with the central stimulus position was decoded to a unimodal stimulus distribution . The sign of the orientation with the highest probability was compared with the sign of the target . If they were the same , performance on that trial was considered correct . We measured performance over 100 trials for varying target tilts . Based on these data , 75%-correct performance thresholds were determined . This procedure was repeated for different values of N . Input stimuli consisted of a ±10° tilted target without a flanker ( condition 1 ) , with a 30° tilted foveal flanker ( condition 2 ) , or a 30° tilted foveal flanker ( condition 3 ) . Flanker contrast and the size of both the target and flankers were set to 1 . For all three conditions , 75%-correct target contrasts were estimated for a range of target-flanker spacings . Threshold elevations TEfoveal and TEperipheral were defined as described in the main text . The input stimuli consisted of a set of oriented bars , comprising three contours within a field of randomly oriented bars . The circle contour consisted of 35 equally spaced segments , was centered at ( 0 , 10 ) degrees of eccentricity and had a radius of 4 degrees of visual angle . The other four contours consisted of 23 line segments each , with a spacing of 0 . 7 degrees of visual angle between every two neighboring segments . The randomly oriented line segments were placed on a grid with a radius of 18 degrees of eccentricity and a grid spacing of 2 . 0 degrees . The contrast and size of all line segments was set to 0 . 8 . | Visual crowding refers to the phenomenon that objects become more difficult to recognize when other objects surround them . Recently there has been an explosion of studies on crowding , driven , in part , by the belief that understanding crowding will help to understand a range of visual behaviours , including object recognition , visual search , reading , and texture recognition . Given the long-standing interest in the topic and its relevance for a wide range of research fields , it is quite surprising that after nearly a century of research the mechanisms underlying crowding are still as poorly understood as they are today . A nearly complete lack of quantitative models seems to be one of the main reasons for this . Here , we present a mathematical , biologically motivated model of feature integration at the level of neuron populations . Using simulations , we demonstrate that several fundamental properties of the crowding effect can be explained as the by-product of an integration mechanism that may have a function in contour integration . Altogether , these results help differentiate between earlier theories about both the neural and functional origin of crowding . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"neuroscience/sensory",
"systems",
"computational",
"biology/computational",
"neuroscience"
]
| 2010 | A Neurophysiologically Plausible Population Code Model for Feature Integration Explains Visual Crowding |
In recent years , the primate malaria Plasmodium knowlesi has emerged in human populations throughout South East Asia , with the largest hotspot being in Sabah , Malaysian Borneo . Control efforts are hindered by limited knowledge of where and when people get exposed to mosquito vectors . It is assumed that exposure occurs primarily when people are working in forest areas , but the role of other potential exposure routes ( including domestic or peri-domestic transmission ) has not been thoroughly investigated . We integrated entomological surveillance within a comprehensive case-control study occurring within a large hotspot of transmission in Sabah , Malaysia . Mosquitoes were collected at 28 pairs households composed of one where an occupant had a confirmed P . knowlesi infection within the preceding 3 weeks ( “case” ) and an associated “control” where no infection was reported . Human landing catches were conducted to measure the number and diversity of mosquitoes host seeking inside houses and in the surrounding peri-domestic ( outdoors but around the household ) areas . The predominant malaria vector species was Anopheles balabacensis , most of which were caught outdoors in the early evening ( 6pm - 9pm ) . It was significantly more abundant in the peri-domestic area than inside houses ( 5 . 5-fold ) , and also higher at case than control households ( 0 . 28±0 . 194 vs 0 . 17±0 . 127 , p<0 . 001 ) . Ten out of 641 An . balabacensis tested were positive for simian malaria parasites , but none for P . knowlesi . This study shows there is a possibility that humans can be exposed to P . knowlesi infection around their homes . The vector is highly exophagic and few were caught indoors indicating interventions using bednets inside households may have relatively little impact .
The success story of reducing malaria worldwide [1] has been marred by a few notable exceptions where bulk of disease is caused by zoonotic “neglected” malaria species with atypical transmission that makes them less easy to control . Zoonotic malaria , such as Plasmodium knowlesi from the long tailed macaque ( Macaca fascicularis ) in SE Asia [2] , and P . brasilanum from NewWorld monkeys in South America [3] , are a growing public health problem . In South East Asia , the long tailed macaque harbours at least five simian malarias , namely , P . coatneyi , P . inui , P . fieldi , P . cynomolgi and P . knowlesi [4 , 5] . Plasmodium knowlesi is presently the main zoonotic malaria with the greatest public health importance , especially in Sabah , Malaysian Borneo which has recorded the highest growing number of P . knowlesi cases in humans , and most of these cases are clustered within one district , Kudat in the North eastern region [6 , 7] . Plasmodium knowlesi is morphologically similar to P . malariae and had been misdiagnosed as such for a long time [2] . A first case of naturally acquired human infection with P . cynomolgi has also been reported from peninsular Malaysia [8] . Thus it is a possibility that other primate parasites may also be soon contributing to human cases as previously predicted [9] . In Sabah , it has been confirmed that An . balabacensis is the primary vector [10] and the long tail macaques , which are the natural reservoir hosts for simian malaria parasites are present . Furthermore , Kudat district has many secondary forest areas surrounded by hilly areas , oil palm estates and rubber plantations which in general serve as habitats not only for long-tail macaques but also for Anopheles species . Close interaction between monkeys , mosquitoes and human increases the chances of being infected with P . knowlesi . The vectors of P . knowlesi malaria in Malaysia comprise of five Anopheles species of the Leucospyrus group namely: An . hackeri , An . latens , An . cracens , An . introlatus and An . balabacensis [10–15] . In Vietnam Anopheles dirus of the Dirus group was recorded as the vector [16 , 17] . These vectors are found mainly in the forests and are outdoor biters , and likely to have low susceptibility to frontline control strategies which typically involve use of insecticides in homes . In Malaysia , the National Malaria Eradication Program was launched in 1967 , followed by state-wide malaria control programs during the 1970s and 1980s . Consequently , great reductions in malaria prevalence were recorded , from 240 , 000 in 1961 to around 50 , 000/year during the 1980s [11 , 18] . The success of the eradication programs was also reflected in Sabah , East Malaysia where malaria notifications decreased sharply , from peak notifications of 33 , 153 and 15 , 877 during 1994–1995 for P . falciparum and P . vivax respectively to 605 and 628 respectively in 2011 . Similarly notifications of P . malariae/P . knowlesi also fell from a peak of 614 in 1994 to <100/year in the late 1990s/early 2000s [6] . Although Malaysia has shown considerable success in the control of human malaria and is on target towards elimination of malaria by 2020 [19] , notifications of suspected P . knowlesi cases have increased from 59 notifications in 2004 to 996 in 2013 , an overall increase of over 16-fold [6] . According to the Malaysian Ministry of Health , P . knowlesi is the predominant species occurring in the country comprising 62% of the cases in 2013 [20] . It was suggested that the increase in number of P . knowlesi notification in Kudat maybe be due to high awareness of knowlesi infection among physicians and availability of better diagnostic tools to identify this malaria parasite [20] . In other words what is reported now represents the true infection rate as compared in the late 1990s and early 2000s when P . knowlesi cases were misdiagnosed as P . malariae . However , recent findings by Fornace et al . had demonstrated a clear link between land use change and P . knowlesi incidence , which strongly supports the idea that this is not just a problem of poor diagnosis/changing awareness , but a real epidemiological change [21] . The frontline vector control methods practiced in Malaysia under the malaria elimination programme are same as those used in several other endemic settings within the region i . e . the application of insecticides in houses either through use of Long Lasting Insecticide Treated Nets ( LLINs ) or Indoor Residual Spraying ( IRS ) . However there is little evidence to support that this is effective against P . knowlesi vectors . Thus there is a need for more detailed entomological investigation to assess the relative importance of exposure to mosquito vectors at or away from home , and to design control measures accordingly . Our working hypothesis is that given An balabacensis is the primary vector , we would expect that infection risk is higher when they are present . Towards this end , a case-control entomological study was conducted to determine if P . knowlesi infection risk is linked to exposure to vectors in domestic and peri-domestic settings . Specific comparisons were included to test for differences in vector abundance , species composition , biting time and infection rate at case and control households . Further aims were to assess and characterize the biting behavior of P . knowlesi vectors near homes ( time and place of biting e . g outdoors vs indoors ) , and to evaluate the potential for spread of other primate malarias in domestic settings . These findings will be useful for the control programme in designing vector control measures .
The study was conducted in Kudat district which is located in the northeasternn tip of Sabah ( 6°53'14 . 35" N 116°49'25 . 10" E ) and covers an area of approximately 1 , 300km2 with a population of 84 , 000 people of predominantly Rungus ethnicity ( 2010 National Census ) . The climate is tropical and the area is mainly coastal , with a maximum elevation of 250 metres above sea level . Forest cover is highly fragmented and substantial deforestation has occurred through conversion of forest to agricultural land [21 , 22] . The majority of the population lives in small villages ( mean population 160±15 , S1 Table ) and the main livelihood activities are small scale farming and plantation work . Plasmodium knowlesi is the main cause of human malaria in Kudat and , due to this relatively high incidence , this area is the focus for a number of interdisciplinary studies on the biomedical , environmental and social risk factors for P . knowlesi ( http://malaria . lshtm . ac . uk/MONKEYBAR ) . This includes a case control study , in which clinical malaria cases were recruited from the district hospital and visited at their homes within two weeks of initial diagnosis [22] . As there is mandatory reporting and referral of all malaria cases to the district hospital , the majority of symptomatic cases are captured by hospital systems . Approximately 180 P . knowlesi cases were identified through this active case surveillance between 2013–14; of which we randomly selected a subgroup of 28 for further entomological follow up ( representing cases reporting February , July , 2014 from 23 different villages: Fig 1 , S1 Table ) . The cases came from the age group 19–74 years old with a mean of 44 . There was a preponderance of males amongst the cases , and many ( 78 . 6% ) worked in agricultural sector , taking at least about 10–30 minutes to walk to their work place ( Table 1 ) Additionally , a matched “control” household was recruited in the vicinity of the case household for study which shared similar environmental characteristics in terms of surrounding vegetation and terrain , but where no occupants had reported with any malaria infection within the study period as indicated by records from the hospital and interviewing the residents . From the group of potential “control” households identified in the vicinity , one was randomly selected using a random table . The final choice was also dependent on the owner’s agreement . Within two weeks of the case detection , the selection of control house was accomplished and entomological work initiated . Data on the types of crops or vegetation surrounding households was collected , as well as the distance between each pair of case and control household ( S2 Table ) . The mean distance between the case and the control houses was 255±48 m ( 18–1000 m ) . As the villages were generally small in area , occasionally a control house could be located in a neighbouring village . Of the 28 case-control household pairs , five pairs occurred within the same village . For all pairs of households , mosquito sampling was conducted by four workers each at case and control simultaneously on the same nights . At these pairs , sampling was conducted for one to three nights depending on the owner’s permission and logistic constraints . Indoor collections were conducted at one station in the living room of houses ( H ) , whereas outdoor collections were conducted at three selected stations ( S1 , S2 & S3 ) within the garden area surrounding the house . The distance of the stations from the house was 24±1 . 7 m for the case and 19±0 . 8 m for the control household . These outdoor stations were selected based on information provided by the family members about where they were most likely to spend time outdoors in the evening . Mosquitoes were baited using human landing catch ( HLC ) method , but only Anopheles spp were collected for further analysis , whilst other species were killed and discarded at the site . Here a volunteer collected mosquitoes by exposing his lower legs ( from knee downwards ) and collecting all mosquitoes that land upon them in a plastic specimen tube ( 2 cm diameter X 6 cm ) which had a small piece of moist tissue at the bottom . Each station was manned by one person who would collect Anopheles for 12 hours straight ( 1800 to 0600 hr ) , and there was rotation of workers . The HLC workers at each case and control house were regularly monitored by a supervisor . The Anopheles were kept individually in a tube with a label which had information on the place , date , hour , location ( indoors vs outdoors ) and station of collection . The mosquito samples collected were recorded by hour in order to estimate the biting profile over course of night . The next morning the samples were taken to the laboratory to be processed . Anopheles specimens were identified the next morning in the laboratory based on morphology characters using published identification keys . The key of Sallum et al . [23] was used for Leucosphyrus group , whereas keys developed by Rattanarithikul et al . [24] were used for other groups . The identified specimens were kept individually in a sterile 1 . 5mL microfuge tube and stored in -20°C until used for molecular analysis . Each Anopheles specimen was cut into two parts: head-thorax and abdomen , and placed separately inside an autoclaved mortar and the tissue homogenized using pestle . The total DNA was extracted from each part following the method of Phillips and Simon [25] and stored in -30°C until PCR analysis . Detection of malaria parasites in the Anopheles specimens was performed using the nested PCR Plasmodium genus-specific method described by Singh et al . [26] . When a sample was found positive for malaria parasites , a second nested PCR was performed to determine the Plasmodium spp . using species specific primers in singleplex PCR [4 , 8 , 26 , 27] . Primers of nine species of Plasmodium namely P . coatneyi , P . inui , P . fieldi , P . cynomolgi , P . knowlesi , P . falciparum , P . vivax , P . malarie and P . ovale were used in this study . All these species have been recorded in Malaysia although P . ovale is an imported species , while P . knowlesi is the prevalent simian malaria infecting man . Both PCR 1 and PCR 2 were performed with 25μl final volume . The reaction components were prepared by mixing 5 . 0μl of 5X PCR buffer ( Promega ) , 0 . 5μl of dNTPs ( 10mM ) mixture ( Promega ) , 3 . 0μl of 25mM MgCl2 ( Promega ) , 1 . 0μl each of 10μM forward and reverse primers , 0 . 3μl of Taq DNA polymerase ( 5U/μl ) , 2 . 0μl of DNA template and sterile dH2O up to 25μl final volume . After the first PCR was completed , 2 . 0μl of the first PCR product was used as a template in the second PCR . The PCR conditions used were: an initial denaturation at 95°C for 5 min , followed by 35 cycles of 94°C for 1 min , annealing for 1 min and 72°C for 1 min , and a final extension at 72°C for 5 min . The annealing temperature was set based on the optimum temperature of the primers ( S3 Table ) . Statistical analysis was conducted using R programming language for statistical analysis ( version 3 . 2 . 2 ) . Generalised linear mixed models ( GLMM ) were constructed to test for variation in the abundance of Anopheles between case and control houses , and indoor and peri-domestic settings . In the analysis , household type ( case or control ) and location ( indoor or outdoor ) were considered as fixed effects , while month , night and sampling station ( site ) as random effects . To identify the best model , both negative binomial and Poisson distributions , interaction between type and location , as well as zero inflation were fitted . Tukey's Post Hoc test was used to compare mean between fix effects ( household type and location ) as well interaction between these two effects . We also analysed the proportion of mosquitoes that were caught feeding outdoors ( Po ) , and the proportion of human exposure to An . balabacensis ( Pe ) . The proportion of mosquitoes that were caught outdoors ( Po ) was calculated as Po=O18−06 h ( O18−06 h+I18−06 h ) ( 1 ) where O and I are respectively the number of mosquitoes caught biting outdoors and indoors during 6 pm– 6 am . From interviewing the residents and observation , more than 50% of the villagers would be indoors by 8 pm , and out the next morning by 5 am as they go to the plantations to work . The proportion of human exposure to An . balabacensis ( Pe ) is thus calculated as the proportion of bites that happen outdoors during the time when people are likely to be outdoors , out of the sum of bites expected throughout the night as humans move between indoor and outdoor areas of their home: Pe=O18−20 , 05 h ( O18−20 , 05 h+I20−05 h ) ( 2 ) where O18-20 , 05 h represents the mosquitoes caught biting from 6–8 pm , and 5-6am , and I20-5h represents the number caught indoors between 8 pm– 5 am . This would give a comparison between the proportion of bites people exposed to when outdoors between case and control households . GLMM with a binomial distribution and a logit link function was used to obtain the binary estimates of Po and Pe . In these models , household type ( case or control ) was fitted as a fixed effect , while sampling station of the case as a random effect . This project was approved by the NMRR Ministry of Health Malaysia ( NMRR-12-786-13048 ) . All volunteers who carried out mosquito collections signed informed consent forms and were provided with antimalarial prophylaxis during participation . House owners also gave permission to use their houses for collection of mosquitoes .
Among those who worked in the agricultural sector ( Table 1 ) , more than double the number cases were employed on rubber or oil palm plantations than controls , while more controls worked in the vegetable farms than cases . A total of 793 Anopheles belonging to 12 species were caught during the period of study , with An . balabacensis being the dominant species ( 81% of total ) , followed by An . maculatus , An . barbumbrosus , and An . donaldi ( Table 2 ) . Overall , more An . balabacensis were caught at case ( total 392 or 1 . 81 bites per man per night ) than control houses ( total 249 or 1 . 15 bites per man per night ) . Ten and 9 different Anopheles species were collected at case and control houses respectively , compared to only 6 and 2 indoor collections from case and control . Higher numbers were recorded at Kg . Tinukadan Laut ( CC24 ) , Kpg . Paradason B ( CC26 ) , Kpg . Nangka ( CC5 ) ( S1 Fig ) . GLMM analysis indicated that the negative binomial distribution gave a better fit than Poisson distribution . The log-likelihood values were for negative binomial and Poisson distribution -513 vs -529 respectively , while the Akaike information criterion ( AIC ) values were 1043 vs 1068 . Adopting the model with a negative binomial distribution , the abundance of An . balabacensis was found to vary significantly between case and control ( case 0 . 28±0 . 194 vs control 0 . 17±0 . 127 , z = 4 . 62 , p<0 . 001 ) , and between the surrounding peri-domestic area and inside the house ( 0 . 56± 0 . 394 versus 0 . 09±0 . 063 , z = 9 . 09 , p<0 . 001 ) ( Fig 2 ) . The interaction between house type and location was not significant ( z = 0 . 8 , P>0 . 05 ) More than 50% of An . balabacensis mosquitoes were caught biting in the early evening ( 6pm - 9pm ) with the peak hour between 7pm - 8pm ( Fig 3 ) . After 8pm , the number rapidly decreased with approximately 84% of the total nightly catch being accumulated by midnight . Nevertheless , one or two individuals of An . balabacensis could still be caught until dawn . In general , the proportion of bites taken by An . balabacensis outdoors ( Po ) was very high ( >95% ) , and did not vary between case and control households ( Table 3 ) . Similarly the proportion of human exposure to bites ( Pe ) did not vary between case and control households . A total of 793 Anopheles individuals were tested by molecular method and only ten An . balabacensis ( out of 641 or 1 . 56% ) were found to be positive for malaria parasites . Seven of them were caught at case houses ( 6 outdoors and one indoors ) , and 3 at control houses ( all outdoors , Table 4 ) . All these mosquitoes were found to be positive for simian malaria parasites ( P . coatneyi , P . inui , P . cynomologi & P . fieldi ) ; but none were with either the dominant zoonotic parasite reported in the area ( P . knowlesi ) or any human-specific Plasmodium . Ninety percent of infected An . balabacensis was caught biting outdoors between 6pm - 10pm ( 9 out of 10 ) , and with one infected individual caught between 1am - 2am . The proportion of infected An . balabacensis caught from case houses ( 7/392 or 1 . 79% ) was slightly higher than at control houses ( 3/249 or 1 . 20% ) . However , the sample sizes were too low for any robust statistical analysis of differences .
We conducted a randomized case-control field study to test the hypothesis that there is an association between P . knowlesi infection risk and higher exposure to mosquito vectors in peri-domestic ( outdoors surrounding houses ) and within domestic ( inside house ) settings . Although 12 Anopheles species were caught by HLC , only 4 were detected in reasonably high abundance: An . balabacensis , An . maculatus , An . barbumbrosus and An . donaldi . Of these , only An . balabacensis and An . donaldi have been previously implicated as malaria vectors of human malaria in Sabah [18 , 28] . In contrast , An . maculatus is the main malaria vector of human malaria in peninsular Malaysia [29] . Anopheles balabacensis appears to be a widespread species found in almost all sites , although significantly high numbers were caught in Kpg Tinukadan Laut ( CC24 ) especially in those areas near to forest fringes . About 95% of An . balabacensis was caught outdoors , similar to what was previously recorded ( a ratio of outdoor:indoor catch of 24:1 ) in Kuala Penyu , another district in Sabah [28] . A recent study conducted in Banggi Island situated north of Sabah and in Kg Paradason in the interior of Kudat district [10] also revealed that An . balabacensis was the predominant species collected , followed by An . donaldi in both sites . However , the next most abundant species was An . vagus , in Banggi , but An . barbirostris group in Kg Paradason . Anopheles maculatus was not caught in Banggi . GLMM analysis indicated a significant difference between the number of vectors caught at case and control houses , and between outdoor and indoor catches . The primary vector of P . knowlesi in the area , An . balabacensis , was present at higher abundance at households where cases were reported , which would suggest a higher risk at the case houses . Furthermore , as 90% of the infected mosquitoes were caught outdoors , it is likely that peri-domestic infection is an important risk factor . Although the indoor number of infective mosquitoes caught was small , getting infected indoors cannot be discounted . In Sarawak , it was postulated that humans were likely to acquire infection of P . knowlesi from being bitten by infected An . latens while hunting in the forest or as they return from the farm around dusk since in their study no infective mosquitoes were obtained from the village [13] . However , in Sabah clustering of cases among family members have been reported and they postulated that people could be infected around their homes [30] . A recent study also in Sabah showed the presence of asymptomatic cases of P . knowlesi occurring among the community in Kudat [31] . Thus , the result of this study seems to support the hypothesis that it is also possible for people to be infected in and around their homes . Although An . balabacensis is highly exophagic with only one infected individual found indoors we should not dismiss the fact that the possibility of indoor infection does exist . Thus we need to possibly expand our paradigm about transmission of P . knowlesi to include the possibility of peri-domestic infection , and conduct further studies to evaluate simultaneously the infection risk in and around households , as well as in forest areas , so the relative contribution of all these routes could be formally quantified . Many areas in Kudat district have undergone deforestation and clearance of vegetation for crop plantations , but it appears that An . balabacensis has remained the dominant species , with the exception of Kinabatangan area of Sabah where An . balabacensis was replaced by An . donaldi as main malaria vector as a result of deforestation and malaria control activity [18] . This suggests that the abundance of An . balabacensis in Kudat district was not greatly affected by the environmental changes . The impact of forest disturbance such as logging has been shown to increase the abundance of this disease vector and may partly explain the rapid rise in P . knowlesi cases in Sabah [32] . The feeding time of An . balabacensis appears to have changed since late 1960s when most of the area in Kudat district was still covered with forest . A study conducted then [29] showed that An . balabacensis was actively biting human at late night ( 10pm onwards ) , compared to early night with peak hour between 7pm to 8pm recorded now . In fact , this species starts biting human outdoors as soon as it starts to get dark . This change in feeding time could be due to An . balabacensis adapting to more people staying closer to forest fringe as more forested areas are cleared for crop plantation and housing . This could also be due to the introduction of insecticide treated bednets . Further study will be needed to confirm this . Although we did not obtain an An balabacensis individual infected with P . knowlesi , as only ten individuals were found Plasmodium positive albeit for other simian malarias , this could be a sampling error . As such , we are unable to make a conclusive prediction about infection risk at case households . Given the generally low rates of P . knowlesi infection in the vector ( eg 13/1482 or 0 . 88% , data also collected in Kudat ) [10] , thousands of samples are needed to obtain strong evidence to show that P . knowlesi was not present , and/or to compare infection rates between case and control households . Furthermore , as the densities of vectors in these settings are generally low , it would not have been feasible to achieve this sample size within the one year time span that the case control study was running . What data collected here do show however is that the primary vector is present at higher abundance in peridomestic settings where cases are reporting , on which basis the possibility of peridomestic transmission cannot be dismissed . This also indicates that people are routinely exposed to a variety of different primate malarias around their home; but that to date , only a couple ( knowlesi and cynomologi ) seem capable to causing any clinical infection . More research needs to be carried out to determine why these two primate malarias succeeded where the others fail so we can be proactive in the fight against future new simian malarias infecting man . The difference between bites per man per night between case and control houses is 0 . 66 ( 1 . 81–1 . 15 ) which works out to be 241more bites per person in a year for the case house . Since the infective proportion of vector is 0 . 88 [10] , this is equivalent to a higher entomological inoculation rate ( EIR ) of 2 . 12 . In addition , more than double the number cases worked in rubber or oil palm plantations than controls . Perhaps these two factors may help explain why there was P . knowlesi infection in the case houses . However more research is needed to validate this . As most P . knowlesi cases have been recorded from villages close to where macaques abound , and given that the primary vectors species , An balabacensis bite outdoors , a new paradigm in managing this malaria is needed . More attention should be focused on the ecology and biology of An . balabacensis in order to develop more effective control methods if the control or elimination of P . knowlesi malaria in Kudat district is to be successful . The current malaria control programme using ITNs might not have the desired impact as this species is mainly an exophagic species , and infection is more likely to occur outdoors in peri-domestic settings , in plantations and forest . | The primate knowlesi malaria has now emerged in human populations throughout South East Asia . Our limited knowledge of where and when people get exposed to the vector ( Anopheles balabacensis ) has resulted in poor control measures , although it is assumed that exposure occurs primarily when people are working in forest areas . We investigated the role of peri-domestic ( outdoors but around the household ) and domestic transmission . Mosquitoes were collected at 28 pairs households composed of one where an occupant had a confirmed knowlesi malaria infection and an associated “control” where no infection was reported . Most of the vectors were caught outdoors from 6pm - 9pm . The vectors were also significantly more abundant in the peri-domestic area than inside houses ( 5 . 5-fold ) , and also higher at case than control households . Ten Anopheles ( out of 641 ) were found positive for primate malaria parasites . This study shows that humans can be exposed to knowlesi infection around their homes . Given the vectors are mainly outdoor biters , interventions using insecticide treated bednets inside households may have relatively little impact . A paradigm shift in control methods is required to reduce infection of this primate malaria . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"parasite",
"groups",
"plasmodium",
"tropical",
"diseases",
"vertebrates",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"animals",
"parasitology",
"mammals",
"primates",
"apicomplexa",
"protozoans",
"molecular",
"biology",
"techniques",
"infectious",
"disease",
"control",
"insect",
"vectors",
"old",
"world",
"monkeys",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"monkeys",
"malarial",
"parasites",
"artificial",
"gene",
"amplification",
"and",
"extension",
"epidemiology",
"molecular",
"biology",
"disease",
"vectors",
"insects",
"macaque",
"arthropoda",
"mosquitoes",
"polymerase",
"chain",
"reaction",
"biology",
"and",
"life",
"sciences",
"malaria",
"amniotes",
"organisms"
]
| 2016 | Investigating the Contribution of Peri-domestic Transmission to Risk of Zoonotic Malaria Infection in Humans |
Neuroprosthetic brain-computer interfaces are systems that decode neural activity into useful control signals for effectors , such as a cursor on a computer screen . It has long been recognized that both the user and decoding system can adapt to increase the accuracy of the end effector . Co-adaptation is the process whereby a user learns to control the system in conjunction with the decoder adapting to learn the user's neural patterns . We provide a mathematical framework for co-adaptation and relate co-adaptation to the joint optimization of the user's control scheme ( "encoding model" ) and the decoding algorithm's parameters . When the assumptions of that framework are respected , co-adaptation cannot yield better performance than that obtainable by an optimal initial choice of fixed decoder , coupled with optimal user learning . For a specific case , we provide numerical methods to obtain such an optimized decoder . We demonstrate our approach in a model brain-computer interface system using an online prosthesis simulator , a simple human-in-the-loop pyschophysics setup which provides a non-invasive simulation of the BCI setting . These experiments support two claims: that users can learn encoders matched to fixed , optimal decoders and that , once learned , our approach yields expected performance advantages .
When the BCI system is closed-loop such that both the user can learn and the decoding algorithm can adapt , we have a setting which permits co-adaptation [3 , 4] . Changes in system performance may be driven by distinct types of adaptation: ( 1 ) All of the adaptation may occur due to the decoder and the user might not learn on relevant timescales [5] . ( 2 ) The user may learn in response to a fixed decoder [6 , 7] . ( 3 ) Both the user and the decoder might change , but the end result might perform no better than if only either the decoder or the user had learned . ( 4 ) In the most fortunate case , co-adaptation might be able to permit some synergistic result where both the user and the decoder learn in a collaborative fashion which yields high performance . Results in the literature hint suggestively at possibility ( 4 ) , yet it is difficult to distinguish this from possibility ( 3 ) because even if both decoder and user adapt , it may not be clear which drove the performance improvements [4] . Here we investigate this open question . While previous work has examined how decoder parameters should adapt in closed-loop settings in order to improve over static decoders [5 , 8 , 9] , there has not been an emphasis on doing so while explicitly considering how the user adapts . We define the decoder as the external system which decodes neural activity into end effector ( BCI ) control signals . The encoder then is the user’s internal encoding of intention into neural activity . As a clarifying example , neurons with simple tuning curves imply a linear encoding model in some cases , and a user’s adaptation of the encoder could correspond directly to tuning curve shifts [4 , 6 , 7 , 10] . For a given , fixed decoder , different encoding models will perform differently and better encoders will be those that , in some sense , match the decoder well . We will show that it is possible to compute the “best” encoding model for a given decoder—that is , some encoding model exists that minimizes the mean squared error ( MSE ) performance for a given decoder subject to signal-to-noise ratio ( SNR ) constraints . Similarly , for a given encoding model , there exists a best decoder in the MSE sense [11] . When a given decoder is presented to a user , optimal performance will be obtained if the user adapts such that encoding is optimally matched to the decoder . With this knowledge , one might imagine it useful to attempt to shape the user’s neural tuning properties [12] or otherwise attempt to guide the user to a certain encoding scheme which has been determined will optimize performance . Conceptually motivated by co-adaptation and BCI-user learning , we propose a framework for jointly optimizing the encoding model and decoder of a BCI under a minimum mean square error ( MMSE ) objective . The central premise of this work is that closed-loop co-adaptation is a special case of this joint encoder-decoder optimization problem , so an optimal decoder can be computed in advance , effectively circumventing the co-adaptation process . The core of our approach amounts to “pre-computing” the limit of an idealized co-adaptation process in which the user ( hypothetically ) optimally learns in order to obtain an optimized encoder-decoder pair . Instead of focusing on the temporal dynamics of learning in closed-loop [11] , we abstract to the space of encoder-decoder pairs and characterize how well different encoder-decoder pairs perform relative to one another . We pre-compute an optimal decoder which we can present as a fixed decoder to the user; given our modeling assumptions , the performance of this decoder by definition should not be surpassed by one obtained through co-adaptation . We emphasize that closed-loop learning by the user will still be critical to learn to control the fixed decoder , but that the decoder will not also need to be adapted . In a very general sense , learning by both the user and the decoder is equivalent to an optimization of performance with respect to these components . This perspective is essential , as it allows us to critically investigate co-adaptation as a mathematical optimization procedure , rather than an ill-defined “synergistic” training process . Co-adaptation then amounts to a specific approach to optimizing this objective over time—a coordinate descent approach where the user updates the encoder , and then the BCI system updates the decoder , iteratively . Seen from this perspective , it becomes clear that this objective function over encoder-decoder pairs could instead be descended using some other optimization strategy , and then a fixed , pre-optimized decoder could be presented to the user and learned . Such a setting would obviate co-adaptation . This approach could break down in a few key places: ( 1 ) The optimization could be intractable . ( 2 ) We may not be able to characterize the user objective . ( 3 ) We may not be able to characterize the constraints on user learning . It is worth emphasizing that these issues are present for co-adaptation as much as any generic optimization approach . Co-adaptation may not do a good job of obtaining optima of the objective function , and without knowing the user’s objective or constraints on user learning , co-adaptive decoder updates may be very difficult to tune or suboptimal . In the remainder of this work , we specify a MSE objective , and we work through the joint encoder-decoder optimization framework for the case when encoding is linear and the decoder is a steady state Kalman Filter ( SSKF ) . We assume that the SNR of the neurons is constrained and that it is difficult for the user to learn encoding models in which the correlation structure of the neural activity must change too much . We also assume that the user is aware of the MSE objective and is motivated to optimize this objective . Our model form is broadly consistent with other Kalman Filter decoding schemes employed in contemporary BCI . However , in particular settings , the objective function and the constraints on learning of the encoding model could be specialized . We return to the opportunity for specialization in the discussion after presenting our results . Finally , we validate the pre-computed decoder in computer simulations as well as in an online prosthesis simulator ( OPS ) [13] , a psychophysics platform which can serve as a test-bed for BCI . The OPS demonstrations are fully transparent ( e . g . allowing us to specify neural signal and noise characteristics ) so they allow us to gain insights into how the approaches we are studying work , and we show that our approach provides decoders which are both plausibly learnable and permit performance improvements on point-to-point reaching tasks .
Allowing for changes to both the decoder and encoder , we here show how to obtain encoder-decoder pairs which theoretically yield better performance than would be obtained either by learning an arbitrary , fixed decoder or adapting the decoder when the user is not learning . Although conceptually applicable generally , we apply our framework to the conventional Kalman filter ( KF ) , which serves as a reasonable choice of decoder for BCI system . KF decoding approaches have been made to have adaptive parameters in various ways [4 , 5 , 8 , 14] . However , no previous work has directly considered the co-adaptation problem as a joint optimization problem over pairs of encoders and decoders . Here we review the KF and then show how to derive optimal encoder-decoder pairs . Having provided the framework and described the encoder-decoder optimization methods , we validate our approach in two settings . We first examine the coding schemes arising from simulations and verify that our approach yields interpretable solutions . Following the simulation results , we show the results of psychophysics experiments implementing our approach on an online prosthesis simulator ( OPS ) , a system which tracks overt human body movements and generates synthetic , closed-loop data .
Our framework points to some qualitative insights into co-adaptation . In presenting our framework , we pointed out a non-obvious property of KF-decoder updates—updates to the decoder that are designed to be optimal for the current encoding model can allow for improvement if the user learns further ( i . e . adapts the encoding model ) . This is effectively why co-adaptation can serve as a coordinate descent algorithm . Implicitly , this property , coupled with the recognition that the user encoding can change , has permitted various approaches that attempt to track these changing encoders , using Kalman Filtering ( e . g . CLDA [4] ) as well as reinforcement learning ( e . g RLBMI [9] ) . In both settings , the decoder is adapted in a principled fashion which permits tracking as a user adapts , so user and decoder collaboratively improve task performance ( via a coordinate-descent-like method ) . These approaches may adapt the decoder slowly over many days . However , there is an intrinsic trade-off between lag and noise when using adaptive methods—the decoder can either adapt slowly when it has enough information to make good steps or adapt quickly but be highly susceptible to noise in the updates [33] . Slow updating induces delays in decoder adaptation which may make exploration by the user difficult because the user may attempt to explore on timescales shorter than the lag . We can contextualize our pre-computation approach relative to standard co-adaptive methods , by viewing pre-computed decoders as an attempt to push towards a certain logical extreme , essentially with the decoder “adapting” entirely before any user learning takes place . Leveraging available information about the statistics of the recorded signals , the decoder can change pro-actively to prompt the user , who may not be learning very efficiently , to move in a direction in which better performance is expected . This is effectively what our algorithm was designed to do . A core intuition in assessing the relationship between these approaches is the observation that if one dimension of neural activity has less noise than another , optimal decoding will rely more heavily on the less noisy dimension . If noisy and non-noisy dimensions initially have non-zero weight , perhaps co-adaptation can be used to tune the relative strengths . However , this tuning can also be computed directly given estimates of signal and noise statistics . We believe our framework helps clarify the performance opportunities available when using fixed or adaptive decoders . Our approach is broadly consistent with the literature on joint source-channel coding , but we are not aware of neural modeling being performed using our formulation . In the control literature , others have derived solutions to the related problem of optimizing time-varying observation models for the time-varying Kalman filter , but their result does not extend to the steady state case and concludes with a numerical analysis showing that their solution can be very suboptimal for our setting [34] . There also exists other research which has looked into control schemes for “body-machine interfaces” , for example using a sensor-wired glove with many degrees of freedom ( DOF ) to control a cursor-on-a-screen [35–37] . This research is generally consistent with our approach , and also uses the term co-adaptation in a way which refers to situations where a user learning over time can be complemented by the decoder adapting to the new encoding of the user . The body-machine interface literature discusses various options for selecting the encoder , but they do not model the effects of noise or the relative differences between different encoder choices . One approach uses principal component analysis ( PCA ) on calibration data and then maps the first n principal components to the n effector DOFs by an arbitrary mapping which can be relatively straightforward for the user to then learn by exploration [38 , 39] . Our approach resembles the PCA approach if noise and temporal correlations are ignored , since we also rely on unsupervised collection of second order statistics . However , neural noise plays a significant role for us . Consider that in our 1D demo the total covariance has high variance in noisy dimensions . A naïve application of the PCA-based decoding scheme will result in a decoder which most heavily relies on the noisiest dimensions simply because they have more variance . Alternatively , when an adaptive decoder is used , the adaptation proceeds from a reasonable initialization [36 , 37]—for the neural case , our approach helps find a reasonable scheme without depending upon any well chosen initialization . There are a few points to be aware of when interpreting the results of this paper . We optimize encoder-decoder pairs for the target tracing task , and we rely on similarity between tracing and the pinball tasks for the derived optimal solutions to hold in the pinball case . Additionally , we still have a free parameter in our optimization , λ , for tuning the magnitude of the neural activity constraints . For OPS experiments , we did not optimize performance over λ; rather , we tuned it very crudely to the parameters of the synthetic neural activity and held it fixed across subjects . A limitation of the present study is that learning in an OPS setting may be different from learning in real BCI or across classes of BCI [40 , 41]—for example , myoelectric interfaces engage later stages of a motor hierarchy , so timescales and constraints of learning may be different [40] . Consequently the extent of learnability and timescale of learning of our fixed , optimized decoder may vary by setting ( OPS vs BCI type ) . While the OPS may serve as a relevant proxy of BCI in many respects [13] , ultimately experiments in specific settings will be required to test the utility of our approach to each . That said , both the decoding approach and the OPS can be modified to incorporate other features of the neural activity such as neural dynamics . While not conventionally modeled in BCI , recent research has revealed dynamical components to neural activity in motor cortices [42] . For such a class of neural activity used for the neuroprosthesis , one could consider the extensions y t = A x t + B x ^ t − 1 + D y t − 1 + n o i s e as a model of the neural activity . Moreover , although we used kinematic signals as the inputs to our simulated neurons , other user-controllable signals ( e . g . muscle activity ) could be used as inputs into the encoding model , possibly affecting the results . The inputs to the OPS are merely intended to serve as reflections of user intention and not necessarily what the real neural activity encodes in normal , non-BCI behavior . We lastly emphasize that our methods require estimation of the neural noise covariance , so a calibration session used to examine neural activity is of crucial importance . For real neural applications , the neural signal and noise components could be estimated using various neural factor analysis methods [43 , 44] . We have made an initial attempt to build learnable decoders using a new joint optimization framework . We have focused on the linear-Gaussian setting which is implicit when KF decoders are employed . There will certainly be model mismatch , and insofar as this model proves too simple , the way forward is to incorporate additional features of the encoding model and constraints on user learning . While our approach is very clear about the extent to which it relies upon the accuracy of these assumptions , it is worth emphasizing that co-adaptation approaches are equally ( if implicitly ) dependent on these assumptions . For example , the structure and timescale of decoder updates in co-adaptive approaches directly impact how well users can learn [36] or whether the user learns at all [5] . By making these assumptions explicit , we hope that we can improve the adaptive decoder engineering process . Additionally , while KF decoders are still widely used in BCI , future improvements may require nonlinear decoding methods , so we may wish to extend our framework to certain classes of nonlinear encoder-decoder pairs . For example we could consider Poisson-observation encoding with point-process decoders or otherwise parameterize the neural noise covariance to scale with the magnitude of the signal ( i . e . signal-dependent noise as in [45] ) . Generally , any processes ( even discrete ones ) could be specified for x and y . Following the conceptual approach proposed here , the objective function over encoder and decoder parameters could be specified and optimized subject to constraints . Furthermore , a full decoding system would certainly incorporate additional components such as modifications for increased stopping reliability [46] and approaches for longer-timescale nonstationarity tracking as in [47 , 48] . In our OPS experiments the modeling assumptions hold by construction , so how significantly results will deviate from those of this paper in physiological experiments is open . Also , we admittedly focused on low dimensional control in order to gain conceptual ground , but this is simpler than high dimensional control . It remains to be explored how much more difficult user learning is in high dimensional settings , and what modifications will be required to assist users in learning optimal encodings . Additionally , more detailed characterization of the learning process may become relevant . If applying our framework to human BCI , there are a number of practical opportunities . For example , it may be interesting to develop interfaces which indicate to the user how much control is available for some number of DOF and permit the user to select how to map control to the BCI effector . Users may have specific preferences to invert certain dimensions of control or request more precision on a certain effector DOF even at the expense of precision for other DOF . Such user-specific modifications are entirely amenable within our approach .
Human protocols were approved by the Columbia University Morningside Institutional Review Board—all subjects read an IRB-approved informed consent document and provided verbal consent ( Protocol Number: IRB-AAAM6407 ) . The Online prosthesis simulator ( OPS ) is an experimental platform that tracks overt movements of the user , uses the movements to simulate noisy neural data , and then decodes the synthetic data by a brain-computer interface ( BCI ) decoding algorithm [5 , 13] . We captured the overt movements of able-bodied users with a Microsoft Kinect . The Kinect interfaces to a standard computer via usb and signals were analyzed in real time in matlab ( middleware for Kinect-to-Matlab follows [5] ) . Kinect data for our experiments was collected at approximately 10Hz . Movement data was used to simulate neural signals by an arbitrary scheme , selectable by the experimenter—we chose to generate neural activity by de-meaning the movement signals , linearly mixing them , and adding independent Gaussian noise to each neural channel . This simulated neural data was then analyzed by real-time BCI algorithms during experiments . The decoded results of the simulated data were presented as feedback to the user so the user could change their movements . See Fig 8 for a visual depiction of this system . The user’s kinematic intention ( xt ) should correspond to an attempt to control the cursor towards a target presented on the screen . The overt arm movements of the user are used to simulate a population of synthetic , noisy neural units ( yt ) . The encoding model ( A ) is the estimated linear mapping from xt to yt . During the task , the neural activity is combined with the decoded estimate of the previous timestep ( x ^ t − 1 ) to determine the current cursor position ( x ^ t ) . The decoded cursor position comes from the use of the optimal steady state Kalman filter ( using F and G via Eq 3 ) . In some high level sense , the user could explore the space of mappings from their behavior to visible feedback . While the user could not change how their overt movements map to neural activity , the user could change the relationship between “kinematic intention” and simulated neural activity—this is the relationship we call the encoding model . That is , changes in how the user made overt movements when intending to obtain certain decoded kinematics will cause changes in the encoding model ( estimated parameters A ) . We emphasize that these overt movements are not supposed to correspond to overt movements in a BCI setting; rather , these overt movements are simply supposed to provide arbitrary volitional signals which the BCI system will exploit to perform a task . BCI permits the use of any neural signals which are correlated with user volition—for example , even imagined speech has been used to control a 1D cursor [49] . The OPS could be driven with arbitrary control signals , and overt movement signals seem a reasonable model of the class of signals which may in part be used to drive a real BCI . To analyze optimal encoder-decoder pairs , we have been considering the objective function of a “tracing” task ( a . k . a . pursuit tracking ) , in which the user has a time-varying kinematic intention and the decoder attempts to optimally estimate the kinematics . This objective is standard , physically meaningful , and the methods developed from it yield smooth priors . We validated our methods using the OPS on a “pinball” task ( a . k . a . point-to-point reaching ) where the user controls a smoothly moving cursor towards targets , as this is more realistic for applications . In this work , we present the results of a 1D pinball task . We present data from 4 subjects ( none of whom are authors of the study or otherwise directly affiliated with this project ) . Subjects were explained the rules of the “game” they were to play but were not described details of simulated neural activity ( some had performed prototype versions of the experiment previous to participation in the experiments reported here ) . Before beginning the task , we collected calibration data for each user—the user was instructed to move around freely in an unstructured way . During this time , the user saw real-time video ( RGB image frames ) of themselves with skeletal tracking superimposed , in order to see how well the Kinect tracking worked . This unstructured phase was to get an estimate of the empirical neural activity covariance ( i . e . estimated as the sample covariance of the neural responses ) . The subjects participated in one session for each of two decoders ( “motor-imitation” and “pre-computed” , described more below ) in each of three “signal cases” . Signal cases consisted of specifications of how strongly the simulated neural channels were driven by different tracked body movements of the user ( described more below ) . Between each of these six conditions , subjects were given a break until they felt ready for the next . For each signal case , the two decoder sessions were presented with the motor-imitation trial first , followed by a trial where the cursor was controlled by the pre-computed decoder . These trials were independent and the decoders were unrelated so the order mattered only so the control strategy the user learned when using the pre-computed decoder would not affect the motor-imitation control scheme . For block-order validation see S1 Fig . Each session consisted of two key phases , “free exploration” and “testing . ” During free exploration , the user controlled a cursor on a screen via closed-loop OPS control , and there was no target . During this time , the user was prompted to explore and try to learn as well as possible how to control the cursor—this phase gave the subject an opportunity to learn , and nothing about the decoder was changed . Exploration rapidly provided information about how the user’s intentions relate to the movement of the cursor . The user can adjust their control strategy , in turn , adjusting the tuning of the simulated neurons—this process constitutes learning in this task . After a few minutes of exploration , a target appeared and the user was prompted verbally to move to the testing phase when ready by holding the cursor at the target for a few seconds ( subject 1 moved directly to the testing phase without having to hold a target first , and this subject reported having had adequate time in free exploration ) . During testing , a single target at a time was presented , surrounded by a visible halo . The user attempted to control their cursor to the target and attempted to “acquire” the target by holding the cursor in the halo-region ( holding is indicated by a color change of the region inside the halo ) . After a short period of time , if the cursor is held within the halo ( ≈1s ) , the target is acquired , disappears , and a new target is randomly placed at a new location . Targets that were not acquired after ≈15s were treated as misses , and a new target replaced them . The subject was instructed to acquire as many targets as possible in the ≈3min duration of the testing phase . Before beginning the task , the entire trial structure was verbally described to each subject , and verbal reminders of trial-stage ( at transitions ) were provided by the experimenter . A video depicting the experimental procedures is included as S1 Movie . The two decoders tested were selected either by “motor-imitation” initialization or by pre-computing the optimal encoder-decoder pair and presenting the optimal decoder . Trials with “motor-imitation” decoders began with an extra phase ( at the beginning of the free exploration phase ) , during which a target moved smoothly on the screen . The user was instructed to “trace” the target with their hand movements , thereby providing the “native” control mapping of making rightwards and leftwards movements of their hands to control a cursor along the horizontal axis ( without visual feedback , i . e . open-loop ) . This OPS phase provided supervised data which permitted estimation ( by recursive least squares—an online algorithm for linear regression ) of a motor-imitation encoding model , a control scheme which is as close as possible to user default control . The motor-imitation decoder refers to the optimal decoder corresponding to this encoding model , which can be computed directly from the estimated encoding model . Alternatively , for the pre-computed decoder trials , we used covariance information determined from the calibration data to estimate an optimal encoder-decoder pair according to our joint optimization approach presented in this paper , and we presented the pre-computed optimal decoder to the user . This second approach does not require labeled data; rather it merely optimizes the encoder-decoder pair such that good task performance results after the user learns how to use the scheme . Selection of the decoder is unsupervised , and this approach relies on the user to adapt to match the fixed decoder . We note that P and Q ( i . e . the prior dynamics parameters of Eq 1 ) are the same for both cases , with P selected just under 1 ( i . e . . 99 ) to discourage drift and Q selected to be small ( . 01 ) , which was determined by the sample rate and the distances in the task to permit appropriate movement rates for this task . Strictly , Q is a free parameter which affects smoothing , but it should be selected in a task-dependent manner matched to the speed of target movement ( or distance between targets ) . Using the OPS , we are able to arbitrarily specify mappings of user movements to the simulated neural data as well as setting noise levels across channels—this corresponds to the blue box of the OPS flow diagram in Fig 8 . To compare results across subjects , we selected three pre-defined mappings by which overt movement would drive simulated neural channels ( k = 6 ) along with corresponding neural noise structure . These could be thought of either as a small number of electrodes or pre-processed neural data ( e . g . after dimensionality reduction ) . Noise is the same order of magnitude as the signal—this is realistic for single units . For each channel , noise is set to a fixed high or low value , with high noise set to be approximately the same magnitude as the signal of a heavily used channel and low noise set to be less than the signal power of a heavily used channel ( per channel ) . Optimal encoder-decoder pairs necessarily take into account user calibration data so it is not possible to know in advance precisely what the optimal encoder-decoder pair will be . With that said , the optimal coding scheme should leverage the highest SNR dimensions . For comparison , the motor-imitation scheme which the user initializes will consist of rightwards and leftwards movements to control the cursor rightwards and leftwards respectively . Case 1: Simulated neural activity is linearly modulated by positions of the right hand in horizontal and vertical dimensions—one channel driven by horizontal movement , another by vertical movement . Noise is low in the vertical dimension channel and high in the horizontal dimension channel , so the optimal encoder-decoder pair will predominantly encode the movement axis in the y movements of the right hand . Case 2: Simulated neural activity is linearly modulated by horizontal position of both hands together as well as distance between hands—one channel driven by the sum of the two hands’ horizontal position , another channel by their difference . The channel driven by distance between hands has low noise so the optimal encoder-decoder pair will predominantly encode the movement axis in channel corresponding to the distance between hands . Case 3: Simulated neural activity is linearly modulated by positions of each hand in horizontal and vertical dimensions independently—separate channels are driven by each of these variables . Noise is low on the simulated activity driven by vertical dimension of each hand , so subjects should move both hands together vertically . We compared all subjects on all three cases of signal-noise structure using both motor-imitation initialized decoders and pre-computed decoders . The above cases were selected with the intent of highlighting differences between naïve , motor-imitation schemes and schemes which permit better performance by coding in dimensions that are not natively leveraged . The intention was to see if optimization of the decoder yielded performance differences and how learnable the schemes were . In the results , we made use of estimated user encoding models , comparing them with optimal encoding models . To estimate the user encoding models during the pinball task , we simply performed linear regression from target positions to neural activity during the pinball phase . This assumes that the user is attempting to control the cursor such that it moves directly towards the target . | Brain-computer interfaces are systems which allow a user to control a device in their environment via their neural activity . The system consists of hardware used to acquire signals from the brain of the user , algorithms to decode the signals , and some effector in the world that the user will be able to control , such as a cursor on a computer screen . When the user can see the effector under control , the system is closed-loop , such that the user can learn based on discrepancies between intended and actual kinematic outcomes . During training sessions where the user has specified objectives , the decoding algorithm can be updated as well based on discrepancies between what the user is supposed to be doing and what was decoded . When both the user and the decoding algorithm are simultaneously co-adapting , performance can improve . We propose a mathematical framework which contextualizes co-adaptation as a joint optimization of the user’s control scheme and the decoding algorithm , and we relate co-adaptation to optimal , fixed ( non-adaptive ) choices of decoder . We use simulation and human psychophysics experiments intended to model the BCI setting to demonstrate the utility of this approach . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | Encoder-Decoder Optimization for Brain-Computer Interfaces |
Clostridium difficile is a Gram-positive spore-forming anaerobe and a major cause of antibiotic-associated diarrhoea . Disruption of the commensal microbiota , such as through treatment with broad-spectrum antibiotics , is a critical precursor for colonisation by C . difficile and subsequent disease . Furthermore , failure of the gut microbiota to recover colonisation resistance can result in recurrence of infection . An unusual characteristic of C . difficile among gut bacteria is its ability to produce the bacteriostatic compound para-cresol ( p-cresol ) through fermentation of tyrosine . Here , we demonstrate that the ability of C . difficile to produce p-cresol in vitro provides a competitive advantage over gut bacteria including Escherichia coli , Klebsiella oxytoca and Bacteroides thetaiotaomicron . Metabolic profiling of competitive co-cultures revealed that acetate , alanine , butyrate , isobutyrate , p-cresol and p-hydroxyphenylacetate were the main metabolites responsible for differentiating the parent strain C . difficile ( 630Δerm ) from a defined mutant deficient in p-cresol production . Moreover , we show that the p-cresol mutant displays a fitness defect in a mouse relapse model of C . difficile infection ( CDI ) . Analysis of the microbiome from this mouse model of CDI demonstrates that colonisation by the p-cresol mutant results in a distinctly altered intestinal microbiota , and metabolic profile , with a greater representation of Gammaproteobacteria , including the Pseudomonales and Enterobacteriales . We demonstrate that Gammaproteobacteria are susceptible to exogenous p-cresol in vitro and that there is a clear divide between bacterial Phyla and their susceptibility to p-cresol . In general , Gram-negative species were relatively sensitive to p-cresol , whereas Gram-positive species were more tolerant . This study demonstrates that production of p-cresol by C . difficile has an effect on the viability of intestinal bacteria as well as the major metabolites produced in vitro . These observations are upheld in a mouse model of CDI , in which p-cresol production affects the biodiversity of gut microbiota and faecal metabolite profiles , suggesting that p-cresol production contributes to C . difficile survival and pathogenesis .
Clostridium difficile is a Gram-positive spore-forming enteric pathogen and the leading cause of antibiotic-associated diarrhoea worldwide[1] . C . difficile infection ( CDI ) ranges from self-limiting diarrhoea to severe and life threatening pseudomembranous colitis[2] . C . difficile spores are the aetiological agent of CDI transmission and are resistant to desiccation , environmental stress , disinfectants and heat[3 , 4] . These spores , present in both hospitals and the environment are transmitted via the faecal-oral route , contributing to both nosocomial and community acquired CDI [3] . Infection with C . difficile is frequently preceded by treatment with broad-spectrum antibiotics , which eliminate discrete taxa of the commensal intestinal microbiota resulting in dysbiosis and permitting colonisation by C . difficile . Certain bacterial taxa have been highlighted as important in the prevention of C . difficile colonisation[5–7] . Since restoration of microbial diversity can resolve recurrent infections , faecal transplantation is viewed as an effective treatment strategy[8] . However , a greater understanding of how C . difficile is able to influence the gut microbiota and disrupt intestinal homeostasis is a current imperative . Human intestinal bacteria have been shown to ferment dietary-derived carbohydrates[9] and proteins[10] , producing short chain fatty acids ( SCFA ) , as well as an array of metabolites via fermentation of aromatic amino acids[11] . The secondary metabolites of this highly diverse microbial community have the potential to either positively or negatively influence many aspects of human health [12] , with some demonstrated to possess toxic and carcinogenic properties [11 , 13 , 14] . Aromatic amino acids such as phenylalanine , tryptophan and tyrosine are important sources of phenyl metabolites . These metabolites can be absorbed in the small intestine or pass through to the colon[15] to be excreted in faeces . One such fermentation product , phenylacetic acid ( PAA ) , is the most commonly detected secondary metabolite in healthy human faeces , with reported concentrations of 479 μM[15] . C . difficile ferments tyrosine , via p-hydroxyphenylacetate ( p-HPA ) , to produce p-cresol . Para-cresol is a phenolic compound [16] that has been demonstrated to inhibit the growth of a range of bacterial species and other microorganisms[17 , 18] . To date , the capacity to produce p-cresol has only been demonstrated in a select number of organisms[19 , 20] , including eighteen intestinal commensal species[11] . However , the in vitro production of p-cresol by these species was relatively low ( ranging from 0 . 06–1 . 95 μg/ml ) [11] . Furthermore , C . difficile can tolerate relatively high concentrations ( 1 mg/ml ) of p-cresol[21 , 22] . As such , the ability to synthesise and tolerate high concentrations of p-cresol has led to the hypothesis that it may provide C . difficile with a competitive advantage over other microorganisms . The enzyme responsible for the decarboxylation of p-HPA is a member of the glycyl radical family , 4-hydroxyphenylacetate decarboxylase , which is encoded by three genes hpdB ( CD630_01530 ) , hpdC ( CD630_01540 ) and hpdA ( CD630_01550 ) , which are co-transcribed in an operon . The hpdBCA operon is highly conserved in all the sequenced C . difficile isolates . We have previously shown that disruption of any of the three genes renders C . difficile incapable of synthesising p-cresol[22] . In this study , we demonstrate that production of p-cresol by C . difficile confers a fitness advantage over other intestinal bacteria both in vitro and in vivo , specifically those with a Gram-negative cell envelope . The treatment of human faecal samples with exogenous p-cresol significantly modified the cultivable bacteria therein , in a species-specific manner . Furthermore , a p-cresol deficient mutant showed a modest but significant reduction in viable counts in a relapse mouse model of CDI . Comparisons of the metabolic and 16S rRNA profiles identified variation in the biochemical and bacterial composition between mice infected with the C . difficile strain 630Δerm and the p-cresol deficient mutant ( hpdC::CT ) following infection and relapse . This is the first study to show that p-cresol production is a mechanism by which C . difficile confers a competitive advantage over other gut bacteria .
It has been hypothesised that p-cresol production provides C . difficile with a selective advantage over competitors in the human gut . To investigate this , we assessed the effect of exogenous p-cresol on the in vitro growth dynamics of selected intestinal commensal species ( S1 Table ) compared to C . difficile strain 630Δerm ( Fig 1 & S2 Table ) . The data shows a clear pattern whereby sensitivity to p-cresol correlated with bacterial cell envelope structure . We observed that Gram-positive bacteria were significantly more tolerant to p-cresol than Gram-negative bacteria ( Coefficient of variance ( COV ) = 0 . 599 , p<0 . 001 ) . Growth of the Gram-negative species , including members of the Bacteroidaceae ( Bacteroides thetaiotaomicron ) and Enterobacteriaceae ( Escherichia coli , Klebsiella oxytoca and Proteus mirabilis ) families were inhibited by the addition of exogenous p-cresol in a dose-dependent manner ( Fig 1 & S3 Table ) and demonstrated a significant decrease in cell growth compared to C . difficile ( p<0 . 005 ) . In contrast , the Gram-positive species including those from the Bifidobacteriaceae ( Bifidobacterium adolescentis ) , Enterococcaceae ( Enterococcus faecium ) and Lactobacillaceae ( Lactococcus fermentum ) families displayed no significant reduction in growth rate , even at 0 . 1% ( v/v ) p-cresol ( Fig 1 ) . Interestingly , E . faecium displayed greater tolerance to p-cresol than C . difficile itself ( COV = 0 . 6 p = 0 . 002 , S3 Table ) . We had previously constructed a ClosTron inactivation mutant in the hpdC decarboxylase gene ( strain hpdC::CT ) , which renders C . difficile unable to produce p-cresol[22] . To investigate whether production of p-cresol contributes to fitness in vitro , we performed co-culture assays with 630Δerm and hpdC::CT cultured with a selection of intestinal commensal species , supplemented with exogenously added p-cresol . Brain heart infusion media with yeast extract ( BHIS ) was chosen for these co-culture experiments , as we have previously shown that intrinsic production of p-cresol under these conditions is negligible[22] , therefore any observed effect could be attributed to the exogenously added p-cresol . We observed no difference in growth rate between 630Δerm and hpdC::CT in these conditions [22] . To establish the comparable growth conditions , each species was normalised to the same starting optical density ( OD595 0 . 5 ) and starting CFU/ml was determined ( S4 Table ) . The competitors were mixed in a 1:1 ratio at matched OD , and were grown for 24 hours in media supplemented with 0 . 05% ( v/v ) p-cresol . Viable counts for each species were determined by plating serial dilutions onto media supplemented with and without D-cycloserine and cefoxitin , facilitating differentiation between C . difficile and the competitor . When C . difficile 630Δerm was grown in co-culture with E . coli in the absence of exogenous p-cresol , E . coli was the dominant organism , represented by a significantly higher CFU/ml than C . difficile ( 8:1 ratio of E . coli to C . difficile ) ( COV = 1 . 02 , p = 0 . 003; Fig 2 , S5 Table ) . However , when the medium was supplemented with exogenous p-cresol , the relative proportion of C . difficile increased to a ratio of 1:1 , representing an 8-fold increase in the number of viable C . difficile ( Fig 2A ) ( COV = -1 . 38 , p<0 . 001 ) . A similar profile was observed when E . coli was co-cultured with the hpdC::CT mutant ( Fig 2B ) ( COV = -0 . 27 , p = 0 . 882 ) . This suggests that 630Δerm and the hpdC::CT mutant displayed comparable fitness when grown in co-culture with E . coli . When C . difficile was grown in co-culture with a Gram-positive bacterium , E . faecium , C . difficile was significantly less abundant compared with the competitor ( COV = 3 . 41 , p<0 . 001 ) . Here , we observed a ratio of 1:10 of C . difficile to E . faecium ( Fig 2C & S5 Table ) . The growth dynamics of the C . difficile hpdC::CT mutant and E . faecium ( Fig 2D ) were also indistinguishable from the 630Δerm grown in competition with E . faecium ( COV = -0 . 33 , p = 0 . 283 ) . However , when the medium was supplemented with p-cresol , the relative proportion of E . faecium increased significantly ( COV = 1 . 44 , p = 0 . 010 ) . This suggests that the growth conditions were more permissive for E . faecium . However , this was not the case for all Gram-positive species tested . The relative ratio in co-culture of C . difficile ( 630Δerm and hpdC::CT mutant ) to L . fermentum was not significantly altered by exogenous p-cresol ( COV = -0 . 058 , p = 0 . 818 ) ( Fig 2E & 2F and S5 Table ) . These data indicate that p-cresol had a range of effects on growth dynamics depending on the Phylum of bacteria and their susceptibility to p-cresol . We observed no difference in competitive fitness between wild type and p-cresol mutant when grown in BHIS supplemented with 0 . 5% ( w/v ) p-cresol . Therefore , we developed an additional in vitro competition assay to determine whether intrinsic p-cresol production by C . difficile conferred a competitive advantage over other intestinal commensal species . To achieve this we measured the growth rate of E . coli and K . oxytoca in monoculture and compared this to that of C . difficile ( S1A Fig ) . Under these conditions , C . difficile reached exponential growth at a later time point than the other species tested . Furthermore , we have previously shown that p-cresol is detected in C . difficile cultures at around 4 hours ( or OD595 0 . 5 ) [22] . In order to limit the dominance of competitor species , and ensure optimal p-cresol production , we grew C . difficile to exponential phase ( OD595 0 . 6 ) before inoculating the medium with the competitor ( at OD595 0 . 05 ) . We also supplemented the growth medium with p-HPA to drive production of p-cresol . To determine a concentration of p-HPA that resulted in inhibitory p-cresol production , competitive co-culture experiments with C . difficile and E . coli were performed in a range of p-HPA concentrations ( 0 . 1% , 0 . 2% or 0 . 3% ( w/v ) ) ( Fig 3 ) . When the medium was supplemented with 0 . 1% ( 6 . 5 mM ) p-HPA we observed a ratio of 13:1 ( E . coli:C . difficile ) ( Fig 3A ) . When the concentration of p-HPA was increased to 0 . 2% ( 13 . 1 mM ) , we observed a significant difference ( p<0 . 001 ) in the ratio of E . coli: C . difficile ( 1:1 ) , compared to the ratio in 0 . 1% p-HPA ( 13:1 ) ( Fig 3A ) . Further increasing the concentration of p-HPA to 0 . 3% ( 19 . 7 mM ) resulted in culture conditions that favoured C . difficile , reflected by a ratio of 1:4 ( E . coli:C . difficile ) ( p<0 . 001 ) ( Fig 3A ) . Thus , we observed a positive correlation between the proportion of p-HPA supplemented in the growth medium and the survival of C . difficile compared to E . coli ( Fig 3A ) . To determine whether this effect was linked to the level of p-cresol production , we quantified p-cresol in these culture supernatants by High Performance Liquid Chromatography ( HPLC ) ( Fig 3B ) . Fig 3B demonstrates that increasing p-HPA concentration correlated with a significant increase in p-cresol production ( p<0 . 001 ) . We observed 25 ±0 . 04 mM p-cresol when the growth medium was supplemented with 0 . 3% p-HPA ( Fig 3B ) . Next , we investigated whether the p-cresol mutant displayed reduced fitness when grown in competitive co-culture with other gut competitor species . Furthermore , we constructed a complement by expressing the hpdC and hpdA genes from a tetracycline-inducible promoter using a plasmid based system ( generating strain hpdC::CT::phpdCA ) . We compared the growth of C . difficile strains 630Δerm , hpdC::CT and the complement ( hpdC::CT::phpdCA ) , in competition with E . coli , K . oxytoca or B . thetaiotaomicron in media supplemented with 0 . 2% p-HPA ( Fig 4 ) . The number of viable counts for each species was determined as outlined above . When 630Δerm was grown in co-culture with E . coli , we observed a 1:1 ratio of C . difficile to E . coli ( Figs 4A & 3A ) . This was consistent with the co-culture assays supplemented with exogenous p-cresol ( Fig 2 ) . However , competitive co-culture between hpdC::CT and E . coli resulted in a decrease in the relative proportion of C . difficile to ca . 25% of the total culture ( 1:4 , C . difficile:E . coli ) . This indicates that the mutant was significantly less viable than the wild type ( COV = -1 . 06 , p<0 . 001 ) . 630Δerm demonstrated comparable relative fitness in competitive co-culture with K . oxytoca ( 1:1 ratio ) ( Fig 3B ) . However , we observed proportionally fewer CFUs of 630Δerm when grown in co-culture with B . thetaiotaomicron ( 1:4 ratio ) ( Fig 3C ) . By contrast , hpdC::CT displayed reduced fitness relative to 630Δerm when grown in competition with both K . oxytoca ( COV = -1 . 40 , p<0 . 001 ) and B . thetaiotaomicron ( COV = -0 . 79 , p = 0 . 001 ) . This fitness defect was restored when the complement was grown in competition with K . oxytoca and B . thetaiotaomicron ( Fig 3 ) . However , complementation of the hpdC mutation , did not restore C . difficile fitness to wild-type levels in competitive co-cultured with E . coli . Therefore , we quantified both p-cresol production and p-HPA utilisation by HPLC ( Figs 4D , 3B & S2 ) . Quantification of 630Δerm supernatants grown in both monoculture and competitive co-culture supplemented with 0 . 2% p-HPA revealed an average p-cresol concentration of 13 . 3 ±0 . 1 mM . In contrast , the concentration in supernatants of the complemented mutant ( at 0 . 2% p-HPA ) was only 4 . 8 ±0 . 2 mM , representing a significant 2 . 7 fold reduction ( p<0 . 01 ) . Therefore , we conclude that under competitive co-culture conditions , 4 . 8 ±0 . 2 mM p-cresol was sufficient to have a deleterious effect on the growth of K . oxytoca and B . thetaiotaomicron , but not on E . coli . Increasing the concentration of the transcriptional inducer ( anhydrotetracycline ) and p-HPA resulted in increased p-cresol production ( from 4 . 8 ±0 . 2 mM to 15 . 6 ±3 . 9 mM ) by the complement and restoration of the phenotype ( S2 Fig ) . Here , the level of p-cresol production directly correlated with the concentration of the transcriptional inducer ( S2C and S2D Fig ) . As expected , this suggests that complementation was more greatly influenced by transcript expression rather than availability of the p-HPA precursor . Furthermore , we observed no difference in growth rate between the three C . difficile strains at any tested concentration of anhydrotetracycline ( S1 Fig ) . Taken together , these data suggest that production of p-cresol by C . difficile confers a competitive growth advantage over susceptible bacterial species ( specifically , Gram-negative species ) under our in vitro conditions . To further understand the effect of p-cresol production on the interaction between C . difficile and these intestinal species , we characterised the metabolic content of both mono-culture and competitive co-culture supernatants using 1H nuclear magnetic resonance ( NMR ) spectroscopy . We analysed the culture supernatants described in Fig 4 and performed principal component analysis ( PCA ) to identify metabolic variation across the profiles . The scores plot from the PCA model comparing all profiles showed that the largest variation in the metabolic data ( Principal Component 1 ( PC1 ) ) was between the C . difficile strain 630Δerm samples and those from the hpdC::CT strain ( Fig 5A ) . The metabolic profiles from the complement samples clustered between 630Δerm and hpdC mutant samples . The loadings for PC1 describe the metabolites varying between the strains . This indicated that culture supernatants from 630Δerm contained significantly greater amounts of p-cresol and alanine compared to the other strains , but lower amounts of p-HPA , butyrate and isobutyrate ( Fig 5B ) . This was consistent with the notion that p-HPA is being depleted in order to produce p-cresol . The 630Δerm samples clustered together regardless of whether the bacteria were grown in mono-culture or co-cultured with E . coli , K . oxytoca or B . thetaiotaomicron . In contrast , hpdC::CT and complement ( hpdC::CT::phpdC-A ) strains were separated in the second principal component ( PC2 ) based on the competitive co-culture conditions . The loadings for PC2 indicated that the mono-cultured C . difficile and B . thetaiotaomicron competitive co-culture samples contained lower amounts of acetate compared to the competitive co-cultures from E . coli and K . oxytoca . This metabolic variation between strains and competitive co-culture conditions is summarised in the clustergram shown in Fig 5B , which was constructed from the Z-scores of 1H NMR peak integrals measured for each metabolite across all samples . The dendrogram shows that the 630Δerm metabolic profiles clearly cluster away from those of the other two strains ( Fig 5 ) and the variation in p-cresol production between samples is apparent in the 1H NMR spectrum ( S3 Fig ) . The dendrogram also showed that p-cresol and alanine clustered together as did butyrate and isobutyrate ( Fig 5B ) . We also assessed the effect of altered p-HPA concentration on metabolic profiles . The PCA of metabolites produced in media supplemented with 0 . 1% and 0 . 2% p-HPA demonstrated no clear metabolic variation between these samples . However , we did observe clustering within the 0 . 1% p-HPA samples , driven by increased p-cresol and alanine ( S4 Fig ) . These data suggest that under these growth conditions p-cresol is one of the most abundant metabolites in culture supernatants . This is reflected by both metabolic profiling and HPLC quantification , which correlates to p-cresol susceptibility observed in both competitive co-culture and monoculture of Gram-negative bacteria . Our results have demonstrated that p-cresol production confers a fitness advantage over discrete bacterial species in vitro . Therefore , we sought to determine whether this was also true in vivo . Individually caged C57BL/6 mice were infected in parallel with 1x104 spores of C . difficile strain 630Δerm ( n = 5 ) or the hpdC::CT mutant ( n = 5 ) and compared to uninfected naïve control mice ( n = 5 ) , in a relapse model of infection . Mice were given cefoperazone in their drinking water for 10 days to stimulate gut dysbiosis , before infection by oral gavage with C . difficile spores ( Fig 6A ) [23 , 24] . Stool samples were collected throughout the experiment for analysis . Twenty-eight days post-infection , mice were treated with vancomycin in their drinking water for 7 days to encourage recurrence of infection ( Fig 6A ) . Infection was monitored by enumeration of spores isolated from faeces on C . difficile selective media ( Fig 6B ) . We observed no significant difference in the number of spores recovered from faeces of mice infected with either 630Δerm or hpdC::CT following cefoperazone treatment , indicating that the hpdC::CT mutant and 630Δerm were equally competent at initial colonisation ( Fig 6C ) . This was consistent with the notion that these strains demonstrate similar resistance to cefoperazone and vancomycin ( S6 Table ) and no differences in sporulation in vitro ( S5 Fig ) . However , at day 7 post-infection we observed a modest difference in colonisation , represented by significantly fewer 630Δerm CFUs compared to the hpdC::CT mutant ( p<0 . 05 ) . Relapse was detected by enumeration of C . difficile spores post-vancomycin treatment . Three days following removal of vancomycin ( D3R ) , spores were detected in all but one 630Δerm infected mouse and three out of five hpdC::CT mutant infected mice . By day 4 post-relapse ( D4R ) C . difficile spores were recovered from the faeces of all mice and we observed a significant reduction ( p<0 . 05 ) in the number of spores recovered from the faeces of mice infected with the hpdC::CT mutant relative to 630Δerm infected mice ( Fig 6D ) . Both infections followed a broadly comparable progression , however , we observed modest but significant differences in the number of spores recovered both pre- and post- relapse between 630Δerm infected and hpdC::CT infected mice at discrete time points . Interestingly during relapse , the spore density remained lower in hpdC::CT compared to 630Δerm , despite an initially higher CFU at D7 post-infection , indicating that these differences are a result of in vivo fitness . Given that the hpdC::CT mutant displayed an altered colonisation profile during relapse and that p-cresol displays bacteriostatic properties against a number of species , this led us to postulate that production of p-cresol may alter the composition of the intestinal microbiota in such a way that favoured C . difficile re-colonisation . We isolated bacterial DNA from four key time points during the relapse model of CDI; day 7 post-cefoperazone treatment , immediately upon cessation of vancomycin treatment ( D0R ) , day 2 post-relapse ( D2R ) and day 4 post-relapse ( D4R ) , when all the mice were colonised with C . difficile ( 1 . 6 x 106 WT and 2 . 3 x 105 hpdC::CT mutant spores/g faeces ) . To assess the community structure of the microbiota , 16S rRNA sequencing was performed by sequencing the V5-V7 region of 16S rRNA gene . The data was grouped with distance-based similarity of 97% into operational taxonomic units ( OTUs ) , using Greengenes and associated summaries and diversity analyses were performed in QIIME . Consistent with previous studies , the microbiota of cefoperazone-treated mice was dominated by Lactobacillaceae ( Fig 7A and S7 Table ) [24–26] , which comprised 39 . 7% ( mean relative abundance ) of the total microbiota in 630Δerm infected mice . The microbiota was also populated by Bacteroidetes , including members of the S24-7 ( an uncultured commensal of homeothermic animals[27] ) ( 17 . 6% ) and Paraprevotellaceae ( 1 . 2% ) families . Furthermore , Firmicutes , including Staphylococcaceae ( 12 . 75% ) , other Clostridiales ( 7 . 2% ) , Lachnospiraceae ( 4 . 2% ) , Erysipelotrichaceae ( 3 . 7% ) , Ruminococcaceae ( 2 . 9% ) , Enterococcaceae ( 2 . 2% ) , Turicibacteraceae ( 1 . 2% ) , and Actinobacteria including Bifidobacteriaceae ( 1 . 9% ) ( Fig 7A ) also contributed to the microbiota composition . However , animals infected with the hpdC::CT mutant demonstrated a significant increase in microbial diversity at D7 ( ANOVA p<0 . 05 ) , compared to 630Δerm infected and naïve mice ( Fig 7B and S7 Table ) , which is also upheld with an ANOSIM population analysis p<0 . 05 ( S6 Fig ) . Consistent with the notion that p-cresol prevents outgrowth of Proteobacteria , the majority of the families that were only present in the hpdC mutant infected animals were members of the Proteobacteria Phylum ( S7 Table ) , albeit at low abundance . Treatment with vancomycin significantly reduced diversity of both uninfected ( naïve ) and infected mice ( 630Δerm and hpdC::CT ) ( Fig 7B & 7C ) , resulting in a dramatic increase in the relative representation of Lactobacillaceae , specifically the Lactobacillus genus , which constituted ≥85% of the microbiota of all the mice examined ( 90% 630Δerm , 87% hpdC::CT mutant and 85% naïve at D0R ) . Principle component analysis demonstrated clustering of D0 and D2 post relapse ( Fig 7C ) . At D2R , the diversity of the microbiota remained low . At D4R , partial recovery of the microbial diversity was observed ( Fig 7A and 7B ) , which coincided with the detection of C . difficile spore in faeces ( Fig 6D ) . At D4R there were distinct differences in population composition in the intestinal bacteria of these animals ( Fig 7D ) . We observed an increase in spread on the PCA plot at D4R ( Fig 7C ) , compared to the clustering observed at D0R and D2R . Although the mean relative proportions of Lactobacillaceae were similar ( 73 . 9% in 630Δerm and 72 . 3% in the hpdC::CT mutant infected animals ) , there were clear differences in the representation of other taxa , including other Firmicutes , Proteobacteria and Bacteroidetes ( Fig 7D ) . In 630Δerm infected mice there was an increase in representation of Firmicutes from the Erysipelotrichales ( 16 . 5% 630Δerm and 0 . 1% hpdC::CT mutant ) , Bacillales ( 1 . 9% 630Δerm and 0% hpdC::CT mutant ) and Clostridiales orders , and the Bacteroidetes ( Fig 7D ) . Conversely , in the p-cresol mutant infected mice , we observed an increase in the representation of Proteobacteria , specifically , the Gammaproteobacteria of the Pseudomonadales ( 5 . 5% 630Δerm and 18% hpdC::CT mutant ) and Enterobacteriales ( 0% 630Δerm , 6 . 75% hpdC::CT mutant ) order , and the Betaproteobacteria of the Burkholderiales order ( Fig 7D ) . Consistent with the notion that Gram-negative species were more susceptible to the effects of p-cresol , Gammaproteobacteria formed 26 . 2% of the total microbiome in hpdC::CT infected animals D4R , compared with 5 . 5% in 630Δerm infected mice ( COV = 9 . 37 , p = 0 . 023 ) , suggesting that p-cresol may inhibit their outgrowth following treatment with vancomycin . Our data suggest that p-cresol production by C . difficile influenced the composition of the mouse faecal microbiota . Therefore , we investigated whether these differences resulted in an altered biochemical profile . Stool samples collected throughout the duration of the initial mouse infection ( at day 2 ( D2 ) , 4 ( D4 ) and 7 ( D7 ) ) and during relapse ( at days 0 ( D0R ) and 4 ( D4R ) ) were analysed using 1H NMR spectroscopy ( Fig 8 ) . The PCA scores plot identified biochemical variation between the faecal profiles collected D2-D7 versus D0R-D4R in the control mice and those infected with the hpdC::CT mutant . The D2-D7 samples contained greater acetate compared to the relapse time points , and lower amounts of an unknown metabolite ( δ 3 . 59 , singlet ) . Vancomycin induced perturbations in the metabolic activity of the intestinal bacteria are likely to underlie these changes[28] . The faecal profiles from mice infected with the 630Δerm strain showed similar metabolic alterations to the control mice and those infected with the mutant strain during the initial infection ( D2-D7 ) . However , the response was different 4 days after relapse . At D4R , the faecal profiles from mice infected with the 630Δerm strain were more variable than the uninfected mice and those infected with the mutant strain and were similar in composition to the initial infection profiles ( Fig 8 ) . We have shown that p-cresol production has deleterious effects on the outgrowth of Gram-negative bacterial species both in vitro ( Figs 1–4 ) , and in an in vivo mouse infection model ( Fig 7 ) . Thus , we sought to determine the effect of exogenously added p-cresol on biodiversity of the human microbiome . Healthy human stool samples were taken from donors ranging from 60–65 years old , who had not received antibiotic treatment in the last 3 months , eliminating possible perturbations by antibiotic therapy . We measured the effect that exogenously-added p-cresol ( at 0 . 1% and 0 . 3% ) had on faecal microbiota , compared to a Phosphate Buffered Saline ( PBS ) control . Differential plating revealed that the facultative anaerobes were particularly sensitive to p-cresol at both 0 . 1% ( COV = -0 . 61 , p = 0 . 006 ) and 0 . 3% ( COV = -1 . 82 , p<0 . 001 ) , represented by a significant reduction in viable counts ( Fig 9 ) . The Bacteroides fragilis group was also significantly reduced after exposure to both 0 . 1% ( COV = -1 . 29 , p = 0 . 009 ) and 0 . 3% ( COV = -4 . 39 , p<0 . 001 ) p-cresol ( Fig 9 ) . The total anaerobes and lactose-fermenting Enterobacteriaceae were also significantly reduced after exposure to 0 . 3% p-cresol ( COV = -1 . 48 , p<0 . 001 , COV = -2 . 36 , p<0 . 001 , respectively ) ( Fig 9 ) . Consistent with the mouse model of CDI , p-cresol at 0 . 1% had a limited effect on the survival of Lactobacillus ( COV = -0 . 045 , p = 0 . 890 ) and Bifidobacterium species ( COV = -0 . 100 , p = 0 . 642 ) . However , a significant decrease in survival was observed for both groups when they were incubated in 0 . 3% p-cresol ( p<0 . 01 ) . In line with our in vitro co-culture data ( Fig 2 ) , Enterococcus species present in human faecal samples were not adversely affected by the addition of p-cresol ( Fig 8 ) , even at the highest concentrations tested ( COV = 0 . 48 , p = 0 . 873 ) . Given that we observed a clear distinction in the nature of species that displayed tolerance to p-cresol , we reasoned that the cell envelope would be an obvious target for its mode of action . Phenolic compounds that target membranes typically induce a rapid loss of low molecular weight compounds from within the cell as a result of increased membrane permeability[29–31] . Thus , we used the release of inorganic phosphate ( Pi ) as a metric for determining membrane integrity in the presence of p-cresol . Initially , we compared the release of phosphate from E . coli and C . difficile strains ( 630Δerm and hpdC::CT ) in increasing concentrations of p-cresol ( Fig 10A ) . We observed a significant increase in the amount of phosphate released by E . coli compared to C . difficile ( COV = 0 . 868 , p = 0 . 005 ) . Only 16% of the total intracellular phosphate of C . difficile was released upon contact with p-cresol . Furthermore , the hpdC::CT mutant displayed a similar phosphate release profile to 630Δerm C . difficile ( COV = 0 . 201 , p = 0 . 444 ) , which was not significantly different ( Fig 10A ) . This indicates that disruption of p-cresol production had little bearing on p-cresol tolerance , under these conditions . Our data suggests that cells can tolerate p-cresol up to a threshold level ( ~0 . 4% v/v ) , after which the amount of phosphate released becomes saturated . Therefore , we selected a concentration of 0 . 3% ( v/v ) to determine phosphate release over time in a selection of Gram-positive and Gram-negative gut bacteria . Membrane integrity was measured in the presence of p-cresol by comparing the amount of p-cresol induced phosphate release , to the total intracellular phosphate , which was determined by boiling cell suspensions for 15 minutes . Fig 10 demonstrates that species with a Gram-positive cell envelope display greater tolerance to p-cresol than Gram-negative species , represented by significantly less phosphate release ( COV = -2 . 478 , p<0 . 001 ) , ( Lactobacillales: E . faecium ( p = 0 . 005 ) and L . fermentum ( p = 0 . 003 ) , the Bifidobacteriales: B . adolescentis ( p = 0 . 01 ) and the Clostridiales: C . difficile ( p<0 . 01 ) ) corroborating previous observations . Conversely , the Gram-negative Gammaproteobacteria: P . mirabilis , E . coli and K . oxytoca released their total intracellular pool of phosphate over the course of the assay . P . mirabilis and K . oxytoca released 68% and 60% of their total phosphate respectively immediately upon contact with p-cresol ( Fig 10B ) . Both species released >90% of their total phosphate pool following 30 minutes contact with p-cresol . In contrast , no Gram-positive species analysed released their total intracellular phosphate pool over the course of the assay ( Fig 10C ) . However , B . adolescentis released 63% of its total phosphate at 30 minutes compared to 20% for E . faecium , 27% for L . fermentum and 33% for C . difficile , indicating that B . adolescentis is more sensitive to p-cresol than other Gram-positive species . Prolonged exposure to p-cresol resulted in other Gram-positive species releasing a greater portion of their intracellular pool of phosphate , however , the level of Pi released by C . difficile never exceeded its initial level of release ( Fig 10 ) . In conclusion , our data demonstrate a clear correlation between bacterial cell envelope structure and susceptibility to p-cresol .
The indigenous microbiota has been shown to form an ecological barrier that prevents the ingress of pathogenic bacteria such as C . difficile [32] . However , the specific components of the intestinal microbiota that facilitate colonisation resistance are only recently becoming clear [5–7 , 25 , 33–36] . Both the treatment with broad-spectrum antibiotics and the availability of specific metabolites has been shown to play a role in the expansion of particular bacterial species within the human microbiota [37 , 38] . Here , we present compelling evidence that C . difficile may directly modify the intestinal microbiota through production of p-cresol . We demonstrate that C . difficile displays a greater degree of tolerance to p-cresol compared to other common intestinal species , including the Gammaproteobacteria: E . coli , K . oxytoca and P . mirabilis , as well as the Bacteroidetes , B . thethaiotaomicron . We show that these bacterial species are susceptible to the effects of both endogenous and exogenous p-cresol , which was reflected in reductions of viable counts when these intestinal microbiota species were grown in competitive co-culture with C . difficile . Using a plasmid based complementation system to restore the expression of the p-HPA decarboxylase , we have shown that p-cresol production by C . difficile must exceed 5 mM to elicit a significant alteration in competitive growth dynamics . We have shown that C . difficile is able to utilise all the available p-HPA supplemented in the growth medium , which results in the production of up to 25 ±0 . 04 mM p-cresol in vitro ( Fig 3 ) , which is 1000-fold more than the amount of p-cresol produced from tyrosine metabolism by other organisms cultured from the intestinal microbiota ( range 0 . 06–1 . 95 μg/ml ) [11] . There is evidence that p-HPA is present in the human colon and detected in healthy human stool samples at 19 μM[15] , therefore C . difficile can potentially utilise free tyrosine and p-HPA to produce p-cresol in vivo . We expanded our investigation of the influence of p-cresol on the growth of other bacterial species , to identify other metabolites influencing growth . In particular , alanine , p-cresol , acetate , butyrate , isobutyrate and p-HPA were the six main metabolites that were differentially modulated in mono-culture and co-culture of C . difficile with intestinal bacteria . The abundance of these metabolites in vitro was altered in the presence of the p-cresol mutant compared to C . difficile strain 630Δerm . Acetate and butyrate are the most common end products of fermentation in the gut[39] . C . difficile can use amino acids as the sole energy source via Stickland fermentation , in which amino acid acceptors ( such as glycine , proline and hydroxyproline ) are reduced in a paired metabolism with electron donors ( such as leucine , isoleucine or alanine ) . This can result in the conversion of alanine to acetate [40] . However , in C . difficile monoculture , we did not observe an inverse association between acetate and alanine ( Fig 5B ) , suggesting that C . difficile is not utilising alanine in stickland fermentation under nutrient rich conditions ( in BHIS media ) . This suggests that the competitor species , may have been responsible for the increased utilisation of alanine in co-culture with the p-cresol mutant and complement , where the competitor is more abundant . The reduction of the Stickland acceptors glycine and proline in C . sticklandii and C . difficile requires two selenium dependant reductases , glycine reductase and D-proline reductase [40 , 41] , highlighting the importance of selenium in growth and metabolism in C . difficile , particularly in glycine reduction and selenocysteine production [40 , 42] . Nutrient availability has been linked to virulence in C . difficile in a number of different ways , via the global transcriptional regulators CodY , CcpA , PrdR and Rex , which are involved in overlapping cellular processes including toxin production , amino acid biosynthesis , stickland fermentation , nutrient transport , fermentation and cell membrane components[43] . The hypervirulent C . difficile strains ( RT027 and RT078 ) have also developed the ability to metabolise low concentrations of trehalose , via acquisition of a single point mutation in the trehalose repressor ( treA ) , which increases virulence of these ribotypes in vivo[44] . C . difficile is also capable of utilising ethanolamine as a carbon source[45] and the ethanolamine genes are upregulated in vivo in the presence of B . thetaiotaomicron when animals were fed on a standard polysaccharide diet [38] . Cysteine is involved in amino acid and energy metabolism in C . difficile [46] , modulating processes such as carbon transfer , electron transport , butyric acid and butanol production . Cysteine results in increased levels of intracellular tyrosine [47] . Cysteine also down-regulated 4p-hydroxyphenylacetate-3-hydroxylase [48] , which may reduce p-HPA availability for the p-HPA decarboxylase and thus would decrease flux to p-cresol . Therefore , cysteine-regulated pathways may result in increased p-cresol production . This is consistent with the notion that the production of butyrate and p-cresol are inversely regulated . C . difficile and other opportunistic gut bacteria have developed metabolic strategies that differ in response to environmental signals , one such strategy is the production of short chain fatty acids [38] . Studies using a simplified gnotobiotic mouse model , have shown that succinate produced by B . thetaiotaomicron is used by C . difficile to produce butyrate , boosting C . difficile titres [38] . However , in our experimental conditions ( nutrient rich conditions ) succinate was detected at very low levels by 1H NMR spectroscopy . Butyrate has anti-inflammatory properties [49] and is produced by a diverse array of bacterial phyla [50] . Yet , butyrate stimulates C . difficile toxin production in the absence of rapidly metabolised carbohydrates ( e . g . glucose ) [48] . The production of butyrate from acetyl-CoA or succinate by C . difficile is suppressed by the transcriptional regulators CcpA , CodY and Rex [51] , however , if proline is limited then alternative pathways for NAD+ regeneration are used including glycine reductase , alcohol dehydrogenase and butyrate production from acetyl-CoA and succinate are induced [51] . We observed lower butyrate and isobutyrate concentrations in the 630Δerm cultures ( mono- and co-cultures ) , which implies that C . difficile is unable to synthesise butyrate under these conditions . However , we observed high butyrate and isobutyrate concentrations in co-culture with the p-cresol mutant and complement , where the competitor is more abundant , suggesting that the competitors are responsible for the increase in butyrate in co-culture . Inter-C . difficile strain variation in the metabolic profiles included altered abundance of alanine , isobutyrate , p-cresol and p-HPA . We observed an increase in the production of p-cresol in monocultures and co-cultures containing the C . difficile 630Δerm and an absence of p-cresol in all cultures with the hpdC::CT mutant , consistent with the inability of the mutant to synthesise p-cresol ( Fig 5B ) . The complement produced an intermediate amount of p-cresol ( Figs 5B and S3 ) and therefore clustered between the wild type and mutant in the PCA plot , suggesting that p-cresol and p-HPA were the main metabolites driving separation between C . difficile strains . This observation was corroborated when p-cresol production was increased in co-culture by increasing the supply of p-HPA , which resulted in a decrease in the competitor relative to wild-type C . difficile . In this study , we present evidence that p-cresol production by C . difficile prevents outgrowth of discrete taxa of bacteria in vitro and that p-cresol production may modulate composition of the mouse microbiota . Our data showed clear differences in the composition of the microbiota of mice infected with the p-cresol mutant compared with the 630Δerm strain pre- and post-relapse in our infection model ( Fig 7 ) . Recovery of the microbial community to its pre-dysbiotic state is often a slow process and , consequently , susceptibility to C . difficile colonisation can be increased for weeks and even months following cessation of antibiotics[52 , 53] . Both the 630Δerm and the hpdC::CT mutant successfully colonised mice at the initial infection stage . However , we observed increased microbial diversity in hpdC::CT infected mice , at day 7 post-infection . This diversity was largely driven by OTUs that each constituted <0 . 1% of the total microbiota . Despite relatively low abundance of these OTUs , they could have important consequences for the microbial ecosystem . These Families include Corynebacteriaceae , Propionibacteriaceae ( both Actinobacteria ) , Bradyrhizobiaceae , Burkholderiaceae , Comamonadaceae , Oxalobacteraecea , Rhodocyclaceae , Bdellovibrionaceae and Enterobacteriaceae ( all Proteobacteria ) . Consistent with the notion that p-cresol prevents outgrowth of Proteobacteria , the majority of these Families were members of the Proteobacteria Phylum . Vancomycin treatment reduced microbial diversity , altered the metabolic content of the stool samples , and resulted in a microbiome that was susceptible to relapse with C . difficile . The remaining microbial community was dominated by Lactobacillaceae ( consistent with previous publications [24] ) , which was insufficient to restore colonisation resistance following vancomycin withdrawal . Upon cessation of vancomycin treatment , we observed an expansion of microbial diversity ( D2R and D4R ) . There were clear differences in microbiota composition at D4R between animals infected with the p-cresol mutant and 630Δerm C . difficile . The second most abundant class present in the microbiome of hpdC mutant-infected mice was Gammaproteobacteria ( 26 . 2% ) . In contrast , the second most abundant class in animals infected with 630Δerm C . difficile was the Erysipelotrichia ( 16 . 5% ) . Other studies have shown that without FMT , dysbiosis is maintained in mice with two main OTUs , Lactobacillus and Turicibacter ( Erysipelotrichia order ) [24 , 35] , which we observed in the 630Δerm infected mice , but not mice infected with the p-cresol mutant . We have shown in vitro that Gammaproteobacteria , including K . oxytoca , E . coli and P . mirabilis are more sensitive to p-cresol than C . difficile , while Gram-positive bacteria from the Lactobacillales family are more resistant to p-cresol . This is particularly pertinent as the majority of the microbiota post-vancomycin treatment was comprised almost exclusively of Lactobacillus and an increased expansion of the Gammaproteobacteria was only seen in p-cresol mutant-infected mice . The faecal metabolic profiles from all animals post-vancomycin treatment ( D0R ) were clearly distinct from those collected post-infection ( D2 , D4 and D7; Fig 7 ) . However , the metabolic profiles of mice infected with the 630Δerm strain 4 days after withdrawal of vancomycin were more variable than those infected with the mutant strain . The 630Δerm infected mice also had a metabolic signature more similar in composition to those samples collected post-infection ( D2 , D4 and D7 ) . This suggests that the biomolecular perturbations following re-establishment of infection with 630Δerm were more closely related to those observed with the initial infection . In contrast , the hpdC::CT-infected mice had profiles more closely related to the uninfected mice . To complement the in vitro co-culture assay and mouse model of CDI we assessed the effect of exogenous p-cresol on the human microbiota using ex vivo healthy human faecal samples . We observed a reduction in the number of viable total anaerobes , facultative anaerobes and lactose fermenting enterobacteriacea ( LFE ) . The LFE are comprised of the Gammaproteobacteria E . coli , Klebsiella spp , Enterobacteria spp , Citrobacter spp and Serratia spp . This observation corroborates previous findings that demonstrate a significantly reduced viability of Gram-negative bacteria by in vitro growth kinetic analysis , competitive co-culture and in a mouse model of CDI . In contrast , the Gram-positive bacteria isolated from the ex vivo healthy human faecal samples ( Bifidobacteriaceae , Lactobacillales and Enterococcaceae ) were consistently less sensitive to p-cresol in the assays we performed . In this study , we have shown that C . difficile displays a greater degree of tolerance to p-cresol when compared to a selection of other common intestinal bacterial species . Our data suggest a clear distinction between the fundamental properties of the organisms susceptible to the p-cresol , whereby Gram-positive species displayed greater tolerance than Gram-negative species . We demonstrate that p-cresol affects the integrity of surface barriers resulting in a concentration-dependant leakage of small molecules such as phosphate . Similar effects have been observed with m-cresol and chloro-cresol on bacterial cell membranes[29] . p-cresol was recently shown to inhibit proliferation of colonic epithelial cells and induce necrotic leakage of protons through the inner mitochondrial membrane[54] . Pseudomonas putida strain P8 , which has the capacity to degrade p-cresol , modifies its fatty acid composition by increasing the abundance of 9-trans hexadeconoic acid and decreasing the abundance of 9-cishexadeconoic acid when grown in the presence of sub-lethal concentrations of phenol[30 , 55 , 56] . Previous work has demonstrated that sublethal concentrations of phenolics , including p-cresol , resulted in an increase in the degree of saturation of cell membrane lipids , which is thought to counteract the increase in membrane fluidity [30 , 57] . In conclusion , we demonstrate that the production of p-cresol by C . difficile alters the composition and recovery of diversity in the intestinal microbiota . A p-cresol deficient mutant has a reduced ability to compete with other intestinal microbiota species in vitro . We have shown that the effect of p-cresol is more detrimental to the growth of Gram-negative bacteria , differentially inhibiting proliferation of various bacterial Phyla . Exposure to p-cresol resulted in release of cellular phosphate , suggesting that it disrupts cell envelope integrity . This study provides evidence that p-cresol production by C . difficile provides it with a competitive survival advantage over other intestinal bacterial species .
C . difficile strains 630Δerm[58] and hpdC::CT[22] have been previously described . The intestinal microbiota species used in the study were obtained from Mark Wilcox and Simon Baines at the University of Leeds isolated from a gut soup model of CDI ( S1 Table ) . All bacteria were cultured in pre-reduced Brain Heart Infusion ( BHI ) ( Oxoid ) , supplemented with 0 . 5% ( w/v ) yeast extract ( BHIS ) and 0 . 05% ( w/v ) L-cysteine ( Sigma ) , at 37°C and under anaerobic conditions . For growth rate analysis , our collection of gut bacteria was grown in 100 ml tissue culture flasks with shaking at 50 rpm , 37°C and under anaerobic conditions . Pre-reduced growth media was supplemented with 0 . 1% , 0 . 05% and 0 . 01% ( v/v ) p-cresol as indicated . OD595 was determined every hour for 8 hours with a final reading at 24 hours , growth curves were performed in triplicate . A p-cresol complement strain ( hpdC::CT::phpdCA ) was made , using an inducible plasmid based system derived from pRPF185[59] ( S2 Table ) . The hpdCA genes were PCR amplified and cloned downstream of a tetracycline inducible promoter ( ptet ) in pRPF185 [59] to produce the plasmid phpdCA . This was then conjugated into the hpdC::CT mutant to create a complement[60] . This was performed alongside an empty plasmid pLDempty , which was derived from pRPF185[59] , to contain the ptet promoter , but without a gene ( S2 Table ) . This was transferred into C . difficile using competent E . coli CA434[60] into both 630Δerm and hpdC::CT mutant as controls . Linear regression analysis was performed using Stata15; data was transformed using Log10 to approximate a normal distribution . The data was mined to determine if there was a significant difference in; a ) strains , the growth of all bacteria strain compared to the reference strain C . difficile strain 630 , b ) p-cresol concentration compared to the BHIS untreated control , c ) the Gram-negative bacteria compared to the Gram-positive bacteria . The COV indicates whether the growth is higher ( positive number ) or lower ( negative number ) than the reference and the p-value indicates the probability , a minimum cut off of p<0 . 05 was used throughout for significance ( S3 Table ) . Culture supernatant from mono-culture and co-culture experiments in media supplemented with 0 . 1 and 0 . 2% ( v/v ) p-HPA were filter sterilised . Samples were diluted into 400 μL of phosphate buffer ( pH 7 . 4 , 100% D2O , 3 mM of NaN3 , 1 mM of 3- ( trimethyl -silyl ) -[2 , 2 , 3 , 3-2H4]-propionic acid ( TSP ) for the chemical shift reference at δ0 . 0 ) according a 1:2 ratio . Samples were transferred to 5 -mm tubes for 1H nuclear magnetic resonance ( NMR ) spectroscopic analysis , which was performed on a Bruker 600 MHz spectrometer ( Bruker Biospin , Karlsruhe , Germany ) at 300K ( 26 . 85° ) . The parameters of the acquisition were as previously reported for urine[61] . Each spectrum was acquired with 4 dummy scans followed by 32 scans . Spectra were automatically phased , baseline corrected and calibrated to the internal standard ( TSP ) using Topspin ( Bruker Biospin , Karlsruhe , Germany ) . The processed spectral data was imported into Matlab ( version R2014a , The Mathworks Inc . ) . The region δ4 . 84–4 . 76 was removed to eliminate the residual water signal . Principal Components Analysis ( PCA ) was performed using pareto scaling , due to the significant intensity of the acetate signal ( δ 1 . 92 , single . Based on the PCA loadings , spectral peaks contributing to the principal components were integrated using an in-house script . These metabolite peak integrals were used to construct a clustergram in Matlab using the clustergram script . Primary cultures were inoculated from a single colony of C . difficile strains ( 630Δerm , hpdC::CT , hpdC::CtphpdCA ) into pre-reduced BHIS broth and grown to an OD595 0 . 3 on a shaking platform at 50 rpm . These were inoculated 1/100 into pre-equilibrated BHIS broth which was incubated statically for 72 h under anaerobic conditions at 37°C . Total counts ( vegetative cells and spores ) and spore counts were then determined using CFU assays in 1X PBS ( 1/10 dilutions from 0 to -5 ) . All dilutions were plated onto BHIS plates supplemented with 1% taurocholate . The spore counts were performed by heat inactivation of vegetative cells at 65°C for 20 minutes , these were then serially diluted and CFU counts determined on BHIS taurocholate plates . All experiments were performed with duplicate technical replicates and triplicate biological replicates . All data was analyzed in Excel , plotted in GraphPad Prism 7 and statistical analysis was performed in Stata15 using regression analysis p<0 . 05 were considered significantly different ( S5 Fig ) . Female C57BL/6 mice ( Charles River; 7–9 weeks old ) were kept in independently ventilated cages under sterile conditions . As outlined by Theriot et al . [62] , mice were treated with cefoperazone in the drinking water ( 50 mg/litre ) for 10 days to disrupt their normal microflora , rested for two days , before they were infected with 104 C . difficile spores by oral gavage . After 28 days , vancomycin was added to the drinking water for 7 consecutive days ( 400 mg/litre ) to induce relapse of CDI . Fresh faecal samples from individually infected mice were collected throughout the time course to be utilised for determining the C . difficile load , 16S rRNA sequencing of the microflora and metabolite profiling by 1H spectroscopy . Stool samples were plated onto C . difficile selective plates to determine the bacterial load ( CFU/g ) . Statistical analysis was performed using a one tailed Mann Whitney U test , p<0 . 05 were considered significantly different . All animal procedures were performed at Royal Holloway in accordance with the Home Office project license PPL 70/8276 , that enables work to be conducted under the UK “Animal ( Scientific Procedures ) Act 1986” . This work was approved by the Royal Holloway , University of London Ethics Committee . Healthy human donor faecal samples were collected and processed using different healthy donors who had not received antibiotic treatment in the preceding 3 months in accordance with the University of Hertfordshire Ethics committee guidelines and approval ( UH Ethics Approval Number: aLMS/SF/UH/00103 ) . All donors provided informed written consent . DNA was extracted from faecal samples using a combined method based on phenol:chloroform:isoamyl alcohol extraction , ethanol precipitation and FastDNA SPIN kit for soil ( MPBiomedicals ) . Briefly , an equal mass of faecal material was suspended in 50 mM Tris-HCl pH7 . 5 , 10 mM EDTA , homogenised and bacterial cells were lysed using a FastPrep-24 Classic Instrument ( MPBiomedicals ) . Nucleic acid was extracted using a standard phenol:chloroform:isoamyl alcohol procedure , followed by ethanol precipitation and was suspended in nuclease free dH2O . Faecal DNA was subsequently purified using the DNA binding matrix from the FastDNA SPIN kit for soil with minor modifications of the manufacturer’s instructions . Briefly , DNA samples were added to sodium phosphate buffer , MT buffer and protein precipitation solution supplied in the kit and this was added directly to the binding matrix . DNA was subsequently purified according to the manufacturer’s instructions and eluted in DNase-free water . Library preparations for the MiSeq were performed as outlined in Rosser et al[63] . Briefly , an amplification step was used to add Illumina compatible adaptors , with a unique 12 bp individual barcodes for each sample , with an extra pad and linker sequence . The V5-7 regions of the 16S rRNA genes were then amplified using 785F: 5ʹ-GGATTAGATACCCBRGTAGTC-3ʹ , 1175R: 5 ʹ-ACGTCRTCCCCDCCTTCCTC-3ʹ primers , where the reverse primer ( 1175R ) contained the individual error-corrected barcode: 25 μl reactions were comprised of 1x Molzym PCR buffer , 0 . 025 μM Moltaq ( Molzym ) , 200 μM dNTPs ( Bioline ) , 0 . 4 μM forward and reverse primer , 2 μl DNA and nuclease free water ( Bioline ) [63] . Cycling parameters for each reaction were 94°C x 3 min , then 30 cycles of 94°C x 30 s , 60°C x 40 s , 72°C x 90 s and final extension at 72°C for 10 min . Samples were purified and normalised using a SeqPrep normalisation plate kit ( Invitrogen ) , and quantified using a Qubit2 . 0 ( Life technologies ) , and further purified using 0 . 6 X Agencourt AMPure Beads ( Beckman Coulter ) , a selection of samples were run on an Agilent high sensitivity DNA chip ( Agilent Technologies ) , samples were quantified again using a Qubit 2 . 0 ( Life Technologies ) , and were pooled in equimolar solution , then diluted to a 2 nM library , with 10% PhiX control and loaded into the MiSeq run cartridge in accordance with the manufacturer’s instructions ( Illumina ) . The MiSeq runs produced 250 bp paired end reads , with a 12 bp individual index for each sample . The sequence reads generated were de-multiplexed and quality filtered using QIIME ( version 1 . 9 . 1 [64] ) following the standard pipeline to assign Illumina reads to operational taxonomic units ( OTUs ) using the Greengenes database [65] ) . Associated summaries and diversity analyses were also performed in QIIME . Subsequent analyses were performed in R [66] and visualised with ggplot2 [67] . We selected families to include in our 16S plots ( Fig 7A ) if they had mean proportion of greater than 1% in any of the 12 day/type phenotype combinations . Box plots ( Fig 7B ) and PCA ( Fig 7C ) were calculated from the full 16S rRNA sequence dataset at the family level . ANOSIM analysis was used to identify variation in species abundance and composition between strains 630Δerm and hpdC::CT , as well as between time points D7 , D0R , D2R and D4R . Significant differences were indicated with a circle p<0 . 001 . Faecal samples were defrosted and mixed with 400 μL of phosphate buffer ( pH 7 . 4 , 100% D2O , 3 mM of NaN3 , 1 mM of 3- ( trimethyl-silyl ) -[2 , 2 , 3 , 3-2H4]-propionic acid ( TSP ) for the chemical shift reference at δ0 . 0 ) and Zirconium beads ( 0 . 45 g ±0 . 1 ) . The samples were vortexed and then homogenised with a FastPrep-24 Classic Instrument ( MP BIOMEDICALS ) ( 30 sec per cycle , speed 6 . 0 , 2 cycles ) . After a centrifugation ( 13 , 000 xg , 15 min ) , 180 μL of the supernatants were collected and transferred in 3-mm tubes for 1H nuclear magnetic resonance ( NMR ) spectroscopic analysis , which was performed on a Bruker 600 MHz spectrometer ( Bruker Biospin , Karlsruhe , Germany ) at 300K ( 26 . 85° ) . The parameters of the acquisition were as previously reported for urine[61] . 4 dummy scans followed per 64 scans were acquired for each spectrum which were then imported into Matlab ( version R2014a , The Mathworks Inc . ) . The region δ4 . 82–4 . 76 was removed to eliminate residual water signal . All spectra were normalised according probabilistic quotient method and automatically aligned . Principal Components Analysis ( PCA ) was performed with mean-centring and Pareto scaling . Frozen culture supernatants were defrosted on ice and were mixed in a 1:1 ratio with methanol: water , transferred to HPLC tubes and processed immediately by HPLC . Mouse faecal samples were defrosted , and added to a 2 ml screw cap tube containing 2 mm beads . These were weighed before and after addition of the faecal sample . To these , 400 μl 1:1 methanol:water was added to the pellet , then ribolysed twice using a FastPrep-24 Classic Instrument at speed 6 . 0 m/s for 30 sec . Tubes were transferred to ice and centrifuged at 14000 xg for 20 minutes . 250 μl of the supernatant was transferred to a clean sterile HPLC tube and were transferred immediately for HPLC . Each experiment was performed in triplicate . All HPLC equipment , software , solvents , columns and vials were from Thermo Fisher Scientific , UK . Separations were performed utilising an Acclaim 120 , C18 , 5 μm Analytical ( 4 . 6 x 150 mm ) and the mobile phase consisting of ammonium formate ( 10 mM , pH 2 . 7 ) and menthol ( v/v; 40:60 ) at a flow rate of 2 ml/min . p-HPA and p-cresol were detected by the photo-diode array detector ( UV-PDA; DAD 3000 ) set at 280 nm . Peak identity was confirmed by measuring the retention time , spiking the sample with commercially available p-HPA and p-cresol and determination of absorbance spectra using the UV-PDA . A calibration curve of each compound was generated by Chromeleon ( Dionex software ) using known amounts of the reference standards ( 0–100 mg/ml ) in methanol/water ( v/v; 1:1 ) injected onto the column to determine the amount in the samples . The lower limit of detection was determined for p-HPA to be 0 . 03 mg/ml , for p-cresol to be 0 . 02 mg/ml . The concentration in mM was determined in Excel , using the molecular weight of the compounds and the quantity in mg/ml . The data was analysed in GraphPad Prism7 and statistical analysis was performed in Stata15 using linear regression analysis . Healthy human donor faecal samples were collected and processed using three different healthy donors who had not received antibiotic treatment in the preceding 3 months . Faecal samples ( 5 g ) were emulsified in sterile pre-reduced PBS ( 50 ml ) and faecal material was coarse filtered by passing the 10% emulsion through sterile muslin cloth to remove larger particle matter and leave a bacterial suspension . Faecal emulsions were incubated for 1 hour and 30 minutes in 1X PBS , or PBS containing 0 . 1% ( v/v ) p-cresol or 0 . 3% ( v/v ) p-cresol . Samples were then sedimented by centrifugation at 14000 x g for 5 minutes and the supernatant were removed . Pellets were resuspended in 1 ml 1X PBS and viable counts ( CFU/ml ) were performed on differentially selective agar , both anaerobically and aerobically . Each experiment was performed in triplicate . Serial 10-fold dilutions of re-suspended faecal emulsions in sterile pre-reduced peptone water were inoculated onto: fastidious anaerobe agar ( total anaerobes ) , nutrient agar ( total facultative anaerobes ) , kanamycin aesculin azide agar ( Enterococci ) , LAMVAB agar ( Lactobacilli ) , Beeren’s agar ( Bifidobacteria ) , MacConkey agar number 3 ( lactose-fermenting Enterobacteriaceae ) , Bacteroides bile aesculin agar ( B . fragilis group ) , and total viable counts were determined in triplicate , and normalised to the starting CFU . Release of cellular phosphate was investigated using a Colorimetric Phosphate Assay Kit ( Abcam ) . The assay involved treating samples with ammonium molybdate and malachite green which forms a chromogenic complex with phosphate ions which can be detected at a wavelength of 650 nm . An overnight culture of each bacterial strain was sedimented by centrifugation and re-suspended in Tris-buffered saline ( TBS , 50 mM Tris-HCl pH7 . 5 , 150 mM NaCl ) and subsequently washed two further times to remove traces of the growth medium . OD595 was determined and cell suspensions were normalised to an OD595 of 1 . 0 . Five hundred microliter aliquots of cell suspension were sedimented by centrifugation and re-suspended in either TBS alone or TBS + p-cresol . Cell suspensions were incubated for the indicated time , cells were sedimented and 30 μl of supernatant was removed and added to 170 μl H2O and 30 μl ammonium molybdate and malachite green reagent . Absorbance was read at 650 nm in a 96-well microtitre plate reader . Phosphate release was determined by normalising the optical density from cell suspensions incubated with p-cresol against cell suspensions that were incubated with TBS alone . The assay was performed under anaerobic conditions except for spectrophotometry and sedimentation steps , for which tubes and flasks were sealed with parafilm to prevent oxygen infiltration . The maximum intracellular phosphate pool was determined by boiling a 500 μl cell suspension ( OD595 1 . 0 ) for 15 minutes . All assays were performed in triplicate . | Clostridium difficile is a bacterium responsible for causing the majority of antibiotic associated diarrhoea outbreaks world-wide . In the United States of America , C . difficile infects half a million people annually . Antibiotics disrupt the natural protective gut microbiota , rendering people susceptible to C . difficile infection , which leads to potentially life-threatening disease and complications . C . difficile is transmitted by spores , which are able to survive in harsh environments for long periods of time . After initial treatment for C . difficile , up to 35% of patients develop the disease again , thus requiring additional and more successful treatment . Here , we use novel techniques to show that C . difficile produces a compound , p-cresol , which has detrimental effects on the natural protective gut bacteria . We show that p-cresol selectively targets certain bacteria in the gut and disrupts their ability to grow . By removing the ability of C . difficile to produce p-cresol , we show that it makes C . difficile less able to recolonise after an initial infection . This is linked to significant alterations in the natural healthy bacterial composition of the gut . Our study provides new insights into the effects of p-cresol production on the healthy gut microbiota and how it contributes to C . difficile survival and pathogenesis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"bacteriology",
"gram",
"negative",
"bacteria",
"medicine",
"and",
"health",
"sciences",
"gut",
"bacteria",
"microbiome",
"chemical",
"compounds",
"phosphates",
"nmr",
"spectroscopy",
"microbiology",
"animal",
"models",
"model",
"organisms",
"metabolites",
"experimental",
"organism",
"systems",
"bacteria",
"microbial",
"genomics",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"clostridium",
"difficile",
"spectrum",
"analysis",
"techniques",
"medical",
"microbiology",
"chemistry",
"mouse",
"models",
"gastrointestinal",
"tract",
"biochemistry",
"anatomy",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"genomics",
"metabolism",
"organisms"
]
| 2018 | Para-cresol production by Clostridium difficile affects microbial diversity and membrane integrity of Gram-negative bacteria |
New frontier settlements across the Amazon Basin pose a major challenge for malaria elimination in Brazil . Here we describe the epidemiology of malaria during the early phases of occupation of farming settlements in Remansinho area , Brazilian Amazonia . We examine the relative contribution of low-density and asymptomatic parasitemias to the overall Plasmodium vivax burden over a period of declining transmission and discuss potential hurdles for malaria elimination in Remansinho and similar settings . Eight community-wide cross-sectional surveys , involving 584 subjects , were carried out in Remansinho over 3 years and complemented by active and passive surveillance of febrile illnesses between the surveys . We used quantitative PCR to detect low-density asexual parasitemias and gametocytemias missed by conventional microscopy . Mixed-effects multiple logistic regression models were used to characterize independent risk factors for P . vivax infection and disease . P . vivax prevalence decreased from 23 . 8% ( March–April 2010 ) to 3 . 0% ( April–May 2013 ) , with no P . falciparum infections diagnosed after March–April 2011 . Although migrants from malaria-free areas were at increased risk of malaria , their odds of having P . vivax infection and disease decreased by 2–3% with each year of residence in Amazonia . Several findings indicate that low-density and asymptomatic P . vivax parasitemias may complicate residual malaria elimination in Remansinho: ( a ) the proportion of subpatent infections ( i . e . missed by microscopy ) increased from 43 . 8% to 73 . 1% as P . vivax transmission declined; ( b ) most ( 56 . 6% ) P . vivax infections were asymptomatic and 32 . 8% of them were both subpatent and asymptomatic; ( c ) asymptomatic parasite carriers accounted for 54 . 4% of the total P . vivax biomass in the host population; ( d ) over 90% subpatent and asymptomatic P . vivax had PCR-detectable gametocytemias; and ( e ) few ( 17 . 0% ) asymptomatic and subpatent P . vivax infections that were left untreated progressed to clinical disease over 6 weeks of follow-up and became detectable by routine malaria surveillance .
Malaria is one of the major tropical infectious diseases for which decades of intensive control efforts have met with only partial success in Brazil [1] . With nearly 243 , 000 slide-confirmed infections , this country contributed 52% of all malaria cases reported in the Region of the Americas and the Caribbean in 2012 [2] . Most transmission in Brazil occurs in open mining enclaves , logging camps and farming settlements across the Amazon Basin , a region that currently accounts for 99 . 9% of the country-wide malaria burden [3] . Since the early 1970s , official and informal colonization projects in densely forested areas of Amazonia have attracted migrant farmers from the malaria-free South and Southeast regions , originating a number of new frontier agricultural settlements [4] , [5] . Initial land clearing for slash-and-burn agriculture and extensive logging can induce major changes in vector biology , by creating or expanding mosquito breeding habitats , as well as in vector species composition , with a marked increase in the abundance of the highly competent local malaria vector Anopheles darlingi [6]–[10] . Not surprisingly , recent frontier settlements , where ongoing deforestation and the immigration of non-immune pioneers favor transmission , constitute malaria hotspots until these communities become more stable and endemicity declines [6] . Plasmodium falciparum and P . vivax infections are widespread across Amazonia , with rare and focal P . malariae transmission [11]–[14] . A clear change has been recently observed in the relative proportion of the two main species . Similar proportions of slide-confirmed infections were due to P . falciparum and P . vivax until 1990 , but transmission of the latter species maintained an upward trend while that of P . falciparum declined steadily throughout the next decade [15] . Plasmodium vivax now accounts for 85% of the malaria burden in Brazil [2] . These trends may be explained by factors such as the presence of dormant liver stages ( hypnozoites ) and the early circulation of sexual stages ( gametocytes ) in peripheral blood , which may render P . vivax less responsive than P . falciparum to available control strategies based on early diagnosis and treatment of blood-stage infections [16] , [17] . Here we describe the epidemiology of malaria and associated risk factors during the early phases of occupation of frontier agricultural settlements in the Amazon Basin of Brazil . We observed a major decline in P . vivax prevalence , with vanishing P . falciparum transmission , over 3 years of malaria surveillance . Risk of both infection and P . vivax-related disease decreased with increasing cumulative exposure to malaria , consistent with anti-parasite and anti-disease immunity being acquired by this population . We discuss the challenges of controlling and eliminating malaria , especially that caused by the resilient parasite P . vivax , in low-endemicity areas where most infections are asymptomatic and parasite densities are often below the detection threshold of conventional microscopy .
Once a sparsely populated rubber tapper settlement ( seringal ) situated in southern Amazonas state , northwestern Brazil , Remansinho ( average population , 260 ) now comprises five farming settlements ( Figure 1 ) . The main settlement is situated along the final 40 km of the Ramal do Remansinho , a 60 km-long unpaved road originating from the BR-364 interstate highway , while the other four are situated along secondary roads ( known as Ramal da Linha 1 , Ramal da Castanheira , Ramal dos Seringueiros , and Ramal dos Goianos ) originating from this main unpaved road ( Figure S1 ) . The farming settlements along Ramal da Linha 1 and Ramal da Castanheira were opened in the late 1990s , whereas the colonization of the other areas started only in 2007 . Most houses have complete or incomplete wooden walls and thatched roofs; just a few of them have brick walls and are covered with asbestos , cement or zinc shingles . With a typical equatorial humid climate ( annual average temperature , 26 . 4°C ) , Remansinho receives most rainfall between November and March ( annual average , 2 , 318 mm ) , but malaria transmission occurs year-round . The main local malaria vector is the highly anthropophilic and exophilic An . darlingi [18] . Most families currently living in Remansinho have resettled from other areas within Amazonia , and are now involved in subsistence agriculture and logging . There is a single government-run health post in Remansinho , which provides free malaria diagnosis and treatment , but a small proportion of locally acquired infections are diagnosed and treated in the nearest village ( Nova Califórnia; population , 2 , 600 ) , situated along the BR-364 highway , about 60 km south of Remansinho ( Figure 1 ) . There is no electricity or piped water supply in the area . A population-based prospective cohort study was initiated in March 2010 to estimate the prevalence and incidence of malaria parasite carriage in Remansinho , by combining microscopy and molecular diagnosis , and to characterize risk factors for malaria infection and clinical disease in the local population . This ongoing study comprises periodic cross-sectional malaria prevalence surveys of the entire population , every four months between March 2010 and March 2011 and every six months thereafter , complemented with clinical and laboratory surveillance of febrile illness episodes between the cross-sectional surveys . Here we analyze data collected from March 2010 to May 2013 . During this period , we enrolled 584 participants belonging to 205 households . Dwellings were geo-localized using a hand-held 12-chanel global positioning system ( GPS ) receiver ( eTrex Personal Navigator , Garmin , Olathe , KS ) , with a positional accuracy within 15 m . Nearly all ( 98 . 8% ) study subjects were recruited during house-to-house visits in Ramal do Remansinho ( 376 or 65 . 2% ) , Ramal da Castanheira ( 85 or 14 . 7% ) , Ramal da Linha 1 ( 57 or 9 . 9% ) , Ramal dos Goianos ( 32 or 5 . 5% ) , and Ramal dos Seringueiros ( 27 or 4 . 7% ) ; only 7 ( 1 . 2% ) subjects , who were enrolled at the local health post , had their settlement of origin undetermined . Each cross-sectional survey comprised a population census and the entire population found during the census was considered eligible to participate in the study . During the first ( March–May 2010 ) , 165 inhabitants identified during the census in Ramal do Remansinho and Ramal dos Goianos were invited to participate . Subsequent surveys , which also included subjects living in the other three settlements , were carried out between May and July 2010 ( survey 2 ) , October and November 2010 ( survey 3 ) , March and April 2011 ( survey 4 ) , October and November 2011 ( survey 5 ) , April and May 2012 ( survey 6 ) , October and November 2012 ( survey 7 ) , and April and May 2013 ( survey 8 ) . Most surveys were carried out either at the beginning or the end of the rainy season , except for survey 2 , which took place during the dry season . Total numbers of subjects present in the study area during each survey are given in Table 1 . A baseline questionnaire was applied to all study participants in March–May , 2010 , to collect demographic , health , behavioral and socioeconomic data . Cumulative exposure to malaria was estimated using the duration of residence in Amazonia as a proxy . We used a structured questionnaire [19] to determine the presence and intensity of 13 malaria-related signs and symptoms ( fever , chills , sweating , headache , myalgia , arthralgia , abdominal pain , nausea , vomiting , dizziness , cough , dyspnea , and diarrhea ) up to seven days prior to the interview . Information on selected household assets , access to utilities , infrastructure , and housing characteristics was used to derive a wealth index , from which socioeconomic status was estimated . We combined discrete ( i . e . , yes or no ) ownership information ( for power generator , chainsaw , radio , sofa set , shotgun , bicycle , car , motorcycle , and well ) and continuous data ( i . e . , total number of items , for beds , rooms and bedrooms present in the household , and number of pigs , cattle , chickens , ducks , and horses owned ) . Principal component analysis , carried out using statistical software STATA 12 . 1 , was used to weight each variable [20] . The first principal component explained 18% of the variability and gave the greatest weights to ownership of beds , number of rooms , number of bedrooms , sofa set , and chickens . Lowest weights were given to ownership of horses , ducks or cattle . The scores were summed to give a wealth index for each household . Wealth indices were then used to stratify households into quartiles in increasing order ( first quartile , 25% poorest ) . A shorter version of the baseline questionnaire was used in all subsequent cross-sectional surveys to update demographic and clinical data . All inhabitants in the study area aged more than 3 months were invited to contribute either venous ( 5-ml ) or finger-prick blood samples for malaria diagnosis , irrespective of any clinical symptoms , Duffy blood group genotyping , and other laboratory assays , such as hemoglobin measurements and ABO and Rh blood group typing . The participation rates ranged between 96 . 3% in survey 1 ( 159 of 165 inhabitants ) and 70 . 3% in survey 5 ( 204 of 290 ) ( Table 1 ) . Nearly all study participants provided venous blood samples in all but one survey; the exception was survey 3 , during which finger-prick capillary blood was preferentially collected from all participants for logistic reasons . Reasons for not providing blood samples included temporary absence from the study area , age below 3 months , inability to perform venous puncture , and refusal to participate . Given the high mobility of the study population , only 21 subjects ( 3 . 6% of the study population ) contributed blood samples in all cross-sectional surveys; 529 subjects ( 90 . 6% ) participated in two or more surveys . All study participants , either symptomatic or not , who provided either venous or finger-prick blood samples during cross-sectional surveys and had malaria diagnosis confirmed by onsite microscopy were treated according to the malaria therapy guidelines published by the Ministry of Health of Brazil in 2010 [21] . Briefly , P . vivax infections were treated with chloroquine ( total dose , 25 mg of base/kg over 3 days ) and primaquine ( 0 . 5 mg of base/kg/day for 7 days ) , while P . falciparum infections were treated with a fixed-dose combination of artemether ( 2–4 mg/kg/day ) and lumefantrine ( 12–24 mg/kg/day ) for 3 days . Infections that were missed by onsite microscopy but later confirmed by polymerase chain reaction ( PCR ) were left untreated because the results of molecular diagnosis were not available at the time of the cross-sectional surveys . To quantify clinical malaria episodes diagnosed between the cross-sectional surveys , we examined all records of slide-confirmed infections diagnosed between March 2010 and November 2013 at the government-run health posts in Remansinho and in the nearest village , Nova Califórnia . Local malaria control personnel performed both active and passive detection of febrile cases during the study period . Blood samples were collected and examined for malaria parasites whenever febrile subjects visited the health posts in Remansinho or Nova Califórnia or were found during monthly house-to-house visits carried out by field health workers in Remansinho . This strategy is assumed to detect virtually all clinical malaria episodes in cohort subjects between the cross-sectional surveys , since there are no other public or private facilities providing laboratory diagnosis of malaria in the area . Microscopic diagnosis is required to obtain antimalarial drugs in Brazil , which are distributed free of charge by the Ministry of Health and cannot be purchased in local drugstores . Laboratory diagnosis of malaria was based on microscopic examination of thick smears and PCR . A total of 1 , 541 thick blood smears were stained with Giemsa in our field laboratory in Acrelândia ( 120 km southwest of Remansinho ) . At least 100 fields were examined for malaria parasites , under 1000× magnification , by two experienced microscopists , before a slide was declared negative . We additionally used quantitative real-time PCR ( qPCR ) that target the 18S rRNA genes [22] to detect and quantify P . vivax and P . falciparum blood stages in 1 , 476 clinical samples ( Methods S1 ) . Because microscopy is poorly sensitive for detecting circulating gametocytes [23] , we used a quantitative reverse transcriptase PCR ( qRT-PCR ) that targets pvs25 gene transcripts [24] , [25] to detect and quantify mature gametocytes in 55 laboratory-confirmed P . vivax infections diagnosed during cross-sectional surveys 4 , 5 , and 6 ( Methods S1 ) . Since co-infection with multiple parasite clones has been suggested to either increase or reduce the risk of clinical falciparum malaria , we sought to determine whether the presence of multiple-clone P . vivax infections was associated with malaria-related morbidity . To this end , we amplified two highly polymorphic single-copy markers , msp1F1 ( a variable domain of the merozoite surface protein-1 gene ) [26] and MS16 ( a P . vivax-specific microsatellite DNA marker with degenerate trinucleotide repeats ) [27] , using the nested PCR protocols of Koepfli and colleagues [28] . DNA samples from 85 qPCR-confirmed P . vivax infections ( all of them isolated from venous blood samples ) were tested for the presence of multiple clones; 47 were from asymptomatic and 38 from symptomatic parasite carriers . PCR products were analyzed by capillary electrophoresis on an automated DNA sequencer ABI 3500 ( Applied Biosystems ) , and their lengths ( in bp ) and relative abundance ( peak heights in electropherograms ) were determined using the commercially available GeneMapper 4 . 1 ( Applied Biosystems ) software . The minimal detectable peak height was set to 200 arbitrary fluorescence units . We scored two alleles at a locus when the minor peak was >33% the height of the predominant peak . Plasmodium vivax infections were considered to contain multiple clones if one or both loci showed more than one allele . Since Duffy blood group polymorphisms modulate the ability of P . vivax merozoites to invade human red blood cells ( reviewed by [29] ) , we used TaqMan assays ( Applied Biosystems ) to genotype two major Duffy polymorphisms: the T-33C substitution in the red blood cell-specific GATA1 transcription factor binding motif ( rs2814778 ) , which suppresses Duffy expression on the erythrocyte surface ( Fy phenotype , associated with FY*BES allele homozygozity ) , and the G125A polymorphism ( rs12075 ) , which defines the FY*B ( wild-type ) and FY*A ( mutated ) alleles associated with the Fyb and Fya phenotypes , respectively . The primers and probes ( labelled with VIC and FAM ) were designed and synthesized by Applied Biosystems ( assay ID , C__15769614_10 and C__2493442_10 ) [30] . We used a Step One Plus Real-Time PCR System ( Applied Biosystems ) for genotyping , with a template denaturation step at 95°C for 10 min , followed by 40 cycles of 15 sec at 95°C and 1 min at 60°C , with a final step at 60°C for 30 sec . DNA samples from 487 study participants were genotyped . A laboratory-confirmed malarial infection was defined as any episode of parasitemia detected by thick-smear microscopy , qPCR , or both . Subpatent or submicroscopic infections were defined as infections confirmed by qPCR but missed by microscopy . We defined clinical malaria as a laboratory-confirmed infection , irrespective of the parasite density , diagnosed in a subject reporting one or more of the 13 signs and symptoms investigated , at or up to seven days before the interview . No attempt was made to calculate pyrogenic thresholds of parasitemias in our heterogeneous group of study participants . Subjects with laboratory-confirmed infection , irrespective of the parasite density , who reported no signs or symptoms at or up to seven days prior to the interview , were classified as asymptomatic carriers of malaria parasites . A database was created with SPSS 17 . 0 software ( SPSS Inc . , Chicago , IL ) and all subsequent analyses were performed with R statistical software [31] . For the purposes of explanatory data analysis , proportions were compared using standard χ2 , Mantel-Haenzel χ2 for stratified data , or χ2 tests for linear trends . Correlations between parasite densities , which had an overdispersed distribution in the population , and other continuous variables were evaluated using the nonparametric Spearman correlation . Median parasitemias were compared with the nonparametric Mann-Whitney U test . Statistical significance was defined at the 5% level and 95% confidence intervals ( CI ) were estimated whenever appropriate . Separate regression models were built to describe independent associations between potential risk factors and two outcomes: ( a ) P . vivax infection and ( b ) clinical ( i . e . , symptomatic ) vivax malaria . Due to the small number of P . falciparum infections detected in the community no attempt was made to characterize risk factors for infection with this species . Dependent variables were assumed to follow a binomial distribution with a logit link function , being fitted with a logistic regression . We considered the nested structure of the data , intrinsic to the study design , when building regression models; we have repeated observations ( up to 8 observations over 3 years of study; grouping variable , “survey” ) nested within subjects ( grouping variable , “individual” ) who are clustered within households ( grouping variable , “household” ) . This clustered sampling scheme introduces dependency among observations that can affect model parameter estimates . Consequently , we used mixed-effects regression models that include the grouping variables as random factors to handle nested observations . Our modeling strategy further considered the hierarchical levels of independent variables ( Methods S1 ) . The effects of distal determinants , such as demographic , social and environmental factors , on malaria risk are often not direct , but mediated by more proximate determinants , such as occupational and behavioral factors [32] . Variables within each level of determination were introduced in the model in a stepwise approach , and only those that were associated with the outcome at a significance level of at least 20% were retained . Most subjects with missing observations were excluded from the final model , except those with missing values for the following four variables: Duffy genotype , wealth index , whether bedroom windows were left open at night , and main occupation . These were maintained in the model by creating a new missing-value category . All models were adjusted for the timing of the survey ( months elapsed since the beginning of the study in March 2010 ) . Three variables in the model were time-dependent: age , years of residence in Amazonia , and timing of the survey . The final models comprised 1 , 242 observations from 442 individuals grouped into 159 households ( outcome: P . vivax infection ) , and 1 , 237 observations from 438 individuals grouped into 158 households ( outcome: vivax malaria ) . Alternative logistic models , which excluded Duffy-negative subjects ( 88 observations from 31 subjects ) , examined the association between Duffy-positive genotypes ( FY*AFY*BES , FY*AFY*A , FY*AFY*B , FY*BFY*BE , and FY*BFY*B ) and risk of P . vivax infection and vivax malaria . To account for the hypothesis that age at the beginning of exposure to malaria affects the rate at which antimalarial immunity is acquired by migrants [33] , we further tested for an interaction between age and years of residence in Amazonia . In addition , we fitted mixed-effects Poisson regression models to the data , but the random-effects variances associated with the estimates were substantially higher than those obtained with the logistic models described above . As a consequence , here we only present results derived from the logistic regression analysis . In addition , we used a mixed-effects Cox proportional hazards model [34] to compare the risk of slide-positive vivax malaria between the surveys in two sub-cohorts of asymptomatic subjects: ( a ) carriers of subpatent P . vivax infections at baseline that were left untreated and ( b ) control subjects who were parasite-negative at baseline by both microscopy and qPCR . Subjects who were symptomatic but parasite-negative at baseline were excluded from the uninfected sub-cohort because they might harbor ongoing low-grade infections , causing malaria-related symptoms , which were missed by our laboratory methods . At each survey , eligible study participants were assigned to either sub-cohort and followed up until the next survey at which their clinical and infection status was reassessed . Time at risk was defined as either the interval between two consecutive surveys in which the subjects participated ( the first survey in the pair was defined as the baseline survey ) or the interval between the baseline survey and the date when subjects left the study , whatever came first . Analysis was adjusted for subjects' age ( stratified as <15 years and ≥15 years ) , Duffy blood group negativity , and years of residence in Amazonia . The clustering of repeated observations within individuals was modeled as a random effect [34] . As required for all observational studies published by PLoS Neglected Tropical Diseases , this article includes the STROBE ( STrengthening the Reporting of OBservational studies in Epidemiology ) checklist to document its compliance with STROBE guidelines ( Checklist S1 ) . Study protocols were approved in early 2010 by the Institutional Review Board of the University Hospital of the University of São Paulo ( 1025/10 ) and by the National Human Research Ethics Committee of the Ministry of Health of Brazil ( 551/2010 ) . The ethical clearance has been renewed annually by the Institutional Review Board of the University Hospital of the University of São Paulo . Written informed consent was obtained from all study participants or their parents/guardians .
Of 584 people living in Remansinho who participated in at least one cross-sectional survey , 333 ( 57 . 0% ) were male and 251 ( 43 . 0% ) were female , with a median age of 23 . 0 years . Nearly all ( 94 . 3% ) adult subjects aged more than 18 years were migrants , 42 . 2% of them originating from malaria-free areas outside Amazonia . Only 31 subjects ( 6 . 4% ) were homozygous FY*BES carriers , with the P . vivax-refractory Duffy-negative ( Fy ) phenotype; 127 ( 26 . 1% ) had the Fya phenotype ( 70 FY*AFY*BES heterozygotes and 57 FY*A FY*A homozygotes ) , 142 ( 29 . 2% ) had the FyaFyb phenotype ( FY*A FY*B heterozygotes ) , and 187 ( 38 . 4% ) had the Fyb phenotype ( 91 FY*BFY*BES heterozygotes and 96 FY*B FY*B homozygotes ) . Polyethylene bed-nets treated with 2% permethrin ( Olyset Net , Sumitomo Chemical , London , United Kingdom ) were distributed , free of charge , to the entire study population in August 2012 , as a component of malaria control activities in Brazilian Amazonia . In October–November 2012 ( survey 7 ) , 74 . 4% of the study participants reported having slept the previous night under an Olyset net; the corresponding figure for April–May 2013 ( survey 8 ) was 84 . 5% . No other insecticide-treated bed nets were available in the community . A total of 1 , 541 blood samples were examined for malaria parasites by microscopy , qPCR , or both . Of these , 141 ( 9 . 1% ) were positive ( by one or both methods ) for P . vivax , 40 ( 2 . 6% ) for P . falciparum and 10 ( 0 . 6% ) for both species . Over the entire study period , 191 ( 12 . 4% ) samples examined tested positive for malaria parasites; 10 P . vivax and 2 P . falciparum infections were only diagnosed by microscopy , since DNA samples were not available for qPCR or qPCR yielded negative results . In addition , 61 . 8% of all infections diagnosed by qPCR , regardless of the infecting species , and 49 . 6% of the qPCR-confirmed single-species P . vivax infections , were missed by conventional microscopy and thus defined as subpatent . The last P . falciparum infections in Remansinho were diagnosed ( by qPCR only ) in March–April 2011 . These figures , however , changed over time . The numbers of malaria infections , either symptomatic or not , diagnosed by conventional microscopy and qPCR in each cross-sectional survey are shown in Table 1 . The proportions of qPCR-confirmed single-species P . vivax infections that were subpatent varied widely across surveys , ranging from 73 . 1% in the surveys with the lowest P . vivax prevalence rates ( surveys 4 , 6 , 7 , and 8 combined; 26 qPCR-confirmed infections ) to 43 . 8% in those with the highest prevalence rates ( surveys 1 , 2 , 3 , and 5 combined; 105 qPCR-confirmed infections; Yates' corrected χ2 = 6 . 02 , 1 degree of freedom [df] , P = 0 . 014 ) . The numbers of P . falciparum and mixed-species infections were too small for a similar comparison . Microscopy thus had a better diagnostic performance for vivax malaria when overall parasite prevalence rates were higher , consistent with a recent meta-analysis of P . falciparum data showing lower proportions of submicroscopic infections in areas with greater malaria transmission [35] . Overall , 17 . 1% of the study subjects ( ranging between 12 . 3% in survey 6 and 39 . 6% in survey 1 ) interviewed during the cross-sectional surveys reported one or more malaria-related signs and symptoms up to seven days prior to the interview ( Table 1 ) . However , reported clinical signs and symptoms were neither sensitive nor specific for malaria diagnosis . On the one hand , almost two thirds ( 64 . 5% ) of all qPCR-confirmed malaria infections by any species , and 56 . 6% of those due to P . vivax , were asymptomatic; on the other hand , only 26 . 7% of subjects reporting symptoms had a malaria infection ( by any species ) confirmed by microscopy , qPCR , or both . All carriers of mixed-species infections ( all of them confirmed by qPCR but missed by microscopy ) were asymptomatic ( Table 1 ) . Most P . vivax-infected subjects harbored few parasites , with densities estimated by qPCR on 129 samples ranging between 2 . 1 and 38 , 390 parasites/µL ( median , 49 . 1 parasites/µL; interquartile range , 10 . 0–483 . 1 parasites/µL; data were missing for 2 qPCR-confirmed infections ) . We found no evidence for decreasing P . vivax densities with increasing cumulative exposure to malaria in this population . In fact , individual P . vivax parasitemias did not show a negative correlation with the subjects' length of residence in Amazonia , a proxy of cumulative exposure to malaria ( Spearman correlation coefficient rs = −0 . 046 , P = 0 . 600 ) , or with their age ( rs = −0 . 068 , P = 0 . 427 ) . We next tested whether differences in the diagnostic sensitivity of conventional microscopy across cross-sectional surveys might be explained by higher average parasite densities found at times of increased malaria transmission [35] . Parasitemias appeared slightly higher in qPCR-positive samples ( P . vivax only ) obtained during surveys 1 , 2 , 3 and 5 ( high prevalence ) , with a median of 55 . 7 parasites/µL ( interquartile range , 10 . 4–597 . 6 parasites/µL; n = 103 ) , than in those obtained during surveys 4 , 6 , 7 , and 8 ( low prevalence ) , with a median of 19 . 8 parasites/µL ( interquartile range , 5 . 8–65 . 4 parasites/µL; n = 26 ) , although the difference did not reach statistical significance ( Mann-Whitney U test , P = 0 . 057 ) . The proportion of symptomatic P . vivax infections correlated positively with increasing parasite density ( χ2 for trend = 7 . 99 , 1 df , P<0 . 005 ) . Only 30 . 6% of the subjects carrying less than 10 parasites/µL , but 73 . 9% of those carrying more than 1 , 000 parasites/µL , reported one or more malaria-related symptoms ( Figure 2 ) . Consistent with previous observations from Amazonia [36] , [37] , more than half ( 53 . 9% ) of the asymptomatic infections with this species confirmed by qPCR were missed by conventional microscopy ( Table 1 ) . Overall , 32 . 8% of the 131 single-species , qPCR-confirmed P . vivax infections for which complete data were available were both subpatent and asymptomatic ( Figure 3 ) . Only one P . vivax infection was diagnosed by qPCR , but missed by conventional microscopy , among 88 samples collected from Duffy-negative study participants during the 8 cross-sectional surveys . The only reported symptom during this subpatent P . vivax infection in a Duffy-negative subject was a chronic myalgia; parasite density was very low ( 9 . 9 parasites/µL of blood ) . To estimate the relative contribution of asymptomatic parasite carriage to the total P . vivax biomass in the host population , we summed up all individual qPCR-derived P . vivax densities and calculated the fraction corresponding to asymptomatic infections . Assuming that , on average , asymptomatic and symptomatic subjects have similar whole blood volumes , we concluded that most ( 54 . 4% ) P . vivax blood stages circulating in Remansinho at the time of the surveys were found in apparently healthy subjects who were unlikely to have their infection diagnosed through case detection strategies targeting only febrile subjects . The number of P . falciparum and mixed-species infections was too small for similar analyses . Next , we examined whether P . vivax gametocyte carriage was similarly frequent in symptomatic and asymptomatic infections . We detected pvs25 gene transcripts , consistent with mature P . vivax gametocytes circulating in the bloodstream , in all 32 symptomatic infections , and in 21 of 23 ( 91 . 3% ) asymptomatic infections from which cryopreserved blood samples were available for RNA extraction . Interestingly , 25 of 27 ( 92 . 6% ) subjects with subpatent P . vivax parasitemia , and all 28 subjects with patent infection , had pvs25 gene transcripts detected by qRT-PCR . Therefore , qRT-PCR failed to detect pvs25 transcripts in only 2 ( 3 . 6% ) of 55 samples tested ( Figure 3 ) , both of them collected from asymptomatic carriers of low parasitemias ( 6 . 3 and 11 . 0 parasites/µL ) . Not surprisingly , the number of pvs25 transcripts per µL of blood , measured by qRT-PCR , correlated positively with the qPCR-derived overall parasite density ( rs = 0 . 445 , P<0 . 0001 ) . The mixed-effects logistic regression model showed that the risk of P . vivax infection decreased with increasing cumulative exposure to malaria , consistent with anti-parasite immunity being acquired in this population ( Table 2 ) . Each additional year of residence in Amazonia decreased the average odds of being infected by 2% ( Figure 4 ) . There was no significant interaction between age and years of residence in Amazonia ( P = 0 . 9008 ) , suggesting that the subjects' age when exposure started did not affect , in this migrant population , the rate of decline in P . vivax infection risk with increasing time of residence in Amazonia . Calendar time was also a major determinant of infection risk; each month elapsed since March 2010 was associated with a 7% decrease in the odds of being infected ( Table 2 ) . Moreover , the grouping variable “survey” accounted for 99 . 9% of the random effect variance in the mixed-effects model , with minor contribution of individual- and household-level grouping . Interestingly , adjusting for more proximate determinants affected the association between age and risk of infection with P . vivax in multivariate models . Age under 15 years was a protective factor of borderline significance ( partially adjusted OR = 0 . 616; 95% CI , 0 . 33–0 . 95 , P = 0 . 057 ) in the first model , which also adjusted for sex , years of residence in Amazonia , Duffy blood group genotype , months elapsed since the beginning of the study , and wealth index quartiles . However , after adjusting for main occupation , the effect of age on infection risk became non-significant ( fully adjusted OR = 1 . 169; 95% CI , 0 . 63–2 . 19 , P = 0 . 624 ) . These results indicate that young age per se is not protective , but young subjects are less likely to engage in activities such as logging and fishing in the fringes of the rain forest , which are potentially associated with increased risk of infection ( Table 2 ) . Not surprisingly , Duffy-negativity emerged as a protective factor against P . vivax infection in this community ( Table 2 ) . However , additional logistic regression models including only Duffy-positive subjects showed that , compared to FY*A FY*B heterozygotes , neither FY*A FY*BES heterozygotes ( OR = 0 . 864; 95% CI , 0 . 43–1 . 73 ) and FY*A FY*A homozygotes ( OR = 0 . 921; 95% CI , 0 . 48–1 . 75 ) were protected against P . vivax infection , nor FY*B FY*BES heterozygotes ( OR = 1 . 226; 95% , 0 . 69–2 . 17 ) and FY*B FY*B homozygotes ( OR = 0 . 588; 95% CI , 0 . 32–1 . 09 ) were at increased risk of infection . These results are consistent with a protective role of FY*BES heterozygosity , but not of FY*A allele carriage , against P . vivax infection in this population . The risk of clinical P . vivax malaria decreased with increasing cumulative exposure to malaria ( Table 2 ) ; each additional year of residence in Amazonia decreased the odds of having vivax malaria by 3% , again with no significant interaction between age and length of residence in Amazonia ( P = 0 . 863 ) . These findings are consistent with similar exposure-dependent rates of acquisition of anti-parasite and anti-disease immunity in this community . Calendar time was the only other major determinant of malaria risk; each month elapsed since the beginning of the study was associated with an 8% decrease in the odds of having clinical vivax malaria ( Table 2 ) . Due to the small sample size , Duffy-negativity emerged as a protective factor of borderline significance ( OR = 0 . 16 , 95% CI , 0 . 02–1 . 29 , P = 0 . 084 ) against clinical vivax malaria . In Brazil , malaria is only treated if blood smear microscopy is positive; subpatent malaria parasitemia as determined with qPCR is not accepted as the basis for treatment . Of 53 asymptomatic subpatent P . vivax infections diagnosed at baseline , 9 ( 17 . 0% ) progressed to clinical malaria over the following 6 weeks , being diagnosed by onsite microscopy and treated ( Figure 5 ) . During this 6-week period , only 2 . 5% of the subjects in the uninfected cohort experienced an episode of slide-confirmed vivax malaria , but at the end of the follow-up period similar proportions of subjects in each sub-cohort had experienced vivax malaria episodes confirmed by microscopy ( Figure 5 ) . A Cox proportional hazards model revealed no significant difference , between the two sub-cohorts , in overall risk of vivax malaria episodes , after controlling for potential confounders ( hazard ratio = 1 . 07; 95% CI , 0 . 52–2 . 22 , P = 0 . 840 ) . Most subpatent asymptomatic infections cleared spontaneously ( or , at least , became undetectable by qPCR ) , since only 5 of 44 ( 11 . 4% ) carriers who remained untreated were again P . vivax-positive in the next survey . Therefore , few asymptomatic and subpatent P . vivax infections eventually became patent and symptomatic ( and therefore detectable by routine malaria surveillance ) over the following weeks . We conclude that untreated , low-density , and asymptomatic P . vivax parasitemias may persist for several weeks without progressing to clinical disease , and thus constitute a major infectious reservoir for continued transmission in the community . By typing two highly polymorphic markers , we found more than one genetically distinct clone in 25 of 85 ( 29 . 4% ) P . vivax infections analyzed . Although multiple-clone infections were more frequent in symptomatic ( 13 of 38 , 34 . 1% ) than asymptomatic ( 12 of 47 , 25 . 5% ) carriers , this difference did not reach statistical significance ( Yates' corrected χ2 = 0 . 762 , 1 df , P = 0 . 382 ) . Because average P . vivax densities were lower in asymptomatic infections and detecting minority clones may be more difficult in samples with low-level parasitemias , we re-analyzed the data after stratifying parasite densities into quartiles . Again , stratified analysis yielded negative results ( Mantel-Haenzel χ2 = 0 . 004 , 1 df , P = 0 . 991 ) . Therefore there was no observable association between multiplicity of P . vivax infection and the presence of symptoms in this community .
This longitudinal study in newly opened frontier settlements provides further evidence that carriers of low-density parasitemias , who are often missed by conventional microscopy , contribute significantly to ongoing P . vivax transmission in rural Amazonia . Results from this and other studies in Amazonia [12]–[14] , [36] , [37] challenge the often persisting view that subjects in low malaria transmission settings are unlikely to harbor low parasitemias , due to the lack of acquired immunity . To the contrary , average parasite densities decreased , with higher proportions of P . vivax infections being missed by microscopy , as malaria prevalence decreased in the community . Interestingly , our findings for P . vivax are consistent with a recent meta-analysis of 106 P . falciparum prevalence studies worldwide that combined microscopy and molecular methods [35] . Because the risk of P . vivax infection ( confirmed by microscopy , qPCR , or both ) correlated negatively with cumulative exposure to malaria , we suggest that our study population has developed over time some degree of anti-parasite immunity , in line with recent findings from traditional riverine communities in Amazonia [13] , [37] . Finally , we show that nearly all subpatent blood-stage P . vivax infections comprise mature gametocytes detected by a highly sensitive molecular technique [24] . We thus conclude that subpatent infections constitute a major P . vivax reservoir in rural Amazonia and possibly in other low-transmission settings . Our findings also challenge classical views regarding asymptomatic infections in low-endemicity populations . Prior to the molecular diagnosis era , nearly all laboratory-confirmed episodes of malarial infection , even those with low parasite densities , were thought to elicit clinical disease in pioneer settlements across the Amazon Basin [38]–[40] . More recent surveys , however , demonstrated that subclinical infections are common in agricultural settlements [12] , [14] and traditional riverine communities [13] , [36] , [37] , [41] , but most of them are missed by microscopy . Interestingly , the high proportion of infections found to be asymptomatic in the present study must be interpreted as a conservative estimate . We may have misclassified some episodes of parasite carriage in subjects reporting any of the 13 symptoms investigated , which may or may not be caused by the current infection , as symptomatic malaria infections , overestimating the proportion of symptomatic infections . Not surprisingly , however , we found very low P . vivax densities in most subclinical infections in Remansinho . Conventional microscopy missed 54% of them , suggesting that previous microscopy-based studies failed to detect asymptomatic parasite carriage in rural Amazonians because they missed a large proportion of low-density infections . Mathematical models identified asymptomatic infections as a crucial target for P . falciparum malaria eradication efforts in Africa [42] , but no similar analyses are available for other endemic areas and other human malaria parasite species [43] . The following findings argue for a major role of asymptomatic infections in maintaining ongoing P . vivax transmission in Remansinho: ( a ) apparently healthy subjects accounted for half of the total P . vivax biomass found in the local population , ( b ) nearly all asymptomatic infections comprised mature gametocytes , and ( c ) few untreated asymptomatic infections became symptomatic ( and thus detectable by routine surveillance ) over the next few weeks of follow-up . We were unable to measure the average duration of untreated , asymptomatic infections in our population; there is a recent estimate of 194 days of duration for untreated P . falciparum infections in Ghana [44] , but no comparable estimate is currently available for P . vivax . Specific studies to quantify the transmissibility of subpatent parasitemia to vector mosquitos via direct and membrane feeding assays are ongoing ( JMV and colleagues , unpublished data ) . Who are at risk of malaria in Remansinho ? Migrants from malaria-free areas ( 54 . 5% of the adults in the community ) constitute a major risk group , with each year of residence in Amazonia decreasing their risk of P . vivax infection and clinical vivax malaria by 2–3% . In some Amazonian communities , malaria has been associated with forest-related activities such as logging , fishing and mining , which typically involve young male adults [12] , [39] , [45] , [46] . However , we show that housekeeping and forest-related activities were associated with similar risks for infection and disease in Remansinho . We hypothesize that nearly all adolescents and adults of both genders engage to some extent in farming activities , especially harvesting , in the forest fringes close to their dwellings , although only young males are often involved in logging and land clearing in more densely forested areas . We are currently using high-resolution satellite images to measure the distance between dwellings and forest fringes to further explore the association between proximity to the forest environment and risk of malaria in Remansinho . Interestingly , malaria transmission appears to be relatively homogeneous across all settlements in the area , equally affecting the poorest and least poor people of both sexes , with no differences in risk according to main house characteristics . Whether the vectorial capacity of An . darlingi is spatially homogeneous is a key question to be answered by ongoing vector biology studies in this site . Detection of gametocytes , through pvs25 gene transcripts , in nearly all qPCR-confirmed P . vivax infections tested is somewhat surprising , since recent studies have found much lower proportions of gametocyte-positive infections in Southeast Asia [47] , [48] and Papua New Guinea [49] . Since gametocytes comprise only 2% of circulating blood stages [50] , microscopists are likely to miss gametocytes in population-based studies where low-density infections are often sampled [23] . Furthermore , we argue that even molecular methods may be poorly sensitive if suboptimal techniques for sample storage and RNA extraction are used under field conditions . For RNA isolation , we cryopreserved venous blood samples at −70°C or in liquid nitrogen a few hours after collection , since our previous attempts to amplify pvs25 transcripts from RNA isolated from classic FTA microcards ( Wathman ) , QIAcards ( Qiagen ) , 903 protein saver cards ( Whatman ) , and 3MM filter papers ( Whatman ) impregnated with P . vivax-infected blood and kept at ambient temperature had all failed [24] . Storing filter papers impregnated with blood in TRizol reagent ( Qiagen ) may improve RNA yield , but almost two thirds of the bloodspots from PCR-confirmed P . vivax infections tested by Wampfler and colleagues [49] were negative for pvs25 transcripts by TaqMan assays , despite previous TRizol reagent treatment . Long-term asymptomatic carriage of P . falciparum has been suggested to protect against subsequent malaria-related disease in Africa [51] , [52] , possibly by reducing the risk of superinfection with more virulent strains . An explanation for this finding is premunition , originally defined by Sergent and Parrot ( 1935 ) as the protection against new infections resulting from immune responses to the existing infection [53] . Alternatively , ongoing blood-stage infection might arrest the development of subsequently inoculated sporozoites in the liver . Such an inhibition of superinfection appears to be mediated by the iron regulatory hormone hepcidin , produced in response to blood-stage parasitemia [54] . However , an opposite effect ( i . e . , increased risk of subsequent disease in asymptomatic P . falciparum carriers ) has also been described , suggesting that a proportion of asymptomatic infections will eventually reach the host's pyrogenic threshold [55] . Here we found no significant association between asymptomatic carriage of low-density P . vivax infection and protection from subsequent malaria morbidity , suggesting that treating individuals with asymptomatic P . vivax infections would not render them more vulnerable to clinical malaria over the next few weeks or months . Although we have identified challenges for malaria control that are not currently addressed by routine surveillance , malaria transmission in Remansinho has declined dramatically over 3 years of surveillance , and P . falciparum was found only during the first four surveys . Factors that may have contributed to this decline include drastic environmental changes resulting from logging and land clearing for farming , variation in climate , the widespread use of insecticide-treated bed-nets since August 2012 , and the implementation of research activities in the area . To address the first two hypotheses , we are now analyzing high-resolution satellite images to track environmental changes over time . Consistent with the third hypothesis , two studies have provided evidence that insecticide-treated bed-nets are effective for malaria control in Amazonia . The first was a case-control study in Colombia that showed more than 50% reduction in malaria , relative to no net use , although the advantage of impregnated over non-impregnated nets was not statistically significant [56] . The second study , a randomized trial of lambdacyhalothrin- versus placebo-treated nets in the Amazonas State of Venezuela , showed a protective efficacy of 55% [57] . Whether insecticide-tread bed-nets alone can reduce malaria incidence rates throughout the Amazon Basin remains uncertain , mostly due to the highly variable biting behavior of An . darlingi across the region [58] , with strong evidence of significant blood-fed and exophilic host-seeking behavior [59]–[61] . In addition , the decline in transmission in Remansinho preceded the distribution of bednets . Finally , the presence of a research team continuously working in the area for over 3 years may affect positively both diagnostic and treatment practices . The external slide revision routinely carried out by our team provides an example of intervention that may have enhanced the diagnostic skills of local microscopists . Moreover , active case detection during 8 consecutive surveys allowed for the early diagnosis and prompt treatment of several slide-positive asymptomatic infections that would have been missed by routine passive surveillance . Eliminating residual foci when malaria is nearly disappearing , but remains entrenched in a few hotspots , is the next major goal in Remansinho and many other similar endemic settings . Case detection strategies in areas approaching malaria elimination often target only subjects presenting with fever or with a history of recent fever , who are screened for malaria parasites by conventional microscopy or rapid diagnostic tests ( RDT ) and receive prompt antimalarial treatment if found to be infected [62] . These strategies overlook asymptomatic infections that might be detected by periodic cross-sectional surveys of the entire population at risk [63] , as we did in Remansinho . Nevertheless , the cost-effectiveness of mass blood surveys for detecting and treating these residual infections decreases proportionally as malaria transmission declines , since: ( a ) large populations must be screened to diagnose relatively few asymptomatic carriers , and ( b ) diagnostic techniques available for large-scale use , such as microscopy and RDT , are not sensitive enough to detect low-grade infections that are typical of residual malaria settings [64] . As an alternative , we are currently testing a reactive case detection strategy that has been tailored for the relapsing parasite P . vivax to detect new infections in the neighborhood of malaria cases diagnosed by routine surveillance in frontier settlements similar to Remansinho . Evaluating this and other strategies of active surveillance to cope with asymptomatic infections in residual P . vivax foci is a top research priority in the context of current malaria elimination efforts worldwide . | Despite decades of control efforts , malaria remains a major public health concern in Brazil . A large proportion of the 243 , 000 cases diagnosed per year originate from areas of recent colonization in the densely forested Amazon Basin . This population-based longitudinal study addresses the epidemiology of malaria during the early stages of colonization of frontier settlements in Remansinho area , rural Amazonia . We documented a major decline in the prevalence of P . vivax infection , from 23 . 8% to 3 . 0% , between March–April 2010 and April–May 2013 . Up to 73 . 1% of the P . vivax infections were missed by microscopy as malaria transmission declined and most ( 56 . 6% ) of these infections caused no clinical signs or symptoms . Few ( 17 . 0% ) asymptomatic P . vivax infections that were left untreated eventually progressed to clinical disease , becoming detectable by routine malaria surveillance , over 6 weeks of follow-up . Moreover , nearly all P . vivax infections that were undetected by microscopy had gametocytes , the parasite's blood stages responsible for malaria transmission to mosquito vectors , detected by molecular methods . These findings indicate that apparently healthy carriers of low-density parasitemias , who are often missed by conventional microscopy , contribute significantly to ongoing P . vivax transmission and may further complicate residual malaria elimination in Remansinho and similar endemic settings . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"and",
"health",
"sciences",
"epidemiology",
"tropical",
"diseases",
"malaria",
"parasitic",
"diseases"
]
| 2014 | Epidemiology of Disappearing Plasmodium vivax Malaria: A Case Study in Rural Amazonia |
Evidence is mounting that the evolution of gene expression plays a major role in adaptation and speciation . Understanding the evolution of gene regulatory regions is indeed an essential step in linking genotypes and phenotypes and in understanding the molecular mechanisms underlying evolutionary change . The common view is that expression traits ( protein folding , expression timing , tissue localization and concentration ) are under natural selection at the individual level . Here , we use a theoretical approach to show that , in addition , in diploid organisms , enhancer strength ( i . e . , the ability of enhancers to activate transcription ) may increase in a runaway process due to competition for expression between homologous enhancer alleles . These alleles may be viewed as self-promoting genetic elements , as they spread without conferring a benefit at the individual level . They gain a selective advantage by getting associated to better genetic backgrounds: deleterious mutations are more efficiently purged when linked to stronger enhancers . This process , which has been entirely overlooked so far , may help understand the observed overrepresentation of cis-acting regulatory changes in between-species phenotypic differences , and sheds a new light on investigating the contribution of gene expression evolution to adaptation .
The evolution of gene expression has become a subject of intensive research in the last years , sparking debates upon its role in adaptive evolution and speciation [1–6] . Clearly , protein folding , expression levels , timing and tissue localization of expression are important regulatory traits under selection as they are essential steps in the genotype-to-phenotype map . Gene expression is regulated at each step along the pathway from DNA to protein . Among them , transcription initiation is a crucial step responsible for a large proportion of the variation in expression profiles . Here we will focus on regulatory regions controlling this transcription initiation . Changes in gene expression are often caused by mutations in cis-regulatory elements ( CREs ) and trans-regulatory elements ( TREs ) [7] . Cis and trans-regulators control transcription of genes located on the same chromosome , or on both homologous chromosomes , respectively . Several recent technological breakthroughs [7–10] have considerably improved our ability to study these regulatory sequences and their associated expression profiles . They revealed that gene expression profiles were highly variable and heritable within species [2 , 11 , 12] , quickly diverging among species [8] . Furthermore , many studies have shown how changes in gene expression contribute to adaptive changes , by natural selection at the individual level [4 , 6 , 13–17] . From a theoretical standpoint , several models have also been developed to understand the evolution of gene expression by individual level selection [18–22] . In this paper we investigate a new and different selective phenomenon also acting on the evolution of regions controlling transcription initiation . This phenomenon may contribute to the fast divergence of regulatory networks between closely-related species , but unlike the usual view , it is not rooted in individual-level selection , and hence does not necessarily increase individual fitness . We term this selective process ‘ER‘ ( for ‘enhancer runaway’ ) . It results from competition for expression between homologous cis-acting regulatory sequences in diploid organisms . With this gene-level selection , these regulatory sequences behave as self-promoting genetic elements . Transcription initiation is determined by the binding of a suitable RNA-polymerase to a Transcription Start Site ( TSS ) . This binding involves a complex machinery of often over 30 partner proteins [23] . It depends on the interaction of CREs and TREs . CREs are non-coding sequences located on the same chromosome as the regulated gene . They include core promoters located around the TSS , which integrates the regulatory inputs [24] . They also include enhancers , which influence transcription initiation rate independently from their orientation or localization on this chromosome [9] . TREs are coding sequences , located anywhere in the genome , that produce transcription factors ( TFs ) , which bind to CREs on both homologs . They can also produce cofactors , which bind other proteins ( including other TREs ) [7] . In order to show how the ER process works , we use population genetic models with both protein-coding sequences ( the ‘gene ( s ) ’ ) and regulatory sequences ( enhancers or transcription factors ) . For generality , we do not specify precisely how regulatory mutations change regulatory networks . We simply assume that mutations occur in enhancers and TFs , which impact expression levels . Indeed , there are many ways for an enhancer mutation to modify expression levels [2 , 7 , 9 , 25 , 26] . It may change for instance the binding site affinity towards TFs , the number and/or spacing between binding sites , or the nucleosome conformation around the enhancer region , which strongly impacts DNA accessibility for the transcriptional machinery . Because enhancers act in cis , the presence of different alleles at the enhancer locus creates a pattern of allele-specific expression ( imbalance in chromosomal origin of transcripts ) for the protein-coding gene [27 , 28] . Indeed , in heterozygous individuals , the ‘weaker’ enhancer contributes less to protein expression than its ‘stronger’ counterpart on the other chromosome , causing imbalanced expression of homologous gene alleles . TF mutations may also alter expression levels in various ways . For instance , they can change its DNA-binding domain ( modifying its affinity for different binding sites ) or its protein-protein binding domain ( modifying its interactions with e . g . cofactors or histones ) . However , they act in trans , and do not generate allele-specific expression . To obtain a good understanding and broad evaluation of the significance of the ER process , we investigate several complementary models . In a first model , we study the ER process in the absence of individual level selection for expression level . We then incorporate individual level selection on expression . Many form of individual level selection on expression levels could have been chosen , and there has been indeed much debate over the different selection pressures acting on expression levels . While some argue that regulatory polymorphism is mainly neutral or quasi-neutral [29 , 30] , others suggest that regulatory evolution is mostly shaped by stabilizing selection [31–33] and occasionally by directional selection [8 , 34–36] . In a second and third model , we thus chose to introduce stabilizing selection on expression levels ( due to a relative gene dosage constraint in model 2 , or an absolute expression constraint in model 3 ) . In these two models , other regulatory regions can evolve in concert ( other enhancers in model 2 , TFs in model 3 ) . Finally , since the ER process may act very differently in genomes exhibiting inbreeding and low heterozygosity , we investigate how it is influenced by the mode of reproduction ( inbreeding in model 4 ) . Overall we show that , in a small genomic region around genes , there is a selection pressure on enhancer to increase expression levels . This phenomenon leads to an open-ended escalation in enhancer strength ( a ‘runaway’ ) . This process is not halted by inbreeding , or by stabilizing selection on expression levels , as long as enhancers can evolve in concert with other regulatory sequences ( enhancers or TFs ) involved in the same regulatory network . Enhancer runaway is not a highly specific or idiosyncratic process: it is expected to occur at variable intensities for all genes in nearly all eukaryotic diploid organisms . This widespread phenomenon may significantly shift our current understanding of gene regulatory regions , and opens a wide array of possible tests and comparisons .
To illustrate how competition for expression works , we first present a two-locus model in an infinite , diploid , sexual population . The first locus , the ‘gene’ , codes for a protein . We suppose that it undergoes recurrent deleterious mutations ( with fitness effect s and dominance h ) at a rate u , but the argument would apply equally well with beneficial mutations ( see methods ) . This locus quickly reaches the usual deterministic mutation–selection equilibrium , with deleterious mutation at frequency u/hs . The second locus , the ‘enhancer’ , is located at a recombination distance r and controls the expression of the gene in a cis-regulatory fashion . We wish to determine the selection pressure acting on mutations that modify the strength of this enhancer . In model 1 , we consider that the overall expression level is tightly controlled , for instance because of trans-acting regulatory factors producing a negative feedback loop . Such a feedback loop is not relevant to all genes , but is a particularly useful starting case . It allows investigating how the ER process works without the complication of additional selection pressures acting , at the individual level , on protein expression . We thus assume that overall expression levels are constant ( due to the feedback loop ) , such that only the relative contributions of each homologous alleles vary due to mutations on enhancers . Labelling e1 and e2 the strengths of the two enhancer alleles of an individual , the gene associated with enhancer of strength e1 contributes a fraction e1/ ( e1+e2 ) of proteins produced . As a consequence , the gene allele linked to a stronger enhancer contributes a larger share of proteins . In double heterozygotes , if the deleterious allele of the gene is linked to the weaker enhancer , less than 50% of the proteins produced will be of the defective form , and thus its effect on fitness will be less than predicted based on its dominance coefficient h ( which in effect reduces its dominance to h1<h ) . In contrast , if it is linked to the stronger enhancer , its deleterious effect on fitness will be stronger than predicted by h ( which in effect increases its dominance to h2>h ) . The fitness effect of linkage with a specific enhancer can thus be modeled as a change in h that occurs only when the enhancer is heterozygote , while dominance is necessarily h as long as an enhancer allele is fixed , as illustrated in Fig 1 . Because deleterious mutations are usually partially recessive [37] , the relationship between the fraction of defective proteins and fitness is necessarily nonlinear , monotonously decreasing and concave ( Fig 2 ) . Thus , we have h1 + h2 > 2h ( applying Jensen’s Inequality to the strictly concave fitness function ) . In other words , a polymorphism at the enhancer locus necessarily increases average dominance ( the mean of h1 and h2 is greater than h ) , which increases the strength of selection against heterozygous deleterious mutations , but also reduces the fitness of individuals carrying these mutations . The deterministic change in frequency at the enhancer locus can be computed from standard population genetics equations , considering a generic diploid life cycle with four steps: diploid selection , meiosis with recombination , mutation and syngamy . This frequency change can be decomposed into two terms that correspond to ‘masking’ and ‘purging’ , as introduced in previous related models [38 , 39] ( see derivation in methods ) . The ‘masking’ term does not depend on recombination , and is frequency-dependent . It always disfavors the enhancer allele when rare , independently of its strength . The reason is that a rare enhancer will most often be represented in double heterozygotes ( since deleterious mutations often are rare too ) , which have lower average fitness , since they have higher average dominance ( this is similar to Fisher’s argument for the evolution of lower dominance [40] ) . In other words , rare enhancer alleles pay the cost of unmasking deleterious alleles . This term is strongly conservative , as it prevents any new enhancer to enter the population . The ‘purging’ term , in contrast , is frequency-independent , always favors stronger enhancers , and increases when recombination between the enhancer locus and the gene decreases . The reason is that genes that are linked with stronger enhancers are more exposed to selection and more efficiently purged from deleterious mutations . Hence , stronger enhancers are disproportionately found on –and hitchhike with– favorable genetic backgrounds . Overall , combining these two effects , weaker mutant enhancers are always disfavored , while stronger mutant enhancers are favored if sufficiently tightly linked to the gene for the ‘purging’ effect to overcome the ‘masking’ effect . The recombination distance where the two effects balance each other depends however on the strength difference of the two enhancer alleles . For enhancers that are very similar in strength , the range of recombination distances where the runaway can occur is larger , but the intensity of selection on the stronger enhancer is weaker ( S1 Fig ) . This is due to the fact that , for small differences in enhancer strength , Δh becomes vanishingly small compared to δh ( and thus the masking term in front of the purging term , see Eqs 3 and 5 in methods ) . Overall , stronger runaway is expected close to genes because selection intensity on new enhancers is stronger at small recombination distances and because the runaway only concerns enhancers of small effects at larger recombination distance . In regulatory regions close to the gene , selection thus favors enhancers contributing more to protein production: homolog enhancers compete for expression . This outcome is illustrated on Fig 3 , where this analysis is checked against –and agrees with– stochastic numerical simulations reporting the fixation probabilities of new enhancers in finite populations . This finding indicates that stronger and stronger cis-regulatory elements should evolve in an open-ended fashion in the vicinity of genes , as long as that does not influence the total amount of proteins produced . Competition for expression may thus be responsible for a runaway process of enhancer strength . During this process , allele-specific expression would transiently occur , but expression balance would be restored once the new enhancer reaches fixation . This process shares many similarities with the endless occurrence and spread of new segregation distorters that transiently bias Mendelian ratio while they sweep [41] . It also shares similarities with models of Fisher runaway in the context of sexual selection , where female mating preference drives the open-ended evolution of extravagant traits in males [42] . Interestingly , the spread of such stronger enhancers occurs even though it temporarily decreases population mean fitness ( see methods ) . Indeed , stronger enhancers spread because they find themselves on better backgrounds , but at the expense of temporarily increasing mean dominance and hence unmasking deleterious mutations . Overall the ER process does not optimize mean individual fitness in the population . Gene expression regulation is not necessarily embedded in a negative regulatory loop , as considered above . Mutations on enhancers will often alter overall expression levels . When this occurs , stabilizing selection on overall gene expression will interfere with competition for expression . We developed two additional models ( model 2 and 3 ) with such interactions ( see methods ) . In these models , mutations on enhancers alter both relative contribution to expression , e1/ ( e1+e2 ) , and total expression levels , e1+e2 . We assume that total expression levels undergo stabilizing selection with different intensities that reflect the functional diversity of genes ( over- or under-expression may be more costly for some genes than others ) . In model 2 , we assume that stabilizing selection on expression levels stems from gene dosage , such that the optimal amount of a given protein depends on the amounts of another protein coming from another loci , as occurs for instance in enzymatic or metabolic pathways . This produces the strongest constraint on enhancer strength evolution when only two proteins are concerned ( see methods ) . In this case , any increase in the expression of gene 1 causes a departure from optimal dosage for gene 2 , which reduces fitness . Results show that such stabilizing selection fails to prevent enhancer strengths from escalating . This is because enhancers of both genes coevolve: their strengths increase in parallel , allowing maintenance of the correct protein dosage . However , stabilizing selection tends to decrease escalation rates ( longer doubling times on Fig 4 , see methods ) . Strength-increasing mutations of large effects are indeed counter-selected , since they lead to large deleterious departures from optimal gene dosage . However , strength-increasing mutations of small effects lead only to small enough departures from optimal gene dosage for the genotype to survive until optimal gene dosage is restored by compensatory mutations . Stabilizing selection needs to be relatively strong ( stronger than the selection at the gene locus ) for it to significantly alter escalation rates . In model 3 , we considered a situation where stabilizing selection acts directly on the absolute expression levels of a gene , but transcription factors influence expression levels as well as enhancers . We designed a three-locus model with a gene , an enhancer locus , and a TF locus , where both the strength of the enhancer and TF combine to determine expression levels ( see methods ) . Here again , as shown on Fig 4 , stabilizing selection slows down but does not stop the ER process , due to coevolution between regulators ( here , enhancers and TFs ) . As enhancer strength increases , TF strength decreases in proportion , which maintains approximately constant and optimal total expression levels . In this model , escalation rates are systematically lower than the ones obtained with model 2 ( except for I = 0 ) . This is to be expected as two identical enhancer loci are exposed to the ER process in model 2 , while only one enhancer is exposed to the ER process in model 3 ( the TF locus only responds to stabilizing selection ) . As a consequence , in model 3 , stabilizing selection importantly decrease escalation rates at lower intensities ( lower than the intensity of selection at the gene locus ) . Variation in enhancer strength only makes a phenotypic difference in double heterozygotes . In situations where such heterozygotes are less frequent than under random mating , the pace of the ER process is expected to be slower . Inbreeding is a typical and common situation decreasing heterozygote frequency . It may be caused by various processes ( e . g . selfing in hermaphroditic plants or animals , population structure ) . To determine whether ER indeed differs among species exhibiting e . g . different mating systems , and to quantify the extent of this reduction , we introduced partial selfing in the first model . In model 4 , at each generation , individuals were considered to have probability ps of selfing ( i . e . with probability ps , the same parent is used to sample the second gamete ) . As expected the ER process slowed down as selfing rate increased . Fig 5 illustrates this behavior . Results show that relatively high levels of self-fertilization are needed for the ER process to be significantly slowed down .
The ER process we describe sheds a new light on the evolution of gene regulatory regions in diploids . It should be a widespread phenomenon , applicable to all genes in diploid eukaryotes as the theory only involves generic assumptions ( recessivity of deleterious mutations [37] , genetic variation at enhancer loci [43] , CREs—TREs coevolution [44] , stabilizing selection on expression levels [31–33] ) . This process has been acting probably since early unicellular diploids , more than a billion years ago , before the evolution of complex combinatorial or developmental regulatory pathways . As we already stressed in the introduction , it does not undermine the idea that expression regulation is also subject in contemporary eukaryotes to other important selective effects ( direct selection for expression level , timing and localization ) . Before discussing the implications and predictions associated with this theory , it is useful to relate it to previous models involving variation in dominance , ploidy , mutation load and selection for modifiers . It shares with models of ploidy evolution the fact that increased haploid expression leads to more efficient purging of deleterious mutations , which benefits tightly linked modifiers [38 , 39] . It shares with models of dominance evolution [45–48] the limit that selection on modifiers is weak , of the order of the mutation rate ( even if this issue may be alleviated if migration is the source of deleterious alleles [48 , 49] ) . Such a weak selection is usually seen as a limit in the case of dominance evolution because genetic drift in small populations or small pleiotropic effects of modifiers should easily overwhelm the selection pressure for dominance modification [50] . A distinctive feature of our theory is that the ER process is not strongly limited by such pleiotropic effects: as we showed , direct selection against suboptimal expression levels can be readily compensated by CREs-CREs or CREs-TREs coevolution , without halting the ER process . A critical difference with previous models is that dominance does not evolve: the dominance of deleterious mutations is h before the sweep of a stronger enhancer , and is still h after its fixation . Similarly , ploidy or the number of gene copies stays constant and does not evolve ( with the consequence that , unlike in models involving gene duplicates or ploidy variation , the mutational load is not permanently changed by a change in gene number [38 , 47] ) . In our models , selection is frequency-dependent at leading order and leads to a runaway escalation not seen in models of ploidy or dominance evolution . Our models specifically focus on cis- or trans-acting modifiers , mimicking the actual genetic variation occurring on regulatory regions . Cis-regulation introduces naturally an asymmetry in the fitness of the two double heterozygotes ( EA/ea and Ea/eA ) , which is also not a feature present in models of ploidy or dominance evolution . This theory leads to a series of predictions that can be further tested . We highlight eight of them below using capital letters A-H . The ER process should occur for almost all genes in diploids , but with different , and possibly very different rates depending on their specific evolutionary constraints ( regulatory loops , dosage relationships , intensity of stabilizing selection ) . For instance , as we illustrated with our first model , this runaway is fastest for enhancers that are embedded in a downstream negative regulatory loop ( prediction A ) . Such negative feedback loops have been extensively described and are often thought to largely contribute to phenotypic robustness ( e . g . [51–53] ) . This runaway should be slower for genes that are not regulated with such loops and exposed to direct stabilizing selection on relative or absolute expression levels . The pace of the ER process depends in those cases on the form and intensity of the stabilizing selection , as well as on the opportunities for CREs and TREs to coevolve to maintain optimal total expression levels , which may differ among genes . Our results show a strong impact of the recombination rate between the enhancer and the gene on the ER process ( see Fig 3 ) . There are two regimes: ( 1 ) for large recombination rates , the ER process does not occur ( large and medium effect enhancers are selected against and small effect enhancers are nearly neutral ) , while ( 2 ) at shorter genetic distances , the runaway occurs and its rate increases as the recombination rate decreases . Everything else being equal , the ER process should cause a positive correlation between enhancers’ strength and their proximity to the gene , but it should not concern enhancers located at large genetic distances ( prediction B ) . This prediction would be consistent with the observation that most CREs remain close to the gene they regulate [54] . The critical genetic distance delimiting the two ER regimes increases as the intensity of selection on the gene increases . As a result , genes undergoing stronger purifying selection would be expected to have a larger surrounding region where enhancers increasing expression may arise and compete for expression . Consequently , we predict that genes experiencing stronger purifying selection should exhibit a larger surrounding regulatory region , and that , for a given genetic distance to such genes , enhancers should exhibit a faster ER process ( prediction C ) . Such qualitative predictions may be altered if there is an inherent physical tendency for CREs strength to covary with physical map distance . However , if , as is most likely , CRE strength decreases with physical distance along the chromosome , the qualitative pattern will remain identical ( since decreased CRE strength and increased recombination both prevent the fixation of new enhancers ) . There are several reasons , not included in our models , why the ER process could slow down or stop . First , many mutations on enhancers are likely to have pleiotropic effects on timing or localization of expression . These features are not included in our model , and probably limit the availability of enhancer mutations that can contribute to the ER process in higher eukaryotes . Second , mutations may not be able to increase enhancer strength indefinitely . For instance , when a binding motif sequence is optimized for a particular TF , there may be little room for further improvement . Similarly , there is a limit to the number of binding motifs that can be packed in a regulatory region , etc . These constraints could be revealed by studying the effect of random mutations on regulatory regions . If enhancer strength is maximized by the ER process , no ( or very few ) random mutations on the enhancer will be able to increase its strength , such that the strength of those enhancers should be biased downward by random mutations . Conversely , if a TF has evolved to compensate for the enhancer runaway ( i . e . by increasingly repressing transcription ) , random mutations on this TF should on average increase expression levels ( prediction D ) . The evolution of regulatory networks’ complexity is not well understood and controversial ( some argue that the evolution of complex regulatory networks stem from adaptive processes [18 , 20 , 55] , while others consider a non-adaptive origin [56 , 57] ) . We saw that the ER process depends on the shape of regulatory networks ( e . g . presence of feedback loops , of gene dosage etc . ) . It may also be facilitated with more complex regulatory architecture , as highly degenerate and complex regulatory architectures [58] could provide more ‘degrees of freedom’ for CREs—TREs ( co ) evolution . Interestingly , the ER process may also contribute to the evolution of complex regulatory networks . For instance , an increase of enhancer strength may be achieved e . g . by locally duplicating TF binding motifs ( and hence increasing the complexity of regulatory architecture ) . The duplicated motifs may further diverge to attract a larger diversity of TFs if this happens to be a route to further increase of enhancers’ strength: combinatorial regulation ( where expression specifity results from a particular combination of several TFs , as in e . g . [59] ) may thus evolve from the ER process . A positive feedback loop may thus occur between the ER process and the evolution of regulatory complexity . Much of these effects will rely on the contingency of mutational variation on enhancers , but can produce a level of architecture complexity much beyond the level expected under individual selection alone . Species where the ER process is expected to be faster ( e . g . outcrossing vs . selfing diploids , diploid versus haploid eukaryotes ) should exhibit more complex regulatory architectures ( prediction E ) . Like for other theories for the evolution of complexity [56 , 57] , complexity would emerge here as a by-product of another process ( here ER ) and is not a direct target of selection . It is generally envisioned that many regulatory networks can be functionally equivalent . For instance , many CREs—TREs combinations can perform the same signaling , and different global regulatory wirings among TREs and CREs can achieve the same regulatory pattern . Like with a key / lock or signal / receiver mechanisms , the central functional requirement is the reciprocal recognition of interacting regulatory elements . Such a situation produces a fitness landscape with a ridge , along which ‘evolutionary freedom’ allows for substantial neutral divergence [60] . This process is often thought to drive relatively fast divergence of regulatory networks among species without modifying expression patterns much [61] . In our models , the same process occurs ( coevolution between regulators occurs without modifying much expression levels ) , but , in addition , selection for stronger proximal CREs leads to faster-than neutral divergence ( prediction F ) . Using again the fitness landscape metaphor , it means that the ridge is actually not flat , but is slightly sloping: evolution between networks with the same expression pattern is not neutral , but rather directed to favor networks involving stronger proximal enhancers . Moreover , this faster-than-neutral divergence is expected to be accentuated for CREs ( compared to TREs , prediction G ) . Indeed , the ER process necessarily involves CREs , and optionally TREs . For instance , cis- cis- coevolution illustrated in model 2 does not involve TREs evolution . This may partly explain why CREs are usually ( but not always , e . g . [62] ) found to contribute more than TREs to expression regulation divergence among species ( e . g . in Saccharomyces [63] , Drosophila [64] , Gasterosteus [16] and Mus [65] ) . As shown in model 4 , the ER process is slower in presence of inbreeding . In particular , the ER process is expected to be faster in outcrossing than in self-fertilizing species ( prediction H ) . One way to test for this prediction would be to compare CRE strengths in hybrids between closely related outcrossing and self-fertilizing species . Indeed , in the hybrid , expression level differences between alleles can only be explained by CRE variation , as TREs from both parents are shared [66] . Such method could be used in e . g . Arabidopsis [67] or Capsella [68] . The expectation would be that CREs derived from the self-fertilizing species should be weaker on average , biasing expression pattern in F1s towards the outcrossing parent alleles . Such test would require however to control for the direction of the cross ( distinguishing the species effect from maternal versus paternal effects ) . Several genetic oddities that have been observed in some species may greatly limit the ER process . For example , in somatic cells of Diptera , homolog chromosomes pair and expression is largely influenced by a phenomenon referred to as ‘transvection’ [69] . With transvection , “CREs” impact regulation of both homolog chromosomes ( i . e . are not really behaving as cis-regulatory elements as they also regulate in trans ) . Somatic pairing and transvection in these taxa certainly reduce , or even eliminate competition for expression . Similarly , repulsion of homologs , as found in mammals’ nucleus [70] , ( i . e . the fact that the chromosome territories of homologs tend to be further apart within the nucleus compared to a random pair of chromosomes ) could also strongly reduce competition for transcription factors between competing CREs , as the transcription of homologs likely involves different and spatially segregated ‘transcription factories’ [71] . In both cases , reduced competition for expression between homologs could strongly limit the ER process . Finally some recombination hotspots are located close to genes in several species [72 , 73] . This could also strongly limit the ER process by breaking linkage disequilibria between CREs and genes . Whether some of these genetic oddities evolved as suppressors of the ER process is an intriguing possibility , especially given that their evolutionary significance remains elusive .
We firstly consider a diploid two-locus model: the first locus , referred to as ‘the gene’ is a protein-coding locus undergoing mutations and diploid viability selection; the second one , referred to as ‘the enhancer’ , is a CRE locus controlling the expression of the gene . The gene is exposed to recurrent deleterious mutations , at a rate u per individual per generation , changing A alleles into a alleles . In the analytical derivation below , we focus on this bi-allelic case , but this assumption is relaxed later for numerical simulations . We define the relative fitness of genotypes as 1 , 1 –h s and 1 –s for AA , Aa and aa genotypes , respectively , where s is the selection coefficient against the a allele and h its dominance . The enhancer is at a recombination distance r from the gene . We consider two alleles ( E1 / E2 ) , which differ by their ability to promote expression of the gene located in cis ( i . e . on the same chromosome ) . This ability is referred to as their ‘strength’ and noted e1 and e2 . Biologically this strength depends on many parameters , such as sequence affinity , chromatin state , binding network or intracellular signals [7] . Furthermore , different mechanistic models have been put forward to describe enhancer effects: they are thought to change either promoter activation levels or the probability that a promoter will be activated to initiate transcription [74] . Here , we will not assume a particular mechanism but cover all these possibilities by simply supposing that enhancer sequences intrinsically differ by their ability to activate gene transcription . We assume that the gene on the same chromosome than an enhancer with strength e1 , facing on the homolog chromosome an enhancer with strength e2 , contributes to a fraction e1/ ( e1 + e2 ) of protein produced ( see Fig 1 ) . As a consequence , the gene associated with a stronger enhancer contributes a larger share of proteins . A major feature of this first model is that the total protein expression level is tightly regulated: the total amount of protein produced is assumed to be constant , and does not depend on enhancers’ strength ( 6 proteins are always produced on Fig 1 ) . Such a situation would occur , for instance , when the amount of transcription factors , or a repressor is regulated through a negative feedback loop ( e . g . through the total amount of protein produced or by the concentration of a downstream metabolite resulting from protein activity ) . Besides representing this fairly common biological situation , this model is also important to understand the evolution of enhancer strength independently from the selection pressure acting on the amount of protein in a given cell at a given time . When the gene locus is homozygous , the fitness of individuals does not depend on the enhancer alleles , as 100% of the proteins are of the same type ( the fitness is 1 or 1 –s for AA and aa genotypes respectively ) . However , different situations arise in individuals that are heterozygous at the gene locus . When they are homozygous at the enhancer locus , they equally express each type of proteins ( 50% each ) , and their fitness is 1 –h s ( genotypes ( a ) and ( d ) in Fig 1 ) . When they are heterozygous at the enhancer locus they express more ( or less ) of the defective protein if the stronger ( or weaker ) enhancer is associated with the deleterious allele ( genotypes ( b ) and ( c ) in Fig 1 ) . This changes dominance at the gene locus . When fewer defective proteins are produced , the deleterious effect will be lower , which means lower dominance , noted h1 such that h1 < h ( genotype ( c ) in Fig 1 ) . Conversely , if more defective proteins are produced , the deleterious effect will be larger , which means a higher dominance , noted h2 such that h2 > h ( genotype ( b ) in Fig 1 ) . The values of h1 and h2 will depend on the strengths e1 and e2 of E1 and E2 alleles . In order to be more specific about this relationship , we note that phenotype-to-fitness relationship must verify two major properties: ( 1 ) the fitness must decrease as the proportion of deleterious proteins expressed increases and ( 2 ) the relationship between the fitness and the proportion of defective proteins expressed must be concave , as deleterious mutations are most often partially recessive [37] , which means that fitness effects of deleterious mutations are lower than what would be expected in an additive ( linear ) situation . As a consequence , the relationship between fitness and the fraction of defective proteins expressed must be similar to the situation illustrated on Fig 2 . A generic way to formulate these properties is to express fitness as a function of f[a] , the fraction of defective proteins , h and s with a concave monotonic power function: W=1−sf[a]−Log ( h ) Log ( 2 ) , ( 1 ) which conveniently converges to the additive situation for h = 1/2 . With such a relationship , and assuming that e2 > e1 , we have: { h1= ( e1e1+e2 ) −Log ( h ) Log ( 2 ) h2= ( e2e1+e2 ) −Log ( h ) Log ( 2 ) ( 2 ) Noting Δh = ( h1+h2 ) /2 –h , one can use Jensen’s Inequality on the strictly concave fitness function to show that Δh > 0 . The commonly accepted assumption that deleterious mutations are partially recessive ( i . e . h is on average around 0 . 25 for mildly deleterious mutations , [37] ) leads in our model to the outcome that polymorphism at the enhancer locus increases the mean dominance . To study the evolution of enhancer strength , we are first looking for the frequency variation of the stronger E2 enhancer allele after one generation ( noted as Δp ) . To do so , a four-step life cycle is implemented , with diploid selection , meiosis , mutation and syngamy . The exact recursion is then linearized to leading orders , defining a parameter ξ << 1 and assuming ( 1 ) weak selection on the gene ( h s is of order ξ ) , ( 2 ) very small mutation rate ( u is of order ξ2 ) and ( 3 ) small recombination rate r between enhancer and gene loci ( of order ξ ) . Under those fairly generic assumptions , the frequency of the deleterious allele a ( noted pa ) is small ( of order ξ ) and the linkage disequilibrium ( DEA ) between the enhancer and the gene is small ( of order ξ ) . DEA is defined as positive when E2A and E1a haplotypes are over-represented ( meaning that DEA is positive when stronger enhancers are associated with beneficial alleles ) . We obtain , noting p the frequency of E2 and q the frequency of E1 , the leading order of the frequency variation of E2: Δp=−2Δhpapq ( 1−2p ) s ( 3a ) +DEA[4hpq+ ( 1−2p ) ( h2q−h1p ) ] s+o ( ξ2 ) ( 3b ) The first term Eq ( 3a ) corresponds to a direct selection on enhancers while the second term Eq ( 3b ) is an indirect selection proportional to DEA . The direct selection is negative when E2 is rare ( since Δh > 0 ) and positive when it is frequent . Thus , it favors the most common allele at the enhancer locus . To understand this effect , it is useful to derive the mutation-selection equilibrium of pa , paEq . Using the same linearizing assumptions than for Δp , assuming that the mutant stronger enhancer has just entered the population ( thus neglecting DEA ) and noting δh = h2−h1 , one obtains: paEq=uh¯s+o ( ξ ) , ( 4 ) where h¯=h+2 Δh p q is the average dominance in the population . Note that the polymorphism at the enhancer locus increases average dominance ( since Δh > 0 ) , which reduces the frequency of deleterious mutations . The direct selection term ( 3 . a ) actually stems from this effect on average dominance: fixation at the enhancer locus is similar to reducing average dominance , which is favorable . Like in models for the evolution of ploidy [38 , 39] , direct selection favors the masking of deleterious mutations , but here this effect is frequency-dependent . We refer to this term as the ‘masking’ term . The indirect selection is relatively straightforward since the expression within brackets in ( 3 . b ) can be shown to be always positive . That means that indirect selection has the same sign than DEA: it is positive when the stronger enhancer ( E2 ) is preferentially associated with beneficial gene alleles ( DEA > 0 in this situation ) . The question turns now to determine the sign of DEA . To do so , we use a quasi-linkage equilibrium ( QLE ) approximation that requires that the forces causing frequency changes are weak relative to recombination rate [75] . This approximation usually breaks down at low recombination rates . However , because we are at mutation selection balance for the gene , and because the frequency change at the enhancer locus is small , an accurate approximation can be obtained by linearizing under the same assumptions as above , and keeping terms in both rDEA and sDEA [76] . We obtain DEAQLE: DEAQLE=p q pa[2Δh ( 1−2p ) +δh]s2r+[h1p2+2pqh+h2q2]s+o ( ξ ) ( 5 ) Note that this quasi-linkage equilibrium value does not diverge for small recombination rate . Noting that Δh is at most half as large as δh , it is straightforward to show that DEAQLE>0 , indicating that the indirect selection term is positive and favors stronger enhancers . This effect stems from the fact that E2 carrying chromosomes are more exposed to selection because of their increased average expression . They are thus purged more efficiently from deleterious mutations: as a consequence E2 alleles are most often found on , and hitchhike with , beneficial genetic background ( they are associated to A alleles ) , which is beneficial for E2 alleles . We refer to this term as the ‘purging’ term . The same model can be made with beneficial instead of deleterious mutations . In this case , we cannot assume that pa << 1 , since a alleles sweep to fixation . As a consequence the expression of Δp , involves more terms depending on pa . However , Δp can still be partitioned into a ‘masking’ and a ‘purging’ term like in Eq 3a and Eq 3b , respectively . Provided that beneficial mutations are dominant , and for any pa value , the ‘masking’ term still favors the most frequent alleles , whereas the ‘purging’ term still favors stronger enhancers . Qualitatively , the model with beneficial dominant mutations and the model with deleterious recessive mutations give similar results concerning enhancer alleles’ frequency dynamics . Considering partially dominant deleterious mutations would lead to a moderately different outcome ( the term 3 . a will switch sign , causing more enhancer polymorphism ) , but this scenario is biologically much less relevant . When a stronger enhancer spreads , it is favored by the purging term , but at the cost of unmasking deleterious mutations . Mean fitness equals 1–2u when there is no variation at the enhancer locus ( p = 0 or p = 1 ) . However , during the spread of an enhancer , mean fitness W¯ decreases as can be readily seen from: ∂W¯∂p=−4Δh ( DEA+ ( 1−2p ) pa ) s , ( 6 ) which is negative around p < ½ and positive around p > ½ . This is due to the fact that deleterious mutations are unmasked when average dominance increases , before they have time to reach their lower mutation—selection equilibrium frequency . A temporary genetic burden appears: deleterious mutations are too frequent given their new mean dominance . Because the selection coefficient on a new enhancer allele is frequency dependent ( 3 . a ) , it is useful to obtain a more integrated measure of selection on enhancer alleles . One solution is to compute their probability of fixation U ( which accounts for all frequency trajectories ) and compare it to a neutral expectation . We computed it , for a mutant enhancer initially at frequency p0 , using a diffusion approximation [77]: U ( p0 ) =∫p=0p0e−∫4NpopΔppqdpdp/∫p=01e−∫4NpopΔppqdpdp ( 7 ) As the numerical integration calculated from Eq 7 relies on some assumptions , we use numerical simulations of mutant enhancer fixation or loss in a finite population to check the corresponding results . Fig 3 illustrates ratio of the probability of fixation of new enhancer alleles relative to 1/2Npop , the probability of fixation of a neutral allele . Numerical simulations were performed using a C++-program of an individual-based stochastic version of the model described above . There are Npop individuals in the population . Each individual has two loci , enhancer and gene , with two alleles each . Alleles at the enhancer locus are encoded by a real value representing enhancer strength . Alleles at the gene locus carry either the wild-type allele or a deleterious allele of fitness effect s . The fitness of the diploid genotype is given by w1—hi ( w1—w2 ) , were hi is the dominance in this individual , which can vary depending on the genotype at the enhancer locus ( following Eq 2 ) and where w1 is the fitness of the fittest of the two alleles ( w2 the fitness of the other allele ) . At generation 0 , all individuals are homozygote for the same enhancer allele and the wild-type gene allele . Then , 2000 generations of the life cycle ( diploid selection , meiosis with recombination , mutation and syngamy ) are performed . During these generations , there is no mutation on the enhancer locus . At each generation , the number of mutations on the gene is sampled in a Poisson Distribution with expectation 2Npopu A corresponding number of alleles ( sampled randomly in the population ) is then assigned to be deleterious . We then generate the population of Npop individuals at the next generation , accounting for selection , meiosis and random mating . For each individual in the next generation , we first determine its two parents . Two individuals of the current generation are sampled randomly . When chosen , each candidate is accepted with a probability equal to its fitness , or resampled . Once the two parents are identified , we sample one gamete in each of them ( recombination occurring at a rate r between the two loci ) . After the 2000 generations , the deleterious allele frequency is close to the mutation-selection balance u/hs . A chromosome is then randomly chosen in the whole population and we assign it a new enhancer allele differing in strength . The new enhancer allele is then monitored until fixation or loss . Fixation probabilities were computed from 10000 runs of such simulations for each set of parameter values . The results presented on Fig 3 in the main text are obtained dividing those probabilities of fixation by that of a neutral enhancer in the same conditions ( i . e . a mutant enhancer allele having the same strength than the resident allele ) . Results showed on Fig 3 were obtained using the following parameters values: ( 1 ) dominance coefficient h = 0 . 25 , ( 2 ) gene mutation rate u = 10−3 , ( 3 ) population size Npop = 103 , ( 4 ) selection coefficients s = 0 . 1 or 0 . 01 and ( 5 ) enhancer mutant strength three times larger or smaller than the resident allele . The 0 . 25 value of dominance is the most biologically plausible value for deleterious mutations [78] . Results illustrated are valid for other combination of u and Npop provided Npopu = 1 . In other situations , the results would be magnified about 1 by a factor Npopu . For instance in a very large population Npop = 108 with weaker gene mutation rate u = 10−5 , we have Npopu = 103 , so that tightly linked stronger enhancers would be ∼2 . 103 more likely to fix than a neutral allele ( instead of ∼2 more likely as illustrated on Fig 3 for Npopu = 1 ) . Fig 6 illustrates this behavior , where the fixation probability of stronger enhancers , relative to the neutral expectation , scales linearly with Npopu . A simplifying assumption of the model described above is to consider a population with only two enhancer alleles . We also designed an infinite-allele version of the model to check the robustness of the conclusions and to study long-term enhancer strength evolution . The goal here is to study the evolution of the mean strength of a population of enhancers undergoing recurrent unbiased mutations . In order to be consistent , we also considered the gene locus as an infinite-allele locus undergoing recurrent deleterious mutations of various effects . The simulations performed on this model are designed as the fixation probability simulations with an individual-based stochastic model , except that we do not study fixation probabilities but long-term trait evolution . At each new generation , the number of mutations on the enhancer locus is drawn from a Poisson distribution with mean 2NpopuE , uE being the mutation rate on the enhancer locus per individual per generation . We denote zi = Log ( ei ) , the logarithm of enhancer strength of allele i . We consider that mutations alter the trait zi such that after mutation it becomes zi + ε , where ε is a normal deviate ε∼N ( 0 , σE2 ) . We considered additive mutational effect on the logarithm of enhancer strength , in order to avoid that the mutational variance vanishes on mean enhancer strength when trait value increases . Indeed , in the model , only relative enhancer strength matters in the competition for expression . As a consequence , if mean enhancer strength increases in the long run , a constant mutational variance on the trait ( not its Log ) would tend to produce less and less differences between enhancers such that relative enhancer strength ratios will eventually tend to 1 . Mutation as described above avoids this artifact and also ensures that enhancer strength always remains positive . To model mutations on the gene , the number of mutation events is drawn for each generation from a Poisson distribution with mean 2NpopuA , uA being the mutation rate on the gene locus per individual per generation . For each mutation event , the fitness effect of the new mutant allele is drawn from a negative exponential distribution with mean s . In order to measure the rate of the ER process , we follow the population mean of z through time . This mean increases linearly with time and we use the slope of this linear increase to compute the mean doubling time ( i . e . the time needed for mean enhancer strength to double ) . To obtain the mean doubling time , we first store at regular time points z1 ( i , j , t ) and z2 ( i , j , t ) , the logarithms of the strengths of both enhancers of individual i at generation t during iteration j ( simulations are repeated 100 times ) . For each generation sampled , we calculate mean z over the whole population and over the different iterations: z¯ ( t ) =1Nit∑j=1Nit∑i=1Npopz1 ( i , j , t ) +z2 ( i , j , t ) 2Npop ( 8 ) As z¯ increase linearly with time , we estimate its rate of increase using a linear regression . Doubling times T×2 is computed as: T×2=Log ( 2 ) a ( 9 ) where a is the slope of this regression . Results of this model are illustrated on Fig 4 ( red curves ) and were obtained using the following parameter values: ( 1 ) dominance coefficient h = 0 . 25 , ( 2 ) mutation rates uA = uE = 10−3 , ( 3 ) selection intensity s = 0 . 1 , ( 4 ) recombination rate between the gene and its enhancer r = 10−6 , ( 5 ) initial strength of enhancers e0 = 1 , ( 6 ) population size Npop = 5000 , ( 7 ) number of iteration Nit = 100 , ( 8 ) number of generations Ngen = 100000 . Note that this model is a special case of model 2 and 3 presented below . In model 4 , we want to study modifications in escalation rates resulting from variation in the mating system . Here , we use the same assumptions than in model 1 except that each individual gets a probability ps to self-fertilize . Self-fertilization is modelled by sampling the second gamete from the same diploid parent than the first . Doubling times are calculated as previously and are reported on Fig 5 . | With the advent of new sequencing technologies , the evolution of gene expression regulation is becoming a subject of intensive research . In this paper , we report an entirely new phenomenon acting on the evolution of gene regulatory sequences . We show that in a small genomic region around genes there is a selection pressure to increase expression , such that stronger enhancers are favored . This leads to an open-ended escalation of enhancer strength . This outcome is not a particular case and we expect it to occur for all genes in nearly all eukaryotic diploid organisms . We also show that this escalation is not stopped by stabilizing selection on expression profiles . Indeed , regulators may coevolve to maintain optimal phenotypes despite the enhancer strength escalation . This widespread phenomenon can significantly shift our understanding of gene regulatory regions and opens a wide array of possible tests . | [
"Abstract",
"Introduction",
"Model",
"Discussion",
"Methods"
]
| []
| 2015 | Enhancer Runaway and the Evolution of Diploid Gene Expression |
C . elegans inhabit environments that require detection of diverse stimuli to modulate locomotion in order to avoid unfavourable conditions . In a mammalian context , a failure to appropriately integrate environmental signals can lead to Parkinson’s , Alzheimer’s , and epilepsy . Provided that the circuitry underlying mammalian sensory integration can be prohibitively complex , we analyzed nematode behavioral responses in differing environmental contexts to evaluate the regulation of context dependent circuit reconfiguration and sensorimotor control . Our work has added to the complexity of a known parallel circuit , mediated by interneurons AVA and AIB , that integrates sensory cues and is responsible for the initiation of backwards locomotion . Our analysis of the galanin-like G-protein coupled receptor NPR-9 in C . elegans revealed that upregulation of galanin signaling impedes the integration of sensory evoked neuronal signals . Although the expression pattern of npr-9 is limited to AIB , upregulation of the receptor appears to impede AIB and AVA circuits to broadly prevent backwards locomotion , i . e . reversals , suggesting that these two pathways functionally interact . Galanin signaling similarly plays a broadly inhibitory role in mammalian models . Moreover , our identification of a mutant , which rarely initiates backwards movement , allowed us to interrogate locomotory mechanisms underlying chemotaxis . In support of the pirouette model of chemotaxis , organisms that did not exhibit reversal behavior were unable to navigate towards an attractant peak . We also assessed ionotropic glutamate receptor GLR-1 cell-specifically within AIB and determined that GLR-1 fine-tunes AIB activity to modify locomotion following reversal events . Our research highlights that signal integration underlying the initiation and fine-tuning of backwards locomotion is AIB and NPR-9 dependent , and has demonstrated the suitability of C . elegans for analysis of multisensory integration and sensorimotor control .
Neuronal circuits integrate various environmental stimuli to alter behavioral phenotypes . Characterization of the mechanisms underlying multisensory integration is essential to understanding how organisms perceive their environment . Disruptions to neural signaling pathways have been linked to a variety of disorders including epilepsy and schizophrenia [1 , 2] . Although mammalian models have provided useful insights , the complexity of the vertebrate nervous system coupled with limitations in cell specific analysis have impeded the characterization of neural circuits . Alternatively , C . elegans has proven to be a model organism for neuroscience due its established connectome , relatively simple nervous system of 302 neurons , and an ability to respond to an array of environmental conditions [3–7] . Individual neuronal pathways have been identified for distinct nematode behaviors , yet our understanding of how multiple environmental cues are integrated and can reconfigure signaling pathways is limited . For instance , nematodes respond slower to volatile odorants when a food source is not available [6] . Evaluation of behavior in the presence of multiple environmental cues provides a means to analyze how distinct cues can modify neuronal circuits . Indeed , mammalian studies have identified that seemingly unrelated cues ( auditory and visual ) have pronounced effects on the perception of the observer [8] . Environmental stimuli alter nematode locomotory patterns , e . g . backwards and forwards locomotion , via unique pathway [6 , 9] . For instance , mechanical and odorant stimuli are detected by the two ASH sensory neurons which modulate the activity of interneurons AIB and AVA to promote the initiation of backwards locomotion , i . e . a reversal [6 , 9] . Despite the shared neuronal contributions , the signaling pathways are differentially modulated by alternative environmental cues that are simultaneously integrated . More specifically , the presence of food alters neuronal activity and responsiveness to an odorant , while signaling dynamics underlying mechanical stimulation remain unaffected [6 , 10] . Understanding the mechanisms , i . e . neurons and signaling molecules , that allow for differential modulation of sensory pathways is essential to the field of multisensory integration . Interneurons AVA and AIB function in parallel to trigger the initiation of backwards locomotion upon their activation [9] . However , AIB expresses serotonin , glutamate , and neuropeptide receptors , while AVA is only known to integrate glutamatergic signaling [11–13] . Mutants that lack a functional vesicular glutamate transporter , encoded by eat-4 , are largely defective for the initiation of a reversal in response to environmental stimuli [14] . However , multiple neurotransmitters and neuropeptides are released to regulate circuit activity [15 , 16] . Consequently , AIB has the unique potential to integrate diverse , context-dependent signals from sensory neurons . AIB is also densely connected to the interneuronal circuitry highlighting its capacity for a broad role in the regulation of downstream or parallel pathways ( Table 1 ) [3] . AIB has been identified as a hub for the integration of signals derived from sensory neurons ASH , AWC and ASE in the regulation of octanol responses on and off food [17] . Within this study , we have expanded the range of behaviors analyzed in order to determine if AIB serves as a broad integration hub or is specific to octanol . Since glutamate is an essential signaling molecule of many behaviors and has been shown to regulate AIB activity in response to environmental cues , we evaluated glr-1 cell specifically across a range of behaviors [9 , 12 , 17] . Interactions between galanin and glutamate signaling pathways have been highlighted in mammalian models [18] . Hence , we also investigated a galanin-like G-Protein Coupled Receptor ( GPCR ) , namely NeuroPeptide Receptor-9 ( NPR-9 ) . NPR-9 is expressed exclusively in AIB and modulates spontaneous locomotion . Loss of function alleles of npr-9 produce an increase in “dwelling” behavior characterized by a lack of forward locomotion , while over-expression of npr-9 increases forward locomotion [13] . Loss of function of npr-9 alters spontaneous reversal frequencies on and off food , while also producing abnormal off food octanol responses . glr-1 knock-down ( KD ) specifically in AIB only interferes with off food octanol responses , however glr-1 ( KD ) in an npr-9 ( LF ) background alters spontaneous locomotion frequencies . Over-expression of npr-9 abolishes the initiation of a reversal in response to nose touch , octanol , copper , or during spontaneous locomotion . Ablation of either AVA or AIB reduces spontaneous reversal frequency by ~50% . Ablation of both interneurons nearly abolishes reversals suggesting that AVA and AIB function in parallel pathways to initiate a reversal [9] . Over-expression of npr-9 mimics the dual ablation of AVA and AIB . Provided that npr-9 is only expressed in AIB , the npr-9 ( GF ) phenotype suggests that NPR-9 can regulate both pathways . Moreover , identification of a mutant that rarely reverses has also allowed us to interrogate the behavioral mechanisms underlying long-range chemotaxis . The weathervane model of chemotaxis purports that nematode turning behavior mediates chemotaxis , while the pirouette model suggests that reversals are responsible [19–21] . Mutants exhibiting abnormal turning behavior exhibit normal chemotaxis responses , which has cast doubt on the weathervane model [22] . Here we report that the pirouette model is the primary mode of nematode chemotaxis . Collectively , our research indicates that NPR-9 is a key regulator of AIB , which serves to integrate signals from multiple sensory neurons and coordinate the interneuronal circuitry to control locomotion .
Food availability serves as an environmental cue to modify spontaneous and evoked behavioral patterns . Nematodes move slower in the presence of food and exhibit less reversals compared to organisms moved to environments absent of food [23 , 24] . These behavioral changes are associated with food availability and are largely regulated by serotonin signaling , i . e . serotonin signaling is up-regulated while organisms are on food [23] . Indeed , the mutant tph-1 , which is unable to biosynthesize serotonin , exhibits abnormal locomotory patterns during the transition from a well-fed to starved state [24 , 25] . Inhibitory serotonin receptor MOD-1 is expressed in AIB and plays a role in the food-based serotonergic regulation of locomotion [11 , 17 , 26] . AIB has been implicated in mediating the transition from well-fed to starved behavioral states [24 , 27] . Manipulations to NPR-9 dependent signaling alter general roaming behavior , however the effect on reversal frequency remains uncharacterized . In order to isolate the specific locomotory defect of npr-9 mutants , we evaluated reversal frequency . Behavioral analysis with and without food has aided in the characterization of context dependent circuit reconfiguration . Wildtype C . elegans exhibit spontaneous reversals on food 3 times per minute and 6 times per minute off food [28 , 29] ( Fig 1A ) . In contrast , npr-9 ( LF ) animals exhibit a decreased reversal frequency off food and an increased reversal frequency on food ( Fig 1A and 1B ) . The abnormal reversal frequency of npr-9 ( LF ) mutants indicates that the neuropeptide receptor coordinates circuit activity yet promotes differing locomotory patterns based upon environmental cues . Decreases in glutamatergic signaling , as observed in mutants defective for glutamatergic transmission ( e . g . loss of glutamate receptor GLR-1 ) , decreases reversal frequency both on and off food [29 , 30] . Interneuron specific expression of glr-1 suppresses locomotory defects in glr-1 mutants indicating that glutamate regulates interneuron activity to control locomotion [29 , 30] . The similar on and off food phenotypes suggest that glutamatergic regulation is independent of nutritional state . However , a metabotropic glutamate receptor within AIB is essential in the regulation of starvation responses [27] . Consequently , the role of glutamate in regards to contextual regulation of locomotion remains unclear . In order to evaluate the role of GLR-1 cell specifically , we knocked-down expression of glr-1 within AIB . Organisms that lacked AIB glr-1 expression exhibited no change in reversal frequency . Provided that a parallel glutamatergic pathway is mediated by AVA and AIB , the lack of glr-1 expression is likely compensated by GLR-1 activity in AVA . Interestingly , AIB or AVA ablation reduces reversal frequency by ~50% , yet the lack of a stimulatory glutamate receptor in AIB has no effect [9] . This suggests that alternative receptors are able to maintain endogenous AIB activation patterns . Ablation of either AIB or AVA reduces on food reversal frequency by approximately 50% , but does not abolish reversals entirely . Collective ablation of these interneurons reduces reversal frequency by ~80% indicating that they promote reversals independently . AIB activation leads to inhibition of motor neuron RIM to trigger the initiation of backwards locomotion [9] . Organisms with ablated AIBs also display decreased reversal frequencies off food suggesting that AIB activation promotes reversals in either context [24] . Over-expression of npr-9 led to more than a 90% reduction in reversal frequency both on and off food ( Fig 1A and 1B ) . Under standard locomotion assay conditions , AIB ablation or AIB-expression of cell death gene egl-1 reduces reversal frequency by roughly 50% [9 , 17 , 24] . Consequently , the severe lack of reversals exhibited by npr-9 ( GF ) organisms suggests that NPR-9 regulates AVA in a cell non-autonomous manner ( Fig 1A and 1B ) . Although AVA and AIB work in parallel to initiate a reversal , regulatory interactions have been identified between the two pathways [3 , 24] . Regulation of AVA is largely mediated by glutamatergic signaling while AIB activity is dictated by a number of signaling molecules suggesting that AIB and AVA circuits are not redundant [11–13 , 30] . Cross-talk between AVA and AIB mediated circuits would ensure that AIB signal integration is incorporated to fine-tune locomotion . Such cross-talk could include synaptically transmitted glutamate or peptides that are expressed in interneurons . Omega turns are a component of C . elegans locomotion in which the nematode exhibits a 135° change in direction resembling an omega ( Ω ) symbol . These turns are characteristically observed after off-food reversals of three or more head swings , termed “long reversals” [24] . Null glr-1 mutants exhibit a decrease in the number of omega turns per minute [31] . Similar to the nose touch pathway , glutamate is likely the major coordinator of omega turns . Since omega turns occur at the end of a reversal , the regulation of omega turns is likely independent from the signaling pathway underlying the initiation of backwards locomotion . Analysis of omega turn frequency thus allows for the interrogation of neuronal fine-tuning independent of reversal initiation . Organisms with ablated AIB neurons exhibit a reduction in omega turns indicating that AIB promotes omega turn behavior , while AVA ablations do not alter omega turn patterns [24] . Prior analyses of omega turns have characterized the behavior by measuring either the frequency of omega turns or the absolute number [24 , 31] . For instance , an organism that exhibited 6 reversals and 3 omega turns in three minutes would be calculated as 50% with our research parameters , while absolute measurements would present the data as 1 omega turn/minute . However , an organism that reversed 30 times and exhibited 3 omega turns would still be represented as 1 omega turn per minute , while we would present that as 10% . We calculate omega turn frequency relative to the total number of reversals in order to determine how frequently an omega turn occurs after a reversal as opposed to counting omega events without consideration for reversal frequency . Considering omega turns are coupled to reversals , the number of omega turns per minute is dependent on reversal frequency . Thus , measuring the absolute number of omega turns simultaneously analyzes the initiation and termination of reversals , which are distinctly regulated [17] . In order to solely interrogate the circuitry regulating omega turns , we have evaluated glr-1 KD , glr-1 and npr-9 mutants for omega turn frequency . Wildtype organisms perform an omega turn after roughly 40% of reversals off food ( Fig 2 ) . glr-1 null organisms exhibited an increased omega frequency ( Fig 2 ) . To the contrary , loss of glr-1 within the AIB produced a deceased omega turn frequency ( Fig 2 ) . Consequently , GLR-1 function within AIB serves to promote omega turns while alternative neurons inhibit omega turns via GLR-1 . Despite a role in regulating off-food spontaneous reversals , npr-9 null animals exhibited no significant change in omega turns ( Fig 2 ) . Surprisingly , the omega turn frequency of npr-9 ( LF ) ;glr-1 KD did not differ from N2 ( Fig 2 ) . This could suggest that AIB regulation of omega turns via GLR-1 is dependent on functional NPR-9 signaling . Since omega turns occur after the initiation of reversals , termination events , e . g . omega turns , are likely dependent on the upstream signaling dynamics . Over-expression of npr-9 decreased omega turn frequency likely due to the absence of reversals ( Fig 1 ) . We suspect that npr-9 ( GF ) exhibits infrequent omega turns due to inhibition of the upstream reversal initiation . Collectively , the data indicates that NPR-9 is essential for the initiation of a reversal , while GLR-1 fine-tunes the termination of a reversal via omega turns . The initiation of backwards locomotion following mechanostimulation to the nematode nose is known as the “nose touch response” . In contrast to the signaling pathway underlying spontaneous locomotion , serotonin does not regulate the nose touch response [32] . Defective nose touch responses amongst glutamate receptor mutants ( glr-1 and glr-2 ) indicate that glutamate is the primary signaling molecule regulating the nose touch circuitry [33] . The initiation of backwards locomotion evoked from nose touch is dependent on the AVA/AIB parallel pathway that governs spontaneous locomotion reversal frequency . GLR-1 activity in either AVA or AIB is sufficient to restore the nose touch response in a glr-1 mutant background [9] . Although AVA and AIB interneurons regulate reversal patterns , similar to spontaneous locomotion , the absence of a role for serotonin suggests that the circuit is differentially modulated . We evaluated nose touch responses of npr-9 and glr-1 KD mutants to delineate signaling differences between spontaneous locomotion and reversals provoked by nose touch . In agreement with the parallel pathway , glr-1 KD in the AIB exhibited no defect in nose touch responses ( Fig 3 ) . Despite the reduction in spontaneous reversal frequency , npr-9 ( LF ) and npr-9 ( LF ) ;glr-1 KD organisms exhibited wildtype nose touch responses ( Figs 1A , 1B and 3 ) . To the contrary , AIB ablation results in a reduced , although not abolished , nose touch response which suggests that abnormal AIB regulation should affect signal dynamics [3] . However , npr-9 ( GF ) resembled the glr-1 null mutant and rarely responded to nose touch stimulation ( Fig 3 ) . Despite the distinct pathways underlying spontaneous and nose touch reversals , npr-9 ( GF ) fails to reverse in either context which suggests that signals downstream of NPR-9 broadly inhibit multiple circuits . Octanol is a volatile odorant that stimulates backwards locomotion . In the presence of food , organisms respond to dilute ( 30% ) octanol within 3–5 seconds . Without food in the environment , nematodes initiate backwards locomotion within 8–10 seconds [34] . Application of exogenous serotonin , in the absence of a bacterial lawn , can restore on food responses [34] . A complex network of neuropeptides and neurotransmitters regulate dilute octanol responses on and off food [15 , 16] . Sensory neuron AWC basally stimulates AIB via GLR-1 , while ASE derived glutamate acutely inhibits AIB activity AVR-14 . Moreover , MOD-1 functions to basally inhibit AIB activity while organisms are on food [17] . Collectively these results indicate that AIB is an integration hub for signals derived from differing sensory neurons that act antagonistically to regulate AIB activity . The integration of multiple signals allows for “fine tuning” of backwards locomotion , i . e . dictating the trajectory of movement after reversal termination . Since npr-9 null animals have been shown to display hyper-aversive responses to dilute octanol off food , we evaluated octanol response [15] . In agreement with previously published research , glr-1 KD and npr-9 ( LF ) organisms displayed a hyper-aversive off food response ( Fig 4B ) . However , npr-9 ( LF ) ;glr-1 KD responded slower than npr-9 ( LF ) on and off food , but did not differ from glr-1 KD in either condition ( Fig 4A and 4B ) . The hyper-aversive npr-9 ( LF ) and glr-1 KD octanol phenotype suggest that NPR-9 and GLR-1 inhibit the initiation of reversals during off food octanol conditions . Loss of both receptors in AIB continues to produce a hyper-aversive phenotype that resembles glr-1 KD . This suggests that signaling downstream of NPR-9 activation is partly dependent on GLR-1 regulation of AIB activity in the octanol pathway . In agreement with an inhibitory role for NPR-9 , npr-9 ( GF ) mutants did not respond to 30% octanol on or off food ( Fig 4A and 4B ) . The lack of reversals exhibited by npr-9 ( GF ) could be due to the inhibition of signaling pathways that promote backward locomotion or simply due to impaired muscle contraction . AIB is pre-synaptic to AVA , which could facilitate the transmission of inhibitory signals after NPR-9 activation to inhibit reversals [3] . Thus , we sought to analyze npr-9 ( GF ) for a behaviour that was not mediated by interneurons that are post-synaptic to AIB . The harsh touch response , in which prodding an organism with a wire pick anterior to the vulva induces a reversal , is primarily mediated by AVD , while an AVA ablation produces no defect [35] . AIB does not exhibit gap junctions or chemical synapses with AVD [3] . To determine if the npr-9 ( GF ) defect is due to a physiological or signaling abnormality , we evaluated the mutant for the harsh touch response . Application of harsh touch to npr-9 ( GF ) mutants induced reversals at wildtype levels ( Fig 5 ) . The positive harsh touch response suggests that the lack of reversals in response to differing stimuli is reflective of circuit miscoordination rather than a physiological inability to reverse . Moreover , this result reinforces the hypothesis that if NPR-9 is exclusively expressed in the AIB then it likely inhibits multiple interneurons via the transmission of signaling molecules through synaptic connections . Similar to the harsh touch response , the AVD is the primary mediator of reversals evoked from the plate tap [21 , 36] . At low concentrations , diacetyl is a chemoattractant to C . elegans [37] . odr-10 is a GPCR expressed in AWA that is essential for diacetyl chemotaxis [38] . Analysis of the interneurons that are involved in diacetyl chemotaxis has revealed a complex network of regulation . Inhibition or activation of interneurons AIB , AIY , AIZ , and AIA can alter an organism’s ability to exhibit chemotaxis to high or low concentrations of diacetyl [39] . These results indicate that complex interneuronal regulation is essential for the regulation of locomotory responses underlying general chemotaxis behavior . In order to determine the importance of circuit regulation in chemotaxis , we analyzed npr-9 ( LF ) and glr-1 KD organisms . Despite the irregular locomotory patterns , npr-9 ( LF ) , glr-1 KD , and npr-9 ( LF ) ;glr-1 KD all responded to diacetyl ( Fig 6 ) . In the context of diacetyl , it appears that loss of npr-9 and/or glr-1 in the AIB does not interfere with the integration of diacetyl-derived signals . C . elegans chemotaxis is also mediated by locomotory behaviours; however , the behavioral mechanism underlying attraction is an issue of debate . Two theories have been proposed to explain chemotaxis: the weather vane model and the pirouette model [19–22] . The weathervane model purports that nematode attraction is mediated by gradual turns that allow a nematode to sense concentration changes and alter direction accordingly . On the other hand , the pirouette model suggests that chemotaxis is mediated by sharp changes in direction and reversals , i . e . pirouettes . To determine the role of reversals and the pirouette model in chemotaxis , we evaluated npr-9 ( GF ) for diacetyl chemotaxis ( Fig 6 ) . The lack of reversals within npr-9 ( GF ) mutants translated to a defective chemotaxis response in support of the pirouette model of chemotaxis . Although npr-9 ( GF ) mutants rarely reverse , they do exhibit turning behavior , thus the chemotaxis defect is most likely due to the reversal abnormality [13] . As previously mentioned , a plate tap can induce the initiation of backwards locomotion in npr-9 ( GF ) mutants . To further determine the importance of reversal behavior as a mediator of chemotaxis , we incorporated a mechanosensory stimulus , i . e . a plate tap , into the diacetyl assay . If the npr-9 ( GF ) diacetyl defect is derived solely from a lack of reversals , the induction of reversals via an alternative stimulus should restore diacetyl responsiveness . In support of this hypothesis , consistent administration of plate taps throughout the diacetyl assay increased chemotaxis responses in npr-9 ( GF ) mutants ( Fig 7 ) . Since the diacetyl defect of odr-10 is due to an inability to detect low concentration diacetyl , the induction of reversals should not alter the phenotype . As expected , odr-10 mutants remain defective despite evoked reversal behavior ( Fig 7 ) . Observational analyses have noted that nematodes are less likely to initiate a reversal when moving towards a higher concentration of an attractant [20] . The introduction of a plate tap during diacetyl chemotaxis assays with large sample sizes likely triggers backwards locomotion in nematodes that are moving towards and away from an attractant . Consequently , worms that are already oriented towards an attractant could be negatively affected by inducing a reversal . We introduced plate taps in assays that evaluated individual nematode responses as they moved towards and away from an attractant . N2 and npr-9 ( GF ) organisms exhibited no difference in their ability to re-orient towards an attractant regardless of their initial trajectory ( Fig 8 ) . We also established that reversals increase the likelihood that organisms reorient towards an attractant; however , a reversal did not guarantee that an organism will immediately re-orient to an attractant . The behavior is likely repeated throughout the course of chemotaxis until a nematode reaches the attractant peak . Ultimately , locomotory patterns , i . e . turning and reversals , must be a component of chemotaxis as they are integral to nematode locomotion . However , in regards to the initial detection of diacetyl , reversal behavior alone is sufficient for wildtype chemotaxis . ASH neurons also mediate copper aversion; G protein GPA-3 mediates responses to copper while ODR-3 transduces signals related to osmotic and mechanical avoidance [40–42] . The introduction of copper in the presence of food leads to increased depolarization of ASH relative to off food activity . Such changes in electrical activity are reflected behaviorally; organisms respond more robustly to copper if food is present [32] . npr-9 ( GF ) organisms do not exhibit behavioral changes in response to food suggesting that the integration of food derived signals is inhibited ( Figs 1 , 2 and 3 ) . A modified food race assay was utilized to evaluate if npr-9 ( GF ) organisms can reach and maintain themselves on a bacterial food patch over the course of a 4 hour assay [43] . Moreover , a repulsive stimulus , i . e . copper , was also incorporated into the assay to evaluate if the abnormal aversive behavior of npr-9 ( GF ) organisms would persist for extended periods of time . Coupling of attractant and antagonistic sensory cues has proven useful in characterizing the molecular underpinnings of behavioral choice [44] . Intuitively , the requirement for nutrition , i . e . food , is more important than exposure to potentially harmful environmental conditions if an organism is forced to choose between the two . As a consequence , the copper food race assay evaluates behavioral choice between an attractant ( food ) and an aversive cue ( copper ) over an extended period of time [44] . Wildtype organisms are able to cross the copper barrier , find the food patch , and alter their locomotory behavior to maintain themselves on food ( Fig 9A ) . Without food , nematodes rarely cross the barrier despite experiencing starvation conditions for 4 hours ( Fig 9B ) . This indicates that nematodes can sense the presence of food across the copper barrier and that stimulation via food over-rides the aversive response as worms become more starved . We observed that npr-9 ( GF ) animals would cross the barrier shortly after transfer to the plate and reach the food source . However , they did not maintain themselves on the food patch and would even re-cross the copper barrier rather than stay in a food-rich environment . Despite the ability to cross the copper barrier , npr-9 ( GF ) organisms were incapable of modulating locomotory patterns to maintain themselves on the food patch ( Fig 9A ) . Moreover , the absence of food did not alter aversive behaviors of npr-9 ( GF ) mutants ( Fig 9B ) . Provided that food should stimulate a higher aversive response , this suggests that upregulation of NPR-9 dependent signaling interferes with the integration of food cues [32] .
The pathway underlying spontaneous reversal frequency is regulated in a tonic fashion [17] . Acute inputs , reflective of novel environmental conditions , can modify AIB activity to shift from tonic to phasic regulation . The mechanisms that regulate such transitions in mammalian models have been impeded by the complexity of signaling pathways [46] . Analysis of AIB regulation in C . elegans has provided a more comprehensive view of the processes that mediate regulatory shifts [17] . Interneurons AVA and AIB function in parallel to initiate a spontaneous reversal on food; however , AVA regulation is primarily mediated via glutamate receptors , while AIB integrates serotonin , glutamate , and neuropeptide signals [9 , 11–13] . Activation of AVA or dis-inhibition of AIB can lead to a reversal [9] . In mammals , a similar sensorimotor circuit has been identified within the basal ganglia known as the ‘direct and indirect pathways model’ [47 , 48] . Initially , the direct and indirect pathways were believed to be distinctly regulated and have opposite effects on movement , however recent evidence has suggested that these circuits are structurally and functionally intertwined [48] . The presence of a synaptic connection from AIB to AVA suggests that these two circuits are structurally intertwined . Functional interactions between the AVA and AIB circuitries have also been identified in the regulation of locomotory responses to osmotic changes [9] . On food , AIB activity is inhibited via serotonin-gated chloride channel MOD-1 and dis-inhibition events trigger a reversal [17] . The presence of food and subsequent increased serotonin has no modulatory effect on the npr-9 ( GF ) phenotypes indicating that excessive NPR-9 signaling interferes with the integration of serotonergic food cues . Similarly , serotonin pathways can reconfigure mammalian circuits and excessive galanin inhibits the integration of serotonin signals [49 , 50] . After a reversal event , nematodes recommence forward locomotion , alter their trajectory with minor turns , or perform an omega turn to facilitate a large directional change . Accordingly , the prevalence of omega turns is dictated by neuronal fine-tuning following reversal initiation . Fine-tuning of neurons in complex behaviors has also been highlighted in optic motor control and neuronal responses to opioids [51 , 52] . Such neuronal fine-tuning provides a mechanism that allows for a small population of neurons to mediate diverse and complex behaviors . Recent evidence has highlighted that inhibitory and stimulatory glutamatergic inputs coordinate AIB activity to dictate behaviors following a reversal . However , up-regulation of NPR-9 can impede AIB dis-inhibition related to reversals and omega turns despite unique regulatory mechanisms underlying each behavior . Ablations to interneurons AIY , RIB , RIM , and RIV also alter omega turn frequencies indicating that the omega turn circuitry is likely regulated by outputs from a number of interneurons [24] . In agreement with a general inhibitory role , galanin broadly inhibits neuronal activity in mammalian models [53–55] . GLR-1 activity within either circuit , i . e . the activation of AVA or AIB , is sufficient for a nose-touch evoked reversal . The lack of reversals observed in npr-9 ( GF ) animals suggests that signals from AIB can inhibit reversals initiated by AVA . Existing synaptic connections from AIB to AVA could mediate the transmission of modulatory signals [3] . Unlike spontaneous locomotion and octanol , the presence or absence of food does not modulate nose touch evoked backwards locomotion [10] . Moreover , the touch response is instantaneous , indicating that acutely stimulated pathways can operate on different timescales . Despite the different timescale and signaling modalities , excessive galanin receptor activation impedes multiple circuitries to inhibit backwards locomotion . Analyses of on and off food octanol responses allows for the dissection of acute pathways that are reconfigured by the presence of food , i . e . context dependent circuit reconfiguration [56] . Similar circuit reconfiguration has also been shown to modulate vertebrate locomotory pathways [57] . Reversals stimulated via octanol are regulated by a complex signaling pathway that is coordinated by tyramine , octopamine , serotonin , glutamate , dopamine , and neuropeptides [15 , 17 , 58] . Octopamine is analagous to norepinephrine and its role in octanol signaling been likened to noradrenergic inhibition of nociception , i . e . pain , in mammals [59] . Inhibition of AIB , via a histamine-gated chloride channel , reduces spontaneous reversal frequency , yet produces a more rapid octanol response [17] . Despite the complexity of the octanol pathway , abnormal regulation of one neuron , i . e . AIB , via up-regulation of NPR-9 , prevents the integration of any signal that could modify locomotory phenotypes . Similarly , up-regulation of galanin signaling in a mammalian context has been shown to impede the integration of food related environmental cues associated with Alzheimer’s disease and has been linked to antinociception in mammalian models [60 , 61] . Responses to aversive stimuli facilitate the avoidance of unfavourable habitats . However , organisms exhibit less robust aversive responses the longer they are removed from food [6 , 62] . A reduction in serotonin signaling reconfigures neuronal networks to facilitate exploratory behavior to find a food source , while ignoring aversive cues [24] . Despite starvation conditions and altered neuronal regulation , upregulation of galanin signaling prevents locomotory modulation upon nematodes reaching a food source . Interneurons AVA/AVD/AVE collectively regulate the initiation of some reversals , however each neuron has unique synaptic connectivity and receptors suggesting unique roles [3 , 11–13] . For example , AVD ablation does not differ from mock ablations relative to spontaneous reversals , but produces a reversal defect for harsh touch responses [9 , 36] . AIB is pre-synaptic to AVA and AVE , but is not pre-synaptic to AVD [3] . Provided that excessive galanin signaling does not inhibit harsh touch responses , the regulatory of influence of AIB is likely facilitated via synaptic transmission rather than extrasynaptic mechanisms . C . elegans sense fluctuations in odorant concentration rather than absolute concentration , thus locomotory patterns must somehow accommodate changes in direction . Changes in reversal behavior ( pirouette model ) and/or changes in turning behavior ( weathervane model ) have been proposed to explain how modulation of locomotion can enable the detection of different attractant concentrations [19–21] . Conversely , nematodes that exhibit a severely defective turning behavior are not defective for chemotaxis suggesting that turning behavior is non-essential for attractant detection [22] . The lack of chemotaxis in npr-9 ( GF ) , a mutant that turns but does not reverse , reinforces that only reversals are essential to diacetyl chemotaxis . Similar to the pirouette model , the biased random walk facilitates bacterial chemotaxis [63] . In our research we have shown that npr-9 , a galanin-like GPCR , is a key regulator of reversal frequency in response to diverse stimuli . The analysis of multiple behaviors has highlighted the essential role of NPR-9 in regulating the interneuronal circuitry and has reinforced that AIB is a key hub for the integration of diverse sensory outputs . Moreover , regulation via NPR-9 can be circumvented in neurons that are not post-synaptic to AIB ( e . g . AVD ) . Our study illustrates that excessive galanin signaling impedes the integration of environmental cues in AIB and that NPR-9 broadly coordinates the interneuronal circuitry despite expression limited to a single interneuron .
C . elegans were maintained at 20°C on NGM agar plates seeded with OP50 E . coli according to standard protocols . The following strains were used: wild-type strain N2 , IC683 npr-9 ( tm1652 ) , IC836 quIS20 [npr-9::npr-9; sur-5::gfp; odr-1::rfp] , KP4 glr-1 ( n2461 ) , MT6308 eat-4 ( ky5 ) , CX3410 odr-10 ( ky225 ) . Germline transformations were performed according to standard protocol . For glr-1 KD , 50 ng of odr-1::RFP plasmid was used as the co-injection marker with 50 ng of the glr-1 KD plasmid . For each assay L4 hermaphrodites were picked 16–24 hours prior to the start of an assay . All assays were performed at 20–23°C . Nematodes that were damaged during transfer or roamed off of the plate were not included in the data set . For on food reversal frequency assays , plates were seeded with 200μl OP 50 E . coli bacteria . The bacteria was then distributed to cover the entire plate . Seeded plates were incubated overnight and allowed to dry prior to the assay . Reversal frequency assays were performed as previously described [64] . Briefly , a young adult worm was transferred to a seeded an assay plate and allowed to acclimate for 1 minute prior to the start of the assay . Reversals were counted manually over a three minute period . Off food reversal frequency measurements were performed similarly; however , worms were transferred to an intermediate plate with no food for 1 minute prior to the assay start time to ensure that bacteria would not be carried over . A nematode was transferred from an intermediate plate to an assay plate with M9; excess M9 was removed delicately with a kim-wipe . Nose touch , octanol , and diacetyl assays were performed as previously described [6 , 65–67] . Nose touch assays were carried out in the absence of food . A hair was placed in front of a forward moving worm so that the organism collides with the hair at a perpendicular angle . Each organism was tested 10 times for nose touch and data was recorded as % responding . For octanol assays , responsiveness was determined by placing a hair dipped in 30% octanol in front of a forward moving organism . On food assay plates were prepared as previously mentioned . Off food octanol assays were performed after the nematode remained off food for ten minutes . Diacetyl assays were performed with a minimum of 50 worms , as dispersal timing can be population density dependent . Diacetyl assay plates were prepared as follows: 4 equal quadrants were marked on the underside of the plate with a 0 . 5 cm radius circle marked in the middle of the plate designated the “origin” . A point was marked in each quadrant that was equidistant from the origin and other quadrant points . Two points were designated as controls ( EtOH and 0 . 5M sodium azide ) or test ( 0 . 5% diacetyl diluted with EtOH and 0 . 5M sodium azide ) . 50–250 nematodes were washed in S Basal buffer and transferred to the origin with test or control solutions added to their designated points after transfer . The number of organisms outside the origin were counted to calculate the chemotaxis index . The “diacetyl tap assay” is a modified version of the previously described diacetyl assay; the modification involves the use of a plate tap to induce a reversal . Plate taps are performed once every 15 seconds throughout the one-hour diacetyl assay . The plate is rotated 90 degrees after each plate tap to prevent directional bias . The individual analysis of diacetyl responses to tap utilized a previously described chemotaxis assay , in which a radial gradient was established and a gradient peak is clearly marked on the plate [21] . Plate taps were incorporated into this assay and administered throughout . Single nematodes were tested for the ability to orient towards the attractant peak after tap application while moving away or towards the peak . A minimum of 5 trials per organism were used to evaluate tap responses . The copper-modified food race assay was designed based on a previously utilized food race assay and behavioral choice assay [43 , 44] . Plates were seeded with OP50 E . coli with a food patch that was positioned to one side of the plate rather than central . After the seeded plates were incubated overnight to allow for bacterial growth , approximately 100 μl of a 0 . 5M copper ( II ) sulfate solution was applied to an agar plate to create a barrier that divided the agar plate in two sections . This solution was also applied around the edge of the plate to prevent organisms from leaving the assay plate entirely . Once the copper ( II ) sulfate solution had dried , organisms were transferred to the section of the plate with no food with M9 . Excess M9 was removed with a kim-wipe . Every 30 minutes worms were positively scored if they crossed the copper barrier and were observed on food . In the absence of food , positive responses were scored if nematodes crossed the copper barrier . | Multiple environmental cues are sensed by an organism in order to coordinate behavioral responses . Consequently , organisms must be able to simultaneously detect and integrate multiple external stimuli in order to appropriately modify their behavior . Identifying the unique circuits mediating the response to individual stimuli and points of overlap is essential to understanding how multiple signals can be integrated for a coordinated behavioral response . In order to analyze individual circuits , we have used the model organism C . elegans . We have identified that a C . elegans neuropeptide receptor ( NPR-9 ) and a glutamate receptor ( GLR-1 ) function in a single interneuron to play a broad regulatory role in multiple neural circuits . Our research has identified that interneuron AIB is involved in the integration of signals from numerous sensory neurons . Moreover , regulation of AIB via a neuropeptide receptor ( NPR-9 ) and a glutamate receptor ( GLR-1 ) coordinates AIB activity in the context of multisensory integration . Long-range chemotaxis behavior , in which an organism alters locomotory patterns based on odorant sensation , is also regulated by NPR-9 . Our analysis indicates that reversals , and thus the pirouette model , are sufficient for chemotaxis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"invertebrates",
"cell",
"motility",
"medicine",
"and",
"health",
"sciences",
"neurochemistry",
"nose",
"face",
"pathology",
"and",
"laboratory",
"medicine",
"caenorhabditis",
"nervous",
"system",
"pathogens",
"neuroscience",
"biological",
"locomotion",
"biomechanics",
"animals",
"animal",
"models",
"caenorhabditis",
"elegans",
"model",
"organisms",
"interneurons",
"neurotransmitters",
"research",
"and",
"analysis",
"methods",
"animal",
"cells",
"neural",
"pathways",
"head",
"chemotaxis",
"foodborne",
"organisms",
"glutamate",
"biochemistry",
"cellular",
"neuroscience",
"neuroanatomy",
"cell",
"biology",
"anatomy",
"physiology",
"neurons",
"nematoda",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
]
| 2016 | NPR-9, a Galanin-Like G-Protein Coupled Receptor, and GLR-1 Regulate Interneuronal Circuitry Underlying Multisensory Integration of Environmental Cues in Caenorhabditis elegans |
Data assimilation is a valuable tool in the study of any complex system , where measurements are incomplete , uncertain , or both . It enables the user to take advantage of all available information including experimental measurements and short-term model forecasts of a system . Although data assimilation has been used to study other biological systems , the study of the sleep-wake regulatory network has yet to benefit from this toolset . We present a data assimilation framework based on the unscented Kalman filter ( UKF ) for combining sparse measurements together with a relatively high-dimensional nonlinear computational model to estimate the state of a model of the sleep-wake regulatory system . We demonstrate with simulation studies that a few noisy variables can be used to accurately reconstruct the remaining hidden variables . We introduce a metric for ranking relative partial observability of computational models , within the UKF framework , that allows us to choose the optimal variables for measurement and also provides a methodology for optimizing framework parameters such as UKF covariance inflation . In addition , we demonstrate a parameter estimation method that allows us to track non-stationary model parameters and accommodate slow dynamics not included in the UKF filter model . Finally , we show that we can even use observed discretized sleep-state , which is not one of the model variables , to reconstruct model state and estimate unknown parameters . Sleep is implicated in many neurological disorders from epilepsy to schizophrenia , but simultaneous observation of the many brain components that regulate this behavior is difficult . We anticipate that this data assimilation framework will enable better understanding of the detailed interactions governing sleep and wake behavior and provide for better , more targeted , therapies .
Great strides have been made in understanding the physiological basis for sleep regulation [1] in terms of the interacting neuronal cell groups and their neurotransmitter mediated interactions . This physiology is now increasingly being embodied into complex mathematical models of sleep dynamics [2]–[6] . But the limits to which these models are either validated or otherwise utilized for insight and prediction is currently limited . Due to physical and technological constraints , simultaneous measurement of the physiology embodied in the models - such as cell group firing rates and neurotransmitter concentrations - is not feasible in freely behaving animals or people . We demonstrate here that such models of the sleep-wake regulatory system can be put into a data assimilation framework that allows for reconstruction and forecasting of unobserved dynamics from limited noisy measurements . We anticipate these tools will help shed light on core brain circuitry implicated in sleep disorders as well as sleep-related neurological disorders such as epilepsy [7] , bipolar disorder [8] , and generalized anxiety disorder [9] . Data assimilation is an iterative process that couples and synchronizes mathematical models to observed system dynamics with the purpose of estimating both noisy observed and unobserved variables , as well as forecasting the future system state . Data assimilation algorithms for nonlinear systems often employ the ensemble Kalman filters [10] . One such ensemble filter is the unscented Kalman filter ( UKF ) , used in an iterative prediction-correction scheme in which model-generated predictions are corrected to agree with or track experimental observations [11] . The objectives of this article are to demonstrate data assimilation applicability within the context of relatively high-dimensional nonlinear biological models of the sleep-wake regulatory system , and to investigate the observability properties of these models [4] , [12] . In the Materials and Methods section , we introduce these models , as well as the basic mathematics of the UKF and parameter estimation algorithms . In the Results section , we demonstrate the use of the UKF to reconstruct data generated from these models . We introduce a reconstruction quantification that allows one to gauge the relative observability of the model variables . We demonstrate how this empirical observability coefficient can be used to optimize UKF parameters such as model covariance inflation , as well as how to select the optimal variables for measurement . We then demonstrate a method for optimizing model parameters for tracking slowly varying dynamics . Finally , we demonstrate that we can use measurements of discretized sleep-state generated from the model , instead of specific model variables , to reconstruct unobserved model dynamics .
Recent advances in single and multi-unit recordings have contributed to the growing knowledge of the mammalian sleep-wake regulatory system . The current prevailing hypothesis includes a flip-flop switch that regulates transitions between non rapid-eye-movement sleep ( NREM ) and wakefulness ( Wake ) [1] . Gamma-aminobutyric acid ( GABA ) -ergic ventrolateral preoptic nucleus ( VLPO ) neurons in the hypothalamus promote NREM . Monoaminergic cell groups in the brainstem , including the noradrenergic locus coeruleus ( LC ) and the serotonergic dorsal raphe ( DR ) neurons , promote Wake . Mutual inhibition between these two groups causes each to promote its own activity by inhibiting the other's . McCarley and Hobson [13] described transitions between NREM and rapid eye movement sleep ( REM ) arising from predator-prey like interactions between cholinergic cell-groups in the brainstem , including the laterodorsal tegmentum ( LDT ) and pedunculopontine tegmentum ( PPT ) , and the monoaminergic cell-groups in LC and DR . For a more in-depth overview of the literature , including controversial hypotheses for REM regulation , see [14] . More recently , orexin and adenosine have been implicated in further regulation of the sleep-wake system . Orexin producing neurons in the lateral hypothalamus have descending projections to all aforementioned monoamergic and cholinergic cell groups and reinforce arousal , for a review see [15] . Extracellular adenosine has been found to increase during prolonged wakefulness in several cortical and subcortical regions [16] , and has been proposed as a homeostatic accumulator of the need to sleep [17] . These dynamics are further modified by the circadian drive [18] , regulated by the suprachiasmatic nucleus ( SCN ) in the hypothalamus , which sets a roughly 24-hour cycle affecting sleep and many other physiological functions . The SCN has indirect projections to the VLPO in the hypothalamus which results in inhibition of sleep during the day [19] . Here day is subjectively defined by species' dependent diurnal behavior , and refers roughly to 12-hour periods consisting mostly of active-wake behavior . The SCN clock can be modulated by afferent cortical inputs in response to a variety of external cues . Food restriction studies have shown entrainment of the circadian cycle to food availability [20] . Light input from the melonopsin expressing ganglion cells in the retina can also affect the SCN [21] . Retrograde trace studies have shown that a number of central nervous system sites innervate the SCN in the rat [22] , though further study is needed to fully elucidate the involved circuitry . For instance , it is well known that lesions of the temporal lobe leading to epileptic seizures also affect the circadian clock [23] , [24] , but the relevant brain circuitry has yet to be determined . The DB model [4] , depicted in Fig . 1A , describes interactions among five distinct neuronal populations: two Wake-active groups , LC and DR; two groups in the LDT/PPT , one that is REM-active , denoted R; one active both in Wake and REM , denoted W/R; and one group active during NREM in the VLPO . As illustrated in Fig . 1A , these cell groups communicate through various transmitters: LC transmits norepinephrine ( NE ) , DR transmits serotonin ( 5-HT ) , the two groups in the LDT/PPT transmit acetylcholine ( ACh ) , and VLPO transmits GABA . Excitatory thalamic input is modeled by the variable and the brain's homeostatic sleep drive is represented by . Sample output of this model's sleep-wake cycles , as well as mutual inhibition between Wake and sleep-active regions is shown in Fig . 1B . Each cell group is described by its firing rate and the concentration of the neurotransmitter that it releases to post-synaptic populations . Cell group firing rates are a function of their input neurotransmitter concentrations , and evolve according to: ( 1 ) Here is a weighted sum of neurotransmitter into cell group firing rate , with coupling constants ; ( 2 ) In addition , is a first order process time constant . The steady state firing rate , , a function of input neurotransmitter , is given by maximum firing rate parameter , times a sigmoidal function with midpoint and slope : ( 3 ) where is a constant for all cell groups except VLPO where is proportional to the homeostatic sleep drive . The concentration of neurotransmitter released by each cell group to the post-synaptic space also evolves according to a first order process with time constant and steady state neurotransmitter concentration given by: ( 4 ) ( 5 ) where is an adjustable scale parameter . Because ACh comes from both the R and W/R cell groups , the total ACh concentration in the post-synaptic space is the sum of the ACh concentrations generated individually from these groups . Random excitatory projections from thalamocortical circuits to the Wake-active populations LC and DR are modeled as Poissonian impulses with a rate of 0 . 003 Hz , which through a leaky integrator form another input concentration denoted with a decay constant of seconds: ( 6 ) In addition to firing rate and neurotransmitter concentration variables , the homeostatic drive variable regulates the duration of sleep and wake bouts by changing , the threshold for firing of the NREM-active VLPO cell group . The accumulation of during Wake , and dissipation during sleep , is given by: ( 7 ) where is the Heaviside function , is the threshold parameter for the onset of increase or decrease in , and and determine the rate of accumulation and dissipation . Typical output of the DB model is shown in Fig . 1B . The top three traces are the time dynamics of the firing rates for the Wake-active ( LC ) , NREM-active ( VLPO ) , and REM-active ( LDT/PPT ) cell groups . Note that , following [4] we denote firing rate for the REM-active LDT/PPT cell group as . The state of vigilance ( SOV ) , or sleep state , shown as a hypnogram in the fourth trace , is determined by the rank-ordered comparison of these cell group activities , with LDT/PPT activity dominating the definition . Fleshner , Booth , Forger and Diniz Behn [12] introduced an extension of the DB model that includes the SCN as an additional cell group with GABA as its associated neurotransmitter [12] , depicted in Fig . 1C . The firing rate of the SCN cell group follows the same dynamics as the cell groups in the DB model ( Eq . 1 ) . The SCN has an inherent 24-hour circadian cycle ( ) , with higher activity during the 12-hour light phase and lower activity during the 12-hour dark phase . The projections from the sleep-wake network to the SCN provide dynamical feedback that increases the SCN's activity during both Wake and REM and decreases its activity during NREM . The SCN receives 5-HT and ACh synaptic inputs from the core sleep-wake regulatory system through the variable . This is modeled by composing from the sum of and . Although the amplitude of is smaller than that of , its oscillation time-scale is faster , typically on the order of minutes . ( 8 ) ( 9 ) ( 10 ) Here hours . We have shifted the phase of from [12] by adding to make the light period ( high ) start at 6 am . Feed-forward projections of the SCN on the sleep-wake network are mediated through GABAergic transmission , modeled by the additional neurotransmitter concentration , which adds into the dynamics of the LC , DR , VLPO , and R firing rates , modifying Eq . ( 2 ) from the DB model to become: ( 11 ) Typical output of the FBFD model on short time-scales is similar to that of the DB model . But , as is typical for real rats , on diurnal time-scales the typical duration times in different states , as well as cycle times through states , changes . The hypnogram of the output of this model is shown in Fig . 1D for a 36 hour period . Rats are nocturnal . In the model , REM and NREM are primarily observed during the putative light phase , while long periods of Wake are observed during the putative dark phases . The Kalman filter estimates the state of a system from noisy , sparsely measured , variables . Kalman's initial filter derivation [25] was for linear systems . The unscented Kalman filter is an ensemble version developed to tolerate nonlinearities without linearization [26] . The details of the UKF algorithm can be found in many standard textbooks [27] , [28] . We present here an overview , along with the key equations needed to understand details presented later in this manuscript . State estimation with the UKF is carried out recursively using a prediction-correction scheme . Each iteration starts with a best estimate of the current state at iteration time . Included is an estimate of the current uncertainty in state . A prediction or forecast is then generated by iterating an ensemble of points near , called sigma points , through the nonlinear model dynamics . Given a dimensional state space for , we choose sigma points such that they have covariance to represent the state uncertainty . We denote the sigma point prior to iteration , and after iteration . The model prediction is then the mean of the forward iterated sigma points: ( 12 ) The prediction uncertainty is then the covariance of these points plus an additive covariance inflater matrix . ( 13 ) is nonzero only on the diagonal , and is added to account for underestimates of the forecast error , from the covariance of the sigma points , due to process noise and inadequacies in the filter model [29] , [30] . The prediction is then corrected to account for a measurement at time . need not contain the same number of variables as . The correction factor weights the observation and prediction according to the Kalman gain : ( 14 ) where denotes the prediction mean from the estimated sigma points for the observed variables , : ( 15 ) The Kalman gain is formed from the ratio : ( 16 ) where and are formed from averages over the sigma points , either in the full dimensional space of or in the subspace spanned by the measurements: ( 17 ) ( 18 ) where is the measurement uncertainty . The Kalman gain is also used to correct - and ideally collapse - the prediction uncertainty: ( 19 ) Within this recursive scheme , the UKF synchronizes model state to measurements and thereby improves the estimate of the experimentally inaccessible variables . The limit to which this succeeds depends in part on the relative observability of the reconstructed model variables from the measured variables . We discuss below an empirical method for assessing this relative observability . We note that , the uncertainty of the measurement process usually can be estimated , using the assumption that measurement noise is normally distributed [31]–[33] . On the other hand , the additive covariance inflation parameter is less clearly defined . Some methods have been proposed to estimate its values under the limited case that its source is an additive process noise [34]–[38] . Within our results - we demonstrate that even with identical system and model dynamics , non-zero improves tracking , and present a method of optimally choosing the values of for tracking and prediction . One approach for parameter estimation within the UKF framework is to solve the dual problem of estimating parameters and states at the same time , for instance via an augmented state space approach [11] . The alternative approach is to separate state reconstruction from parameter estimation by iteratively alternating between the two [39] . We found that dual estimation did not work well for our high-dimensional sleep-models , likely in part due to the many degrees-of-freedom when neither parameters nor variables were fixed , and especially because in nonlinear systems the sensitivity of the dynamics to particular parameters can be highly dependent on location in state space . We therefore estimate parameters iteratively over windows of length that are longer than a typical sleep-wake cycle of the dynamics . Within our method , hidden states are first reconstructed with the UKF using a filter model with initial best-guess parameters . The full-state reconstruction over is then used in a parameter estimation step which yields an updated parameter set . This updated parameter set is then passed to the UKF for the next iteration . This process is repeated until the parameter estimate has stabilized . The parameter estimation step is essentially an application of a multiple shooting method [11] , [40] . Within each window , we estimate parameters by creating an average cost-function that quantifies the divergence between short model-generated trajectories and the UKF-reconstructed trajectories for the measured variables . We then minimize this cost-function with respect to the parameter of interest . In order to prevent the model-generated trajectories from diverging too far from the reconstructed ones , we reinitialize on the reconstructed trajectories at regular intervals : ( 20 ) We then calculate a cost-function averaged over the window using the divergence between the model-generated trajectories and : ( 21 ) where denotes a matrix with non-zero elements on diagonal positions corresponding to measured elements . In order to properly weight the errors for each variable , the non-zero elements of are set to the inverse of the standard deviation of the associated variable . For our current implementation , we perform a minimization with respect to parameters by explicitly computing for test parameters and then choosing the one with the minimum . We use a constant value of , and restrict our parameter update maximally to per iteration . Though somewhat computationally intensive , this method yields a stable approach to a local minimum in . This also limits the resolution to which the parameter can be estimated . For estimation of non-stationary parameters , we use overlapping windows , with an update period of . We note that should be greater than the maximal expected rate of change of the parameter of interest , to ensure that parameter dynamics are estimated with reasonable fidelity .
We can accurately reconstruct unmeasured variables of the DB model of sleep with the UKF framework . To demonstrate this , we generate data from this model , then apply a noisy observation function - the output of which is a noisy subset of the variables - to mimic experimental conditions . We then reconstruct the unobserved variables with the UKF . Finally , we validate this reconstruction by comparing to the original data set . An example of this procedure is shown in Fig . 2 . Long multivariate time series of sleep-wake data were generated from the DB model . The observation function yielded a noisy univariate version of the firing rate of the Wake-active LC region . Explicitly we added random , normally-distributed , zero mean noise with variance of 4% that of the variance of to the true values . We provided the framework the parameters used to generate the original data , and either default values ( left panels ) or optimized values ( right panels ) for the covariance inflation parameter . Default values of were chosen as times the typical variance of each variable . Additionally , the initial conditions of the model state , , were arbitrarily chosen in each case . Shown in Fig . 2 are the reconstructed ( red ) and true ( black ) values of the NREM-active firing rate variable , the REM-active firing rate variable and the stochastic thalamic noise variable . In both cases of tracking , reconstruction of the observed variable is good . This can be seen from the closeness of the reconstructed traces to both the observation points , shown in blue , and the true values in black . Likewise , the reconstruction of also tracks the true state quite well . However , for the default values of , is not reconstructed as well , and is not reconstructed at all . These errors extend to lower reconstruction fidelity of and even of . On the other hand , when we use optimized values the reconstruction of is improved . In addition , much of the thalamic noise input through - which is stochastically driven and receives no input from the other variables - is now represented . For these reconstructions , we initialized the model state far from that of the true system state . Therefore , there is a transient period during which reconstruction is poor . In our experience , once the model state comes close to that of the true system , this data assimilation framework keeps the model relatively close to the system state . We can find optimal framework parameters , such as the UKF covariance inflation parameter matrix , by maximizing values . Although the matrix only has nonzero diagonal terms , for the full DB model including the thalamic noise output variable , there are 12 's . So blind simultaneous optimization is inefficient . But we can use the full , and the ranked partial observability , as a guide to this optimization . Note from Eq . 13 that adds to the diagonal elements of the covariance of the sigma points . This inflation has the effect of widening the sigma points on the next iteration step , which results in an increase in the Kalman gain . Larger values for the Kalman gain bias the correction towards the measurements . Our general rule therefore is that if measurement of a variable yields poor reconstruction of other variables - i . e . low values of down a column - then we should favor measurement derived values of other variables over model derived ones , and therefore should use increased values of . On the other hand , if a variable is not reconstructed well from other variables - i . e . low values of across the row - we should favor model derived values over measurement derived values for this variable by decreasing . We iteratively compute the , then choose the variable with the lowest scores down a row or column , and change its corresponding appropriately . We then recompute the and repeat . This prevents us from optimizing with respect to 's that have only modest impact on reconstruction fidelity . There is a finite usable range for . As an inflater for the covariance matrix , must be greater than or equal to zero . The standard deviation and range of the dynamics of variable are two natural scales that can be used to define the usable range of . We use the square of the former , multiplied by a proportionality constant , as a default starting value for . The square of the latter forms the maximum for . We now demonstrate this algorithm to optimize reconstructions given measurements of , as in Fig . 2B . The full for the DB model - including - with default is shown in Fig . 4A . From the far right column , we observed that no variables are reconstructed well from measurements of . This is understandable , since the dynamics of receive no input from any of the other variables . Therefore we start our optimization of by adjusting , and explicitly expect to increase it . Shown in Fig . 4D is as a function of increasing . Although only the trace for is shown , increases for most variables as a function of increasing . We pick optimal values for based on the average peak reconstruction of all variables from measurement of , found with a value of . The matrix after this first optimization iteration is shown in Fig . 4B . Notably , although is the variable measured from the real system , its reconstruction improves when is increased . This effect can be further understood by inspection of Fig . 2 . The brief increases in from its low value - interpreted behaviorally as brief awakenings that correlate with spikes in in Fig . 2 , are better reconstructed with optimized . Indeed , the matrix values overall , shown in Fig . 4B , have increased with increasing . Now the row/column with the lowest values , on average , corresponds to reconstruction of . Therefore we expect to need to decrease to improve reconstruction . Reconstruction fidelity of from measurement of , as measured by is shown in Fig . 4E as a function of . Reconstruction improves with decreasing values over the potential usable range . As shown by the black horizontal line , the best reconstruction is achieved using the minimum value of 0 for , although values of smaller than the default value of result only in marginal reconstruction improvement . This second optimization step yields only marginal improvement in the overall matrix shown in Fig . 4C . In part , this lack of improvement in reconstruction is due to the poor observability of REM dynamics through other variables as apparent from the row marked in Fig . 3 . We investigated pairings of two or more variables with respect to their relative partial observability . We found that for the DB and FBFD models , the empirical observability of variable given measurements of variables is always at least as good as the individual : . We also observed that good reconstruction of all variables requires some measurement of both Wake and REM dynamics . These states are readily observed from real biological systems from external physiological measurements such as power bands in the EEG , muscle tone , and eye movement . Therefore , for the subsequent computations , we assimilate noisy measurements of both Wake-active and REM-active dynamics , and use them to reconstruct the full system state and derive parameter values . As applied here , the UKF framework requires both a model for the dynamics as well as the model's parameters . We have implemented a version of a multiple shooting method [11] for optimizing the choice of parameters . The performance of this method is illustrated in Fig . 5 . For illustration purposes we generated data with fixed parameters and assimilated noisy measurements of and to reconstruct the dynamics . Initially , all model parameters in the UKF were set to the same values used to generate the true data set - except the parameter that couples ACh into dynamics . To this we supplied an arbitrary initial value . Parameter estimation is performed by minimizing the distance between the UKF reconstructed traces and short model-generated trajectories that originate on the reconstructed traces . For these computations , we set the length of these short trajectories at 2 minutes . This is long enough that differences in parameters result in measurable divergence between the short computed trajectories and the reconstructed dynamics . Here measurable is much larger than the measurement noise , but not so large that the distance between the computed and reconstructed trajectories becomes comparable to the range of the state space . To sample the full state space , each step of this minimization averages this divergence over time windows longer than the sleep-wake cycle time of the dynamics . As seen in Fig . 5A , our estimation of converges to the true value . In Fig . 5B , we plot trajectories for the short model-generated ( magenta ) , reconstructed ( red ) , and true ( black ) dynamics for different periods of the convergence of . Note that initially , for significantly different than the true value , the short trajectories diverge quickly from the reconstructed values , and the reconstructed values of are different from the true ones . When approaches the true value , both the short model-generated and reconstructed trajectories approach the true dynamics . As coded , the parameter estimation essentially optimizes short model-generated forecasts . To investigate the effect on reconstruction fidelity , we compute the normalized mean square reconstruction error for each variable , averaged over each parameter estimation window . This is shown for variables , , and the homeostatic drive . We note that for initial values of this parameter , even reconstruction of the measured variable is quite poor - with typical errors of its standard deviation . As a reference point , the initial measured data - a noisy version of - has a normalized mean squared error , shown as a horizontal blue line . As the estimated parameter converges , falls well below , and the reconstruction metric improves for all variables . We can also estimate parameters which change slowly over time . We demonstrate this by using a slightly modified DB model , which lacks any circadian dynamics , to reconstruct dynamics observed from the expanded FBFD model which specifically includes SCN driven circadian oscillations . We use this modified DB model to assimilate noisy measurements of and from the full FBFD model , and use it within the multiple shooting method to estimate the value of . An example of the output is shown in Fig . 6 for a 1 . 5 day period . We have skipped the initial 12 hours which includes a transient period of convergence of both the filter and the parameter estimate . The effect of the SCN is to modulate the overall sleep cycles , with frequent sleep periods that include REM in the light period and dominant , longer Wake periods in the dark period . Short example time series for and are shown in the panels in Fig . 6A for different phases of the circadian cycle . The filter model in the UKF , also used in parameter estimation , is missing these SCN associated variables and the fast feedback oscillations resulting from their interaction with the sleep network . However , we replace the input contribution of the SCN's feed forward GABAergic projections on to the sleep network to a single quasi-static parameter that gets added to other neurotransmitter variables in Eq . ( 11 ) . We then estimate this parameter which represents the presumed SCN drive . The estimated value for ( magenta ) is shown in Fig . 6B , along with the true input from SCN in the generating model ( black ) . Though the reconstructed parameter is an estimate with inherent averaging over half-hour periods , and therefore does not reproduce the fast dynamics of the real SCN input , it tracks the mean value quite well . In addition , it yields good reconstruction of the model variables . Examples of the normalized reconstruction error , averaged over the fitting windows , are shown in Fig . 6C for sample variables . Here again , as a reference point , we plot the mean squared error for the noisy measurement of ( blue line ) in the top panel of Fig . 6C . Note that even reconstruction of the homeostatic sleep drive , which has no direct coupling to the observed variables , is quite good over most of the day . So far , we have implemented the data assimilation framework using measurements that amount to noisy versions of the true variables . In real applications , when one uses observations from real systems , the actual system measurements might only remotely resemble variables in the tracking model . But even in this case , data assimilation methods can still be used . To this end , we demonstrate that we can use measurements of state-of-vigilance ( SOV ) generated from the model and illustrated in Fig . 1B , to reconstruct the unobserved model dynamics with reasonable fidelity . The method we have implemented is illustrated in Fig . 7 . We sleep-score the model-generated data , also used in Figs . 2–5 , by assigning an SOV to each point as a function of time . The SOV is determined based on relative values of , , and . We then take the filter model , and generate example data , which we also sleep-score . From this scored filter-model data , we compute the probability distribution functions ( pdf ) for the variables , , and conditioned on SOV . These are illustrated in Fig . 7A . Note that these state-dependent distributions are highly skewed , and have small variance around the mean . The observation function from the measured values - here SOV as a function of time shown in Fig . 7B - must provide values and error estimates of variables in the filter model to the UKF . To translate the observed SOV to inputs to the UKF , we use the state-conditioned medians from the above-generated pdfs , and then use the state-conditioned standard deviations as the measurement uncertainties . In this way , we use observations of SOV to infer observations of the model variables . We then use these observations , shown for in blue in Fig . 7C , as inputs to the UKF . Note that in this case , the measurement noise estimates are time dependent . After a short transient convergence time , the reconstructed dynamics converge close to the true dynamics . However , certain details such as brief awakenings and transitions into NREM are not reconstructed well . We can likewise apply all the other tools described here to assimilation of SOV data through this inferred observation function . Shown in Fig . 8 is the same parameter estimation procedure as shown in Fig . 5 , with the same initial conditions for unknown parameter . Although the convergence is not as good as with direct observation of , the estimated parameter does approach the parameter used to generate the data . The reconstruction error in decreases as the parameter approaches its correct value , however neither converge all the way . This can be understood because the UKF attempts to constrain the observed variables to the median values mapped from the SOV . Likewise , the parameter estimation algorithm attempts to minimize the error between model forecasts and reconstructions for the observed variables . As a supplemental performance metric , we also consider the reconstruction error if we simply use the median observation map for all variables as our reconstruction . These are plotted as horizontal dashed lines for each variable in Fig . 8C . The UKF reconstruction error for the observed variable improves beyond this reference point as parameter estimation improves . In contrast , the UKF reconstruction errors for unobserved variables such as and are overall far better .
Data assimilation is a valuable tool in the study of any complex system , where measurements are incomplete , uncertain , or both . It enables the user to take advantage of all available information including experimental measurements and short-term model forecasts of a system . Since the introduction of the UKF to neuronal dynamics by Voss et al . in 2004 [11] , a few investigators have applied these methods to the study of biological systems [43]–[48] . Other data assimilation techniques have also been successfully applied to study neuronal dynamics [49] . Nevertheless , the sleep modeling community has yet to utilize these resources . Several important advantages of data assimilation in sleep modeling are : 1 ) access to unmeasured variables to create a more complete estimate of model state 2 ) subject-specific parameter estimation even when the parameter is associated with an unobserved variable 3 ) allowance for uncertainty in model structure or measurements and 4 ) prediction of future dynamics . Not all variables perform equally in reconstructing the state space . In biological experiments utilizing data assimilation it would be beneficial to have some insight into the relative performance of each variable so that we can choose the best one or ones for measurement . A natural metric to guide this choice is the observability of each variable . Letellier et al . showed in [50] that observability and ability to synchronize are related . Since the UKF is basically a synchronization scheme , it follows that reconstruction-performance by any variable should be a function of its observability . Thus we propose using observability based metrics in the study of partially observed biological systems . Analytical methods to determine observability for nonlinear systems are mathematically rigorous , require rational models , and generally do not produce graded values for partial observability . Letellier et al . [51] , [52] proposed a simple algebraic solution to rank all variables of a system according to their relative partial observability . Although their approach works well for low-dimensional systems , we found it problematic for our high-dimensional sparsely connected system , where many variables are directly coupled to just one or two other variables , and where the coupling is effectively on only in highly localized regions of state space . Inspired by their work , we developed an empirical metric , the , to rank the partial observability of each variable based on reconstructed error . The can be used to select the optimal observed variable to obtain the best estimate of a particular unobserved variable . The absolute optimal observed variable receives as input to its dynamics unambiguous invertible information about the state of the unobserved variable . Here invertible implies a one-to-one ( bijectvie ) relationship between the unobserved and observed variables . In complex networks , this observability is modulated by the number and relative weights of additional unobserved variables in the system that couple into the dynamics of the observed variable [50]–[52] . Because the is a measure of reconstruction fidelity , we demonstrate that the reconstruction framework parameters can be optimized by improving it . Importantly , we described an intuitive approach to use the to optimize the covariance inflation parameters . Although some analytical methods have been proposed for this task in nonlinear systems [34]–[38] , we are unaware of an observability-based metric for covariance inflation optimization . Correct parameter estimates aid the prediction of future dynamics and model selection and verification and can provide useful biological information . A common method for parameter estimation in nonlinear models utilizes a feedback-synchronization scheme , developed by [53] and extended by [54]–[58] and many others . Within such a scheme , two identical - except for unknown parameters - systems are unidirectionally coupled , and are continuously synchronized through error feedback . The parameters of the responder are allowed to vary - often using a gradient-decent approach - to minimize a cost function based on driver-responder distance . Although these methods have been shown to work well for systems with smooth variables we found that the sharp transitions in our firing rates , and the highly variable sensitivity of the dynamics to particular parameters as a function of position in state space , resulted in unstable and inaccurate parameter estimates . We therefore adopted a multiple shooting parameter estimation method [11] , [40] that estimates divergence of short model forecasts from the UKF reconstructed trajectories over time windows long enough to explore the state space . This estimation step involves the minimization of a least-squared error , and can therefore be cast as a maximum-likelihood step . This is done in an iterative fashion to update parameter estimates by minimizing divergence of trajectories reconstructed using previous parameter estimates . Therefore this method becomes an expectation-maximization method , with all the associated global optimization implications [59] , [60] . Estimation of one or more parameters with any parameter estimation method will be inherently limited by the identifiability of the state space . Identifiability is a structural property of a model defined as the ability to identify a unique set of parameter values given error-free observations of the dynamics [61] . A comparable experimental or empirical version of identifiability has also been discussed by[62]–[65] . If some parameters are not structurally identifiable no parameter estimation method will prevail . Our experience and expectation is that the multiple shooting method will converge reasonably for combinations of identifiable parameters , but the convergence time increases with the number of parameters . A key advantage of using the UKF for state reconstruction is allowance for uncertainties in the model and/or measurements . As noted , the Kalman filter is an iterative prediction-correction scheme . By altering the elements of the covariance inflation parameters and measurement uncertainty , we can guide the Kalman filter to favor either the observations or model predictions . Higher values of downgrade the model-based forecasts during the correction step . We utilized this when developing the method for optimizing choice of values based on the . For those variables that are poorly observed from others , we more heavily weight prediction over measurement; for those that yield poor reconstruction of other variables , we more heavily weight measurements . Furthermore , as we showed in Fig . 6 , inadequate models - which omit the full dynamics of certain variables - can be used to successfully assimilate experimental data and estimate unknown dynamics . In this example , we used a model that lacked any circadian dependencies to correctly estimate a 24-hour cycle and the mediated interaction with the SCN . Therefore our data assimilation framework can tolerate inadequate models and uncover dynamics outside the scope of the model's governing equations . Several issues must be considered for assimilation of biological measurements . First , initial values for the filter parameters should be estimated off-line via the iterative reconstruct state/estimate parameter approach . During this off-line learning process , non-arbitrary initial values for the covariance matrices as well as model parameters can be determined . Second , we will not have access to many of the state variables for validation . Previously , we developed a system that can automatically stage the behavioral state of a freely moving animal in real time [66] , based on measurements of EEG and head acceleration with a resolution of a few seconds . This process can validate the UKF's predictions of sleep-state transitions . We can also use the scored behavioral state to infer the value of the Wake-active , NREM-active , and REM-active firing rate variables . As we have shown in Fig . 7 and Fig . 8 , we can then use these inferred measurements to reconstruct hidden variables and estimate unknown parameters . It is technically feasible to measure extracellular neurotransmitter concentrations using either dialysis or electrochemical sensors . Dialysis measurements do not have the temporal resolution to resolve REM dynamics , which occur on the order of one minute or less in the rodent , or the spatial specificity to localize dynamics to sleep-wake nuclei in the rodent brain . However these measurements could be used to track and validate slow systemic dynamics such as the circadian variations that modulate the sleep-wake nuclei . In contrast , off-the-shelf electrochemical sensor technology [67] allows for highly localized measures of neurochemicals such as ACh and 5-HT with sub-second temporal resolution and sub-mm spatial resolution . Such measurements can and should be used to establish and validate models used within the data assimilation framework . In addition , they can be used to identify the subset of measurements that can be accurately reconstructed from less costly observations . An appropriate cost-function for biological data assimilation would balance the degree of reconstruction inaccuracy against the cost of obtaining risky or hard-to-access measurements . We also note that this framework could potentially be used to choose among model dynamics . Our parameter estimation methods rely on a minimization of prediction error . A similar metric or cost function could be utilized to differentiate between UKF-based tracking and prediction of system dynamics utilizing different models with such a filter framework . In conclusion , we have presented a data assimilation framework for combining sparse measurements together with a relatively high-dimensional nonlinear computational model to reconstruct unmeasured variables , and have demonstrated its use in the context of a model of the sleep-wake regulatory system . We have demonstrated with simulation studies that once the tracked state approaches the true system state , it reliably reconstructs the unobserved system state ( Fig . 2 ) . We have introduced a metric for ranking relative partial observability for computational models ( Fig . 3 ) that allows us not only to assess reconstruction performance based on choice of measurement , which can serve as a guide to which system variables to measure , but also provides a methodology for optimizing filter framework parameters such as the covariance inflation ( Fig . 4 ) . In addition , we have demonstrated a parameter estimation method ( Fig . 5 ) that allows us to track non-stationary model parameters and accommodate slow dynamics not included in the UKF model such as circadian-dependent input from the SCN ( Fig . 6 ) . Finally , we have demonstrated that we can even use observed discretized SOV , which is not one of the model variables , to successfully reconstruct model state ( Figs . 7–8 ) . These key features will aid in the transition of this framework to the experimental bench . Our long-term plan is to develop an observer-predictor system that will track and predict sleep-wake cycles as well as the underlying state of the neural cell groups and their neurochemical environment . Because these system dynamics are implicated in and interact with numerous neurological diseases from epilepsy to schizophrenia , we anticipate that these tools will enable better understanding of the detailed interactions and provide for better , more targeted , therapies . | Mathematical models are developed to better understand interactions between components of a system that together govern the overall behavior . Mathematical models of sleep have helped to elucidate the neuronal cell groups that are involved in promoting sleep and wake behavior and the transitions between them . However , to be able to take full advantage of these models one must be able to estimate the value of all included variables accurately . Data assimilation refers to methods that allow the user to combine noisy measurements of just a few system variables with the mathematical model of that system to estimate all variables , including those originally inaccessible for measurement . Using these techniques we show that we can reconstruct the unmeasured variables and parameters of a mathematical model of the sleep-wake network . These reconstructed estimates can then be used to better understand the underlying neuronal behavior that results in sleep and wake activity . Because sleep is implicated in a wide array of neurological disorders from epilepsy to schizophrenia , we anticipate that this framework will enable better understanding of the link between sleep and the rest of the brain and provide for better , more targeted , therapies . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"control",
"theory",
"sleep",
"applied",
"mathematics",
"anatomy",
"and",
"physiology",
"physiological",
"processes",
"control",
"engineering",
"mathematics",
"computational",
"neuroscience",
"biology",
"nonlinear",
"dynamics",
"systems",
"biology",
"control",
"systems",
"physiology",
"computational",
"biology",
"engineering"
]
| 2012 | Reconstructing Mammalian Sleep Dynamics with Data Assimilation |
Migraine can be sub-classified not only according to presence of migraine aura ( MA ) or absence of migraine aura ( MO ) , but also by additional features accompanying migraine attacks , e . g . photophobia , phonophobia , nausea , etc . all of which are formally recognized by the International Classification of Headache Disorders . It remains unclear how aura status and the other migraine features may be related to underlying migraine pathophysiology . Recent genome-wide association studies ( GWAS ) have identified 12 independent loci at which single nucleotide polymorphisms ( SNPs ) are associated with migraine . Using a likelihood framework , we explored the selective association of these SNPs with migraine , sub-classified according to aura status and the other features in a large population-based cohort of women including 3 , 003 active migraineurs and 18 , 108 free of migraine . Five loci met stringent significance for association with migraine , among which four were selective for sub-classified migraine , including rs11172113 ( LRP1 ) for MO . The number of loci associated with migraine increased to 11 at suggestive significance thresholds , including five additional selective associations for MO but none for MA . No two SNPs showed similar patterns of selective association with migraine characteristics . At one extreme , SNPs rs6790925 ( near TGFBR2 ) and rs2274316 ( MEF2D ) were not associated with migraine overall , MA , or MO but were selective for migraine sub-classified by the presence of one or more of the additional migraine features . In contrast , SNP rs7577262 ( TRPM8 ) was associated with migraine overall and showed little or no selectivity for any of the migraine characteristics . The results emphasize the multivalent nature of migraine pathophysiology and suggest that a complete understanding of the genetic influence on migraine may benefit from analyses that stratify migraine according to both aura status and the additional diagnostic features used for clinical characterization of migraine .
Migraine is one of the most common and debilitating neurological disorders and its clinical presentation can be quite variable [1] . Even when the diagnosis of migraine meets consensus criteria and can in most cases be clearly distinguished from other types of headaches ( e . g . tension-type headache ) , phenotypic heterogeneity in migraine persists [2] , [3] . The most pronounced heterogeneity in migraine is the dichotomous sub-classification according to the presence ( MA ) or absence ( MO ) of aura , which most commonly manifests as a visual disturbance that generally precedes an attack of headache fulfilling the criteria for migraine . Other characteristics that may be used to sub-classify migraine are features of the migraine attack , including pulsatile pain character , unilateral pain , photophobia , phonophobia , attack duration , nausea , aggravation by physical activity , severity that inhibits daily activities , and finally the frequency of attacks . The International Classification of Headache Disorders ( ICHD ) acknowledges all these characteristics either as diagnostic criteria for migraines or to distinguish different forms of migraine [4] . Although our understanding of the migraine and aura pathophysiology has substantially improved [5] , many details of migraine aura and the role of other migraine features remain unclear . The heterogeneity of migraine characteristics raises both a challenge and opportunity for using genetics to understand migraine pathophysiology . While the power to detect genetic associations will be degraded by potential misclassification due to the heterogeneity of the clinical presentation , associations that are selective for migraine with certain characteristics may help reveal detailed biological causes of migraine , and anticipate the potential of gene-based migraine classification and treatment . Genetics is known to be an important determinant of migraine with heritability estimated in the range 30–60%; and the heritability for MA is estimated somewhat higher than for MO [6]–[10] . Recent reports have described a greater number of highly significant common genetic variants for MO than MA in genome-wide analyses , as well as only partial overlap between the sets of identified genes [11]–[13] . One possible explanation of the apparent discrepancy between heritability estimates and yield of genome-wide significant associations may be different genetic contributions to MA v . MO with , for example , the former possibly characterized by genetic variants that are rarer or more population specific , or more heterogeneous compared with the latter [14] , [15] . Similarly , it is possible that a dichotomy in the genetic architecture may underlie the additional features that often accompany migraine headache , i . e . nausea , photophobia , etc . Here we apply a likelihood-based analytic framework [16] to explore the possibility of preferential associations with sub-classified migraine in a population-based cohort of women for 12 single nucleotide polymorphisms ( SNPs ) arising in recent genome-wide association studies ( GWAS ) for migraine overall , MO , or MA [13] . Enforcing strict significance thresholds , we find that five SNPs are associated with migraine in our cohort among which four had selective association with sub-classified migraine . At suggestive significance , 11 loci were associated with migraine and all but one displayed selective association with sub-classified migraine . However , none of the patterns of selective association according to aura status or the other characteristics was shared by more than one SNP . The findings suggest that the recently reported genetic variants influence the underlying pathophysiology of migraine in very different ways .
Among the Women's Genome Health Study ( WGHS ) participants of European ancestry with available genetic data , there were 3 , 003 women who reported active migraine at baseline , defined as migraine experienced in the year prior to enrollment , compared with 18 , 108 who had never experienced migraine ( Table 1 ) . An additional 2 , 119 reported having experienced migraine previously but not in the preceding year and thus were not sub-classified according to migraine characteristics . These participants were excluded from the main analysis . Compared with non-migraineurs , active migraineurs tended to be younger , have higher BMI , be more likely to use hormone replacement therapy but less likely to smoke . Thirty nine percent ( N = 1 , 177 ) of the WGHS participants with active migraine reported aura , compared with 61% ( N = 1 , 826 ) who did not ( Table 2 ) . The prevalence of features associated with migraine ranged from 34% for pain aggravated by physical activity to 78% for duration of 4–72 hours . Applied to 12 SNPs ( Table S1 ) recently reported with genome-wide association [13] , the statistical model selection procedure with the Bayesian information criterion ( BIC ) penalty ( see Methods for model selection approach ) identified six SNPs with evidence for migraine association in the WGHS displaying selectivity according to one or more of the characteristics accompanying a migraine attack ( Table 3 ) . To estimate the significance of the models selected for each combination of SNP and migraine sub-classification , an empirical approach was used to control for multiple hypothesis testing ( see Methods ) . In particular , permutation of genotype assignments to individuals was used to estimate: 1 ) for each combination of SNP and migraine sub-classification , the probability of selecting a “non-null” model among six models tested with the BIC ( Table S2A ) , 2 ) for each combination of SNP and migraine sub-classification , an empirical p-value for the analytic log-likelihood ( LLR ) test of the BIC selected model ( Table S2B ) , and 3 ) an overall p-value for each SNP correcting the empirical p-value in 2 ) for model selection across all 10 migraine sub-classifications ( Table S2C ) . The empirical p-values selecting a non-null model for each combination of SNP and migraine subtype were in the range 0 . 0002–0 . 0066 . For 2 ) , the six SNPs with “non-null” BIC-selected model for at least one migraine sub-classification had LLR tests with a maximum empirical p-value 0 . 0032 before correction for multiple testing , among which the five excluding rs13208321 ( FHL5 ) were significant after correcting for testing across the 10 migraine sub-classifications ( “*” symbol , Table 3 ) . SNP rs7577262 ( TRPM8 ) was significantly associated with migraine , correcting for testing across all 10 sub-classifications , but displayed no selectivity for any of the migraine-associated characteristics . The remaining four significant SNPs were selective for one or more migraine associated characteristics . For example , SNP rs1172113 ( LRP1 ) was preferentially associated with the migraine without aura , i . e . MO ( “inverse subset” ) , and also for migraine with duration 4–72 hours ( “subset” ) . Models distinguishing status of the other characteristics except pulsatile pain were selected for at least one of the SNPs meeting significance thresholds for multiple testing . However , none of the characteristics showed selective association shared by all SNPs . Model selection with the Akiake information criterion ( AIC ) penalty was less stringent , identifying selective associations according to migraine characteristics for all SNPs except rs7577262 ( TRPM8 ) ( Table 4 ) . The greater number of non-null models could be explained by the more permissive model selection found in permutation analysis that estimated the fraction of non-null models by chance in the range 0 . 16–0 . 35 ( Table S3A ) . Nevertheless , all SNPs except rs10915437 ( near AJAP1 ) had at least one model with nominally significant empirical p-value for the LLR test ( i . e . <0 . 05 ) , and the same five SNPs that had empirically significant LLR tests in the BIC model selection were also significant in the AIC model selection , although in some cases different models were selected ( Tables 4 & S3 A , B , C ) . Thus , for SNP rs10504861 ( near MMP16 ) , the AIC selected the “inverse subset” model for aura ( i . e . MO ) and a “basic” model for migraine characterized by aggravation by physical activity , inhibition of daily activities , or attack frequency ≥6/year compared with the BIC selected “null” model for these characteristics . Similarly , SNP rs13208321 ( FHL5 ) was identified as “null” with BIC model selection but as “inverse subset” for aura by the AIC as well as “subset” for other features . Additional differences at the five SNPs included selection of “general” rather than “subset” models for phonophobia and migraine attack frequency ≥6/year at rs12134493 ( TSPAN ) , and “general” rather than “basic” or “subset” models respectively for phonophobia and aggravation by physical activity at rs2651899 ( PRDM16 ) . Some of the remaining SNPs had AIC-selected models with nominally significant empirical LLR p-values ( Table S3B ) , although none of these models was significant after correction for multiple testing ( Table S3C ) . Nevertheless , the nominally significant selective models highlighted additional differences compared with the BIC penalized analysis , among which “inverse subset” models for aura ( i . e . MO ) were selected at rs9349379 ( PHACTR1 ) and rs6478241 ( ASTN2 ) . Using the same BIC and AIC model selection methodology , there were few differences in the SNP associations between the 3 , 003 active migraineurs and the 2 , 119 former migraineurs who were excluded from the current analysis due to lack of information related to migraine sub-classification ( Table 5 ) . With the BIC penalty , four SNPs were assigned “non-null” models , all of which were of the “basic” type , implying no statistical difference in SNP association between active and former migraine status . With the AIC penalty , five additional SNPs were assigned “non-null” models and only one , rs10504861 ( near MMP16 ) , displayed preferential association suggesting stronger association with active migraine . To examine the model selection results in more detail , the association effects of each SNP for migraine sub-classified according to presence or absence of each characteristic were estimated by logistic regression ( Table S4 ) and depicted in Figure 1 . To aid in presentation of the results , SNPs were ordered according to clustering based on the normalized differences in the association effects for migraine accompanied with or without the characteristics ( Fig . S1 ) . The clustering thus juxtaposed SNPs with approximately similar patterns of selectivity across aura status and the other migraine-associated characteristics . At the top of Figure 1 , SNPs rs7577262 ( TRPM8 ) , rs11172113 ( LRP1 ) , rs6478241 ( ASTN2 ) , rs10915437 ( near AJAP1 ) , and rs9349379 ( PHACTR1 ) form a cluster with relatively less pronounced differences in association by stratum status of the migraine-associated characteristics . In particular , associations with SNP rs7577262 ( TRPM8 ) displayed associations essentially undifferentiated by stratum status , as reflected also by exclusively BIC-selected “basic” models ( Table 2 and Fig . 1 , boxes with heavy dotted outline ) . SNP rs11172113 ( LRP1 ) in this group had mostly undifferentiated associations , except for stratum-specific associations according to aura status ( for MO , beta [SE] = 0 . 14 [0 . 036] , p = 8 . 3×105 compared with MA , beta [SE] = 0 . 057 [0 . 043] , p = 0 . 19 ) and migraine duration 4–72 hours ( beta [SE] = 0 . 12 [0 . 032] , p = 0 . 00012 ) but not duration under four hours ( beta [SE] = 0 . 054 [0 . 058] , p = 0 . 34 ) , as reflected also by “inverse subset” and “subset” models with the BIC ( boxes with heavy solid outline in Fig . 1 ) , respectively . In the middle of Figure 1 , SNPs rs10504861 ( near MMP16 ) , rs13208321 ( FHL5 ) , rs2651899 ( PRDM16 ) , and rs12134493 ( near TSPAN2 ) all show significant associations for migraine sub-classified according to one or more of the migraine characteristics . These findings are consistent with corresponding “subset” and or “inverse subset” models from the BIC ( Table 4 ) . The remaining three SNPs have a mixture of stratum independent and stratum specific associations that differentiate them from the other two clusters . Throughout the array additional differences in the association effects according to stratum status are observed for many SNPs as again reflected also in the “subset” or “inverse-subset” models from the AIC model selection ( Table 5 ) , for example the differences according to aura status at rs9349379 ( PHACTR1 ) and rs6478241 ( ASTN2 ) as above rs10915437 ( near AJAP1 ) , and rs13208321 ( FHL5 ) . Only two SNPs , rs6790925 ( near TGFBR2 ) and rs2274316 ( MEF2D ) , were found to have “null” models for stratification according to aura status , and inspection of the beta coefficients ( Fig . 1 , Table S4 ) suggested additional qualitative differences from the other SNPs . In spite of the very small effects on MA and MO , and essentially no association with active migraine overall , both SNPs had appreciable and significant associations for migraine accompanied by one or more of the characteristics , even showing significant protective association for rs6790925 and migraine without nausea ( beta [SE] = −0 . 095 [0 . 048] , p = 0 . 047 ) and without photophobia ( beta[SE] = −0 . 044[0 . 037] , p = 0 . 23 ) , or for rs2274316 and migraine without pulsation ( beta[SE] = −0 . 062[0 . 042] , p = 0 . 14 ) , although only the first of these combinations was significant . These patterns of association are consistent with the “general” models ( dashed line , Fig . 1 ) that were selected with the AIC penalty and are characterized by different SNP allele frequencies in all three sub-groups , i . e . unaffected individuals as well as migraineurs either accompanied or not with the characteristics .
Examining the 12 SNPs recently discovered for association with migraine , we demonstrated significant preferential associations with MO compared to MA at high stringency for one SNP ( rs11172113 , LRP1 ) and at lower stringency for five SNPs . Of these , only rs10504861 ( near MMP16 ) , had been discovered initially in an association analysis specifically targeting MO . Four additional SNPs had no evidence of selectivity for aura status in their associations with migraine . SNP rs7577262 ( TRPM8 ) in particular was highly significant for association with active migraine but not selective for aura or any of the other characteristics . It is perhaps relevant that TRPM8 , the candidate gene for this SNP , is thought to mediate the sensation of pain rather than specific neurological or vascular functions that might more directly differentiate the pathophysiology of the migraine sub-classes [17] . The final two SNPs were not associated with active migraine , MA , or MO but were associated with migraine accompanied by one or more of the other migraine-specific characteristics , implying that these characteristics may be more relevant to the underlying pathophysiologic consequences of these genetic variants than aura status . Among the candidate functions for loci other than TRPM8 , PRDM16 has roles in cardiac development [18] and directing developmental cell fates toward brown fat or skeletal muscle [19] , SNPs near the matrix metalloproteinase gene MMP16 have been associated at genome-wide significance with psychiatric conditions [20] and non-syndromic cleft lip [21] , and LRP1 , encoding the LDL receptor-related protein 1 with molecular functions in endocytosis in several settings , has been implicated by GWAS in lipid homeostasis [22] , lung function [23] , abdominal aortic aneurysm [24] , and transport of beta-amyloid in the brain [25] . The function of TSPAN2 , the final candidate gene with BIC-selective association in the WGHS , is largely unknown , but it belongs to the tetraspanin family that has been linked to signal transduction [26] . Thus , the genetic architecture of migraine appears to reflect a multivalent pathophysiology and , from the dual perspectives of statistical power and understanding biology , association strategies that rely on the conventional dichotomy according to aura status may not be the sole or even best approach for genetic dissection of migraine . Instead , the patterns of selective association with migraine accompanied by the non-aura characteristics may be at least as important to understanding migraine pathophysiology as the selectivity toward aura . All of the SNPs ( except rs7577262 ) displayed some selectivity according to the non-aura characteristics that commonly accompany migraine , at least with the AIC penalty , and there was at least one SNP with a selective association for each characteristic even with the BIC penalty , which enforced a very high stringency in model selection . No pair of SNPs shared an identical patter of subtype associations . The selective associations are likely not to reflect , trivially , the contribution of the WGHS population to discovery of the candidate SNPs since sub-classification of migraine was not used in the original discovery . Moreover , the WGHS contribution to the main discovery analysis in the previous study [13] included a total of 5 , 122 migraineurs , as the combination of the 3 , 003 active migraineurs analyzed here and an additional 2 , 119 WGHS participants who reported having had migraine in their life but not in the year prior to enrollment . Thus , the migraineurs in the present analysis are a subset of those used in the published meta-analysis , even as the WGHS contributed approximately one-fifth of the total cases in that study . Only rs10504861 ( near MMP16 ) in the previous analysis was identified exclusively in a previous sub-analysis restricted to MO and therefore including only the 1 , 826 WGHS MO cases rather than the 5 , 122 cases with history of any migraine . In contrast , rs10915437 ( near AJAP1 ) and rs6790925 ( near TGFBR2 ) were discovered exclusively among clinic-based samples , excluding the WGHS altogether . However , rs11172113 ( LRP1 ) rather than rs10504861 was identified as an MO-specific in the analysis using the BIC penalized model selection . Thus , the design of the previous discovery meta-analysis is expected to confer only minimal bias at most in the selectivity of associations presented here , especially selective associations identified with the BIC penalty and those identified for sub-classifications other than the MA v . MO dichotomy . There has been a suggestion that migraineurs who remit , i . e . appear to no longer experience migraine , may have different underlying pathophysiology from those who do not . This difference may have a partial genetic basis that might extend to differences in the selective associations reported here . Our data do not allow exploration of this possibility directly since migraineurs who did not report active migraine in the WGHS at baseline were not sub-classified according to aura status or the other migraine characteristics . However , we note that model selection for SNP associations with active compared with former migraine status suggested that the associations with overall migraine in the two groups were largely similar . The exception was rs10504861 ( near MMP16 ) that displayed selective associations using the stringent BIC ( Table 3 ) and also a preferential association with active compared with former migraineurs ( Table 5 ) . One potential , though ultimately not robustly supported , explanation of the subtype associations might be that they simply reflect associations among individuals suffering a greater severity of migraine rather than selectivity for specific features . If this were the case , then one would expect that selectivity patterns might be highly correlated , perhaps especially with associations according to strata for migraine attack frequency , one measure of migraine severity . However , these patterns were not observed . First , the patterns of selective associations are not shared by any of the SNPs . Second , some SNPs show no selectivity ( i . e . “basic” model ) for migraine attack frequency ≥6/year but selectivity ( i . e . “subset” or “inverse subset” models ) for other characteristics , for example aura status ( rs11172113 [LRP1] , rs6478241 [ASTN2] , rs13208321 [FLH5] , rs10504861 [near MMP16] ) . Finally , among SNPs where there is a selective association with migraine characterized by attack frequency ≥6/year , there are few similarities among the associations with sub-classification based on the other characteristics . For example , rs12134493 ( near TSPAN2 ) , which is highly selective for migraine with attack frequency ≥6/year , also shows selectivity for unilateral pain , phonophobia , and photophobia , but not for pulsation , duration of 4–72 hours , aggravation by physical activity , and inhibition of daily activities , all features that show selective association with other SNPs , including SNPs that are also selective for high frequency migraine . Several strengths and limitations should be considered when interpreting our results . Strengths include the large , homogeneous population-based sample of middle-aged women of European ancestry who were apparently healthy at study entry , as enforced specifically by a lack of overt CVD or cancer at baseline . Thus , the WGHS is very well-powered for the migraine sub-classification analysis presented here and further represents an age range in which migraine is relatively prevalent . Limitations include the self-reported nature of migraine and sub-phenotypes , which may result in misclassification . Other , comparably ascertained and well-powered cohorts that also include ascertainment of migraine sub-phenotypes and genotype information are not readily available and this circumstance limited our ability to replicate the analysis . Instead , we used a permutation procedure to establish significant thresholds consistent with multiple hypothesis testing . The study also does not address genetic associations with sub-classified migraine in other groups including women younger than 45 , men , or children , nor does it address explicitly the genetic underpinnings of sub-classification in migraine with a strong familial inheritance pattern . Further targeted studies are warranted to address these issues . The selective genetic associations with sub-classified migraine provide a glimpse into the future possibility of resolving some of the heterogeneity in migraine . Sub-classification of migraineurs according to combinations of migraine-associated characteristics potentially representing more clinically homogeneous sub-groups has been suggested as one approach [2] , [27] . However , sub-groups of migraineurs cannot be unambiguously defined based on discrete patterns of co-occurrence of migraine-associated characteristics . Because of this ambiguity , applying the present statistical methodology to test for selective genetic associations with such sub-groups is much more complex than analyses based on individual migraine characteristics . In considering alternative approaches to solving the complex presentation and pathophysiology of migraine , ongoing research experience in the genetics of psychiatric disorders may be relevant . Psychiatric disorders are notoriously difficult to diagnose , a challenge that also extends to devising optimal treatment . Attempts to classify psychiatric disorders on the basis of clinical symptoms alone , as for example by the updated diagnosis criteria in the recently published DSM-5 , are controversial [28] , [29] . At the same time recent genome-wide genetic analyses have revealed both different and shared causal genetic loci across multiple psychiatric disorders with distinct diagnoses on the basis of clinical presentation alone [30] , [31] . It is not hard to imagine that increasingly detailed clinical and genetic characterization may ultimately coalesce into integrated and more reliable diagnostic criteria for these psychiatric conditions . Such combined clinical and genetic strategies for improved classification may be imagined also for migraine , although they would likely require establishment of a larger number of genetic loci than the 12 robust loci explored in the current analysis . Nevertheless , improved classification of migraine may help identify the most important pathophysiological pathway ( s ) in a given migraine patient and may allow for prioritization of treatment options . In this respect , discovery of more loci and therefore genes relevant to migraine in future genome-wide studies may provide further understanding of the complete set of biological interactions that underlie migraine in its various forms . Knowledge of these interactions may guide development of novel therapeutic strategies . The same knowledge may also be translated toward the ultimate clinical goal of delivering the most individually targeted therapy in treating migraine . | Migraine is among the most common and debilitating neurological disorders . Diagnostic criteria for migraine recognize a variety of symptoms including a primary dichotomous classification for the presence or absence of aura , typically a visual disturbance phenomenon , as well as others such as sensitivity to light or sound , and nausea , etc . We explored whether any of 12 recently discovered genetic variants associated with common migraine might have selective association for migraine sub-classified by aura status or nine additional migraine features in a population of middle-aged women including 3 , 003 migraineurs and 18 , 180 non-migraineurs . Five of the 12 genetic variants met the most stringent significance criterion for association with migraine , among which four had selective association with sub-classified migraine , including one that was selective for migraine without aura . At suggestive significance , all of the remaining genetic variants were selective for sub-classifications of migraine although no two variants showed the same pattern of selectivity . The selectivity patterns suggest very different contributions to migraine pathophysiology among the 12 loci and their implicated genes . Further , the results suggest that future discovery efforts for new migraine susceptibility loci would benefit by considering associations with sub-classified migraine toward the ultimate goals of more specific diagnosis and personalized treatment . | [
"Abstract",
"Introduction",
"Results",
"Discussion"
]
| [
"neuroscience",
"neurobiology",
"of",
"disease",
"and",
"regeneration",
"medicine",
"and",
"health",
"sciences",
"signs",
"and",
"symptoms",
"pathology",
"and",
"laboratory",
"medicine",
"migraine",
"gene",
"regulatory",
"networks",
"genetic",
"association",
"studies",
"neurology",
"genetics",
"biology",
"and",
"life",
"sciences",
"human",
"genetics",
"genetics",
"of",
"disease",
"computational",
"biology",
"headaches",
"gene",
"function"
]
| 2014 | Selectivity in Genetic Association with Sub-classified Migraine in Women |
Five species of Ebola virus ( EBOV ) have been identified , with nucleotide differences of 30–45% between species . Four of these species have been shown to cause Ebola hemorrhagic fever ( EHF ) in humans and a fifth species ( Reston ebolavirus ) is capable of causing a similar disease in non-human primates . While examining potential serologic cross-reactivity between EBOV species is important for diagnostic assays as well as putative vaccines , the nature of cross-reactive antibodies following EBOV infection has not been thoroughly characterized . In order to examine cross-reactivity of human serologic responses to EBOV , we developed antigen preparations for all five EBOV species , and compared serologic responses by IgM capture and IgG enzyme-linked immunosorbent assay ( ELISA ) in groups of convalescent diagnostic sera from outbreaks in Kikwit , Democratic Republic of Congo ( n = 24 ) , Gulu , Uganda ( n = 20 ) , Bundibugyo , Uganda ( n = 33 ) , and the Philippines ( n = 18 ) , which represent outbreaks due to four different EBOV species . For groups of samples from Kikwit , Gulu , and Bundibugyo , some limited IgM cross-reactivity was noted between heterologous sera-antigen pairs , however , IgM responses were largely stronger against autologous antigen . In some instances IgG responses were higher to autologous antigen than heterologous antigen , however , in contrast to IgM responses , we observed strong cross-reactive IgG antibody responses to heterologous antigens among all sets of samples . Finally , we examined autologous IgM and IgG antibody levels , relative to time following EHF onset , and observed early peaking and declining IgM antibody levels ( by 80 days ) and early development and persistence of IgG antibodies among all samples , implying a consistent pattern of antibody kinetics , regardless of EBOV species . Our findings demonstrate limited cross-reactivity of IgM antibodies to EBOV , however , the stronger tendency for cross-reactive IgG antibody responses can largely circumvent limitations in the utility of heterologous antigen for diagnostic assays and may assist in the development of antibody-mediated vaccines to EBOV .
The genus Ebolavirus , family Filoviridae , has five identified ( including one proposed ) viral species [1] . Of these , four viral species , Zaire ebolavirus ( ZEBOV ) , Sudan ebolavirus ( SEBOV ) , Côte d'Ivoire ebolavirus ( CIEBOV ) , and Bundibugyo ebolavirus ( BEBOV ) are known to cause Ebola hemorrhagic fever ( EHF ) in humans , and in previous large outbreaks due to ZEBOV , SEBOV , and BEBOV , case fatality has ranged from 32 to 90% [2] , [3] , [4] , [5] , [6] , [7] , [8] , [9] . A fifth viral species , Reston ebolavirus ( REBOV ) , has been shown to cause severe disease in non-human primates [10] , [11] , [12] and can infect swine [13] . Similarly , evidence of human infections with REBOV have been documented serologically , however , no human disease has been associated with REBOV [13] , [14] , [15] . Despite the common characteristic of severe pathogenic potential in humans or non-human primates , genomic sequencing indicates relatively high divergence between Ebola viruses , with nucleotide differences ranging from 30–45% between species [16] . The role of antibody response in viral clearance and protective immunity against Ebola viruses in humans is not fully understood , however samples from individuals with acute ZEBOV infection have demonstrated antibodies titers that peak relatively early among those who survive , whereas low or absent antibody titers are commonly present in those with a fatal outcome [17] , [18] . Similarly , others have reported the presence of detectable anti-EBOV antibodies in humans during acute EHF ( in some instances with concurrent detectable viremia [16] , [19] , [20] ) , as well as in asymptomatic individuals shortly after exposure [21] , [22] , again suggesting that antibody response may be a correlate of protective immunity to EHF . EHF outbreaks commonly occur in remote locations , and often there is a significant lag between the occurrence of initial illnesses and subsequent diagnostic sample collection . As a result , diagnostic samples are frequently collected from individuals following clearance of viremia , only allowing serologic diagnosis of EHF . Adding to the challenge in EHF diagnosis , is the near geographic overlap of at least three pathogenic EBOV species ( ZEBOV , SEBOV , and BEBOV ) in central Africa [23] , [24] . While we previously have had success in the serologic diagnosis of EBOV infection using heterologous antigen ( for instance , BEBOV was initially identified by IgM reactivity to ZEBOV antigen [16] ) , the overall genetic divergence between EBOV species remains a concern , and previous data has suggested potential differences in serologic reactivity to different EBOV species in humans with EHF [25] , [26] . In order to examine the extent of serologic cross-reactivity of EBOV , as well as assess the utility of heterologous viral antigen for diagnosis of EBOV infection , we generated non-recombinant , infectious virus-based antigen preparations for the five known EBOV species , and examined the IgM and IgG responses against all five viruses in human sera collected from previous outbreak responses , associated with ZEBOV , SEBOV , BEBOV , and REBOV .
All samples were collected as part of public health diagnostic activities , were pre-existing relative to the start of the study , and were examined as anonymous samples . Ethical review of the study protocol was performed by the CDC Investigational Review Board and study approval was obtained following review , from the CDC Human Research Protection Office . Samples for this current study were previously collected as part of EHF outbreak responses , for 24 individuals infected with ZEBOV ( Kikwit , Democratic Republic of Congo , 1995 [5] ) , 20 individuals infected with SEBOV ( Gulu , Uganda , 2000 [19] ) , and 33 individuals infected with BEBOV ( Bundibugyo , Uganda , 2007 [9] ) ( table 1 ) . In addition , we assessed antibody responses in 18 samples that were collected from humans in the Philippines and sent to CDC for confirmatory testing , following the 2008 detection of REBOV in swine [13] . During diagnostic testing at CDC , the Philippines samples were found positive for REBOV-reactive IgG antibodies; the date of onset , or even previous occurrence of illness in individuals from whom these samples were obtained is unknown . While the time of sample collection , relative to disease onset differed between outbreaks ( with samples from Gulu tending to be from earlier stages post-infection than samples from Bundibugyo or Kikwit ) , all samples were from individuals who survived EBOV infection , and diagnostic testing at the time of outbreak response demonstrated the absence of viremia ( by PCR or antigen detection ELISA ) and the presence of IgG antibodies in each the samples included in this study . Each sample included in this study is from a discrete individual . Antigen preparations for IgM and IgG assays were performed as described previously [17] , [27] . Briefly , viral antigens for IgM and IgG ELISA were prepared by viral culture in Vero E6 cells , and harvested when at least 90% of cells had evidence of infection by immunofluorescence assay . Infected cells were processed by lysis of cells and supernatant for slurry antigen preparations ( IgM ) or by detergent basic buffer extraction of infected cells for lysate antigen preparations ( IgG ) , as described previously [17] , [27] . While the approach for antigen preparation does differ in terms of antigen concentration between IgM and IgG assays , the viral antigenic components are similar between both approaches . The decision to use these specific approaches is based on previously optimized protocols , which have been applied in numerous diagnostic settings . Viral antigen preparations were developed for each of the five known EBOV species , using viral isolates the following outbreaks: Kikwit , Democratic Republic of Congo , 1995 ( ZEBOV ) [5] , Gulu , Uganda , 2000 ( SEBOV ) [19] , Bundibugyo , Uganda , 2007 ( BEBOV ) [9] , the Philippines ( isolate from swine tissue sample submitted to USDA ) , 2008 ( REBOV ) [13] , and Tai Forest , Côte d'Ivoire , 1994 ( CIEBOV ) [6] . Mock-infected control antigens for IgM and IgG assays were prepared in similar manners , respectively , in the absence of virus . Western blots on viral antigen preparations were performed as described previously [28] , individually using hyperimmune mouse ascitic fluid ( HMAF ) poly-clonal antibodies [29] against ZEBOV , SEBOV , REBOV , and CIEBOV , and rabbit poly-clonal antibodies against ZEBOV , SEBOV , and REBOV , for detection of EBOV proteins; secondary antibodies were goat anti-mouse IgG horseradish peroxidase conjugate and goat anti-rabbit IgG horseradish peroxidase conjugate , respectively . IgM capture and IgG ELISAs were performed using Clinical Laboratory Improvement Amendments ( CLIA ) certified protocols that have been used for diagnostic testing of EBOV since 1990 [17] , [27] . For IgM assay , we used goat anti-human IgM antibody ( 1∶500 ) for antigen capture , slurry antigen preparations ( 1∶1000 ) , an HMAF poly-clonal antibody mixture , raised against ZEBOV , SEBOV , REBOV , and CIEBOV as detector antibody ( 1∶2000 ) , and anti-mouse IgG horseradish peroxidase conjugate ( 1∶8000 ) and ABTS substrate . For IgG assay , we used lysate antigen preparations ( 1∶1000 ) and mouse anti-human IgG horseradish peroxidase conjugate ( 1∶4000 ) and ABTS substrate . ELISAs were performed for samples , using both viral antigen and mock-infected antigen , at dilutions of 1∶100 , 1∶400 , 1∶1600 , and 1∶6400 . Adjusted optical density ( OD ) values represent the OD value ( at 410 nm ) of an individual sample dilution , after subtracting the OD value of mock infected antigen from the viral antigen for that dilution . The adjusted sum OD represents the sum of adjusted OD values of the four dilutions for an individual sample . For diagnostic assessment of antibody responses , individual sample dilutions with an adjusted OD of ≥0 . 1 ( IgM ELISA ) or ≥0 . 20 ( IgG ELISA ) were considered positive at that respective dilution , and an antibody response was considered positive for a sample if the sample had a positive titer of at least 1∶400 plus an adjusted sum OD of ≥0 . 45 ( IgM ELISA ) or ≥0 . 95 ( IgG ELISA ) . Cut-off values for the adjusted OD and adjusted sum OD for both assays correspond with diagnostic criteria currently used for EHF rule-out testing by CDC , and are based on previous evaluation of the distribution of values from thousands of negative serologic samples . For statistical comparison of adjusted sum OD values between autologous ( reaction to the same EBOV species that the individual was infected with ) and heterologous ( reaction to different EBOV species that the individual was infected with ) virus antigen preparations , we selected all samples from a single outbreak and performed Wilcoxon rank sum tests , for a non-parametric paired sample comparison of adjusted sum OD values . That is , for an individual outbreak , for each sample we calculated the difference in adjusted sum OD values between autologous antigen and a single heterologous antigen and tested whether the distribution of differences for all samples for that autologous-heterologous antigen pair was different from zero , by Wilcoxon rank sum test .
We produced non-recombinant infectious virus-based slurry ( for IgM ) and lysate ( for IgG ) antigen preparations for each of the five EBOV species . In order to confirm the presence of EBOV antigen in each viral lysate and slurry antigen preparations , we performed Western blots , using HMAF poly-clonal antibodies , raised against ZEBOV , SEBOV , REBOV , and CIEBOV ( figure 1A ) and rabbit poly-clonal antibodies , raised against ZEBOV , SEBOV , and REBOV ( figure 1B ) , as detector antibodies . Although neither detector antibody mixture contained antibodies specifically raised against BEBOV , the presence of reactive nucleoprotein ( NP ) bands near the 100 kilodalton weight marker , both in the lysate and slurry preparations , indicates the presence of viral antigen in both of the preparations . While only a faint NP band was detected in the SEBOV lysate preparation using the HMAF detector antibody mixture , the presence of antigen was apparent using the rabbit polyclonal antibody mixture . This may suggest an issue in reactivity of the HMAF antibody mixture against the SEBOV lysate preparation . However , we noted the presence of a strong NP band in the SEBOV slurry preparation , using the HMAF antibody mixture , indicating the utility of the HMAF as a detector antibody for the SEBOV IgM ELISA assay . Samples for this study are convalescent specimens collected as part of diagnostic activities for outbreaks due to ZEBOV ( Kikwit , Democratic Republic of Congo , 1995 [5] ) , SEBOV ( Gulu , Uganda , 2000 [19] ) , BEBOV ( Bundibugyo , Uganda , 2007 [9] ) , and REBOV ( Philippines [13] ) ( table 1 ) . We quantitatively examined the IgM antibody reactivity to autologous versus heterologous virus antigen by comparing adjusted sum OD values for each of the individual virus slurry antigen preparations , among outbreak samples . While many of the samples , particularly from Kikwit and Bundibugyo , had low IgM titers , overall adjusted sum OD IgM values tended to be higher to autologous than heterologous virus antigen preparations ( figure 2 ) . For instance , adjusted sum OD values for samples from the Kikwit outbreak were significantly higher against ZEBOV antigen , than against SEBOV , BEBOV , and REBOV . Similar trends are also apparent for samples from the Gulu and Bundibugyo outbreaks . Interestingly we note that samples from the Kikwit outbreak had significantly higher adjusted sum OD values against CIEBOV than against ZEBOV , and additionally samples from Bundibugyo had higher ( although not significantly different ) values against CIEBOV than BEBOV antigen . All samples from the Philippines were demonstrated to be IgM negative during diagnostic testing and thus adjusted sum OD values were not examined in this study . We additionally examined IgG antibody reactivity of autologous versus heterologous virus antigen by comparing adjusted sum OD values for each of the individual virus lysate antigen preparations among samples collected from each of the outbreaks . Owing to the convalescent stage at which most samples were collected , overall IgG adjusted sum OD values were mostly higher than IgM values ( figure 3 ) . Similar to trends observed for IgM responses , samples collected from Gulu and Bundibugyo outbreaks had significantly higher adjusted sum OD IgG values against autologous antigen than against heterologous antigen ( with the exception of samples from Bundibugyo having higher values against CIEBOV than against BEBOV ) . In contrast , adjusted sum OD values for samples from Kikwit did not differ between ZEBOV and SEBOV , BEBOV , or REBOV , and had higher values for CIEBOV in comparison to ZEBOV antigen . Interestingly , samples from the Philippines had higher adjusted sum OD values against ZEBOV , SEBOV , and CIEBOV , than against autologous REBOV antigen . We examined the kinetics of antibody development , for samples from Kikwit , Gulu , and Bundibugyo , by plotting the adjusted sum OD to autologous antigen for each of the sets of samples , relative to time post symptom onset ( figure 4 ) . The combined data for samples from these three outbreaks indicated early presence of IgM antibodies ( earliest samples for this study were at 14 days post symptom onset ) . While sample collection dates varied for the Kikwit , Gulu , and Bundibugyo samples , adjusted sum OD values peaked between 30–50 days , and largely declined by 80 days post symptom onset . As with IgM , IgG antibodies were present , even in most early samples , however , adjusted sum OD values remained high over the full course ( as long as 117 days ) of sample collection post-symptom onset . While the adjusted sum OD measure allowed us to quantitatively compare serologic cross-reactivity between autologous and heterologous antigens using a continous variable measure , we additionally wanted to examine the performance of heterologous antigen from a discrete ( positive or negative ) diagnostic standpoint . In order to assess the utility of heterologous antigen for the serologic diagnosis of EBOV infection by IgM ELISA , we selected all individuals with positive IgM antibody responses to respective autologous antigen from Kikwit , Gulu , and Bundibugyo outbreaks , and examined the sensitivity of the heterologous antigens for serologic diagnosis of EBOV in these samples ( table 2 ) . While the overall sensitivity of heterologous pairs varied widely , many heterologous virus combinations had low sensitivity for detection of positive IgM antibody responses . For instance , SEBOV , BEBOV , and REBOV antigen preparations had a sensitivity of less than 40% for all combinations of heterologous outbreak samples . We similarly examined the diagnostic utility of heterologous antigen for the serologic diagnosis of EBOV infection by IgG ELISA . In contrast to the above results for the IgM assay , heterologous antigens had a high sensitivity in the detection of IgG antibodies ( table 3 ) . With the exception of samples from Gulu , which displayed a diagnostic sensitivity of 74% with ZEBOV and REBOV antigen , all heterologous antigen pairs displayed at least 95% sensitivity for detection of IgG antibodies , and for many combinations , heterologous antigen detected positive results for 100% of samples .
The precise nature of antibody cross-reactivity between EBOV species has not been fully characterized . Some studies have reported detectable antibody reactivity to heterologous antigen in serum from humans or animals [25] , [27] , [30] , [31] , [32] , as well as noted potential differences in the cross-reactivity between autologous and heterologous antigen [27] , [32] , [33] . However , interpretation of these results remains difficult , owing the differences in antigen ( whole virus versus recombinant antigen ) and overall sample size , for many previous studies . In this study , among samples from the Kikwit , Gulu , and Bundibugyo outbreaks , we consistently observed higher adjusted sum OD values for IgM antibody responses against autologous antigen than against heterologous antigens . While IgM antibody responses were low for many samples ( in contrast to IgG responses ) , when we limited our analysis to those samples that were positive to autologous antigen on the basis of diagnostic IgM criteria , we observed low sensitivity of the IgM ELISA to heterologous antigen . Although some samples did react to heterologous antigen , our data indicate a species-specificity of IgM antibody responses in individuals infected with EBOV . In contrast to IgM antibody responses , IgG antibodies consistently displayed cross-reactivity to heterologous antigen , as demonstrated by the high adjusted sum OD values to heterologous antigens , as well as the high sensitivity of the IgG ELISA as a diagnostic assay . Previous serosurveys in Gabon , Central Africa Republic , and Democratic Republic of Congo have reported prevalence of anti-EBOV antibodies in rural populations ranging from ( 5–15% ) , which were presumed as indicative of previous infection with ZEBOV [34] , [35] , [36] . Owing to the high degree of IgG cross-reactivity we observed in this study , it is possible the relatively high seroprevalence of anti-EBOV antibodies reported in these studies may be the result of exposure to an unknown EBOV species , with lower pathogen potential than ZEBOV . The kinetics of antibody response to EBOV in humans has been best described for ZEBOV . Ksiazek et al reported early onset and peaking ( ∼18 days ) of IgM responses , which largely diminished by 60 days post-infection , while IgG antibodies were also present early post-onset and persisted for months following infection , in survivors [17] . Similar observations were reported by Baize et al . [18] and recent data from Wauquier et al . indicated that ZEBOV-reactive IgG antibodies persist for years following EHF [37] . While the samples examined in this study are not uniform with regard to time post-symptom onset , relative the EHF outbreak , our data do suggest the above observations can be extended for other EBOV species . An unexpected finding in this study was the overall high level of seroreactivity of heterologous samples to CIEBOV antigen . For instance , IgM adjusted sum OD values for samples from Kikwit , and IgG adjusted sum OD values from Kikwit , Bundibugyo , and the Philippines , were all significantly higher for heterologous CIEBOV antigen than for autologous antigen . The reason for this observation is unclear , however these are likely not the result of higher concentrations of CIEBOV antigen in lysate and slurry preparations , as demonstrated by the similar antigen concentration of CIEBOV antigen to the other antigen preparations in Western blot . It would be of interest to compare the cross-reactivity of human anti-CIEBOV sera , between autologous and heterologous EBOV antigen , however because of the scarcity of identified human infections ( only one patient diagnosed ) with to CIEBOV , we were unable to address this question . In addition , for samples from the Philippines , adjusted sum OD values for IgG tended to be higher against heterologous antigens than again REBOV antigen . We do not have an explanation for this observation , however , owing to the absence of detectable IgM responses in any samples from the Philippines , and the apparent lack of symptomatic disease in humans exposed to REBOV , these samples could represent later stage serologic responses in comparison to the other groups of samples , and may potentially include individuals with boosted immune responses due to multiple previous exposures to REBOV . Our observations indicate limitations in the utility of IgM ELISA , for diagnosis of EHF , prior to identification of the virus species . However , previous studies have reported early development of IgG antibodies in surviving EHF cases [17] , [18] ( and similarly supported by temporal data from this study ) . Because the IgG ELISA detected positive IgG antibody responses for the majority of samples with heterologous antigen , IgG ELISA assays in late-acute or early-convalescent samples may effectively circumvent the limitations in IgM ELISA , for diagnosis of EHF when the viral species is not known . We do note limitations of our study . The limited availability of diagnostic sera prohibited the opportunity to examine antibody cross-reactivity to specific EBOV proteins , or to specific epitopes . While the antigenic preparations used for ELISA in our study may be modestly enriched for NP , this approach does not preclude other proteins ( as demonstrated by Western blot in figure 1 ) and antigen preparations were used at high concentrations for the ELISAs [17] , [27] . For instance , in a recent study , Becquart et al . used the same IgG ELISA ( including ZEBOV antigen produced in the same manner ) as this current study to identify a large number of seropositive individuals , and confirmed EBOV-specific antibody responses in 138 individuals by Western blot . All individuals reacted to at least one viral protein , however , only 56% displayed antibody reactivity to NP [36] . Secondly , while we quantified IgM and IgG antibody levels , these do not necessarily represent the presence or quantity of neutralizing antibodies . The kinetics of development and functional role of neutralizing antibodies in viral clearance and protection in humans in not well understood . Currently most known neutralizing antibodies to EBOV target epitopes in the viral glycoprotein ( GP ) , and data suggest GP as an important protein for viral neutralization [38] , [39] , [40] , [41] , [42] . Interestingly , in studies involving humans with evidence of asymptomatic infection [22] and humans seropositive to EBOV [36] , the most common seroreactive proteins by Western blot were VP40 and NP; only a minority of individuals displayed evidence of reactive antibodies to GP . Although it is possible that antibody responses have a limited role in protective immunity to EBOV in humans , data from these studies ( living individuals with evidence of previous EBOV infection ) as well as from outbreak studies [17] , [18] , support the notion that antibody responses are an important correlate of immunity to EBOV in humans . In summary , we assessed the cross-reactive nature of IgM and IgG antibodies from groups of human survivors who were infected with four different species of EBOV . We observed cross-reactivity of IgG antibodies to heterologous antigen , however , overall reactivity to IgM and IgG antibodies tended to be stronger for autologous than heterologous antigen . Some experimental vaccines have suggested limited cross-protection of heterologous EBOV antigen [43] . Hensley et al . recently reported cross-protection against BEBOV infection in cynomolgus macaques vaccinated with DNA/rAd5 vaccine expressing GP of ZEBOV and SEBOV , although concluded that protection was the result of cellular immunity [33] . Our data suggest potential utility of heterologous vaccine for protection against EBOV , should IgG antibody responses prove to be an effective mediator of immunity to EBOV in humans . | Ebola virus ( EBOV ) is a highly pathogenic virus , capable of causing Ebola hemorrhagic fever in humans and non-human primates . Five species of EBOV have been identified . To examine whether infection with one EBOV species results in antibodies that cross-react with other EBOV species , we selected groups of human diagnostic samples from four outbreaks , which were each due to a different EBOV species , and compared IgM and IgG responses by ELISA to each of the five EBOV species . For samples from an individual outbreak , we found limited IgM reactivity to species of EBOV other than the virus species the individual was infected with . In contrast , for all groups of outbreak samples we observed strong cross-reactive IgG antibodies to all EBOV species . Our study demonstrates that IgG antibody responses tend to be more cross-reactive than IgM antibody responses in people infected with EBOV , a finding that has implications for the development of diagnostic assays and vaccines to EBOV . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"viral",
"hemorrhagic",
"fevers",
"infectious",
"diseases",
"neglected",
"tropical",
"diseases",
"viral",
"diseases",
"ebola",
"hemorrhagic",
"fever"
]
| 2011 | Serologic Cross-Reactivity of Human IgM and IgG Antibodies to Five Species of Ebola Virus |
HIV-1 particle production is driven by the Gag precursor protein Pr55Gag . Despite significant progress in defining both the viral and cellular determinants of HIV-1 assembly and release , the trafficking pathway used by Gag to reach its site of assembly in the infected cell remains to be elucidated . The Gag trafficking itinerary in primary monocyte-derived macrophages is especially poorly understood . To define the site of assembly and characterize the Gag trafficking pathway in this physiologically relevant cell type , we have made use of the biarsenical-tetracysteine system . A small tetracysteine tag was introduced near the C-terminus of the matrix domain of Gag . The insertion of the tag at this position did not interfere with Gag trafficking , virus assembly or release , particle infectivity , or the kinetics of virus replication . By using this in vivo detection system to visualize Gag trafficking in living macrophages , Gag was observed to accumulate both at the plasma membrane and in an apparently internal compartment that bears markers characteristic of late endosomes or multivesicular bodies . Significantly , the internal Gag rapidly translocated to the junction between the infected macrophages and uninfected T cells following macrophage/T-cell synapse formation . These data indicate that a population of Gag in infected macrophages remains sequestered internally and is presented to uninfected target cells at a virological synapse .
The human immunodeficiency virus type 1 ( HIV-1 ) Gag polyprotein precursor , Pr55Gag , plays an essential role in virus assembly and release . Its expression alone is able to generate virus-like particles ( VLPs ) [1] , [2] . All four domains of Pr55Gag–matrix ( MA ) , capsid ( CA ) , nucleocapsid ( NC ) and p6–play important roles in particle assembly and release [1] , [3] . The MA domain regulates the association of Gag with the host cell plasma membrane ( PM ) ; this membrane-binding activity is provided primarily by a myristic acid moiety covalently attached to the N-terminus of MA and a highly basic patch of amino acid residues that interacts with acidic phospholipids , including phosphatidylinositol- ( 4 , 5 ) -bisphosphate [PI4 , 5 ) P2] on the inner leaflet of the PM [4] , [5] , [6] , [7] . CA and NC promote Gag-Gag interactions during assembly [8] , in part through the ability of NC to interact with nucleic acid [2] , [9] . Finally , the p6 domain of Gag stimulates virus release by interacting with components of the cellular endosomal sorting machinery [10] , [11] , [12] . Although significant progress has been made in elucidating the viral and cellular factors necessary for Gag membrane binding , Gag multimerization , and virus release , the subcellular location of HIV-1 assembly has been the subject of controversy and the itinerary of Gag trafficking to the site of assembly remains to be defined . Mutational studies have shown that the viral determinants for Gag targeting to the PM reside in the MA domain of Gag . A large deletion in MA redirects HIV-1 assembly to the endoplasmic reticulum [13] , [14] , whereas point mutations , particularly in the highly basic domain of MA , shift the site of assembly from the PM to internal compartments [15] , [16] , [17] defined as late endosomes or multivesicular bodies ( MVBs ) [18] . HIV-1 was long assumed to follow the classically defined “C-type” pathway in which Gag assembly and release take place at the PM [2] . This dogma was challenged by a number of studies suggesting that HIV-1 assembly takes place in an endosomal compartment and that particle release from the infected cell follows the “exosomal” pathway in which virus-containing endosomes fuse with the PM to release their contents [19] , [20] , [21] , [22] , [23] , [24] . This endosomal model was then subsequently contested by several studies showing PM-based HIV-1 assembly and release [25] , [26] , [27] , [28] , [29] . The nature of the HIV-1 assembly site in primary monocyte-derived macrophages ( MDMs ) has been a matter of particular interest [30] . Early electron microscopy ( EM ) observations in HIV-1-infected MDMs revealed an abundance of virions assembling and budding into intracellular vacuoles [31] , [32] . In later studies , it was observed that the virus-containing internal compartments in MDMs bore markers characteristic of late endosomes or MVBs; e . g . , major histocompatibility complex II ( MHC II ) and tetraspanins CD63 , CD81 , and CD82 [18] , [33] , [34] . Furthermore , virions derived from MDMs packaged late endosome/MVB markers , suggesting that these virions originated from a late endosomal compartment [33] , [35] , [36] . In an intriguing refinement of the model that HIV-1 assembles in MVBs in primary macrophages , it was demonstrated that at least some of the virus-positive , “intracellular” structures in MDMs were actually connected to the PM . These apparently internal structures may therefore represent PM invaginations that are positive for tetraspanin markers [37] , [38] . Elucidating the virus assembly pathway in primary MDMs is highly significant since this cell type represents one of the major targets for HIV-1 infection in vivo [39] . One of the difficulties in evaluating previous studies focused on defining the Gag assembly/release pathway in MDMs is the absence of live-cell imaging data in this cell type that allow the trafficking of Gag to be visualized in real time . To this end , we developed a system for visualizing in living cells the localization and trafficking of Gag expressed in the context of a fully infectious and replication-competent HIV-1 molecular clone . We used the biarsenical-tetracysteine labeling method first described by Tsien and colleagues [40] , [41] , [42] . This system is based on the insertion of a small tetracysteine ( TC ) motif into a protein of interest . Cells expressing the TC-tagged protein are treated with a membrane-permeable biarsenical dye [e . g . , green ( FlAsH ) or red ( ReAsH ) ] that fluoresces upon binding to the TC tag . The advantages of this method are that the TC tag is very small and that labeling occurs immediately upon binding of the dye to the TC tag . Recently , this system was used to label Gag expressed from non-infectious clones in HeLa , Mel Juso and Jurkat T cells [23] , [29] . We introduced the TC tag near the C-terminus of the MA domain of Gag in the context of the full-length infectious HIV-1 molecular clone pNL4-3 . Insertion of the TC tag had no significant effect on HIV-1 Gag function . By using VSV-G-pseudotyped viruses , we were able to infect and follow Gag trafficking in primary MDMs . Our data indicate that in MDMs Gag accumulates both at the PM and in an apparently internal MVB-like compartment . Although we obtained no evidence for constitutive movement of the internal Gag to the PM , or for internalization of the PM-localized Gag to apparently internal structures , we observed rapid relocation of the internal population of Gag to the site of cell-cell contact following addition of susceptible T cells to the infected macrophage cultures . These findings support a model whereby newly assembled virus particles are sequestered in infected macrophages and then efficiently presented to susceptible target cells following synapse formation .
The MA domain of HIV-1 Gag performs several important functions in virus assembly and release [1]; however , deletion of a number of C-terminal residues ( amino acids 116-128 ) [43] , or the insertion of a Myc or green fluorescent protein ( GFP ) tag near the C-terminus of MA [44] does not block virus assembly and release , suggesting that the C-terminus of MA is relatively insensitive to mutation . Thus , to facilitate the study of HIV-1 Gag trafficking , we deleted codons 121–128 of MA and inserted a TC tag in the full-length molecular clone pNL4-3 to generate pNL4-3/MA-TC ( Fig . 1A ) . To determine the effects of the TC tag on virus replication , we transfected the Jurkat T-cell line with WT pNL4-3 or with pNL4-3/MA-TC and monitored virus replication over time by measuring the levels of reverse transcriptase ( RT ) activity in the medium ( Fig . 1B ) . We observed that replication of NL4-3/MA-TC was comparable to that of WT in Jurkat T-cells . To test the replication of MA-TC in primary MDMs , the MA-TC tag was introduced into the macrophage-tropic pNL4-3 derivative pNL ( AD8 ) [45] , [46] . Virus stocks were prepared and used to infect MDMs . As indicated in Fig . 1B , the NL ( AD8 ) /MA-TC virus replicated with kinetics indistinguishable from those of WT NL ( AD8 ) in this physiologically relevant primary cell type . The ability of MA-TC virus to replicate efficiently in both T-cell lines and primary MDMs suggested that the insertion of the TC tag near the C-terminus of MA does not affect HIV-1 assembly or release . To test this directly , we transfected HeLa cells with WT pNL4-3 or pNL4-3/MA-TC . One day posttransfection , the cells were labeled for 5 minutes with or without FlAsH , washed with ethanedithiol ( EDT ) , and metabolically labeled for 2–3 hrs with [35S]Met/Cys . Cell and virion lysates were prepared , immunoprecipitated with anti-HIV immunoglobulin ( HIV-Ig ) , subjected to SDS-PAGE , and bands were quantitated by phosphorimager analysis ( Figure 1C ) ( see Materials and Methods ) . The results indicated that insertion of the MA-TC tag had no significant effect on virus particle production and that the FlAsH dye caused no measurable disruption of HIV-1 particle production . We note that insertion of the TC tag in MA resulted in increased labeling of the MA protein with [35S]Met/Cys due to the additional Cys residues ( Fig . 1C ) . We also compared the single-cycle infectivity of WT and MA-TC Gag in the TZM-bl indicator cell line [47] and observed no effect of the MA-TC tag on virus infectivity ( data not shown ) . Together , these data demonstrate that the MA-TC tag does not disrupt normal HIV-1 Gag function . We previously reported a number of mutations within the MA domain of Gag that alter normal HIV-1 Gag trafficking and localization [15] , [16] , [18] . For example , mutation of the site of Gag myristylation ( 1GA; [15] ) results in a diffuse cytosolic Gag localization . Mutations in the MA highly basic domain ( e . g . , 29KE/31KE ) retarget Gag to MVBs [16] , [18] . To validate further the TC labeling approach , we sought to confirm that the effect of these mutations on Gag localization in the context of otherwise WT Gag would be recapitulated in the context of MA-TC Gag . We introduced the 1GA and 29KE/31KE MA mutations into MA-TC and analyzed Gag localization within cells using a rapid FlAsH labeling method . Transfected HeLa cells were labeled for 5 min with FlAsH and washed for 20 min in EDT . Cells were then fixed and either mounted or processed further for antibody labeling . Similar to our previous results obtained with antibody labeling [16] , [48] , MA-TC Gag was found primarily in a punctate pattern at the cell surface . In contrast , MA-TC/1GA was diffusely localized throughout the cytosol and MA-TC/29KE/31KE was found in internal compartments ( Fig . 2 , top ) . We previously observed that the internal compartment to which 29KE/31KE localizes in HeLa cells is positive for the MVB marker CD63 [18] . To verify that this was also the case in the context of MA-TC Gag , we examined the colocalization of the 29KE/31KE-TC mutant , labeled with ReAsH , with CD63 in transfected HeLa cells . We observed nearly complete colocalization between 29KE/31KE Gag and CD63 ( Fig . S1 ) . These results confirm the biochemical experiments indicating that the addition of the TC tag had no effect on Gag trafficking , assembly , or release in HeLa cells . Since our goal was to visualize Gag trafficking in physiologically relevant primary cells , we analyzed the localization pattern of MA-TC Gag in infected MDMs . Cells were infected with VSV-G-pseudotyped virus stocks obtained from transfected 293T cells . Infected MDMs were then labeled with FlAsH and fixed 24 to 72 hours post-infection . Infection efficiencies , as determined by Gag staining , typically ranged between 2 and 10% . MA-TC Gag localized both to the PM and to an apparently internal compartment ( Fig . 2 , bottom ) . In agreement with our previous results obtained by antibody labeling [16] , [18] , [49] , the localization of MA-TC-derived 1GA and 29KE/31KE Gag in MDMs was similar to that observed in HeLa cells: 1GA-TC was diffusely distributed throughout the cytoplasm , and 29KE/31KE-TC was almost exclusively found in apparently internal compartments ( Fig . 2 , bottom ) . To provide a clearer visualization of the internal localization of 29KE/31KE-TC Gag in MDM , we obtained a z-series reconstruction by using the Maximum Intensity Projection mode from the image processing software OsiriX ( Video S1 ) . The results presented in Fig . 2 demonstrate that introduction of the TC tag near the C-terminus of MA ( MA-TC ) allows HIV-1 Gag to be readily visualized in infected primary MDMs at early time points postinfection . We and others have previously reported that the apparently internal vesicles to which HIV-1 Gag localizes in MDMs bear tetraspanin markers , suggesting that they are MVBs or MVB-like structures [18] , [33] , [34] . To define the site of Gag localization in MDMs at early time points postinfection using TC-tagged Gag , we infected MDMs and examined the localization of Gag and tetraspanins ( CD63 and CD81 ) at 20 hrs postinfection . As previously observed with fully WT Gag [18] , MA-TC Gag displayed a localization pattern that partially overlapped with that of CD63 ( Fig . 3A ) . The colocalization pattern in these cells was very heterogeneous , with some cells displaying a high degree of Gag/CD63 colocalization ( Fig . 3A , top panels ) and other cells showing a lower level of colocalization ( Fig . 3A , lower panels ) . 29KE/31KE-TC Gag also overlapped with a subset of CD63 in infected MDMs ( Fig . 3B ) . Both MA-TC ( Fig . 3C ) and 29KE/31KE-TC Gag ( data not shown ) showed much more extensive colocalization with CD81 than with CD63 . We note that some cells displayed a high level of Gag and CD81 costaining at the PM ( Fig . 3C , lower panel ) , consistent with HIV-1 assembly occurring in tetraspanin-enriched microdomains at the cell surface [19] , [50] . To quantitatively compare the degree of Gag/CD63 vs . Gag/CD81 colocalization in MDMs , we measured the Pearson correlation coefficient ( R ) values ( see Materials and Methods ) for these two sets of colocalizing proteins in a total of 75 cells . The results confirmed the higher degree of Gag/CD81 compared to Gag/CD63 colocalization ( Fig . S2 ) . We observed that 71% of cells displayed a Gag/CD63 R-value of <0 . 6 , whereas 91% of cells showed a Gag/CD81 R-value of >0 . 6 ( Fig . S2 ) . As indicated in Fig . 3 , WT Gag was localized both to an internal tetraspanin-positive compartment and to the PM in infected MDMs . Very few cells showed exclusively PM staining; instead , the vast majority of cells showed either an internal localization or both PM and internal staining . To determine whether the distribution changed over time , we classified cells as displaying uniquely PM , intracellular , or both PM and intracellular Gag localization at 20 , 24 , 48 , 72 , and 96 hrs postinfection . The percentage of cells within these three categories remained essentially unchanged over time ( Fig . 4A ) . To visualize Gag movement in living MDMs , cells were infected with MA-TC virions pseudotyped with VSV-G and were labeled for 5 min with FlAsH 24 to 72 hrs post-infection . After washing , labeled MDMs were immediately placed in a microscope chamber ( 37°C/5% CO2 ) and imaged over time . Interestingly , no clear movement of Gag between PM and apparently intracellular compartments was observed during the time course ( Fig . 4B ) ; i . e . , no obvious internalization of Gag from the PM was visualized , nor was there clear movement of internal Gag puncta to the PM . These results suggest that Gag can assemble both at the PM and in internal compartments in infected MDMs . As stated in the Materials and Methods , prior to 20 hrs post-infection we were not able to definitively distinguish between specific Gag staining and the diffuse , low-level background . During the course of our analyses , we frequently observed concentrated Gag staining at the contact sites formed between infected and uninfected MDMs ( Fig . 5A ) . 3D z-stack reconstructions illustrating this phenomenon are presented in Figs . S3A and Video S2 ) . These Gag-enriched cell-cell junctions also displayed a high degree of staining for the tetraspanin markers CD81 and CD82 ( Fig . 5B and data not shown ) . Analogous junctions have been reported to form between HIV-1-treated dendritic cells and T-cells; because these junctions bear markers ( e . g . , tetraspanins and adhesion molecules ) found at immunological synapses [51] , [52] , [53] they have been named “infectious” or “virological” synapses [54] , [55] , [56] , [57] , [58] . A concentration of budding and released virions was also observed in the vicinity of cell-cell contact sites by transmission electron microscopy ( EM ) ( Fig . 5C ) . To quantify the localization of Gag at the synapse observed in our EM analysis , we counted the number of virus particles and budding structures at synapse vs . non-synapse regions of the plasma membrane . More than 60 cells were scored for this analysis . The results indicated a markedly ( 5-6-fold ) higher density of particles and budding events at synapse vs . non-synapse regions of the cell surface , consistent with the immunofluorescence data presented above . To extend the analysis of Gag concentration at the cell-cell synapse to include junctions formed between infected MDMs and uninfected T-cells , we performed the following analysis: infected MDMs were labeled with FlAsH and then incubated at 37°C for 2 hours with Jurkat T-cells . The cells were then fixed and , when necessary , labeled with anti-CD81 antibodies . Gag was frequently detected at the synapses between infected MDMs and uninfected Jurkat T-cells ( Fig . 6A ) . 3D z-stack reconstructions are provided in Videos S3 and S4 . Furthermore , as we observed for MDM/MDM junctions , MDM/T-cell synapses also displayed a high degree of colocalization between Gag and tetraspanin markers ( Fig . 6B ) . We also observed that Gag concentrated at synapses formed between infected MDMs and primary T cells ( data not shown ) . Overall , these data show that HIV-1 Gag , along with CD81 , are recruited to the synapses formed between infected macrophages and uninfected macrophages or T-cells . To quantify the concentration of Gag at the synapse , we used the ImageJ software to determine the pixel intensity for Gag staining at the MDM/MDM and MDM/T-cell synapses compared to the overall pixel intensity in each infected cell . The results confirmed a high degree of Gag concentration at cell-cell junctions , with approximately 80% of the total Gag signal localized to the synapse ( Figs . 6C , S3A ) . To analyze further the process of Gag recruitment to the synapse in infected MDMs , we determined whether Gag was recruited to cell-cell junctions in the context of proviral clones carrying additional mutations . We first examined the localization of Gag in the absence of Env expression by using the Env ( - ) MA-TC mutant KFS/MA-TC . Examining a possible role for Env in Gag recruitment to the MDM synapse was of interest as it has been reported that Env is required for synapse formation between infected and uninfected T cells [59] and also plays a role in the formation of filopodial bridges that can facilitate transfer of retroviruses between cells [60] . In contrast to these prior findings in non-monocytic cell types , we observed that Gag was efficiently localized to both MDM/MDM and MDM/T-cell synapses in the absence of Env expression ( Fig . 7A and B ) . This concentration of Gag to the synapse was quantified as described above , confirming the high degree of localization of Gag to the cell-cell junction independent of Env expression ( Fig . 7C , S3B ) . 3D z-stack reconstructions are provided in Video S5 . The data indicated no statistically significant difference between Gag concentration at the MDM/T-cell vs , MDM/MDM synapse , or in the presence or absence of Env expression ( compare Figs . 6C and 7C ) . We previously reported that mutations in the highly basic domain of MA ( e . g . , 29KE/31KE ) redirect Gag to MVBs [18] . Here , we observed that in MDMs the 29KE/31KE-TC mutant displayed nearly complete localization to an apparently internal compartment that stained positive for CD63 and CD81 ( Figs . 2 and 3 and data not shown ) . In contrast , MA-TC Gag displayed a mix of PM and internal staining ( e . g . , Fig . 2–4 ) . It was therefore of interest to examine whether 29KE/31KE-TC Gag could redistribute from its normally internal site of localization to the cell surface upon synapse formation . Interestingly , we observed that in contrast to MA-TC Gag , 29KE/31KE-TC Gag did not relocalize to either MDM/MDM ( Fig . 8; Video S6 ) or MDM-T-cell ( data not shown ) synapse . Instead , in cells expressing 29KE/31KE-TC Gag , both Gag and CD81 remained deep within the infected cell ( Fig . 8 ) . In four independent experiments with 29KE/31KE-TC , Gag accumulation was never observed at the synapse . These data suggest the possibility that the apparently internal compartments to which WT and 29KE/31KE Gag localize are distinct . The data presented above using fixed infected cells and EM techniques demonstrate the accumulation of Gag and virus particles at the junction between infected MDMs and uninfected MDMs or T-cells . To visualize the movement of Gag to the cell-cell contact site , we used FlAsH labeling and live-cell imaging in infected MDMs . For these experiments , infected MDMs were labeled with FlAsH for 5 minutes 24 to 72 hours post-infection , washed , and imaged over time . When visualizing MDM/T-cell junctions , Jurkat T-cells were added to the infected cells post-FlAsH labeling and imaged under the same conditions . After addition of the Jurkat cells , incubation periods of approximately 30–60 min were required for stable MDM/T-cell synapses to form . As mentioned above , we observed no clear evidence of movement of apparently internal Gag to the PM , or vice versa , in MDMs not actively engaged in cell-cell contact . Interestingly , however , upon addition of Jurkat T-cells to the infected MDM cultures , we observed rapid movement of apparently internal Gag to the MDM/T-cell synapse . The infected macrophage ( “M1” ) in Fig . 9A is surrounded by uninfected macrophages ( e . g . , “M2” ) and Jurkat T-cells ( “T1” and “T2” ) . In this particular time course , 40 min after adding Jurkat T cells to the MDMs ( t = 0 min ) , Gag has already accumulated at the contact site between M1 and T1 . Gag-containing compartments are also rapidly recruited to the site of M1/T2 contact . Movement of other Gag-containing compartments toward the site of MDM-MDM ( M1/M2 ) contact can be observed starting at time t = 25 min and is complete at t = 40 min . These data demonstrate that Gag present in internal compartments can be rapidly redistributed to the site of contact with uninfected cells . After its movement to the MDM/MDM synapse , Gag can be seen moving along the surface of macrophage M2 ( e . g . , at 45 and 50 min ) . A movie of Gag movement to the synapse can be viewed at Video S7 . Intriguingly , we frequently observed an apparent preference for MDM/T-cell synapse formation at sites close to high levels of Gag concentration ( Fig . 9B ) . In this gallery , time t = 0 represents cells 90 min post-FlAsH labeling and 70 min after addition of T-cells . One of the surrounding T-cells ( “T “ ) at time t = 0 min extends on top of the infected MDM toward the site of Gag accumulation . This resulted in the movement and attachment of the T-cell with the infected MDM near the site of Gag accumulation .
Most studies that have examined HIV-1 Gag trafficking have used non-infectious constructs in which codon-optimized Gag is fused to fluorescent proteins such as green or red fluorescent protein ( GFP or RFP ) . Although these studies provided important insights , disadvantages of using GFP and its derivatives in protein trafficking analyses include the large size of the fluorescent protein and the fact that achieving their fluorescent state requires time-dependent chromophore maturation [61] , [62] . We have also observed that Gag expressed from some codon-optimized constructs assembles relatively inefficiently and forms perinuclear cytosolic aggregates not typically observed with WT Gag ( unpublished results ) . In this study , we describe the application of the biarsenical labeling system to visualize HIV-1 Gag trafficking in primary MDMs . We show here that the TC tag that serves as the binding site for the biarsenical dye FlAsH is remarkably well tolerated with respect to preserving Gag function when introduced near the C-terminus of the MA domain . The MA-TC tagged Gag produces virus particles with WT efficiency , and these particles are fully infectious in both single-cycle assays and in spreading infections . The MA-TC Gag can be readily delivered to primary cells as a VSV-G pseudotype . Overall , this system provides a rapid and efficient method for observing WT Gag trafficking in living cells . The biarsenical labeling system employed in this study allowed us to examine Gag localization after ∼20 hrs postinfection , the earliest time point at which Gag expression could be readily and consistently visualized . At this early time point , we observed a mix of PM and apparently internal Gag staining . Between 20 and 96 hrs postinfection , we did not observe a shift in the percentage of cells displaying PM , intracellular , or PM+intracellular staining ( Fig . 4 ) nor did we observe a time-dependent accumulation of internal Gag . These results differ from those of a recent study in which intracellular Gag-GFP staining increased over time [27] . We note that the TC-tagged Gag in the current study is fully functional for particle assembly and release and produces infectious virions . Furthermore , our TC-tagged Gag is expressed in the context of a full-length molecular clone that encodes all the HIV-1 accessory proteins including Vpu . Indeed , elimination of Vpu expression led to a time-dependent accumulation of Gag in internal compartments ( unpublished results ) , consistent with recent reports [26] , [28] . In agreement with the study of Jouvenet et al . [27] , we did not observe an effect on virus release of treating infected cells with U18666A , a drug that arrests endosome motility ( unpublished results ) . This observation supports the hypothesis that release of HIV-1 in MDMs occurs from PM-assembled VLPs . Although the FlAsH method can be accompanied by high background staining , we observed that this problem is largely mitigated by using very brief labeling periods . We also observed that background staining in MDM is less evident in MDM than in HeLa cells . The most significant finding in this study is the visualization of apparently internal Gag moving to the site of cell-cell contact after synapse formation with uninfected T cells . It therefore appears that the tetraspanin-rich , apparently internal compartment in which HIV-1 assembles in MDMs can serve as a storage compartment for rapid presentation of virus particles at cell-cell junctions . These findings have clear implications for HIV-1 transmission from MDMs . In this regard , it is interesting to note that the virions in these internal vesicles reportedly remain infectious for weeks postinfection [63] and that virus transmission between infected MDMs and T cells is extremely rapid [64] . In several respects , our observations with infected MDMs are similar to those made with dendritic cells treated with HIV-1 . Binding of HIV-1 virions to dendritic cells can lead to transfer of virus to uninfected T-cells through the formation of a virological or infectious dendritic cell/T-cell synapse without the dendritic cell itself being productively infected [55] , [56] , [57] , [58] , [65] , [66] . Virions bound to the dendritic cell are reportedly internalized into an internal compartment that is strongly positive for CD81 but only weakly positive for CD63 . Synapse formation induces the redistribution of virus particles and CD81 to the site of cell-cell contact , presumably facilitating transfer of virus to the T cell . The internal virus-containing compartment in dendritic cells is weakly acidic , as also reported for the virus-positive compartment in MDMs [67] . Thus , it appears that HIV-1 has evolved to subvert a pathway in both MDMs and dendritic cells that allows infectious virus particles to be retained in an apparently internal compartment and then redistributed to the cell surface following infectious synapse formation . Transfer of HIV-1 between T cells also involves the formation of a synapse that bears tetraspanin markers [68]; however , there is currently no evidence for long-term retention of infectious virus particles within an internal compartment in T cells . Interestingly , whereas the generation of T-cell/T-cell infectious synapses [59] and formation of cell-cell filopodial bridges that allow intercellular transfer of virus particles [60] have been reported to require Env expression , we found no such Env dependence for Gag translocation to the synapse formed between MDMs and T cells ( Fig . 7A ) . We previously reported that mutations in the highly basic domain of MA ( e . g . 29KE/31KE ) induce a shift in Gag localization in HeLa cells and T cells from PM to MVBs [18] . In MDMs , both WT and 29KE/31KE Gag localize to an apparently internal , tetraspanin-positive compartment [18] . This finding is confirmed here ( e . g . , Fig . 2 and 8 ) . Interestingly , while WT MA-TC Gag rapidly translocates to the MDM/T-cell junction after synapse formation , we did not observe significant movement of 29KE/31KE-TC Gag to the MDM/T-cell synapse ( Fig . 8 ) . These results imply that WT and 29KE/31KE Gag localize to distinct tetraspanin-positive compartments in MDMs . A possible interpretation of these observations is that 29KE/31KE Gag localizes to “true” MVBs which do not move to the synapse , whereas WT Gag localizes to a compartment that is apparently internal but is connected to the PM [37] , [38] . It is this PM-connected tetraspanin-positive compartment that undergoes a shift in localization following synapse formation , thereby allowing virus particle movement to the site of cell-cell contact . Surprisingly , we frequently observed that T-cells made contact with regions of the MDM PM under which Gag was concentrated , and in many cases the T-cells formed pseudopodia to contact this site ( e . g . , Fig . 9B ) . These results imply that the T cell can “sense” regions of the PM that overlie the putative invaginations in which assembled virus particles are concentrated . These regions of the PM may be enriched in lipid rafts and/or tetraspanin-enriched microdomains . These observations are somewhat reminiscent of previous studies on the recruitment of uninfected T cells into infected cell syncytia [69] . A future challenge will be to characterize in greater detail the membrane composition at the site of MDM/T-cell contact and elucidate the signals that induce the movement of the newly assembled , internally sequestered virus particles to the infectious synapse .
Plasmids pNL4-3/MA-TC and pNL ( AD8 ) /MA-TC were constructed as follows: for pNL4-3/MA-TC , nucleotides 1250–1273 ( encoding DTGNNSQV Gag codons 121–128 ) were deleted in the MA-coding region of the full-length HIV-1 molecular clone pNL4-3 [70] and the TC tag GSMPCCPGCCGSM was inserted in its place using overlap-extension PCR [71] . The MDM-tropic pNL ( AD8 ) /MA-TC clone was constructed by exchanging the EcoRI-XhoI fragment of pNL4-3/MA-TC with that from the CCR5-tropic clone pNL ( AD8 ) [45] . Construction of molecular clones expressing pNL4-3 MA mutants 1GA and 29KE/31KE was described previously [15] , [16] . The molecular clones pNL4-3/1GA-TC and pNL4-3/29KE/31/KE-TC were constructed by exchanging the BssHII-SphI fragments of pNL4-3/1GA or pNL4-3/29KE/31KE with the corresponding fragments from MA-TC . To construct the Env ( - ) pNL4-3 construct , pNL4-3/KFS-TC , we exchanged the EcoRI-XhoI fragment from the Env ( - ) molecular clone pNL4-3/KFS [72] with the corresponding fragment from pNL4-3/MA-TC . Finally , we constructed pNL4-3/Vpu ( - ) /MA-TC by replacing the BssHII-EcoRI fragment from Vpu-DEL-1 [73] ( kindly provided by K . Strebel ) , with the corresponding fragment from pNL4-3/MA-TC . VSV-G-pseudotyped virus stocks were prepared by transfecting 293T cells with the Gag/Pol expression vector pCMVNLGagPolRRE [74] , the VSV-G expression vector pHCMV-G [75] , and the indicated HIV-1 molecular clones by using Lipofectamine 2000 ( Invitrogen ) , according to the manufacturer's protocol . HeLa and Jurkat T cells were cultured as previously described [15] . MDMs were prepared by culturing elutriated monocytes [45] in RPMI-1640 medium , supplemented with 10% fetal bovine serum , for 5 to 7 days on ultra-low attachment plates ( Costar ) . HeLa cells were transfected by using the calcium phosphate method , as previously described [15] . Jurkat T-cells were transfected by using the DEAE-dextran procedure as previously reported [15] . Infection of MDMs was performed as follows: MDMs were detached from the ultra-low attachment plates ( Fisher Scientific , Pittsburgh , PA ) and plated onto tissue culture dishes or microscope culture chambers Fisher Scientific , Pittsburgh , PA ) . Virus stocks , pseudotyped with VSV-G , were incubated with MDMs for 5-6 hours . 2×106 counts/minute ( cpm ) of reverse transcriptase ( RT ) activity was used per well of 4-well Nunc chambers , 106 RT cpm/well for 8-well Nunc-chambers , and 4×106 RT cpm/well for 6-well plates . Virus replication assays in the Jurkat T-cell line were performed as previously described [15] . Briefly , Jurkat cells were transfected in parallel with WT pNL4-3 or MA-TC using the DEAE-dextran method . Cells were split 1∶3 every two days and an aliquot of medium was reserved at each time point for RT assay [76] . MDMs in 6-well plates were infected with 2×106 RT cpm/well with WT pNL ( AD8 ) or MA-TC ( AD8 ) virus stocks . Medium in the infected MDM cultures was changed every two days and an aliquot was reserved for RT activity . For single-cycle infectivity assays , 4×105 HeLa-derived TZM-bl cells [47] ( obtained from J . Kappes through the NIH AIDS Research and References Reagent Program ) per well were infected with 2×105 RT cpm virus stocks . Infection efficiency was determined by measuring luciferase activity 2 days post-infection , as described previously [77] . Adherent cells cultured in Lab-Tek chamber slides ( Nunc ) or 6-well plates were labeled 24–72 hours posttransfection/infection . All labeling steps were performed at 37°C in the dark . The cells were washed twice with Opti-MEM I ( Invitrogen , Carlsbad ) . For each experiment , biarsenical labeling solutions were freshly prepared immediately prior to use . Wash solutions of 300 µM and 100 µM 1 , 2-ethaneditiol ( EDT ) ( Aldrich Chemical Company , Inc . , Milwaukee ) were prepared in phosphate-buffered saline ( PBS ) and 0 . 2 mM Lumio Green ( FlAsH ) was prepared in dimethyl sulfoxide ( DMSO ) ( Sigma-Aldrich , Inc . , St Louis ) . Before labeling , 2 µl of 1 mM EDT was mixed quickly with 4 . 7 µl of 0 . 2 mM FlAsH or Lumio Red ( ReAsH ) and immediately added to 400 µl Opti-MEM I . This solution was added to cells , which were incubated for 5 min at 37°C . After the 5 min biarsenical labeling , cells were washed with 300 µM EDT/PBS for 8 min and 100 µM EDT/PBS for 10 min at 37°C . Cells were then washed further 3X with PBS and either fixed with 3 . 7% formaldehyde prior to antibody labeling or incubated with Opti-MEM I for live cell imaging or addition of Jurkat T cells . The levels of background in MDMs and in Hela cells were greatly reduced with the addition of EDT in our labeling solutions . We also observed that keeping the biarsenical labeling time short ( 5 min ) was enough to obtain a strong Gag signal , while limiting non-specific background staining . As early as 20 hrs post-infection , specific Gag staining could be readily detected . However , at earlier time points , the cytosolic Gag signals were too low to be clearly distinguishable from background fluorescence , and therefore no data were acquired before 20 h postinfection . For fluorescence microscopy , 24–48 hours post transfection/infection , cells were labeled using the biarsenical method and either fixed using 3 . 7% formaldehyde/PBS for 20 min or Jurkat T-cells were added to the MDMs ( in Opti-MEM I ) for 2 hours , then fixed with formaldehyde . The cells were then permeabilized with 0 . 1% Triton-X100/PBS and incubated with 0 . 1 M glycine for 10 min at room temperature to quench free aldehyde groups . The cells were then blocked with 3% bovine serum albumin ( BSA ) /PBS , incubated with either mouse monoclonal anti-CD63 ( Santa Cruz Biotechnology ) or mouse monoclonal anti-CD81 ( BD Pharmingen ) for 1 hr at room temperature , washed and incubated with Alexa-594 or 488-conjugated anti-mouse IgG ( Invitrogen ) for 30 min at room temperature . The cells were then washed and mounted with Aqua Poly Mount ( Polysciences Inc . , Warrington , PA ) . For live cell imaging , the labeled cells were imaged in a temperature-controlled chamber ( 37°C/ 5% CO2 ) in Opti-MEM I . For both fixed and live-cell microscopy , the cells were imaged using an Olympus 1X-71 inverted deconvolution microscope and analyzed with Delta Vision software ( Applied Precision Inc . , Seattle , WA ) . To quantify the degree of relative colocalization , we obtained the Pearson correlation coefficient ( R ) values , which are standard measures of colocalization [78] . The R values were calculated using the softWORx colocalization module which generates a “colocalized” image from two channels . A scatter plot of the two intensities on a pixel-by-pixel basis is then plotted and the R value is calculated by dividing the covariances of each channel by the product of their standard deviations . For EM , infected cells were fixed and processed as previously described [15] . Metabolic radiolabeling , preparation of cell and viral lysates , and immunoprecipitation assays were performed as previously described [15] . Briefly , transfected HeLa cells , or infected MDMs , labeled with the biarsenical dyes or DMSO ( control ) were metabolically labeled with [35S] Met/Cys for 2 hours , 24-48 hours posttransfection/infection , and released virions were pelleted by ultracentrifugation . Cell and virus lysates were immunoprecipitated with HIV immunoglobulin ( HIV-Ig ) , obtained from NABI and the National Heart Blood and Lung Institute through the NIH AIDS Research and Reference Reagent Program . Immunoprecipitates were subjected to SDS-PAGE followed by fluorography . Quantitative analysis of the bands visualized by radioimmunoprecipitation was performed using a Bio-rad phosphorimager . | The viral Gag protein is both necessary and sufficient for the assembly of a new generation of virus particles . There has been a significant amount of debate in recent years regarding the site in the cell at which HIV-1 assembly takes place . Of particular interest has been the site of assembly in macrophages , a cell type that serves as an important target for HIV-1 infection in vivo . In this study , we examine the site of Gag localization and virus assembly in primary human macrophages in living cells by using biarsenical dyes that become fluorescent when they bind a small target sequence introduced into HIV-1 Gag . We observe Gag localization both at the plasma membrane and in an apparently internal compartment that bears markers characteristic of multivesicular bodies ( MVBs ) . Significantly , when infected macrophages are cocultured with uninfected T cells , the apparently internal Gag moves rapidly to the contact site , or synapse , between the macrophage and the T cell . These findings support the hypothesis that infected macrophages sequester assembled HIV-1 particles in an internal compartment and that these particles move to synapses where cell–cell transmission can occur . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"virology/immunodeficiency",
"viruses"
]
| 2008 | Real-Time Visualization of HIV-1 GAG Trafficking in Infected Macrophages |
For cells to function , the concentrations of all proteins in the cell must be maintained at the proper levels ( proteostasis ) . This task – complicated by cellular stresses , protein misfolding , aggregation , and degradation – is performed by a collection of chaperones that alter the configurational landscape of a given client protein through the formation of protein-chaperone complexes . The set of all such complexes and the transitions between them form the proteostasis network . Recently , a computational model was introduced ( FoldEco ) that synthesizes experimental data into a system-wide description of the proteostasis network of E . coli . This model describes the concentrations over time of all the species in the system , which include different conformations of the client protein , as well as protein-chaperone complexes . We apply to this model a recently developed analysis tool to calculate mediation probabilities in complex networks . This allows us to determine the probability that a given chaperone system is used to mediate transitions between client protein conformations , such as folding , or the correction of misfolded conformations . We determine how these probabilities change both across different proteins , as well as with system parameters , such as the synthesis rate , and in each case reveal in detail which factors control the usage of one chaperone system over another . We find that the different chaperone systems do not operate orthogonally and can compensate for each other when one system is disabled or overworked , and that this can complicate the analysis of “knockout” experiments , where the concentration of native protein is compared both with and without the presence of a given chaperone system . This study also gives a general recipe for conducting a transition-path–based analysis on a network of coupled chemical reactions , which can be useful in other types of networks as well .
Protein homeostasis ( proteostasis ) is essential for the viability of an organism . The disruption of protein homeostasis involving the misfolding and subsequent aggregation of proteins is implicated in many diseases , including Down's syndrome , type-II diabetes , Alzheimer's , Parkinson's and Huntington's disease [1]–[4] . In addition , inherited mutations that lead to excessive degradation of proteins can lead to loss-of-function diseases , such as cystic fibrosis and Gaucher disease [2] , [5] . Thus , in every living cell , a system of chaperones – called the proteostasis network – has evolved to help proteins fold , correct or clear misfolded protein , and prevent ( or even reverse ) the formation of protein aggregates . Proteostasis networks can be broken down into chaperone subsystems ( such as the Hsp60 , Hsp70 and Hsp90 systems in eukaryotes ) [4] , and these systems can be studied individually . Much work has focused on cataloguing the proteins that are clients of these different chaperone systems , and examining their structural features [6]–[9] . The molecular mechanisms of interaction of chaperones with client proteins in each system has been studied [10]–[15] . Data from experiment has been synthesized into theoretical models , which describe the passage of client proteins through a given chaperone system [16]–[18] . However , different chaperone systems do not operate in isolation in vivo . Most chaperone activity relies on the consumption of ATP , which is derived from a shared source . Also , chaperones can have more than one function; DnaK in prokaryotes can bind both to unfolded or misfolded protein in order to prevent aggregation , or it can help prepare aggregates for binding to another chaperone , ClpB , for disaggregation . Complicating matters is that proteins are not selective in the chaperone system to which they bind: it has been shown that there is significant overlap in the sets of client proteins whose solubilities increase under the action of different chaperone groups [19] . Furthermore , experimental studies have shown the synergistic action of multiple chaperone groups [9] , [19] , [20] . Thus , chaperones form complex networks of interaction in the cell . This motivates a holistic , systems-based approach , in which experimental data from a variety of contexts is synthesized to study the proteostasis network in its entirety . This is precisely the goal of FoldEco [21]: a recently presented tool that describes the proteostasis network in Escherichia coli . FoldEco synthesizes previously established models of various chaperone systems into a single network of reactions , whose rates are parameterized using experimental data . Dynamics on the FoldEco network describe the synthesis , folding , unfolding , misfolding , aggregation and degradation of a client protein , as well as the passage of the client protein through three chaperone systems , which work to correct misfolded structures , prevent aggregation and maintain a population of functional native protein . The FoldEco program uses a set of initial conditions and reaction rates to propagate the concentrations of the different species in the network forward in time . However , lost in this approach is the ability to track the trajectories of single molecules , which would allow us to answer fundamental questions such as: how often is a given chaperone system used to get from one point in the network to another ? We have developed a network analysis technique that quantitatively determines mediation probabilities in complex networks ( i . e . , how often a state A is found on transition paths from B to C ) . This analysis was previously used to detect hub-like activity in protein configuration space networks , and was referred to as “hub scores” [22] , [23] . Here , we show how mediation probabilities can be calculated from the output of the FoldEco program by constructing transition matrices for states of the client protein . We show that these probabilities can provide insight into proteostasis networks by revealing how often different competing pathways , involving different chaperone systems , are used to connect different regions of the network , such as the misfolded and unfolded states of the client protein . For client proteins , we choose four characteristic biophysical protein profiles based on the Monte Carlo results of Powers et al [21] , which demonstrate a range of characteristic behaviors . We calculate the relative probabilities of taking transition paths through each chaperone system for the four different protein types , and demonstrate how these probabilities change as a function of system parameters , such as the protein synthesis rate , and the total chaperone concentration .
In the FoldEco model , there are a large number of parameters that can be adjusted in order to more accurately describe the activity of a particular protein . As exploring this entire parameter space is infeasible , in Powers et al . [21] a subset of six variables were chosen , and the resulting six-dimensional space was explored using a large number of points chosen by Monte Carlo sampling . The six variables comprise folding rate constants ( and ) , misfolding rate constants ( and ) , and well as two parameters that control the aggregation propensity of the protein ( and ) . We instead study a small number ( four ) of proteins in depth , which are chosen to demonstrate a range of preferences for the different chaperone systems . Using the results of the Monte Carlo study , Powers et al . determined biophysical profiles of optimal substrate proteins for the GroELS system while the KJE system is present , and vice versa . They were chosen based on the percent increase of native concentration upon the addition of either the GroELS or the KJE chaperone system ( in the presence of the other ) . We choose parameters from the distributions of the optimal GroELS and KJE substrates to define two of our client proteins . As the optimal GroELS substrates characteristically showed slow folding rates , we refer to the GroELS protein as “Slow Folder” . Similarly , the optimal KJE substrates characteristically show high misfolding rates , and we refer to the KJE protein as “Bad Folder” . We also define a protein using the average biophysical profile of a set of proteins that were found to aggregate at a low synthesis rate ( ) [21]; we refer to this protein as “Aggregator” . Finally , a fourth protein is defined using values of the six parameters that are intermediate between the three proteins defined so far . Since these values are close to the default values given by the FoldEco program , we refer to this protein as “Default” . The values of these 6 parameters are given for each of the four proteins in Table 1 . Using mediation probabilities , we determine how often certain transition pathways are used between two given states in the FoldEco network . An overview of the network is given in Figure 1 , and more information on its construction and the FoldEco model is given in Methods . In this section , we study how the network corrects misfolded states ( transitions in Figure 1 ) . With all three chaperone systems active , there are five distinct transition paths possible from the misfolded state to the unfolded state . These transitions are: direct , GroELS-mediated , KJE-mediated , B+KJE-mediated , and degradation followed by re-synthesis . We study how often these pathways are visited as a function of synthesis rate . As the synthesis rate increases , correcting misfolded states becomes increasingly important if aggregation is to be prevented . The synthesis rate is controlled through the ribosome activation rate ( rate constants and in Powers et al . [21] ) , which takes on the values of and , resulting in synthesis rates of and , respectively . We initialize the simulation with no client protein , and a fixed concentration of chaperone species ( given in Table 2 ) . The simulation is stopped at and the concentrations at that time are used to construct the rate matrix used for analysis as described in Methods . We found that after , the native protein concentrations have reached equilibrium and the protein dynamics are approximately steady state in systems that do not feature runaway aggregation . The relative pathway probabilities for the four characteristic proteins are shown in Figure 2 . The Default , Bad Folder , and Aggregator proteins show similar behavior: at low synthesis rate , the clearance of misfolded protein is mostly governed by the KJE system , and this responsibility is shifted gradually to the B+KJE system as the synthesis rate increases . This is expected , since the KJE and B+KJE systems produce unfolded protein , via similar mechanisms , from misfolded and aggregated protein , respectively . More aggregated protein is present at higher synthesis rates , which causes the fraction of misfolded protein cleared by B+KJE to increase ( Figure S1 shows native , unfolded , misfolded and aggregated protein concentrations ) . We test this hypothesis by comparing the ratio of activity of the KJE and B+KJE systems with the ratio of the concentrations in the misfolded and aggregated states ( Figure 3 ) . The ratio can be fit to the function , where is the ratio of the concentration of misfolded and aggregated protein . This is consistent with the hypothesis that the relative probability of the two pathways is governed by the relative concentrations of their starting points . The proportionality constant , , indicates that if the concentration ratio of misfolded to aggregate protein is , then the KJE pathway ( from M to U ) is preferred over the B+KJE pathway by a factor of about . The crossover from the KJE pathway to the B+KJE pathway is shifted to lower and lower synthesis rates as we move from Default to Bad Folder to Aggregator . This can be explained by the values of the three proteins: and , respectively . For a given synthesis rate , the relative population of aggregates grows with larger , which would lead to increased usage of the B+KJE system , as shown in Figure 3 . The Slow Folder protein shows markedly different behavior . The highest probability pathway is the GroELS system , and the probabilities are roughly constant as a function of synthesis rate . As the Slow Folder protein was chosen as the ideal client for the GroELS system , this is not surprising , but the question remains as to specifically why the GroELS pathway is favored . The main entries from the misfolded state into the KJE and GroELS systems , respectively , are the ( misfolded protein associated with ATP-bound DnaK ) and ( misfolded protein associated with ATP-bound∶GroEL ) states . As such , we compare the entrance rates into these states from the misfolded state at the lowest synthesis rate ( Figure 4a ) . Although the entrance rate into the GroELS cycle is about higher for Slow Folder as compared to the others , this is not sufficient to explain the drastic difference in path preference shown in Figure 2 . Figure 4b shows committor probabilities starting from the and states . These are the probabilities of reaching the unfolded state before the misfolded state given a starting point in either or , and are computed from the matrices described in the Methods section “Getting mediation probabilities” , where states and in this context are the misfolded and unfolded states . For Slow Folder , transition paths from have a much higher likelihood of reaching the unfolded state ( ) as compared with the other three proteins ( ) . This is due to the fast transitions in the GroEL cavity ( the rates of which are set to be the same in solution ) , equal to and for the Default , Bad Folder and Aggregator proteins respectively , and for the Slow Folder protein . Figure 4b also reveals why the KJE pathway is favored for the other proteins . Although the entrance rate into the GroELS cycle is higher than the KJE cycle , the committor probability of reaching the unfolded state is to times higher for KJE . This underscores the importance of folding kinetics to the efficiency of the GroELS cycle . These results give different information than a more conventional “knockout” analysis wherein a particular chaperone system is disabled and the effects on a particular observable is measured – usually either representative of the concentration of native species , or the concentration of aggregated species . To demonstrate this we choose a particular protein and synthesis rate for analysis that is particularly interesting: the Default protein at a ribosome activation rate of . As shown in Figure 2 , at this synthesis rate the Default protein uses the chaperone systems GroELS , KJE and B+KJE with approximately equal probability . We then study this protein with the GroELS system knocked out , achieved by setting the initial concentration of GroEL and GroES to zero . The native , unfolded , misfolded and total aggregate concentrations at are shown in Figure 5a . We see that the native concentration is approximately unchanged , and the amount of aggregated species is still negligible ( the percent of insoluble protein is and with and without GroELS , respectively ) , indicating that knocking out the GroELS system would not result in a measurable change in the observables corresponding to the native species or aggregated species concentrations . Figure 5b shows the contributions to the flux by the GroELS , KJE and B+KJE chaperone systems both with and without the GroELS system present . We see that the absence of the GroELS system is more than compensated for by an enhanced contribution of the B+KJE system . Even though the GroELS system is used to clear over of the misfolded protein under these conditions , its removal had no effect on the native state concentration . This example highlights the advantages of using a transition-path based analysis over knockout experiments when multiple chaperone systems are present . It is also interesting to see that the contribution along the KJE pathway also decreased as GroELS is removed , even though the concentration of its starting state ( the misfolded state ) increased . This is because the B+KJE pathway uses the chaperones DnaK , DnaJ and GrpE , leaving a lower concentration available for the KJE pathway . In order to see if binding affinities can govern chaperone preferences , we vary the binding rates to the chaperones DnaK and GroEL and observe the impact on the relative probabilities of each chaperone pathway . The coefficients are varied using a multiplicative factor . When , we recover the binding rates used above ( which are for X to , for X to GrL∶X , and for X to , where X is either U or M ) . The GroEL binding rates are multiplied by , and the DnaK binding rates are divided by , which allows to act as a tuning variable that encourages usage of either the KJE or the GroELS chaperone systems . In Figure 6a , we use the Default protein at the lowest ribosome activation rate ( ) , which has a natural preference for the KJE pathway . As increases , the pathway preference smoothly switches from KJE to GroELS , crossing over between and . In Figure 6b , we use the Slow Folder protein at the lowest synthesis rate , which has a preference for the GroELS pathway . As decreases ( from right to left ) , the KJE probability initially increases , and then decreases for . This behavior is counter-intuitive: why would increasing the DnaK binding rate discourage usage of the KJE pathway ? Figure 6c demonstrates that even though the binding rate increases with decreasing , the committor probability from to the unfolded state decreases . This is due to the decreasing concentration of free DnaJ , which is spent by the formation of and complexes which can result from unproductive binding of unfolded protein to DnaK ( Figure S2 ) . Nevertheless , substantial switching of path preferences can again be achieved by adjusting the binding affinities by a factor between and . This suggests , for the M to U transition , that relative usage of the two chaperone systems can be controlled through modest adjustments of the relative binding affinities to the two chaperones . We now study mediation of the ( or folding ) transition . This is less complicated than the transition in that there are only two possible pathways: direct and GroELS-mediated . We construct an analytical model of GroEL-mediated folding in supplemental file Text S1 . The performance of the GroELS chaperone system depends on both the capacity of the system ( quantified by the concentration of total GroEL and total GroES ) , and the demands on the system ( quantified by the concentrations at the entry points to the GroEL system – the unfolded and misfolded states ) . We study the percentage of GroELS-mediated folding trajectories as a function of the system capacity over the set of four proteins , each of which have different system demands . Figures 7a–d show the pathway flux through the GroELS system ( solid bars ) as well as the direct flux ( transparent bars ) from the unfolded state to the folded state . The total concentrations of GroEL and GroES are varied together by a multiplicative factor ranging from to ( plotted on the horizontal axes ) , where results in the concentrations used in the previous section . For all four of the proteins , both the absolute and percentage usage of the GroEL-mediated folding pathway goes down with decreasing total GroEL concentration . For three of the proteins ( Default , Slow Folder , and Bad Folder , shown in panels a , b and c , respectively ) , this decrease is mostly compensated for by an increase in usage of the direct pathway . Of these , Slow Folder has the largest decrease in total folding flux , resulting in a decrease in native yield of , while Default and Bad Folder have decreases in native yield of and , respectively . The compensation for lack of GroEL folding flux occurs by accumulation of unfolded protein that would otherwise enter the GroELS system ( as the direct folding flux is given by ) ( Figure 7e ) . Therefore , GroEL is not needed for folding , but these proteins will take the GroEL pathway ( almost exclusively ) if it is available . As the compensation for the lack of GroELS occurs by building up population in the unfolded state , we note that without the GroELS system these proteins will be more vulnerable to degradation , and we expect to see a stronger dependence of native protein yield on GroELS at higher concentrations of the protease Lon . In contrast , the folding flux for the Aggregator protein is highly dependent on the available concentration of GroEL . This is because the GroEL system acts as a “holder” to keep proteins from aggregating , and the extra unfolded protein resulting from its removal does not accumulate , but is transferred to the misfolded state , and subsequently aggregates . The Aggregator protein can thus be seen as a “class-III substrate” in that the total folding flux ( and thus the concentration of the native state ) is dependent on the availability of GroEL [7] . The other three proteins ( Slow Folder , Default and Bad Folder ) can be seen as “class-I” or “class-II substrates” of the GroEL system ( see Text S1 ) , in that they do not strictly require the GroELS system to fold . It is striking that the Slow Folder protein is not a “class-III substrate” , even though it was parameterized to be an optimal substrate of the GroELS system . We note that this is done using parameters at a lower synthesis rate . At this higher synthesis rate , the GroELS system primarily serves to rescue proteins from aggregation , as opposed to degradation , and as such the optimal clients of the GroELS system misfold and aggregate easily [21] . The biophysical profiles of top GroELS substrates in the presence of KJE at this synthesis rate are similar to that of the Aggregator protein ( see Figure S4B of Powers et al . [21] ) . It is important to note that the simulations conducted here only take into account one client protein at a time , whereas in vivo , there are about different proteins that act as GroELS substrates , which compete to bind to a shared pool of GroELS chaperones [7] . It is then easy to see how competition can arise between class-I/II and class-III substrates , as strong-binding class-I/II substrates would lower the effective concentration of total GroEL , reducing the yield of class-III substrates without increasing their yield of native protein . There should thus be an evolutionary drive to increase the binding affinity of class-III substrates in comparison to class-I/II substrates . In Text S1 we examine whether increasing the GroEL binding affinity for the Aggregator protein can compensate for lower concentrations of GroEL chaperone , and we find that it cannot . This underscores the importance of increasing binding affinity from the perspective of inter-protein competition .
We have used transition path analysis in combination with the FoldEco program to study the proteostasis network in E . Coli . The analysis reveals features of the network dynamics that are undetectable by observing concentrations of network components alone . For the misfolded to unfolded transition , we find that the usage of the KJE vs B+KJE systems depends mostly on the relative concentrations of the misfolded and aggregated states . We also observe that the efficiency , and hence the pathway probability , of the GroELS cycle depends mostly on folding kinetics of a client protein within the GroELS cycle . If the folding kinetics within the GroEL-GroES chaperone complex are the same as in bulk , in order for the GroELS system to increase native yields , either the degradation or aggregation processes need to be competitive with folding . We have also shown that modest adjustments in the binding affinities to the two chaperones DnaK and GroEL can control which chaperone system is used to correct misfolded states at a given synthesis rate . This study serves as a proof of principle that a transition path analysis can be applied to proteostasis-type networks with little complication . We expect that this analysis will become more valuable as networks become larger and more interconnected , since the behavior of the transition paths will become less intuitive . The computational cost of the analysis is dominated by multiplications of matrices that are approximately size by , where is the number of states in the network . Although matrix multiplication scales as , GPU architectures allow fast multiplications of large matrices ( a two-GPU cluster can multiply matrices at a speed of [24] , hence matrices of size by can be multiplied on a two-GPU cluster in about minutes ) . This would make the analysis presented here feasible on networks up to about states with current hardware . For larger networks , rather than multiply matrices it would be easier to generate a large number of “psuedo-trajectories” using the state-to-state transition probabilities , and calculate mediation probabilities directly from the trajectories , as done in our previous work [23] . Mediation probabilities would be extremely challenging to measure experimentally , since they would rely on the tracking of single molecules in vivo . For instance , to determine the relative fraction of GroELS- , KJE- and B+KJE-mediated misfolded to unfolded transitions , one would need to distinguish between misfolded , unfolded , GroELS-bound , DnaK-bound and aggregated states in real time . However , even without verifying the mediation probabilities directly , the overall proteostasis network model can be verified by comparing the concentrations of species in the model with those from experiment over a range of system parameters ( as is done for firefly luciferase in Powers et al . [21] ) . This prescribes a complex synergy between theory and experiment , where experiment is first used to parametrize the reactions , theory is used to construct a network , experiment then used again to validate the network , and theory used again for the network analysis described here .
There are four main chaperone systems acting to maintain proteostasis in E . coli . The first is the Hsp70-like system , consisting of chaperones DnaK , DnaJ and GrpE ( the KJE system ) . DnaK has a hydrophobic pocket that preferentially binds to unfolded peptides with exposed stretches of hydrophobic residues [9] , [12] . Bound peptides can be locked in through the motion of a helical lid domain that is closed by the hydrolysis of bound ATP , which is regulated by the binding of co-chaperone DnaJ . After DnaJ unbinds , the binding of GrpE catalyzes ADP release , and subsequent ATP rebinding results in lid opening and release of the peptide . Because the protein is kept unfolded throughout the cycle , the KJE system can allow misfolded , aggregation-prone proteins to return to an unfolded state . A large part of the E . coli proteome ( at least proteins ) binds to DnaK [9] , making the KJE system extremely important in preventing aggregation [4] , [25] . The second is the Hsp60-like GroEL/GroES chaperonin system ( GroELS ) , which is the only chaperone system that is absolutely necessary for the viability of an E . coli cell [26] . GroEL exists as two stacked seven-membered rings which form a cylindrical complex that is capable of encompassing a single protein , acting as an infinite-dilution cage . GroES forms a single seven-membered ring that acts as a cap to the cylinder , enclosing the protein . It has been shown that enclosure within the GroEL∶GroES complex can increase folding rates [27] , although the chaperonin system works to prevent aggregation even when folding kinetics are unchanged . The unbinding of the GroES cap is mediated by allosteric ATP binding , and occurs after seconds [4] , which gives the peptide time to fold in a sterically-confined environment that is isolated from other misfolded copies of the peptide that encourage aggregation . Discharged protein that is not folded can be rapidly rebound , and consequently many proteins are known to undergo many GroELS cycles before folding [27]–[29] . GroEL binds to a wide variety of proteins , comprising at least to of cytosolic proteins under normal growth conditions [6] . An in vitro study by Kerner et al . shows that of about proteins that interact with GroEL , about are absolutely dependent on the chaperonin system to fold [7] . However , a more recent study has shown that only of these are strictly dependent ( or “obligate” ) on GroELS in vivo [8] . The KJE system can also cooperate with the Hsp104-like chaperone ClpB to pull monomers from amorphous aggregates ( the B+KJE system ) [11] , [13]–[15] , [25] . ClpB is an oligomeric , ring-like machine that uses the energy from ATP hydrolysis to exert mechanical force on protein aggregates . Both DnaK and DnaJ are used to prepare aggregates for ClpB which then can extract monomers from the aggregate [30] . Two mechanisms have been proposed for the disaggregation mechanism of ClpB: one in which ClpB acts as a “crowbar” to break apart an aggregate [13] , the other in which ClpB threads a single monomer through a central pore [14] . The last chaperone is trigger factor , which can bind to translating polypeptides and protect them from aggregation [7] , [19] . As trigger factor only acts as a holder chaperone , trigger-factor–bound states cannot act as intermediates on transition paths between the major client protein states ( e . g . native , unfolded , misfolded ) . We thus exclude trigger factor from our transition path analysis , and focus on the first three chaperone systems mentioned above . The FoldEco program was recently introduced to study the proteostasis of a client protein . It describes , in a holistic fashion , the synthesis , folding , misfolding , aggregation , degradation and recovery of misfolded and aggregated proteins through the KJE , GroELS and B+KJE chaperone systems . It uses coupled kinetic equations that evolve a particular set of initial conditions ( which are the concentrations of each species in the system ) forward in time , using reactions that are parameterized from in vitro experimental data . In Figure 1 we show the network of client protein states used here . The nodes in the network are particular configurations of a single protein molecule , and mostly describe the formation and destruction of complexes with different chaperones in the proteostasis network . This can be compared with Figure 1 of Powers et al [21] , where there is more information about the nature of the transitions , but does not explicitly include all of the connections between client states . We note that FoldEco also describes reactions that do not involve the client protein , such as the binding and unbinding of ATP from DnaK . We omit these from Figure 1 since they are not part of the network of client states . To simplify our analysis , we also connect the processes of degradation and re-synthesis through a “null” state . This does not affect our results , and allows us to examine the steady-state dynamics of a single protein traversing the network . The rate constants for the transitions between the states in this network are determined from a large body of experimental literature . In theory these rate constants can be tailored in a protein-specific fashion to more accurately connect with experiment , although for simplicity we fix all but six rate constants , and the values for these fixed constants are given in Table S4 of Powers et al [21] . The same table also describes the initial concentrations of the chaperone species used here , which are reproduced in Table 2 . Although FoldEco is a powerful tool for synthesizing experimental data , there are some simplifications used by the model that affect our analysis . Firstly , it does not account for the effect of bacterial growth , which would lead to the dilution of proteins as they are being synthesized . One effect of this is that steady-states reached by FoldEco tend to have much larger concentrations of protein than are observed in experiment . Thus , in our analysis we do not analyze the networks at steady-state , we instead choose a common analysis time for each system ( ) . FoldEco also does not take into account the presence of the background proteome , and does not describe competition for binding to chaperones . Above , we study this competition indirectly by lowering the concentration of GroEL and GroES that is accessible to the client protein . We note that both of these limitations are planned to be addressed in future versions of FoldEco [21] . The kinetic equations in FoldEco are formulated as a set of equations that describe the time evolution of the concentrations of different client and non-client species in the system . For our analysis , we wish to convert this into a master equation of the form , where is a vector of the concentrations of different states in the model , and is a time-independent rate matrix , the elements of which describe the rate of transition from state to state . This allows for the transitions of a single tagged protein molecule to be tracked from state to state , and for the analysis of its dynamical properties . The first complication to arise is that some of the states in the network involve multiple copies of the client protein . For instance , a GroEL-GroES complex can accommodate two client proteins , one in the cis ring and one in the trans ring . It is important to maintain a distinction between proteins in the same complex if they are in different states ( e . g . , one is folded and the other is unfolded ) . We thus artificially separate the multi-client complex states into two states , depending on which protein is the tagged protein . This allows us to track the tagged protein in a continuous fashion once the complex has dissolved . Aggregated structures are also multi-client states , but to rigorously keep track of a specific monomer in a large aggregate would be unfeasible: describing aggregates up to monomers in length would require over states for each aggregate-containing species in the network . This is because a monomer of size would require distinct states that distinguish the position of the monomer within the aggregate . Furthermore , it is not clear whether or not the order in which monomers are added to the aggregate affects the order in which they would be removed by ClpB . We thus assume , upon removal of a monomer from an aggregate of size , that the probability of the tagged protein being removed is equal to . The assumption of the indistinguishability of monomers is reasonable for amorphous aggregates , and would also be reasonable for highly ordered aggregates if ClpB could act via a “crowbar”-type mechanism [13] , allowing monomers to be extracted from the middle of a structure . We note that for beta amyloid , both amorphous ( preamyloid ) and fibrillar aggregates can be observed in vitro , depending on the conditions [31] . After a proper set of states is established , a second complication arises in building the master equation , as some of the differential equations are nonlinear ( e . g . , the rate of aggregation of two monomers depends on the square of the monomer concentration ) , and many of the rates depend on the concentrations of non-client protein species , which are changing over time . We instead cast the problem as , where the elements of depend on the average concentrations of the network components , and thus on the time . In principle , at long times the concentrations of the different species in the model will reach steady state , and the elements of will be time-independent , but this limit does not always exist , especially for conditions in which runaway aggregation occurs . In order to employ the same protocol for a broad range of simulation conditions , we instead choose an analysis time of interest ( ) and use the rate matrix calculated at that time for analysis . We then assume that the elements of this matrix are approximately constant on the timescale of the transition paths . This assumption is addressed in Text S1 . Once we have obtained the rate matrix as calculated at the observation time , we use a method similar to that proposed in our previous work [22] to obtain mediation probabilities . Previously , modified rate matrices were diagonalized in order to determine infinite-time behavior . As we have found some of the matrices here to be unstable to diagonalization , we instead use a slightly modified approach . Firstly , the rate matrix is converted into a transition probability matrix , in which the elements are equal to the probabilities of transition between states in a given amount of time , . To find an appropriate we find the fastest rate in the system ( ) , and then set . The equation , yields , for small ( 1 ) where is the identity matrix , and denotes the value of the vector at time . The nondiagonal elements of the transition matrix are then equal to , and the diagonal ones to . This transition matrix is then modified in the same manner as in our previous work [22] . Here we describe the method in brief , focusing on the parts that differ from our previous implementation . Consider a modified transition probability matrix with “sinks” at two states and , created by setting the nondiagonal elements in columns and uniformly to zero , and setting the diagonal elements to . The long-time dynamics using this new matrix , , reveals committor probabilities for each state as follows: ( 2 ) where is the probability of reaching before , given starting in state , and . In order to determine mediation probabilities , we calculate “conditional committors” , such as , which is the probability , when starting in , of reaching before , having gone through a third state , . The probability of reaching before having not gone through is denoted . Conservation of probability now gives the equality . These conditional committors are computed using an extended probability matrix , where two ensembles of states are used: one in which has been visited so far along the trajectory , the other in which has not been visited . Sinks are put into the matrix at and , in both ensembles , and conditional committors are determined using the elements of the infinite-time extended probability matrix , in a manner similar to Equation 2 . In practice , is computed by iteratively squaring the matrix until the probability in the non-sink states is less than . The number of iterations required is less than or equal to for all matrices used here . | To maintain proper amounts of folded , functional proteins , cells use systems of chaperones to correct misfolded proteins , disassemble aggregates , and provide sheltered environments in which proteins fold to their native structure . Typically , an individual system is studied in isolation , and its effects on a given protein are studied using “knockouts” , where the amount of native protein is compared with and without the active chaperone system . However , when multiple chaperone systems are operating simultaneously , knockouts can fail to reveal chaperone activity , as different chaperone systems can compensate for one another . We use a previously introduced computational model of chaperone systems in Escherichia coli , in combination with our transition-path analysis methods for networks , to analyze paths of individual proteins through the set of possible chaperone-bound and -unbound states . Our analysis allows us to answer questions that are inaccessible to knockout experiments , such as: How often will a given chaperone system be used to rescue a protein from a misfolded state ? This approach provides a clear view of how the different systems of chaperones cooperate and compete under varying conditions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| []
| 2013 | Quantifying Chaperone-Mediated Transitions in the Proteostasis Network of E. coli |
Mycolactones are a family of polyketide-derived macrolide exotoxins produced by Mycobacterium ulcerans , the causative agent of the chronic necrotizing skin disease Buruli ulcer . The toxin is synthesized by polyketide synthases encoded by the virulence plasmid pMUM . The apoptotic , necrotic and immunosuppressive properties of mycolactones play a central role in the pathogenesis of M . ulcerans . We have synthesized and tested a series of mycolactone derivatives to conduct structure-activity relationship studies . Flow cytometry , fluorescence microscopy and Alamar Blue-based metabolic assays were used to assess activities of mycolactones on the murine L929 fibroblast cell line . Modifications of the C-linked upper side chain ( comprising C12–C20 ) caused less pronounced changes in cytotoxicity than modifications in the lower C5-O-linked polyunsaturated acyl side chain . A derivative with a truncated lower side chain was unique in having strong inhibitory effects on fibroblast metabolism and cell proliferation at non-cytotoxic concentrations . We also tested whether mycolactones have antimicrobial activity and found no activity against representatives of Gram-positive ( Streptococcus pneumoniae ) or Gram-negative bacteria ( Neisseria meningitis and Escherichia coli ) , the fungus Saccharomyces cerevisae or the amoeba Dictyostelium discoideum . Highly defined synthetic compounds allowed to unambiguously compare biological activities of mycolactones expressed by different M . ulcerans lineages and may help identifying target structures and triggering pathways .
The macrolide exotoxin mycolactone is a key virulence factor of M . ulcerans and plays a central role in the pathogenesis of Buruli ulcer [1] . Mycolactones have been shown to act both in vivo and in vitro on various mammalian cell types , including fibroblasts [1]–[6] , adipocytes [7] , keratinocytes [8] , myocytes [9] , [10] , macrophages [2] , [6] , [11]–[14] , dendritic cells [15] and T-cells [16]–[18] . Effects caused by mycolactones include induction of apoptosis/necrosis , cytoskeletal rearrangements , impaired cytokine production and interference with cellular signaling . The polyketide synthases required for mycolactone biosynthesis are encoded on the extrachromosomal plasmid pMUM [19] , [20] . M . ulcerans has evolved from a common M . marinum ancestor by acquisition of this plasmid and has subsequently diverged into two principal mycolactone-producing lineages [21] , [22] . The “classical” lineage includes M . ulcerans isolates associated with Buruli ulcer from Africa and Australia . The “ancestral” lineage includes both Buruli ulcer isolates from Japan , China and Mexico and isolates from fish and frogs previously also designated M . pseudoshottsii , M . liflandii or M . marinum [23] , [24] . Mycolactones are composed of a 12-membered macrolide core and two attached side chains; a short upper , C-linked side chain ( comprising C12–C20 ) and a longer lower , C5-O-linked polyunsaturated acyl side chain . While the macrolide core structure and upper side chain are conserved , mycolactone populations from different M . ulcerans sub-lineages vary in the length , the number and localization of hydroxyl groups and in the number of double bonds of the lower side chain . M . ulcerans strains may produce several molecular variants of mycolactone , with one or more species dominating [25] . The mycolactone repertoire seems to be highly conserved within a defined geographical sub-lineage of M . ulcerans [25] . Mycolactone A/B is produced by strains of the classical M . ulcerans lineage found in Africa and is regarded as the most potent toxin . Australian classical lineage strains produce - in addition to mycolactone A/B - mycolactone C , which lacks one hydroxyl group . Mycolactone D with an additional methyl group is produced by Chinese strains belonging to the ancestral lineage . M . ulcerans ancestral lineage isolates from fish and frogs have been found to produce the mycolactone variants E and F . Mycolactones have previously been prepared from M . ulcerans cultures by a two-step extraction procedure , yielding preparations of acetone soluble lipids predominantly containing mycolactone . These extracts can be further purified by chromatographic methods [26]; nevertheless , the use of extracted mycolactones for comparative studies may be compromised by the heterogeneity of preparations . Therefore , biological studies with highly defined synthetic mycolactones represent an attractive alternative . Based on the established synthesis of the mycolactone core [27]–[30] , different synthesis strategies have been pursued for the stereoselective partial and total synthesis of mycolactones [31]–[33] . In addition , simplified C8-desmethyl-mycolactone analogues have been synthesized , which were analyzed for their cytopathic potency by using cell rounding as a parameter to compare cytotoxic activities [34] . No systematic structure-activity relationship studies on larger sets of synthetic mycolactones have been published so far . Results of individual studies cannot be reliably compared , since different readouts , such as cell rounding [22] , [34] , [35] , cytokine production [36] or flow cytometric parameters [35] , [36] have been employed . Furthermore , different cell lines ( such as Jurkat T-cells [36] , murine fibroblasts [22] , [34] , [35] and sets of human tumor cell lines [37] have been used and cytotoxic activity has been assessed after different times , such as 24 hours [36] or 24 and 48 hours [35] . Most of these structure-activity studies have been limited to mycolactone A/B and a limited number of derivatives . A comparison of the activity of eight C8-desmethyl mycolactone analogues is hampered by the fact that lack of the C8-methyl substituent reduces the cytopathic activity by a factor of 125 [34] . Here , we have performed more systematic comparative studies using synthetically produced natural toxins and additional structural mycolactone variants that are not found in nature .
We recently reported the synthesis of mycolactone A/B [38] . Details of the syntheses of mycolactones C and F and of the six non-natural mycolactone derivatives will be published elsewhere ( Gersbach et al . , manuscript in preparation ) . Briefly , all mycolactones discussed here were prepared by the same overall strategy that we had previously developed for the synthesis of mycolactone A/B . Thus , a modified Suzuki coupling was employed to establish the C12–C13 bond and elaborate the full upper side chain and a Yamaguchi type acylation reaction was used to attach the lower side chain . All final products used for biological testing were purified by RP-HPLC; they were generally obtained as mixtures of ( interconverting ) double bond isomers . Analytical data for the synthetic compounds are provided as supplementary information ( S1 ) . For biological testing , 0 . 5 mg/ml stock solutions of the mycolactones were prepared in cell culture grade DMSO ( Sigma ) . Stock solutions were aliquoted and stored frozen at −20°C . Murine L929 fibroblasts were grown in RPMI medium ( Gibco ) supplemented with 10% FCS ( Sigma ) , 2 mM glutamine ( Gibco ) and 0 . 05 mM β-mercaptoethanol ( Gibco ) and incubated at 37°C and 5% CO2 . Cells were passaged 3–6 times prior to use in cytotoxicity experiments . For flow cytometry analysis 24 , 000 cells were seeded into 24-well plates ( Falcon ) and allowed to adhere o/n . Medium was then aspirated and replaced by 500 µl medium containing different concentrations of mycolactone and 0 . 12% DMSO ( vol/vol ) . After incubation for 24 , 48 or 72 h , cells were detached from the culture plates by repeated gentle flushing through a pipette tip without use of Trypsin-EDTA . Harvested cells were centrifuged for 10 min at 1 , 200xg , resuspended in 300 µl binding buffer with 0 . 2 µg/ml Annexin V-FITC ( AnnexinV kit , Calbiochem ) and incubated for 30 min at 4°C . The cells were spun again and pellets were resuspended in 300 µl staining buffer containing 0 . 3 µg/ml propidium iodide ( AnnexinV kit , Calbiochem ) . Cell suspensions were analyzed by flow cytometry using a BD FACS Calibur Flow Cytometer ( Becton Dickinson ) and apoptotic ( A+/PI− ) and necrotic ( A+/PI+ ) cell populations were determined using the CellQuest Pro Software ( Becton Dickinson ) . The experiments were set up in triplicates and performed at least twice . Mycolactone A/B in a concentration range of 3 . 75 to 120 ng/ml was included as control in all experiments . The mycolactone concentration at which 50% of the cells were killed ( LC50 ) was determined by plotting the percentage of affected cells ( sum of A+/PI− and A+/PI+ cells ) against the log concentration of the individual mycolactones . In Fig . 1–3 only data for concentrations close to the LC50 are shown . To measure proliferation of L929 fibroblasts , 24 , 000 cells were seeded into 24-well plates ( Falcon ) and allowed to adhere o/n . The medium was aspirated and replaced by 500 µl medium containing 60 ng/ml of mycolactone and 0 . 06% DMSO ( vol/vol ) . At time point 0 and after 24 , 48 and 72 h fibroblasts were harvested , diluted 1∶100 in isotonous solution and measured using an automated cell counting device ( Casy®TT , Schärfe System ) . The experiment was set up in triplicates and performed twice . 24 , 000 cells were seeded on four-chamber glass slides ( BD Falcon ) with complete RPMI medium and allowed to adhere for 24 hours . The medium was aspirated and replaced by 500 µl medium containing different concentrations of mycolactone . After the specified incubation period , cells were washed once in PBS and fixed with 4% formaldehyde ( Medite ) for 20 min . Fibroblasts were washed again in PBS prior to permeabilization in Triton X-100 ( 0 . 1% in PBS ) for 20 min . Cells were rinsed in PBS and blocked by incubation in 4% FBS in PBS for additional 20 min . The actin cytoskeleton was stained by incubating the cells for 1 h at room temperature with Texas Red-X phalloidin ( 3 units/ml , in blocking solution , Molecular Probes ) . Cells were washed in blocking buffer , then in PBS . ProLong Gold antifade reagent ( Invitrogen ) containing diamidino-2-phenylindole ( DAPI ) was used for nuclear counterstaining . Cover slips were mounted onto the slides and cell rounding as well as the staining of nuclei and actin cytoskeleton was qualitatively analyzed by fluorescence microscopy ( Leica DM 5000 B ) . Mycolactone-induced changes in the pattern and intensity of the Texas Red-X phalloidin staining of the cytosolic actin cytoskeleton as well as in the uniform , round and clear-edged DAPI staining of nuclei in healthy cells were observed . Metabolic activity of mycolactone-treated L929 fibroblasts was analyzed by performing Alamar Blue assays . Seeding of cells and addition of mycolactones were performed as described for the flow cytometry-based cytotoxicity assays . After mycolactone treatment , alamarBlue® reagent ( Invitrogen ) was added to the wells ( 10% v/v ) and the cells were further incubated for 1 hour at 37°C and 5% CO2 . Fluorescence intensities were measured using a SpectraMax Gemini XS ( Molecular Devices ) and the values were calculated referring to the DMSO control ( 0 ng/ml mycolactone ) set at 100% . The experiments were set up in triplicates and performed at least twice . The concentration at which the metabolic activity of cells was inhibited by half ( IC50 ) was determined by plotting the fluorescence intensity against the log concentration of the individual mycolactones . Antimicrobial activity of mycolactone on Streptococcus pneumoniae ( SP1 , P1577 ) , E . coli ( DE ( 3 ) ) and Saccharomyces cerevisiae was tested by applying the disk agar diffusion ( Kirby-Bauer ) method . Bacteria were pelleted , resuspended in PBS and spread on blood agar/LB agar . The plates were dried for 30 min and sterile paper disks were distributed circle-like onto the agar . Mycolactone A/B solutions of different concentrations ( 0 . 003 µg/ml to 10 µg/ml ) were applied on the paper disks ( 40 µl ) . The agar plates were incubated o/n at 37°C and then analyzed for potential zones of inhibition . The effect of mycolactone A/B on the growth of Dictyostelium discoideum DH1-10 was assessed by performing an Alamar Blue assay in a 24-well format . 2 , 000 cells were seeded in 500 µl medium containing mycolactone in the concentration range of 0 . 16 to 500 ng/ml . As controls , DMSO and blasticidin were used . After an incubation period of 3 days at room temperature , alamarBlue® reagent ( Invitrogen ) was added and the plate was further incubated for 18 hours at room temperature .
Recently we described a novel strategy for the synthesis of mycolactone A/B that is based on the stereoselective construction of the macrolactone core by ring-closing olefin metathesis and subsequent incorporation of the C- and O-linked side chains by suitable fragment couplings [38] . Taking this synthesis approach a set of natural mycolactones ( mycolactone A/B , mycolactone C , mycolactone F ) and additional derivatives displaying modifications in the lower or upper side chain ( PG-119 , PG-120 , PG-155 , PG-157 , PG-165 and PG-182 ) were produced ( see Figures 1 , 2 and 3 ) for biological testing . The biological activity of these synthetic compounds on the murine L929 fibroblast cell line was assessed by flow cytometry . After treatment with different concentrations of synthetic mycolactones , cells were stained both with FITC-labeled annexin-V and with propidium iodide . Annexin-V binds to exposed phosphatidylserine residues translocated from the inner to the outer leaflet of the plasma membrane in cells undergoing apoptosis . Propidium iodide intercalates into the DNA of cells that have lost nuclear membrane integrity , serving as a marker for necrosis . Quadrant analysis was performed to determine apoptotic ( A+/PI− ) and necrotic ( A+/PI+ ) cell populations . While first signs of mycolactone A/B-mediated cell death were already detectable after 24 hours , significant effects were only observed after 48 hours [38] . For comparison with mycolactone A/B , the lethal concentration of mycolactone analogues at which 50% of the cells were affected ( LC50 ) was therefore determined after 48 hours ( Table 1 ) . As described previously [38] , mycolactone A/B was highly potent ( Figure 1 ) with a LC50 of 12 nM ( Table 1 ) . For the two naturally occurring structural variants mycolactone F and mycolactone C , the LC50 values were 29 nM and 186 nM , respectively ( Table 1 ) . Mycolactone C differs from mycolactone A/B in lacking the hydroxyl group at position C12 of the lower side chain . Mycolactone F has a shorter side chain with also only two hydroxyl substituents ( Figure 1 ) . While these natural mycolactones retained cytotoxic activity , compound PG-155 , a non-natural structural variant devoid of all hydroxyl groups in the lower side chain , showed only minor activity with a LC50 of 4550 nM ( Table 1 ) . Apart from these mycolactone variants with modifications in the lower side chain , also analogues with modifications in the upper side chain were synthesized and tested ( Figure 2 ) . Introduction of a hydroxyl group at C20 in compound PG-165 had no major effect , since PG-165 had only a slightly higher LC50 ( 15 nM ) than mycolactone A/B ( Figure 2 , Table 1 ) . In addition , derivatisation of this hydroxyl group into an acetate ( PG-157 with a LC50 of 45 nM ) or into a bulky butyl carbamate ( PG-182 with a LC50 of 50 nM ) reduced cytotoxicity only about three-fold ( Figure 2 , Table 1 ) . Thus , the upper side chain turned out to be relatively tolerant to a significant extension in length and to the presence of polar linker elements between the natural side chain and the extension module . PG-120 , a derivative with a significantly truncated lower side chain , showed some residual cytotoxic activity ( LC50 = 3426 nM ) , whereas PG-119 , a derivative with an acetyl residue as the lower side chain , showed no activity within the concentration range tested ( Figure 3 , Table 1 ) . For all compounds , except PG-120 , concentrations required for cytotoxic activity ( as measured by flow cytometry ) , reduction of metabolic activity in an Alamar Blue-based assay , changes in the intensity and pattern of phalloidin-staining of the actin cytoskeleton and changes in the round , clear-edged and uniformly stained nuclear morphology of normal cells were in the same range . While the IC50 value for PG-120 ( 171 nM; Table 1 ) , was twenty-fold lower than the LC50 , the LC50/IC50 ratios of all other compounds with widely varying toxic potency ranged between 1 . 5 and 3 . 2 ( Table 1 ) . Furthermore , at such sub-lethal PG-120 concentrations a marked reduction in cell proliferation ( Figure 4 ) , and a transient effect on the actin cytoskeleton accompanied by the rounding up of the cells , without changes in nuclear morphology was observed ( Figure 5A ) . A similar activity was not observed for PG-119 ( Figures 4 and 5 ) . When analyzed for antimicrobial activity , mycolactone A/B was found inactive against all microbial species tested , including Gram-positive ( Streptococcus pneumoniae ) and Gram-negative ( Neisseria meningitis , Escherichia coli ) bacteria; it was also inactive against yeast ( Saccharomyces cerevisae ) and amoeba ( Dictyostelium discoideum ) .
Our flow cytometric analyses of murine fibroblast L929 cells treated with a series of synthetic mycolactones reconfirmed that changes in the O-linked lower side chain can profoundly affect the biological activity . Activity of the synthetic mycolactone A/B was in the range reported for mycolactone preparations extracted from M . ulcerans cultures [1] . Mycolactone F was about two times less active and mycolactone C about 15 times less active than mycolactone A/B , respectively . For extracted mycolactone C an even far more pronounced difference in activity compared to mycolactone A/B has been described in assays determining L929 fibroblast rounding at 24 h and loss of monolayer at 48 h [25] . In addition to mycolactone C , Australian M . ulcerans strains also produce mycolactone A/B . Our data indicate that this mycolactone A/B portion may be more important for the pathogenesis caused by these strains than mycolactone C . In accordance with our findings , only a slightly lower activity was observed , when extracted mycolactone F was compared to mycolactone A/B in a L929 cell apoptosis assay at 24 h [3] . When the inhibition of IL2 production by activated Jurkat T-cells instead of cell death was used as readout , both mycolactones F and C were dramatically less potent than mycolactone A/B [36] . While we have investigated different types of modifications for the lower and upper side chains , it is clear that both the incorporation of polar substituents at C20 and the extension of the upper side chain by up to 7 heavy atoms , in contrast to most of the modifications of the lower side chain , does not lead to a substantial loss in cytotoxicity . It remains to be seen how the removal of hydroxyl groups from the upper side chain or its overall shortening would affect potency . It has been proposed that mycolactones enter mammalian cells via passive diffusion and interact with cytosolic target ( s ) [39] . Reduced or abolished activity of structural variants of mycolactone may thus be related to lack of binding to target structure ( s ) , inefficient triggering of activation pathways or reduced translocation across the cell membrane . Studies using isotopically labeled rather than fluorescence labeled structures with altered biophysical properties are required to gain better insight into mechanisms that allow mycolactones to cross biological membranes . Our findings with the truncated mycolactone PG-120 shows that different biological effects of mycolactone can be dissociated by using structural variants . In line with these observations , sub-lethal doses of mycolactone A/B have been shown to alter trafficking and cytokine production of lymphocytes and macrophages [40] , [41] . It remains to be elucidated whether different pathways and target structures are involved in the triggering of the biological effects of mycolactone . Since a number of macrolides have antibiotic activity against a broad spectrum of bacteria it has been speculated that mycolactone secreted by M . ulcerans during active disease may prevent superinfection of BU wounds . However , synthetic mycolactone A/B showed no antimicrobial activity against any of the microorganisms tested here . In line with this observation , superinfection of Buruli ulcer lesions seems to be much more common than traditionally anticipated ( Yeboah-Manu et al . , personal communication ) . Much has still to be learnt about the biophysical properties of mycolactones , their distribution and stability in biological systems , their target structures and triggering pathways in mammalian cells . Synthetic natural mycolactones , isotopically labeled derivatives and structural variants represent valuable tools to address these open questions in future . | Buruli ulcer is a chronic necrotizing skin disease caused by Mycobacterium ulcerans . The characteristic histopathological features of Buruli ulcer , severe destruction of subcutaneous tissue with minimal inflammation in the core of the lesion , are primarily attributed to the cytotoxic activity of mycolactone , the macrolide exotoxin of M . ulcerans . Different geographical lineages of M . ulcerans produce different structural variants of mycolactone . By using highly defined synthetic mycolactones , including both naturally occurring molecular species and additional non-natural variants , we have assessed the influence of the structure of the C-linked upper side chain and the lower C5-O-linked polyunsaturated acyl side chain on biological activity . Changes in the lower side chain affected the cytotoxic activity against mammalian cells more profoundly than changes in the upper side chain . Mycolactone A/B had no antimicrobial activity against Gram-positive and Gram-negative bacteria and was also inactive against Saccharomyces and Dictyostelium . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"biochemistry",
"biology",
"microbiology",
"molecular",
"cell",
"biology"
]
| 2013 | Structure-Activity Relationship Studies on the Macrolide Exotoxin Mycolactone of Mycobacterium ulcerans |
Adenosine-to-inosine modification of RNA molecules ( A-to-I RNA editing ) is an important mechanism that increases transciptome diversity . It occurs when a genomically encoded adenosine ( A ) is converted to an inosine ( I ) by ADAR proteins . Sequencing reactions read inosine as guanosine ( G ) ; therefore , current methods to detect A-to-I editing sites align RNA sequences to their corresponding DNA regions and identify A-to-G mismatches . However , such methods perform poorly on RNAs that underwent extensive editing ( “ultra”-editing ) , as the large number of mismatches obscures the genomic origin of these RNAs . Therefore , only a few anecdotal ultra-edited RNAs have been discovered so far . Here we introduce and apply a novel computational method to identify ultra-edited RNAs . We detected 760 ESTs containing 15 , 646 editing sites ( more than 20 sites per EST , on average ) , of which 13 , 668 are novel . Ultra-edited RNAs exhibit the known sequence motif of ADARs and tend to localize in sense strand Alu elements . Compared to sites of mild editing , ultra-editing occurs primarily in Alu-rich regions , where potential base pairing with neighboring , inverted Alus creates particularly long double-stranded RNA structures . Ultra-editing sites are underrepresented in old Alu subfamilies , tend to be non-conserved , and avoid exons , suggesting that ultra-editing is usually deleterious . A possible biological function of ultra-editing could be mediated by non-canonical splicing and cleavage of the RNA near the editing sites .
Post-transcriptional modification of RNA molecules increases the complexity of the transcriptome and constitutes an additional mechanism for controlling gene activity . One of the most frequent modifications in primates is Adenosine-to-Inosine ( A-to-I ) RNA editing of pre-mRNA . Since inosine is later translated as guanosine ( G ) , A-to-I editing can lead to recoding of protein sequences . A-to-I editing , mediated by adenosine deamisnase proteins acting on double-stranded RNA ( ADARs ) [1]–[4] , is crucial for normal life and development [5] , [6] and was found to play a role in human disease , especially brain related [7] , [8] . Editing affects gene expression , both globally and in a gene-specific manner [9]–[14] , and enhances the cell's capacity of information processing and evolvability [15] , [16] . Inosine is recognized as guanosine also during sequencing; editing can therefore be detected as a G in an RNA sequence with an A in the corresponding genomic DNA . Systematic surveys of cDNA and EST libraries [17]–[25] , as well as experimental genome-wide screens [26]–[29] , have so far detected about 40 , 000 human editing sites [30] . Known A-to-I editing sites can be roughly classified into two categories . In the first type , specific sites are edited in coding sequences . This type of editing usually modifies a protein sequence and potentially its function , and is therefore highly selective: in each gene , only one or few , specific , usually conserved sites are edited , in a regulated manner . Only few tens of such editing sites are currently known [2] . In the second category , which encompasses the bulk of the sites , adenosines at repetitive elements are indiscriminately hyper-edited , mostly in Alu elements [31] in UTRs or introns [17]–[21] . Due to the large number of Alu repeats in the human genome , adjacent , reversely oriented Alus can form double stranded RNA ( dsRNA ) structures that serve as targets for ADAR proteins . Editing of repetitive elements is highly promiscuous and ranges between a few to tens of nucleotides . The biological role of hyper-editing is mostly elusive . However , a few functions were proposed . For example , a hyper-edited RNA was shown to be retained in the nucleus [10] and to be released upon cleavage [14] . Inosine-containing synthetic dsRNAs were shown to be cleaved at specific sequences [32] , to globally down-regulate gene expression [13] , and to suppress apoptosis [33] . Changes in the RNA sequence , even if outside coding sequences , can also be functional , if , for example , they occur at splice sites [34] , [35] or at miRNA targets [36] . A particularly interesting class of hyper-edited RNAs , which we refer to here as ‘ultra’-edited RNAs , represents molecules that underwent editing of an extremely large fraction of their adenosines ( for a precise definition see Materials and Methods ) . Although it is known that long synthetic dsRNAs are ultra-edited [37]–[39] , not much is known about such endogenous RNAs— except for a small number of ultra-edited RNAs that were occasionally discovered ( e . g . , in [17] , [19] , [40]–[42] ) , ultra-editing was usually overlooked in systematic RNA editing detection screens . These methods work by aligning candidate RNA sequences to the reference genome and searching for clusters of A-to-G mismatches . However , for extensively edited RNAs , the alignment to the genome suffers from so many mismatches that the RNA is likely to be discarded . Based on this observation , on the preliminary evidence for ultra-edited RNAs , and on the large amount of cellular inosine [43] , we suspected that many more ultra-edited RNAs exist . In this paper , we devised and applied a computational pipeline to identify ultra-edited RNA . We started with RNA sequences that previously could not be aligned to the genome , and realigned them after reducing the genomic DNA and RNA sequences to three letters by an A→G transformation . This way , mismatches in ultra-edited RNAs due to A-to-I editing were masked and fast alignment algorithms could be employed to detect the genomic origins of these RNAs . Whenever a transformed RNA has successfully aligned to the transformed genome , the original sequences were recovered and the mismatches were examined . A particularly large number and density of A-to-G mismatches indicated that the RNA was ultra-edited . We detected , with high confidence , 760 ultra-edited RNAs edited in over 14 , 000 editing sites , most of which were previously unknown . Comparison of the ultra-edited elements with sites of moderate editing suggested that , as expected , ultra-editing is preferred in repeat-rich regions with potential for particularly long fold-back dsRNA structure .
We queried the UCSC Genome Browser [44] ( http://genome . ucsc . edu ) for long ( >250 bp ) human ESTs or mRNAs from GenBank that did not align to the genome , and downloaded their sequences from NCBI Batch Entrez ( http://www . ncbi . nlm . nih . gov/sites/batchentrez ) . The 458 , 124 sequences were filtered to discard possible low-quality sequences: ESTs or mRNAs with particularly large ( >60% ) or small ( <10% ) percentage of a single nucleotide , with over 10% of ambivalent nucleotides ( non-[ACGT] ) , or with over 50% simple repeats content . We also aligned ( MEGABLAST [45]; http://www . ncbi . nlm . nih . gov/blast/megablast . shtml ) the remaining 438 , 807 sequences to the genome ( GRCh37/hg19 ) and eliminated each sequence that aligned with ≥98% identity ( along ≥90% of its length ) . The remaining 334 , 344 candidate sequences were sent to downstream analysis . Since the number of full-length mRNAs was relatively small ( ∼2% ) , we refer henceforth to our candidate sequences as ESTs , or just RNAs , interchangeably . A-to-I ultra-edited RNAs harbor a large number of A-to-G mismatches ( A in the DNA , G in the RNA ) , but no ( or very few ) mismatches of any other type . Therefore , an ultra-edited RNA would generate a good alignment to the genome ( and therefore be detected ) if A-to-G mismatches will be specifically ‘masked’ . To this end , we transformed every A to G both in the genomic DNA sequence and in the candidate RNA sequences . As demonstrated in Figure 1B , ultra-edited , high-quality , transformed RNA sequences will align perfectly to the transformed DNA . Low-quality , erroneous RNA sequences will not align well even after the transformation . A-to-I editing always takes place on the sense strand . However , the actual sequenced DNA and RNA strands are arbitrary . Therefore , to detect all ultra-edited RNAs , all strand combinations must be separately aligned ( DNA+/RNA+ , DNA+/RNA− , DNA−/RNA+ , DNA−/RNA−; see Table S1 ) . For genuine ultra-edited RNA , exactly one strand combination will produce a good alignment after the transformation . Note that additional information on transcription direction ( e . g . , a polyA tail , protein sequence , splicing signals , etc . ) is required to rule out the possibility that the A-to-G mismatches are due to a T-to-C editing event ( see also Table S1 ) . With A→G transformation , we detect clusters of A-to-G mismatches , but also clusters of G-to-A . The G-to-A clusters serve as a negative control , because we expect such clusters to result from a sequencing error . The same holds true for other types of mismatches; we therefore created additional transformations: A→C ( ×4 strand combinations ) , G→C ( ×2 ) , and A→T ( ×2 ) . For G→C and A→T , it is sufficient to align the ( + ) DNA to the ( +/− ) RNA , as the other two combinations ( with ( − ) DNA ) are equivalent to the first two . The 12 transformations are summarized in Table S1 . To speed up the computation of the alignments , we uploaded the candidate RNA sequences and the human genome to a commercial cloud computer ( http://aws . amazon . com/ec2 ) . We performed the transformations listed above and aligned , in parallel , the 12 transformed RNA and DNA pairs using MEGABLAST [45] . We retained only the best alignment , and only when it was particularly convincing ( E-value≤10−50 , percent identity≥95% , length≥100 bp ) . The number of successful alignments was 690 , 495 , ∼17% of the number of possible alignments ( 334 , 344 candidate sequences ×12 transformation/strand combinations ) . For each aligning RNA and DNA pair , we realigned the original , 4-letter sequences and recorded all mismatches . Consider , for example , alignments coming from the A→G transformed sequences . We designated an RNA as ultra-edited if it satisfied the following conditions: A similar procedure was used to search for RNAs with other possible types of ‘editing’ ( G-to-A , A-to-C , etc . ) . The values of the cutoffs were chosen to roughly match the expected number of mismatches in an EST aligning to an Alu element that was discarded by UCSC ( 4% dissimilarity×300 bp Alu length = 12 mismatches , which are ∼20% of the ∼60 adenosines in the consensus Alu ( Repbase [46] ) ) . However , as there is no clear-cut boundary between ultra-edited RNAs and other edited RNAs , other values could have been selected as well . RNAs passing the above criteria were further filtered to remove the following cases , where apparent editing is likely an artifact . In total , 760 RNA sequences containing 14 , 538 unique editing sites survived the cleanup procedure to constitute our final set of A-to-I ultra-edited RNAs . A complete list of the ultra-edited RNAs , along with some of their properties ( e . g . , GenBank accession , genomic coordinates , location of mismatches , sequence context , etc . ) , can be found in Dataset S1 . A list of the ultra-editing sites formatted as a UCSC genome browser track is given in Dataset S2 . Clearly , the pool of ESTs we analyzed contains many RNAs which are hyper-edited , even if not ultra-edited according to our strict definition . Rerunning our screen exactly as above , but allowing for less than 12 editing sites ( but at least five ) , we discovered 280 additional ESTs containing 2 , 286 unique editing sites . Although a detailed analysis of these ESTs is beyond the scope of this paper , we report their coordinates and basic individual and genome-wide properties in Dataset S3 , Dataset S4 and Text S1 .
We speculated that published cDNA sequences that could not be confidently aligned to the genome include some ultra-edited RNAs . We therefore extracted , from the UCSC genome browser , ∼450 , 000 ESTs whose genomic origin could not be confidently established , from which we removed ∼100 , 000 sequences with potential sequencing errors ( e . g . , long single-nucleotide stretches ) . We masked A-to-I editing sites by transforming every A to G in the RNA sequences and in the genome , and then aligned the transformed RNA and DNA sequences using MEGABLAST . We repeated the transformation and alignment for all possible strand combinations , and for other types of possible ‘editing’ ( e . g . , A-to-C ) as a control ( see Table S1 ) . For ESTs for which a good alignment was found , the original ( non-transformed ) sequences were recovered and the mismatches were examined . We designated an EST as ultra-edited if the number of A-to-G mismatches was at least 12 , and at least 90% of all mismatches , and if the fraction of edited adenosines to all adenosines was at least 20% . Finally , we discarded seemingly ultra-edited RNAs whose alignment was suspicious ( e . g . , too many gaps , high repeat content , strand ambiguous or inconsistent with other ESTs , etc . ) . More details on the computational procedure appear in Materials and Methods ( see also Figure 1 ) . At the termination of the computational pipeline , we remained with a final set of 760 ( A-to-G ) ultra-edited ESTs . Four typical cases of ultra-edited ESTs are presented in Figure 2 . The distributions of the number of editing sites and the editing rates ( fraction of edited As/number of As ) are shown in Figure 3 . Additional 280 ESTs were found when we allowed for a smaller number of editing sites in each EST . These ESTs are reported and analyzed in Dataset S3 , Dataset S4 and Text S1 , but are not further discussed here . The number of ultra-edited ESTs of each type of mismatch is shown in Figure 4A . The number of ultra-edited ESTs of type A-to-G is more than five times the number of ‘edited’ ESTs of all other types combined ( 760 vs . 138 ) . The largest class of non-A-to-G editing is G-to-A , containing 75 ESTs . To explain the origin of these 75 ESTs , we plot in Figure 4B the number of ESTs in which the edited RNA strand was ( + ) ( the sequenced strand ) or ( − ) . Since ESTs are derived from double-stranded cDNA clones , the strand that was sequenced is usually arbitrary ( relative to the sense strand ) , and we expect to see roughly equal numbers of ( + ) and ( − ) ESTs . However , as can be seen in Figure 4B , all but one of the G-to-A ultra-edited ESTs are from the ( + ) strand . This indicates that the source of these mismatches is possibly a technical sequencing error [21] . In support of this hypothesis , we note that the vast majority ( 63/75 ) of the G-to-A ESTs came from NCI-CGAP ( National Cancer Institute – Cancer Genome Anatomy Project ) libraries , as opposed to just 99/760 for A-to-G . Additionally , 65/75 of the ESTs were sequenced in the year 1997 , compared to only 114/760 for A-to-G . It is thus conceivable that most of the G-to-A clusters are due to isolated cases of technical faults . The total number of A-to-G editing sites discovered by our screen is 15 , 646 , of which 14 , 538 are unique . This the same order of magnitude as discovered in former editing screens [17]–[19] . Almost all sites ( 13 , 668 , 94% ) are novel: they did not appear in DARNED [30] , the most up to date database of RNA editing in humans . The 760 ultra-edited ESTs map to 695 distinct genomic regions , 647 of which are covered by one ultra-edited EST , 41 by two ESTs , and one ( chr3:183879216–183879642+ , intron of DVL3 gene ) by 11 ESTs ( all from the lung EN0096 library ) . Only 42 sites ( 0 . 29% ) overlap with genomic SNPs . Figure 5 shows the frequency of nucleotides upstream and downstream of the editing sites , as well as the frequencies of their combinations . The sequence preference of all previously known editing sites ( as listed in DARNED ) is also presented . As expected [17]–[20] , [26] , [27] , guanosines are depleted upstream and overrepresented downstream of the editing sites . The frequencies of the other nucleotides differ slightly between ultra-editing and DARNED , particularly for upstream As and Ts . Comparison of all dinucleotide combinations between the ultra-editing sites and the DARNED sites reveals that ultra-editing is relatively more common than DARNED at AAA , GAA , and GAG ( the middle A is the editing site ) and is less common than DARNED at CAC , AAG , and TAG . The latter two are ADAR2 motifs [39] , suggesting that ultra-editing is mediated mostly by ADAR1 . We next characterized the conditions under which ultra-editing has occurred . A list of the ultra-edited tissues and health states , sorted by the number of edited ESTs , is given in Table 1 . The most surprising observation is the large amount of ultra-edited ESTs in the liver . Further investigation revealed that 305 of these ESTs are from a single library named “Human liver regeneration after partial hepatectomy” ( Library ID:18893 ) . We believe that these ESTs represent bona fide A-to-I editing events for the following reasons . First , the fraction of ESTs not aligning to the genome ( http://genome . ucsc . edu/ ) in the liver library is neither exceptional nor even the largest . The fraction of non-aligning ESTs that are ultra-edited is also not the largest . Next , the sequence context of the liver ultra-editing sites is the one expected from ADAR targets , namely , a deficit of G upstream and an excess of G downstream of the editing site . Finally , all but seven of the liver ultra-edited ESTs overlap with an Alu element . We thus speculate that the ultra-edited liver library has been generated under experimental conditions of ADAR overexpression , perhaps due to induction by interferon [47] . Of the other tissues , brain is the most ultra-edited , followed by lung , thymus , and eye . In Table 1 , we also report the enrichment factor of each tissue , that is , the number of ultra-edited ESTs in the tissue divided by the expected number . The tissues most enriched are thymus , spleen , muscle , and brain . Ultra-editing in cancer tissues is infrequent [48] . As expected , almost all ultra-edited RNAs overlapped with an Alu element ( 693/760 ) , and only six did not overlap with any repeat . An important question raised by our finding of ultra-edited RNAs is whether these RNAs have any distinct properties . To address this question , we compiled , using DARNED , a list of all previously known A-to-I editing clusters that are not ultra-edited , by grouping adjacent editing sites ( separated by less than 300 bp , the Alu length ) and eliminating clusters with a single site or with 12 or more sites . This resulted in a set of 4456 “short clusters” to which we compared our ultra-edited ESTs . In Table 2 , we report the fraction of edited RNAs originating from each major Alu sub-family ( AluJ , AluS , and AluY ) . Most notably , ultra-edited ESTs are underrepresented in AluJ elements ( P<10−14 , χ2-test comparing AluJ elements to all others ) . In comparison , the number of DARNED's short clusters found in AluJ elements is roughly what is expected based on the genome-wide distribution of these elements ( P = 0 . 64; χ2-test ) . As AluJ is the oldest Alu sub-family , these results suggest that ultra-editing sites were eliminated from relatively old Alu sub-families . The strand of an Alu element within a transcript can be either sense or antisense . We found that ultra-edited Alu elements have a clear strand preference: 630 ultra-edited Alus are sense ( 77% ) , compared to only 186 antisense ( 23% ) . In DARNED's short clusters , there is almost no strand preference: 2382 sense ( 53% ) vs . 2141 antisense ( 47% ) . The explanation of this result is likely the composition bias of the Alu elements: even without the terminal polyA tail , the consensus sense strand Alu ( Repbase [46] ) has 59 As compared to only 46 Ts . We speculated that ultra-editing occurs at particularly long or stable dsRNA structure [37] , [38] , [40] , [49] . We therefore calculated the maximum possible length of dsRNA structure in the edited regions . We used two measures: the total number of matching base pairs when aligning the edited region and its reverse complement , and the maximal length of the stem in the RNA secondary structure , as predicted by RNA Fold [50] . Indeed , the putative dsRNA length is significantly longer , according to both measures , in the ultra-edited regions than in DARNED's short clusters ( Table 3 , properties 1 , 2 ) . The reason for the increased dsRNA length is likely the dramatic overabundance of repeats in the ultra-edited flanking regions ( Table 3 , property 3 ) . Specifically , the ultra-edited regions have a larger number of inverted pairs of Alu repeats than the short clusters ( Table 3 , property 4 ) , and a smaller distance between the edited Alu and the nearest inverted Alu ( Table 3 , property 5 ) . Most ultra-edited RNAs overlap with genes ( 547/760 ESTs ( 72% ) ; the overlap is with 460 genes; gene annotation is from the UCSC genome browser ) . Among these , 61 ( 8% ) overlap with exons: 38 with 3′UTRs , four with 5′UTRs , 17 with non-coding RNA , and two with coding sequences ( DW412140 with GSK3B and DA857874 with OLR1 ) . The other 486 ESTs overlap with introns . The higher level of editing in 3′UTRs compared to 5′UTRs , which has been previously observed [17] , [20] and is also observed in the DANRED database , is most probably due to their larger sizes ( mean ∼525 bp , compared to ∼145 bp for the 5′UTR [51] ) . DARNED's short clusters have only slightly larger overlap with genes ( 75% ( 3359/4456 ) ; P = 0 . 02 , binomial test ) , but significantly larger overlap with exons ( 1239/4456 ( 28% ) ; P = 10−42; binomial test ) . A list of the ultra-edited ESTs overlapping with exons is given in Dataset S5 . A functional classification of the ultra-edited genes appears in Dataset S6 . Among the ultra-edited genes , 19 are related to stress response , 14 to apoptosis , and three to hematopoiesis ( also listed in Dataset S6 ) , which could be related to the known role of ADAR1 in these processes [5] , [52]–[56] . Hyper-edited RNAs can be specifically cleaved [32] , [57] , [58] , and hundreds of putative hyper-editing sites were shown to be non-canonically ( NC ) spliced out of UTRs [59] . To find out if ultra-edited regions are also cleaved or NC-spliced , we searched for ultra-edited regions that overlap with both a UTR and an intron . We found 31 such ESTs , listed with their genes in Dataset S7 . We manually inspected the splice variants of these genes to identify cleavage or NC-splicing . Cleaved RNAs appear as properly spliced sequences , up to a certain point where an exon extends abnormally until it is cleaved at the ultra-edited region . NC-spliced RNAs also appear to be normally spliced , except for an additional short intron in their 3′UTR , whose boundaries overlap with the ultra-edited Alus but lack the GT-AG canonical splicing signals . We identified ten cleaved and five NC-spliced mRNAs in regions of ultra-editing ( indicated in Dataset S7 ) , including one that was previously shown [59] . We note that few of the cleavage sites may be alternatively explained as premature polyadenylation at the Alu sequence [60] , [61] . Ultra-editing substrates are more abundant in introns and in new Alu sub-families than the short clusters , indicating their general adverse effect . We hypothesized that ultra-edited genomic regions are also less conserved . Therefore , we extracted for each edited region ( with flanking 500 bp upstream and downstream ) , the average primate PhyloP [62] conservation score , which is a measure of the acceleration or reduction of the rate of nucleotide substitution . The ultra-edited regions are less conserved ( average score 8*10−3 ( ±2*10−3 standard error of the mean ) ) compared to the short clusters ( ( 15±1 ) *10−3; P = 0 . 004 ( t-test ) ) . We note though that when comparing an alternative conservation score ( PhastCons [63] , which is the probability the entire region is conserved ) , no difference is observed between ultra-editing sites and short clusters . We selected two ultra-edited RNAs , for which no editing was known before , for experimental validation . The first EST , DA098819 , was derived from an AluSx element in the intron of the ZNF83 gene ( chr19:53120521–53121009− ) . It was generated from a normal brain and had 34 A-to-G mismatches . The second EST , DA364252 , came from an AluSq element in the intron of ING5 ( chr2:242643522–242644012+ ) . It was also generated from a normal brain and had 25 mismatches . We amplified genomic DNA and cDNA from a brain of a single donor for these two targets ( details on experimental procedures are given in Text S2 ) . The genomic DNA was sequenced , and the cDNA PCR product was cloned . We selected and sequenced several clones ( 14 for DA098819 , 13 for DA364252 ) and searched for A-to-G mismatches when compared to the genomic DNA . For DA098819 , the average number of A-to-G mismatches per clone was 19 , with the most heavily edited clone having 36 mismatches . The total number of editing sites we found ( over all clones ) was 45; these sites cover 33/34 of the sites seen in the EST . For DA364252 , the average number of sites was 14 , with 22 sites in the most edited clone . Over all clones , 38 sites were found , covering 19/25 of the sites of the EST . A histogram of the number of clones with each number of editing sites for the two targets is presented in Figure 6 . The alignment of the clones to the genomic DNA , annotation of the editing sites , and additional statistics appear in Dataset S8 .
Previous screens to detect RNA editing systematically overlooked RNA sequences that poorly aligned to the genome . We conjectured that many of these sequences are in fact highly edited and therefore attempted to realign them . To improve the chances of obtaining a successful alignment , we masked the A-to-I editing sites by an A→G transformation . Indeed , we discovered more than 700 ESTs ultra-edited in over 14 , 000 sites , which is about a third of the number of currently known editing sites . We deposited the coordinates of our sites in DARNED , the database of RNA editing . We also experimentally validated two of the targets . As many apparent editing sites could really be sequencing errors , we applied stringent cutoffs and various cleaning procedures to ensure the sites we detected are genuine . The high confidence we have in our ultra-edited RNAs stems from the extremely small number of mismatch clusters of types other than A-to-G , because if our sites had resulted from a sequencing error , we would have observed a similar number of mismatch clusters of all types ( or at least transitions ) . More evidence for the authenticity of the ultra-edited RNAs comes from their sequence motif , which is typical to editing by ADAR , and the localization of the editing sites in Alu elements . We believe that with relaxation of some of our strict detection thresholds , even more sites will be detected . Characterization of the ultra-edited ESTs revealed that with the exception of a single liver library , the most edited tissue is the brain . However , this is to some extent because of the high coverage of the brain transcriptome; in terms of enrichment , the thymus , spleen , and muscle tissues are more ultra-edited , in agreement with previous observations [17]–[19] . Muscle tissue is ultra-edited in a couple of libraries despite the low expression of ADARs in that tissue [43] , [64] , [65] . Ultra-editing in muscle could thus be a result of induction of ADAR1 , perhaps due to stress , as observed in [54] . The extreme number of ultra-edited RNAs from a regenerating liver library may also indicate induction of ADAR1 due to stress , possibly a viral infection [8] . However , the precise reason for ADAR's extreme hyperactivity in that sample remains to be elucidated . The biological function of ultra-editing is still cryptic . Some of our findings ( weak degree of sequence conservation , localization in new Alu subfamilies and in introns ) may suggest that ultra-editing is generally undesirable , and that its major effect , if any , is gene-independent . In the latter case , the large amount of inosines in the transcriptome could affect gene expression globally , as recently shown [13] , [33] . The other option is that ultra-editing affects the expression of specific genes . This could be mediated by modification of the RNA secondary structure ( dsRNA destabilization ) , RNA nuclear retention , and cleavage/non-canonical splicing at the edited nucleotides . We demonstrated possible instances of the latter mechanism . The direct sequence changes induced by editing ( A-to-G ) do not seem to have an important function , in agreement with the large variation in the usage of editing sites that we experimentally observed ( see Dataset S8 ) . We did however find one ultra-edited RNA with five editing sites in a protein coding region ( OLR1 ) , four of which are non-synonymous . If more coding sequences are similarly ultra-edited , this could serve as an extremely powerful mechanism that ( reversibly ) diversifies protein sequences . Specific ultra-edited genes of interest are 17 genes involved in apoptosis and hematopoiesis , because of the role of ADAR1 in these processes [5] , [52] , [53] , [56] . Regardless of the function of ultra-editing , the edited regions are characterized by potential to create particularly long , stable dsRNA structure , as expected from experiments with synthetic dsRNA [37] , [38] . The stability of the dsRNA seems to be facilitated by a large frequency of repetitive elements , Alu and others , near the editing sites . It could however be that the editing efficiency is also affected by other factors , yet to be discovered . Finally , our findings raise the intriguing question of how rare ultra-editing is . We detected a number of ultra-edited RNAs of the same order of magnitude as in previous genome-wide screens; as each ultra-edited RNA accommodates , by definition , a large number of sites , it could be that ultra-editing is responsible for a significant fraction of the cellular inosines . On the other hand , ultra-editing could be incidental , occurring sporadically in a stochastic manner . To decisively resolve this issue , editing must be studied in a transcriptome covered in depth . However , current technology and computational methods permit such studies only in small-scale [26] , [66] , [67] . We tend to adopt the view that ultra-editing is rare , for the following reasons . First , only 0 . 4% ( 3/695 ) of the ultra-edited regions are covered by four or more ESTs , compared to 10 . 6% ( 173/1637 ) in a previous genome-wide screen [17] , [41] . Second , only 2/27 clones in our study , and 3/69 clones in [41] , are far more edited than other clones . Third , Alu editing is , to a good approximation , a stochastic process where each site is edited independently with a given rate [41] , [66] . Under this model , the probability to encounter an ultra-edited RNA is exponentially small . In the ultra-edited RNAs that we discovered , the editing rate was probably sufficiently large ( due to e . g . , particularly long dsRNA structure , specific induction of ADAR1 , etc . ) that ultra-editing was visible even with the current shallow coverage . | The traditional view of mRNA as a pure intermediate between DNA and protein has changed in the last decades since the discovery of numerous RNA processing pathways . A frequent RNA modification is A-to-I editing , or the conversion of adenosine ( A ) to inosine ( I ) . Since inosine is read as a guanosine ( G ) , A-to-I editing leads to changes in the RNA sequence that can alter the function of its encoded protein . In recent years , tens of thousands of human A-to-I editing sites were discovered by computationally comparing RNA sequences to the human genome and searching for A-to-G mismatches . However , previous screens usually ignored RNA sequences that were edited to extreme , because the large number of A-to-G mismatches carried by these RNAs obscured their genomic origin . We developed a new computational framework to detect extreme A-to-I editing , or ultra-editing , based on masking potential editing sites before the alignment to the genome . Our method detected about 14 , 000 editing sites , with each edited molecule affected , on average , in more than 20 nucleotides . We demonstrated that the likely reason for the ultra-editing of those sequences is their potential to fold back into a particularly long double-stranded structure , which is the preferred target of the editing enzymes . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"sequence",
"analysis",
"genome",
"expression",
"analysis",
"genome",
"analysis",
"tools",
"genomics",
"rna",
"processing",
"gene",
"expression",
"genetics",
"biology",
"computational",
"biology",
"genetics",
"and",
"genomics",
"transcriptomes"
]
| 2011 | Identification of Widespread Ultra-Edited Human RNAs |
Intercellular communication in adjacent cell layers determines cell fate and polarity , thus orchestrating tissue specification and differentiation . Here we use the maize stomatal apparatus as a model to investigate cell fate determination . Mutations in ZmBZU2 ( bizui2 , bzu2 ) confer a complete absence of subsidiary cells ( SCs ) and normal guard cells ( GCs ) , leading to failure of formation of mature stomatal complexes . Nuclear polarization and actin accumulation at the interface between subsidiary mother cells ( SMCs ) and guard mother cells ( GMCs ) , an essential pre-requisite for asymmetric cell division , did not occur in Zmbzu2 mutants . ZmBZU2 encodes a basic helix-loop-helix ( bHLH ) transcription factor , which is an ortholog of AtMUTE in Arabidopsis ( BZU2/ZmMUTE ) . We found that a number of genes implicated in stomatal development are transcriptionally regulated by BZU2/ZmMUTE . In particular , BZU2/ZmMUTE directly binds to the promoters of PAN1 and PAN2 , two early regulators of protodermal cell fate and SMC polarization , consistent with the low levels of transcription of these genes observed in bzu2-1 mutants . BZU2/ZmMUTE has the cell-to-cell mobility characteristic similar to that of BdMUTE in Brachypodium distachyon . Unexpectedly , BZU2/ZmMUTE is expressed in GMC from the asymmetric division stage to the GMC division stage , and especially in the SMC establishment stage . Taken together , these data imply that BZU2/ZmMUTE is required for early events in SMC polarization and differentiation as well as for the last symmetrical division of GMCs to produce the two GCs , and is a master determinant of the cell fate of its neighbors through cell-to-cell communication .
The development of the stomatal complex in maize ( Zea mays ) , consisting of a pair of dumbbell-shaped guard cells ( GCs ) flanked by two subsidiary cells ( SCs ) , provides an excellent model system to study the signals controlling fate determination of adjacent cells by intercellular communication . Unlike the two kidney-shaped GCs in eudicots , the four-celled stomatal complex in grasses may facilitate a faster response to environmental cues in order to optimize photosynthesis and water use [1 , 2] . Highly differentiated GCs and SCs in grasses are generated by two asymmetric divisions [3] . The first asymmetric division generates a guard mother cell ( GMC ) , which produces the extrinsic cues conveyed to the laterally adjacent subsidiary mother cells ( SMCs ) . Upon perception of the signal , SMCs become spatially polarized with respect to the GMC , by localized accumulation of cortical F-actin at the SMC/GMC interface and migration of the pre-meiotic SMC nuclei to each interface [4] . The extrinsic cue eventually triggers an asymmetrical division of the SMCs , in an orientation that positions the smaller daughter cell , the SC , adjacent to the GMC [5 , 6] . Finally , the GMC produces the two GCs by symmetrical division . Several intrinsic factors that specify cell fate during the development of GCs have been identified . Moreover , an extrinsic signaling cascade has been dissected in the control of the number and plane of asymmetric divisions [7] . In Arabidopsis , three of basic helix-loop-helix ( bHLH ) transcription factors , SPEECHLESS ( SPCH ) , MUTE and FAMA sequentially modulate the initiation , transition and differentiation events , respectively , regulating the cell lineage involved in the production of stomata [8–13] . The earliest acting bHLH protein , SPCH , functions in the transition from protoderm cells ( PDCs ) to meristemoid mother cells ( MMCs ) [10 , 12] . Inhibition of SPCH function predominantly affects MMCs and meristemoids and their ability to divide asymmetrically [9 , 10 , 14] , suggesting that SPCH is necessary for the asymmetric divisions that establish the stomatal lineage . The second bHLH protein , MUTE , which is required to terminate asymmetric division and stimulate transition from meristemoid to GMC identity , upregulates a series of cell-cycle genes to drive symmetric division of stomata [12 , 15 , 16] . Meristemoids of mute mutants undergo excessive amplifying divisions and fail to transition to a GMC [12] . Interruption of cell to cell signaling can instigate an extension or reduction of meristemoid division , consistent with the observation that epidermal cell number , including stomata , is a plastic trait that is monitored and adjusted based on internal and external cues [8 , 12 , 17 , 18] . The third bHLH protein , FAMA , controls the final fate transition from GMC to GC , consisting of two distinguishable events: symmetric cell division of the GMC and GC transition [11] . Recently , stomatal initiation in the grass Brachypodium distachyon has been shown to involve the orthologs of Arabidopsis stomatal bHLH regulators . However , it appears that the function and behavior of the individual components and their interacting regulatory networks have diverged , acquiring specific functions for the unique stomatal development of grasses [19 , 20] . BdMUTE , an ortholog of Arabidopsis MUTE in B . distachyon , is expressed in GMCs before moving to neighboring cells , and this mobility is a protein-intrinsic feature [21] . In maize , a leucine-rich receptor-like kinase ( LRR-RLKs ) , PANGLOSS1 ( PAN1 ) , has been found to participate in transmission of the signals for establishing polarity in SMCs [22] . About 40% of SMCs in loss-of-function pan1 mutants had unpolarized nuclei and developed defective stomatal complexes , resulting from abnormal asymmetric cell division and SC patterning [22 , 23] . Interestingly , PAN2 is required for the subsequent polarization of PAN1 and the Rho family GTPases ( ROPs ) [24 , 25] . After PAN2 polarization , PAN1 and ROP2/9 are polarized , an actin patch forms at the SMC/GMC contact sites , and the SMC nuclei migrate to this site in an actin-dependent manner . pan2 , pan1 , rop2 and rop9 mutants have abnormal SCs , and their SMCs are defective in actin patch formation and nuclear migration [26 , 27] . It has been proposed that the GMC may send a cue to the neighboring cells inducing the asymmetric cell division of the SMCs at adjacent sites to produce the SCs [21 , 22] . However , the molecular mechanisms by which GMCs control polarization and cell fate of SMCs in maize remain unclear . Here we identify BZU2/ZmMUTE ( Bizui2 , abbreviated to BZU2 ) gene , encoding a bHLH transcription factor , an ortholog of Arabidopsis MUTE protein . Loss-of-function BZU2/ZmMUTE mutants show no SCs and are defective in stomatal complexes . In addition , bzu2-1 displays defects in symmetric GMC division and lacks SMC polarization . We found that BZU2/ZmMUTE controls the expression of several essential genes involved in the formation of the stomatal complex , and binds to the promoters of PAN1 and PAN2 for amplifying the expression of the polarity program in SMCs . Our finding suggests that BZU2/ZmMUTE functions as an important player both acting to regulate the GMC to GC fate transition , and serving as an intercellular signal required for polarization and recruitment of the SMCs .
To isolate stomatal response deficient mutants , we produced an extensive collection of mutagenized maize lines . Pollen from the maize inbred line Mo17 was treated with ethyl methane sulfonate ( EMS ) and the M1 progeny were self-pollinated to produce M2 mutant seeds . We screened ~45 , 000 of these M2 at the seedling stage for drought-sensitivity , using a far infrared thermal imaging approach [28 , 29] . We identified a number of mutants displaying an unusually high leaf-surface temperature and designated these as bizui which means closed mouth ( abbreviated to bzu ) . One of these recessive mutants , viz . bzu2-1 , was studied in detail . It constitutively displays a hot leaf surface phenotype as compared to the wild-type ( Mo17 ) ( Fig 1A ) . In addition , the average leaf temperature of bzu2-1 mutants was ~1°C higher than wild-type , which reflects an inability of bzu2-1 leaves to appropriately regulate transpirational water loss from their surfaces ( Fig 1B ) . The bzu2-1 mutant produced pale , highly hydrated , translucent leaves , and the seedlings died about 14 days after germination ( Fig 1C ) . Upon examination of leaf structure , we observed that the normal mature four-celled stomatal complexes , comprising two dumb-bell shaped GCs adjacent to two SCs , were absent from bzu2-1 mutants ( Fig 1D ) . As a consequence , the water loss from the leaf surfaces of bzu2-1 mutants was significantly slower than that of the wild-type ( Fig 1E ) . We subsequently characterized chlorophyll fluorescence emission , comparing the maximum quantum yield ( Fv/Fm ) of wild-type and mutant plants . We found that leaves of the bzu2-1 mutant have a lower Fv/Fm value than that of wild-type ( Fig 1F and 1G ) , which implies that the photosynthetic activity is dramatically decreased in 8-day-old seedlings of the bzu2-1 mutant . These observations indicate that the bzu2 mutant lacks the normal physiological functions of stomata , and is consistent with the lethal seedling phenotype . As illustrated in Fig 2A , in the wild-type , protodermal cells ( PDC ) undergo an asymmetric division producing a smaller daughter cell , which is the precursor of an apical GMC , and a larger basal interstomatal cell . Following the differentiation and elongation of the GMCs , the two flanking cells acquire SMCs identities prior to the formation of the GCs . The SMCs then undergo another asymmetric division , producing the two SCs prior to a final division of the GMC , which divides symmetrically and longitudinally to produce the two GCs . The stomatal complexes with dumbbell-shaped GCs immediately adjacent to two triangular SCs are linearly separated by rectangular interstomatal cells . In bzu2-1 mutants , the early phase of stomatal development seems to be normal , with PDCs being able to produce GMCs . However , the longitudinal symmetric division of GMCs is altered in the bzu2-1 mutants , with asymmetric or irregular directional divisions occurred ( a transverse or diagonal longitudinal division ) , resulting in the production of short columns of elongated cells , which lack hallmarks of GC morphology ( Fig 2A ) . Subsequently , none of GMCs develop into normal stomatal complexes ( Fig 2B ) . Instead , in the bzu2-1 mutant , 48% , 27% and 25% of GMCs were further divided into 2- , 3- or 4-cell groups of undifferentiated cells , respectively ( Fig 2C ) . Our data imply that one function of BZU2 may be to suppress ectopic or premature guard mother cell divisions . Interestingly , in bzu2-1 mutants , GMCs also failed to induce flanking cells to form SMCs , resulting in an almost complete absence of SCs . Examination of 802 stomata in bzu2-1 mutants showed that ~95% of the abnormal GCs had no SCs , and ~5% remaining had one single abnormal SC ( S1 Fig ) . The lack of SCs suggests that absence of BZU2 may abrogate the acquisition of cell fate of SMC precursors adjacent to GMCs ( Fig 2A and S1 Fig ) . We further examined the spatial interactions between the GMCs and SMCs during stomatal development via light and fluorescence microscopy . In the wild-type , GMCs induce the formation of SMCs , which exhibit polarity as defined by the anisotropic positioning of the SMC nucleus with respect to the SMC/GMC contact surface . In this process , an important step in SMCs polarization is the formation of a dense patch of F-actin , which presumably mediates nuclear migration or anchoring during cell division [22] . In the wild-type , when actin-patches appeared at the SMC/GMC contact sites , the SMCs nucleus will migrate to the contact sites , and then the SMCs undergo one asymmetric division to form SCs ( Fig 3A ) . Our data demonstrated that 91 . 3% of the SMCs nuclei migrated to the SMC/GMC contact surface in the wild-type , the cells therefore being polarized and highly anisotropic . In contrast , 84 . 4% of the SMCs examined in bzu2-1 mutants were non-polarized , with the nucleus being centrally located ( Fig 3A and 3B ) . These results show that BZU2 either acts as an intrinsic signal in SMC development , or controls the expression of intrinsic factors that determine SMC polarity and , ultimately , cell fate . In order to characterize BZU2 at the molecular level , we crossed the bzu2-1 mutant ( carried in the Mo17 background ) to the inbred line B73 , thereby generating reciprocal F1 hybrid progeny . The F2 population resulting from F1 self-crossing was screened based on the bzu2-1 mutant phenotype ( S2 Fig ) . Initial mapping indicated that the BZU2 locus co-segregated with the simple sequence repeat ( SSR ) markers bnlg1863 ( recombination rate 2 . 1% ) and umc1858 ( recombination rate 12 . 5% ) in Bin 8 . 03 of chromosome VIII . Several rounds of fine-mapping narrowed this locus to a 0 . 69 Mb region between SSR markers 79M15 ( 79 . 01 M ) and 79M45 ( 79 . 70 M ) containing fifteen genes ( Fig 4A ) . After sequencing , we identified the BZU2 gene as GRMZM2G417164 ( Fig 4B ) , which encodes a bHLH transcription factor exhibiting sequence similarity to Arabidopsis , rice and Brachypodium MUTE ( Fig 4D and S3 Fig ) . bzu2-1 contains a 4 nucleotide ‘AGCT’ insertion at the position 390 bp downstream of the start of transcription generating a premature STOP codon ( Fig 4B and 4C ) . The bzu2-1 mutant phenotype was further confirmed by targeted gene knockouts of BZU2/ZmMUTE using the CRISPR/Cas9 system . Stomatal phenotypes of three independent CRISPR/Cas9-mutated lines ( bzu2-2 , bzu2-3 and bzu2-4 ) are identical to bzu2-1 , confirming BZU2/ZmMUTE as GRMZM2G417164 ( Fig 4E and 4F and S4 Fig ) . When a GFP- BZU2/ZmMUTE fusion chimera was transiently expressed in tobacco ( Nicotiana tabacum L . ) , we were able to show accumulation of GFP-BZU2/ZmMUTE protein within the nucleus ( S5 Fig ) , which is consistent with the putative function of BZU2/ZmMUTE as a transcription factor . The interesting role of BZU2/ZmMUTE in controlling neighbor cell fate prompted us to further investigate the behavior of BZU2/ZmMUTE protein in the early developmental stages of SCs and GCs . In the transgenic reporter plants , the fluorescence of ZmMUTEp:YFP-ZmMUTE was first detected during the stomatal development of early GMCs , and then moved to SMCs in the SMC establishment stage . The YFP-ZmMUTE signal remains strong until the young GCs become mature GCs ( Fig 5 ) . Unexpectedly , as compared to the expression pattern of BdMUTE , BZU2/ZmMUTE is more specifically expressed in the SMC establishment stage ( Fig 5 ) . To further test the capability of ZmMUTE to move , YFP-fused BZU2/ZmMUTE , as well as its homologues , BdMUTE , OsMUTE , and AtMUTE , were expressed in rice ( Oryza sativa L . ) . OsMUTEp:nls-YFP was only expressed in the GMC at the developmental stages leading from GMC formation to SMC division ( S6A Fig ) , indicating that OsMUTE promoter is active during development of GMC and SMC [13] . Interestingly , as for OsMUTEp:YFP-BdMUTE , OsMUTEp:YFP-OsMUTE and OsMUTEp:YFP-ZmMUTE , the fluorescence of the YFP-fused MUTEs were first detected in the early GMCs , then appeared in the SMCs ( S6B–S6D Fig , white arrows ) . Similar to ZmMUTEp:YFP-ZmMUTE ( Fig 5 ) , after the division of the GMCs , the signals of OsMUTEp:YFP-ZmMUTE were also observed in young GCs and SCs , finally disappearing from the GCs and SCs at maturity ( S6C Fig ) . In contrast , the fluorescence of OsMUTEp:YFP-AtMUTE was very weak in the early GMCs and not detected in SMCs ( S6E Fig ) . Meanwhile , we compared the expression patterns of these different YFP-fused MUTE coding sequences driven by the AtMUTE promoter in Arabidopsis . AtMUTEp:AtMUTE-YFP was observed exclusively in Arabidopsis GMCs ( Fig 6A and 6F ) , but AtMUTEp:YFP-ZmMUTE and AtMUTEp:YFP-BdMUTE were observed both in GMCs and in neighboring cells ( Fig 6B , 6C and 6F ) . Since divergent functions of MUTE have been reported in different species [10 , 21] , we performed sequence alignments , the results of which indicate that MUTE proteins from different species have a variable C-terminal region , whereas the N-terminal region is relatively conserved ( Fig 4D ) . To further verify the characterization of the C-terminal regions of BZU2/ZmMUTE and BdMUTE , we generated AtMUTEp:YFP-ZmMUTE-ΔC and AtMUTEp:YFP-BdMUTE-ΔC transgenic plants in Arabidopsis which lack the 190–219 and 208–237 amino acids in ZmMUTE and BdMUTE , respectively . The fluorescence of AtMUTEp:YFP-ZmMUTE-ΔC is only detected in the GMCs and expressed in both nucleus and cytoplasm . In contrast , AtMUTEp:YFP- BdMUTE-ΔC is detected in the GMCs and restricted in nucleus , which is similar to the pattern seen with AtMUTEp:AtMUTE-YFP ( Fig 6D–6F ) . At the same time , we checked the expression patterns of OsMUTEp:YFP-ZmMUTE-ΔC and OsMUTEp:YFP-BdMUTE-ΔC in rice . As shown in S6F and S6G Fig , YFP-ZmMUTE-ΔC and YFP-BdMUTE-ΔC are only located in the early GMCs , and not in the SMCs and young SCs . However , it is obvious that the localization ( for ZmMUTE-ΔC , Fig 6D and S6F Fig ) , or the intensity ( for BdMUTE-ΔC , Fig 6E and S6G Fig ) , does not correspond to that of the wild-type , suggesting that the protein does not properly interact with its binding partner for normal localization and/or stability . To further confirm whether a truncated BZU2/ZmMUTE protein , from which the 30 amino acids of the C-terminus had been deleted , can still bind the E-box motifs within the promoters of PAN1 ( -27 ) and PAN2 ( -187 , -200 ) , the yeast one-hybrid assay and electrophoretic mobility shift assays ( EMSA ) were used to test the activity of ZmMUTE-ΔC binding to short nucleotide fragments ( 26–37 bp ) containing the E-box motifs . The results of both assays show that BZU2/ZmMUTE lacking these C-terminal amino acids is similar in behavior to BZU2/ZmMUTE ( S7 Fig and S8 Fig ) . These data imply that the C terminus of BZU2/ZmMUTE and BdMUTE might be necessary for their characteristic mobility . Together , these results support that BZU2/ZmMUTE is mobile and necessary for SMC formation and asymmetric division in normal development of the maize stomatal complex . In order to further assess the role of BZU2/ZmMUTE in stomatal development and , in particular , define its interactions with other cellular components , we performed RT-qPCR to compare , for bzu2-1 mutants and wild-type , the transcript levels of genes previously reported to be involved in stomatal and leaf development ( Fig 7A ) . PAN2 is polarized in premitotic SMCs [24] . After PAN2 polarization , PAN1 and ROP proteins are polarized , and an actin patch forms at the GMC/SMC interface [22 , 23] . ROP2/9 functions downstream of PAN1 to promote the premitotic polarization of SMCs [25] , and the premitotic SMC nucleus migrates to this site in an actin-dependent manner . Liguleless1 ( LG1 ) accumulation at the site of ligule formation and in the axil of developing tassel branches , functions in the leaf shape and tassel architecture [30] . Low levels of the PAN1 , PAN2 and ROP2/9 transcripts were observed in bzu2-1 mutant plants as compared to wild-type . Furthermore , these genes related to stomatal development are down-regulated at very early seedling stages ( S9 Fig ) , since stomata production in grass initiates at the leaf base with a longitudinal gradient of development and differentiation toward the tip . This therefore suggests that the reduction of the transcript levels of these genes is not simply due to the leaves dying . For example , ROP2 transcript abundance in the wild-type was 6 . 7 times higher than that in bzu2-1 . Transcripts of the Brick1 ( BRK1 ) and BRK3 genes , required for the formation of epidermal cell lobes as well as for actin-dependent cell polarization events of subsidiary mother cells [26 , 31 , 32] , were also found at lower levels in bzu2-1 plants . The expression of SCARECROW ( SCR ) gene in rice and maize was observed in leaf primordia and in young leaves , which is required for asymmetric cell divisions of GMCs [13 , 26 , 33] . The transcript level of SCR1 was significantly reduced in bzu2-1 as compared to wild-type . However , the transcript levels of LG1 , a SQUAMOSA PROMOTER-BINDING proteins 1 and 2 like gene , in bzu2-1 mutants was comparable to that in the wild-type . More importantly , it is also known that the function of FAMA and OsFAMA is conserved between dicots and monocots in the regulation of the final symmetric division of the GMCs [13 , 34] . In rice , c-osfama ( OsFAMA mutant in rice , generated by CRISPR/Cas9 ) mutants showed the stomatal complex consisted of four swollen cells , two GCs and two SCs , occasionally a stoma lacking one SC was observed , however in c-osfama the entry division and GMC differentiation stage even the stomatal density is the same compared with wild-type . OsFAMA controls the cell fate transition from GMCs to GCs and SMCs to SCs and affected SMC asymmetrical division [13] . We noticed that , in bzu2-1 , the expression of ZmFAMA is reduced significantly ( Fig 7A ) , similarly to the manner in which OsFAMA is dramatically downregulated in the c-osmute mutant [13] . Therefore , the absence of ZmFAMA or OsFAMA is simply a result of the lineage abortion . These data clearly demonstrate that the genes of asymmetric and symmetric division in bzu2-1 mutants are impaired in stomatal development , which is consistent with the proposed role of BZU2/ZmMUTE as a master regulator of stomatal differentiation . Therefore , it is speculated that the abnormal formation of SCs and the symmetric division of GMCs in bzu2-1 mutants might be specifically due to downregulation of PAN1 , PAN2 , and ZmFAMA ( Fig 7A and S9 Fig ) . Previous studies have shown that bHLH transcription factors can specifically bind to the E-box cis-element and regulate the expression of targets ( Fig 7C ) [35] . Therefore , we performed motif enrichment analysis by MEME [36] , found that five and three E-box cis-elements are located in the regions within -750 - -1 in the promoter of PAN1 and PAN2 , respectively ( Fig 7C ) . This prompted us to assess the DNA binding specificity of BZU2/ZmMUTE . First , we obtained transgenic plants of BZU2/ZmMUTEp:YFP-BZU2/ZmMUTE , and used for ChIP-qPCR experiments using anti-GFP ( ab290 , Abcam ) polyclonal antibody ( the expression of YFP-BZU2/ZmMUTE was confirmed by Western blot using anti-GFP antibody ) ( S10 Fig ) . Our results showed a strong enrichment in promoter fragments for PAN1 and PAN2 , as compared to negative controls ( the wild-type in the presence of anti-GFP ) . However , no enrichment was observed in the negative control using the regions lacking E-boxes in the promoters of PAN1 and PAN2 ( Fig 7B ) . These data are consistent with the ChIP-qPCR results using the native antibody of BZU2/ZmMUTE ( S11 Fig ) . Second , we employed the yeast one-hybrid assay to directly examine interactions between BZU2/ZmMUTE and 3–4 repeats of short nucleotide fragments ( 26–37 bp ) containing the E-box motifs from the promoters of PAN1 ( -27 ) and PAN2 ( -187 , -200 ) . Activation of BZU2/ZmMUTE the LacZ reporter gene was seen for PAN1 sequence motifs , but was not detected using sequences from a second putative E-box contained in the PAN2 promoter ( -645 ) ( Fig 7C–7E ) . Mutation of the E-box motif eliminated LacZ expression , providing confirmation of the binding specificity of BZU2/ZmMUTE . Finally , the experiment of EMSA further indicated that BZU2/ZmMUTE binds to the E-box motifs in vitro ( Fig 7F ) , which is consistent with the yeast one-hybrid data . Taken together , our results suggest that PAN1 and PAN2 are two direct targets of BZU2/ZmMUTE .
In this study , we found that BZU2/ZmMUTE encoding a bHLH transcription factor is an ortholog of AtMUTE . AtMUTE plays an essential role in the transition of the meristemoid to GMC by repressing the stem cell activity of the meristemoid and inducing guard mother cell formation [10 , 12] . Loss-of-function bzu2-1 mutants can form normal GMCs , but fail to undergo a symmetric division to generate two guard cells , as in wild-type . This indicates that the GC precursors in bzu2-1 are similar to wild-type , with the exception of stomatal development . Consequently , eight-day-old seedlings of bzu2-1 displayed etiolated phenotypes and , at the stage of three leaves , the seedlings subsequently died ( Fig 1 ) . Unlike AtMUTE mutants , the longitudinal symmetric division of the GMC in bzu2-1 mutants is replaced by an asymmetric or irregular directional division ( transverse or diagonal longitudinal division ) , resulting in the production of 2–4 short columns of elongated cells ( Fig 2B and 2C ) . In bzu2-1 mutants , the stomata are defective and thus the functions of gas exchange and water loss are impaired . Our data suggest that BZU2/ZmMUTE acts as a switch that controls stomatal development . Normally , in maize , premitotic SMCs polarize toward the GMC in response to hypothetical cues coming from the adjacent GMCs . This process involves migration of the nucleus toward the GMCs and a distinct enrichment of cortical F-actin at the point of interaction of the GMCs and SMCs when the cell files are forming . Subsequently , SMCs divide asymmetrically to produce subsidiary cells flanking the GMC , which in turn divides to produce a guard cell pair to form stomatal complex [4] . In bzu2-1 mutants , a lack of nuclear polarization and/or actin accumulation at the GMC/SMC interaction area was observed . Thus , the GMCs failed to induce the flanking cells to form SMCs . bzu2-1 mutants share similar features with BdMUTE mutants of B . distachyon , but are different in several respects ( Fig 2 ) . Both mutations of MUTE in Zea mays and B . distachyon result in an abnormal SMC formation and polarization . BdMUTE mutants fail to recruit SCs and instead produce dicot-like two-celled stomata [21] . As compared to BdMUTE , where partial guard cells form without SCs , bzu2-1 is completely defective in guard cell and SMC formation . These data suggest that the function of BZU2/ZmMUTE transcription factor in maize is different from BdMUTE , since absence of the former appears to disrupt SC and GC formation in a more direct and severe manner than does absence of BdMUTE . In addition , our results suggest the C-terminal is necessary for the mobile nature of both BdMUTE and BZU2/ZmMUTE ( Fig 6 ) . Even though previous work in B . distachyon has established the mobile nature of BdMUTE , the molecular mechanisms of how BdMUTE controls SCs formation are still unknown [21] . In fact , the mechanisms governing SMC polarization to allow establishment of asymmetric division are also largely unknown . We provide several lines of evidence suggesting that BZU2/ZmMUTE participates in the regulation of SMC development , and of PAN1 and PAN2 , which are early regulators of SC precursor and of SMC polarization by cooperatively promoting polarization of the actin cytoskeleton and nuclei in these cells [22] . Firstly , our data show that BZU2/ZmMUTE can bind to the E-box of the PAN1 and PAN2 promoters ( Fig 7 and S7 Fig ) ; consistent with this , transcript levels for PAN1 and PAN2 are severely downregulated in bzu2-1 mutants as compared to the wild-type . Unexpectedly , yeast one-hybrid and EMSA data show that BZU2/ZmMUTE is not able to activate P3 of the promoter of PAN2 in vitro , but the ChIP-qPCR data indicate that BZU2/ZmMUTE can activate P3 of the PAN2 promoter in vivo . We speculate that BZU2/ZmMUTE combines with additional unknown factors to activate P3 of the PAN2 promoter in vivo ( Fig 7 , S7 Fig and S8 Fig ) , by analogy to the observation that the DNA binding specificity of AtMUTE depends on its dimerization partner ICE1 or SCRM2 [37] . Furthermore , a number of genes involved in the regulation of stomatal development are down-regulated in bzu2-1 mutants as compared to the wild-type ( Fig 7A ) . Finally , mutations in BZU2/ZmMUTE disrupted the actin-based patch attachment of GMCs with SMCs , which can mis-orient the deposition of new cell walls ( Fig 3A and 3B ) [38–41] . Actin plays an important role in the spatial regulation of asymmetric cell division . For example , actin-dependent relocation of the nucleus during G1 to a defined cortical site of SMCs during stomatal complex formation is one of the early events of asymmetric cell division in Tradescantia [42] which is followed by formation of a dense actin patch at this site [43 , 44] . Combining the existing knowledge and the results presented , here we propose a model in which BZU2/ZmMUTE plays a role partially similar to what has been previously described , but that includes some different roles in maize stomatal development ( Fig 8 ) . In the base of maize leaves , protodermal cells undergo one asymmetric division to form the GMCs . At a specific stage of GMC development ( Fig 5 ) , these cells could send BZU2/ZmMUTE as an extrinsic cue to neighboring cells connecting non-sister GMCs and SMCs , thereby acting as an important regulator of the early players in SC development ( e . g . PAN1 , PAN2 ) . After BZU2/ZmMUTE-mediated induction , PAN1 and PAN2 ( and possibly other additional factors ) accumulate at the SMC/GMC interface , working together with F-actin to induce SMC polarity and nuclear migration towards the GMC proximal site [26] . Following polarization , SMCs undergo one asymmetric division to form a SC and an epidermal cell . In the final stage of stomatal development , BZU2/ZmMUTE performs an additional role controlling the symmetric division of the GMCs to produce the two GCs . It is important to note that the role of BZU2/ZmMUTE in the biogenesis of SCs is not limited to the control of PAN1/2 expression , since SC formation is not entirely disrupted in pan1 or pan2 mutants , whereas SCs are completely absent in bzu2-1 mutants [22] . Thus , these imply that BZU2/ZmMUTE , acting as a GMC-derived polarizing signal , moves to neighboring cells ( Fig 5 ) . This , in turn , initiates the expression of the polarity program by regulating the expression of the genes required for nuclear polarization and polarized actin accumulation at the GMC contact sites ( Fig 7A ) . These data indicate that BZU2/ZmMUTE may play a role in the modulation of gene expression at earlier stages of SC precursor development . In summary , our data support a critical role for BZU2/ZmMUTE in the regulation of SC development and GC maturation . Further studies will be required to dissect the functions of BZU2/ZmMUTE in the determination of GMC fate and in the initiation of intercellular signaling required for the recruitment of the SMCs during stomatal development . More insights could emerge from characterization of the additional mutants obtained in our screen that show different defects in GC and SC development .
Maize plants were grown at the experimental station of Henan University in the Kaifeng experimental field , Henan Province , and the Sanya experimental field , Hainan Province . To isolate stomatal development deficient mutants , we established an extensive collection of EMS-mutagenized maize plants , as follows: the pollen of maize inbred line Mo17 was treated with EMS , T1 population seeds were sowed in the soil , and T1 plants were self-crossed to obtain T2 population seeds . We screened the T2 mutagenized population for the phenotype of an altered leaf surface temperature , using a far-infrared imaging instrument . In 3 , 000 lines , represented by ~45 , 000 T2 population seeds , we found a single lethal mutant , and named bzu2-1 ( called bizui , closed mouth , bzu2-1 ) . Genetic analysis indicates BZU2 is a qualitative trait gene , with a segregation ratio of 3:1 in the F2 generation ( Table 1 ) . The homozygous bzu2-1 ( -/- ) mutant is lethal . We use the heterozygous bzu2-1 ( +/- ) crossed with B73 for generation of the reciprocal F1 population . The F2 population resulting from the self-crossed F1 , and a map-based cloning population , was screened for the bzu2-1 phenotype from the F2 population ( S2 Fig ) . Preliminary mapping of BZU2 used 306 plants from the F2 population , derived from a cross between Mo17 and B73 . The 384 SSR markers ( SIGMA catalog number M4193 ) selected from the Maize Genetics and Genomic Database cover the entire genome with an average of 20 cM units of map distance between every two SSR markers . More SSR markers that were genetically mapped on IBM2 2008 Neighbors Frame were used , and BZU2 was mapped between SSR markers bnlg1863 ( recombination rate 2 . 1% ) and umc1858 ( recombination rate 12 . 5% ) . Based on Maize B73_RefGen_v3 ( http://www . maizesequence . org ) , the BZU2 was mapped to the chromosome VIII between bins 8 . 03 and 8 . 04 . Therefore , all BAC contigs in bins 8 . 03–8 . 04 were exploited to develop new polymorphic markers . To develop more SSR markers , SSR Hunter 1 . 3 [45] was used to search for SSR sequences present in bins 8 . 03–8 . 04 . SSRs and their flanking sequences about 150 bp were then aligned with NCBI nucleotide BLAST ( http://www . ncbi . nlm . nih . gov ) ( high-throughput genomic sequences: HTGSs ) . Only single sequences were used as SSR markers and amplified by PCR . In fine mapping , ~7 , 000 plants from F2 population deprived from a cross between Mo17 and B73 , 63 polymorphic markers were used , BZU2 was mapped in SSR markers 79M15 ( 79 . 01 M ) and 79M45 ( 79 . 70 M ) . Further analysis and sequencing confirm that GRMZM2G417164 was located between these two markers , AGCT insert in the + 390 bp of GRMZM2G417164 . CRISPR constructs were designed using the vector system and following the design protocol [46] , and was done by Genovo Biotechnology Co ( Xi’an , Shanxi ) . The PAM sites were chosen at the + 289 bp downstream of the start codon of BZU2/ZmMUTE genome region , because there is no intron in BZU2/ZmMUTE . The sequence of the gRNA is CCTGTCATGATCAAGGAGCTCGC ( S1 Table ) . To genotype CRISPR-induced mutations , we amplified a 668-bp fragment including the guide RNA site by PCR from the genome of transgenic seedlings [47 , 48] , and the PCR products were sequenced . We obtained three CRISPR/Cas9 mutant lines: bzu2-2 , bzu2-3 and bzu2-4 . A 5 bp and 25 bp deletion was detected behind PAM site in bzu2-2 line and bzu2-3 respectively , while a 1 bp insertion was detected behind PAM site bzu2-4 line . In T0 mutant plants , the phenotype of homozygous mutants lines was comparable to bzu2-1 ( Fig 4C ) . The heterozygous T0 mutants lines were planted in the field to get seeds . In the T1 populations , homozygous mutants lines also showed a similar phenotype to that of bzu2-1 . Reporter constructs to be transformed into Oryza sativa Japonica were generated using the In-fusion cloning with the monocot binary expression vector pIPKb003 and to be transformed in Arabidopsis thaliana were generated using the Gateway Recombination Cloning Technology with plant expression vector pGWB504 . AtMUTE promoter and genome sequences were amplified from the Arabidopsis ( Col-0 ) genome , OsMUTE promoter and genome sequences were amplified from the Japonica genome , and BdMUTE sequence was amplified from the Brachypodium distachyon genome . All genomic DNA samples were produced using the Plant Genomic DNA Kit ( TIANGEN ) . RNA samples were produced using the TRIzol extraction method and corresponding cDNA was obtained by M-MLV reverse transcriptase ( M1705 , Promega ) . The Pri1 and Pri2 primers were used to clone the OsMUTE promoter from the rice genome . We then amplified tag-YFP using forward primers Pri3 and Pri4 ( for AtMUTE ) , Pri3 and Pri7 ( for OsMUTE ) , Pri3 and Pri10 ( for BZU2/ZmMUTE ) , and Pri3 and Pri14 ( for BdMUTE ) , respectively , which have sequences homologous with the OsMUTE promoter , and using four reverse primers ( Pri4 , Pri7 , Pri10 , Pri14 ) to get four tag-YFP PCR products which has homologous sequence with the AtMUTE , OsMUTE , BZU2/ZmMUTE and BdMUTE . Then we used primers Pri5 and Pri6 to clone the AtMUTE ORF carrying a STOP codon , primers Pri8 and Pri9 to clone the OsMUTE ORF with a STOP codon , and the primers Pri11 and Pri12 to clone the ZmMUTE ORF with a STOP codon . Primers Pri11 and Pri13 were used to clone the ZmMUTE ORF with a deletion of 30 AA in the C terminal replaced by a STOP codon , and primers Pri15 and Pri16 to clone the BdMUTE ORF with a STOP codon . Primers Pri15 and Pri17 were used to clone the BdMUTE ORF with a deletion of 30 AA in the C terminal replaced by a STOP codon . The three PCR products were employed for in-fusion cloning using the vector pIPKb003 . OsMUTEp:YFP-AtMUTE , OsMUTEp:YFP-OsMUTE , OsMUTEp:YFP-BdMUTE and OsMUTEp:YFP-BZU2/ZmMUTE were produced according to the procedures described in the manual of the Clone Express® MultiS One Step Cloning Kit ( Vazyme Biotech Co . , Ltd , China ) . Pri18 and Pri19 were used to clone the whole genomic fragment of AtMUTEp:AtMUTE from the Arabidopsis thaliana ( Col-0 ) genome . The tag-YFP fragment was amplified using Pri20 and Pri21 , Pri20 bearing homology to AtMUTE and Pri21 providing an attB2 site . The fusion gene AtMUTEp:AtMUTE-YFP was then generated by overlap PCR using primers Pri18 and Pri21 , and then BP recombined into pDONR207 , next LR recombined into destination vector pGWB504 . For AtMUTEp:YFP-BZU2/ZmMUTE and AtMUTEp:YFP-BdMUTE , primers Pri18 and Pri22 were used to clone the AtMUTE promoter from the Arabidopsis ( Col-0 ) genome . Pri22 provided sequence homologous to tag-YFP . With primer Pri23 and reverse primers ( Pri24 , Pri25 ) , this amplified the tag-YFP PCR products all carrying sequences homologous to BZU2/ZmMUTE and BdMUTE . We then used primers Pri26 and Pri27 to clone the BZU2/ZmMUTE ORF carrying a STOP codon , and primers Pri26 and Pri28 to clone the BZU2/ZmMUTE ORF with 30 AA at the C terminal being replaced with a STOP codon . Primers Pri29 and Pri30 were similarly used to clone the BdMUTE ORF with a STOP codon , and primers Pri29 and Pri31 to clone of BdMUTE ORF with 30 AA at the C terminal being replaced with a STOP codon . All reverse primers amplified ORFs carrying Gateway attB2 sites . The inframe gene fusions AtMUTEp:YFP-BZU2/ZmMUTE , AtMUTEp:YFP-BZU2/ZmMUTE-ΔC , AtMUTEp:YFP-BdMUTE , and AtMUTEp:YFP-BdMUTE-ΔC were generated by overlap PCR using the forward primer Pri18 and four reverse primers ( Pri27 , Pri28 , Pri30 , Pri31 ) , then BP recombined into pDONR207 , followed by LR recombined into the destination vector pGWB504 . AtMUTEp:AtMUTE-YFP , AtMUTEp:YFP-BZU2/ZmMUTE , AtMUTEp:YFP-BdMUTE , AtMUTEp:YFP-BZU2/ZmMUTE-ΔC and AtMUTEp:YFP-BdMUTE-ΔC were produced through Gateway Recombination Cloning ( Invitrogen ) . BZU2/ZmMUTEp:YFP-BZU2/ZmMUTE was inserted into the maize genome via Agrobacterium mediated transformation ( see below ) . The promoter of BZU2/ZmMUTE being amplified from the wild-type ( Mo17 ) genome using primers Pri32 and Pri33 , the YFP ORF lacking the stop codon amplified using primers Pri34 and Pri35 , and the ORF of BZU2/ZmMUTE including the stop codon amplified from the cDNA of Mo17 with primers Pri36 and Pri37 . The three PCR products were employed for In-fusion cloning with the pCM3300 vector . The primers used in this work are listed in S1 Table . Reporter constructs of OsMUTEp:YFP-AtMUTE , OsMUTEp:YFP-OsMUTE , OsMUTEp:YFP-BdMUTE , OsMUTEp:YFP-BZU2/ZmMUTE , OsMUTEp:YFP-BZU2/ZmMUTE-ΔC and OsMUTEp:YFP-BdMUTE-ΔC were transformed into Oryza sativa Japonica calli with EHA105 Agrobacterium , as previously described [49] . AtMUTEp:AtMUTE-YFP , AtMUTEp:YFP- BZU2/ZmMUTE , AtMUTEp:YFP-BdMUTE , AtMUTEp:YFP-BZU2/ZmMUTE-ΔC and AtMUTEp:YFP-BdMUTE-ΔC were transformed into Arabidopsis ( Col-0 ) using Agrobacterium strain GV3101 [50] . Maize transgenic lines were produced via Agrobacterium-mediated transformation [51] . Maize seedlings were grown in the greenhouse or in the experimental field . Ears containing immature embryos , between 1 . 0 to 2 . 0 mm in length along the axis and optimal for transformation , were collected 8 to 13 days after pollination . The immature embryos were submerged in an Agrobacterium tumefaciens suspension contained in a 2 . 0 mL Eppendorf tube at room temperature for 1 h . The solution was then removed , and the embryos transferred onto fresh co-cultivation solid medium with the scutellum face up , and were incubated in darkness at 25°C for 2–3 days . After that , the calli were transferred onto fresh screen solid medium , screening three times for a period of 2 weeks each . The Type I calli that further proliferated were transferred to shoot regeneration medium , and incubated under continuous illumination ( 5 , 000 lux ) at 25°C for 14–30 days . The emerging shoots were transferred onto root regeneration medium , and were incubated under continuous illumination ( 5 , 000 lux ) at 25°C for 14–30 days . The rooted seedlings were transferred to pots containing appropriately supplemented soil for growth in the greenhouse for 3–4 months to collect progeny seeds . We typically analyzed at least three independent lines in the T0 generation ( depending on how many independent lines were recovered upon regeneration ) and confirmed the observed expression pattern in at least three T1 individuals if the transgenics were fertile and produced seeds . T1 maize transgenic plants were used in this study . The images were acquired using a Leica SP8 confocal microscope , the cellular membranes being counterstained with propidium iodide ( PI , red ) in maize and Arabidopsis , and FM4-64 ( red ) in rice . To determine the stomatal phenotype in wild-type and bzu2-1 mutants ( Fig 1D ) , epidermal strips were peeled from mature leaves of wild-type and bzu2-1 plants , and were observed using a Zeiss Axioskop II microscope equipped with differential interference contrast optics . To obtain images characterizing the process of stomatal complex development ( Fig 2A ) , around ~1 . 5 cm of segments from the leaf base were excised from 8-day-old seedlings . The tissues were cleared in Herr's solution ( lactic acid:chloral hydrate:phenol:clove oil:xylene ( 2:2:2:2:1 , by weight ) ) [52] . The stomatal development process was studied using Zeiss Axioskop II microscope equipped with differential interference contrast optics . For F-actin observation , ~1 . 5 cm ( the section of stomatal complex develops from stomatal lineage cell to mature stomatal complex is about 1 . 5 cm base in the leaf from maize seedling root node , indicated by the analysis of stomatal complex development using microscopy ) of basal leaf segments , excised from 8-day-old seedlings , were cut into 0 . 2 cm wide x 0 . 5 cm long strips , and fixed for 30 min at room temperature in a solution comprising 4% paraformaldehyde , dissolved in 50 mM PEM ( 50 mM PIPES , 2 . 5 mM EGTA , 2 . 5 mM MgCl2 ) . The strips were washed three times for 5 min in 50 mM PEM , and were permeabilized by submersion 20 mins in 50 mM PEM containing 5% DMSO and 1% Triton X-100 . After three further washes in 50 mM PEM , the sections were incubated for 1 . 5 h in 50 mM PEM solution containing 90 nM AlexaFluor 488-phalloidin ( dissolved in DMSO , Invitrogen/Molecular Probes ) at room temperature . Images were acquired using a Zeiss LSM710 confocal microscope . For observation of the polarized nuclei of the GMCs , ~1 . 5 cm of second or third basal leaf segments excised from 8-day-old seedlings was cut into 0 . 2 cm wide x 0 . 5 cm long strips directly stained with 4 μg/mL Hoechst 33258 ( 94403-1ML , SIGMA ) dissolved in water for 15 min at room temperature . Images were acquired using a Leica SP8 Confocal Microscope . A peptide corresponding to amino acids 192–206 of the BZU2/ZmMUTE protein ( GQDTAEQKPQAEENH ) was synthesized , conjugated to KLH , and used for polyclonal antibody production in rabbits by the Hanlin Biotechnology Co . ( Shijiazhuang , Hebei ) . ChIP-qPCR experiments were carried out using the Magna ChIP kit ( MAGNA001 , Millipore ) with minor modification [53 , 54] . Samples ( 0 . 4 g ) of the stomatal development zones of 8-day-old seedlings were collected , and then immersed in buffer A ( 0 . 4 M sucrose , 10 mM Tris [pH 8] , 1 mM EDTA , 1 mM PMSF ) containing 1% formaldehyde , and were subjected to four eight-minute cycles under vacuum , until the materials became translucent . The materials were then transferred to fresh buffer A containing 0 . 1 M glycine , and incubation was continued for 16 min at 4°C . The materials were then washed , and frozen in liquid nitrogen . Samples ( approximately 0 . 4 g ) of the materials were ground for each immunoprecipitation , and were resuspended in 1 mL lysis buffer ( 50 mM HEPES [pH 7 . 5] , 150 mM NaCl , 1 mM EDTA , 1% Triton X-100 , 0 . 1% deoxycholate , 0 . 1% SDS , 1 mM PMSF , and 1 x Roche Protease Inhibitor Cocktail ) , followed by immunoprecipitation with an anti-GFP antibody ( ab290 , Abcam ) . Immunoprecipitated products were resuspended in 50 μL elution buffer . Approximately 3–5 μL was used for ChIP-qPCR . Each immunoprecipitation was performed three times independently , with the wild-type being used as the control . The primers for ChIP-qPCR are listed in S1 Table . The plasmids pB42AD-BZU2/ZmMUTE and pB42AD-BZU2/ZmMUTE-ΔC , and wild-type and mutated PAN1-P1:LacZi , PAN2-P2:LacZi , PAN2-P3:LacZi were co-transformed into yeast strain EGY48 using standard transformation techniques . Transformants were grown on dropout plates containing X-gal ( 5-Bromo-4-chloro-3-indolyl-β-D-galactopyranoside ) for blue color development . pB42AD-SPL9 ( squamosa promoter binding protein-like 9 ) reacting with DFR ( dihydroflavonol reductase ) [55] was used as a positive control , and combinations with the empty pB42AD vector were used as negative controls . The PAN1 and PAN2 primers are listed in S1 Table . The EMSA was conducted using the LightShift™ Chemiluminescent EMSA Kit ( Thermo Fisher Scientific ) according to the manufacturer’s protocol . The coding sequence of BZU2/ZmMUTE was cloned into pGEX-2TK . Recombinant BZU2/ZmMUTE protein was expressed and purified from BL21 E . coli . The probes of the PAN1/2 promoter were obtained by gene synthesis and biotin-labeled at their 5’ terminal . Biotin-unlabeled probes of the same sequences were used as competitors . The probe sequence is described in S1 Table . The amino acid sequences of BZU2/ZmMUTE and other homologous proteins were retrieved from NCBI ( https://www . ncbi . nlm . nih . gov/ ) . Alignments of BZU2/ZmMUTE ( GRMZM2G417164 ) ( in wild-type and bzu2-1 ) , BdMUTE ( LOC100843821 ) , OsMUTE ( Os05g51820 ) and AtMUTE ( At3g06120 ) were conducted using the MUSCLE and BOXSHADE programs . Construction of phylogenic trees using the neighbor-joining method and confirmation of tree topology by bootstrap analysis ( 5 , 000 replicates ) was performed with MEGA6 software ( using default settings except for the replicates of bootstrap value ) [56] . Samples were taken from ~1 . 5 cm of the base of leaves of 8-day-old maize seedlings for RT-qPCR analysis . Total RNA was isolated using the TRIzol reagent ( Life Technologies ) according to the manufacturer’s protocol . Reverse transcription into cDNA was done with 2 μg total RNA in 25 μL reverse transcription mixture , using M-MLV Reverse Transcriptase ( M1705 , Promega ) . The cDNA was diluted to 100 μL , and 1 μL diluted cDNA was used as the template for quantitative RT-PCR analysis . The maize ZmUbiquitin 2 gene was used as an internal standard to normalize expression of the tested genes . Quantitation was performed using at least three independent biological replicates [57] . The primers used for RT-qPCR are listed in S1 Table . The coding sequences of BZU2/ZmMUTE was cloned into pGreen0280 ( 35S:BZU2/ZmMUTE-GFP ) for the analysis of subcellular localization . 35S:BZU2/ZmMUTE-GFP , empty vector , and 35S:H2B-mCherry were cotransfected into tobacco leaves . Green and red fluorescence was imaged using the Zeiss LSM710 confocal microscope 24 h after Agrobacterium-mediated infiltration of tobacco ( Nicotiana benthamiana ) leaves [58] . The primers used are listed in S1 Table . Previous studies indicated that the E-box cis-element is the conserved CANNTG . The cis-element was produced by the Multiple Expectation Maximization for Motif Elicitation MEME Suite web server [36] . | In the grasses , individual stomatal complexes comprise a pair of dumbbell-shaped guard cells associated with two subsidiary cells and the pore , which together play essential roles in the exchange of CO2 and O2 , in xylem transport , and in transpiration . However , little is known about grass stomatal complex development . We have uncovered and characterized a key factor ( BZU2/ZmMUTE ) determining the formation of guard cells in Zea mays . Our data suggest that BZU2/ZmMUTE has a dual role , both as an important player in determining the formation of the guard mother cell , as well as being required for polarization and recruitment of the subsidiary mother cells . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"biotechnology",
"plant",
"anatomy",
"stem",
"anatomy",
"engineering",
"and",
"technology",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"brassica",
"stomata",
"plant",
"science",
"model",
"organisms",
"experimental",
"organism",
"systems",
"plant",
"genomics",
"sequence",
"motif",
"analysis",
"seedlings",
"plants",
"research",
"and",
"analysis",
"methods",
"arabidopsis",
"thaliana",
"bioengineering",
"sequence",
"analysis",
"grasses",
"animal",
"studies",
"bioinformatics",
"maize",
"plant",
"genetics",
"leaves",
"eukaryota",
"plant",
"and",
"algal",
"models",
"cell",
"biology",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"plant",
"biotechnology",
"organisms"
]
| 2019 | BZU2/ZmMUTE controls symmetrical division of guard mother cell and specifies neighbor cell fate in maize |
Protein promiscuity is of considerable interest due its role in adaptive metabolic plasticity , its fundamental connection with molecular evolution and also because of its biotechnological applications . Current views on the relation between primary and promiscuous protein activities stem largely from laboratory evolution experiments aimed at increasing promiscuous activity levels . Here , on the other hand , we attempt to assess the main features of the simultaneous modulation of the primary and promiscuous functions during the course of natural evolution . The computational/experimental approach we propose for this task involves the following steps: a function-targeted , statistical coupling analysis of evolutionary data is used to determine a set of positions likely linked to the recruitment of a promiscuous activity for a new function; a combinatorial library of mutations on this set of positions is prepared and screened for both , the primary and the promiscuous activities; a partial-least-squares reconstruction of the full combinatorial space is carried out; finally , an approximation to the Pareto set of variants with optimal primary/promiscuous activities is derived . Application of the approach to the emergence of folding catalysis in thioredoxin scaffolds reveals an unanticipated scenario: diverse patterns of primary/promiscuous activity modulation are possible , including a moderate ( but likely significant in a biological context ) simultaneous enhancement of both activities . We show that this scenario can be most simply explained on the basis of the conformational diversity hypothesis , although alternative interpretations cannot be ruled out . Overall , the results reported may help clarify the mechanisms of the evolution of new functions . From a different viewpoint , the partial-least-squares-reconstruction/Pareto-set-prediction approach we have introduced provides the computational basis for an efficient directed-evolution protocol aimed at the simultaneous enhancement of several protein features and should therefore open new possibilities in the engineering of multi-functional enzymes .
Proteins are capable to perform molecular tasks with impressive efficiency and , often , with exquisite specificity . Nevertheless , many proteins possess weak promiscuous functions , which are more or less related to the primary activity , but involve different substrates or different chemical alterations [1]–[5] . Protein promiscuity has been extensively studied in recent years due to its important biotechnological applications [6]–[12] , to its role in adaptive metabolic plasticity [13]–[15] and also because of its fundamental connection with molecular evolution . Indeed , promiscuity in modern proteins is plausibly a vestige of the broad specificity of primordial proteins [1] . Furthermore , as briefly elaborated below , promiscuity likely plays an essential role in the development of new functions through divergent evolution [3] , [5] , [8] , [16]–[20] . Development of new functions does occur during evolution , sometimes with impressive speed . In most cases , the process involves gene duplication as a necessary step . It has been repeatedly noted , however , that random accumulation of mutations in a gene is unlikely to create a new function . It is generally assumed , therefore , that a sufficient level of the new ( initially promiscuous ) activity must be present before the duplication event . In this way , natural selection can act on one of the gene copies to enhance the new function , while the original function is retained by the other copy . However , optimization of a functional site for a given molecular task likely interferes with the efficient performance of the protein for a different task based on the same site . Consequently , enhancement of the promiscuous activity prior to gene duplication may be expected to cause a decrease in primary activity that could conceivably compromise organism survival . As a solution to this conundrum , a “weak trade-off” scenario has been proposed [5]: enhancement of the promiscuous activity is assumed to be accompanied with only a moderate decrease in primary function and , therefore , a generalist protein ( significant levels of both activities ) can be formed prior to gene duplication without seriously impairing organism fitness . This weak trade-off explanation is certainly supported by a number of laboratory evolution experiments [5] . However , the possibility that natural evolution may actually avoid or bypass primary/promiscuous activity trade-offs ( i . e . , a “no trade-off” scenario as opposed to a “weak trade-off” scenario ) should be seriously taken into account , since bifunctional enzymes with the capability to catalyze efficiently two different biochemical reactions based on the same active site are known and have been recently characterized in detail [21] , [22] and experimental studies have supported that the trade-off between high activity and tight specificity can be greatly relaxed [23] . Here , we aim at assessing the patterns of primary/promiscuous activity modulation in the mutational space actually explored by natural evolution when recruiting a promiscuous activity for a new function . The approach we propose involves essentially three steps: A unique global optimum cannot be defined when dealing with a multi-objective optimization problem , such as , for instance , enhancing a promiscuous activity while keeping the level of the primary activity as high as possible . However , a set of several optimal solutions can be defined using the Pareto criterion: a solution ( protein variant in this case ) belongs to the set of optimal solutions ( the so-called Pareto set ) if it is not dominated by any other solution . The dominance relationship is defined as follows: a solution a dominates a solution b if it shows enhanced performance for all optimization objectives . In the specific case of interest here , variant a dominates variant b if primary-activity ( a ) >primary-activity ( b ) and simultaneously promiscuous-activity ( a ) >promiscuous-activity ( b ) . The construction of the Pareto set of non-dominated solutions is illustrated with a simple example in Figure 1A . The Pareto set includes the solutions with optimal trade-offs between the different objectives and has been used extensively in economics , while its application to protein design has only been explored in recent years [29]–[31] . In the specific case of interest here , determination of the Pareto set should immediately clarify the main features of the modulation of the primary and promiscuous activities within a given mutational space . For instance , if the starting variant ( the “wild-type” protein , for instance ) already belongs to the Pareto set , enhancement of the promiscuous activity necessarily implies a decrease in primary function and the trade-off will be weak ( Figure 1B ) or strong ( Figure 1C ) depending of the general slope of the Pareto set in the plot of promiscuous activity versus primary activity . On the other hand , if the starting variant does not belong to the Pareto set , simultaneous optimization of both activities is in principle feasible ( Figure 1D ) and primary/promiscuous trade-offs can be avoided to some extent . To test the approach proposed , we have chosen the three basic activities associated with the thioredoxin fold: reduction of disulfide bridges , formation of disulfide bridges and isomerization ( reshuffling ) of disulfide bridges . The two latter activities are linked in vivo to protein folding processes [32] , [33] ( oxidative folding and rescuing of proteins with incorrect disulfide bridges ) and , in the periplasm of bacteria , are performed by different proteins: DsbA and DsbC , respectively . By contrast , in the endoplasmic reticulum of eukaryotic cells , both disulfide-linked folding processes are catalyzed by the same protein: protein disulfide isomerase ( PDI ) . PDIs are multidomain proteins that contain thioredoxin-fold domains [32] . Processes of disulfide reduction in vivo ( obviously unrelated with folding ) are typically catalyzed by single-domain thioredoxins , which may also show low in vitro levels of the protein folding activities associated to disulfide bridge formation and reshuffling . Indeed , it is tempting to speculate that low levels of these activities were already present in primordial thioredoxins and that , at some evolutionary point , were recruited for new-function development leading to the proteins involved in disulfide-bridge-linked protein folding [32] . The processes of thioredoxin-domain catalyzed reduction , formation ( oxidation ) and reshuffling of disulfide bonds are all dependent on the active-site CXXC motif . Reduction ( see Figure 2 ) starts with the reduced enzyme and involves the nucleophilic attack of the thiolate form of the amino-terminal cysteine on the disulfide bridge of the substrate [34] , [35] . The mixed-disulfide thus formed is resolved by the nucleophilic attack of the carbonyl-terminal cysteine . Oxidation , on the other hand ( see Figure 3 ) , involves a nucleophilic attack of the substrate on the disulfide bridge of the oxidized enzyme and the mixed-disulfide intermediate is resolved by attack from the free cysteine ( in the thiolate form ) of the substrate [36] . It is relevant that the oxidation and reduction processes involve opposite chemical changes in the substrate ( break-up and formation of disulfide bridges ) as well as different mechanisms for the resolution of the mixed-disulfide intermediate . Furthermore , the two processes may be expected to be linked to different values of the redox potential ( as suggested by the redox potentials of thioredoxin and PDI: see Figure 6 in Hatahet & Ruddock [36] ) and , as it has been extensively discussed in the literature , they likely have different molecular requirements in terms of the conformational changes during catalysis , the stability of cysteine thiolates and the modulation of the pK values of the catalytic groups [33] , [36]–[39] . Clearly , disulfide reduction and catalysis of oxidative folding ( involving formation of disulfide bridges ) may be expected to strongly trade-off . Contrary to disulfide reduction and formation ( Figures 2 and 3 ) , disulfide-bridge reshuffling in misfolded proteins to yield the correctly folded state does not involve a net change in the oxidation state of the substrate and could in principle occur through cycles of catalyzed reduction/oxidation [36] , [40] . Alternatively , the initial attack of the enzyme on a substrate disulfide bridge may yield a free cysteine that could attack another disulfide bridge thus starting a cascade of disulfide-bond rearrangments leading to the most stable configuration [36] , [40] . The specific protein system we use in this work is E . coli thioredoxin , an enzyme involved in multiple reduction processes in vivo [35] , [41] which , besides this primary ( i . e . reductase ) activity is able to catalyze , albeit with very low efficiency , disulfide-bridge-linked protein folding processes [42] , [43] ( promiscuous activities of E . coli thioredoxin ) . We apply the approach proposed ( steps 1–3 above ) to E . coli thioredoxin with the catalysis of oxidative folding as the promiscuous activity . Nevertheless , the variants thus obtained are also tested for the disulfide reshuffling activity .
To find a set of positions likely linked to the emergence of disulfide-bridge-linked folding functions in the thioredoxin fold , we have used statistical coupling analysis ( SCA ) which works by comparing the amino acid distributions at different positions in a multiple sequence alignment ( MSA ) with the corresponding ones in a given sub-alignment [24] . To apply SCA to the study of primary/promiscuous activity modulation we propose selecting the sub-alignment on the basis of a function-related criterion related with the promiscuous activity . We thus start with a MSA derived from a sequence-database search using the E . coli thioredoxin sequence as query and select as sub-aligment those sequences belonging to thioredoxin-fold domains in proteins involved in protein folding in vivo . Actually , this selection step is made straightforward by the fact that these domains contain the active-site CGHC sequence , while thioredoxin reductases contain the CGPC sequence . In fact , the P34H mutation on the E . coli thioredoxin background has been shown to enhance significantly its “PDI-like” promiscuous activities [43] . The MSA we have used contains indeed a significant number of sequences with a histidine at position 34 ( E . coli thioredoxin numbering ) that , in most cases , belong to thioredoxin-fold domains of eukaryotic PDI's . We thus obtained the statistical free-energies using the P→H substitution as the perturbation at position 34 ( see Methods for details ) and we expect these values to reveal networks of residues related with the emergence of the protein folding activities in the thioredoxin fold . However , SCA is based upon the perturbation of the 20 amino acid distribution at each position and actually provides a list of coevolving positions ( see Figure 4A ) , while we are interested in specific mutations at these positions . Therefore , we included an additional layer of statistical analysis . For each given position , we considered the mutation from the amino acid present in E . coli thioredoxin ( the “Ec” aminoacid ) to the amino acid “X” defined as the amino acid different from “Ec” that has the highest frequency ( largest number of occurrences in the sequence alignment ) when there is a histidine at position 34 . Then , for each of the 13 positions with the highest coupling free energies from the SCA analysis ( Figure 4A ) , we calculated the following score: ( 1 ) where f is the frequency of occurrence of the amino acids in the sequence alignments and subscripts “H” and “P” refer to the condition for the calculation of the frequencies ( histidine or proline at position 34 , respectively ) . Large positive values for the score indicate that the P34H substitution shifts the statistics strongly towards amino acid X . We retained for experimental analysis the 10 positions ( and the corresponding Ec→X mutations ) for which the score was positive ( see Figure 4B ) : I4V , D26E , W28Y , E30P , I38L , K57A , N59D , D61T , I75Y , L94R . It is important to note that the 10 positions selected form a well-defined , connected network surrounding the active site ( see Figure 4C ) , a fact fully consistent with their likely role in the development of the new functions related with protein folding catalysis . We also wish to emphasize at this point that the immediate purpose of this work ( see section below ) is to probe the interplay between primary and promiscuous activities in the mutational space defined by the 10 mutations selected . Specifically , the derivation of a molecular-level picture of what each of these mutations is doing ( a task that would require extensive structural work due to the potential non-additivity of the mutation effects ) , is beyond the scope of this work . A simple visual examination ( see Figure 4D ) of the sequences of the MSA used that include histidine at position 34 ( most of them belonging to eukaryotic PDI's ) shows that different combinations of the 10 mutations selected occur in extant PDI's . We conclude that natural selection does efficiently explore the mutational space defined by combinations of the 10 mutations . To assess how the interplay between the primary and promiscuous activities is modulated in this mutational space , we prepared a combinatorial library spanning the 10 mutations ( i . e . , 210 = 1024 variants ) on the P34H background and determined the reductase and the catalysis of oxidative folding activities for 29 randomly selected variants . The primary activity has been probed by following the standard reduction assay for DTT-reduced thioredoxin which uses insulin as a model substrate ( see Methods for details ) . Clearly , this assay reflects the reduction process of Figure 2 . The catalysis of oxidative folding ( promiscuous activity in E . coli thioredoxin ) has been assessed using fully reduced ribonuclease A as substrate ( see Methods for details ) . This assay might include some contribution from the reshuffling process since the first disulfide bridges formed need not be the correct ones . Nevertheless , it is expected to probe mainly the oxidation pathway ( Figure 3 ) , an expectation that will be supported by the results reported here . Most of the 29 variants screened show increased levels of the promiscuous activity with respect to both the background P34H variant and wt E . coli thioredoxin ( Figure 5A ) . This result is consistent with the proposed role of the selected mutations in the emergence of the protein folding activities of the thioredoxin fold . What may perhaps be surprising , however , is that some of the variants also show an increased level of the primary activity , indicating the possibility of the simultaneous enhancement of the primary and promiscuous functions . In order to assess the full range of function modulation achieved by the combinatorial library we have carried out a fit of the experimental data based on the equation: ( 2 ) followed by a reconstruction of the entire library using the values of the fitted parameters . The meaning of the symbols in equation 2 is as follows: Ak is the dependent variable ( activity ) with k being a label that identifies the type of activity ( i . e . , k = “primary” or k = “promiscuous” ) = 1; δi is an independent variable that may take the values 0 or 1 , corresponding to the absence o presence of the mutations at position i; pik is a measure of the effect of the mutation at position on the activity Ak; δij = δi·δj is an independent variable that takes a value of 1 when mutations at positions i and j occur simultaneously ( and takes a value of zero otherwise ) ; pijk is a measure of the effect of the coupling between mutations at positions i and j on the activity k . Equation 2 embodies a comprehensive model that includes the effects of individual mutations ( pik values ) as well as the possibility that mutation effects are non-additive ( pijk≠0 ) . It involves , however , 110 fitting parameters ( 10 pik parameters and 45 pijk parameters for each of the two activities ) , while the number of experimental values to be fitted is only 58 ( i . e . , the values of the two activities –primary and promiscuous- for the 29 library variants studied ) . Having more fitting parameters than dependent variable values is a common occurrence in chemometrics , often addressed using partial least-squares [27] , [28] ( PLS ) , a dimensionality reduction approach akin to principal component analysis . Indeed , the widespread usefulness of the PLS approach is often credited to its ability to handle a large number of independent variables ( i . e . , fitting parameters ) ( see chapter 7 in Livingstone [44] . PLS thus uses latent variables ( latent vectors ) : orthogonal combinations of the original variables that explain most of the variance in the original independent variable set and are also constructed to maximize their covariance with the dependent variables . The original variables may then survive the PLS dimensionality reduction , but they are combined in a few relevant latent vectors . In the case of interest here , once a PLS fit of equation 2 to the experimental data for the 29 variants studied ( Figure 5A ) has been performed ( see Methods for details ) , it is straightforward to calculate the expected primary and promiscuous activity data for the whole library of 1024 variants . Of course , there remain two important issues related to the assessment of the uncertainty associated to such full-library reconstruction and to its experimental validation . To assess reconstruction uncertainty , we have used a bootstrapping approach involving PLS fits to replica sets obtained by randomly re-sampling from the original experimental set ( see Methods for details ) . Full-library reconstructions resulting from the PLS analyses of 20 such replicas are given in Figure 5B . They clearly suggest that the mutation set derived from the statistical coupling analysis potential has a huge potential for modulating both , the primary and promiscuous activities . Experimental validation of the reconstruction ( based on experimental measurements on the predicted Pareto set of optimal variants ) is described in the following section . We derived an “optimistic” prediction of the Pareto set of optimal primary/promiscuous activities ( Figure 5C ) as the set of non-dominated solutions in the ensemble of reconstructions shown in Figure 5B . The 11 variants in this predicted set were prepared and their activities determined experimentally . There is an excellent qualitative agreement between prediction an experiment , in the sense that , for all the 11 variants , increased levels of both activities were found ( Figure 5D ) . This agreement validates the reconstruction carried out on the basis of the PLS analysis of the 29-variants set . It is relevant to note at this point that the PLS-reconstruction/Pareto-prediction analysis leads to an expansion of our experimental variant set ( from 29 variants to 40 variants ) but in a manner that is not random . Actually , the 11 variants added to the experimental set allow us to move in the space of primary/promiscuous activities in the general direction of the simultaneous enhancement of both activities , as is visually apparent in Figures 5E and 5F . The Pareto set from the experimental data for the 29+11 = 40 variant set ( Figures 5E and 5F ) is still only an approximation to the Pareto set for the whole library , since additional cycles of PLS-reconstruction/Pareto-prediction could in principle lead to further enhancements in both activities . However , PLS-reconstruction starting with the 40-variant experimental data set suggests that additional improvements are expected to be small ( see Figure 6 ) , supporting that 40-variant Pareto set is likely close to the Pareto set of the full library . Furthermore , the main result of the analysis is already apparent with the 40-variant set: E . coli thioredoxin , as well as the background P34H variant for library construction , does not belong to the Pareto set and , therefore , simultaneous enhancement of the primary and promiscuous activities is feasible ( and has been experimentally achieved: Figures 5E and 5F ) . Note that , in addition to the targeted simultaneous enhancement ( implying enhanced levels for both activities ) , the experimental data set ( as well as the PLS reconstructions of the full combinatorial library ) indicates that the two activities can be modulated in an independent-like manner and includes “specialist” variants with a high level for one activity and low value for the other . Figure 7A highlights the modulation ranges experimentally achieved for the reductase and catalysis of oxidative folding activities ( about 33-fold and 7-fold , respectively ) . One important issue is whether these ranges ( in particular that of the promiscuous activity ) are to be considered large or small . The answer to this question depends largely on the relevant context . Certainly , the modulation ranges we have found are much smaller than those reported in some protein design studies ( consider , for instance , the 200-fold enhancement in engineered Kemp eliminase activity reported by Baker , Tawfik and coworkers [45] ) and the ranges typically considered relevant in a biotechnological application context . It is important to note , however , that the approach we have used is aimed at assessing the patterns of primary/promiscuous activity modulation in the mutational space actually explored by natural evolution when recruiting the promiscuous activity for a new function . That is , the mutations included in our combinatorial library are those expected to be associated to the emergence of folding catalysis in the thioredoxin scaffold during the course of natural evolution . If the approach is successful , promiscuous activities approaching the levels of natural thioredoxin-scaffold folding catalysts should be reached . In an evolutionary/biological context , therefore , the promiscuous activity modulation achieved should be compared with the evolutionary significant modulation range estimated on the basis of activity data for a protein disulfide isomerase . Experimental data for bovine PDI are included in Figure 7 and indeed show an acceptable level of congruence with the Pareto set for the 40-variant variant set ( Figure 7A ) as well as with the corresponding PLS reconstructions ( Figure 7B ) . Clearly , the modulation range achieved for the catalysis of oxidative folding is significant from a biological/evolutionary point of view . Interestingly , this does not appear to hold for the other folding-related activity of thioredoxin domains: the disulfide-reshuffling activity responsible for the rescue of misfolded proteins with incorrect disulfide bridges . Figure 7C is a plot of reshuffling activity ( measured using disulfide-scrambled ribonuclease A as substrate: see Methods for details ) versus catalysis of oxidative folding , including the 40-variant set , wild-type thioredoxin from E . coli , the P34H background variant and bovine PDI . Figure 7D is a similar plot including the PLS reconstructions based on the experimental data of Figure 7C . These two plots suggest that the combinatorial library used ( based on the mutation set derived from SCA analysis: Figure 4 ) spans the evolutionary relevant range for the catalysis of oxidative folding , but not the corresponding range for the reshuffling activity . This result is actually consistent with some known features of the structure-function relationship in protein disulfide isomerases . PDIs have a multidomain structure usually described [46] in terms of four distinct domains ( a b b′ a′ ) , two of which ( the a and a′ domains ) display the thioredoxin-fold structure and the CXXC active site motif responsible for the catalysis of disulfide-linked process . The isolated a and a′ domains have been shown to introduce efficiently disulfide bridges into proteins [47] , while additional domains are required for efficient catalysis of disulfide bond reshuffling in folded proteins [48]–[50] , perhaps because the “inactive” b and b′ domains play a role in facilitating steps that involve difficult conformational changes [48] . Obviously , this “multidomain” effect cannot be reproduced by engineering based on a single-domain thrioredoxin scaffold . The primary/promiscuous plots presented so far ( Figures 5 , 6 and 7 ) , employ logarithmic activity scales in order to emphasize the order of magnitude of the modulations achieved . However , using linear scales in these plots ( Figures 8A and 8B ) reveals a surprisingly simple pattern: a linear-like Pareto set and a roughly triangular shape for the “cloud” of experimental data points below the Pareto set . This pattern is robust , being observed in the 40-variants data set and in the PLS reconstructions of the full combinatorial library . Note also that the observed pattern implies that essentially all the experimental data points are at or below a line connecting the expected maximum values for the primary and promiscuous activities and , therefore , that the experimental data points populate an area in the primary/promiscuous activity plot which is about half the maximum area accessible . The probability of this happening by chance if there is no correlation between the primary and promiscuous activities is on the order of ( 1/2 ) ∧NDP being NDP the number of experimental data points . This gives a negligible probability for NDP = 1024 even for NDP = 40 . Finally , as we have already pointed out , the second-round of the library screening process was sharply focused to the Pareto set and that , as a result , the Pareto set of the 40-varaints experimental set is likely to be close to the Pareto set of the full library . We conclude from all this reasoning that the simple experimental pattern in Figures 8A and 8B is robust and is unlikely to have arisen by chance . It is natural then to seek a simple interpretation for such a simple , but intriguing pattern . As we elaborate below one simple explanation is provided by the so-called conformational diversity hypothesis . The conformational diversity hypothesis posits native proteins may exist in solution as different conformations in equilibrium and provides a plausible structural rationale for the existence of protein promiscuity [5] , [51] , [52] . In very simple terms , the most populated ( i . e . , dominant ) conformation is responsible for the primary activity while alternative , low-population conformations perform the promiscuous activities . Mutations can shift the equilibria between the different conformations and thus modulate the balance between the primary and promiscuous activities . A linear-like Pareto set could thus be explained in terms of two optimal conformations , each being responsible for catalyzing efficiently only one of the activities . For instance , one conformation would achieve molecular optimization for the substrate reduction process ( Figure 2 ) when the active-site disulfide is reduced , while an alternative conformation would achieve optimization for substrate oxidation ( Figure 3 ) when the active-site disulfide is oxidized . Obviously , data points below the Pareto set would correspond to significant population of other conformations that are suboptimal in terms of activity . This interpretation is clarified below with a simple illustrative example . Consider three protein conformations: a0: a conformation with no activity; a1: the conformation responsible for the primary activity; a2: the conformation responsible for the promiscuous activity . The mol fractions of the three conformations must add up to unity: ( 3 ) Mutations may change these mol fractions and , obviously , the optimal primary/promiscuous activity situations will be achieved when X ( a0 ) = 0 and , ( 4 ) Since activities should be proportional to the corresponding mol fractions , equation 4 defines a straight line in a plot of promiscuous versus primary activity . We refer to this line as the “trade-off line” . In the same plot , suboptimal situations in which X ( a0 ) ≠0 will necessarily be represented by points in a triangular area defined by the trade-off line and the plot axes ( see Figure 8D ) . Certainly , the plot in Figure 8D ( showing the optimal trade-off line and a shaded triangular region corresponding to suboptimal situations ) is an idealized representation . In practice , we must consider the possibility that the mutations used are unable to completely shift the equilibria towards the active conformations ( i . e . , they might be unable make the mole fraction of the inactive conformation strictly equal to zero ) . To provide an illustration of this situation , we have carried out a stochastic simulation of a 40-variant data set , assuming that conformation populations are proportional to statistical weights derived from flat distributions . That is , the mol fraction of a given conformation ai is given by wi/Σwi where wi is its statistical weight ( derived from a random number generator in the [0 , 1] interval ) and Σwi is the sum of the statistical weights for all the conformations . The result of this simulation ( Figures 8E ) is a roughly triangular-shaped cloud of data points with a linear-like Pareto set that approaches the trade-off line . For simplicity and illustration , the simulations included in Figure 8 assume that there is only one sub-optimal conformation and that it has zero primary and promiscuous activity levels . It is important to note , however , that the general result of the simulations is robust and it is obtained with different conformation models ( see Figures 9A–C ) including several sub-optimal conformations with non-zero activity levels . Apparently , all that is required for a linear-like Pareto set to be obtained in these simulations is a model with two optimal conformations with quite different capabilities to catalyze the primary and promiscuous processes . Actually , if the two optimal conformations are efficient at catalyzing both activities ( and , therefore , there are no trade-offs ! ) , the pattern of a linear-like Pareto set with a triangular-like data points cloud is not obtained ( see Figure 9D for a representative simulation ) . Certainly , the simulations discussed above ( Figures 8 and 9 ) are not meant to be taken as direct evidence in support of the conformational diversity . In fact , obtaining such direct evidence would require extensive structural and dynamic characterization ( see , for instance [53] and [54] ) which is beyond the scope of this work . Because of this we cannot rule out that other kinds of models ( based , for instance , on modeling the mutation effects on the activities ) could also explain the experimental results founds . Nevertheless , it is clear that the conformational diversity hypothesis provides a simple , Occam-razor explanation ( since modeling of specific mutation effects is not involved ) for an equally simple , but otherwise intriguing , experimental modulation pattern . The simultaneous enhancement of the primary and promiscuous activities we have discussed in the preceding sections is only one aspect ( albeit a prominent one ) of a general property of the set of mutations derived from the function-based statistical coupling analysis: the potential for originating a multiplicity of mutational paths leading to different types of function modulation patterns . To illustrate the idea ( Figure 10 ) we use one of the full-library reconstructions derived from the PLS analysis of the 40-variants experimental set . Each of the mutational paths shown in Figure 10 has been constructed using the following procedure: a ) An initial variant is chosen; b ) the variants connected to the chosen one by single mutations are tested for a given activity-related condition; c ) one variant among those that pass the test is randomly selected; d ) the cycle a-c is repeated until no mutational steps are available . The paths shown in Figure 10A start with the background variant ( i . e . , the variant with no mutations ) and mutational steps are allowed if promiscuous activity ( catalysis of oxidative folding ) is increased while the primary activity ( reductase ) is maintained above a certain threshold . These simulations illustrate the case in which there is selection for enhanced promiscuous activity while maintaining a level of the primary activity that does not compromise fitness . Several paths lead to a variant with increased promiscuous activity and still a significant level of primary activity . Interestingly , PDIs show a significant level of reductase activity ( see published work [47] , [48] , [50] and Figure 7 ) , perhaps because the catalysis of disulfide-linked folding likely involves steps in which incorrect disulfide bridges must be broken up . It is thus tempting to speculate that the no-trade-off paths in Figure 10A illustrate some of the actual function changes taking place in the evolution of these disulfide-linked folding catalysts . It is also interesting that some of the intermediate variants in the paths of Figure 10A have significantly increased levels of both activities , again emphasizing the possibility of the simultaneous enhancement of the primary and promiscuous functions . The Paths shown in Figure 10B use as starting point a variant with comparatively high values of both , the primary activity and the promiscuous activity ( actually , a member of the Pareto set of optimal solutions ) and mutational steps are allowed if the primary activity is decreased while the promiscuous activity remains above a given threshold . The paths in Figure 10B lead to a specialist protein with high promiscuous activity and low primary activity . These paths could be viewed as illustrating the molecular changes that could in some cases occur in one of the gene copies arising from the gene duplication event involved in the emergence of a new function . According the so-called balance hypothesis [55] , single-gene duplication may actually be harmful because it immediately leads to a very large excess of a given protein , which may be deleterious . If imbalance is associated to an excessive level of the primary function , then the mutational paths in Figure 10B illustrate a potential of the mutational space explored by natural selection to efficiently restore balance . A final clarification should be made . The mutational paths in the illustrative simulations summarized in Figure 10 have been obtained assuming that all the single-mutation steps can be readily achieved , although some of them cannot be realized with a single-base substitution . However , these amino acid substitutions do occur during natural evolution ( involving an intermediate amino acid ) as clearly shown by the sequences in Figure 4D . In connection with this , it is important to note that the mutational space we have characterized is very likely a subspace of the full mutational space explored by natural selection in the evolution of disulfide-linked folding catalysts ( the latter involving additional positions and several mutations at each position ) . Obviously , this fact only reinforces our conclusions . Current views on the relation between primary and promiscuous protein activities are derived to a significant extent from laboratory evolution experiments aimed at enhancing promiscuous functions . Many of these studies have found a decrease ( often moderate ) in primary activity concomitantly with the increase in promiscuous function , suggesting that the two activities trade-off . In this work , we have introduced an approach to determine how the interplay between the primary and promiscuous activities of a protein is modulated in the mutational space evolutionary linked to the emergence of a new function . Application of this new approach to the emergence of folding catalysts reveals a hitherto unexplored scenario: diverse patterns of primary/promiscuous activity modulation may occur as response to different types of evolutionary pressure , including no-trade-off paths involving the simultaneous enhancement of both activities . Some general remarks related with this result are appropriate:
BLAST2 ( 1996–2003 , W . Gish http://blast . wustl . edu ) was used to search the TrEMBL sequence database of October-2007 ( http://www . ebi . ac . uk/trembl ) using the sequence of E . coli thioredoxin as query . The resulting sequences were aligned with the query sequence using the Smith-Waterman algorithm and only those with sequence identity with the query of 0 . 3 or higher were retained for further analysis . We made no further attempt to correct or filter the alignment , since the results obtained from its analysis made clear sense from both , the structural and functional viewpoints ( see Figures 4 and 5 ) . Of the1440 sequences in the alignment used , 1264 had a proline at position 34 and 132 had a histidine at that position . Essentially all the sequences with histidine at position 34 belonged to eukaryotes and most of them were actually annotated as protein disulfide isomerases . Statistical coupling analysis of the sequence alignments based on the P34→H perturbation were performed using homemade programs , but in a manner identical to that described by Lockless and Ranganathan [24] . The robustness of this analysis is supported by the fact that the positions with high values for the statistical coupling energy also rank high in a simple covariance analysis [58] of the sequence alignments ( see Figure S1 in Supporting Information ) . The combinatorial library of thioredoxin variant sequences on the P34H background was prepared by using gene assembly mutagenesis as we have previously described [59] . For ease of protein purification , the genes encoded a His6 tag at the N-terminal end ( i . e . , at a position roughly opposite to the active-site region ) . Purification of the thiredoxin variants , assessment of their purity and concentration measurements , were performed as previously described [59] . Bovine PDI was purchased from Sigma and used without further purification . Reductase activity of the thioredoxin variants was determined at 37°C by a turbidimetric assay of the thioredoxin catalyzed reduction of insulin [60] . Briefly , thioredoxin-variant ( or PDI ) solutions at pH 6 . 5 ( phosphate buffer 0 . 1 M ) in the presence of 2 mM EDTA and 0 . 5 mg/mL insulin were prepared . The reactions were initiated by addition of DTT to a 1 mM final concentration and monitored by measuring the absorbance at 650 nm ( A650 ) as function of time . Activity is calculated as the maximum value of the change of A650 with time , i . e . , the maximum value for the derivative dA650/dt ( see Figures S2 and S3 in Supporting Information for representative examples ) . Typically , for each variant , 3–4 experiments at different thioredoxin-variant concentrations ( within the 0–5 µM range ) were carried out and the specific activity values , together with its associated standard errors , were determined from linear fits to the activity versus concentration profiles: see Figure S4 in Supporting Information for representative examples and for further details . The catalysis of oxidative folding activity was determined by following the recovery of ribonuclease A ( RNase A ) activity from completely reduced RNase A following the procedure described by Lundström et al . [43] . Briefly , nitrogen-saturated solutions of thioredoxin variants ( or PDI ) in 0 . 1 M phosphate buffer pH 7 in the presence of 1 mM EDTA and 100 µM oxidized glutathione were prepared . The reaction was initiated by addition of RNase A from a stock solution to a final concentration of 0 . 4 mg/mL . After 1 hour incubation at 37°C , the RNase A activity was determined using the standard assay based on the hydrolysis of 2′-3′-cCMP . Typically , for each variant , 4 experiments at different thioredoxin-variant concentrations ( within the 0–15 µM range ) were performed and the specific activity values , together with its associated standard errors , were determined from linear fits to the recovered RNase A activity versus concentration profiles . Assays for the disulfide reshuffling activity were carried out in the same way , except that disulfide-scrambled RNase A was used and 100 µM reduced glutathione was included in the reaction solution . Fully-reduced and scrambled RNase A were prepared as described by Lündstrom et al . [43] See Figures S5 and S6 in Supporting Information for representative examples of the disulfide-linked folding assays and for further details . PLS analyses were carried out with the program Unscrambler X from CAMO software using the NIPALS algorithm . In all cases , the dependent variables were the logarithms of the values for the primary and promiscuous activities and were auto-scaled ( i . e . , they were subjected to mean subtraction followed by division by the standard deviation ) prior to the analysis . Leave-one-out cross-validation was used and the number of latent variables retained was the optimum value suggested by the Unscrambler program on the basis of the mean square error of cross-validation . Actually , the PLS analyses were carried with 20 replica sets constructed from the original set through random re-sampling ( bootstrapping ) and the number of latent variables retained did depend on the replica set used; typical values , however , were on the order of 3–11 ( i . e . , much smaller than the numbers of dependent and independent variables involved ) . Illustrative plots experimental versus predicted activities are given in Figure S7 in Supporting Information . | Interpretations of evolutionary processes at the molecular level have been determined to a significant extent by the concept of “trade-off” , the idea that improving a given feature of a protein molecule by mutation will likely bring about deterioration in other features . For instance , if a protein is able to carry out two different molecular tasks based on the same functional site ( competing tasks ) , optimization for one task could be naively expected to impair its performance for the other task . In this work , we report a computational/experimental approach to assess the potential patterns of modulation of two competing molecular tasks in the course of natural evolution . Contrary to the naïve expectation , we find that diverse modulation patterns are possible , including the simultaneous optimization of the two tasks . We show , however , that this simultaneous optimization is not in conflict with the trade-offs expected for two competing tasks: using the language of the theory of economic efficiency , trade-offs are realized in the Pareto set of optimal variants for the two tasks , while most protein variants do not belong to such Pareto set . That is , most protein variants are not Pareto-efficient and can potentially be improved in terms of several features . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Methods"
]
| [
"biology"
]
| 2012 | Probing the Mutational Interplay between Primary and Promiscuous Protein Functions: A Computational-Experimental Approach |
Atrial fibrillation ( AF ) is the most common cardiac arrhythmia , but our knowledge of the arrhythmogenic substrate is incomplete . Alternans , the beat-to-beat alternation in the shape of cardiac electrical signals , typically occurs at fast heart rates and leads to arrhythmia . However , atrial alternans have been observed at slower pacing rates in AF patients than in controls , suggesting that increased vulnerability to arrhythmia in AF patients may be due to the proarrythmic influence of alternans at these slower rates . As such , alternans may present a useful therapeutic target for the treatment and prevention of AF , but the mechanism underlying alternans occurrence in AF patients at heart rates near rest is unknown . The goal of this study was to determine how cellular changes that occur in human AF affect the appearance of alternans at heart rates near rest . To achieve this , we developed a computational model of human atrial tissue incorporating electrophysiological remodeling associated with chronic AF ( cAF ) and performed parameter sensitivity analysis of ionic model parameters to determine which cellular changes led to alternans . Of the 20 parameters tested , only decreasing the ryanodine receptor ( RyR ) inactivation rate constant ( kiCa ) produced action potential duration ( APD ) alternans seen clinically at slower pacing rates . Using single-cell clamps of voltage , fluxes , and state variables , we determined that alternans onset was Ca2+-driven rather than voltage-driven and occurred as a result of decreased RyR inactivation which led to increased steepness of the sarcoplasmic reticulum ( SR ) Ca2+ release slope . Iterated map analysis revealed that because SR Ca2+ uptake efficiency was much higher in control atrial cells than in cAF cells , drastic reductions in kiCa were required to produce alternans at comparable pacing rates in control atrial cells . These findings suggest that RyR kinetics may play a critical role in altered Ca2+ homeostasis which drives proarrhythmic APD alternans in patients with AF .
Atrial fibrillation ( AF ) is currently the most common cardiac rhythm disorder , posing a significant medical and economic challenge for the US health care system [1] , [2] . This burden is likely to increase as the population ages and AF prevalence rises [3] . Effective prevention and treatment of AF depends upon advances in our understanding of underlying disease mechanisms . Although several features of AF electrophysiological remodeling have been identified over the past decades [4] , [5] , our knowledge about the arrhythmogenic substrate remains incomplete . Beat-to-beat alternation in the shape of cardiac electrical signals , a phenomenon called alternans , has been observed in the atria of AF patients , but the mechanism underlying these alternans is not known [6]–[11] . Narayan et al . reported differences in the rate dependence of action potential duration ( APD ) alternans in patients , with APD alternans occurring at pacing rates near rest in AF patients but only at fast pacing rates in controls [8] . Narayan et al . also found that APD alternans always preceded AF initiation , indicating that alternans may play an important role in establishing the arrhythmogenic substrate and creating vulnerability to AF . Thus , a better understanding of AF arrhythmogenesis will likely depend upon identification of the mechanism driving atrial alternans at heart rates near rest . Interestingly , in AF patients the slope of the APD restitution curve was <1 during APD alternans onset at slow pacing rates . This suggests that a cellular mechanism other than voltage-driven instability underlies APD alternans at heart rates near rest [9] . Altered Ca2+ handling in atrial myocytes is known to play a crucial role in the generation of AF triggers and in AF maintenance [12] , [13] . Ca2+ cycling instabilities have been shown to underlie ventricular alternans in heart failure [14] , [15] , as well as atrial alternans in several non-AF animal models [16]–[18] . However , it is unknown whether these represent a plausible mechanism for atrial alternans in AF patients , particularly at heart rates near rest . We therefore sought to determine , using a computer model of human atrial tissue , whether Ca2+ handling abnormalities , or other electrophysiological changes that occur in AF , lead to APD alternans . We identified a critical change in the kinetics of the ryanodine receptor ( RyR ) that was responsible for APD alternans onset at slower pacing rates , and subsequently aimed to elucidate the mechanistic relationship between this disruption in RyR kinetics and alternans onset . To this end , we employed single-cell clamping of ionic model parameters and iterated map analysis in order to dissect the mechanisms which drive alternans in atrial tissue , as well as to provide important insights into the pathophysiological changes that contribute to the development of alternans in AF patients .
In order to investigate ionic mechanisms in human AF that contribute to the generation of atrial APD alternans at the tissue level , we created a computer model of human atrial tissue incorporating ionic remodeling associated with chronic AF ( cAF ) , as described in Methods . The sensitivity of APD alternans to ionic model parameters was evaluated by varying parameters one at a time and applying the clinical pacing protocol used by Narayan et al . to induce APD alternans in AF patients [8] ( see Table 1 and Methods ) . For control , a model of normal human atrial tissue was also simulated . We then assessed the magnitude and onset pacing cycle length ( CL ) of APD alternans by analyzing voltage traces from the recording electrode ( Fig . 1A ) , as outlined in Methods . In the control model , significant APD alternans did not occur before loss of capture at 260 ms CL ( Fig . 1B ) . However , in the cAF-remodeled tissue preparation , significant APD alternans appeared at a CL of 240 ms ( Fig . 1B ) . Varying the RyR inactivation rate constant ( kiCa ) had the greatest effect on alternans onset CL in the human cAF-remodeled tissue ( Fig . 2A ) . In fact , only reduction of kiCa resulted in alternans onset at CLs of 300–500 ms ( Fig . 2B ) , matching alternans onset CLs observed in AF patients [8] . When other ionic model parameters were varied from their original cAF values , APD alternans either did not appear in the tissue model at CL≥300 ms ( Fig . 2A , blue areas ) , appeared only at CL≤350 ms ( Fig . 2A , red areas ) , or did not appear before loss of capture or conduction block occurred in the tissue ( Fig . 2A , white spaces ) . These results suggest that altered RyR kinetics is the critical cellular component underlying the occurrence of APD alternans in AF patients at pacing rates near rest , and that kiCa plays a key role in this process . We also tested whether differences between left and right atrial electrophysiology affect alternans susceptibility using a right atrium ( RA ) version of the cAF model [19] in tissue simulations . Results for RA tissue were very similar to those for the left atrium ( LA ) , demonstrating that modulation of kiCa could reproduce alternans observed at pacing rates near rest in both the LA and RA of AF patients [8] ( S2 Figure ) . When kiCa was decreased by 50% in the cAF model ( we refer to this as the cAFalt ionic model ) , APD alternans onset data from the human AF tissue model agreed well with data from persistent AF patients . Significant APD alternans began at 400-ms CL ( Fig . 1B , dotted red line ) , mean APD at onset was 229 ms , and APD alternans magnitude at onset was 27 ms ( Fig . 1C , dotted red line ) . These metrics were each within one standard deviation ( SD ) of clinical observations [8] ( Fig . 3 ) . The cAFalt model also displayed noticeable alternans in intracellular Ca2+ ( [Ca2+]i ) at the onset CL ( Fig . 1D ) . For both the cAF and cAFalt models , mean APDs were shorter than in the control model ( Fig . 1B–C ) , and diastolic and systolic [Ca2+]i were lower than in control ( Fig . 1D ) . At 400-ms CL in the cAFalt model , on the odd ( long ) vs . the even ( short ) beat ( Fig . 4 , blue vs . red ) , there was higher sarcoplasmic reticulum ( SR ) Ca2+ load before release ( 0 . 288 vs . 0 . 273 mM ) , higher peak RyR open probability ( RyRo ) ( 9 . 0e-4 vs . 4 . 7e-4 ) , a larger intracellular Ca2+ transient ( CaT ) amplitude ( Δ[Ca2+]i = 0 . 13 vs . 0 . 067 µM ) , similar L-type Ca2+ ( LCC ) current ( integrated over one beat: 144 vs . 140 mC/F ) , and increased Na+/Ca2+ exchanger ( NCX ) current ( INCX , integrated over one beat: 98 . 4 vs . 74 . 5 mC/F ) . The positive coupling between transmembrane potential ( Vm ) and Ca2+ , with INCX as the primary electrogenic current , is consistent with experimental findings [20] . Since the magnitude and onset of APD alternans in the cAFalt model provided the best agreement with clinical APD alternans data ( Fig . 3 ) , we chose to use this model for subsequent investigations into the underlying causes of alternans occurrence . Since APD alternans throughout the homogenous cAFalt tissue preparation were concordant and of similar magnitude ( S3 Figure ) , electrotonic effects and CV restitution were excluded as factors influencing these alternans . Indeed , APD and CaT alternans in the cAFalt tissue model were very similar to alternans in the isolated single-cell cAFalt model ( Fig . 5 , left column vs . Fig . 4 , top row ) . We therefore concluded that cellular mechanisms gave rise to alternans in the cAFalt tissue model and decided to utilize single-cell simulations in order to investigate these mechanisms . We first used the ionic model variable clamping protocol described in detail in Methods . The percent change in APD and CaT alternans magnitudes , when each ionic model variable was clamped to its trace from either the even ( short ) or odd ( long ) steady-state beat at the alternans onset CL ( 400 ms ) , are summarized in Fig . 6 ( right column: state variables , left column: currents and fluxes ) . Variables which resulted in >99% reduction in APD and CaT alternans magnitudes for both even and odd beat clamps were considered essential for alternans . Clamping Vm resulted in −61 . 8% change in CaT alternans magnitude for even beat clamps and +6 . 6% for odd beat clamps , demonstrating that the alternans were not voltage-driven ( see even and odd beat clamps depicted in column 2 of Fig . 5 and S4 Figure , respectively ) . Clamping [Ca2+]i enhanced APD alternans ( +55 . 2% and +75 . 8% for even and odd beat clamps , respectively , column 3 of Fig . 5 and S4 Figure ) . However , when SR Ca2+ ( [Ca2+]SR ) was clamped to either the even or odd beat waveforms , alternans in both APD and CaT were eliminated ( <−99% ) , demonstrating that the alternans were driven by SR Ca2+ instability ( column 4 of Fig . 5 and S4 Figure ) . In addition , four other variables could be clamped to the even or odd beat waveforms to eliminate APD and CaT alternans: RyR inactivated probability ( RyRi ) , RyR open probability ( RyRo ) , junctional Ca2+ ( [Ca2+]j ) , and SR Ca2+ release flux ( JSRCarel ) ( Fig . 6 , and S5 and S6 Figures ) . All five of these variables were therefore critical for enabling alternans to occur at the onset CL . Furthermore , these variables directly impact SR Ca2+ release , implicating SR Ca2+ release as the underlying source of alternans in the cAFalt model . There were two ionic model components which greatly reduced but did not eliminate alternans when clamped: sub-sarcolemmal Ca2+ ( [Ca2+]sl ) and sub-sarcolemmal Na+/Ca2+ exchanger current ( INCXsl ) . Clamping [Ca2+]sl to the even beat eliminated all alternans; clamping to the odd beat greatly reduced APD and CaT alternans ( −95 . 8% and −96 . 2% , respectively ) , although large alternation in SR load persisted ( Fig . 6 and columns 1–2 of S7 Figure ) . Similarly , clamping INCXsl to the even beat waveform resulted in elimination of APD but not CaT alternans ( +72 . 9% ) , while clamping to the odd beat waveform resulted in elimination of all alternans ( Fig . 6 and columns 3–4 of S7 Figure ) . Hence , the SR Ca2+-driven instabilities produced alternans in Ca2+ cycling which were positively coupled to voltage through INCXsl and [Ca2+]sl . Increased steepness of the SR release-load relationship is a well-known mechanism for CaT alternans [21] , [22] . The importance of SR Ca2+ release variables for APD and CaT alternans , as demonstrated by the results in Fig . 5 , 6 , and S4 , S5 , S6 Figures , led us to hypothesize that such a mechanism might give rise to Ca2+-driven alternans in the cAFalt model at pacing rates near rest . To test this , we compared the cAF and cAFalt ionic models under action potential ( AP ) voltage clamp conditions so that changes in CaT alternans would be due solely to changes in Ca2+ homeostasis rather than bidirectional coupling between Vm and Ca2+ . After clamping each ionic model at a CL of 400 ms until steady state was reached , we perturbed [Ca2+]SR and tracked SR load and SR Ca2+ release on the subsequent clamped beats ( see Methods for details ) . The SR release-load relationships for the cAF ( black ) and cAFalt ( red ) ionic models are depicted in Fig . 7 ( left column , row 1 ) . The slope of the release-load relationship in the cAFalt model ( = 3 . 1 ) was much greater than the slope in the cAF model ( = 1 . 7 ) , confirming our hypothesis that differences between the cAF and cAFalt ionic models led to a steepening of the SR Ca2+ release slope . To better explain the differences between the cAF and cAFalt ionic models that gave rise to different SR Ca2+ release slopes , we first compared [Ca2+]SR , RyRo , [Ca2+]j , and cumulative Ca2+ release for the two models at steady state ( Fig . 7 , left column , rows 2–5 , solid lines ) . In the cAFalt model , [Ca2+]SR at steady state was 19 . 7% lower than in the cAF model as a result of increased RyR opening ( Fig . 7 , left column , rows 2 and 3 , red vs . black solid lines ) . Although this led to a 15 . 2% decrease in peak [Ca2+]j in the cAFalt model , the duration of the release event was prolonged ( Fig . 7 , left column , row 4 , red vs . black solid lines ) . Consequently , though cumulative Ca2+ release in the cAFalt model initially lagged behind , at t≈90 ms it actually surpassed the cumulative release in the cAF model , ultimately resulting in a 3 . 4% increase in total release by the end of the beat ( Fig . 7 , left column , row 5 , red vs . black solid lines ) . To illustrate how these differences between the cAF and cAFalt ionic models impacted SR release slope , we applied a large perturbation to [Ca2+]SR ( +20 µM ) at the beginning of a clamped beat and compared the unperturbed ( steady state , solid line ) and perturbed ( dotted line ) traces for each model ( Fig . 7 , left column , rows 2–6 ) . Higher SR load at the beginning of the beat led to increased SR release flux due to luminal Ca2+ regulation of the RyR ( causing more opening ) , as well as to the increased concentration gradient between the SR and junctional compartments . In both the cAF and cAFalt models , these changes led to increased peak [Ca2+]j ( +54 . 4% and +100% , respectively ) and RyR opening ( +64 . 6% and +129% , respectively ) as a result of more Ca2+-induced Ca2+ release ( Fig . 7 , left column , rows 2–4 ) . The positive feedback relationship between [Ca2+]j and RyR opening was strong enough such that when SR load was increased ( Fig . 7 , left column , row 2 , dotted vs . solid lines ) , this actually resulted in a lower minimum [Ca2+]SR during release ( −3 . 6% and −13 . 3% for cAF and cAFalt models , respectively ) . However , the amount of positive feedback differed between the cAF and cAFalt ionic models . Positive feedback amplifies changes in release inputs , such as SR load; therefore , in the cAF model , where [Ca2+]j is higher and positive feedback is stronger , the increase in [Ca2+]SR produced a slightly greater change in release ( compared to the unperturbed , steady state simulation ) during the rising phase of [Ca2+]j ( t<48 ms ) than in the cAFalt model ( Fig . 7 , left column , row 6 , black vs . red ) . By contrast , termination of release occurs through a negative feedback process , with RyRs inactivating upon the binding of junctional Ca2+ . Negative feedback attenuates changes in release so that robust , fast termination of release is achieved even when a disturbance ( such as a transient increase in SR load ) occurs . In the cAFalt model , negative feedback is decreased both directly , via reduction of kiCa , and indirectly , via reduction in [Ca2+]j that occurs as a result of decreased SR load . This causes prolongation of the Ca2+ release event and a larger peak [Ca2+]j ( Fig . 7 , left column , row 4 , red vs . black dotted lines ) . Consequently , when SR load was increased by the same amount in the cAF and cAFalt models , although the cAFalt model had a lesser initial change in release because of weaker positive feedback , it also had a greater final change in release , i . e . a steeper SR release-load relationship , because of weaker negative feedback ( Fig . 7 , left column , row 6 , red vs . black ) . The results in column 1 of Fig . 7 demonstrate how the steeper SR release slope in the cAFalt ionic model ( as compared to the cAF ionic model ) depends upon RyR inactivation by junctional Ca2+ . However , recent work suggests that termination of release does not rely on direct Ca2+-dependent inactivation of the RyR but rather on local SR Ca2+ depletion [23]–[26] . In order to test whether steepening of the SR release slope could occur in the cAF model by an alternative release termination mechanism , we implemented a version of the cAF model in which the RyR Markov model was replaced with that of Sato and Bers and the SR was divided into junctional ( JSR ) and network ( NSR ) compartments [27] ( see Table 2 and S1 Text ) . Termination of release in this alternative RyR model relies on calsequestrin ( CSQN ) binding to the RyR , which occurs as luminal [Ca2+] decreases causing changes in RyR opening and closing rates . The effects of decreased RyR termination in the Sato-Bers RyR model are shown in the right column of Fig . 7 . When the CSQN-bound RyR closing rate k34 ( analagous to the inactivation rate kiCa in the original model ) is decreased from 100% to 50% ( cAFalt ) , steady-state Ca2+ concentrations change modestly as compared to the original RyR formulation ( Fig . 7 , black vs . red solid lines ) , but nevertheless display similar trends: [Ca2+]JSR decreases by 1 . 5% ( vs . 19 . 7% , row 2 ) , peak [Ca2+]j is reduced by 10 . 5% ( vs . 15 . 2% , row 4 ) and delayed , and total release increases by 3 . 6% ( vs . 3 . 4% , row 5 ) . When [Ca2+]NSR is perturbed in the Sato-Bers models by +20 µM , Ca2+ release increases more in the cAFalt model than in the cAF model ( Fig . 7 , right column , row 6 , red vs . black dotted lines ) . Consequently , the SR Ca2+ release slope is steeper in the cAFalt model ( = 3 . 7 vs 1 . 9 , Fig . 7 , right column , row 1 ) . Thus , although changes in SR Ca2+ release slope in the original cAF model are caused by altered junctional Ca2+-dependent inactivation , altered SR Ca2+-dependent mechanisms of release termination can produce such changes in SR Ca2+ release slope as well . Although SR Ca2+ release slope is an important component of Ca2+ homeostasis , other aspects of Ca2+ cycling , such as SR Ca2+ uptake , could also have a significant impact . In order to understand how both SR release and uptake contribute to CaT alternans onset at slow pacing rates in human cAF cells , we used an iterated map analysis for investigating Ca2+ cycling stability under AP voltage clamp conditions . Three factors affecting Ca2+ cycling stability were included in the analysis: SR release , SR uptake , and cellular Ca2+ flux across the sarcolemma . The latter factor was included because Ca2+ content in the human atrial cell model varied significantly enough to affect alternans threshold predictions . For each version of the human atrial cell model ( cAF and control ) , we calculated the SR Ca2+ release slope ( ) , the SR Ca2+ uptake factor ( ) , and the cellular Ca2+ efflux factor ( ) [28] , [29] for a range of kiCa values and pacing rates and compared the value of to the threshold for alternans . For a typical range of parameter values ( , see S1 Text ) , the threshold value of required for alternans is given by the following equation: ( 1 ) Theoretical analysis predicts that the system is stable when . Eq . 1 is graphed for a range of values in Figs . 8A–C ( dotted lines ) . Each curve represents the boundary between stable ( no alternans ) and unstable ( alternans ) Ca2+ cycling in the - plane for a particular value of . As increases ( Fig . 8A–C , dark blue to dark red ) , the threshold curve steepens , indicating that increased Ca2+ extrusion from the cell has a protective effect , helping to restore Ca2+ content back to steady state following a perturbation . Thus , a higher value of is required to reach alternans threshold for higher values of . Note that in this theoretical approach , increased Ca2+ efflux ( κ ) has the opposite effect as in Qu et al . [29] , suppressing rather than promoting Ca2+ alternans . The effects of changing CL and changing kiCa are explored for the cAF model in Fig . 8A . At the default kiCa value ( 100% ) , as CL is decreased from 700 ms to 200 ms ( −10 ms increments ) , decreases , increases , and the system approaches the alternans threshold given by Eq . 1 . The change in values is non-monotonic , initially decreasing ( orange to green ) and then increasing ( green to orange ) as CL is decreased . However , the change in has a minimal effect at small values , since the threshold curves for different values converge at . At CL<220 ms , the cell begins to display alternans in Ca2+ cycling , coinciding with the iterated map parameter values residing very close to the theoretically predicted boundary given by Eq . 1 ( Fig . 8A , orange X's ) . When kiCa is set at 50% of the default cAF value ( cAFalt model ) , a similar trend is observed . However , the 50% kiCa cAF model reaches threshold at a lower pacing rate ( CL = 390 ms for the 50% kiCa cAF model vs . 210 ms for the 100% kiCa cAF model , Fig . 8A , X's ) . This is primarily due to increasing as kiCa is decreased , illustrated by the trajectory of the system in the - plane as CL is held constant at 390 ms but kiCa is decreased from 100% to 50% ( Fig . 8A ) . We next performed the same iterated map analysis for the control atrial cell model with varying CL and kiCa values ( Fig . 8B ) . When kiCa is at 100% , decreases as CL is decreased . However , unlike in the cAF model , in the control case the value of undergoes a net decrease as CL shortens from 700 to 200 ms . Ultimately , since both and decrease as CL is shortened , the control atrial cell ( with kiCa at 100% ) fails to reach threshold and remains in the stable , no alternans region . This suggests that alternans in control patients , which occur at CL<250 ms [8] , are driven by voltage rather than Ca2+ . As in the cAF model , the alternans threshold CL in the control model can be adjusted by modulating the value of kiCa ( Fig . 8B , CL = 390 ms ) . However , in the control model , kiCa must be decreased much more than in the cAF model in order to reach at a CL of 390 ms ( kiCa reduced to 16% vs . 50% ) . The need for dramatic and possibly unrealistic reductions in kiCa to produce alternans at slow rates in control is consistent with the absence of alternans observed in control patients at CL≥250 ms [8] . To explain the difference in Ca2+ cycling properties of the cAF and control models , we examined the effects of cAF cellular remodeling on iterated map parameters . Stochastic ionic model parameter variation and regression analysis [30] ( see S1 Text ) predicted that of the ten model parameters altered in the control model to construct the cAF model , seven would have significant effects on alternans threshold CL ( these are gCaL , gKur , koCa , IbarNCX , gto , gK1 , and gNa , see S8 Figure ) . Of these seven parameters , three are involved in Ca2+ handling ( gCaL , koCa , and IbarNCX ) . The effects of changing these three parameters from control to cAF values is depicted sequentially in Fig . 8C: starting with the default values for the control cell at a CL of 390 ms , first gCaL is decreased and then IbarNCX and koCa are increased to cAF values , resulting in an overall decrease in and . Finally , when kiCa is decreased to the cAFalt value ( 50% ) , the large increase in causes the system to reach and alternate ( Fig . 8C , red X ) . This illustrates why the control cell is less susceptible to CaT alternans than the cAF cell: at a given kiCa value and pacing rate , SR uptake efficiency ( ) is higher in the control model , thus requiring a large increase in the pacing rate ( which decreases ) and/or a large decrease in kiCa ( which increases ) in order to reach . Of the three cAF parameters which decrease , however , gCaL is the most important for alternans onset , since remodeling of IbarNCX and koCa decreases , while remodeling of gCaL increases . When gCaL is remodeled and IbarNCX and koCa remain at control values , only a 28% decrease in kiCa is required to reach ( Fig . 8C , green X ) .
The first goal of this study was to identify the electrophysiological changes in human atrial cells that are responsible for the occurrence of APD alternans at heart rates near rest , as observed in AF patients . Using parameter sensitivity analysis , we found that of the 20 electrophysiological model variables tested , only changes in the RyR inactivation rate constant ( kiCa ) could produce APD alternans at relatively slow pacing rates in a tissue model of persistent/chronic AF . In particular , decreasing kiCa by 50% ( the cAFalt model ) produced a good match to clinical data . We next aimed to provide mechanistic insight into why disruption of RyR kinetics , together with other electrophysiological changes occurring in AF , leads to alternans onset at pacing rates near rest . We established that alternans in the cAFalt model at the onset CL were Ca2+-driven rather than voltage-driven , and that they depended upon SR Ca2+ release . Furthermore , CaT alternans occurred in the cAFalt model at relatively long CLs because of steep SR Ca2+ release slope and decreased SR Ca2+ uptake efficiency . Lastly , we demonstrated that the ability to generate alternans at slower pacing rates by modulating kiCa depended upon the negative feedback properties of SR Ca2+ release . This study is the first to identify a possible mechanism for alternans occurring at slow heart rates in AF patients . Our novel findings show that alternans at slow rates is Ca2+-driven , brought about by AF-associated remodeling of the Ca2+ handling system in atrial cells . Clinical and experimental research has shown that atrial alternans is associated with disease progression in AF patients [8] and with increased AF susceptibility after myocardial infarction [31] , [32] and atrial tachycardia [33] , [34] in animal models . Additionally , CaT alternans have been studied in animal atrial myocytes [17] , [18] , [35] and in the intact atria of AF-prone mice [36] . However , the precise cellular mechanism underlying alternans at heart rates near rest in the remodeled human atria has not been previously identified , and a direct relationship between human AF and CaT alternans in the atria has not been established until now . Elucidating the mechanism driving alternans at slow rates is particularly important because APD oscillations appear to be closely linked to AF initiation [8] . If APD alternans play a direct role in AF initiation , the onset of alternans at slower pacing rates would indicate an increased susceptibility to arrhythmia in AF patients , consistent with clinical observations [8] . Identification of this mechanism would thus provide a significant scientific and clinical benefit , improving our understanding of arrhythmogenesis and aiding in the development of new targeted therapies for AF . In this study , we demonstrate how different aspects of AF remodeling contribute to Ca2+-driven alternans onset at slower heart rates using a theoretical analysis of Ca2+ cycling . This analysis allowed us to quantitatively assess CaT alternans threshold under AP voltage clamp conditions in a detailed electrophysiological model , providing valuable insights into the effects of AF electrophysiological remodeling on Ca2+ handling and alternans . Furthermore , we identify a critical aspect of SR Ca2+ release—inactivation of the RyR—which is necessary for CaT alternans to occur at slow heart rates . These findings extend mechanistic insight about proarrhythmic ventricular Ca2+ remodeling [15] , [37] , [38] to the atria and may inform new therapeutic strategies to target the RyR and suppress Ca2+-driven alternans in the atria for the purposes of preventing or treating AF [36] , [39] . The RyR has been the focus of several studies concerning trigger-mediated AF . In particular , disruption of RyR regulation has been shown to promote AF through increased RyR open probability , diastolic SR Ca2+ leak , and delayed afterdepolarizations [12] , [39] , [40] . Here we identify an additional pathological consequence of the disruption of RyR regulation in AF: Ca2+-driven alternans . Similar to what has been demonstrated with regards to Ca2+ sparks and triggered activity [39] , we found that CaT alternans is coupled to voltage primarily through upregulated INCX , thus driving the generation of APD alternans . The RyR's central role in both alternans and triggers has important clinical implications , given the proarrhythmic consequences of interaction between ectopic activity and the arrhythmogenic substrate created by voltage alternans [41] . New drug treatments to restore the normal function of the RyR and NCX , and thereby prevent arrhythmogenic triggers and alternans , have the potential to provide more effective alternatives to current AF drug therapies which target voltage-gated ion channels and often have proarrhythmic side effects [39] . The signaling pathways involved in RyR dysfunction in AF have been the focus of much active research over the past several years [39] , [40] . Possible molecular mechanisms which could account for reduced RyR inactivation include RyR hyperphosphorylation by CAMKII and PKA and dissociation of the RyR subunit FKBP12 . 6 , which have been shown to increase RyR open probability and promote arrhythmia [42] , although the exact role of these mechanisms in RyR dysregulation are still debated [43] . Calmodulin has also been shown to interact directly with the RyR to decrease its open probability [44] . Metabolic factors may play a role , since modulation of the RyR as a result of glycolytic inhibition has been linked to atrial alternans in non-AF animal models [16] , [17] , [35] . Such metabolic impairment is thought to contribute to profibrillatory remodeling in the atria [45]–[47] . The cAFalt model , with its reduction in kiCa , can be considered a phenomenological representation of the various signaling pathway disruptions leading to alternans , which were not represented in the original cAF model . As more information becomes available , incorporation of these signaling mechanisms into computational models may provide additional insights into how reduction in RyR inactivation leads to Ca2+-driven alternans at slow heart rates in AF patients . There is debate over whether CaT alternans depend primarily on SR Ca2+ load alternation or on RyR refractoriness [21] , [41] , [48] . Recent experiments [18] , [49] and simulation studies [50]–[53] have shown that RyR refractoriness can drive CaT alternans under conditions where near-identical SR loads produce different amounts of SR release . In some simulation studies , this phenomenon was restricted to limited parameter values , clamping conditions , and cycle lengths [51] , [52] , while in a more recent modeling study focusing on atrial cells , SR load-independent alternans occurred over a broad range of pacing rates when the number of t-tubules was reduced [53] . Of note is the fact that many of these studies [51]–[53] utilized the same RyR gating scheme as this current study , yet they identified various mechanisms for CaT alternans . This demonstrates that the relative importance of the various mechanisms , whether SR load-driven , RyR refractoriness-driven , or otherwise , is highly context-dependent . Although exploring the issue of SR load vs . RyR refractoriness was beyond the goals of the current study , our results suggest that in human cAF , both SR load alternation and RyR refractoriness are involved in alternans genesis at slower pacing rates . In our cAFalt model , alternation in all SR Ca2+ release variables , including [Ca2+]SR , RyR open probability , and RyR inactivated probability , was necessary for alternans at the onset CL of 400 ms ( Fig . 6 ) . In addition , SR uptake flux ( Jserca ) enhanced alternans when clamped ( Fig . 6 ) and therefore suppressed alternans under normal pacing conditions , suggesting that SR load is indeed an important driver of CaT alternans in cAF and that upregulation of the SERCA pump may be an important therapeutic strategy for diminishing alternans . We also showed that CaT alternans occurred in the cAFalt model at slow pacing rates because decreased RyR inactivation resulted in steepening of the SR release-load relationship . Together , these results indicate that the interplay between SR load and RyR kinetics is responsible for alternans onset in human AF . The mechanisms for human atrial alternans susceptibility are likely to encompass a range of complex interactions at multiple scales of biology , which extend beyond the cellular-level mechanisms found here . In this study we examined the behavior of an atrial cell with well-developed t-tubules [19] . Research has shown that rat atrial cells have variable levels of t-tubule organization [54] . Such variation , if present in human atrial cells , would result in subcellular Ca2+ gradients which could make cells more susceptible to alternans [17] , [55] , [56] . Models of atrial myocytes incorporating detailed spatial descriptions [57] and local control of Ca2+ [58] will aid in future investigations of the subcellular mechanisms of cAF-related alternans . In addition , the complex structure of the atria , including its normal conduction pathways [59] and fibrotic remodeling in AF [60] , [61] , may promote heterogeneity and discordant alternans , which significantly affect alternans dynamics and reentry initiation [9] , [62] . Consideration of these factors in the future will further enrich the mechanistic insight gained from this current study and will advance our understanding of the role that alternans play in AF arrhythmogenesis . In many cell models , the effective refractory period ( ERP ) is not consistent with ERP at the tissue level [63] . Electrotonic effects in tissue and the whole heart can shorten or lengthen APD depending on which structures and cell types are coupled to each other . Furthermore , alternans in single cell models may not be predictive of alternans in tissue , where conduction alternans can occur . This was the case for the control atrial tissue model , in which loss of capture occurred at a CL of 260 ms before reaching the very fast pacing rates at which APD alternans were observed in human control patients ( CL = 218±30 ms ) [8] . However , alternans onset at clinically observed rates occurred in the single-cell control model ( 200 ms CL , S9 Figure , black curve ) and when kiCa was reduced by 5% ( 230 ms CL , S9 Figure , red curve ) . This suggests that the ionic model may not be well-constrained for tissue simulations at very fast rates . However , this issue did not affect the study of alternans onset at slower pacing rates , as was observed in AF patients . Our ionic model variable clamping protocol , which involved separately clamping the even or odd beat waveforms , was used to test for model variables which could robustly suppress alternans when clamped to either of two very different waveforms . An alternative approach would be to clamp model variables to the single unstable , non-alternating waveform obtained using a control algorithm [64] . This approach would allow more precise assessment of fixed point stability , since clamping is done at the point of instability rather than during the bistable ( alternans ) endpoint . However , for the purposes of quantifying the most important variables influencing instability , the clamping protocol used in this study was sufficient to identify the central role of SR Ca2+ release , which was later confirmed through iterated map analysis . Recent experimental evidence points towards local SR Ca2+ depletion , rather than Ca2+-dependent RyR inactivation , as the main mechanism of SR release termination [23]–[26] . Although alternans in the cAFalt model relied on Ca2+-dependent RyR inactivation , other termination mechanisms which rely on SR Ca2+ ( used in the Sato-Bers RyR model ) may have similar effects on SR release slope and alternans susceptibility ( Fig . 7 , column 2 ) . However , with the Sato-Bers RyR model , alternans and other complex oscillations began at the baseline pacing rate ( 750 ms CL , S10 Figure ) and did not display the same rate dependence observed in patients [8] . In addition , large oscillations in CaT amplitude did not couple as strongly to voltage as with the original RyR , and oscillations were also attenuated in tissue ( S10 Figure ) . Further work is needed to develop atrial cell models which incorporate current mechanistic understanding of SR Ca2+ release and which can also reproduce AF-related alternans rate dependence in tissue . AF is associated with progressive changes in alternans onset in the human atria , with alternans occurring at slower heart rates as AF severity worsens . We found that the differences in alternans onset between AF and control patients could be accounted for by changes in the inactivation rate of the RyR in a model of human atrial cAF-remodeled tissue . Single-cell simulations revealed that alternans at these slow heart rates were driven by abnormal Ca2+ handling and the development of CaT alternans , and that changes in CaT alternans threshold resulted from steepening of the SR Ca2+ release slope , decreased SR Ca2+ uptake efficiency , and decreased inactivation of the RyR . These findings provide important insight into the mechanisms underlying proarrhythmic APD alternans occurring at slow heart rates in cAF patients . Such insight may aid in the development of targeted therapies and new treatment strategies for AF in the future .
In order to investigate ionic mechanisms in human AF that contribute to the generation of atrial alternans at the tissue level , we created a computer model of human atrial tissue incorporating ionic remodeling associated with cAF . The atrial tissue preparation had dimensions of 0 . 33×0 . 33×9 . 9 mm3 ( Fig . 1A ) , similar to the one used by Krummen et al . [65] Human atrial cell membrane kinetics were represented by a modified version of the Grandi-Pandit-Voigt ( GPV ) human atrial action potential model [19] , which we refer to as the GPVm model . Detailed explanation and justification of the GPVm model modifications are provided in the supplement ( S1 , S2 Texts ) . Different types of human atrial tissue were modeled individually as homogenous tissue preparations , with each incorporating ionic changes appropriate for each tissue type . Both control and cAF-remodeled tissue , as well as left and right atrial tissue , were modeled using the parameter changes specified by Grandi et al . [19] ( see S1 Text ) . The isotropic bulk conductivity value for the tissue was tuned to produce a conduction velocity of 0 . 62 m/s in control tissue [59] , [66] . When cAF ionic remodeling was incorporated , the same bulk conductivity value produced a conduction velocity of 0 . 59 m/s . These values are within the reported ranges for control and AF conduction velocities [67] . We assessed alternans in the human AF tissue model by applying the clinical pacing protocol used by Narayan et al . to induce alternans in AF patients [8] . The tissue model was first initialized at all nodes with steady-state values from a single cell paced at 750-ms CL . The tissue was then paced from the stimulus electrode ( Fig . 1A ) for 20 beats at 750-ms CL and then for 74 beats at each subsequent CL , starting from 500 ms and shortened in 50-ms steps to 300 ms , and then shortened in 10-ms steps , until loss of capture or conduction block occurred . Voltage traces from the recording electrode ( Fig . 1A ) were analyzed for APD alternans . APD was calculated as the time from maximal upstroke velocity to 90% repolarization of Vm from phase II amplitude . Alternans magnitude was quantified as the mean magnitude of change in APD over the last 10 pairs of beats ( 11 beats total ) . APD alternans normalized magnitude ( ANM ) , obtained by dividing the alternans magnitude by the mean APD over the last 10 beats , was used to compare alternans between cells of varying APD . Alternans onset CL was defined as the longest CL for which ANM was greater than 5% [8] . To identify cellular changes which could account for the onset of alternans in AF patients at CLs of 300–500 ms [8] , we explored how ANM varied in human AF tissue models of both the left and right atrium as a result of changes in ionic model parameters . Of the 20 ionic model parameters tested , 10 were parameters altered in the GPVm model to represent cAF [19]; others were associated with L-type Ca2+ current ( ICaL ) , rapidly activating potassium current ( IKr ) , SR uptake , or SR release ( Table 1 ) . We scaled parameter values one at a time to 25–200% of the default left or right atrium values specified by Grandi et al . [19]; for each parameter value within this range , simulations were conducted to determine the presence of alternans ( 282 simulations total ) . In AF patients , average alternans onset CL was>300 ms [8] , so pacing and alternans analysis was restricted to CLs≥300 ms . After identifying conditions under which APD alternans magnitude and onset CL matched clinical observations , we utilized two different clamping approaches in order to investigate the key cellular properties that gave rise to these alternans , as described below . Further explanation of the rationale behind these methods can be found in Results . The monodomain and ionic model equations were solved using the Cardiac Arrhythmia Research Package ( CARP; Cardiosolv , LLC ) [69] . Details on the numerical techniques used by CARP have been described previously [70] , [71] . A time step of 20 µs was used for all simulations . | Atrial fibrillation is an irregular heart rhythm affecting millions of people worldwide . Effective treatment of this cardiac disorder relies upon our detailed knowledge and understanding of the mechanisms that lead to arrhythmia . Recent clinical observations have suggested that alternans , a phenomenon where the shape of the electrical signal in the heart alternates from beat to beat , may play an important role in this process , but the underlying mechanisms remain unknown . In this study , we use computational models to conduct a detailed examination of the causes and contributors to alternans associated with human atrial fibrillation . We find that in atria remodeled by atrial fibrillation , alternans appears near resting heart rates because several aspects of calcium cycling are disrupted in the atrial cells . In particular , the release and uptake of calcium from the cellular storage compartment , the sarcoplasmic reticulum , becomes imbalanced , leading to alternation in calcium signals from beat to beat . These findings provide important insights into the mechanisms of proarrhythmic alternans in human atrial fibrillation which may be used to develop novel therapeutic targets and treatment strategies in the future . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"biotechnology",
"bioengineering",
"biomedical",
"engineering",
"engineering",
"and",
"technology",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"biophysics",
"biophysical",
"simulations"
]
| 2014 | Disrupted Calcium Release as a Mechanism for Atrial Alternans Associated with Human Atrial Fibrillation |
A concept fundamental to viral pathogenesis is that infection induces specific changes within the host cell , within specific tissues , or within the entire animal . These changes are reflected in a cascade of altered transcription patterns evident during infection . However , elucidation of this cascade in vivo has been limited by a general inability to distinguish changes occurring in the minority of infected cells from those in surrounding uninfected cells . To circumvent this inherent limitation of traditional gene expression profiling methods , an innovative mRNP-tagging technique was implemented to isolate host mRNA specifically from infected cells in vitro as well as in vivo following Venezuelan equine encephalitis virus ( VEE ) infection . This technique facilitated a direct characterization of the host defense response specifically within the first cells infected with VEE , while simultaneous total RNA analysis assessed the collective response of both the infected and uninfected cells . The result was a unique , multifaceted profile of the early response to VEE infection in primary dendritic cells , as well as in the draining lymph node , the initially targeted tissue in the mouse model . A dynamic environment of complex interactions was revealed , and suggested a two-step innate response in which activation of a subset of host genes in infected cells subsequently leads to activation of the surrounding uninfected cells . Our findings suggest that the application of viral mRNP-tagging systems , as introduced here , will facilitate a much more detailed understanding of the highly coordinated host response to infectious agents .
At the interface of pathogen infection and host response lies a complex network of regulated interactions . As the host seeks to eradicate the pathogen and maintain survival , the pathogen itself seeks to continue its own proliferation at whatever cost is necessary to the host cell . Therefore , the insult associated with viral infection often involves numerous changes in host gene expression . Fundamental to many viral pathogenesis studies is the investigation of these specific changes within the host cell , or on a more global scale , within a specific tissue , organ , or the entire animal . Although it has been possible in several systems to singularly identify cellular genes that are altered in expression due to infection , these genes most likely represent a very small fraction of all the genes induced or repressed . In an attempt to more fully understand the interactions between pathogen and host , virologists have turned to high-throughput genomic profiling technologies within the past decade to evaluate the status of host gene expression post-infection . Although widely informative , there remains an inherent limitation in applying these analyses to viral pathogenesis studies . In the absence of an acutely susceptible system in which all cells can be uniformly infected , a heterogeneous environment of infected and uninfected cells naturally exists during viral infection . This is particularly true in vivo , where only a minority of cells in a given tissue or organ are infected , even when that tissue is a major target of infection . In traditional gene expression analysis utilizing total RNA isolation , there is an inability to discriminate the population of host mRNAs isolated from the infected cells versus the surrounding uninfected cells . As the percentage of uninfected cells is high in vivo , mRNA from uninfected cells likely creates background signal that skews or masks the analysis from infected cells . Discriminating the direct viral impact on infected cells from the subsequent effects on bystander uninfected cells is critical to fully understanding the pathogenesis of a given virus , yet most analyses lack this distinction . To circumvent this limitation , we have optimized and implemented an innovative mRNP-tagging technique to isolate host mRNA specifically from infected cells following viral infection in cultured cells as well as tissues in vivo . The mRNP-tagging technology was originally developed from a functional genomics approach termed ribonomics , which examines mRNAs functionally clustered in ribonucleoprotein complexes [1 , 2] . The mRNP-tagging system takes advantage of the natural interaction of RNA-binding proteins with cellular mRNA to effectively enrich and isolate messages from a specific minority cell type within a heterogeneous environment . One such interaction that has been used in several systems is the well established strong binding of poly ( A ) binding protein I ( PABP ) to the poly ( A ) tail of cellular mRNAs prior to translation [3–8] . In the mRNP-tagging technique , a unique version of PABP engineered with an epitope tag is expressed in a cell- or tissue-specific manner . The cellular mRNA bound to the tagged-PABP is then co-immunoprecipitated using an anti-epitope antibody , enriching the mRNA from the targeted cell population and separating it from the mRNA of the surrounding cells or tissues [1 , 2 , 9] . Gene profiling methods such as cDNA microarrays or quantitative real-time PCR can then be performed using the enriched mRNA population to assess the gene expression status within the cell or tissue population of interest . This method has been successfully applied to identify tissue-specific mRNA populations in Caenohabditis elegans and Drosophila melanogaster [10–13] , as well as to identify cell type-specific gene expression changes in mixed cell culture models in vitro [14] . Here , we have adapted the mRNP-tagging technique to characterize host gene expression changes following infection with Venezuelan equine encephalitis virus ( VEE ) . VEE is an arthropod-borne , single-stranded ( + ) sense RNA virus associated with periodic epidemics and equine epizootics in the Western Hemisphere , and serves as a leading model for the study of alphavirus pathogenesis [15] . Numerous studies have underscored the dramatic role of virus genetics and the subsequent host defense response in dictating the course and outcome of VEE infection [16–32] . Although infection in the murine model has been well studied for some time , little is known concerning the molecular markers of VEE-induced disease , including the direct effects on host cell gene expression . VEE infection is characterized by two distinct disease phases following infection in humans , horses , and mice: An initial lymphotropic phase characterized by a high serum viremia , followed by invasion of the central nervous system and initiation of a neurotropic phase leading to encephalitis . In horses and mice , progression to the neurotropic phase occurs at very high frequency . Previous studies in our laboratory have carefully examined the progression of pathogenesis in the mouse model , utilizing molecularly cloned infectious VEE as well as an extensive panel of mutants blocked at various stages of infection [18 , 21 , 22 , 25 , 27 , 29] . The draining lymph node ( DLN ) , and in particular the dendritic cells , was subsequently identified as the initial site of viral replication , with infected Langerhans cells migrating there from the site of inoculation in the footpad [20] . It has been hypothesized that the early events within the DLN set the stage for the VEE-specific pattern of virus replication and host response . However , many details of the earliest stages of VEE infection remain largely undefined , with the innate host response likely playing a major role . To define the molecular profile of the early virus–host interactions central to VEE pathogenesis , we took advantage of several tools . One tool paramount to studying the early events in infection are VEE replicon particles ( VRP ) . VRP are propagation-defective vector particles that undergo only one round of infection , as the structural genes which normally drive the assembly of progeny virions are deleted and replaced with a marker gene of interest [33] . As such , VRP replication is limited to the first infected cells , allowing us to model the earliest events of VEE infection . In addition , the application of an mRNP-tagging technology offers an opportunity for a distinct view of the VEE-induced changes in host gene expression . By expressing an epitope-tagged version of PABP from VRP , host messages induced specifically within the first round of infected cells can be fractionated from those of the surrounding uninfected cells . Through co-immunoprecipitation with antibody to the epitope tag , the infected cell host mRNA bound to the VRP-delivered tagged-PABP can be isolated and screened as a discrete mRNA population for changes in host gene expression . This technology enables discrimination of uninfected cells from infected cells , and specifically profiles the changes induced in the infected cell population—a distinction that previously has been difficult to achieve , particularly in vivo where the infected cells may be only a small minority in a given tissue ( e . g . , in the DLN post-VRP infection ) . Using VRP to infect primary dendritic cells in vitro , and to limit infection to the initially infected cells in vivo , we have elucidated gene expression patterns that define the early stages of VEE pathogenesis , including members of the interferon ( IFN ) and proinflammatory host defense pathways . This analysis revealed multifactorial interactions that occur within the virus-infected host , and indicated a two-phase innate response characterized by cytokine and antiviral gene induction profiles in the first infected cells that were distinct from that of uninfected bystander cells within the same tissue . By elucidating these specific and distinct host responses , systems such as the VRP mRNP-tagging approach have the potential to further our understanding of the complex interactions between pathogens and their hosts .
An mRNP-tagging technique has been optimized and applied to isolate mRNA specifically from the infected cell population following VRP infection . The fundamental basis of mRNP-tagging relies on the natural interaction of RNA-binding proteins with RNA , with the synthesis of a uniquely tagged RNA binding protein in target cells enabling the isolation of the specific mRNA population through tag-specific antibodies and co-immunoprecipitation . To apply this technique to a virus infection model , a FLAG epitope-tagged version of PABP was delivered specifically to infected cells by engineering the virus itself to express the unique RNA-binding protein . The PABP coding sequence , with the FLAG epitope fused in frame at the 5′ end , was cloned directly downstream of the 26S promoter of the VEE replicon plasmid ( Figure 1A ) , and the replicon RNA was packaged into VEE replicon particles to generate FLAG-PABP VRP . Upon infection of BHK and L929 cells , the FLAG-PABP VRP programmed the robust expression of the epitope-tagged version of PABP , as determined by anti-FLAG immunoprecipitation as well as by western blotting and from cell lysates ( Figure 1A ) . Robust levels of expression were expected based on the documented high level of transgene expression from the 26S mRNA promoter of VRP [33] , a key element for this VRP mRNP-tagging system . An outline of the VRP mRNP-tagging procedure is shown in Figure 2 . Briefly , following infection with FLAG-PABP VRP at low multiplicity of infection ( MOI ) , the FLAG-tagged PABP molecule is synthesized only in the infected cells as the replicon RNA is expressed . At various times post-infection the cells are lysed , releasing PABP-bound messages . Pre-cleared lysate is mixed with agarose beads coated with anti-FLAG antibody . The mRNA from the infected cells is immunoprecipitated specifically via the FLAG epitope of the PABP bound to the message in the RNP complex , as this form of PABP is only present in VRP-infected cells . The message is subsequently isolated from that complex by proteinase K digestion and phenol-chloroform based extraction . RNase protection assays ( RPAs ) were initially utilized to detect host messages following the anti-FLAG immunoprecipitation from infected L929 cell extracts , and verified that the VRP-supplied tagged-PABP was in fact bound to host messages , and the host mRNA could be isolated and used in gene expression assays . Bead saturation studies were completed to determine the concentration of immunoprecipitating antibodies that would ensure antibody excess in immunoprecipitating the PABP-bound mRNA population ( data not shown ) . An issue that could complicate the precise nature of the profiling involved in mRNP-tagging systems is the potential for promiscuous exchange or reassortment of endogenous and tagged-PABP among mRNAs in cell extracts . To alleviate concern in this matter , several groups have employed the use of formaldehyde crosslinking to increase the stability of the mRNA-PABP interaction during immunoprecipitation [10–13] . However , in studies where this treatment was assessed , it was found that crosslinking had little to no effect on the level of mRNA enrichment from the target populations [9 , 13] . Additionally , the degree of crosslinking that is effective , without irreversibly linking the mRNA to the protein and thus dampening RNA recovery , may be difficult to determine . Published studies demonstrating that mRNA originally bound to PABP in cell lysates can not be displaced by competing excess poly ( A ) RNA or free PABP also alleviated concerns of PABP reassortment [9] . Nonetheless , we designed an experiment to address the degree , if any , of PABP reassortment in our system . L929 cells were pretreated with actinomycin D ( AMD ) for 1 h prior to infection with FLAG-PABP VRP , with infection proceeding under continued AMD treatment . AMD inhibits DNA-dependent RNA transcription , thereby preventing the synthesis of new host RNA during infection . Importantly , the treatment of cells with AMD at or shortly before alphavirus infection does not inhibit viral RNA-dependent RNA synthesis [34] . Therefore , when AMD pretreated cells were subsequently infected with FLAG-PABP VRP , the FLAG-tagged PABP was expressed from the VRP , however no new host messages were available for binding to the tagged-PABP delivered by the VRP . The only remaining source of host RNA available for binding by the VRP-supplied tagged-PABP would be host mRNA that was present prior to AMD treatment and infection—the large majority of which would already have been bound by endogenous PABP . Therefore , unless promiscuous exchange or reassortment of endogenous PABP and tagged-PABP occurred among host mRNAs , there should be little to no host RNA signal detected upon mRNA-tagging analysis of AMD-pretreated versus PBS-pretreated cells . In fact , when the mRNA-tagging technique was used following FLAG-PABP VRP infection to isolate host mRNA from AMD-pretreated cells , either a complete absence or severely reduced levels of host mRNA in anti-FLAG immunoprecipitated lysates was observed in comparison to PBS-pretreated cells ( see Figure S1 ) . Therefore , the VRP-supplied tagged-PABP did not reassort with or out-compete endogenous PABP bound to mRNA during infection . These results , along with the data generated from other groups , indicated that PABP reassortment was not a major concern . The fundamental purpose of applying an mRNP-tagging approach to the examination of the host response during viral infection is to be able to discern changes in host gene expression that occur directly within the infected cells . This is a distinct advantage over traditional profiling techniques , particularly when infected cells are the minority in the overall cell population . The mRNP-tagging technique does isolate a particular subset of mRNA in the cell , namely those that bind PABP , therefore we wanted to verify that this method would yield an mRNA population representative of host transcription following infection . To do so , cells were infected at a high MOI such that the tagging technique and the more traditional total RNA isolation would assess profiles from a similar cell population . L929 cells , a murine fibroblast cell line , were infected with FLAG-PABP VRP at an MOI of 5 . At 6 , 12 , or 24 hours post-infection ( hpi ) , RNA was harvested using each of the following three methods: 1 ) For traditional isolation of total cellular RNA , a commercially available solution of guanidine salts , urea , phenol and detergent ( UltraSpec reagent ) was used . 2 ) To isolate all poly ( A ) RNA bound to PABP in the cell population , an anti-PABP immunoprecipitation assay was performed on a separate set of mock and VRP-infected cells . 3 ) Finally , to isolate poly ( A ) RNA bound to the FLAG-tagged PABP provided by the FLAG-PABP VRP infection , an anti-FLAG immunoprecipitation assay was performed . RNA isolated from all three techniques was used as input RNA in an RPA designed to analyze the expression profiles of host genes relevant to this study . The infected cell profile of two such genes , IRF-1 and IFNβ , relative to that in mock infected cells is shown in Figure 3 , with a comparison of the levels as assayed by the three different RNA isolations . The data shown are representative of two separate experiments , and the specific mRNA signal generated from each isolation technique was normalized to GAPDH signal . During a timecourse of 6 , 12 , and 24 hpi at a high MOI , the message profiles of these two relevant host genes were similar among the various methods that were used to isolate RNA . Therefore , the mRNP-tagging technique was not limiting in terms of the availability or abundance of RNA screened . While the RNA populations examined using the VRP mRNP-tagging system in a high MOI situation lead to informative analyses , the truly advantageous use of the technique is in low MOI situations , where the number of infected cells are a small minority ( i . e . , tissues from VRP-infected animals ) . Therefore it was important to assess the level of sensitivity that could be expected in these low MOI situations . An in vitro experiment was designed to model the in vivo–like condition of a low frequency infected cell population . At 6 hpi , cell lysates were prepared from L929 cells that were either mock infected or infected at an MOI of 5 with FLAG-PABP VRP . The cell lysates were mixed in decreasing ratios of infected cell lysate to uninfected cell lysate , and the mixed lysate was then immunoprecipitated with anti-FLAG antibody , isolating the FLAG-PABP bound mRNA . This mRNA served as template RNA in an RPA , using probes specific for several host mRNAs . The results ( Figure 4 ) for two host mRNAs , IRF-1 and GAPDH , demonstrate that the mRNP-tagging system provides a sensitive measure of VRP-induced host gene expression , as the signal from FLAG-tagged host mRNA immunoprecipitated from the infected cells was detected in samples that contained as little as 1% infected cell lysate . In this particular experiment , the 1% value is approximately equal to 2 , 000 infected cell equivalents . This mRNA signal can be directly visualized in Figure 4 , as well as by the raw pixels plotted for IRF-1 in Table S1 and Figure S2 . It is worthy to note that RPAs do not include an amplification step , and as such the input RNA is directly assayed . Therefore , a higher degree of sensitivity was expected when shifting to profiling methods that include an amplification step , such as real-time PCR . An alternate strategy to assess host gene expression changes specifically in the infected cell compartment might rely on the sorting of the infected cell population from the uninfected cells ( e . g . , by fluorescence-activated cell sorting [FACS] ) , followed by the independent analysis of the RNA isolated from each population . To further assess the level of sensitivity , VRP-induced changes in host gene expression as assessed by the mRNP-tagging method were compared to the host profile derived by a FACS-based method ( Figure 5 ) . L929 cells ( 2 × 106 ) were infected at a low MOI of 0 . 2 with VRP expressing either GFP ( Figure 1B ) or FLAG-PABP . At this MOI , it is estimated that ∼10% of L929 cells were infected ( see Figure S3A ) . At 12 hpi , GFP expression was used as the basis for FACS-facilitated sorting of the GFP-VRP infected and uninfected cells into the two respective populations . The recovered GFP-positive ( infected ) cells were lysed , and all PABP-bound host messages were subsequently isolated by anti-PABP immunoprecipitation and RNA isolation . In parallel , the mRNP-tagging assay was used to sort messages from FLAG-PABP VRP infected L929 cells by lysing the entire monolayer , and using anti-FLAG immunoprecipitation to isolate FLAG-tagged PABP-bound messages specifically from the infected cells in the monolayer . To compare host mRNA levels in each infected cell population , mock treated L929 monolayers also were lysed , and the PABP-bound mRNA isolated by immunoprecipitation . As shown by real-time PCR analysis ( Figure 5 ) , a substantial induction of IFNβ , IP-10 , and IRF-1 mRNA in the L929 infected cell population over mock treated cells was apparent following analysis of both the FACS-based or mRNP-tagging techniques . Importantly , in comparing the two methods , the degree of sensitivity in detecting mRNA from the minority population of infected cells using the tagging technique was at least equal to ( IFNβ , IRF-1 ) , if not enriched ( >5× enrichment in IP-10 ) in comparison to those generated by the FACS-based method . These results further validate the mRNP-tagging system as a powerful tool for the analysis of changes in host gene expression following viral infection . A major advantage in using VRP as opposed to VEE virus in the mRNP-tagging system is the opportunity they provide to study the earliest events in the course of VEE pathogenesis , as VRP infect and replicate only within the first round of infected cells . We have previously demonstrated that dendritic cells represent an important early target of infection in vivo , as VRP target DCs at the site of inoculation following footpad delivery in the mouse model [20] , and have likewise been shown to efficiently transduce human DCs in vitro [35] . In addition , several groups have examined DC-tropic properties for other alphaviruses , such as Sindbis virus [36–39] and Ross River virus [40] . Primary murine bone marrow dendritic cells were therefore chosen as an in vitro model system to study early VRP-induced host responses . To compare the host response profile following traditional total RNA isolation of bone marrow–derived dendritic cells ( BMDCs ) versus that generated using the mRNP-tagging system , primary DCs isolated from 129sv/ev mice were infected at an MOI of 0 . 5 with FLAG-PABP VRP . At this MOI , approximately 4% of the cells are infected ( see Figure S3B ) . At 6 hpi , RNA was isolated from mock treated and VRP-infected BMDC by either 1 ) preparing cell lysates for isolation of PABP and FLAG-PABP bound mRNAs by immunoprecipitation , or 2 ) adding UltraSpec reagent for isolation of total RNA . The mRNP-tagging method specifically isolated mRNA from the infected cells using the bound FLAG-tagged PABP marker , and this population of mRNA was compared to endogenously PABP-bound messages in the mock BMDC culture . In contrast , the traditional RNA isolation lacked this discrimination , and therefore total cellular RNA was isolated from the entire infected and mock treated BMDC cultures for comparison . As shown in Figure 6 ( black bars ) , the gene expression profiles ( IFNβ , IP-10 , IL-6 ) evaluated specifically from the infected cells of the BMDC culture using the mRNP-tagging technique were dramatically enhanced in comparison to profiles from the entire population of infected and uninfected DCs generated using total RNA . The fold induction of IFNβ , IP-10 , and IL-6 mRNA in the infected BMDC cultures were found to be approximately 20- to 200-fold higher than that measured by total RNA ( compare the black bars in Figure 6 ) . It is likely that the high proportion of uninfected cells in the low MOI environment masked the signal from the minority of infected cells when assayed from the total RNA samples . Therefore only once these populations could be assessed separately could the fundamental differences existing between the infected and uninfected cell responses be revealed . All three of the evaluated defense response genes were induced to high levels in the infected DCs at this early timepoint of 6 hpi , suggesting that the host innate response is rapidly initiated following VRP infection of BMDCs . Although this rapid response could be detected in the DC culture as a whole using the total RNA analysis , the mRNP-tagging method was required to reveal the full extent of this early defense response within the infected cell population . The contribution of the IFNαβ system in determining the outcome and severity of VEE disease has been described for over 30 years [41 , 42] . Therefore , in seeking to further characterize the initial stages of VEE pathogenesis , the interferon response was a primary target for study . Accordingly , in the previous experiment , the host response to infection in primary BMDC isolated from IFNαβ receptor knockout ( IFNαβR−/− ) animals also was analyzed at 6 hpi . We hypothesized that in a system lacking IFNαβ receptor signaling , the mRNP-tagging method of specifically profiling the infected cell response should detect a diminished induction in interferon-stimulated genes ( ISGs ) , as the positive autocrine feedback signaling within these cells should be crippled in the absence of the IFNαβ receptor . In this manner , the IFNαβR−/− BMDC also provided another model system for substantiating the VRP mRNP-tagging technique . As demonstrated in Figure 6 , the ISG response in the infected IFNαβR−/− BMDC ( stippled bars ) as measured by the mRNP-tagging assay was much diminished in comparison to the response in wildtype BMDC ( black bars ) . The induction of IFNβ and IP-10 mRNA in the receptor knockout BMDC was diminished by approximately 200- to 1 , 500–fold , respectively , in comparison to levels of induction in the wildtype BMDCs . The IL-6 response also was reduced specifically in the infected cells of the IFNαβR−/− BMDC culture as compared to the wildtype IFNαβR+/+ BMDC culture , with the induction of IL-6 message measured by the mRNP-tagging assay at only background levels in comparison to mock treated IFNαβR−/− BMDC culture . Therefore , the host response in the absence of IFNαβ receptor signaling did in fact demonstrate a diminished induction of ISGs , specifically within VRP-infected cells . In contrast to the infected cell response , when the same cultures were globally assayed for changes in gene expression using total RNA analysis , the loss of the IFNαβ receptor in the culture overall appeared to have little effect on ISG induction at this early time post-infection ( Figure 6 ) . As compared to mock treated cells , the induction of IP-10 and IL-6 mRNA was similar in wildtype and interferon receptor knockout BMDC when measured by total RNA at 6 hpi . However , following VRP infection , IFNβ total RNA message levels were found to be approximately 5-fold higher in the IFNαβR−/− BMDC culture in comparison to the wildtype BMDC culture . This enhancement may be due in part to anti-viral paracrine signaling induced in the uninfected cell majority in an IFNαβ receptor-independent manner . It is important to note that in analyzing this low MOI infection using only total RNA , the high background of uninfected cells in the culture ( ∼94% to 96% of the culture , Figure S3B ) masked the dramatic effect the receptor knockout had on host gene expression in the infected cell population . The decreased host response of IFNβ , IP-10 , and IL-6 message in the infected IFNαβR−/− cells would have gone undetected had the mRNP-tagging system not been utilized . Previous studies by our group and others have examined the succession of events characteristic of VEE pathogenesis in the mouse model . Following inoculation in the footpad , Langerhans-like cells infected at the site of inoculation in the footpad subsequently migrate to the popliteal DLN [20] . It has been hypothesized , based on these observations , that the early events within the DLN set the stage for the specific pattern of virus replication and host response characteristic of VEE pathogenesis . Seeking to focus on the early interactions of virus and host in the DLN , we extended our characterization of the host response to the mouse model , with the hypothesis that combining total RNA profiling with the tagging system would generate a more comprehensive view of the host response post-infection in vivo . Adult BALB/c mice were inoculated in each rear footpad with 106 IU of FLAG-PABP-VRP . At 6 h and 9 hpi , the popliteal DLNs were removed and pooled . Subsequently , either total RNA was isolated using the RNeasy Protect protocol ( Qiagen ) , or RNA specifically from the infected cells was isolated by anti-FLAG immunoprecipitation . cDNA was synthesized from each RNA sample , and Taqman real-time PCR was performed to analyze several target host genes . Two independent DLN samples were analyzed from each group , with GAPDH serving as the internal housekeeping control gene . The expression profiles of three host messages ( IFNβ , IP-10 , and IL-6 ) characterized from the DLN post-infection are shown in Figure 7 . Two distinct views of the response to infection in the DLN were revealed; the global host response in the DLN as a whole , and the specific response within the infected cell population . In examining the response to infection over time within the infected cells , an early robust expression of IFNβ and IP-10 was exhibited in the anti-FLAG isolated RNA at 6 hpi , which waned by 9 hpi . This suggests a very rapid response to VRP infection within infected cells of the DLN . While the response was waning between 6 and 9 hpi within the infected cells , the total RNA profile of the DLN demonstrated that the response in the organ as a whole was increasing during this time interval . By 9 hpi , the majority of the IFNβ and IP-10 host response appears to have shifted to the surrounding uninfected cells of the DLN . This is likely due to a robust activation of the innate immune response directly within the infected cells , which then induces mediators ( e . g . , cytokines ) that are released into the surrounding environment of the DLN to initiate paracrine responses in neighboring uninfected cells . This detailed view of events occurring at early times post-infection has been difficult , if not impossible , to examine previously . The IL-6 host response presents an interesting scenario , as there was a complete lack of signal detected at 6 hpi in the infected cells of the DLN as measured by the mRNP-tagging system ( Figure 7 , asterisk ) . However there was a robust IL-6 response measured at 6 hpi from the total RNA isolated from the entire DLN . This suggests that the infected cells of the DLN are not the main producers of IL-6 initially following VRP infection . Instead , the uninfected bystander cells may be particularly poised to respond to paracrine signals from infected neighboring cells , resulting in IL-6 expression . By 9 hpi , IL-6 expression was detected in both the infected cell population as well as in the DLN as a whole , indicating a continuation of the dynamic interplay between the infected and uninfected cells in this environment . This two-phase innate response would have otherwise gone undetected had the mRNP-tagging technique not been integrated with the traditional profiling , allowing for a uniquely multifaceted examination of the VRP-infected host .
Here we have introduced an innovative approach for assessing gene expression changes following viral infection in vitro and in vivo , addressing a critical parameter that has been difficult , if not impossible , to address previously . By distinguishing changes in the host transcriptional program of infected cells from that of uninfected bystander cells , the mRNP-tagging technology provides an important advancement in gene expression profiling , and promises to increase our understanding of the host response to virus infection , particularly in vivo . Characterization of the VRP mRNP-tagging approach has demonstrated several key aspects of the system . First and foremost is the ability of the system to effectively target and isolate mRNA from the infected cell population . In high MOI cell culture experiments , where all cells are infected , the gene expression profiles generated from total RNA and RNA isolated by mRNP-tagging were similar . Additionally , in low MOI cell culture experiments , where only a minority of cells ( ∼10% ) are infected , the RNA profiles generated from the infected cell population isolated by FACS were similar to the gene expression profiles isolated by the mRNP-tagging technique without prior cell sorting . These results demonstrate that mRNP-tagging yields profiles that are representative of infected cells , even when they are the minority cell population . An equally important aspect of the mRNP-tagging system is that the method is sensitive . In mixing experiments , where the infected cell signal was diluted with mock lysate , the mRNP-tagging approach detected a unique transcriptional profile when as few as 1% of the cultured cells were infected . Moreover , the sensitivity of directly isolating infected cell mRNA via the mRNP-tagging approach proved to be as great , or even enhanced in comparison to analysis of message levels when isolated from infected cells following a commonly utilized FACS-based approach . This ability to effectively analyze a small minority of infected cells , such that the mRNA signal from infected cells is no longer masked by the background of uninfected cells , is a property well suited for application in vivo . Additionally , the VRP mRNP-tagging technique allows this level of specificity without harsh treatment of the infected cell populations prior to RNA isolation . This is in contrast to FACS analysis , which commonly requires physical manipulation of cultured cells or tissues ( e . g . , trypsinization , collagenase digestion ) , and may result in cellular damage to delicate cell types analyzed post-infection , such as the shearing of fragile dendrites from the cell body of DCs . This physical manipulation may affect the host gene expression profile of cells , similar to the profound effects various isolation techniques can have on the maturation and function of DCs [43] . However , since the VRP mRNP-tagging system isolates mRNA from infected cells following lysis directly in the culture vessel or from intact tissue , the cells are not distressed prior to message isolation and may more accurately reflect the host transcriptional program at the time of isolation . Furthermore , avoiding manipulations that result in a long lag time between cells/tissue harvest and RNA isolation may be particularly advantageous when dealing with relatively unstable messages . The finding of 5-fold higher levels of IP-10 by the mRNP-tagging technique in Figure 5 may reflect such a situation . It is possible that the difference in IP-10 signal sensitivity between the GFP-VRP infected group and the FLAG-PABP VRP infected group may be a manifestation of the stability of the IP-10 message versus the other messages examined . In fact IP-10 message has been documented to be relatively unstable and to increase in stability when mRNA binding proteins are bound to it [44] . Lysing and analyzing the infected cell IP-10 message directly from the culture vessel by the mRNP-tagging method may have allowed us to increase the sensitivity in detecting an unstable message . It is also worthy to note that given the well-established roles for PABP in translation initiation and mRNA stability [5 , 7 , 8 , 45] , PABP-associated mRNA may actually be more representative of the cell's actively translated message population or the proteome [1 , 13] . Application of the mRNP-tagging method has revealed several important insights into the host innate response to virus infection . First , the host response is dramatically different in infected and uninfected cells within the same in vitro cell culture or the same tissue in vivo , varying in quantitative , qualitative and temporal terms . Quantitatively , the level of response in each population varied by gene , indicating that the transcriptional programs of infected and uninfected cells within the same culture or tissue are uniquely affected . On a gene by gene basis , comparing the infected cell profile generated by the mRNP-tagging system to the total RNA profile generated from the entire culture or tissue allows the relative contribution of the infected versus uninfected cell populations to be teased apart . In the case where a small percentage of primary BMDC were infected in vitro , the induction of IFNβ , IP-10 , and IL-6 in infected cells as measured by the mRNP-tagging system was up to 200 times that indicated in the total RNA sample for the entire culture . These results suggest that the infected cells may be the primary source of the early IFNβ , IP-10 , and IL-6 response to VRP infection in BMDC cultures . Additionally , had the mRNP-tagging system not been utilized , the induction of IL-6 would likely have been overlooked due to the low total RNA induction measured from the BMDC culture . Therefore , at particular times the level of response in the infected and uninfected cell populations can differ so extensively that their analysis becomes qualitatively different . In other words , a response that is robust in one population may be completely absent in the other , and in this regard , the host response will appear to be quite different when evaluating infected cells alone rather than the entire culture or tissue . Temporally , the combination of the mRNP-tagging approach and total RNA analysis offered a unique vantage point into the kinetics of the infected and uninfected cell responses . In the DLN , this analysis demonstrated two phases of the innate host response . The first apparent phase was a rapid response in the infected cells of the DLN , including the robust activation of key host defense genes , IFNβ and IP-10 , at 6 h following footpad inoculation . While this infected cell response was waning by 9 hpi , an apparent second phase of response was mounting in the surrounding uninfected cells of the DLN , with induction of the same defense genes increasing from 6 h to 9 hpi . In fact , a similar rapid onset of the host innate response has been described previously following virulent VEE infection , with cytokine RNA levels in the DLN of infected mice peaking at 6 to 12 h following footpad inoculation , and waning by 24 to 48 h [46] . However , the use of VRP and the application of the mRNP-tagging system in our study provided the ability to distinguish the events specifically occurring within the first round of infected cells in vivo from the effects on the uninfected bystander cells of the DLN , as well as from the complication of viral cell-to-cell spread , revealing the unique kinetic response to infection occurring in each population . Several signaling events are likely to contribute to the kinetics of this two-phase activation . Replicon particle binding , entry , and replication provide a multitude of signals to initiate the early host defense response directly within the infected cells . It is likely that the rapid innate activation of the infected cell population leads to the secretion of cytokines and other soluble immune modulators . A portion of these mediators likely initiate the cell signaling events in uninfected bystander cells that are responsible for the strong paracrine response in the DLN at later times post-infection . This would include the activation of cells which have homed to the DLN as an active site of viral infection . In fact , a large influx of cells to the DLN has been observed following footpad VRP inoculation , including cells of the proinflammatory response and antigen presenting cells ( J . M . Thompson , A . C . Whitmore , J . L . Konopka , T . P . Moran , and R . E . Johnston , unpublished data ) . These recruited cells would remain uninfected , but would be susceptible to the primed environment of the DLN , and likely contribute to the induction of the second phase of the host innate response in vivo . Our results highlight the role of the innate immune response during VEE infection , particularly the interferon response . Evidence for the major role of the interferon system in controlling VEE replication and spread in vivo has been well established [18 , 19 , 23 , 40–42 , 47–49] , with tremendous levels of soluble , biologically active interferon in the serum , measured at up to 80 , 000 IU/ml following virulent VEE infection [18] . The absence of interferon signaling in vivo results in a significantly shorter average survival time following VEE infection ( 30 h in mice lacking the IFNαβ receptor in comparison to 7 . 7 days in wildtype mice ) , and a 10 , 000-fold increase in virus titers [18] . Here , the role of the interferon response in autocrine and paracrine signaling of infected and uninfected cells was further elucidated in BMDC . While ablating the IFNαβ receptor had no apparent effect on the total RNA induction of IFNβ , IP-10 , and IL-6 genes in the BMDC culture as a whole , a dramatically reduced induction of each of these host response genes was observed in the infected BMDC population . This strongly suggests that autocrine signaling through the IFNαβ receptor on the infected cells plays a critical role in inducing a strong ISG response . Conversely , in the uninfected BMDC population , the absence of IFNαβ receptor signaling had little effect on the induction of the same host response genes , remaining at near wildtype levels . This was a surprising , nonetheless interesting result that suggests interferon-mediated signaling through the IFNαβ receptor is not the primary paracrine mediator leading to the induction of these particular host response genes in uninfected bystander cells . It is important to highlight the fact that this data does represent a temporal snapshot of the host response at the early time of 6 h post VRP infection . However , it is consistent with previous findings showing levels of biologically active type 1 serum interferon are similar in IFNαβ receptor knockout and wildtype mice at early times post VRP infection [18] . Together , these data suggest that at least early in infection , VRP may induce an interferon response mediator that is independent of type 1 receptor signaling , such that surrounding uninfected cells lacking the type 1 receptor are still capable of mounting an antiviral response . It is important to note that BMDC lacking the type 1 interferon receptor are more permissive to infection with VRP than wildtype BMDC ( Figure S3B ) , which may result in increased levels of transgene expression ( Figure S3C ) , and may in part contribute to the higher induction of IFNβ in the receptor knockout BMDC demonstrated in Figure 6 . Such factors will need to be addressed should a separate line of investigation be initiated to explore the identity of this novel mediator ( s ) . Taken together , the data presented here highlight an important additional observation . Namely , reductionist in vitro approaches do not always recapitulate what is occurring in vivo . Often , high multiplicity infections of largely homogenous cultured cells are utilized in vitro to draw conclusions about what occurs in naturally low multiplicity infections of complex heterogeneous tissues in vivo . However , the data presented here strongly argue that the naturally heterogeneous environment of infected and uninfected cells existing during infection in vivo must be appreciated in order to understand the dynamic interactions occurring between these populations . For example , here in cultured BMDC , IL-6 was highly induced in VRP infected cells at 6 hpi . However , this did not appear to be the case in vivo where induction of IL-6 mRNA was first documented in the uninfected bystander population . The application of the VRP mRNP-tagging system offers a multitude of future studies that promise unique perspective on the highly coordinated host response to viral infection . However , it is worthwhile to note that there is room for potential improvement to the overall VRP mRNP-tagging platform . As it stands currently , the mRNP-tagging system specifically grants a temporal snapshot of the infected cell response—a response that differs not only quantitatively but also qualitatively from that of the bystander uninfected cells and from the culture as a whole . While this is a key strength , in this initial application of the system we relied on total RNA isolated by a separate technique to evaluate the “uninfected cell” response . This was possible given an underlying condition of low MOI infections , in that the majority of cells in a given culture or tissue would be uninfected . In evolving the system from here , a major goal will be to achieve a way to more directly compare host RNA responses from uninfected and infected cells in the same culture or tissue . Nonetheless , even as it stands at this current stage , the mRNP-tagging system provides a powerful way to address the dynamic interaction of pathogen and host . A critical future application of the VRP mRNP-tagging system will be the analysis of the host response to infection within the brain , where the most extensive pathogenesis is observed following VEE virus infection . While alphavirus CNS pathogenesis has been studied extensively in several model systems , it has been suggested that multiple host- and virus-specific parameters contribute to VEE-induced pathology in the brain [50–52] . In elucidating such parameters , distinguishing the relative contribution of infected versus bystander cells in the CNS has proven critical following infection with neuroadapted Sindbis virus [53 , 54] , and is also likely to be crucial in understanding VEE-induced neurodegeneration . Developing the mRNP-tagging system using VEE virus with a double 26S subgenomic promoter would also facilitate the characterization of the host response to VEE infection downstream of the DLN , including the impacts of virus budding as well as cell-to-cell spread during infection [20] . The application of mRNP-tagging technology to the study of the host response to viral infection opens avenues of investigation that have previously been difficult to navigate . A better understanding of virus–host interactions may subsequently facilitate the design of improved therapeutics and vaccines . More specifically , gaining a clear profile of the host response to VEE infection promises to further our understanding of the specific virus–host interactions that define alphavirus pathogenesis in vivo .
The construction and packaging of VRP using a split helper system have previously been described [33] . The replicon plasmid constructs used in this study were ( i ) replicons expressing green fluorescent protein ( GFP-VRP ) and ( ii ) replicons expressing an N-terminally FLAG-tagged version of poly ( A ) binding protein I ( FLAG-PABP VRP ) . The production of GFP-VRP has been described previously [20 , 55] . The FLAG-PABP replicon plasmid was generated by the directional cloning of the ORF of PABP I containing an N-terminal FLAG epitope tag ( GACTACAAGGACCACGATGACAAG , kindly provided by J . D . Keene [14] ) , immediately downstream of the 26S mRNA promoter of the pVR21 replicon plasmid . All replicon particles used in this study were packaged in the wildtype ( V3000 ) VEE envelope [33] . Briefly , the replicon RNA genome containing the VEE nonstructural genes and expressing the heterologous gene from the viral 26S promoter , along with two defective helper RNAs providing the wildtype capsid and glycoprotein genes , but lacking the virus-specific packaging signal , were co-electroporated into BHK-21 cells ( ATCC ) . Due to the lack of encoded viral structural genes in the replicon genome , infectious VRP undergo only one round of infection , and the absence of propagating recombinant virus was confirmed by passage in BHK-21 cells . VRP were concentrated from supernatants by centrifugation through a 20% sucrose cushion and resuspended in PBS . BHK-21 titers were determined either by immunofluorescence ( GFP-VRP ) , or immunocytochemistry ( null VRP , FLAG-PABP-VRP ) using sera containing antibody to the VEE nonstructural proteins . ( i ) Infection of L929 cells . L929 murine fibroblasts ( ATCC ) were maintained at 37°C under 5% CO2 in complete alpha minimal essential medium ( αMEM , Gibco ) containing 10% donor calf serum , 10% tryptose phosphate broth , 2 mM L-glutamine , 100 U/ml penicillin and 0 . 5 mg/ml streptomycin . For VRP infection , 106 cells were seeded in 60 mm dishes and incubated overnight . The medium was removed from the monolayer and the cells were infected at a multiplicity of infection ( MOI ) of 5 ( unless otherwise indicated ) in 0 . 2 ml endotoxin-free PBS supplemented with 110 mM Ca2+ , 50 mM Mg2+ , and 1% vol . /vol . donor calf serum . After 1 h of adsorption at 37°C , complete αMEM was added to the monolayer . In studies involving the pretreatment of L929 cells with AMD to inhibit the transcription of new host RNA , cells were pretreated with 4 ug/ml AMD for 1 h prior to VRP infection . ( ii ) Generation of primary murine BMDCs . Breeding pairs of IFNαβR+/+ 129Sv/Ev and IFNαβR−/− mice were kindly provided by Herbert Virgin ( Washington University , St . Louis , MO ) and Barbara Sherry ( North Carolina State University , Raleigh , NC ) , respectively . Mice were bred under specific pathogen-free conditions in the Department of Laboratory Animal Medicine breeding colony facilities at the University of North Carolina , Chapel Hill . To generate primary immature BMDCs [40 , 56] , bone marrow cells from femurs and tibia of 8- to 14-week-old mice were aspirated with RPMI-10 medium ( RPMI 1640 [Gibco] , 10% FBS , 2 mM L-glutamine , 50 uM 2-ME , 100 U/ml penicillin , 100 ug/ml streptomycin sulfate ) . Cells were filtered through a 40 um cell strainer , pelleted , and resuspended in lysis buffer ( 0 . 15 M NH4Cl , 0 . 1 mM Na2EDTA , 1 mM KHCO3 [pH 7 . 2–7 . 4] ) . Following lysis of red blood cells at room temp , 10 ml of RPMI-10 media/mouse was added , cells were again pelleted and resuspended in fresh RPMI-10 media . Cells were seeded in 6-well low cluster plates ( Corning ) in RPMI-10 media supplemented with 20 ng/ml GM-CSF ( Peprotech ) , and incubated at 37°C under 5% CO2 . On day three , RPMI-10 media supplemented with 20 ng/ml GM-CSF and 20 ng/ml IL-4 ( Peprotech ) was added to each well . On day five , additional RPMI-10 media supplemented with 10 ng/ml GM-CSF and 10 ng/ml IL-4 was added to each well . On day seven , the BMDC were harvested and either immediately used for infection , or were cryopreserved at 2–5 × 106/ml in 90% FBS/10% DMSO . ( iii ) Infection of primary BMDCs . Cryopreserved BMDCs were quickly thawed in a 37°C water bath , and gently transferred to a conical tube containing an equal volume of RPMI-10 . The volume of RPMI-10 was brought up to 10 ml , and the cells were pelleted . An additional wash with RPMI-10 was completed , and the cells were resuspended in RPMI-10 , supplemented with 5 ng/ml GM-CSF and IL-4 . BMDCs were seeded at 2 . 5 × 106 cells/well in hydrated six-well low cluster plates , and allowed to recover overnight at 37°C , 5% CO2 . BMDCs were harvested and pooled with a cold PBS wash of each well . After pelleting , cells were resuspended at 106 cells/ml , and 106 cells/well were seeded in a hydrated six-well low cluster plate . BMDC were infected with VRP at an MOI of 0 . 5 in 100 ul PBS supplemented with 1% donor calf serum and Ca+/Mg+ . Following 2 h of absorption at 37°C , 5 ml of RPMI-10 supplemented with 5 ng/ml GM-CSF and IL-4 was added to each well . Seven- to eight-week-old female BALB/c mice were obtained commercially ( Charles River Laboratories ) and allowed to acclimate for 5–7 days . Mice were inoculated in each rear footpad with 106 IU of VRP diluted in 10 ul endotoxin-free PBS containing 1% donor calf serum . Mock-infected animals received diluent alone . ( i ) L929 cells . At indicated times post-infection , media was removed and L929 cell monolayers were washed with cold PBS . The UltraSpec RNA Isolation System was used to isolate total RNA , with 1ml of UltraSpec RNA Reagent added to each 60mm dish of L929 cells per manufacturer's protocol ( Biotecx ) . ( ii ) Primary BMDCs . At 6 or 12 hpi , BMDCs were transferred to a 15 ml conical tube and pelleted ( 1 , 200 rpm , 10 min at 4°C ) , during which time each well was washed with cold PBS . The wash was used to resuspend the pelleted cells , followed by a second spin . Total RNA was harvested from BMDC using the RNeasy Mini Kit , according to the manufacturer's protocol ( Qiagen ) . ( iii ) Lymph nodes . At indicated times post infection , mice were euthanized and both draining popliteal lymph nodes were harvested , washed with cold PBS , and pooled together into 200 ul RNAlater RNA Stabilization Reagent ( Qiagen , Ambion ) . Total RNA was harvested from tissue homogenate prepared using a plastic pestle with a handheld motor and the RNeasy Protect Mini Kit ( Qiagen ) . ( i ) Antibodies . The mRNP-tagging method as applied to cells and animal tissues infected with VRP was developed from general ribonomics/mRNP-tagging protocols previously described by the Keene laboratory [1 , 2 , 9 , 14] . Polyclonal anti-PABP antibody was generously provided by J . Keene ( Duke University Medical Center ) . Additionally , polyclonal anti-PABP H-300 antibody was obtained from Santa Cruz Biotechnology . Monoclonal anti-FLAG M2 antibody was acquired from Sigma-Aldrich . ( ii ) Preparation of mRNP lysate from cultured cells . L929 monolayers ( 106 cells total ) were washed with cold PBS , followed by lysis of the monolayer with 1ml of polysome lysis buffer ( 100 mM KCl , 5 mM MgCl2 , 10 mM Hepes [pH 7 . 0] , and 0 . 5% Nonidet P-40 with 1 mM DTT , 100 U/ml RNaseOUT [Invitrogen] , 0 . 2% vanadyl ribonucleoside complex [New England Biolabs] , and 1 tablet/10 ml Complete Mini Protease Inhibitor Cocktail Tablet [Roche] added fresh at time of use ) . For BMDC lysis , cells were gently pelleted and the media removed . The pellet was washed with PBS and spun again , followed by resuspension and lysis in 500 ul of polysome lysis buffer . Cells were lysed for 10 min , followed by centrifugation at 14 , 000g in a tabletop microfuge for 10 min at 4°C to remove cellular debris . The ∼1 ml total volume of L929 mRNP lysate ( isolated from 106 L929 cells ) was stored at −80°C in 200 ul working aliquots , while the 500 ul total volume of BMDC mRNP lysate ( isolated from 106 BMDC cells ) was stored at −80°C in 250 ul working aliquots . ( iii ) Preparation of mRNP lysate from whole animal tissue ( DLN ) . Freshly dissected DLN were washed with ice-cold PBS . Five DLN were pooled per sample , and coarsely homogenized in 200 ul polysome lysis buffer ( containing RNase and protease inhibitors as described above ) using a plastic pestle and hand-held motor . Samples were frozen at −80°C until use . Upon thawing ( on ice ) , the homogenization and lysis was continued to completion ( as monitored microscopically ) , and the lysate spun at 4°C to pellet any remaining tissue/debris . The supernatant was transferred to a fresh tube on ice , and a second round of lysis/ homogenization was completed on the pellet using 100 ul polysome lysis buffer . Upon centrifugation , this supernatant was pooled with the first ( ∼300 ul total ) and kept on ice until use . ( iv ) mRNP immunoprecipitation . For each immunoprecipitation sample , two 60 ul aliquots of protein beads were prepared; the first aliquot to coat with the specified antibody , and the second to pre-absorb the lysate for removal of non-specific binding . For antibody coating , 60ul Protein-A Sepharose beads ( pre-swollen with 12ml PBS/1 . 5g beads , Sigma ) or fast flow Protein-G Sepharose beads ( commercially pre-swollen , Sigma ) were washed with PBS ( 10 volumes ) , followed by a second wash of 10 volumes NT2 buffer ( 50 mM Tris [pH 7 . 4] , 150 mM NaCl , 1 mM MgCl2 , 0 . 05% Nonidet P-40 ) tumbled end-over-end for 15 min at room temperature . Beads were resuspended in 10 volumes fresh NT2 buffer supplemented with 5% BSA , and tumbled overnight at 4°C with excess immunoprecipitating antibody ( 200 ul anti-PABP antibody; 25 ul anti-FLAG antibody ) . Prior to the immunoprecipitation reaction , the antibody-coated beads were washed with 1 ml NT2 buffer . To pre-absorb the lysate , a second 60 ul aliquot of beads ( per sample ) was prepared as described above through the 15 min NT2 buffer wash . These beads were resuspended in 5–10 volumes NT2 buffer supplemented with 100 U/ml RNaseOUT , 0 . 2% vanadyl ribonucleoside complex , 2 mM DTT , and 25 mM EDTA . The mRNP lysate was added ( ∼200 ul L929 lysate , 250 ul BMDC lysate , or 300 ul DLN lysate ) , along with 1 ul normal animal serum corresponding to the immunoprecipitating antibody ( normal mouse serum for anti-FLAG immunoprecipitation samples; normal rabbit serum for anti-PABP immunoprecipitation samples ) . Samples were tumbled at room temperature for 1 h to remove any material that would non-specifically bind to the beads . This pre-absorbed lysate was recovered by gently pelleting the beads and applying the supernatant to the specific antibody-coated beads , along with 5–10 volumes of NT2 buffer supplemented as described above . This immunoprecipitation slurry was tumbled end-over-end at room temperature for 2 h . ( v ) RNA extraction from immunoprecipitated mRNP complex . Immunoprecipitated samples were gently centrifuged to pellet the beads , and washed four times with ice-cold NT2 buffer ( 10 bead volumes ) . Washed beads were resuspended in 600 ul proteinase K digestion buffer ( 100 mM Tris [pH 7 . 5] , 12 . 5 mM EDTA , 50 mM NaCl , 1% SDS ) , plus 25 ul of 20 mg/ml proteinase K , and incubated for 30 min in a rotating device at 50°C . 600 ul phenol-chloroform-isoamyl alcohol ( Fisher ) was added to the beads , which were vortexed for 2 min and then centrifuged for 5 min at 14 , 000g , 4°C . This was followed by extraction with one volume of RNA-grade chloroform ( Fisher ) , and standard isopropanol precipitation including 8 ul of glycogen . The samples were stored at −80°C until use for gene expression analysis . Upon thawing on ice , the samples were spun for 30 min ( 14 , 000g , 4°C ) , and the RNA pellet washed with 100 ul of 80% ethanol . The RNA pellet was resuspended in RNase-free water ( Ambion ) , or hybridization buffer as necessary . RNase protection assays were used to determine the relative abundance of specific cellular mRNAs in infected and mock infected L929 cells . 32P-labeled RNA probes were synthesized by use of the RiboQuant in vitro transcription kit and a RiboQuant multiprobe custom template set ( BD Pharmingen ) . This custom set included template for the synthesis of radiolabeled probes specific to murine interferon beta ( IFNβ ) and interferon regulatory gene-1 ( IRF-1 ) , as well as mRNAs encoding the murine housekeeping proteins GAPDH and L32 . RNA isolated from the mRNP complexes was used as input RNA for the mRNP-tagging samples , using 100% of the isolated RNA by resuspending the RNA pellet directly in hybridization solution . 2 ug of total RNA to be used as input samples for total RNA analysis was isolated as described above . The custom probe set was mixed with each RNA sample , placed in a pre-warmed heat block at 90°C which was immediately turned down to 56°C , and incubated overnight . The RNA-probe mixtures were treated with RNase according to the RiboQuant RPA kit protocol ( BD Pharmingen ) . The protected dsRNA species were electrophoresed on a 4 . 5% polyacrylamide-8M urea sequencing-sized gel , the gels were dried , and analysis conducted on a Molecular Dynamics Storm phosphorimager with ImageQuant software . Values represent the fold change over mock expression , as normalized to anti-PABP immunoprecipitated GAPDH housekeeping mRNA levels for the mRNP-tagging samples , or total GAPDH housekeeping RNA levels for total RNA samples . NP-40 protein lysate preparations were separated by 8% SDS-PAGE , and transferred to polyvinylidene difluoride membrane at 10 V for 60 min in transfer buffer ( 48 mM Tris , 39 mM glycine , 10% methanol ) . The membranes were blocked overnight in 5% dry milk in TBST ( 50 mM Tris-HCl [pH 7 . 4] , 150 mM NaCl , and 0 . 1% Tween 20 ) , followed by incubation for 1h with either anti-PABP ( Santa Cruz Biotechnology , sc-28834 ) or anti-FLAG ( Sigma-Aldrich , F3165 ) primary antibody diluted in 1% dry milk-TBST . The appropriate horseradish peroxidase ( HRP ) -conjugated secondary antibodies ( Amersham Pharmacia ) were then added and incubated for 1 h , followed by chemiluminescence detection using ECL detection reagents ( Amersham Pharmacia ) . Blots were exposed to film and developed . For quantification of VRP infectivity and 26S transgene expression levels , GFP-VRP or mock infected cells ( BHK , L929 , or BMDC ) were harvested at 12 hpi , and washed once with cold PBS . The cells were fixed with PBS/1% paraformaldehyde before FACS analysis . FACS data were acquired using a Dako CyAn Flow Cytometer and analyzed using Summit software ( Dako ) . To compare the infected cell host message profile generated by the mRNP-tagging approach versus FACS-facilitated sorting , 1 . 5 × 106 L929 cells were mock treated or infected at a low MOI of 0 . 2 with either FLAG-PABP VRP or GFP-VRP . At 12 hpi , cultures were treated in one of two ways to generate a profile of host message levels specifically from the infected cells . The cell monolayers that had been infected with FLAG-PABP VRP were lysed , and FLAG-tagged host messages were immunoprecipitated from the infected cells per the mRNP-tagging protocol described above using 100% of the monolayer lysate . A separate anti-PABP immunoprecipitation was completed from both mock- and FLAG-PABP VRP-infected cell lysates to isolate all PABP-associated mRNA and served as the mock reference with normalization to GAPDH signal levels . To generate a comparative profile of the infected cell host response , cell monolayers that had been infected with GFP-PABP VRP were prepared for FACS-based analysis . The monolayers were trypsinized , washed with media ( αMEM supplemented with 2% FBS , 100 U/ml penicillin , and 0 . 5 mg/ml streptomycin ) and resuspended in 700 ul of fresh media . The UNC-CH Flow Cytometry Core Facility provided cell sorting capability using the Dako Modular Flow ( MoFlo ) Cell Sorter , gating on GFP-positive signal to sort and recover the infected cell population . This GFP-positive cell population was gently pelleted and resuspended in 0 . 5 ml polysome lysis buffer . 100% of the resulting lysate was immunoprecipitated with anti-PABP antibody to isolate all PABP-associated messages specifically from the ( GFP+ sorted ) infected cell population . A separate anti-PABP immunoprecipitation was completed from mock cell lysate and served as the mock reference with normalization to GAPDH signal levels . The host message populations isolated from each technique were analyzed using Taqman real-time PCR ( see below ) . ( i ) cDNA synthesis . A one-tube DNase treatment and reverse transcription protocol was used to generate cDNA , using SuperScript III Reverse Transcriptase First Strand cDNA kit ( Invitrogen ) . 0 . 5–0 . 75 ug of either mRNA-tag isolated or total RNA served as input RNA for the reaction , using RNase-free water to bring the total volume to 10 ul . This was combined with 1 ul 10 mM dNTP mix ( Amersham Biosciences ) , 4 ul 5X SuperScript III reverse transcriptase buffer , 1 ul 0 . 1 mM dTT , 1 ul 40 U/ul RNaseOUT ( Invitrogen ) , and 1 ul RQ1 RNase-free DNase ( Promega ) . The samples were DNase treated at 37°C for 30 min , followed by the addition of 1 ul RQ1 Stop Solution ( Promega ) and heat inactivation of the samples at 65°C for 10 min . Following the addition of random hexamer primers ( 150 ng , Invitrogen ) , reverse transcription of the samples was continued in the same tube , according to the SuperScript III protocol ( Invitrogen ) . cDNA samples were stored at −20°C . ( ii ) Real-time PCR . Real-time PCR was performed to determine the relative abundance of specific cellular mRNAs in infected and mock treated samples . Taqman Gene Expression Assays ( Applied Biosystems ) containing primers and probes for various target host messages were used , with each reaction performed in a 25 ul total volume ( 5 ul cDNA , 12 . 5 ul TaqMan Universal PCR Master Mix without AmpErase UNG [Applied Biosystems] , 1 . 25 ul probe/primer mix [Applied Biosystems] , and 6 . 25 ul RNase-free water ) . The default amplification profile was performed by the ABI Prism 7000 Real-Time PCR System , and the data converted into cycle threshold ( CT ) values by the 7000 Sequence Detection Software ( v1 . 2 . 3 , Applied Biosystems ) . Duplicate samples were amplified from each experimental group with GAPDH serving as the housekeeping control along with each target gene of interest . A negative template control also was performed , with all samples run in parallel on the same plate . ( iii ) Total RNA real-time PCR analysis . Real-time PCR results are presented as fold gene expression in the infected sample over that in the mock sample , as normalized to the GAPDH housekeeping gene . During each reaction , a cycle threshold ( CT ) value was generated for the target gene of interest ( and GAPDH ) , corresponding to the cycle number at which the fluorescence of the PCR product reached significant levels above the background threshold level . Raw CT values generated from total RNA samples were analyzed using the well established delta CT ( ΔCT ) method to generate the fold expression results ( User Bulletin , ABI Prism 7000 Sequence Detection System [Applied Biosystems] ) . Briefly , for each cDNA sample , the GAPDH CT value was subtracted from the CT value of the target gene ( e . g . , cytokine ) of interest , yielding a ΔCT value . The ΔCT value generated for the VRP-infected sample was then subtracted from the ΔCT value of the mock sample , yielding a ΔΔCT value . This widely-used method assumes the target and housekeeping genes were amplified with the same efficiency , thus the increase in host mRNA levels in the infected samples compared to the mock treated samples was calculated as 2− ( ΔΔCT ) . ( iv ) Real-time PCR analysis of mRNA-tagging samples . Prior to this standard ΔCT analysis , raw CT values generated from mRNA-tagging samples were normalized in a manner that was inherently required for this system . The mock signal values were generated from an anti-PABP immunoprecipitation which isolates all PABP-bound messages in the entire cell culture or tissue analyzed . However , the infected sample values were derived using an anti-FLAG immunoprecipitation to specifically isolate the infected-cell minority subset of the population . Therefore , an mRNP-tagging normalization step was utilized to account for two parameters: 1 ) The disparity in the cell population size of the mock and infected samples assayed by the mRNP-tagging system , and 2 ) any difference in the immunoprecipitating antibody strength ( the polyclonal anti-PABP antibody versus the monoclonal anti-FLAG antibody ) . To do so , raw GAPDH CT values were generated from mRNP-tagging samples in the following manner . From FLAG-PABP VRP infected samples , raw GAPDH CT values were generated from both anti-FLAG and anti-PABP immunoprecipitation-derived cDNA , representing GAPDH expression in the infected cell subset or the entire culture , respectively . However from mock samples , raw GAPDH CT values were solely generated from anti-PABP immunoprecipitation-derived cDNA , thus representing the expression of GAPDH in the entire cell population . Therefore , to normalize the mock control values for comparison to the infected samples , a ratio of the anti-FLAG GAPDH signal to the anti-PABP GAPDH signal was applied to account for the difference in cell population size and antibody strength ( r ) : This ratio was then applied to the mock anti-PABP raw GAPDH CT value to generate a normalized mock GAPDH value: This normalized CT value served as the input mock GAPDH value for the standard ΔCT analysis , as described above for total RNA .
The NCBI GenBank ( http://www . ncbi . nlm . nih . gov/Genbank/index . html ) accession and GeneID numbers for genes mentioned in the text are GAPDH ( NM_008084 . 2 , GeneID 14433 ) ; IFNβ ( NM_010510 . 1 , GeneID 15977 ) ; IL-6 ( NM_031168 . 1 , GeneID 16193 ) ; IP-10 ( NM_021274 . 1 , GeneID 15945 ) ; IRF-1 ( NM_008390 . 1 , GeneID 16362 ) . | A major element of viral pathogenesis is the induction of specific changes within the infected host , often reflected in altered gene expression patterns . However , revealing these changes in vivo has been limited by an inability to distinguish changes within the minority of infected cells from that in surrounding uninfected cells . Here we introduce a viral mRNP-tagging system , based on Venezuelan equine encephalitis virus ( VEE ) , that enables the isolation of host mRNA specifically from infected cells in vitro and in vivo , even when they are a small minority . This system allowed us for the first time to monitor the innate response specifically within the cells initially infected in vivo . In combination with simultaneous analysis of the entire tissue response , the result was a multifaceted view of the innate response to VEE in dendritic cells , and in the draining lymph node . The results supported a two-step response in which activation of host genes within infected cells leads to activation of bystander cells , offering insight into the process by which the greater innate immune response to alphaviruses is established in vivo . This system may be employed for a wide variety of pathogens , offering broader implications to the manner in which interactions between pathogens and their hosts are studied . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
]
| [
"viruses",
"infectious",
"diseases",
"virology",
"immunology",
"microbiology",
"mus",
"(mouse)"
]
| 2007 | A Two-Phase Innate Host Response to Alphavirus Infection Identified by mRNP-Tagging In Vivo |
Prediction of response to specific cancer treatments is complicated by significant heterogeneity between tumors in terms of mutational profiles , gene expression , and clinical measures . Here we focus on the response of Estrogen Receptor ( ER ) + post-menopausal breast cancer tumors to aromatase inhibitors ( AI ) . We use a network smoothing algorithm to learn novel features that integrate several types of high throughput data and new cell line experiments . These features greatly improve the ability to predict response to AI when compared to prior methods . For a subset of the patients , for which we obtained more detailed clinical information , we can further predict response to a specific AI drug .
A number of recent large efforts have focused on collecting genomic data from tumors . While these datasets led to several successful studies and insights , in many cases the clinical data available for patients enrolled in these studies is incomplete . This makes it hard to use such datasets for predicting tumor specific outcomes and tailoring treatments to individuals . To develop accurate methods for for predicting treatment responses we need both , a comprehensive genomic dataset profiling the individuals being studied and accurate complimentary clinical information . To date , methods that used the former ( detailed genomic data ) usually were unable to use the latter for a significant number of individuals while methods that only relied on clinical information are limited in their ability to distinguish between tumor responses [1] . Consider , for example , the genomic data that is part of The Cancer Genome Atlas ( TCGA , [2] ) . Several methods have used this data to study general questions related to cancer biology and prognosis . Examples include methods to identify molecular targets for cancer therapy [3] , enhancement / creation of general prognostic classification systems [4–6] , de novo pathway identification via identification of mutually exclusive mutations [7] and identification of genes implicated in cancer via combinations of different data types [8] . In contrast , most efforts for predicting response to specific treatments have been limited to much smaller datasets , usually focused only on specific pathways or classes of mutations , and often only relying on in vitro ( cell line ) experiments which have limited clinical utility [9–11] . Indeed , in many cases a key challenge researchers face when trying to predict such specific response is the lack of detailed and well-curated clinical data to supplement the high throughput molecular data in the large databases . Here we focus on response to aromatase inhibitors ( AIs ) , which block the conversion of androgen to estrogen and thus lower systemic estrogen . AIs show superior efficacy for the treatment of postmenopausal ER+ breast cancer compared to tamoxifen [12] . Despite the significant reduction of recurrence , resistance is common , and remains a tremendous clinical and societal problem . Mechanisms of resistance are very heterogenous [13] , and it is currently not possible to accurately predict response for specific AI treatments . Thus , methods for predicting tumor specific AI responses are urgently needed , especially given availability of choices of endocrine therapy , their potential side effects , and recent findings that extended endocrine treatment benefits a subset of patients [14] . To predict AI response we developed computational methods to construct network smoothed features based on breast cancer genomic data from the Cancer Genome Atlas ( TCGA ) and combined these with manually curated clinical data for a subset of patients in TCGA that were treated at the University of Pittsburgh Medical Center ( UPMC ) . Many previous approaches have been developed to integrate multiple types of omic data using a variety of techniques: multiple kernel learning [15–18] , joint matrix factorization [19 , 20] , latent variable models [21 , 22] , and other network-based data integration methods [23 , 24] , though most of these methods have drawbacks in treatment-specific prediction tasks . Such methods are typically either unsupervised , and therefore intended for general-purpose clustering and stratification of patients , or sacrifice genomic / clinical interpretability . The UPMC clinical data included information on the treatment patients received , its effectiveness and the outcomes . The genomic data we used included sequence variations , expression changes and cell line drug responses all smoothed using general protein-protein interaction networks . We used the clinical and genomic features to predict treatment response and overall survival . Overall we show that by combining genomic and clinical attributes we can obtain high accuracy and predicting cancer survival and slightly improve this accuracy when incorporating functional cell line data . For the more challenging task of predicting treatment outcome we show that both , the addition of the cell line data and the improved clinical data , leads to greater accuracy and improves upon prior methods .
We first tested our method on the set of 590 breast cancer TCGA samples that were either prescribed aromatase inhibitors , or were not considered for this type of treatment given their ER status , for which we assigned artificial “non-response” labels ( see Materials and methods , Classification ) . Figure A in S1 Text shows the univariate predictive performance of individual features we used based on ROC AUC metric , showing the top 20 and bottom 20 features sorted by AUC . As can be seen , while none of the features provide very high accuracy on their own ( the best single feature is the mean across all genes of min{protein targets of arimidex , smoothed differential expression} with an AUC of 0 . 81 ) , several features are still informative in isolation . Overall , the best single features are those using the PCA decomposition of the expression data and those that combine expression and drug target information ( protein targets of aromasin and gene targets of estrogen receptors ) . We also see that the drug targets features from LINCS ( Methods ) related to Arimidex are only weakly informative which may indicate that the specific cell line used for this drug ( HA1E , kidney ) is not enough for extracting general drug response profile for Arimidex . We next trained classifiers using all features to predict general response to aromatase inhibitors . Fig 2a shows cross-validation performance in prediction of aromatase inhibitor response , using probabilistic SVM and Random Forest ( RF ) classifiers , the top two performing methods among those we tested ( See Supplement for the performance of the other classifiers ) . We see that both classification methods lead to high mean ROC AUC ( 0 . 91 ) , demonstrating the advantage of integrating several different types of features . While both methods performed equally well , it is much easier to interpret the RF results and so we focus on these results below . Fig 2b shows the importance of each feature used by RF ( using the scikit-learn package [25] ) . Again , features that combine tumor expression changes with drug target information seem to be the most useful including features based on estrogen receptor targets and targets of aromasin . We also find a high scoring feature that combines tumor mutation information with estrogen receptor target information . Specific genes contributing to these high scoring PCA features are plotted in Figures B , C , and D in S1 Text . These genes include TP53 whose mutations were identified as most significant contributors for the top feature , in line with a recent study by Gellert et al . [26] , in which TP53 mutations were associated with poor response in tumors treated with AIs . Other top genes included CDH1 , which is involved in cancer progression and metastasis [27] , JUN , a transcription factor implicated in cell proliferation and angiogenesis in invasive breast cancer [28] , and KLK4 , which codes for a kallikrein protein that is overexpressed in prostate cancer [29] and is associated with an epithelial-mesenchymal transition-like effect in prostate cancer cells [30] . To obtain additional data for improving the ability of our method to predict tumor response to aromtase inhibitors we performed cell line experiments . In these experiments we grew a selection of ER+ and control ER- breast cancer cell lines in serum estrogen for 5 days and then either kept the estrogen-containing serum for an additional 5 days , or switched to serum free media , thus mimicking the removal of estrogen . Growth measure results for these cells are presented in Fig 3 . As can be seen , for several cells there are significant differences with and without serum estrogen . Since , unlike for the patients , we only have genomic data and no clinical information for these cells , we developed a joint prediction method by combining tumor and cell line derived classifiers ( Materials and methods , Combining cell line and patient derived classifiers ) . The joint prediction combined the predictions of the two separate classifiers ( tumor and cell line based ) by learning a weight for each of them . As expected , given the small number of the cell lines tested compared to the number of patients ( 13 vs . 590 ) , the weight assigned to the cell line predictor was lower ( median γ = 0 . 0276 , mean γ = 0 . 0286 across all leave-one-out cross-validation folds for random forest classifiers ) . As can be seen in Figure Q in S1 Text , with curves in the legend sorted by AUC , the addition of cell line information slightly improves cross-validation performance ( though this difference is not visible when limiting AUC values to two decimal places ) . The cross-validation results presented above correspond to an overall prediction of whether a tumor responds to an aromatase inhibitor . In the “real world” , patients receive one out of three AIs , and within our cohort Arimidex ( anastrozole ) was the most frequently prescribed drug . Given some differences in mechanism of action , side effects and efficacies [31–33] , we next used our method to predict response to Arimidex . Results are shown in Fig 4a . While this is a much more challenging prediction task than overall response to AIs ( reflected in the decreased overall accuracy ) the results still show the predictive power of the features that we compute . These results can be further improved with better clinical data . The discussion so far focused on all TCGA breast cancer samples . For a subset ( n = 151 ) of these patients we also have high-quality , manually-curated patient data , allowing us greater accuracy in identifying and predicting clinical outcomes ( a detailed list of the clinical variables we extracted for this cohort is available on the supporting website ) . We thus examined the difference in performance between training and testing on all TCGA patients and using only University of Pittsburgh/UPMC patients ( n = 62 given anastrozole , and n = 89 which were ER– or given any aromatase inhibitor ) . Results from this analysis are shown in Fig 4b and Figure I in S1 Text . While we do not observe a large performance difference when predicting response to all aromatase inhibitors ( indicating that TCGA clinical features for such analysis are likely good enough ) , we see a larger improvement when predicting of response to anastrozole alone ( ROC AUC 0 . 73 for UPMC samples compared to 0 . 70 for all TCGA samples ) . This indicates that accurate information about the specific drugs used for each patient , switching between drugs and responses and side effects , all present in the UPMC curated data but not in the TCGA data , can greatly help automated methods for feature construction in personalized medicine analysis . To evaluate the usefulness of the features we constructed for this prediction task we first compared the results of using these features to methods that only use the measured expression and sequence data [34–39] . For this we constructed a “naïve” feature set , consisting only of somatic mutations , differential expression , and binary indicator columns for the clinical features ( Methods ) . We repeat our cross-validation analysis using this feature set , using the “raw” binary features , and using the top 8 , 32 , 128 , and 512 components from PCA decomposition/transformation of this matrix . Results are shown in Fig 5 , Figure E in S1 Text and Figure G in S1 Text for all aromatase inhibitors , and Figures F , H , and R in S1 Text for anastrozole . We see that performance of these ‘naïve” feature is comparable for the “all aromatase inhibitor” case ( leave-one-out ROC AUC 0 . 90 for binary features , max 0 . 90 for PCA decomposition , vs . 0 . 91 for our constructed feature set ) , while it is significantly lower for the more challenging task of predicting response to anastrozole ( leave-one-out ROC AUC 0 . 59 for binary features , 0 . 62 for PCA decomposition , vs . 0 . 70 for our constructed feature set ) . We also repeated the University of Pittsbugh/UPMC-only analysis with this ‘naïve” feature set , and again note a large drop in performance ( with ROC AUC dropping from 0 . 70 to 0 . 44 for anastrozole ) . See Figures J and K in S1 Text for full results . We also compared our method to prior methods that used either a network based approach to analyze mutation data [40] , relied on mutually exclusive mutations [7] for prognosis classification , or combined disparate network similarity measures across networks [23] . Results are presented in Fig 5 and Figure R in S1 Text . In general we find that such methods , which only use mutation information , do not perform as well as our methods that integrate several different types of data including expression and drug targets . We have also compared our method to prior methods that are specifically focused on predicting AI response . Turnbull et al . [41] performed feature selection on gene expression data resulting in four genes which were used in a decision tree framework to predict tumor response to AIs . We reimplement their decision tree classifier ( Supporting Methods ) and used it for response predictions . Results are shown in Fig 5 for all aromatase inhibitor prediction , and Figure R in S1 Text for specific prediction of anastrozole non-response . The low accuracy of this method is likely due to the specific parameters used in the original study that are likely not appropriate for a the larger dataset studied in this paper . See Supporting Results for more details . Reijm et al . [42] developed an eight-gene classification system for prediction of AI response , and presented t-statistic values for association of these genes with tumor response . Though there are conceptual differences between t-statistic values and logistic regression coefficients , we nonetheless can use these t-values to produce continuous predictions of tumor response with log-fold gene expression data ( Supplementary Methods ) . Results for this method are presented in Fig 5 and Figure R in S1 Text . We see reasonable performance for prediction of aromatase inhibitor response ( ROC AUC 0 . 76 ) though substantially worse performance for anastrozole-specific response ( ROC AUC 0 . 53 ) .
We combined clinical and high throughput patient data with additional cell line experiments to predict tumor response to aromatase inhibitors . We developed methods for constructing PCA features by smoothing interaction networks overlaid with expression , mutation and drug target data . Our clinical data consisted of abundant ( though less accurate ) data for all TCGA patients and from a more detailed curated dataset for a subset of 151 patients in this set . To further improve the classifiers and the labels , we analyzed electronic medical records within the University of Pittsburgh Medical Center ( UPMC ) system for the subset of TCGA patients that were treated at UPMC . These elements constituted data involving known breast cancer risk factors that either were not included in the TCGA data sets , or it was uncertain how the data was obtained and/or validated . This included reproductive history of patients at the time of breast cancer diagnosis , family history including both first and second degree relatives as well as other malignancy history for the patients if applicable . Data involving comorbid diseases that are common in adult populations including hypertension , diabetes , hyperlipidemia and metabolic information on patient’s weight at the time of diagnoses were also obtained , as these may impact patients’ breast cancer specific survival as well as overall survival . In regards to tumor biology , information from the EMR was obtained to the specific degree of hormone receptor status including H-score or percent staining as well as HER2/neu status . As we show , the models we developed provide accurate general predictions for the success of treatment with aromatase inhibitors . Focusing on treatment with a specific drug , Arimidex , we show that using the more detailed clinical data can lead to much better results when using our methods , greatly improving upon the use of naïve features and on prior methods suggested for this task . For our labels , the analysis of EMRs allowed us to obtain the most up to date survival data . While most TCGA based analysis relies on survival data that was collected several years ago , EMRs are continuously updated and so we were able to use much more up to date information . Finally , we were able to use the EMRs to determine reasons for stopping specific therapy or drug including toxicity . Combined , the new features and improved labels , led to better performance for the challenging task of predicting response to a specific drug as we showed in Results . The top-scoring PCA component in the random forest prediction is strongly influenced by the cell cycle gene CCND1 , overexpression of which correlates with early cancer onset and tumor progression [43 , 44] . It has been well known that proliferation is a strong predictive factor of endocrine treatment response , for example elegantly shown in a series of neoadjuvant short-term pre-surgical studies [45–47] . Another gene that ranked highly in our feature importance score was CDH1 . CDH1 encodes E-cadherin , a calcium-dependent cell-cell adhesion protein , that is frequently mutated in a number of tumor types , including breast cancer . E-cadherin protein is lost in up to 95% of invasive lobular breast cancer , and is one of the hallmark features of this disease , whereas it is lost in less than 5% of invasive ductal breast cancers [48] . We have previously shown that estrogen treatment of breast cancer cells results in downregulation of E-cadherin , potentially contributing to estrogen-mediated activation of migration and motility of cells [49] . In addition , we have shown that ILC cell lines with genetic loss of CDH1 have a unique estrogen response compared to IDC cell lines [50] . Thus the results from this study further suggest a critical role for E-cadherin in response to estrogen and aromatase inhibitors . Much prior work has focused on the prediction of response to endocrine therapy in general and aromatase inhibitors specifically [10 , 51–53] . However , only a few of these studies are directly comparable to this work . Many prior studies use data types unavailable for large clinical datasets ( e . g . proteomic data ) , focus on other organisms such as mice , or are restricted to cell lines only . We therefore focused on comparison to relevant work and on the analysis of the usefulness of the features constructed . As we show , for the more challenging task of predicting specific drug response our method outperforms especially when comparing the results for the more accurate the University of Pittsburgh/UPMC cohort . Our results indicate that , while high throughput datasets are key to constructing accurate prediction methods , it is extremely important to couple these datasets with complete and accurate clinical data . While information on drug prescription and usage is available for all individuals in the TCGA breast cancer dataset , we found several discrepancies between the more detailed UPMC data and the TCGA data for the same individual . This may indicate that data on other patients is noisy as well . We believe that our study provides a strong incentive for additional efforts aimed at curation of such clinical data .
The input to our method consists of genomic and clinical BRCA ( breast cancer invasive carcinoma ) data obtained from TCGA [54] and detailed clinical data for a subset of 151 University of Pittsburgh/UPMC patients ( data description on supplementary website ) . The UPMC data provides specific treatments , reasons for changes in treatments , dates and responses for these patients . Specifically , we used the following data types: We focus on the three aromatase inhibitors prescribed most in BRCA patients: anastrozole ( Arimidex ) , exemestane ( Aromasin ) , and letrozole ( Femara ) . To construct labels for tumors ( response / non response ) for each of these drugs , we examine the treatment information to identify which patients were prescribed that drug , and whether the patient discontinued that drug due to non-response . We then construct a “non-response” vector for each drug , denoting a patient as positive if the patient discontinued that drug or died during treatment with it . We constructed a binary mutation matrix M , a log-fold gene expression matrix E , and a binary differential gene expression matrix D , with samples as rows and genes as columns . We use C ( A ) to denote the set of column labels of matrix A , so that e . g . C ( M ) is the set of genes that appear in the TCGA somatic mutation data . Similarly , we define R ( A ) as the set of row labels of matrix A , corresponding to the distinct samples ( individuals ) present in each data set . The mutation matrices M are defined as M [ i , j ] = { 1 if gene j is mutated in sample i , 0 otherwise ( 1 ) The COSMIC database [55] provides differential gene expression data for TCGA samples , represented as log-fold change between tumor and matched normal samples in the same tumor/tissue . The COSMIC database additionally annotates each log-fold differential expression measurement with “over” , “under” , or “normal” gene expression , for genes with log-fold differential expression outside σ = 2 standard deviations from the mean in each sample . We collect the continuous log-fold gene expression measurements into a matrix E , and collect the normal/over/under expression status into a binary matrix D: D [ i , j ] = { 1 if gene j is over- or under-expressed in sample i , 0 otherwise ( 2 ) The TCGA BRCA data includes somatic mutations in 22 , 232 genes across 1 , 081 samples , and differential expression for 17 , 747 genes in 1 , 079 samples . In addition to the condition specific omics data we also use general interaction datasets . We use the HIPPIE protein-protein interaction network [57 , 58] ( version 2 . 1 , released 2017-07-18 ) , which contains confidence scores for 318 , 757 interactions between 17 , 204 proteins . Additionally , we use gene expression data from the LINCS LDS-1191 assay , which contains measurements of gene expression in cell lines after gene knockouts and introduction of small molecules ( “perturbagens” ) . We use gene expression data in cell lines given the chemotherapy agent Taxol , and the aromatase inhibitor Arimidex . As has been shown in the past , protein interaction networks provide a useful way to overcome data sparsity and noise when predicting cancer responses [4] . Here we use the network propagation/smoothing method described in Vanunu et al . [59] to combine omics data across patients . Given a network G = ( V , E , w ) with V as the set of proteins , E as the set of their interactions , w ( u , v ) representing the reliability of an interaction ( u , v ) ∈E , and a prior knowledge vector Y: V → [0 , 1] , we compute a function F ( v ) ∀v ∈ V that is both smooth over the network and accounts for the prior knowledge about each node . This network smoothing process uses a normalized edge weight matrix W′ , computed via Laplacian normalization of the edge weight matrix W: we first construct a diagonal matrix Δ with Δ[i , i] = ∑j W[i , j] , and compute W′ = Δ−1/2 WΔ−1/2 . Given a prior knowledge vector Y , we then compute the smoothed vector F using the iterative procedure described by Zhou et al . [60] . Starting with F ( 0 ) = Y , we update F at iteration t as follows: F ( t ) = α W ′ F ( t - 1 ) + ( 1 - α ) Y ( 3 ) This procedure is repeated iteratively until convergence; namely we stop when ‖F ( t ) −F ( t−1 ) ‖2 < ϵ . Note that Laplacian normalization produces a W′ with |λ|max ≤ 1 , which is required for this iterative method to converge . When Y is a binary vector , i . e . Y[u]∈{0 , 1}∀u ∈ V , the value F[v] for a gene v in the smoothed vector F naturally corresponds to a continuous measure of network proximity between v and the “selected” genes s ∈ S ⊆ V for which Y[s] = 1 . We therefore this network smoothing method to compute scores of proximity for each gene with respect to multiple gene sets: For each aforementioned gene set S , we construct a binary prior knowledge vector YS: Y S [ s ] = { 1 if s ∈ S ∩ V , 0 otherwise ( 4 ) We then perform network propagation on the vector YS , producing a vector FS . Note that not all genes in the set S are necessarily included in the protein interaction network , and therefore the vectors Y for e . g . somatic mutations in a tumor can differ from rows of the somatic mutation matrix M . For somatic mutations and differential expression , we then collect the smoothed vectors into “propagated” matrices MP and DP , with R ( MP ) = R ( M ) = R ( DP ) = R ( D ) and C ( MP ) = C ( DP ) = V . Intuitively , the propagated matrices MP and DP contain the per-sample binary vectors of M and D smoothed over the network . In biological terms , each row of these matrices represents the network proximity of each gene product to mutated and differentially expressed genes in that sample . Consequently , as illustrated in Fig 1 , the columns of these matrices provide propagated mutation and differential expression profiles for each gene product across all samples , indicating the proximity of the respective gene product to the products of mutated or differentially expressed genes in the respective sample . We next combine the smoothed matrices MP and DP with the smoothed vectors of multiple gene sets S as mentioned above: Protein targets of each drug are obtained from queries to DGIdb [61] , and gene targets of estrogen receptors are obtained from the TRRUST database [62] . Given one of the smoothed “target” vectors described above , denoted as T , we compute a new matrix MP , T: M P , T [ i , j ] = min { M P [ i , j ] , T [ j ] } ( 5 ) That is , for some tumor i and gene j , the value MP , T[i , j] quantifies gene j’s proximity to both somatic mutations in tumor i and the gene set S represented by the smoothed vector T . We compute DP , T similarly , replacing MP with DP in Eq 5 . We use these matrices to compute features for response to treatment in tumor i: In addition to these summary statistics for each tumor , we also perform PCA decomposition of these “minimum” matrices MP , T and DP , T , and use the top 10 PCA components as predictive features . Plots of PCA component scores for genes are shown in Figures B , C , and D in S1 Text—in these figures , genes are sorted by absolute value of there scores as assigned by PCA decomposition , and these absolute values are plotted as the importance of each gene for that PCA component . We obtained data from the LINCS project [56] L1000 LDS-1191 assay , which has profiled the gene expression of many cell lines under normal conditions , after introduction of small molecules ( “perturbagens” ) , and under gene knockouts . We selected the experiments involving the drugs analyzed in this study and identified the DE genes for each of these treatments . Two relevant drugs have been administered to cell lines by the LINCS consortium: Taxol ( a taxane , also known as Paclitaxel and Abraxane ) , and the aromatase inhibitor Arimidex . Each of these two drugs were tested on a single cell line , and we create LINCS features for each tumor by combining that tumor’s continuous log-fold differential expression with the expression change induced by that drug in the appropriate cell line . We compute two features for each ( tumor , drug ) pair: While the two features above are conceptually similar , we note that in addition to the direction of agreement , the dot product also represents the magnitude of change in expression between a tumor and the cell line in question . We use the following categorical variables from the general TCGA clinical data: We expand each categorical variable listed above into 0/1 indicator columns for use in classification methods . We additionally extract the estrogen receptor status of each tumor , used for selecting additional patients based on prior clinical knowledge . With the above features , we perform cross-validation experiments to assess our ability to predict response to aromatase inhibitor treatment . We examine all patients who were given any of the aforementioned drugs: 279 patients were given anastrozole , 51 were given exemestane , and 80 were given letrozole . An additional 180 samples were not considered for aromatase inhibitor therapy due to having ER– tumors , which are known not to respond to AI therapy . In this general aromatase inhibitor response prediction task , we assign a patient a “non-response” label if they were removed from any such drug for clinical reasons , or if the patient died during drug treatment . We also include prior clinical knowledge in this “all aromatase inhibitor” analysis; we integrate this prior knowledge by also computing the above features for the 180 patients who were not given an aromatase inhibitor , but who had estrogen receptor negative ( ER– ) tumors . These tumors are known not to respond to this type of treatment , so we assign these samples “non-response” labels . We use the features discussed above to learn various types of classifiers including logistic regression ( with both L1 and L2 regularization ) , Random Forest and Probabilistic SVMs . For each of these methods and each setting we perform leave-one-out cross-validation . We performed cell line experiments to compare breast cancer cell line growth with and without the addition of estrogen . We initially grow cell cultures for 5 days , with estrogen present , simulating the initial growth of breast cancer in a patient . We then separately grow cultures with a continued supply of serum estrogen , or with replacement cell medium that lacks estrogen—the environment without estrogen simulates the introduction of an aromatase inhibitor . We measure cell line growth with and without serum estrogen after the initial growth period , and from these cell counts we compute measures of how much each cell line responded to the presence of estrogen . We performed this experiment with 12 replicates of each cell line , 6 with and 6 without estrogen after the initial growth period , and used a mixture of ER+ and ER– cell lines ( details in Table B in S1 Text ) . We computed a growth measure for these cells as 1nM E2 GR - 1 no E2 GR - 1 ( 1nM E2 GR - no E2 GR ) ( 6 ) with “GR” denoting growth ratio with or without serum E2 . MCF-7 , BT474 , BT483 , CAMA1 , Uacc812 , ZR75-1 ZR75-30 and T47D breast cancer cell lines were purchased from American Type Culture Collection [ATCC] , Manassas , VA , USA . SUM44PE was purchased from Asterand Bioscience , Detroit , MI , USA , and 600MPE cells were a gift by Dr . Rachel Schiff . For the estrogen removal experiments , the cells were kept for 5 days in IMEM supplemented with 10% charcoal stripped serum ( CSS ) with 1nM E2 , and then plated into 96-well plates with or without 1 nM estradiol . An exception are Sum44PE cells that were kept in IMEM with 2% CSS . After 5 days , cell numbers were measured using Cell-titer Glo ( Promega , Madison , WI , USA ) according to the manufacturer’s instructions . Luminescence was measured with GloMax®multi-Detection System ( Promega , Madison , WI , USA ) , using a VICTOR X4 plate reader ( PerkinElmer , Waltham , MA , USA ) . Bars represent the mean of six biological replicates ± SD . 17β-Estradiol ( E2 ) was obtained from Sigma-Aldrich ( St . Louis , MO , USA ) . We separated a random 10% of patients to use for training weights between tumor and cell line classifiers , and used the remaining 90% of patients for cross-validation analysis . In each cross-validation fold , we fit classifiers to the corresponding training set of patients , and then used those classifiers to produce non-response predictions of the 10% of patients initially set aside . We then computed predictions for those 10% of patients using classifiers trained on cell lines , and chose the optimal convex combination of tumor and cell line predictions in the training set , producing final prediction p = γpc+ ( 1 − γ ) pp , with pc denoting predictions from cell lines and pp denoting predictions from tumors . The validation set predictions then combines the tumor and cell line predictions via the hyper-parameter γ tuned by cross-validation ( note that in this way we can use the full set of features for tumors while still using the cell lines in the prediction algorithm ) . | Breast cancer is the second most common type of cancer in women , with an incidence rate of over 250 , 000 cases per year , and breast cancer cases show significant heterogeneity in clinical and omic measures . Estrogen receptor positive ( ER+ ) tumors typically grow in response to estrogen , and in post menopausal women , estrogen is only produced in peripheral tissues via the aromatase enzyme . Inhibition of aromatase is often an effective treatment for ER+ tumors , but aromatase inhibitor therapy is not effective for all tumors , and causes of this heterogeneity in response are largely not known . In this work , we present a feature construction and classification method to predict response to aromatase inhibitor therapy . We use network smoothing techniques to combine tumor omic data into predictive features , which we use as input to standard machine learning algorithms . We train predictive models using clinical data , including high-quality clinical data from UPMC patients , and show that our method outperforms previous approaches in predicting response to aromatase inhibitor therapy . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"breast",
"tumors",
"cancer",
"treatment",
"protein",
"interaction",
"networks",
"enzymology",
"cancers",
"and",
"neoplasms",
"hormones",
"oncology",
"mutation",
"network",
"analysis",
"aromatase",
"inhibitors",
"pharmacology",
"estrogens",
"enzyme",
"inhibitors",
"computer",
"and",
"information",
"sciences",
"gene",
"expression",
"breast",
"cancer",
"proteomics",
"somatic",
"mutation",
"biochemistry",
"genetics",
"biology",
"and",
"life",
"sciences",
"drug",
"information"
]
| 2019 | Network-guided prediction of aromatase inhibitor response in breast cancer |
Sepsis is a consequence of systemic bacterial infections leading to hyper activation of immune cells by bacterial products resulting in enhanced release of mediators of inflammation . Endotoxin ( LPS ) is a major component of the outer membrane of Gram negative bacteria and a critical factor in pathogenesis of sepsis . Development of antagonists that inhibit the storm of inflammatory molecules by blocking Toll like receptors ( TLR ) has been the main stay of research efforts . We report here that a filarial glycoprotein binds to murine macrophages and human monocytes through TLR4 and activates them through alternate pathway and in the process inhibits LPS mediated classical activation which leads to inflammation associated with endotoxemia . The active component of the nematode glycoprotein mediating alternate activation of macrophages was found to be a carbohydrate residue , Chitohexaose . Murine macrophages and human monocytes up regulated Arginase-1 and released high levels of IL-10 when incubated with chitohexaose . Macrophages of C3H/HeJ mice ( non-responsive to LPS ) failed to get activated by chitohexaose suggesting that a functional TLR4 is critical for alternate activation of macrophages also . Chitohexaose inhibited LPS induced production of inflammatory molecules TNF-α , IL-1β and IL-6 by macropahges in vitro and in vivo in mice . Intraperitoneal injection of chitohexaose completely protected mice against endotoxemia when challenged with a lethal dose of LPS . Furthermore , Chitohexaose was found to reverse LPS induced endotoxemia in mice even 6/24/48 hrs after its onset . Monocytes of subjects with active filarial infection displayed characteristic alternate activation markers and were refractory to LPS mediated inflammatory activation suggesting an interesting possibility of subjects with filarial infections being less prone to develop of endotoxemia . These observations that innate activation of alternate pathway of macrophages by chtx through TLR4 has offered novel opportunities to cell biologists to study two mutually exclusive activation pathways of macrophages being mediated through a single receptor .
Sepsis and septic shock , one of the most common causes of admission in intensive care units results in death of nearly 3 , 50 , 000 people every year only in US and Europe [1] , [2] . The disease is a consequence of systemic bacterial infections that stimulates mediators of inflammation due to hyper activation of phagocytes . Immune cells express pattern recognition receptors ( PRRs ) which recognize immunostimulatory microbial products called PAMPs ( pathogen associated molecular pattern ) and trigger production of inflammatory mediators which assist the host in elimination of infectious agents [3] , [4]; however hyper induction of such mediators by dysregulated innate immune cells leads to sepsis and septic shock [5] , [6] . In sepsis caused by Gram-negative bacteria , endotoxin ( LPS ) activates the immune system through TLR4 and induces activation of macrophages that produce inflammatory mediators [7] , [8] . TLR4 is the signaling receptor for LPS but doesn't directly bind to LPS [9]–[11] . LPS forms a complex with LPS binding protein and CD14 which in turn delivers LPS to MD2 and LPS-MD2 complex activates through TLR4 resulting in dimerization of TLR4 [3] and initiate the signaling process for production of cytokines and other critical molecules needed for hyper-inflammation associated with endotoxemia/sepsis [12] . While effective use of antibiotics has resulted in improved prognosis of sepsis , deterioration of clinical symptoms and mortality has been attributed to persistent inflammatory cascade . Neutralization of inflammation is considered essential for preventing severe consequences of sepsis [13] , [14] . Thus developments of antagonists that block either activation through TLRs or downstream signaling pathways that inhibit the storm of inflammatory molecules are widely pursued by several investigators . Antibodies to TLR4 [15] , [16] , TLR4/MD2 [17] complex or LPS analogs [18] , [19] have been tested in animal models for their efficacy to protect against enditoxemia/Gram-negative sepsis although only LPS analogues have been undergone clinical trials . Nitrate salts [20] , 5c , an inhibitor of sphingosine kinase [21] , oxidized phospholipid [22] , [23] molecules have also offered promising results . Despite all these attempts very few candidate molecules have reached the level of clinical trials . A very recent report of one of the clinical trial for a promising agent was found to be ineffective ( http://clinicaltrials . pharmaceutical-business-review . com/news/eisai-eritoran-fails-to-meet-primary-endpoint-in-phase-iii-trial-250111 ) . In this study we report a novel mechanism that blocks endotoxemia by an approach fundamentally different from those documented so far . We demonstrate that a low molecular weight chito-oligosaccharide , chitohexaose ( chtx ) delicately balances the storm of inflammation induced by LPS while concurrently activating macrophages into a non inflammatory alternate pathway through TLR4 . Administration of chtx protected mice from endotoxemia prophylactically as well as therapeutically . The study also offered evidence for induction of two diverse activation pathways of macrophages through a single receptor , TLR4 . The stimulating ligand appears to determine the activation phenotype viz; classical pathway by LPS and alternate pathway by chtx . We stumbled on these findings while searching for the elusive innate receptors for nematodes .
Our initial studies were designed to identify an innate receptor on murine or human phagocytes that recognize nematodes . Biotinylated somatic extracts of adult stage parasites of Setaria digitata ( FAg ) and Brugia pahangi bound to surface of human monocytes as well as to murine bone marrow macrophages significantly more than the lymphocytes ( Figure 1 A–D ) . Specificity of biotinylated FAg reacting to monocytes was confirmed by competitive inhibition with unlabeled FAg ( Figure 1 B ) . C . elegans , a non-pathogenic nematode also contained a component binding to human monocytes and murine macrophages ( Figure 1C , D ) . A glycoprotein ( AgW ) affinity purified using WGA-Sepharose column ( Figure S1A ) was found to be the active component ( Figure 1E , F ) . Since another filarial glycoprotein ES-62 has been previously reported to interact with macrophage surface through TLR4 [24] , we tested such a possibility in our system . AgW as well as FAg competitively inhibited reactivity of antibodies to TLR4 on surface of murine bone marrow macrophages ( Figure 2A , Figure S 1 B , C , Figure S 2 A , B , C ) and on human monocytes ( Figure 2B , Figure S 1D , E , Figure S 2 D , E , F ) . We sought direct proof by performing a novel solid phase immunoassay developed by us . Soluble TLR4 present in membrane lysates of human PBMCs ( Figure 2C ) and murine bone marrow cells ( Figure S 1F ) reacted with FAg/AgW bound to solid phase . Human and murine TLR4 reacted with extracts of other nematodes also viz; Nippostrongylus brasiliensis , Heligomosomoides polygyrus and Caenorhabditis elegans ( Figure 2C , Figure S 1F ) suggesting the presence of conserved TLR4 binding components in nematodes . Enhanced binding of labelled FAg to jurkat cells over expressing TLR4 further confirmed its ability to interact with TLR4 ( Figure 2D , E ) . Since immunomodulatory properties of helminth products have been shown to depend on glycan moiety [25] we examined the possible role of carbohydrates in FAg-TLR4 interaction described above . Deglycosylated ( Figure S 3A ) or chitinase treated ( Figure S 3B ) FAg failed to competitively inhibit interaction of labeled FAg with monocyte surface suggesting the involvement of carbohydrate residues in such interactions . Susceptibility to chitinase also indicated the role played by chitin or it's oligomers in FAg interacting with TLR4 . This was tested by competitive inhibition using chitin oligomers . Binding of AgW to human monocytes was inhibited by chitosugars of varying size , longer the chain length higher was the degree of inhibition ( Figure 3A ) . Further , the hexasaccharide , chitohexaose ( chtx ) also inhibited reactivity of soluble TLR4 to AgW on solid phase in a dose dependant manner ( Figure 3B ) . These results suggested involvement of chitosugar residues present in FAg/AgW in interaction with TLR4 . In silico analysis using crystal structure of TLR4 and 3D structure of chtx ( http://pubchem . ncbi . nlm . nih . gov/summary/summary . cgi ? cid=197182&loc=ec_rcs ) also indicated significant affinity between the two molecules ( Figure 3C , Figure S 3C ) . Based on these observations further biological characterization were carried out using chtx as described below . The above findings that chtx residues present in FAg or AgW binds to TLR4 opened up the possibility of using the small molecular weight chito-oligosaccharide as a potential TLR4 antagonist to block LPS mediated inflammatory responses . BMDM of normal mice and normal human PBMCs were stimulated with LPS or chtx for 48 hrs and levels of TNF-α , IL-1β , IL-6 and nitrites were quantified in culture supernatants . LPS stimulated significant production of inflammatory mediators while chtx failed to do so in both murine and human systems ( Figure 4 A–E ) . Induction of genes for production of inflammatory molecules by LPS and failure to up regulate such genes by chtx was confirmed by Q-PCR also ( Figure 4F ) . Chtx on the other hand significantly inhibited LPS mediated inflammatory activation of mononuclear cells in both the systems ( Figure 4 A–E ) . We then tested induction of reactive oxygen species ( ROS ) by LPS and chtx . Murine BMDMs upon stimulation with LPS significantly up regulated ROS while chtx failed to do so . Rather chtx significantly inhibited LPS induced up-regulation of ROS . ( Figure 4G ) This observation is significant in the context of a recent report suggesting that induction of inflammatory cytokines is dependent on up regulation of ROS [26] . For in vivo validation of the above observations , BALB/c mice were administered a lethal dose of LPS with and without chtx . Chtx significantly inhibited LPS induced increase in plasma levels of TNF-α , IL-1β , IL-6 and nitrites ( Figure 5 A , B ) while increasing IL-10 levels ( Figure 5B ) . More critically , chtx significantly blocked LPS induced mortality of C57BL/6 mice ( Figure 5C ) suggesting that it can be used as a potential antagonist to block adverse biological consequences of endotoxemia in vivo . Administration of chtx 6 , 24 and 48 hrs after onset of endotoxemia was also effective in blocking mortality of mice suggesting its potential as a therapeutic agent ( Figure 5 D ) . BALB/c mice which are more susceptible to LPS induced endotoxemia than C57BL/6 mice were also protected by chtx ( Figure 5E ) . Alternate activation of macrophages has been well documented in helminth infections [27] . Presence of abundant chitin in helminthes [28] and activation of chitinase family proteins like AMCase , Ym-1 and chitinase-3 during helminth infections [29] suggests that chitin breakdown products from helminthes could be responsible for alternate activation of macrophages . We explored this possibility by stimulating human PBMCs and murine BMDM with chtx . Ym-1 , Arginase-1 and IL-10 ( Figure 6 A–D , Figure S 4 ) were up-regulated by murine BMDM upon stimulation with chtx . Similarly , human monocytes released IL-10 ( Figure 6 E ) and up-regulated intracellular Arginase activity confirming induction of alternate activation of mononuclear cells by chtx in vitro ( Figure 6 D , E ) . The potential of chtx to induce alternate activation of macrophages in vivo was also addressed . BALB/c mice were intraperitoneally administered with chtx or LPS and after 90 minutes the peritoneal cells were harvested and expression of Ym-1 and Arginase-1 in CD14+ve cells were analyzed . Chtx up regulate both Ym-1 and Arginase-1 while such an up-regulation was not observed in LPS administered mice peritoneal macrophages ( Figure S 4 G , H ) . Further , in vivo alternate activation of macrophages by chtx even after onset of endotoxemia was demonstrated . Up-regulation of Ym-1 and Arginase-1 was observed in peritoneal macrophages collected 90 minutes post chtx administration in mice with ongoing endotoxemia . It was observed that chtx induce alternate activation of macrophages even after onset of endotoxemia ( Figure 6 F , G ) . Chitin and chitin breakdown fragments have been reported extensively to activate macrophages but the pathway of activation and receptors involved remain contradictory [30] . The above described results suggested that chtx activates macrophages through alternate pathway using TLR4 . We sought direct biological proof by stimulating BMDMs of C3H/HeJ ( TLR4 mutant mice ) [31] and C3H/OuJ ( wild type mice ) with chtx and LPS and canonical markers of both alternate as well as classical macrophage activation were scored . As expected , LPS induced release of inflammatory cytokines by BMDMs of wild type mice and not by cells of mutant C3H/HeJ mice ( Figure 7 A , B ) . Chtx on the other hand failed to induce classical activation markers viz; TNF-α , IL-1β and nitrite ( Figure 7 A , B ) but up-regulated expression of alternate activation markers viz; Ym-1 ( Figure 7 C , E ) and Arginase-1 ( Figure 7 D , F ) in wild type mice and not in mutant mice . From these observations we make two broad conclusions: a ) classical and alternate macrophage activation could be mediated using the same receptor , TLR4 by LPS and chtx respectively and b ) the mutation in TLR4 gene that results in substitution of proline to histidine in its intracellular domain [32] plays a critical role in both classical as well as alternate activation pathways in macrophages . Finally we addressed the significance of these observations in human lymphatic filariasis . The expression of TLR4 was significantly low on monocytes of infected subjects positive for CFA ( Circulating Filarial Antigen ) in comparison to endemic controls ( negative for CFA ) ( Figure 8A ) . Binding of labeled FAg to monocytes was also significantly less in infected subjects in comparison to endemic controls ( Figure 8B , C ) however when monocytes of infected subjects were incubated in vitro at 37°C for 4 hrs binding of labeled FAg as well as expression of TLR4 were comparable to endemic controls ( Figure 8D , E ) . We interpret these findings to imply that circulating filarial antigens saturate TLR4 on monocytes which get recycled when incubated in culture thus exposing surface TLR4 to bind FAg or to react with anti-TLR4 ( Figure 8E ) . Further , LPS induced inflammatory molecules by PBMCs of infected subjects was significantly decreased in comparison to controls as shown by TNF-α , IL-1β and IL-6 ( Figure S 5A–C ) levels in culture supernatants suggesting that monocytes of subjects with filarial infections are less prone for activation by LPS possibly due to saturation of TLR4 with circulating filarial antigens . This issue was further addressed by incubating monocytes of infected individuals with heterologous ( negative for CFA ) normal plasma and FBS and stimulated with LPS or with LPS+FAg and inflammatory molecules like TNF-α , IL-1β and IL-6 were quantified in culture supernatants . LPS induced inflammatory cytokines by monocytes of infected individuals are comparable with monocytes of endemic controls when incubated with normal heterologous plasma or with FBS suggesting that filarial antigen in infected plasma saturate TLR4 on monocytes thus blocking activation by LPS ( Figure 8 F , G , H ) . Expression of CD23 , CD163 , and CD206 canonical markers of alternate macrophage activation [33] were significantly up-regulated on circulating monocytes of infected subjects in comparison to uninfected controls ( Figure 8 A ) .
Three novel issues stand out from the results being reported in this communication-a ) that a small molecular weight carbohydrate , chtx activates macrophages to a non-inflammatory phenotype through TLR4 and in doing so functions as an LPS antagonist and blocks induction of inflammatory mediators by LPS in vitro ( both in murine macrophages and in human monocytes ) and endotoxemia in vivo , b ) that two diverse pathways of activation of macrophages could be operational by two different ligands viz; LPS and chtx using a single receptor TLR4 on the host cell and finally c ) that glycoproteins of filarial nematodes with chtx as a constituent could be saturating TLR4 on circulating monocytes in infected subjects rendering them refractory to LPS induced inflammatory activation . We have provided evidence for direct binding of AgW to TLR4 by flow cytometry and consequent activation of alternate pathway of monocytes/macrophages . We have also shown by in silico analysis and solid phase immunoassay that chtx also binds toTLR4 and that the phenotype of activation pathway by chtx is similar to that of AgW . It is critical to note that unlike AgW , native LPS does not bind to TLR4 - several investigators have provided convincing evidence to suggest that native LPS does not bind directly to TLR4 but it activates macrophages by forming a complex with LPS binding protein ( LBP ) and CD14 and this complex delivers LPS to MD2 which activates macrophages through TLR4 [9]–[11] . When viewed in this context our findings on direct binding of AgW to TLR4 and activation of macrophages by alternate pathway is fundamentally different from macrophage activation by LPS which does so without directly binding to TLR4 in its native form . Benefits of inhibition of TLR4 activation has been documented in several experimental models of lethal shock . Anti-CD14 antibodies in rabbits , primates and humans [34]–[36] , anti-TLR2 antibodies [37] and antibodies to TLR4 [15] or TLR4/MD2 complex [17] have been tested with a high degree of success . Synthetic LPS antagonists such as Eritoran and Tak-242 have been tested in experimental models of endotoxic shock and also in human disease [19] but Eritoran has been recently reported to be ineffective in phase-III clinical trial ( http://clinicaltrials . pharmaceutical-business-review . com/news/eisai-eritoran-fails-to-meet-primary-endpoint-in-phase-iii-trial-250111 ) . Molecules involved in downstream signaling of TLR such as platelet-activating factor [38] oxidized phospholipids [22] , [23] , nitrate salts [20] , 5c [21] have also been tested with a degree of success . Potential of antibodies to TNF-α , IL-1RA , TNF-α soluble receptors and anti-bradykinin have also tested [39] but it has been observed that treatment with such TLR inhibitors interfere with innate immunity of host against infection and consequently increasing the risk of shock and mortality [39] , [40] , [41] . In this context the results of this study offers significant promise- a non immunogenic inexpensive small molecular weight chito-oligosaccharide can be used as an LPS antagonist . Based on in vitro demonstration of alternate activation of murine macrophages and human monocytes and in vivo activation of murine macrophages into alternate phenotype by chtx leading to inhibition of LPS mediated induction of inflammatory molecules ( such as IL1β , TNF-α , IL-6 etc ) by chtx we conclude that the small molecular weight carbohydrate induces alternate activation of macrophages in vivo and mediates protection against endotoxemia . Thus Chtx appears to protect against endotoxemia by two mechanisms - a ) it competitively inhibits LPS induced activation by binding to TLR4 and/or b ) it activates macrophages by alternate anti-inflammatory pathway . Generation of such an activation appears to have many advantages since alternately activated macrophages are reported to be endotoxin resistant [42] , [43] with increased phagocytic activity [44] and enhanced expression of scavenger receptors and proangiogenic factors [45] make them assist in tissue repair and resolution of inflammation [46] . We are currently testing if monocytes of sepsis patients can be re-programmed to non-inflammatory state by chtx . LNFP-III a complex carbohydrate moiety of S . mansoni , thioredoxin peroxidase of F . hepatica and migration inhibition factor ( MIF ) of B . malayi [47] , [48] are other molecules of helminth origin reported to induce alternate activation of macrophages but the host receptor through which such activation is mediated is still largely unknown . The current study demonstrating the role of TLR4 in chtx induced alternate activation of macrophages has offered insights into the issue of induction of alternate activation by helminth products . Although there have been many studies examining TLR signaling in response to pathogens [49]–[51] fewer studies have examined interaction of multicellular helminth parasites with TLRs on monocytes or macrophages . Helminth products such as excretory secretory ( ES ) product of Necator americana or OV-Asp-1 of Onchocerca volvulus [52] , [53] are known to interact with cells of the innate immune system but the receptor associated with this interaction has not been elucidated . In the present study a filarial glycoprotein designated as AgW has been demonstrated to bind directly to TLR4 . These results are in broad agreement with demonstration of ES-62 , a filarial glycoprotein interacting with TLR4 [24] . Apart from identifying TLR4 as a receptor for a helminth carbohydrate , the current study will be of crucial interest to cell biologists since TLR4 appears to function as a common receptor for both classical as well as alternate activation of macrophages and the nature of ligand determining the phenotype . This clearly offers scope for acquiring insights into molecular events involved in mutually exclusive activation pathways of macrophages . Human filariasis is characterized by chronic persistence of circulating filarial antigens ( CFA ) for several years . TLR4 on monocytes in infected subjects appear to be saturated in vivo with CFA since labeled FAg bound poorly bound to monocytes ex vivo and normal binding of FAg to TLR4 could be achieved by allowing antigen saturated TLR4s to recycle in vitro . The findings reported here suggest that CFA in plasma seem to remain bound to TLR4 on monocyte surface in infected subjects and contribute to sustenance of their alternate activated state . The following observations indicate that inherent defect in monocytes of infected subjects do not contribute to their failure to get activated by LPS: 1 . Monocytes of infected subjects when incubated with infected ( autologous ) plasma , the response to LPS was significantly diminished ( as shown by decreased TNF-α , IL-1β and IL-6 levels in culture supernatants ) when compared with response of same monocytes incubated with FBS or normal plasma; 2 . Monocytes of infected subjects cultured with normal plasma respond poorly to LPS when FAg was added exogenously; 3 . FAg significantly inhibited release of TNF-α , IL-1β and IL-6 by normal monocytes when cultured with autologous plasma and stimulated with LPS; 4 . When monocytes of infected subjects are incubated with FBS or with normal human plasma , the cells get activated well to LPS as shown by higher levels of inflammatory cytokines in supernatants - TNF-α , IL-1β and IL-6 levels are comparable to those observed in normal monocyte cultured with normal ( autologous ) plasma . While our observations that circulating monocytes in filariasis infected subjects display alternate activation markers are similar to an earlier report [54] diminished induction of mediators of inflammation such as TNF-α , IL-1β and IL-6 by LPS treated monocytes of filariasis infected subjects is a novel observation . The possibility that subjects with filarial infections will be regulating hyper inflammation associated with bacterial infections and thus offering protection against endotoxemia associated with sepsis needs further investigation . These findings also suggest interesting evolutionary issues on co-infection of humans with nematodes and gram negative bacteria and their pathogenesis . It is tempting to propose that increasing incidence of sepsis/septic shock in developed countries over the last 100 years ( 2 ) could have been due to eradication of helminth infections , a scenario similar to increased incidence of allergies and autoimmune diseases in developed countries as a consequence of elimination of infectious disease as proposed in ‘Hygiene Hypothesis’ .
Institutional Animal Ethics Committee of Institute of Life Sciences “approved” all the protocols followed for experiments conducted using mice . The study was carried out in strict accordance with the recommendations of the Committee for Prevention of Cruelty and safety of experiments with animals ( CPCSEA ) a regulatory body of Government of India that supervises Care and Use of Laboratory experimentation through their nominees in the Institutional animal ethics committee . Adult Setaria digitata worms from peritoneal cavities were collected from the abattoir attached to a local zoo after obtaining approval from zoo authorities . The animals are slaughtered in the abattoir regularly for feeding wild cats and no animals were slaughtered specifically for the purpose of our study . The study on human filariasis was approved by Institutional Human Ethics Committee of Institute of Life Sciences which operates under the guidance of regulations of Indian Council of Medical Research . Written informed consents were obtained from each of the normal control volunteers , filariasis infected persons and/or their legal guardians before collection of blood samples . Peritoneal dwelling adult female filarial parasites ( Setaria digitata ) were collected from cattle in a local abattoir , attached to the local zoological park at Nandankanan , Bhubaneswar after obtaining necessary approval from zoo authorities . The worms were transported to the laboratory in Dolbecco's Modified Eagles Medium ( DMEM ) ( Sigma ) pH 7 . 00 containing antibiotics [Penicillin Streptomycin solution 1 ml/100 ml of medium] ( Sigma ) and 1% glucose ( Hi-media ) . Aqueous extracts of S . digitata ( designated as FAg ) was prepared by homogenization followed by ultrasonication and the aqueous extract of adult worms was biotinylated using appropriate derivatives of biotin i . e . N-hydroxysuccinamide derivatives ( Sigma ) suitable for protein labeling . One milligram of FAg was passed through WGA-Sepharose ( Sigma ) coloumn and the unbound proteins were washed by passing PBS and glycoproteins bound to WGA were eluted by Glycine-HCl buffer ( pH 3 . 6 ) . The pH of the elutes was adjusted using 0 . 1 M NaOH and dialysed against PBS . Protein concentration of the elute was estimated and stored at −20°C for further use . BALB/c , C57BL/6 , C3H/OuJ and C3H/HeJ mice were obtained from National Institute of Immunology , New Delhi which was originally imported from Jackson laboratories , Germany . Breeding and maintenance were done at the animal facility at Institute of Life Sciences , Bhubaneswar , India . 8–10 weeks old animals were used for this study . Institutional animal ethics committee of Institute of Life Sciences , Bhubaneswar approval was obtained for all the investigations conducted in mice . Mouse bone marrow cells were collected from femoral shafts by flushing with 3 ml . of cold sterile DMEM ( Sigma ) supplemented with 20 mM of L-Glutamine ( ICN ) , antibiotics ( 1 ml penicillin and streptomycin/100 ml of medium ) ( Sigma ) containing 10% FBS . The cell suspension was passed through a sieve to remove large clumps . The cell suspension was washed 2–3 times with sterile DMEM and adjusted to 0 . 5×106 cells/well and cultured in 24 well plates . After 8–10 hrs incubation at 37°C non adherent cells were removed by washing with sterile DMEM and the adherent cells ( more than 96% positive for CD14 ( Figure S-4 and S-6 indicating high purity ) were stimulated with LPS ( Sigma , 055:B5 L-2880 ) with or without FAg or chtx ( Dextra Lab ) at 10 µg/ml concentration . After 48 hrs the supernatants were aspirated , frozen at −80°c and used later for estimation of cytokine levels . The adherent cells were removed using chilled medium and analysed for intracellular Arginase activity by calorimetric assay and intracellular staining using antibodies to Ym-1 and Arginase-1 staining and scored by flowcytometry . 8–10 weeks old BALB/c and C57BL/6 mice were intraperitoneally injected with 15 mg/Kg body wt and 60 mg/Kg body wt . of LPS respectively with and without FAg ( 100 µg ) or AgW ( 50 µg ) or chtx ( 250 µg ) and observed for mortality a period over 168 hrs . These doses were determined by prior titration and the lowest concentration effective in vivo was chosen for experimentation . For analysis of levels of cytokines mice were sacrificed 2 hrs after challenge with LPS or LPS with FAg ( 100 µg ) or LPS with chitohexaose ( 250 µg ) . Blood was collected in heparinised tubes by heart puncture and clear plasma was isolated by centrifugation at 5000 g for 10 minutes and analysed for presence of TNF-α , IL-1β , nitrite and IL-10 as described below . Human PBMCs were isolated from heparinised venous blood samples by density gradient centrifugation method using Histopaque ( Sigma ) . Briefly , the heparinised blood was layered on LSM medium gently in the ratio of 1∶1 and subjected to centrifugation at 100 g for 30 minutes . The white layer representing PBMCs was aspirated out gently and transferred aseptically into sterile centrifuge tubes . The suspension of cells was then washed and cultured in sterile DMEM supplemented with 20 mM of L-Glutamine ( ICN ) , 10% of autologus plasma/FBS and antibiotics ( 1 ml penicillin and streptomycin/100 ml of medium ) ( Sigma ) . The no . of cells was adjusted 0 . 5×106 cells/well in 24 well plates . After 8–10 hrs incubation at 37°C non adherent cells were removed by flushing with sterile DMEM and the adherent cells were stimulated with LPS , FAg or chtx , LPS along with FAg and LPS along with chtx at 10 µg/ml concentration for 48 hrs . after which the supernatants were removed and used for cytokine estimation . The adherent cells were removed and analyzed for intracellular Arginase activity by calorimetric assay as described below . For study of recycling of the receptor , the PBMCs were resuspended in DMEM containing 0 . 1% BSA and incubated at 37°C for 4 hrs . Then cells were incubated with biotinylated FAg at 4°C for 30 minutes followed by staining with streptavidin-FITC and analyzed by FACS . Supernatants from human monocytes or mouse BMDM cultures as well as mice plasma samples were analysed for levels of TNF-α , IL-1β , IL-6 and IL-10 by a sandwich ELISA according to manufacturers instruction using commercially available ELISA kits ( e-Biosciences ) . Human PBMCs or mouse BMDM ( 1×106/ml ) were stained for 30 minutes at 4°C with fluorescence labeled antibodies specific to CD14 , TLR4 ( e Biosciences ) mixed with and without FAg or AgW along with relevant isotype controls . Human cells were also stained with antibodies to CD23 , CD163 , and CD206 ( Santacruz Biotech . ) . The cells were thoroughly washed to remove the unbound antibodies and analysed by FACS ( BD FACS caliber ) . For intracellular Arginase-1 and YM-1 staining , cells were permeabilised with 1× FACS permeabilising solution ( BD biosciences ) and then incubated with rabbit antibodies to mouse YM-1 ( Stemcell technologies ) or goat anti-mouse Arginase-1 antibodies ( Santacruz Biotech . ) followed by staining with anti-rabbit IgG-FITC ( Sigma ) and anti-goat IgG-PE ( Santacruz Biotech . ) respectively and analysed by FACS . Appropriate isotype controls/conjugate controls were used for all flowcytometric assays . Human PBMCs or mouse ( BALB/c ) bone marrow cells ( 2×106 ) were incubated with 2 ml of cell lysis solution ( Sigma ) and a cocktail of protease inhibitors ( Sigma ) for one hr . at 4°C and then ultra-sonicated . The supernatant was collected by centrifuging at 1500 g for 10 minutes and stored for further use . ELISA plates ( Nunc maxisorp ) were coated with 1 µg of PBS extracts of S . digitata , AgW , N . brasiliensis , H . polygyrus or mock BSA , Phosphorylcholine coupled to BSA , GlcNaC-BSA , Mannose-BSA or LPS . After blocking with 1% skimmed milk-PBS ( Hi-media ) , human PBMC lysates or mouse ( BALB/c ) bone marrow cell lysates were incubated for 2 hr at 37°C . The plate was thoroughly washed and were incubated with rabbit anti-human and rabbit anti-mouse TLR4 ( e Biosciences ) respectively . The binding of anti-human and anti-mouse TLR4 was detected by using peroxidase conjugated anti-rabbit IgG ( Sigma ) . The enzyme activity was measured using OPD ( Sigma ) . The intracellular accumulation of ROS was determined using the fluorescent probe ( 2 , 7 , Dichloro Dihydro Fluorescein Diacetate ) H2-DCFDA as described previously [56] . Nitrite level in culture supernatants of BMDM and intracellular Arginase activity of both human monocytes and BMDM lysates were quantified as described elsewhere [57] . Arginase activity was measured in cell lysates . Briefly , cells were lysed using 50 µl of 0 . 1% Triton X-100 . 5 µg Pepstatin and 5 µg aprotinin were used as protease inhibitors during lysis . This mixture was incubated for 30 min at room temperature . 50 µl of 10 mM MnCl2 and 50 mM Tris-HCl were added to lysed cells to activate the enzyme by heating for 10 min at 56°C . Then 25 µl of 0 . 5 M L-arginine , pH 9 . 7 was added and incubated at 37°C for 45 min . The reaction was stopped with 400 µl of H2SO4 ( 96% ) /H3PO4 ( 85% ) /H2O ( 1/3/7 , v/v/v ) . 25 µl of α-isonitrosopropiophenone ( dissolved in 100% ethanol ) was added to the mixture followed by heating at 95°C for 45 min and urea concentration was measured at 540 nm . The X-ray structure of the extracellular domain of TLR4 ( PDB code: 3FXI ) [58] in complex with MD-2 is available . The X-ray structure of the TLR4-MD-2 complex ( 3FXI ) was downloaded from the PDB data base . The TLR4 structure from this complex was extracted and used for docking with chtx . The chemical structure of chtx molecule was extracted from pubchem database ( http://pubchem . ncbi . nlm . nih . gov ) . Structure of the chtx was retrieved into two-dimensional MDL/SDF format and three dimensional coordinates were generated using the ACCELRYS DS modelling 2 . 5 ( Accelrys Inc . San Diego , CA 92121 , USA ) software suite . The missing hydrogen of the structure was fixed and subjected to energy minimization . All energy minimization were carried out using the conjugate gradient method of CHARM force field using the ACCELRYS DS modelling 2 . 5 ( Accelrys Inc . San Diego , CA 92121 , USA ) software suite . Docking studies were carried out using Genetic Optimization for Ligand Docking ( GOLD ) software , version 4 . 1 . 1 ( Cambridge Crystallographic Data Centre , Cambridge , UK ) . The number of run was set to 100 in the standard default settings . The standard default settings , consisting of population size-100 , selection pressure-1 . 1 , niche size-2 , migrate-10 , cross over-95 , number of operations-1 , 00 , 000 number of docking 10 were adopted for GOLD docking . For ligand- protein binding , 10 docking conformations ( poses ) were tested and the best GOLD score were considered for further analysis . The ligand showing maximum interactions with the protein were plotted using the program LIGPLOT . The active site was predicted by using Q-site finder [59] . Total RNA was isolated from stimulated cells using RNAeasy columns from Qiagen , as per the manufacturer's instructions . 1 µg of total RNA from each sample was treated with DNAse I ( Ambion Inc . ) . Synthesis of cDNA was performed by using First Strand Synthesis kit and the Superscript III Reverse Transcriptase ( Invitrogen ) , according to the manufacturer's instructions . All real-time PCR experiments were performed in ABI prism 7900 HT sequence detection system ( ABI ) as described earlier [60] . The PCR conditions were as follows: 95°C for 10 min , 95°C for 15 sec , 58°C for 30 sec and 72°C for 30 sec for 40 cycles . The primers used for each gene are listed in ( Table S 1 ) . Primers were used at a concentration between 1 and 5 pmoles per reaction . All the reactions were analyzed using the software ( SDS 2 . 3 ) provided with the instrument . The relative expression of the genes was calculated by using 2-ΔΔCt formula using GAPDH as a normalizer . The values reported are the mean of two biological replicates . The standard deviation from the mean is shown as error bars in each group . Statistical significance among experimental groups was analyzed by the unpaired Student's t-test using Graph pad prism software ( Prism-5 ) . | Sepsis is one of the leading causes of death contributing to mortality as high as 54 percent in intensive care units across the world . Hyper inflammation induced by bacteria or bacterial products through Toll like receptors leads to sepsis and hence current approaches are directed towards blockade such receptors . While many such candidate antagonists have shown promise they also result in induction of inappropriate innate immune responses thus increasing risk of development of shock leading to death . In this study we describe a novel approach to treat endotoxemia associated with sepsis , fundamentally different from other reports . Chitohexaose a small molecular weight polysaccharide by virtue of its ability to bind to active sites of TLR4 inhibited LPS induced production of inflammatory mediators by murine macrophages and human monocytes . Administration of chitohexaose with LPS blocked endotoxemia leading to mortality of mice . More significantly , Chitohexaose reversed inflammation and protected mice even 24/48 hrs after onset of endotoxemia . Apart from competitively inhibiting LPS induced inflammation chitohexaose also activated alternate pathway of macrophages . Such macrophages are known to display increased phagocytic activity , are resistant to LPS induced activation and associated with resolution of inflammation and tissue repair . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"medicine",
"biology",
"veterinary",
"science"
]
| 2012 | Chitohexaose Activates Macrophages by Alternate Pathway through TLR4 and Blocks Endotoxemia |
Influenza A viral infections have been identified as the etiologic agents for historic pandemics , and contribute to the annual mortality associated with acute viral pneumonia . While both innate and acquired immunity are important in combating influenza virus infection , the mechanism connecting these arms of the immune system remains unknown . Recent data have indicated that the Notch system is an important bridge between antigen-presenting cells ( APCs ) and T cell communication circuits and plays a central role in driving the immune system to overcome disease . In the present study , we examine the role of Notch signaling during influenza H1N1 virus infection , focusing on APCs . We demonstrate here that macrophages , but not dendritic cells ( DCs ) , increased Notch ligand Delta-like 1 ( Dll1 ) expression following influenza virus challenge . Dll1 expression on macrophages was dependent on retinoic acid-inducible gene-I ( RIG-I ) induced type-I IFN pathway , and not on the TLR3-TRIF pathway . We also found that IFNα-Receptor knockout mice failed to induce Dll1 expression on lung macrophages and had enhanced mortality during influenza virus infection . Our results further showed that specific neutralization of Dll1 during influenza virus challenge induced higher mortality , impaired viral clearance , and decreased levels of IFN-γ . In addition , we blocked Notch signaling by using γ-secretase inhibitor ( GSI ) , a Notch signaling inhibitor . Intranasal administration of GSI during influenza infection also led to higher mortality , and higher virus load with excessive inflammation and an impaired production of IFN-γ in lungs . Moreover , Dll1 expression on macrophages specifically regulates IFN-γ levels from CD4+and CD8+T cells , which are important for anti-viral immunity . Together , the results of this study show that Dll1 positively influences the development of anti-viral immunity , and may provide mechanistic approaches for modifying and controlling the immune response against influenza H1N1 virus infection .
Influenza virus type A causes acute respiratory infections that are highly contagious and cause significant morbidity and mortality in humans and animals [1] , [2] . In 2009 , the influenza pandemic caused by the current H1N1 virus affected all the continents of the world [3] . In the United States alone , the 2009 H1N1 influenza virus affected 57 million Americans , with more than 11 , 000 deaths ( CDC report; http://www . cdc . gov/h1n1flu/estimates_2009_h1n1 . htm ) . Although vaccines and other antiviral approaches to control influenza recently have been developed , the disease is by no means under control since these treatments are not available worldwide and their efficacy is less than optimal [1] , [4] . Thus , a better understanding of the molecular mechanisms of pathogenesis and of the host immune response to influenza virus infection is required for the prevention and treatment of influenza . A viral infection is initially sensed by the host innate system , triggering a rapid antiviral response that involves the release of proinflammatory cytokines , and eventually leads to the activation of the adaptive immune response [5] . The first line of defense is initiated when cellular pathogen recognition receptors ( PRRs ) recognize pathogen-associated molecular patterns ( PAMPs ) including influenza virus [6] , [7] . In many PAMPs , RNA virus is recognized not only by PRR Toll-like receptor 3 ( TLR3 ) but also by RIG-I and melanoma-differentiation-associated gene 5 ( MDA5 ) [8] . During the life cycle of influenza virus , these proteins in turn activate the TBK1 and IKKi kinases , which phosphorylate interferon-regulatory factor-3 ( IRF-3 ) and IRF-7 , transcription factors essential for the expression of type-I IFNs [9] . The type-I IFN ( IFN-α/β ) cytokines are vital to the innate immune response and control the expression of>100 gene products , several of which directly reduce viral replication and spreading by conferring the so-called “antiviral state” [10] . IFN-αβ activates these downstream processes by initially engaging the IFN-α receptor ( IFNαR ) and activating the JAK-STAT pathway [11] . This pathway induces a number of early-response , IFN-stimulated genes ( ISGs ) including type II IFN ( IFN-γ ) [12] . Furthermore , IFN-αβ also activates NFκB , which amplifies the IFN response via a positive-feedback loop . This feedback is important for the recruitment of specialized immune cells to the site of injury or viral infection [6] . IFN-αβ is initially produced by leukocytes and fibroblasts , leading to the recruitment of T and NK cells , that produce IFN-γ . IFN-γ induces and activates numerous key antiviral factors , most notably PKR ( RNA-activated protein kinase ) , a serine/threonine kinase induced by both IFN type I and type II stimulation [12] , [13] . Thus , IFN-αβ and IFN-γ affect the activities of macrophages , T cells , dendritic cells ( DCs ) , and NK cells by enhancing antigen presentation , cell trafficking , and cell differentiation profiles , which ultimately enhances antiviral effector functions [13] . In the last decade , it has been demonstrated that Notch signaling pathways contribute to both the hematopoietic and immune systems including a role in the development of embryonic hematopoietic stem cells and a role in multiple lineage decisions of developing lymphoid and myeloid cells [14] . There are 5 mammalian ligands ( Delta-like [Dll]1 , Dll3 , Dll4 , Jagged-1 , and Jagged-2 ) , each of which can activate any of the 4 Notch receptors ( Notch1 , -2 , -3 , -4 ) [15] . Notch signaling during lymphoid development has been extensively studied , and its essential role in specifying cell fate at many stages during T-cell development is well characterized [14] . Moreover , recent data have indicated that the Notch signaling pathway is an important modulator of Tcell-mediated immune responses [15] . For example , Notch signaling is associated with the differentiation of naive CD8+Tcells to cytotoxic T lymphocytes ( CTLs ) , and cytotoxic CD8+T are recruited to kill virus-infected cells by the production of IFN-γ [14] . Another function that was assigned to Notch is the regulation of T helper ( Th ) cell differentiation . The importance of Notch activation has been supported using GSI , which is a pharmacologic inhibitor of Notch signaling pathways , to block the induction of Th1-type responses [14] . Upon recognition of pathogens and presentation of antigen via MHC class II proteins by antigen-presenting cells ( APCs ) such as macrophages and DCs , CD4+Th cells become activated , drive adaptive immunity and induce specific responses to invading microbes [14] . For instance , Th1 cell induction by forced Dll expression on the surface of APCs was shown to induce Th1 cell differentiation , and Dll ligands were thought to inhibit Th2 cell differentiation by interfering with IL-4 receptor signaling [14] . On the other hand , expression of Jagged ligands , but not Dll , on the surface of APCs was shown to induce Th2 cell differentiation [14] . Further , we have demonstrated that Dll4 induction on DCs can specifically promote the generation of Th17 cells [16] . In the present study , we examine the role of Notch signaling during influenza H1N1 virus infection , focusing on APCs because of their central role in driving the immune system to overcome disease . We demonstrate that macrophages , but not DCs , increased Notch ligand Dll1 expression following influenza virus stimulation . Dll1 expression on bone marrow-derived macrophages ( BMDMs ) was dependent on RIG-I induced type-I IFN pathway , and not on the TLR3-TRIF pathway . We also found that IFNαR−/− mice failed to induce Dll1 expression on lung macrophages and had enhanced mortality during influenza virus infection . Our results further showed that specific neutralization of Dll1 during treatment with a Notch signaling inhibitor during influenza virus challenge induced higher mortality , impaired viral clearance , and decreased levels of IFN-γ . Together , the results of this study show that Dll1 positively influences the development of anti-viral immunity , and may provide mechanistic approaches for modifying and controlling the immune response against influenza H1N1 virus infection .
Since we previously demonstrated that Dll4 was upregulated on BM-derived DCs ( BMDCs ) following exposure to certain bacterial antigens including CpG ( TLR9 ligand ) and BCG [16] , we first assessed the gene expression profile of Notch ligands on APCs following influenza virus stimulation . During H1N1 stimulation no Notch ligands were induced on BMDCs ( Fig . 1A ) , while Dll1 mRNA levels were increased in BMDMs ( Fig . 1B ) . Dll3 expression was below detection levels of our assay . In agreement with the data from BMDMs , H1N1 induced the expression of Dll1 on RAW264 . 7 cells , a mouse leukemic monocyte macrophage cell line ( Fig . S1 ) . We next examined protein levels of Notch ligands following treatment with various TLR ligands . No TLR ligands induced expression of Dll1 on BMDCs ( CD11b+CD11c+ ) ( Fig . 1C ) . Though H1N1 failed to induce Dll4 on BMDCs , Dll4 expression was induced on BMDCs following LPS ( TLR4 ligand ) and CpG ( TLR9 ligand ) treatment , indicating that Dll4 induction on DCs is dependent on MyD88 signaling pathway as previously described [17] . When we examined BMDMs ( CD11b+F4/80+ ) , we found that Dll1 expression was induced during H1N1 stimulation as well as by PolyI:C ( TLR3 ligand ) and LPS stimulation , while no Dll4 expression was induced following any of these treatments ( Fig . 1D ) . In addition , ELISA analysis showed that H1N1 stimulation as well as PolyI:C and LPS stimulation , but not CpG stimulation , induced production of type-I IFNs by BMDMs ( Fig . S2 ) . The increased gene expression of both Dll1 and IFN-β were also associated with an increase of the viral load of H1N1 ( Fig . S3 ) . To further investigate the induction mechanism for Dll1 , we examined Dll1 expression using WT , TRIF−/− , MyD88−/− , and IFNαR−/− mice . As shown in Fig . 2A , the mRNA expression levels of Dll1 following H1N1 stimulation in BMDMs from IFNαR−/− mice was completely abrogated , while Dll1 expression in TRIF−/− and MyD88−/− mice was comparable to its expression in WT mice . Further , LPS stimulation of BMDMs from TRIF−/− mice did not increase expression of Dll1 when compared to WT mice . Moreover , BMDMs from IFNαR−/− mice had impaired induction of Dll1 mRNA following each stimulation condition we examined ( Fig . 2A ) . Additionally , when BMDMs were pretreated with anti-IFN-β Ab before treatment with H1N1 and PolyI:C , the expression of Dll1 was significantly decreased ( Fig . 2B ) . Flow cytometry data confirmed that Dll1 protein was not induced in BMDMs from IFNαR−/− mice following H1N1 stimulation ( Fig . 2C ) . These results were also supported by confocal immunofluorescent analysis , which indicated Dll1-positive expression ( red ) on F4/80-positive macrophages ( green ) following influenza virus treatment in WT mice; however , in IFNαR−/− mice , F4/80-positive macrophages ( green ) were Dll1-negative ( red ) following H1N1 stimulation ( Fig . 2D ) . RNA virus can trigger the TLR3-TRIF signaling pathway and/or the RIG-I like pathway , each of which induces type-I IFN . To determine whether these pathways also regulate Dll1 Notch ligand expression , we next examined type-I IFN production and Dll1 gene expression levels during H1N1 stimulation in TRIF−/− mice or by knocking down the RIG-I gene . IFN-α protein levels were significantly lower and IFN-β protein was not detectable in RIG-I siRNA-treated BMDMs compared with control siRNA-treated BMDMs ( Fig . 3 A and B ) . In contrast , levels of type 1 IFN expression were unchanged when BMDMs from TRIF−/− mice were compared to BMDMs from control mice ( Fig . 3 A and B ) . Similarly , the gene expression level of Dll1 was significantly lower in RIG-I siRNA-treated macrophages when compared with control siRNA-treated macrophages , whereas there was no significant difference in Dll1 gene expression between BMDMs from WT and TRIF−/− mice ( Fig . 3C ) . The above studies ( Fig . 2 ) suggested that signaling through IFNαR is critical for Dll1 induction . Thus , we next examined the contribution of the JAK/STAT pathway , which is downstream to IFNαR activation , on Dll1 expression . Following PolyI:C , LPS , H1N1 or rIFN-β stimulation , both STAT1 and STAT2 were phosphorylated and Dll1 was detected in BMDMs from WT mice ( Fig . 3D ) . However , in BMDMs from IFNαR-deficient mice no STAT1/2 phosphorylation and no Notch ligand Dll1 expression were seen . We also demonstrated that BMDMs from STAT1−/− mice and BMDMs from WT mice treated with JAK-I inhibitor failed to induce the expression of Dll1 following stimulation with PolyI:C , H1N1 or rIFN-β ( Fig . 3 E and F ) . In addition , knocking down the STAT2 gene led to significantly lower expression of Dll1 ( data not shown ) . Together these results suggest that phosphorylation of STAT1 and STAT2 are critical for Dll1 expression as is type-I IFN signaling through IFNαR . Thus , H1N1 infection leads to the production of type-I IFN via the RIG-I pathway in BMDMs . Type-I IFNs in turn bind to IFNαR in an autocrine loop and activates the JAK/STAT pathway that results in the transcription of Dll1 . Because we showed that Notch ligand Dll1 was critically regulated by IFNαR in vitro , we next examined whether IFNαR−/− mice infected with influenza virus failed to upregulate Dll1 . First , we monitored the survival of WT , IFNαR−/− , TRIF−/− , and MyD88−/− mice following H1N1 infection up to Day 20 . We confirmed that the absence of IFNαR led to increased mortality after viral challenge when compared to WT mice ( Fig . 4A ) . However , mice deficient for TRIF or MyD88 were not significantly different from WT mice regarding mortality ( Fig . 4A ) . These findings were confirmed in lung histology studies 8 days post infection , that showed a significant increase in lung inflammation in IFNαR−/− mice , as compared to the WT , TRIF−/− , and MyD88−/− mice ( Fig . 4B ) . In agreement with our in vitro BMDM data , we found that Dll1 mRNA levels were increased in whole lungs from WT mice over the study period , while the expression of Dll1 in whole lungs from IFNαR−/− mice was significantly lower on both Day 4 and Day 8 after viral challenge ( Fig . 4C ) . In contrast , no significant difference was observed in expression of Dll4 , Jagged1 , and Jagged2 in lungs from WT and IFNαR−/− mice ( Fig . 4C ) . Dll3 expression was below detection levels of our assay ( data not shown ) . In addition , flow cytometry demonstrated that Dll1 expression on lung macrophages ( CD11b+F4/80+ ) was significantly lower in IFNαR−/− mice when compared with WT mice ( Fig . 4D ) . These results were also confirmed by confocal microscopy , which showed impaired detection of Dll1 ( red ) on F4/80+ macrophages ( green ) in IFNαR−/− mice during H1N1 infection ( Fig . 4E ) . To directly examine the importance of macrophages , we used liposome- Dichloromethylenediphosphonic acid ( DMDP ) to deplete macrophages [18] . Intranasal administration of liposome-DMDP during influenza infection led to higher mortality ( Fig . 5A ) with greater virus load of 50% tissue culture infective dose ( TCID50 ) at both day 2 and day 7 post-infection ( Fig . 5B ) . The gene expressions of influenza H1N1 viral specific mRNA for matrix protein ( M1 ) and nonstructural protein ( NS ) were also significantly higher in liposome-DMDP-treated mice ( Fig . 5 C and D ) . Cellular appearance of bronchoalveolar lavage ( BAL ) cells demonstrated decreased number of macrophages and increased number of neutrophils ( Fig . 5 E and F ) . In addition , most of the remaining macrophages in BAL cells from liposome-DMDP-treated mice that were counted in Fig . 5F had the appearance of dead cells . Moreover , the expression of Dll1 from whole lungs was significantly lower in liposome-DMDP-treated mice ( Fig . 5G ) . We also demonstrated that protein levels of IFN-γ from lungs of macrophage depleted H1N1-infected mice were significantly impaired compared to control liposome-treated mice ( Fig . 5H ) . To directly test the effect of Dll1 against influenza infection , we blocked Dll1 functionality in WT mice by intraperitoneal passive immunization with anti-murine Dll1 Ab . We confirmed the specificity of this antibody with stably transfected OP-9 cell lines for Notch ligands Dll1 , Dll4 , or Jagged1 . The purified antibody was found to react only with the cell line expressing Dll1 ( Fig . 6A ) . Mice were treated intraperitoneally with anti-Dll1 or control IgG antibody ( 1 mg ) on day 0 , 2 , and 4 of viral challenge . We also examined the expression of Dll1 from lung macrophages ( CD11b+F4/80+ ) at Day 7 post-infection to demonstrate whether the Dll1 antibody has an inhibitory effect in vivo . Flow cytometry analysis showed that the protein level of Dll1 after treating H1N1 infected mice with anti-Dll1 antibody was similar to that seen in control PBS-treated mice ( Fig . 6B ) . Treatment with this purified anti-Dll1 Ab led to significantly increased mortality compared to control IgG treated mice ( Fig . 6C ) . Histological assessment showed more severe pneumonia in anti-Dll1-treated mice 7 days post influenza infection ( Fig . 6D ) . Next , viral load was assessed by measuring both TCID50 ( Fig . 6E ) and influenza H1N1 viral specific mRNA for M1 and NS ( Fig . 6 F and G ) . The results showed significantly higher virus load in the lungs of mice that received anti-Dll1 Ab compared to controls at day 7 post-infection . We further demonstrated that the whole lung expression of Hes1 , a downstream transcription factor which is a target of Notch pathways [17] , was significantly lower in the lungs of H1N1 infected mice treated with anti-Dll1 Ab ( Fig . 6H ) . To help elucidate the mechanism underlying the increased mortality and severe inflammation seen in anti-Dll1 Ab treated mice , we examined the cytokine and chemokine profile in whole lungs during H1N1 challenge . Interestingly , the protein level of IFN-β was significantly higher while that of IFN-γ was significantly lower in H1N1-infected whole lungs from anti-Dll1-treated mice compared to lungs from control mice 7 days post-infection ( Fig . 7A ) . Additionally , whole lungs from anti-Dll1 Ab treated mice at day 7 post-infection had significantly higher protein levels of CCL2 and CXCL1 ( Fig . 7A ) , molecules that play a critical role in the recruitment of monocytes/macrophges and neutrophils into inflammatory lesions . The production of CXCL9 and CXCL10 , which support the migration of Th1 cells , was similar in anti-Dll1-treated mice and control mice ( Fig . 7A ) . In agreement with the chemokine and cytokine profile from whole lungs , flow cytometry demonstrated enhanced macrophage and neutrophil recruitment during H1N1 infection in anti-Dll1-treated mice at day 7 post-infection ( Fig . 7B ) . There was no significant difference in the number of T cells ( CD4+ and CD8+ cells ) , NK cells ( NK1 . 1+ ) , myeloid DCs ( mDCs; CD11b+CD11c+ ) , and plasmacytoid DCs ( pDCs; B220+CD11c+ ) ( Fig . 7 B and C ) , whereas the number of IFN-γ+ cells from each subset ( CD4+ , CD8+ , and NK1 . 1+ ) was significantly lower in anti-Dll1-treated mice ( Fig . 7D ) . Cells recovered from draining lymph nodes of H1N1-infected mice after in vitro H1N1 rechallenge , also demonstrated significantly impaired production of IFN-γ compared to control treated mice ( Fig . 7E ) . To directly examine the contribution of Notch signaling during influenza virus infection , we blocked Notch signaling by using GSI , a Notch signaling inhibitor . Intranasal administration of GSI during influenza infection led to higher mortality with excessive inflammation in the lungs compared to the control DMSO-treated group ( Fig . 8 A and B ) . Furthermore , viral load assessed by measuring both TCID50 and influenza H1N1 viral specific mRNA for M1 and NS indicated significantly higher virus load in the lungs of mice that received GSI compared to DMSO controls at day 7 post-infection ( Fig . 8 C–E ) . In addition , the expression of Hes1 from whole lung was significantly lower in the lungs of H1N1 infected mice treated with GSI ( Fig . 8F ) . We also demonstrated significantly impaired production of IFN-γ from lung CD4+ and CD8+ T cells compared to control treated mice ( Fig . 8G ) . Moreover , the data illustrate that there were considerable decreases in IFN-γ production of lungs from GSI-treated mice compared with control DMSO-treated mice ( Fig . 8H ) . To directly test the effect of Dll1 on the T cells , we performed an in vitro lung CD4+ and CD8+ T cell cytokine expression assay with H1N1-stimulated lung-derived macrophages from either WT or IFNαR−/− mice with either addition or deletion of Dll1 . As shown in Fig . 9 A and B , lung macrophages from the IFNαR−/− mice caused a significant decrease in IFN-γ production by T cells when compared to co-cultures with WT macrophages ( white bars ) . Moreover , addition of recombinant ( r ) Dll1 augmented IFN-γ production from T cells isolated from H1N1-challenged lungs and co-cultured with H1N1-treated lung-derived macrophages . The levels of IFN-γ using macrophages from IFNαR−/− or WT mice were comparable in presence of rDll1 . Additionally , anti-Dll1 Ab significantly decreased IFN-γ production from both CD4+ and CD8+ T cells ( Fig . 9 C and D ) . These responses were also seen when using both CD4+ and CD8+ T cells from draining lymph nodes ( Fig . S4 ) . To verify that the addition of Dll1 to co-cultures of macrophages and T cells was activating Notch pathways , we used quantitative real-time PCR to examine Hes1 expression . Cultures receiving Dll1 showed a 3 . 60±0 . 45-fold increase in Hes1 expression over cultures with macrophages and T cells alone . Taken together , our findings suggest that Dll1 is able to skew T cell maturation via Notch signaling pathways .
Our results demonstrate that the Notch signaling pathway and , in particular , the Notch ligand Dll1 is essential in the regulation of influenza H1N1 virus infection . To our knowledge , this is the first report defining this relationship and delineating the underlying mechanisms . Of the five Notch ligands , Dll1 is the only Notch ligand specifically upregulated on macrophages following influenza stimulation , but it is not expressed on DCs . Also , the peak expression of Dll1 on lung macrophages in mice coincides with the period of peak inflammation after H1N1 infection . Our studies confirmed that lung macrophages from in vivo H1N1 infected mice expressed Dll1 . Blocking Dll1 during viral infection led to significantly higher mortality and greater accumulation of inflammatory cells in the respiratory tract . In addition , neutralization of Dll1 during H1N1 infection altered CD4+ and CD8+ T cell activation responses as measured by IFN-γ+ producing cells within the lung . Together , these results have detailed the mechanisms by which the elements of the immune system cooperate and coordinate their efforts to eliminate viral infection . Our understanding of these mechanisms may possibly lead to clinical approaches to fight influenza pandemics . The innate immune response is the first defense of the host to invading pathogens . Once initiated , proinflammatory cytokines and chemokines are released which cause macrophages and neutrophils to migrate to the source of infection [19] . Among the cytokines induced during the innate immune response , activation of type-I IFNs is the most powerful defense mechanism against influenza viral replication and spread [19] . We first demonstrated that macrophages , but not DCs , showed enhanced Notch ligand Dll1 expression in response to influenza virus and to type-I IFN cytokines , which suggested that Dll1 induction is dependent on type-I IFNs . We confirmed this by showing that IFNαR−/−-derived BMDMs completely failed to induce Dll1 . Influenza virus amplifys the type-I IFN response via a positive-feedback loop that activates JAK-1 and Tyk-2 kinases , which leads to the phosphorylation and dimerization of STAT1 and STAT2 proteins [6] , [20] . Our studies also showed impaired Dll1 induction on BMDMs from STAT1−/− mice and BMDMs treated with a JAK-1 inhibitor . PRRs that recognize influenza virus RNA , have been shown to be a key initiator of type-I IFN response in infected cells [6] . These PRRs rely on the RIG-I-like signaling pathway , composed of RIG-I and MDA5 , and also the TLR3-TRIF pathway [5] . Kato et al . demonstrated that mouse fibroblasts lacking RIG-I , but not MDA5 , are defective in the production of type-I IFN in response to influenza virus [21] . Our study also showed that RIG-I-knocked down BMDMs expressed decreased Notch ligand Dll1 with significantly decreased type-I IFN cytokine production following influenza virus stimulation . We also observed that MDA5-knocked down BMDMs expressed levels of Dll1 similar to BMDMs treated with control siRNA ( data not shown ) . In addition , we showed that Dll1 and type-I IFN production in BMDMs was TRIF independent . Thus , our results show that influenza virus-induced type-I IFNs are exclusively RIG-I dependent and that their production is essential for the induction of Dll1 through the IFNαR and the JAK-1/STAT1/2 signaling pathway . Using IFNαR−/− mice , our studies confirmed how critical type-I IFNs are for protection against influenza H1N1 virus in agreement with a recent report using influenza H5N1 virus [22] . Using a liposome-DMDP system [18] , we also demonstrated that macrophages are indispensable for combating influenza virus infection . Though the depletion of macrophage seems incomplete from the number of macrophages remaining , most of the remaining macrophages in BAL cells from liposome-DMDP-treated mice that were counted were likely under going apoptosis . Interestingly , the production of type-I IFNs from whole lungs during influenza virus infection was higher in anti-Dll1-treated mice compared with control Ab-treated mice , suggesting that enhanced type-I IFNs production from anti-Dll1-treated mice might be due to impaired viral clearance . These findings indicate that Dll1 expression on macrophages is crucial for protection against influenza virus . The initial interaction between invading microorganisms and the innate immune system critically influences the development of adaptive antiviral immunity [13] . Although both types of IFNs ( type-I and type-II ) play crucial roles in the immediate innate cellular response to viral infection , the immunomodulatory activities of IFN-γ have a large role in coordinating the adaptive immune response and in maintaining an antiviral state for longer times [12] . In addition , there is increasing evidence that the Notch system is an important bridge between APCs and T cell communication circuits [14] , [15] . Other studies have demonstrated that APCs encountering pathogens that skew the immune response to a CD4+ Th1 cell response , showed an upregulation of Dll1 [14] , [23] . Notch signaling is also associated with the differentiation of naive CD8+ T to cytotoxic T lymphocytes ( CTLs ) [24] . We first demonstrated that depletion of macrophages , a key player in Dll1 induction , induced decreased production of IFN-γ from lung CD4+ and CD8+ T cells with dampening of Dll1 levels during influenza virus infection . The change in Dll1 expression in this model was minor and suggested that we examined Dll1 expression in whole lungs . However , the upregulation of Dll1 returns to naïve mice levels in the absence of macrophages . Our results further showed that specifically blocking Dll1 during influenza infection impaired the survival and inflammatory status in our model with a decreased number of IFN-γ+CD4+ and IFN-γ+CD8+ T cells . Moreover , blocking of Notch signaling by GSI , which has been used in clinical trials as a cancer therapy approach , abrogated the survival and pathogenesis of lung inflammation with a decreased number of IFN-γ+CD4+ and IFN-γ+CD8+ T cells , suggesting the pivotal role of Dll1 through Notch signaling in driving IFN-γ mediated immune response to influenza virus . The expression of Hes1 in lungs was upregulated following influenza virus infection , and the treatment with anti-Dll1 antibody or GSI led to a decreased expression of Hes1 . However , the reduction of IFN-γ from the lungs of influenza-infected mice with these treatments was approximately 30% . This incomplete reduction might be attributed to the immunity of NK cells , one of major producer of IFN-γ during influenza virus infection , to these treatments . APCs , in particular , DCs and macrophages , have a key role in regulating and modulating the immune response [25] . Our findings indicated that induction of Dll1 on macrophages in response to influenza virus specifically regulated IFN-γ production from CD4+ and CD8+ T cells both in vivo and in vitro . Our studies demonstrate that anti-Dll1-treated mice exhibited significantly impaired survival accompanied by an impaired IFN-γ level . Our studies also showed that Dll1 is required for optimal IFN-γ production in response to Ag . Moreover , we demonstrated that GSI-mediated inhibition of Notch signaling attenuated overall IFN-γ production and resulted in fewer numbers of IFN-γ+CD4+ and IFN-γ+CD8+ T cells in our influenza model . Although IL-12 is known to be a strong inducer of CD4+ Th1 cell development , it has been reported that the Th1 response induced by Dll-mediated Notch signaling is IL-12 independent [26] . In our studies , blocking of IL-12 did not alter IFN-γ production from CD4+ and CD8+ T cells in co-culture system of APCs and T cells , and we could not detect IL-12 production in either BMDM or influenza virus-infected lungs ( data not shown ) . Thus , our results show that Notch ligand Dll1 is required to promote IFN-γ production from CD4+ and CD8+ T cells in IL-12 independent manner , a scenario which might be important in the protective immune response against influenza virus . Several studies support our results showing that IFN-γ plays an important role in recovery from influenza viral infection by helping to clear the virus [27] , [28] , [29] . In contrast , using IFN-γ-deficient mice , Graham et al . showed that IFN-γ is not necessary for recovery from influenza virus infection [30] . Possibly , the protective or non-protective role of IFN-γ is dependent on the model system . There may be a balance that is perturbed in some models that inhibits the protective effects of IFN-γ during viral infection . Certainly , given the pleotrophic effects of IFN-γ in the immune response , it is easy to envision that IFN-γ KO mice would experience many different signaling pathway perturbances , masking the protective effects of IFN-γ in a “normal” immune response to virus infection . Thus , in different models an imbalance between inhibitory and activating signals could determine the role of IFN-γ after influenza virus infection , with full activation and signaling through Dll1 overcoming influenza viral-induced-inhibition of IFN-γ . This is also in agreement with the known protective role for protease-activated receptor-2 against influenza virus via IFN-γ dependent pathway [29] . We have not evaluated these ideas in our model and further investigations are needed . Also , lung epithelial cells and fibroblasts play critical roles in influenza infectious models . However , Dll1 expression was not upregulated following H1N1 influenza stimulation in lung epithelial and fibroblast cell lines ( data not shown ) . It is not known what role the Notch system plays in these cells during influenza infection; determining this also may reveal a potential clinical target for fighting influenza virus-induced pneumonia . Neutrophils and macrophages are the dominant leukocytes recruited to the lung during an influenza infection [31] , [32] , and this process is markedly augmented in both IFNαR−/− mice ( data not shown ) and WT mice treated with anti-Dll1 Ab . The recruitment of more inflammatory cells into lungs enhances damage to lung cells and structures , including the respiratory epithelium , which might be related to higher mortality . Importantly , we found significantly higher levels of chemokines CXCL1 and CCL2 in infected IFNαR−/− mice ( data not shown ) and WT mice treated with anti-Dll1 Ab . CXCL1 plays a role in the recruitment of neutrophils , and CCL2 plays a role in macrophage recruitment [33] . It has been previously reported that blocking expression of CXCR2 , the receptor for CXCL1 , resulted in a reduction of neutrophil influx with prolonged host survival during influenza infection [34] . In addition , Dawson et al . showed that CCR2 deficiency , a major receptor for CCL2 , leads to a milder inflammatory response with reduced lung pathology and increased survival rates because of defective macrophage recruitment [35] . The above published reports agree with our findings , which show that higher CXCL1 and CCL2 levels in both lungs from IFNαR−/− mice and lungs with anti-Dll1 Ab might be correlated with not only enhanced neutrophil and macrophage migration into lungs but also impaired survival rate . In summary , we present a comprehensive analysis of Notch ligand Dll1 participation in an infectious model of influenza H1N1 virus . Blockage of Dll1 resulted in accelerated inflammatory responses and decreased IFN-γ levels from CD4+ and CD8+ T cells during influenza infection . Macrophages are indispensable for the protection against influenza virus by their enhancement of Dll1 expression levels during infection . Furthermore , Dll1 expression on macrophages was specifically regulated by type-I IFN . This study supports the concept that an understanding of Notch signaling , especially Dll1 regulation , in the immune response to influenza virus can provide mechanistic approaches that may have clinical applicability .
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 University Laboratory Animal Medicine ( ULAM ) Facility at the University of Michigan Medical School . All animal protocols were approved by ULAM and all efforts were made to minimise suffering . WT C57BL/6 mice , WT 129S6 mice , and STAT1−/− mice ( 129S6 Background ) were purchased from Taconic . C57BL/6 mice lacking the IFNαR gene ( IFNαR−/− ) were provided by M . Kaplan ( University of Michigan Medical School ) . All mice , including female MyD88−/− and TRIF−/− C57 BL/6 mice , were housed in the University Laboratory Animal Medicine ( ULAM ) Facility at the University of Michigan Medical School as described before [36] . All mice were used for experiments at 8–12 week of age . Age- and sex-matched mice were used in these studies . Rat mAbs specific for mouse CD3 ( 17A2 ) , CD4 ( L3T4 ) , CD8 ( 53–6 . 7 ) , CD11b ( M1/70 ) , CD11c ( HL3 ) , CD16/32 ( 2 . 4G2 ) , CD45 ( 30-F11 ) , CD45R/B220 ( RA3-6B2 ) , Gr-1 ( RB6-8C5 ) , NK1 . 1 ( PK136 ) , MHC Class II ( M5/114 . 15 . 2 ) , IL-12 ( C17 . 8 ) , and IFN-γ ( XMG1 . 2 ) were purchased from BD PharMingen . Rat Anti-F4/80 ( CI: A3-1 ) mAb was purchased from Serotec . Hamster anti-Dll1 and anti-Dll4 mAb for flow cytometry were purchased from BioLegend . Antibodies to STAT1 and STAT2 were purchased from Cell Signaling Technology , and Millipore , respectively . PolyI:C was from InvivoGen . LPS from Escherichia coli ( O55:B5 ) was from Sigma-Aldrich . Mouse cytosine-phosphate-guanosine ( CpG ) DNA was from Cell Sciences . Recombinant mouse IFN-α and IFN-β were from PBL InterferonSource . Mouse IFN-β Ab for neutralization was from BioLegend . JAK-I inhibitor and γ-secretase inhibitor ( GSI ) X , a cell-permeable hydroxyethylene dipeptide isostere that acts as a highly specific and a potent inhibitor of γ-secretase were from Calbiochem . DMDP encapsulated liposomes and control plain liposomes were from Encapsula . Mouse cell lines , RAW264 . 7 , M2-10B4 , and LA4 were purchased from the American Type Culture Collection ( ATCC ) . Rabbit anti–mouse Dll1 antibodies were prepared by multiple-site immunization of New Zealand white rabbits with recombinant mouse Dll1 ( R&D Systems ) in CFA and boosted with Dll1 in IFA , as in previously described procedures from our laboratory [17] . Polyclonal antibodies were titered by direct ELISA against Dll1 coated 96-well plates and titered at 107 . Mice were sensitized by intranasal injection of 1 . 0×104 PFU of influenza A virus strain ( strain A/PR8/34; H1N1 isotype: ATCC ) in 30 µl of PBS . PBS was inoculated intranasally into mock-infected mice . In some experiments , mice were treated intraperitoneally with anti-Dll1 or control IgG antibody ( 1 mg ) on day 0 , 2 , and 4 of viral challenge . Lungs and mediastinal lymph nodes ( LNs ) were harvested at the indicated time after influenza infection . Lung left lobe was used for histological assessment , and each right lobe was used for the analysis of mRNA , protein , flow cytometry , and virus infectious titer . Lung homogenates were serially diluted in Minimum Essential Medium Eagle medium ( Sigma-Aldrich ) and virus infectious titers were measured using the 50% tissue culture infectious doses ( TCID50 ) assay based on cytopathic effect as previously described [31] . Individual excised lung lobes were inflated and fixed with 10% buffered formalin for morphometric analysis . For immunofluorescent analysis , lungs were embedded in Tissue-Tek OCT compound , and then frozen in liquid nitrogen . Seven-micron cryostat sections were then fixed in ice-cold acetone , incubated with primary antibodies , followed by the addition of appropriate Alexa-labeled secondary reagents ( Invitrogen Corp . ) . Finally , the sections were analyzed by Zeiss LSM 510 confocal microscope system ( Carl Zeiss Inc . ) . Total RNA was isolated from the cultured cells and whole lungs using RNeasy Mini kit ( Qiagen ) following the manufacturer's instructions and was reverse transcribed in a 25μl volume . Briefly , 1 . 0 µg RNA was reverse transcribed to yield cDNA in a 25-µL reaction mixture containing 1× first strand ( Life Technologies ) , 250 ng oligo ( dT ) primer , 1 . 6 mmol/L dNTPs ( Invitrogen ) , 5 U RNase inhibitor ( Invitrogen ) , and 100 U Moloney murine leukemia virus reverse transcriptase ( Invitrogen ) at 38°C for 60 min; and the reaction was stopped by incubating the cDNA at 94°C for 10 min . The SYBR primer sets for Notch lignads were purchased from Sigma-Aldrich [16] . Real-time quantitative PCR analysis was performed using an ABI 7700 sequence detector system ( PE Applied Biosystems ) . The thermal cycling conditions included 50°C for 2 min and 95°C for 10 min , followed by 40 cycles of amplification at 95°C for 15 s and 55°C for 1 . 5 min for denaturing and annealing , respectively . Quantification of the genes of interests were normalized to GAPDH and expressed as fold increases over the negative control for each treatment at each time point , as previously described [16] . For virus quantification , cDNA was synthesized by using MultiScribe reverse transcriptase and random hexamers ( PE Applied Biosystems ) as previously described [37] . For Real-time quantitative PCR , the following SYBR primers were used: for the M1; forward 5’-CATCCCGTCAGGCCCCCTCA-3’ , reverse 5’-GGGCACGGTGAGCGTGAACA-3’ , for the NS; forward 5’-GGGGCAGCACTCTTGGTCTGG-3’ , reverse 5’-CGCGACGCAGGTACAGAGGC-3’ . BMDMs were lysed in lysis buffer ( Cell Signaling ) , briefly sonicated , kept on ice for 30 minutes , and centrifuged at 15 , 000 g for 15 minutes . The supernatant was collected and stored at −80°C until use . Equal amounts ( 15–30 µg ) of cell lysates were fractionated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis ( Invitrogen ) . Then the proteins were transferred onto a nitrocellulose membrane . After the overnight incubation with appropriate primary antibody , the membrane was counterstained with horseradish peroxidase-conjugated rabbit or mouse IgG antibody and visualized with enhanced chemiluminescence detection reagents ( GE Healthcare ) . A total of 1 . 5×106 BMDMs were transfected with 2 µg of a mixture of RIG-I ( Ddx58 ) -specific , MDA5 ( Ifih1 ) -specific , STAT2-specific , or nontargeting control siRNAs ( Dharmacon ) , using mouse macrophage nucleofector kit ( Lonza ) according to the manufacturer's instructions and plated in a 12-well plate . After 24 hours , cells were used for experiments . Murine cytokine and chemokine levels were measured in 50 µl samples using a Bio-plex bead-based cytokine assay purchased from Bio-Rad Laboratories . IFN-α and IFN-β levels were measured by ELISA according to manufacturer's instructions ( PBL InterferonSource ) . The cytokine levels in lung homogenates were normalized to the protein present in cell-free preparation of each sample measured by the Bradford assay , as described previously [16] . Flow cytometric analyses of lung cells were performed as previously described [16] . In brief , whole lungs were dispersed in 0 . 2% collagenase ( Sigma-Aldrich ) in RPMI 1640 ( MediaTek ) and 5% FBS ( Atlas Biologicals ) at 37°C for 45 minutes to obtain a single-cell suspension . The cells were stained with indicated Abs after 10 minutes of pre-incubation with CD16/CD32 Abs ( Fc block ) and fixed overnight with 4% formalin . For intracellular staining of cytokines , lung cells ( 1 . 0×106 cells/well ) were cultured in 48-well plates containing plate-bound anti-CD3 ( 5 µg/ml ) and soluble anti-CD28 ( 2 . 5 µg/ml ) . After overnight incubation and in the presence of GolgiPlug ( BD Biosciences — Pharmingen ) for the last 2 hours at 37°C and 5% CO2 , the cells were stained for surface markers with FITC-conjugated anti-CD4 , anti-CD8 , or anti-NK1 . 1 Abs , resuspended in fixation/permeabilization solution ( BD Cytofix/Cytoperm Kit; BD Biosciences Pharmingen ) , and stained with PE-conjugated anti–IFN-γ Abs respectively . Cells were analyzed using a Cytomics FC 500 ( Beckman Coulter ) , and data were analyzed by FlowJo software ( Tree Star Inc . ) . BM was harvested from uninfected , normal mice , filtered through nylon mesh . For generation of BMDMs , BM cells were cultured in L929 cell-conditioned medium as described previously [16] . Six days after initial bone marrow culture , BMDM were transferred to well plates overnight . For generation of BMDCs , BM cells were seeded in T-150 tissue culture flasks at 106 cells/ml in RPMI 1640-based complete media with GM-CSF 20 ng/ml ( R&D Systems ) after depletion of erythrocytes with lysis buffer . 6 days later , loosely adherent cells were collected and incubated with anti-CD11c coupled to magnetic beads for isolation of conventional DCs from the GM-CSF cultures ( Miltenyi Biotec ) . The purity of CD11c was more than 94% using flow cytometry . The cells were plated in well plates overnight . The next day , macrophages and DCs were infected with certain stimuli . 7 days after 1 . 0×104 PFU of H1N1 intranasal injection , CD4+ or CD8+ T cells from lungs or mediastinal LN were isolated using a magnetic bead column ( Miltenyi Biotec ) . More than 95% of cells were CD4 or CD8 positive , respectively . For harvest of naïve lung macrophages , the whole lung cells dispersed in 0 . 2% collagenase were washed and resuspended in 10-ml RPMI 1640 , and then incubated in a 100-mm cell culture dish for 2 hour at 37°C and the non-adherent cells were removed . Adherent cells were collected as lung macrophages , and more than 95% were F4/80 positive . Naïve lung macrophages were pulsed with H1N1 ( MOI = 10 ) for 2 hours , and then T cells ( 2×105 cells/well ) were exposed to H1N1-pulsed lung macrophages in 96-well plates at APC:T cell ratio of 1∶5 , and supernatants were harvested 48 hours later for cytokine protein analysis . Plate-bound recombinant Dll1 ( rDll1 ) ( R&D Systems ) was used at a final concentration of 2 . 5 µg/ml , and anti-Dll1 Ab and control IgG were used at a final concentration of 20 µg/ml . Two-tailed Student's t test was performed in Prism ( Graphpad ) in all cases . of p<0 . 05 were considered statistically significant . *P<0 . 05; ***P 0 . 01; ***P<0 . 001 . | Influenza viruses cause annual epidemics and occasional pandemics that have claimed the lives of millions . Both innate and acquired immunity are essential for protection against influenza virus , and Notch and Notch ligands provide a key bridge between innate and acquired immunity . However , the role of Notch system during influenza virus infection is unknown . Here , we show that Notch ligand Delta-like 1 ( Dll1 ) expression was up-regulated in influenza virus H1N1 challenged macrophages , and was dependent on both retinoic-acid–inducible protein I ( RIG-I ) and IFNα receptor ( IFNαR ) -mediated pathways . IFNαR-deficient mice challenged with influenza virus in vivo also display a profoundly impaired Dll1 expression with increased mortality and abrogated IFN-γ production . Treatment of WT mice during influenza infection , with either neutralizing antibodies specific for Dll1 or a γ-secretase inhibitor ( GSI ) , which blocks Notch signaling , resulted in increased mortality , impaired viral clearance , and lower IFN-γ production . In addition , Dll1 specifically regulated IFN-γ production from both CD4+and CD8+T cells in vitro . Together , these results suggest that Notch signaling through macrophage-dependent Dll1 is critical in providing an anti-viral response during influenza infection by linking innate and acquired immunity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"medicine",
"immune",
"cells",
"cytokines",
"respiratory",
"infections",
"antigen-presenting",
"cells",
"immunity",
"to",
"infections",
"immunology",
"pulmonology",
"adaptive",
"immunity",
"infectious",
"diseases",
"inflammation",
"t",
"cells",
"biology",
"infectious",
"disease",
"modeling",
"immune",
"system",
"immunopathology",
"immunity",
"innate",
"immunity",
"viral",
"diseases",
"respiratory",
"medicine"
]
| 2011 | The Critical Role of Notch Ligand Delta-like 1 in the Pathogenesis of Influenza A Virus (H1N1) Infection |
Loiasis is a neglected tropical disease caused by infection with the filarial parasite Loa loa . It is a disease considered by many to be benign . Several reports of trans border importation of the Loa loa worm amongst immigrants and visitors from endemic regions of the world exist . In most cases an adult subconjunctival worm is removed from the patient . An interventional case series is reported and examines the practice of removal of subconjunctival adult Loa loa worms amongst urban dwellers in Nigeria . Four cases of ocular loiasis seen amongst urban dwellers in Nigeria exemplify the different presentations and removal methods of the subconjunctival adult worm . There were 2 males and 2 females aged 35years , 23years , 25years and 30years respectively . Each patient gave a history of having been raised in a rural community in childhood years , during which they were exposed to streams and muddy farm land; and then migrated to the urban community in later years . They all present with the finding of a subconjunctival adult worm , which was successfully removed and identified to be Loa loa . There are more urban dwellers in Nigeria who present with symptoms of foreign body sensation that may be related to the manifestation of a subconjunctival worm and are not recognized . This is because the emphasis on this disease has erstwhile been on the rural , village dwellers and not on urban dwellers . Eye care practitioners working in urban centers need to be aware of the possibility of this presentation , and be ready to remove any subconjunctival worm when it presents .
Ethical approval for this report was sort from the Eye Foundation Hospital Institution Review Board and was waived as this work involved only a retrospective reporting of information documented in the patients case records . The research was performed according to the principles of the Helsinki declaration and adequate informed consent was obtained from each patient prior to the surgical removal of the subconjunctival worm . Also , a verbal consent to report the case was obtained from the patients through a telephone contact before this case series . This consent was documented in the patient’s case records and was approved by the institution review board . Furthermore , all four subjects were adults .
Case 1: A 35 year old male presented with a complaint of sharp pain in the right eye the previous night lasting few minutes with associated itching and photophobia . There was no previous history of similar complaints . The left eye was normal . Ocular examination revealed; unaided visual acuity of 6/5 both eyes , palpebral conjunctival papillae and mild bulbar conjunctival hyperemia in both eyes . He received topical Olopatadine ( a prescription eye drop with mast cell stabilizing and antihistamine effect ) for treatment of the presumed ocular allergic condition . He returned 3 days later with a complaint of a worm moving in the right eye the previous morning . There was no ocular pain or itching and no generalized pruritus , skin rashes , swelling or joint aches . His vision remained unchanged in both eyes , with obvious right eyelid swelling . He had hematological and dermatological investigations , which include; Full Blood Count , peripheral blood film and skin snip test for microfilaria . Results came out as normal , with no eosinophilia noted . However , upon repeat slit lamp examination of the right eye a mobile worm was noticed in the nasal sub conjunctival space "Fig 1" . Dilated funduscopy showed cup-to-disc ratio ( CDR ) 0 . 3 pink , normal macula , vessels and flat retinae . Systemic examination was normal , no evidence of cutaneous lesions , subcutaneous swellings or nodules . Upon further questioning he gave a history of having worked in swampy rural community farmlands as a child . He was taken to the operating room the same day and with a retrobulbar anesthesia , a small conjunctival incision was made inferonasally adjacent to the worm . A white colored live worm was grasped with toothless forceps and extracted carefully , intact . Topical antibiotic and steroid preparations were given post surgery . Oral Albendazole was given as therapy targeting any remaining adult worms , and oral Ivermectin targeting microfilaria . Microscopic evaluation of the specimen done at the histopathology laboratory revealed a 7 cm adult male Loa loa worm "Figs 2 and 3" . Case 2: A 23 year old male , presented with symptoms of crawling sensation and foreign body sensation in his right eye , and had observed a worm in this eye . He had no previous symptoms prior to his presentation . He gave a history of having worked in a cocoa farm plantation during his childhood years and had severally suffered from bites from unknown flies . He had no systemic symptoms nor signs and aside from his ocular complains was healthy . There was no swelling anywhere in the body and no itching . Upon ocular examination his visual acuity was 6/6 in both eyes . The only significant finding was the presence of an actively mobile worm in the nasal subconjunctival space of the right eye "Fig 4" . This worm soon migrated upwards towards the superior fornix and away from view during the examination and before removal could be attempted "Fig 5" . The patient was immediately asked to adopt a face down position and within 30minutes of this time; he could feel a crawling sensation again in the same eye indicating that the worm was back . He was quickly taken to the operating room and the worm was extracted successfully using a local infiltration of the conjunctiva with lignocaine anesthesia . Histological examination revealed it to be an adult Loa loa worm . Cases 3: A 25 year old female who had suffered sensation of movement and foreign body sensation in both eyes for the past 10 years and gave a past history of swimming in rural streams during childhood years . There was no history of swelling on the body and no itching . She had noticed an increasingly frequent occurrence of a worm like movement in both eyes over these years . Following ingestion of diethyl carbamazepine she noticed a sudden appearance of a red patch in the right eye . Upon examination her visual acuity was 6/5 in both eyes . The only significant finding was a localized hyperemic raised lesion on the surface of the right eye . This turned out to be a subconjuctival worm in the inferotemporal subconjunctival space of the right eye . The worm was found to be lifeless and covered by a surrounding cyst wall "Fig 6" . Care was taken to dissect the conjunctival and subtenons tissue away from the encysted worm , which was carefully extracted with a toothless forceps . Conjunctival incision site was closed with interrupted sutures . Histology revealed an adult Loa loa worm . Case 4: A 30 year old female who presented with symptoms of redness in the right eye and seeing a worm moving in the left eye . She also gave a history of exposure to rural streams in childhood . On ocular examination , visual acuity was 6/5 in both eyes . She had a temporal subconjunctival hemorrhage as the only ocular finding . The left eye worm she had seen earlier was no longer present . She was reassured and informed to return to the clinic upon seeing the worm again . She re-presented eleven months later upon seeing the worm again , with symptoms of left eye recurrent redness and feeling of something moving in the eye . This was associated with generalized body itching worse at night times . Ocular examination revealed a mobile worm in the temporal subconjunctival area of this eye . The worm soon migrated to the superior bulbar conjunctiva "Fig 7" . Through a conjunctival incision the life worm was extracted using a forceps and the wound was repaired . Histology confirmed an adult Loa loa . Blood work up including investigations for microfilaria was negative; but there was a positive eosinophilia of 44 . 4% . She was treated with oral Albendazole .
Loa loa is a filarial nematode , which infects millions of people in Western and Central Africa , to cause loiasis . It enters the human body through the bite of the deer fly ( Chrysops ) and explores the subcutaneous tissue of the host undetected , until it enters the subconjunctival space when it becomes symptomatic , visible and presents an opportunity to be physically removed . When present in the subconjunctival space it is freely mobile and the patient complains of a foreign body and , or crawling sensation , seeing a moving worm or tumor , redness and itchy sensation . Its subconjunctival presentation has earned it the name "African Eye Worm . " In some cases it can be associated with significant vision loss and other features of uveitis including pain and photophobia when it migrates into the anterior chamber of the eye . Though L . Loa can be seen in the subconjunctival space , anterior chamber and vitreous cavity , it can present with other well known systemic features and can coexist with other parasitic filarial nematodes including; Wuchereria bancrofti and Brugia spp . ( lymphatic filariasis ) , Onchocerca volvulus ( Onchocerciasis ) and Mansonella spp . ( Mansonelliasis ) [9] . In 1890 Stephen Mckenzie , an Ophthalmologist , identified microfilaria . In 1895 a localized angioedema was observed in Calabar , a coastal town in Nigeria , resulting in the name “Calabar swelling” by Douglas Argyll-Robertson who was a Scottish Ophthalmologist . Microfilaria of L . loa is transmitted by Chrysops . The two important vectors are Chrysops silacea and Chrysops dimidiata; they prefer rainforest-like environments and only exist in Africa . They are popularly known as deerflies and mango flies [10] . The vectors are blood sucking and day-biting flies; the female fly requires a blood meal for production of a second batch of eggs . Humans are the primary reservoir for L . loa . Most reported cases of Loiasis are amongst immigrants from Western and Central African countries , and among travelers to this region . L . loa infestation presenting as the "African eye worm" has been well reported as a trans-border disease imported from endemic regions of the world to other non endemic countries . This is seen in several case reports from Europe , Latin America , Asia and the United States of individuals who have imported the parasite by travelling to or living in endemic regions in Central and West Africa [11 , 12 , 13 , 14] . A review of over a hundred infected individuals identified the patients to have made contact with three countries namely Cameroon , Nigeria and Gabon [15] . The disease in a large number of the cases reported , presented as the eye worm , seen in the subconjunctival space . There have been reports of subconjuctival and intraocular presentation of adult L . loa amongst Nigerians living in urban centers . This is in contrast to expectation and earlier reports of L . loa presenting in rural habitats in Nigeria . Earliest report from Nigeria was by Osuntokun et al in 1975 , of the worm in the anterior chamber . Jain et al [16] in 2008; noticed a subconjunctival worm in a 42 year old Nigerian immigrant in Australia , who had no extra-ocular features . Similarly Shah et al in 2010 reported a case of male L . loa in the subconjunctival space of a 21 year old woman who visited Nigeria 6 years earlier [17] . More recent reports in Nigerians have been by Pedro et al [18] , Hassan et al [19] and Omolase et al [20] . Our case series adds to the existing numbers . Loa loa and other filariasis are established diseases seen amongst the villages and rural communities in endemic areas of Nigeria . Several studies have published on the prevalence of this parasitosis in Nigeria [21] . The ( APOC ) program was specifically targeting rural communities in endemic areas of Africa . This intervention was decisive and judged to have achieved its objectives , having a marked effect on microfilaria including L . loa . Though Loiasis has been regarded as a benign disease , a recent publication demonstrated an association between Loiasis and increased mortality amongst a group of rural Africans [22] . It is therefore not as benign as previously thought . It does appear that awareness of the ocular parasitiosis by L . loa has been greatly reduced . This case series therefore seeks to raise the awareness of its presence in urban and semi urban locations . It is likely that the rural urban drift , which is being experienced in several African countries , will result in patients who have been infected with the parasite in earlier ages but now living in urban centers manifesting the ocular parasitosis . This point is illustrated in all 4 patients who are young educated adults with subconjunctival L . loa , having a past history of living in the rural community at some point in their childhood life . They all gave a history of living in rural community in childhood during which they were exposed to or came in contact with vectors harboring filaria . The 2 males gave a history of working in farmlands at the time of dwelling in the rural community while the 2 females gave a history of exposure to streams . In most cases , Loiasis is asymptomatic . There have been reports of initial clinical presentation of loiasis from as soon as 5 months after infection and in some cases after 13 years [23] . The life span of an adult worm is unknown , but in some cases may exceed 17years[23] . This long life span ought to be borne in mind when confronted with a case of Loiasis , as the patient may have forgotten or only vaguely recall the history of exposure to the vector . Loiasis treatment depends on clinical presentation , and is broadly divided into medical using oral drugs including Diethyl Carbamazine ( DEC ) , Albendazole ( ALB ) , Ivermectin ( IVM ) and surgical; involving removal of the adult worm from the tissue including subconjunctival space and anterior chamber . Surgical removal of the adult worms can be attempted when the worm is within view and antimicrobial can then be used to kill microfilaria and any remaining adult worm . The first successful worm extraction was done in 1778 by Francois Guyot a French surgeon , who removed a worm from a man’s eye among the West African slaves on a French ship to America [24] . A visible subconjunctival L . loa requires removal . This can be safely done by incision of the conjunctiva tissue and extraction of the adult subconjunctival worm with a pair of forceps . This appears to be the commonest situation from several case reports . This was also the situation in the first and the fourth cases in this series and can be safely done with the use of topical anesthesia or local infiltration using an injectable anesthetic agent . The integrity of the adult worm should be preserved for histological examination . If the worm is broken during extraction , it may cause severe local inflammation and provoke a Calabar swelling at any anatomic location . In another case scenario , the subconjunctival L . loa may be dead and encysted in the subconjunctival space . This was seen in case three in which the adult L . loa had died in the subconjunctival space because of the prior ingestion of DEC . This presentation has been previously reported by Lichtinger et al [25] and Carme et al [26] and is a less common presentation . This requires more careful dissection and extraction of the encysted worm without damage for histological review . Case two presents the most challenging situation , in which the adult worm is freely and actively mobile and migrates away from subconjunctival visibility at time of planned removal . In this case the patient was requested to adopt a complete head and face down position and observe a period of waiting . The L . loa worm soon migrated back to visibility in the subconjunctival position . Two methods of removal of such actively mobile worm have been described and include in one case the adult mobile worm was fixed with a forceps through the conjunctiva , then conjunctival incision and grasping the worm was done with a second forceps in the other hand [27] . In another case the authors describe two consecutive patients with suspected migratory nematodes who were treated promptly by strategic placement of a pharmacological barrier in the forniceal conjunctiva using 1% lidocaine with epinephrine to block the routes of retreat and to immobilize the worms for controlled retrieval [28] . Due to this barrier , the L . loa could be successfully removed without opportunity to retreat . To conclude , often the presenting complain amongst patients who may eventually be diagnosed as having a subconjunctival worm include a foreign body like sensation , a red eye , itchy eye and a mobile subconjunctival tumor that showed vermiform movements . An un explained complain of such symptoms should raise a clinicians index of suspicion for a subconjunctival worm . Though methods of removal of subconjunctival worm differ , commonly it can be removed with the use of forceps after a conjunctival incision is made . This requires use of topical anesthesia , but other forms of anesthesia such as , local subconjunctival infiltration and retro bulbar anesthesia have been safely and effectively employed . Several reports also exist of removal using a slit lamp bio microscope . In the 4 cases we have reported the removal was done in the operating room . It is likely that with the rural urban drift and as younger population migrate to the urban areas in search of better living standards in Nigeria and across African this presentation will become more commonly seen by ophthalmologists and eye care providers working in the cities and urban dwellings . Eye care practitioners need to be aware of this and should be prepared to safely extract such worms when they migrate within reach in the subconjunctival position . | Subconjunctival adult Loa loa is one of the ocular manifestations of paraisitosis by Loa loa , which is categorized as a neglected benign disease . Loa loa transmission has predominantly been amongst inhabitants in rural communities in Nigeria , one of the countries in the endemic belt of Africa . However we report a case series of 4 patients presenting with subconjunctival Loa loa who in childhood lived in rural communities but later migrated from rural to urban cities where they presented . The varied presentation of the patients in this series and the technique of worm removal are reported . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"and",
"discussion"
]
| [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"pruritus",
"tropical",
"diseases",
"geographical",
"locations",
"social",
"sciences",
"parasitic",
"diseases",
"animals",
"neuroscience",
"nematode",
"infections",
"neglected",
"tropical",
"diseases",
"eyes",
"africa",
"geography",
"loiasis",
"loa",
"loa",
"head",
"nigeria",
"people",
"and",
"places",
"psychology",
"eukaryota",
"anatomy",
"urban",
"areas",
"earth",
"sciences",
"geographic",
"areas",
"nematoda",
"biology",
"and",
"life",
"sciences",
"ocular",
"system",
"sensory",
"perception",
"organisms"
]
| 2018 | Removal of adult subconjunctival Loa loa amongst urban dwellers in Nigeria |
The genetic code relates nucleotide sequence to amino acid sequence and is shared across all organisms , with the rare exceptions of lineages in which one or a few codons have acquired novel assignments . Recoding of UGA from stop to tryptophan has evolved independently in certain reduced bacterial genomes , including those of the mycoplasmas and some mitochondria . Small genomes typically exhibit low guanine plus cytosine ( GC ) content , and this bias in base composition has been proposed to drive UGA Stop to Tryptophan ( Stop→Trp ) recoding . Using a combination of genome sequencing and high-throughput proteomics , we show that an α-Proteobacterial symbiont of cicadas has the unprecedented combination of an extremely small genome ( 144 kb ) , a GC–biased base composition ( 58 . 4% ) , and a coding reassignment of UGA Stop→Trp . Although it is not clear why this tiny genome lacks the low GC content typical of other small bacterial genomes , these observations support a role of genome reduction rather than base composition as a driver of codon reassignment .
The GC content of bacterial genomes has been known to vary widely since at least the 1950s [1] . Currently sequenced genomes range from 17–75% GC and show a strong correlation between genome size and GC content [2]–[4] ( Figure 1 ) . The tiny genomes of symbionts of sap-feeding insects are extreme exemplars of this relationship: Carsonella ruddii [5] , Sulcia muelleri [6] , and Buchnera aphidicola Cc [7] , which represent three independently evolved endosymbiont lineages , have the smallest and most GC-poor genomes yet reported ( Figure 1 ) . These bacteria have a strict intracellular lifestyle , and this shift from a free-living state to an obligate intracellular one greatly reduces the effective population size of the bacteria , in part by exposing them to frequent population bottlenecks as they are maternally transmitted during the insect lifecycle [2] , [3] , [8] . This population structure leads to an increase in genetic drift , and this increase , combined with the constant availability of the rich metabolite pool of the insect host cell , is thought to explain the massive gene loss and high rate of sequence evolution seen in intracellular bacteria [2] , [3] . Sequence evolution is also likely accelerated by an increased mutation rate , stemming from the loss of genes involved in DNA repair during genome reduction [4] . This loss of repair enzymes may contribute to the AT bias of small bacterial genomes since common chemical changes in DNA , cytosine deaminations and guanosine oxidations , both lead to mutations in which an AT pair replaces a GC pair , if left unrepaired [9] , [10] . Indeed , the properties of all symbiont genomes published to date fit well within this framework ( Figure 1 ) . The UGA Stop→Trp recoding , found in the mycoplasmas and several mitochondrial lineages , is associated with both genome reduction and low GC content [11]–[13] . Under the “codon capture” model , a codon falls to low frequency and is then free to be reassigned without major fitness repercussions . Applying this model to the UGA Stop→Trp recoding , mutational bias towards AT causes each UGA to mutate to the synonym UAA without affecting protein length [14] , [15] . When the UGA codon subsequently reappears through mutation , it is then free to code for an amino acid [14] , [15] . While some have argued that codon capture is insufficient to explain many recoding events [11] , [12] , the fact that all known UGA Stop→Trp recodings have taken place in high AT genomes [11] , [16] makes the argument attractive for this recoding . Here we describe the genomic properties of an α-Proteobacterial symbiont ( for which we propose the name Candidatus Hodgkinia cicadicola ) from the cicada Diceroprocta semicincta ( Davis 1928 ) [17] . We show that at only 143 , 795 bps it has the smallest known cellular genome , but has a high GC content of 58 . 4% and a recoding of UGA Stop→Trp . We hypothesize that gene loss associated with genome reduction is a critical step in this recoding , rather than mutational pressure favoring AT . Specifically , we suggest that loss of translational release factor RF2 , which recognizes the UGA stop , was the unifying force driving the recoding in Hodgkinia as well as in certain other small AT-rich genomes .
Previous work revealed that some cicadas had Sulcia as symbionts [18] , but the identity of other symbionts , if any , was unknown . To identify any coexisting symbionts , we amplified and sequenced 16S rRNA genes from cicada bacteriomes ( organs containing symbiotic bacteria ) . A second bacterial type was discovered and found to have large and irregularly shaped cells ( Figure 2 ) . Unusual cell morphologies have been observed in other bacteria with tiny genomes [5] , [18] , suggesting that this symbiont species might also have a small genome . Preliminary analysis using the Naive Bayesian rRNA Classifier [19] at the Ribosomal Database Project website [20] placed the new 16S rDNA sequence in the α-Proteobacteria with 100% confidence and , more specifically , within the Rhizobiales with 86% confidence . Because all other endosymbiotic α-Proteobacteria with small genomes are members of the Rickettsiales ( e . g . Wolbachia , Rickettsia , and Erhlichia ) , we were interested in obtaining genomic data to further characterize this seemingly strange bacterium . Genome sequencing revealed that Hodgkinia had some properties that were similar to other endosymbiont genomes , such as high coding density and shortened open reading frames ( Table 1 ) . But other aspects of the Hodgkinia genome suggested a highly atypical bacterial genome structure . In particular , the genome was only 144 kb , and thus even smaller than other known symbiont genomes , but had an unusually high GC content of about 58% . To our knowledge , this is an unprecedented combination of genome size and base composition ( Figure 1 ) . Additionally , initial rounds of gene prediction revealed that many protein-coding regions were interrupted by putative stop codons . Our previous experience [6] suggested that this could be due to errors in homopolymeric run lengths predicted by Roche/454 sequencing technology . However , the addition of Illumina/Solexa data indicated that the interrupted reading frames were not caused by sequencing errors . We noticed that computational translation of the genome with the NCBI genetic code 4 ( UGA Stop→Trp ) afforded full-length protein sequences , which immediately suggested that Hodgkinia might use an alternative genetic code . Analysis of the gene complement of Hodgkinia revealed that the genome contains a homolog of prfA , encoding translational Release Factor RF1 , which recognizes the stop codons UAA and UAG , but does not contain a homolog of prfB ( RF2 ) , which recognizes UAA and UGA . RF2 is dispensable if UGA is not used as a stop codon , and the loss of RF2 combined with recoding of UGA Stop→Trp is known in Mycoplasma species [13] , [21] , [22] . Additionally , the anticodon of the sole tRNA-Trp gene in Hodgkinia ( trnW ) has mutated from CCA to UCA , which allows recognition of both the normal tryptophan codon ( UGG ) and the putatively recoded UGA stop codon under Crick's wobble rules for codon-anticodon pairing [23] . This tRNA-Trp mutation has also been observed in mitochondrial genomes that have the UGA Stop→Trp recoding [24] . Additionally , it was observed that UGA codons in Hodgkinia open reading frames correspond to the position of conserved tryptophan residues in homologous proteins of other bacteria ( Figure 3 ) . Cumulatively , these data strongly suggested that UGA encodes tryptophan in Hodgkinia . The long branch lengths for the Hodgkinia lineage in both rDNA and protein trees ( Figure 4 , Figure 5 , and Figure S1 ) indicate a fast substitution rate , a situation typical of reduced bacterial genomes . Because the average percent identity of Hodgkinia proteins to their top hits in the GenBank non-redundant database was only 39 . 5% , it was difficult to rule out other recoding events based solely on sequence comparisons . To eliminate the possibility of other such changes in the genetic code , and to experimentally verify the UGA Stop→Trp recoding , shotgun protein sequencing by mass spectrometry [25] was used to sequence peptides derived from cicada bacteriomes . These peptide sequences ruled out any other codon reassignments , and experimentally confirmed the predicted UGA Stop→Trp code change ( Figure 6 and Table S1 ) . Phylogenetic analysis of 16S rDNA sequences , including two newly acquired sequences from symbionts of other cicada species , shows that the cicada symbionts form a highly supported clade that falls within the α-Proteobacteria but outside of the Rickettsiales ( Figure 4 ) . The complete genome allowed additional phylogenetic analysis to further establish the placement of Hodgkinia within the α-Proteobacteria . Phylogenetic trees based on protein sequences ( Figure 5 and Figure S1 ) support the grouping of Hodgkinia in the Rhizobiales , although the support was not always strong and trees made with some individual protein sequences placed it within the Rickettsiales with weak support ( data not shown ) . We therefore looked for additional evidence in the form of gene order to further resolve the placement of Hodgkinia . The “S10” region ( corresponding to the genomic region flanking ribosomal protein rpsJ ) is a highly conserved cluster of genes that shares blocks of gene order conserved between Bacteria and Archaea [26] . The Rickettsiales have gene rearrangements and broken colinearity in this region that are unique within the α-Proteobacteria ( [27] and Figure 7 ) . Hodgkinia does not share these genomic signatures , instead showing perfect colinearity with genomes in the Rhizobiales and Rhodobacteraceae ( Figure 7 ) . These data rule out Hodgkinia's grouping within the Rickettsiales , but do not entirely preclude a common ancestor with them , as Hodgkinia could have diverged from other Rickettsiales before the S10 region rearrangement . The accurate placement of Hodgkinia within the α-Proteobacteria is confounded by both long branch attraction ( LBA ) and large differences in GC contents between different members of the α-Proteobacteria . LBA is expected to incorrectly associate Hodgkinia with the Rickettsiales , since these two lineages have the longest branches on the tree . Therefore , the fact that most analyses place Hodgkinia outside the Rickettsiales is significant . Conversely , the GC content bias is expected to incorrectly group sequences that are similar in GC content but that are not truly related by ancestry , and this artifact might tend to place Hodgkinia outside of the Rickettsiales , since Hodgkinia and most other non-Rickettsial α-Proteobacteria have high GC contents . We therefore tested all possible permutations in the placement of the Hodgkinia clade shown in Figure 4 under a model that does not assume nucleotide composition homogeneity among taxa [28] , [29] . Hodgkinia did not group with the Rickettsiales in any of the highest scoring trees ( Figure 4 ) , suggesting that Hodgkinia's grouping in the Rhizobiales was not a function of GC content bias . Overall , the results from the phylogenetics of proteins and 16S rDNA , as well as from gene order comparisons , strongly argue for the grouping of Hodgkinia with the Rhizobiales .
All previously confirmed UGA Stop→Trp recoding events have occurred in genomes with low GC content: the mitochondria of Metazoa and Fungi , some Protist mitochondria , and certain bacteria in the Firmicutes [11] . ( This same recoding may have occurred in the nuclear genomes of some Ciliates , but information on those genomes is limited [16] ) . Proposed evolutionary mechanisms for genetic code reassignments fall into three groups: the codon capture hypothesis [14] , [15] , involving the extinction and reassignment of codons; the genome reduction hypothesis , under which the pressure to minimize genome content drives the recoding of some codons , reducing the number of tRNAs [30]; and the ambiguous translation hypothesis , under which a single codon is temporarily read in two different ways , with a subsequent loss of the original meaning of the code [12] , [31] . These hypotheses are not mutually exclusive and may apply more to some recoding events than to others [12] . For example , the pioneering ideas of Osawa and Jukes on this topic [14] involved loss of the corresponding tRNA following the extinction of a codon . Also , ambiguous translation , which is known for Bacillus subtilis [32] , could facilitate a transition through the codon extinction route or the genome reduction route . Codon capture requires the changing of one codon to another synonym though an initial codon extinction step potentially resulting from biases in nucleotide base composition . All previously described cases of UGA Stop→Trp recoding occur in GC-poor genomes , and this recoding has been proposed to result from genome-wide replacement of UGA by UAA , due to AT-biased mutational pressure [14] , [15] . Under this explanation , the extinction of UGA Stop allows UGA to later reappear , recoded as an amino acid . Several arguments weigh against the codon capture hypothesis [11] , [12]; most relevant is the fact that , in mitochondrial genomes , there is no association between the codons that undergo a reassignment and those that are expected to potentially disappear due to GC content bias [12] . Tallying stop codons in α-Proteobacteria with complete genomes also weighs against codon extinction as an initial step in this recoding event: although UGA codons are fewest in small and AT-biased genomes , in no case does UGA approach extinction . Among previously sequenced α-Proteobacteria ( excluding Hodgkinia ) , even the smallest and most AT-biased genomes retain over 100 genes using UGA as Stop ( e . g . , there are 137 UGA Stop codons in the 1 . 11 Mb genome of Rickettsia prowazekii , which has a GC content of only 29% ) . In α-Proteobacteria with GC-rich genomes , UGA is the most frequent of the three stop codons and is typically used in a majority of genes ( typically 50%–70% of coding genes end in UGA ) . Thus , the combination of phylogenetic evidence , which places Hodgkinia in the GC-rich Rhizobiales , and UGA usage patterns in extant α-Proteobacteria weigh strongly against UGA extinction as a causal step in the observed recoding . We suggest an alternative hypothesis , implicating genome reduction as the primary driver of the UGA recoding , to explain the coding change observed in Hodgkinia ( Figure 8 ) . As in the ambiguous translation hypothesis , the recoding would first be enabled by the relaxed codon recognition of a mutated tRNA-Trp as promoted by structural changes in the tRNA [31] ( Figure 8 , step 1 ) . For example , point mutations in either the D- or anticodon-arms of tRNA can induce C-A mispairing at the third codon position [33] , [34] . In the presence of such alternative coding , RF2 is no longer essential and thus can be lost through the ongoing process of genome reduction ( step 2 ) . This is similar to the scenario envisioned in the codon capture hypothesis , except that in our case UGA does not need to have gone extinct before RF2 is lost . The further changes observed in Hodgkinia would evolve readily since they involve single base changes driven by positive selection; these include a change in the tRNA-Trp anticodon ( step 3 ) and shifts in stop codon usage ( step 4 ) . Since UGA Stop→Trp has evolved independently in other small genomes such as Mycoplasma and mitochondria , the case of Hodgkinia weighs in favor of genome reduction , and specifically loss of RF2 , as the common force driving UGA Stop→Trp recoding events . Some of the Mollicutes , including Mycoplasma , and certain mitochondrial lineages are the other clear cases of this recoding event , and these genomes also have been characterized by a history of ongoing gene loss [22] . Of course , some small genomes do not show this recoding , and we do not expect the consequences of genome reduction to be predictable in each case . For example , the highly reduced genome of Carsonella ruddii , which retains UGA Stop and RF2 , exhibits an unusual feature of having many overlapping genes with the most common overlap consisting of ATGA , in which ATG is the start of the downstream genes and TGA is the stop of the upstream gene [35] , a situation that might act to conserve UGA Stop and RF2 in the genome . At the initial loss of RF2 , the additional C-terminal length imposed on UGA-ending proteins might be expected to impose some deleterious effects . It is possible that the functionality of proteins with such extensions could be enhanced in Hodgkinia due to an abundance of protein-folding chaperonins , similar to the high levels of GroEL seen in other symbiotic bacteria with small genomes [36] , [37] . Indeed , analysis of the shotgun proteomic data for Hodgkinia shows that homologs of GroEL and DnaK are the two most abundant proteins in the cell ( Table 2 ) . Additionally , the shortened gene lengths observed in Hodgkinia relative to homologs in other genomes ( Table 1 ) indicate that , if UGA-ending proteins were once extended due to recoding , they have since been reduced in length by the generation of new UAG and UAA stop codons . Other models are possible , such as the loss of RF2 effected by a change in the tRNA-Trp anticodon from CCA to UCA instead of distal mutations . Similarly , it is formally possible that Hodgkinia went through a period of AT bias under which the recoding occurred , with a subsequent shift to GC bias as is seen in the present genome . Because phylogenetic evidence favors placement of Hodgkinia's in the Rhizobiales and not within any group characterized by AT rich genomes , we consider this scenario unlikely . Regardless of the recoding mechanism , however , this example provides a rare case in which the loss of an “essential” gene ( RF2 ) in a highly reduced bacterial genome can be compensated by a few simple steps , namely the adaptive fixation of several point mutations . The mechanisms that give rise to GC-content differences in bacterial genomes are unclear , although variations in the replication and/or repair pathways are often suggested as candidates [38]–[40] . Various lines of evidence support this idea , including a correlation between genome GC content and the types of DNA polymerase III , α subunit ( DnaE ) encoded in a genome [41] and the discovery of point mutations affecting the repair enzyme MutT that can detectably change the GC content of Escherichia coli [38] . One mechanistic clue is the correlation between genome size and GC content , a universal pattern in previously studied bacterial and archaeal genomes ( Figure 1 ) . Until now , this tendency has been especially pronounced in obligate intracellular bacterial genomes . Two ( not necessarily mutually exclusive ) hypotheses have been forwarded to explain this base composition bias in genomes of intracellular organisms . The first is an adaptive argument , based on selection for energy constraints [42]: synthesis of GTP and CTP require more metabolic energy , and ATP is the most common nucleotide in the cell because of its ubiquitous role in cellular processes . Therefore , competition for scarce metabolic resources has been hypothesized to force intracellular genomes to low GC values . The second hypothesis relates to mutational pressure resulting from altered capacity for DNA repair [43] . Small intracellular genomes typically lose many repair genes , and these organisms therefore are expected to be deficient in their ability to repair damage caused by spontaneous chemical changes . This is particularly expected in organisms such as endosymbionts in which genetic drift plays a major role in sequence evolution [43] . Indeed , recent experiments in Salmonella strongly support this hypothesis [44] . Our results weigh against the energetic hypothesis because Sulcia , living in the same bacteriome and presumably exposed to the same metabolite pool , has a GC content of 22 . 6% ( J . P . M , B . R . M , and N . A . M . , unpublished data ) , almost identical to the GC content of 22 . 4% for the previously published Sulcia genome from Glassy-winged sharpshooter [6] . One would expect that if the metabolite pool caused an increase in GC content in Hodgkinia , the same trend would be observed in Sulcia . Additionally , the GC content of the third position in 4-fold degenerate sites ( which should be under little or no selective pressure ) in the Hodgkinia genome is 62 . 5% ( Table S2 ) , consistent with mutational pressure as a cause of elevated genomic GC content . Collectively , these data suggest that the replicative process or mutagenic environment of Hodgkinia differ from those of other small-genome α-Proteobacteria and other small genome insect symbionts . Hodgkinia has only two genes involved in replication ( dnaE , DNA polymerase III , α subunit; and dnaQ , DNA polymerase III , ε subunit ) , implicating them as primary targets for future study of the source of GC bias . Regardless of the mechanisms involved in shifting genomic GC contents , our results indicate that low GC content is not an inevitable consequence of loss of repair enzymes , since Hodgkinia has no detectable repair enzymes ( and is thus more extreme in this regard than previously sequenced symbiont genomes , which show partial loss of repair enzymes ) . Our finding that two other cicada species contained symbionts belonging to the same clade , based on 16S rDNA genes ( Figure 4 ) suggests that this symbiont infected an ancestor of cicadas and subsequently has been transmitted maternally , a typical history for bacteriome-dwelling insect symbionts [45] , [46] . In such cases , the symbiont is restricted to its particular group of insect hosts , and restriction to cicada hosts is highly likely for this case . We propose the candidate name Candidatus Hodgkinia cicadicola for this α-Proteobacterial symbiont of cicadas , with the genus name referring to the biochemist Dorothy Crowfoot Hodgkin ( 1910–1994 ) , and the species name referring to presence only in cicadas . Distinctive features include restriction to cicada bacteriomes , large tube-shaped cells , a high genomic GC content , a recoding of UGA Stop→Trp , and the unique 16S rDNA sequence ACGAGGGGAGCGAGTGTTGTTCG ( positions 535–557 , E . coli numbering ) .
Female cicadas were collected in and around Tucson , Arizona , USA . Tissue for genome sequencing was prepared from bacteriomes dissected in 95% ethanol and cleaned up in Qiagen's DNeasy Blood and Tissue Kit . DNA was prepared for the Roche/454 GS FLX pyrosequencer [47] following the manufacturer's protocols . Sequencing generated 523 , 979 reads totalling 116 , 176 , 938 bases , and these were assembled using the GS De novo Assembler ( version 1 . 1 . 03 ) into 1029 contigs . Contigs expected to be from the Hodgkinia genome were identified by BLASTX [48] against the GenBank non-redundant database and the associated reads were extracted and reassembled to construct the Hodgkinia genome . Eleven contigs with an average depth of 73× were generated representing 143 , 582 nts of sequence with an average GC content of 58 . 4% . The order and orientation of the 11 contigs were predicted using the “ . fm” and “ . to” information appended to read names encoded in the 454Contigs . ace file and these joins were confirmed by PCR and Sanger sequencing . Illumina/Solexa sequencing [49] generated 12 , 965 , 640 reads totalling 505 , 659 , 960 nts . These data were mapped to the Hodgkinia genome using MUMMER [50] ( nucmer -b 10 -c 30 -g 2 -l 12; show-snps -rT -×30 ) to an average depth of 43× . Forty-five homopolymeric nucleotide runs were adjusted in length based on the Illumina data . Annotation was carried out as described previously [6] , except that NCBI genetic code 4 ( TGA encoding tryptophan ) was used to computationally translate the predicted protein-coding genes . The Candidatus Hodgkinia cicadicola genome has been deposited in the GenBank database with accession number CP001226 . D . semicincta bacteriomes were dissected in PBS and gently disrupted with a mortar and pestle . Cells were fixed as described [51] and imaged on a Zeiss 510 Meta microscope . The probe sequences were Cy3-CCAATGTGGGGGWACGC ( Sulcia ) and Cy5-CCAATGTGGCTGACCGT ( Hodgkinia ) . The scale bar in Figure 2 generated by the microscope software was overlaid with a plain white bar for legibility . The PCR conditions used to amplify Magicicada cassini ( Brood XIII , Chicago , Illinois ) and Diceroprocta swalei ( Tucson , Arizona ) 16S rDNA were 94°C for 30 seconds followed by 35 cycles of 94°C 15 seconds , 58°C 30 seconds , 72°C 2 minutes using the primers 10F_ALPHA ( AGTTTGATCCTGGCTCAGAACG ) and 1507R ( TACCTTGTTACGACTTCACCCCAG ) . Amplicons were cloned into Invitrogen's TOPO TA PCR2 . 1 kit and sequenced . The D . swalei and M . cassini 16S rDNA sequences have been deposited in the GenBank database with accession numbers FJ361199 and FJ361200 , respectively . The initial assignment of the Hodgkinia 16S rRNA sequence was based on the Naive Bayesian classifier [19] at the Ribosomal Database Project ( RDP ) [20]; this uses a bootstrapping procedure involving resampling of sequence fragments with replacement and assignment of individual fragments to taxonomic units represented in this large database . The three Hodgkinia 16S rDNA sequences , sampled from bacteriomes of D . semicincta and two additional cicada species ( M . cassini and D . swalei ) , were aligned to the Bacterial 16S rDNA model at the RDP , and the remaining sequences used in the generation of Figure 4 were also obtained from the RDP . The maximum likelihood tree in Figure 4 was generated using RAxML [52] under the GTRGAMMA model of sequence evolution . The clade consisting of the Hodgkinia sequences was moved to all other possible positions on the tree in Mesquite [53] , and the log likelihood of each of these trees was estimated using the non-homogenous model implemented in nhPhyML [29] under a 4 category discrete gamma model using the shape parameter estimated from PUZZLE [54] . The protein sequence used in generating Figure 5 was DnaE ( DNA polymerase III , α subunit ) , and the proteins used in generating Figure S1 were DnaE , InfB ( translational initiation factor IF2 ) , TufA ( translational elongation factor Tu ) , RpoB ( RNA polymerase , β subunit ) , and RpoC ( RNA polymerase , β′ subunit ) . Individual alignments for each gene were generated using the linsi module of MAFFT [55] and ( in the 5-protein alignment ) concatenated . Columns containing gap characters were removed , leaving 861 columns in the DnaE alignment and 4152 columns in the 5-protein alignment . Parameters for a 1 invariant/4 Gamma distributed rate heterogeneity model were estimated using PUZZLE , and maximum likelihood trees were computed with PROML from the PHYLIP package [56] using the JTT model of sequence evolution . One hundred bootstrap datasets were generated using SEQBOOT from PHYLIP , trees were calculated as above , and bootstrap values for these trees were mapped back on the maximum likelihood tree calculated from PROML using RAxML . The family and order names and groupings on Figure 4 , Figure 5 , and Figure S1 were taken from [57] and the RDP website [20] . The genomes used in the phylogenetic analysis were ( the accession numbers noted with asterisks were used in generating Figure 7 ) : Zymomonas mobilis subsp . mobilis ZM4 ( NC_006526 ) , Erythrobacter litoralis HTCC2594 ( NC_007722 ) , Novosphingobium aromaticivorans DSM 12444 ( NC_007794 ) , Sphingopyxis alaskensis RB2256 ( NC_008048 ) , Candidatus Pelagibacter ubique HTCC1062 ( NC_007205 ) , Rickettsia rickettsii str . Iowa ( NC_010263 ) , Rickettsia typhi str . Wilmington ( NC_006142* ) , Neorickettsia sennetsu str . Miyayama ( NC_007798 ) , Wolbachia pipientis ( NC_010981 ) , Wolbachia endosymbiont of Drosophila melanogaster ( NC_002978 ) , Anaplasma phagocytophilum HZ ( NC_007797 ) , Anaplasma marginale str . St . Maries ( NC_004842 ) , Ehrlichia ruminantium str . Gardel ( NC_006831 ) , Ehrlichia canis str . Jake ( NC_007354 ) , Rhodospirillum rubrum ATCC 11170 ( NC_007643 ) , Magnetospirillum magneticum AMB-1 ( NC_007626 ) , Acidiphilium cryptum JF-5 ( NC_009484 ) , Gluconobacter oxydans 621H ( NC_006677 ) , Gluconacetobacter diazotrophicus PAl 5 ( NC_010125 ) , Paracoccus denitrificans PD1222 ( NC_008686/NC_008687 ) , Rhodobacter sphaeroides ATCC 17025 ( NC_009428* ) , Jannaschia sp . CCS1 ( NC_007802 ) , Silicibacter pomeroyi DSS-3 ( NC_003911 ) , Silicibacter sp . TM1040 ( NC_008044 ) , Roseobacter denitrificans OCh 114 ( NC_008209 ) , Caulobacter crescentus CB15 ( NC_002696 ) , Caulobacter sp . K31 ( NC_010338 ) , Maricaulis maris MCS10 ( NC_008347 ) , Brucella melitensis 16M ( NC_003317/NC_003318 ) , Brucella abortus S19 ( NC_010742/NC_010740 ) , Bartonella bacilliformis KC583 ( NC_008783 ) , Bartonella henselae str . Houston-1 ( NC_005956 ) , Mesorhizobium loti MAFF303099 ( NC_002678 ) , Mesorhizobium sp . BNC1 ( NC_008254 ) , Agrobacterium tumefaciens str . C58 ( NC_003062/NC_003062 ) , Rhizobium etli CFN 42 ( NC_007761 ) , Sinorhizobium medicae WSM419 ( NC_009636 ) , Sinorhizobium meliloti 1021 ( NC_003047* ) , Rhodopseudomonas palustris BisA53 ( NC_008435 ) , Rhodopseudomonas palustris BisB18 ( NC_007925 ) , Bradyrhizobium japonicum USDA 110 ( NC_004463 ) , Bradyrhizobium sp . BTAi1 ( NC_009485 ) , Nitrobacter hamburgensis X14 ( NC_007964 ) , Nitrobacter winogradskyi Nb-255 ( NC_007406 ) , Xanthobacter autotrophicus Py2 ( NC_009720 ) , and Escherichia coli str . K12 substr . MG1655 ( NC_000913 ) . Total protein was prepared from the bacteriomes of 10 female D . semicincta by homogenizing in 4 ml Buffer H ( 2% SDS , 100 mM Tris , 2% β-mercaptoethanol , pH 7 . 5 ) followed by centrifugation at 100 , 000×g for 30 min . The supernatant was recovered and precipitated in 12% TCA followed by 3 washes in cold acetone . The resulting protein pellet was resuspended in 150 µl sample loading buffer , and 30 µl ( ∼60 µg ) of this sample was loaded onto a well of a 11 cm×8 cm×1 . 5 mm 10% acryl amide gel . Electrophoresis was performed in a mini cell ( Bio-Rad ) at 130 V . The entire lane was cut into 12 sections , and proteins in each section were identified by LC-MS/MS analysis . The gel bands were washed , homogenized , reduced , alkylated and subjected to overnight in-gel tryptic digests . The peptide mixture was extracted , dried in speed-vac and dissolved in a 15 µl of 5% formic acid . The LC-MS/MS experiments were performed on a Q-TOF 2 mass spectrometer equipped with the CapLC system ( Waters Corp . , Milford , MA ) . The stream select module was configured with a 180 µm ID×50 mm trap column packed in-house with 10 µm R2 resin ( Applied Biosystems , Foster City , CA ) connected in series with a 100 µm ID×150 mm capillary column packed with 5 µm C18 particles ( Michrom Bioresources , Auburn , CA ) using a pressure cell . Peptide mixtures ( 10 µl ) were injected onto the trap column at 9 µl/min and desalted for 6 min before being flushed to the capillary column . The peptides were then eluted from the column by the application of a series of mobile phase B gradients ( 5 to 10% B in 4 min , 10 to 30% B in 61 min , 30 to 85% B in 5 min , 85% B for 5 min ) . The final flow rate was 250 nl/min . Mobile phase A consisted of 0 . 1% formic acid , 3% acetonitrile and 0 . 01% TFA , whereas mobile phase B consisted of 0 . 075% formic acid , 0 . 0075% TFA in an 85/10/5 acetonitrile/isopropanol/water solution . The mass spectrometer was operated in a data dependent acquisition mode whereby , following the interrogation of MS data , ions were selected for MS/MS analysis based on their intensity and charge state +2 , +3 , and +4 . Collision energies were chosen automatically based on the m/z and charge-state of the selected precursor ions . The MS survey was from m/z 400–1600 with an acquisition time of 1 sec whereas the trigged data-dependent MS/MS fragmentation scan was from m/z 100–2000 with an acquisition time of 2 . 4 sec . The peak list was created using the Mascot distiller 2 . 2 software from Matrix Science ( London , UK ) using the default settings for Waters . The Mascot 2 . 2 search engine was used to assist in the search of the combined tandem mass spectra against a custom protein database . The custom protein database consisted of the Hodgkinia proteome , the nearly complete proteome of Sulcia muelleri from Diceroprocta semicincta ( J . P . M . , B . R . M . , and N . A . M . , unpublished ) , and the complete proteome from the pea aphid Acyrthosiphon pisum ( build 1 . 1 ) , the most closely related insect to D . semicincta for which a complete genome is available . The database contained 5 , 508 , 819 amino acids residues in 10 , 887 protein sequences . The parameters used for the searches were as follows: trypsin-specificity restriction with 2 missing cleavage site and variable modifications including oxidation ( M ) , deamidation ( N , Q ) , and alkylation ( C ) . Both MS and MS/MS mass tolerance was set to 0 . 3 Da for the searching . The Mascot significance threshold was set to 0 . 05 , using MudPIT scoring , with a Mowse ion score cutoff of >31 ( the cutoff for a peptide suggesting identity or extensive homology ) . The sequences in the custom proteome database were reversed to generate a decoy database for calculation of a false discovery rate , which was 2 . 6% ( 15 peptides found in the decoy database vs . 576 peptides found in the real database ) . For a peptide to be considered in the calculation of codon coverage ( Figure 6 ) , it had to originate from a protein with at least one other high-quality matching peptide . Eighty-seven ( 87 ) such peptides from 16 Hodgkinia proteins were found ( Table S1 ) . These peptides cover all 62 non-stop codons at least once; the peptides LIWPSAVLQAEEVWAGAR from HCDSEM_044 and VSCLIWTDINR from HisA span recoded UGA codons . | The genetic code , which relates DNA sequence to protein sequence , is nearly universal across all life . Examples of recodings do exist , but new instances are rare . Genomes that exhibit recodings typically have other extreme properties , including reduced size , reduced gene sets , and low guanine plus cytosine ( GC ) content . The most common recoding event , the reassignment of UGA to Tryptophan instead of Stop ( Stop→Trp ) , was previously known from several mitochondrial and one bacterial lineage , and it was proposed to be driven by extinction of the UGA codon due to reduction in GC content . Here we present an unusual bacterial genome from a symbiont of cicadas . It exhibits the UGA Stop→Trp reassignment , but has a high GC content , showing that reduction in GC content is not a necessary condition for this recoding . This symbiont genome is also the smallest known for any cellular organism . We therefore propose gene loss during genome reduction as the common force driving this code change in bacteria and organelles . Additionally , the extremely small size of the genome further obscures the once-clear distinction between organelle and autonomous bacterial life . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"genetics",
"and",
"genomics/microbial",
"evolution",
"and",
"genomics",
"evolutionary",
"biology/microbial",
"evolution",
"and",
"genomics",
"microbiology/microbial",
"evolution",
"and",
"genomics"
]
| 2009 | Origin of an Alternative Genetic Code in the Extremely Small and GC–Rich Genome of a Bacterial Symbiont |
Despite the increased use of vaccination in several Asian countries , Japanese Encephalitis ( JE ) remains the most important cause of viral encephalitis in Asia in humans with an estimated 68 , 000 cases annually . Considered a rural disease occurring mainly in paddy-field dominated landscapes where pigs are amplifying hosts , JE may nevertheless circulate in a wider range of environment given the diversity of its potential hosts and vectors . The main objective of this study was to assess the intensity of JE transmission to pigs in a peri-urban environment in the outskirt of Phnom Penh , Cambodia . We estimated the force of JE infection in two cohorts of 15 sentinel pigs by fitting a generalised linear model on seroprevalence monitoring data observed during two four-month periods in 2014 . Our results provide evidence for intensive circulation of JE virus in a periurban area near Phnom Penh , the capital and most populated city of Cambodia . Understanding JE virus transmission in different environments is important for planning JE virus control in the long term and is also an interesting model to study the complexity of vector-borne diseases . Collecting quantitative data such as the force of infection will help calibrate epidemiological model that can be used to better understand complex vector-borne disease epidemiological cycles .
Despite the increased use of vaccination in several Asian countries , Japanese Encephalitis ( JE ) remains the most important cause of viral encephalitis in Asia in humans [1–3] . A recent review based on updated incidence data estimated that 68 , 000 JE cases occurred annually in the 24 JE-endemic countries , for an estimated incidence of 1 . 8 case per 100°000 people overall [1] . Half of these cases occur in China where expanding vaccination programs should dramatically decrease the incidence of JE in the future . One-fifth occur in areas with no or minimal JE vaccination programme such as Cambodia [1] . Cambodia is a JE high-incidence country with a nascent vaccination programme that should develop into a national program in the coming years [4] . A sentinel surveillance study on Japanese encephalitis in six Cambodian hospitals estimated the clinically-declared JE incidence in 2007 in the country at 11 . 1 cases per 100 000 children under 15 years of age [4] . The epidemiological cycle of JE is complex with different potential host and vector species . JE is considered a predominantly rural zoonosis with a wild cycle involving aquatic birds and Culex mosquitoes and a domestic cycle where pigs are amplifier hosts [5 , 6] . This classical description of JE in which wild ardeids are considered the main reservoir of JE dates back to the 1950s and the first extensive studies of JE epidemiology in Japan [7] . The proximity to irrigated land and in particular paddy fields where JE vectors can breed and the presence of pigs , typical features of most rural areas in Cambodia and other East and South-East Asian countries , have been identified as JE risk factors [8–11] . Several Culex species have been identified as potential JE vectors [5] . The main vectors such as Culex tritaeniorhynchus breed mostly in rural settings , however , other species like Culex quinquefasciatus , an anthropophilic species , could play an a role in JE transmission in periurban or urban areas [12 , 13] . Beyond the aquatic wild birds traditionally suspected to be the main reservoir [5 , 14] , several host species are also thought to be able to play a role in the transmission of the virus such as poultry or non-aquatic wild birds such as passerine birds that experimentally show sufficient viremia to allow virus transmission [15–17] . This means that JE could be transmitted and even maintained in a wide range of environments beyond the typical rural , paddy-fields dominated landscape . JE epidemiology should be rethought depending on the different environments and hosts [17] . With JE expanding [18 , 19] , it is important to understand the range of eco-epidemiological systems in which it could be maintained and transmitted to humans , especially in peri-urban or even urban areas where a growing part of the world population is living . This peri-urban and urban circulation has been observed in Southeast Asia where peri-urban human JE cases have been observed in Bangkok , Thailand and Can Tho , Vietnam [3 , 20] . Similarly , JE may still be actively transmitted in the peripheral part of the Singapore island despite abolishment of pig farming [21] . The main objective of this study was to assess the intensity of JEV transmission to pigs in a periurban environment in the outskirt of Phnom Penh , Cambodia . Specifically , we estimated the force of JEV infection in two cohorts of sentinel pigs by detecting anti-JEV IgG antibodies from the end of the hot dry season to the beginning of the rainy season ( from April to July ) and subsequently from the peak of the rainy season to the beginning of the cool dry season ( from September to January ) . Concomitantly , we captured mosquitoes and tested them for JEV by qRT-PCR during the pig sampling periods to infer JEV potential vector species in this area .
Of 29 pigs that remained in the study ( Pig A03 died of unknown cause during the study , its last blood sample tested negative for JE antibodies ) , 28 seroconverted during the study period ( Fig 1 ) . Test results of the last collected serum were equivocal in the 29th pig ( A09 ) ( Fig 1 ) . Some pigs still had maternal antibodies at the age of two months , but all of them had become seronegative by the age of three months before rapidly seroconverting again between the age of three months and six months during both study periods ( Fig 2 ) . All seroconverted pigs were confirmed positive to JE by SNT . No clinical signs were recorded in any of the pigs during the study ( except for pig A03 ) . qRT-PCR-screened blood samples taken before seroconversion ( n = 106 ) were negative in all but one pig ( pig B14 on 29/09/2014 ( fourth blood sample ) ( Fig 1 ) . The virus isolation attempt was not successful after three passages . The sample was confirmed positive by amplicon sequencing of NS3 gene . For the first cohort , we set May 6 , 2014 ( date of the third blood sample ) as the starting date with ten susceptible pigs ( Pig A03 was removed from the study as it died before seroconverting ) . The model estimated a FOI of 0 . 03192/day ( sd = 0 . 005622/days ) , meaning that during that period , a susceptible pig had a 3 . 19% probability of acquiring JEV infection each day . For the second cohort , we set September 12 , 2014 ( date of the second blood sample ) as the starting date with ten susceptible pigs . The model estimated a FOI of 0 . 04637/days ( sd = 0 . 007973/day ) , meaning that a susceptible pig had a daily probability of 4 . 64% of acquiring JEV infection during that period . Fig 3 shows the fitted model over our data . A total of 11 , 078 mosquitoes were captured , 6 , 692 during the 11 capture sessions between April and July and 4 , 386 during the 14 capture sessions between September and January . Table 1 shows a summary of the mosquito species captured during the study , detailed results are available in S1 Table . Culex tritaeniorhynchus was the most abundant species with around 2/3 of the mosquitoes captured during both study periods , followed by Culex gelidus in April-July and Culex vishnui in September-January ( Table 1 ) . Around 1% of the mosquitoes captured were Culex quinquefasciatus . The number of mosquitoes captured varied greatly during the study with apparent peak of mosquito’s abundance in May , July and December ( Fig 4 ) . A total of 1 , 171 pools were screened for JEV using qRT-PCR . Only 1 pool of Culex tritaeniorhynchus , captured on 12/09/2014 was found positive , i . e . a minimum infection rate ( MIR ) of 11 . 9/ 1 , 000 for Culex tritaeniorhynchus females for this night of capture , a MIR of 0 . 13/ 1 , 000 for Culex tritaeniorhynchus females over the whole study and MIR of 0 . 091/ 1 , 000 for females from all species over the whole study .
Our results provide evidence for intensive circulation of JEV in a periurban area near Phnom Penh , the capital and most populated city of Cambodia . Among 29 pigs , 28 ( 96 . 6% ) had seroconverted before the age of six months , and the last serological result of the 29th pig was equivocal , suggesting that it was seroconverting as most other pigs had a serum tested equivocal prior to the seroconversion . This is in line with results observed in rural Cambodia where 95 . 2% of the pigs older than 6 months were tested seropositive for JEV by IgG ELISA and hemagglutination inhibition tests [22] . This suggests that JEV circulation in periurban areas near Phnom Penh may be as intensive as in rural Cambodia . This intensive JEV circulation in pigs was also observed in a urban environment in the city of Can Tho in Vietnam [20] , pointing out the importance of taking into account the risk of JEV transmission in urban and periurban areas , and not in the typical rural environment only . Populations living in such periurban areas are growing and they should be enrolled in national JEV control programmes . Our protocol allowed us to estimate the FOI of JEV in the sentinel pig population . Results from the two cohorts were close and may suggest a lack of seasonality in transmission but our data are not sufficient to support this point . Indeed , it was not possible to use a simple modified Welch t-test to compare the two FOI because they were estimated using a maximum likelihood approach and the distribution of such estimators can only be approximated to normality when the information about the studied population is almost exhaustive , i . e . for large sample sizes [23] . Furthermore , the relatively large standard deviations associated with the FOI ( coefficients of variation of 17 . 6% and 17 . 2% ) are likely due to our limited sample size and would prevent any interpretation of a non-significant difference between the FOI estimated for the two cohorts . This is , to our knowledge , the first estimation in East and Southeast Asia of the force of JEV infection in pigs . It was estimated in Bangladesh at 20% per year [24] , which is considerably lower than the FOI estimated in our periurban study area ( 3–5% per day ) . This difference may translate a different combination of hosts , vectors and agricultural practices in the two areas , pointing out the importance of taking into account these parameters when planning control programs [17] . In the absence of wild waterbirds in periurban and urban areas , domestic and peridomestic species such as passerine birds may play a role in the transmission and the maintenance of the virus as suggested by the JEV viremia experimentally observed in poultry [15] and in several native North American passerine species [16] . In terms of vectors , Culex tritaeniorhynchus—a species mostly considered rural because it breeds in fresh water such as flooded paddy fields—was the most abundant mosquito species captured , while Culex quinquefasciatus , a “domestic” species , accounted for around 1% only of the total number of mosquitoes captured . As our study was set in a periurban area we could have expected a more balanced proportion of rural and domestic vectors . The location of our traps , close to the pigs , may have influenced our results and led to capture more Culex tritaeniorhynchus , a predominantly zoophilic species , than Culex quinquefasciatus , a more anthropopihlic species . These results suggest that Culex tritaeniorhynchus can be abundant at least in small parts of periurban areas and play a major role in the intensive circulation of JEV , as observed in the urban area of Can Tho city in Vietnam [20] . Intensive circulation of JEV in other urban and periurban areas may then be dependent on the presence of a vector as competent as Culex tritaeniorhynchus . Countries beyond JEV geographical distribution , where Culex tritaeniorhynchus is present , should then implement JEV surveillance in both rural and peri-urban areas [25–27] . Despite an intensive circulation of JEV detected in pigs , only one pool of Culex tritaeniorhynchus tested positive for JEV by qRT-PCR . This low detection rate of JEV in mosquitoes may be related to an actual low infection rate in mosquitoes , as observed for other vector-borne diseases including other flavivirus closely related to JEV such as West Nile virus [28 , 29] , and/or to the dilution effect resulting from pooling the mosquitoes before testing with a molecular method that has its own limit of detection . The MIR of 0 . 091/ 1 , 000 for females from all species over our whole study is low compared to MIR in the range of 1–1 . 2/ 1 , 000 for a JEV study in Can Tho city [20] and two West Nile studies in Florida and Puerto Rico [28 , 29] but is similar to previous studies on JEV in rural areas of Can Tho province ( MIR of 0 . 05 / 1 , 000 ) and in suburban Bangkok ( MIR of 0 . 046 / 1 , 000 ) [30 , 31] . An actual low infection rate of JEV in mosquitoes despite an intensive circulation in pigs could be consistent with the existence of a direct transmission of JEV between pigs as suggested by the results of a recent experimental study showing that oro-nasal virus excretion could last -5-6 days in pigs [32] . The FOI we are estimating in this study would then result from a combination of a vector-borne and a “within-pen” direct transmission . Based on the same set of data , we are currently developing dynamic models to quantify the relative importance of the different transmission routes . Control of JEV in humans has successfully been implemented in several Asian countries over the past decades by introducing vaccination [33] . Mass vaccination campaigns have dramatically decreased the number of clinical acute encephalitis in countries like Japan and South Korea after their introduction [33 , 34] . Since humans are dead-end hosts due to very low viremia , their vaccination does not disrupt the transmission of JEV , and given the complexity of JEV epidemiological cycle , eradicating the disease does not seem realistic . However , other control measures can be combined with human vaccination ( or when vaccination is not available ) to protect humans , which may be especially important if vaccination become less efficient in the future against emergent genotypes [35] . Vaccination can also protect pigs from abortions or orchitis . In Cambodia and in endemic areas with an intensive JEV circulation in general , JEV has little to no impact on pig production since most pigs get infected prior to reaching sexual maturity , as observed in our study or in South Vietnam [20] . But in epidemic areas such as North Vietnam or China , JE is an animal disease and control measures such as vaccination of reproductive pigs can be used . Estimating key parameters such as the force of infection to calibrate models of JEV transmission may then be used to test different measures ( i . e . pig vaccination , banning pig farming near populated areas , rice flooding management ) and optimize JE control according to the local situation in humans and animals . The impact of pig vaccination was for example predicted as an interesting JE control tool in Bangladesh for both animals and humans [24] . Beyond estimating key transmission parameters , the surveillance of JEV with sentinel pigs could also be used to detect JEV emergence . Historically , JEV spread geographically in Asia from the Indonesia-Malaysia region [18 , 36] with a recent emergence in Australia [37] . It may potentially emerge in a diversity of ecosystems including Africa or Europe . As a matter of fact , JEV-RNA like sequences were detected in Italy and Culex tritaeniorhynchus established in Greek paddy fields [26 , 38] . Confirming JEV infection in humans is challenging: direct detection methods such as viral isolation or qRT-PCR have low sensitivity because of transient , early viraemia and diagnosis of JEV infection by IgM detection might be misled by antigenic cross-reaction and by actual secretion of anti-JEV IgM during another neurological infection in patients previously immunised against JEV [39] . Detecting JEV emergence may then be easier in pigs . A first step for detecting JEV emergence could be a routine serological surveillance of pig populations at slaughterhouses in risky areas , followed by the implementation of sentinel pig surveillance in case of positive results . This would help to confirm the emergence , to characterise the virus in pigs and vectors and to quantify the transmission in the emergence area . More generally , with several species of mosquitoes—mostly from the Culex genus , —known as JEV vectors [5] and a large diversity of potential hosts , understanding JEV transmission in different environments is important for planning JEV control in the long term and is also an interesting model to study the complexity of vector-borne diseases . Measuring quantitative data such as the force of infection will help calibrate epidemiological model that can be used to better understand complex vector-borne disease epidemiological cycles and test different strategies of control .
During this study , we followed the World Animal Health Organisation ( OIE ) guiding principles on animal welfare included in the OIE terrestrial Code , Chapter 7 . 8 “Use of Animals in research and education” [40] . In particular , intervals between sampling sessions were 10 days to limit the stress resulting from handling and sampling . At the beginning of the study the pigs were not sampled for two weeks in order to let them acclimate to their new environment . They were separated in three groups of five individuals in separated pens . The study was set in the city of Ta Khmau , in a periurban area located 10 km from the center of Phnom Penh ( 11 . 4739°N , 104 . 9376°E ) , at the interface between a densely populated urban area and a rural landscape dominated by cultivated areas ( Fig 5 ) . Two cohorts of 15 pigs were successively monitored from April to July 2014 and from September 2014 to January 2015 . Pigs were bought at the age of six weeks and kept in a backyard where no domestic animals were usually raised . The pigs were individually identified with ear tags . Blood samples were collected every ten days on every pig from the age of 2 months , when maternal immunity is waning , to the age of six months when pigs are usually sent to the slaughterhouse . Pigs of the first cohort , tagged A01 to A15 were born on February 10 , 2014 , and sampled 11 times from April 9 to July 29 , 2014 . Pigs of the second cohort , tagged B01 to B15 were born on July 7 , 2014 , and sampled 14 times from September the 2nd to January 12 , 2015 . Sera were tested for JEV IgG using an ELISA test adapted from Dong Kun Yang et al [42] . The value of the background absorbance was subtracted from the signal value of all the test reading . The cut off value was calculated based on the mean of the three negative controls ( NC ) : positive if the sample optical density ( OD ) > 4x mean NC , negative if the sample OD < 3 mean NC , and equivocal if the sample OD is within 3x mean NC and 4x mean NC . At the end of the study , the last serum sample of each pig underwent serum neutralization testing ( SNT ) for JEV [43] . BHK-21 cells ( ATCC , CCL-10 ) were initially inoculated at 1 × 106 cells/well in six-well tissue culture plates and propagated for 24 hours at 37°C in a CO2 incubator . Serum samples were inactivated for 30 minutes in a 56°C water-bath and serially diluted ten-fold from 1:10 to 1:1000 in Dulbecco's Modified Eagle Medium ( DMEM ) containing 10% fetal bovine serum ( FBS ) . A 100-μL aliquot of JEV ( JEV SA 14-14-2 ) with 60 plaque-forming units ( pfu ) was mixed with equal volumes of diluted serum samples and incubated for 1 hour at 37°C . Each virus/serum mixture ( total volume 200 μL ) was inoculated onto the BHK-21 cell monolayer after draining the culture medium and was allowed to settle for 1 hour at 37°C in a CO2 incubator . The mixture was removed from the cell monolayer and each well washed once with phosphate-buffered saline ( PBS ) . Then 4 mL of pre-warmed overlay medium consisting of 3% Carboxymethyl cellulose ( Sigma , Cat . C4888 ) and 0 . 9% NaCl ( Sigma , Cat . S6191 ) and 3% FBS in DMEM were poured onto each well . The plates were placed in a CO2 incubator and the overlay medium was removed five days after inoculation . Each well was carefully washed two time with PBS and was stained with 0 . 1% Naphthol Blue Black ( Sigma , Cat . N3393 ) , 25% Isopropanol ( Sigma , Cat . I9516 ) and 10% Acetic acide ( Sigma , Cat . 320099 ) for 30 minutes . Plate wells were slowly washed and dried and the plaques were counted . The neutralizing antibody titer ( PRNT50 ) was defined as the reciprocal of the last serum dilution that showed 50% or more plaque reduction compared with the plaque counts in the virus-only control well . PRNT50 titres ≥1:20 were considered positive . Because IgG antibodies can start to be detected up to several weeks after the infection in pigs [44] , serum samples collected within two weeks prior to the presumed seroconversion date of each pig were tested by quantitative reverse transcriptase polymerase chain reaction ( qRT-PCR ) to detect JEV RNA [45] . The limit of detection of the assay was 10 copy of RNA/reaction . Conventional PCR using primers targeting NS3 region ( S2 Table ) was used on positive samples by qRT-PCR and the PCR products were sequenced for confirmation [46] . One home-made CDC light-trap was placed in the pig open-building during the night preceding each blood sampling of the two cohorts [47] . Mosquitoes captured were identified using a Southeast Asia identification key [48 , 49] . Mosquitoes were counted and pooled by groups of ten individuals of the same species and the same night of capture and subsequently screened for JEV by qRT-PCR . The monitoring of the serological status of the sentinel pigs enabled us to estimate the Force Of Infection ( FOI ) of JEV for each cohort . FOI is the instantaneous probability of a susceptible individual to become infected over a short period of time . As proposed by Heisey et al [50] , FOI can be expressed as a function of the number of susceptible individuals over time: dS ( t ) /dt=−λS ( t ) ( 1 ) With S ( t ) the number of susceptible individuals at time t , and λ the force of infection . ( Eq 1 ) has for solution: S ( t ) =S0* exp ( −λt ) ( 2 ) With S0 the number of susceptible individuals at t = 0 . ( Eq 2 ) can be linearized as: ln ( S ( t ) ) = ln ( S0 ) −λt ( 3 ) In ( Eq 3 ) , FOI can be estimated with a generalized linear model if information is available on the evolution over time of the number of susceptible individuals from a starting date ( t = 0 ) . We estimated the FOI by fitting a generalized linear model to our data depicting the transition of susceptible pigs ( tested negative with our ELISA test ) into non-susceptible pigs ( tested positive with our ELISA test ) . For each cohort we used as the starting date ( t = 0 ) the first date with the highest number of susceptible pigs in order to increase precision and a time step ( dt ) of 1 day . Using this method , we assumed that the FOI was constant over each study period . We used the glm ( ) function in the R software [51] . For our estimation of the FOI , we considered the pigs tested equivocal with the ELISA test as non-susceptible . "The views expressed in this article are those of the author and do not necessarily reflect the official policy or position of the Department of Navy , Department of Defense , nor the U . S . Government . " | Japanese Encephalitis Virus ( JEV ) is the most important cause of viral encephalitis in Asia in humans with an estimated 68 , 000 cases annually . The disease is considered a mainly rural one because it occurs mainly in rural areas dominated by paddy fields where the main mosquito species vector of JEV breed . However , other mosquito species , breeding in urban areas , and a large range of animal hosts can play a role in the transmission of JEV , and JEV could therefore be transmitted in peri-urban and urban areas . Our results show an intensive circulation of JEV in sentinel pigs in a peri-urban area of Phnom Penh Cambodia at two different periods of the year . It shows the potential for JEV to circulate in a large range of landscapes and suggest that JEV control should not be limited to rural areas and that JEV may have the potential to emerge and be and be maintained in new areas . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"invertebrates",
"livestock",
"medicine",
"and",
"health",
"sciences",
"pig",
"models",
"immunology",
"geographical",
"locations",
"vertebrates",
"animals",
"mammals",
"animal",
"models",
"preventive",
"medicine",
"model",
"organisms",
"infectious",
"disease",
"control",
"vaccination",
"and",
"immunization",
"insect",
"vectors",
"research",
"and",
"analysis",
"methods",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"swine",
"geography",
"epidemiology",
"disease",
"vectors",
"insects",
"agriculture",
"arthropoda",
"people",
"and",
"places",
"mosquitoes",
"asia",
"urban",
"areas",
"earth",
"sciences",
"geographic",
"areas",
"biology",
"and",
"life",
"sciences",
"cambodia",
"amniotes",
"organisms"
]
| 2016 | Intensive Circulation of Japanese Encephalitis Virus in Peri-urban Sentinel Pigs near Phnom Penh, Cambodia |
Wuchereria bancrofti , Brugia malayi and Brugia timori infect over 100 million people worldwide and are the causative agents of lymphatic filariasis . Some parasite carriers are amicrofilaremic whilst others facilitate mosquito-based disease transmission through blood-circulating microfilariae ( Mf ) . Recent findings , obtained largely from animal model systems , suggest that polymorphonuclear leukocytes ( PMNs ) contribute to parasitic nematode-directed type 2 immune responses . When exposed to certain pathogens PMNs release extracellular traps ( NETs ) in the form of chromatin loaded with various antimicrobial molecules and proteases . In vitro , PMNs expel large amounts of NETs that capture but do not kill B . malayi Mf . NET morphology was confirmed by fluorescence imaging of worm-NET aggregates labelled with DAPI and antibodies to human neutrophil elastase , myeloperoxidase and citrullinated histone H4 . A fluorescent , extracellular DNA release assay was used to quantify and observe Mf induced NETosis over time . Blinded video analyses of PMN-to-worm attachment and worm survival during Mf-leukocyte co-culture demonstrated that DNase treatment eliminates PMN attachment in the absence of serum , autologous serum bolsters both PMN attachment and PMN plus peripheral blood mononuclear cell ( PBMC ) mediated Mf killing , and serum heat inactivation inhibits both PMN attachment and Mf killing . Despite the effects of heat inactivation , the complement inhibitor compstatin did not impede Mf killing and had little effect on PMN attachment . Both human PMNs and monocytes , but not lymphocytes , are able to kill B . malayi Mf in vitro and NETosis does not significantly contribute to this killing . Leukocytes derived from presumably parasite-naïve U . S . resident donors vary in their ability to kill Mf in vitro , which may reflect the pathological heterogeneity associated with filarial parasitic infections . Human innate immune cells are able to recognize , attach to and kill B . malayi microfilariae in an in vitro system . This suggests that , in vivo , the parasites can evade this ability , or that only some human hosts support an infection with circulating Mf .
Throughout history , parasitic nematode infections have had a major impact on human development , especially of the poorest and most disadvantaged populations . Human diseases associated with these infections include lymphatic filariasis ( LF ) and onchocerciasis . A hallmark of these infections is that they are very long-lasting , with the production of very large numbers of microfilariae ( Mf ) that are able to survive within the host . In order to do this , parasitic nematodes have evolved the ability to modulate and suppress the host immune response via the secretion of a cocktail of proteins , micro RNAs and small molecules [1–4] . The current strategy for the elimination of LF and onchocerciasis as public health problems centers on the prevention of transmission by eliminating the Mf from infected hosts , thus preventing any new infections of the insect vectors and hence , more human hosts [5–7] . At present , this is achieved by mass administration of the effective anthelmintic drugs , albendazole , ivermectin and diethylcarbamizine ( DEC ) . Studies on DEC have indicated that the drug interacts with the host immune system in order to be effective [8–13] and some experiments with animal parasites have suggested that ivermectin has an impact on the ability of polymorphonuclear leukocytes ( PMNs ) and monocytes to attach to and kill Mf [14 , 15] . These results imply that host granulocytes and monocytes have the ability to recognize Mf and possibly to kill them . The innate immune response to parasitic nematodes involves many different cell populations , which include granulocytes such as eosinophils , mast cells and PMNs , as well as monocytes . PMNs are critical for controlling a large variety of pathogens including nematodes . They have previously been implicated in the killing of nematode larvae , including Onchocerca volvulus Mf and L3 [16 , 17] , and have been reported to be a key component of the host innate immune response to nematode infections [18] . For example , increased numbers of PMNs in the skin and blood of infected mice reduced the success of invading L3 of the filarial nematode , Litomosoides sigmodontis [19] . A characteristic feature of PMN responses is the production of DNA-containing neutrophil extracellular traps ( NETs ) [20] . These structures are formed by a unique type of cell death , NETosis , and are characterized by large , extracellular concentrations of expelled cytosolic , granular and nuclear material including DNA , histones , neutrophil elastase and myeloperoxidase [21] . NETosis is frequently , but not always , mediated by NADPH oxidase [21–22] . NET formation is induced by parasitic nematodes but whether these are required for nematode killing is uncertain and may depend on the parasite under study . Despite being trapped by NETs in vitro , the L3 larvae of both Strongyloides stercoralis and Haemonchus contortus were not killed by NETs alone [23–24] although treatment with DNase to destroy NETs did reduce PMN plus macrophage mediated killing of S . stercoralis L3 [23] . In several studies PMNs have been shown to co-operate with monocytes or macrophages in immunity against parasites , including helminths [18 , 24–27] . We have previously shown that PMNs and peripheral blood mononuclear cells ( PBMCs ) from uninfected dogs attach to Dirofilaria immitis Mf in vitro and that this attachment was increased by the addition of ivermectin [14] . We have extended these studies to the human parasite Brugia malayi and investigated the ability of leukocytes purified from presumably parasite-naïve North American human donors to recognize and kill B . malayi Mf isolated from the peritoneal cavity of infected Mongolian gerbils , Meriones unguiculatus . B . malayi is the causative agent of a minority ( roughly 10% ) of cases of LF , however it is the only filarial nematode of humans that can be maintained in a convenient laboratory animal host . Our results provide evidence that PMNs and monocytes of many , but not all , human donors were able to both adhere to and kill B . malayi Mf .
Although NET formation was initially characterized in response to bacteria and protozoa [20 , 27] , it has also been reported for larger multi-cellular pathogens including fungi and three species of parasitic nematodes [19 , 24 , 25] . Our initial experiments co-culturing B . malayi Mf with human neutrophils in the presence of the membrane-impermeable DNA-binding dye SYTOX Orange resulted in Mf becoming tethered in a manner consistent with entanglement in NETs ( S1 Video ) . These observations prompted us to confirm that the structures generated possessed typical NET characteristics . Live Mf were co-cultured with human PMNs and the generation of NETs observed by confocal microscopy after staining with DAPI and antibodies to characteristic NET-associated proteins . Worm-NET aggregates stained positive for citrullinated histone H4 and the granular proteins , neutrophil elastase and myeloperoxidase ( Fig 1 ) . All the NET-associated proteins examined co-localized with extracellular DNA , confirming typical NET morphology . To further characterize Mf-induced NETosis , we developed a confocal microscopy-based extracellular DNA release assay . Live cell fluorescence imaging of PMN and Mf interactions in the presence of SYTOX Orange allowed us to observe extracellular DNA and NET formation over time ( Fig 2A–2C and S2 Video ) . Mean SYTOX Orange intensity values derived from these images were used to quantify total extracellular DNA release ( Fig 2D–2H ) in wells where PMNs were stimulated with either Mf or 25nM phorbol myristate acetate ( PMA ) as a positive control [28] . These data demonstrate that in the absence of serum , B . malayi Mf significantly increased the release of extracellular DNA when compared to the zero Mf controls ( Fig 2D; P = 0 . 041 ) . Both DNase I and the NADPH oxidase inhibitor diphenyleneiodonium ( DPI ) significantly reduced the mean SYTOX Orange intensities derived from Mf treated wells ( Fig 2E; P<0 . 001 for both treatments ) , presumably via the enzymatic breakdown of NET structure and the inhibition of NETosis respectively . Interestingly , in the presence of 5% autologous serum , we did not detect any significant increase in extracellular DNA release within Mf treated wells compared to the zero Mf controls ( Fig 2F ) . This may be due to the inhibitory effects of the autologous sera which impeded DNA release in Mf ( P<0 . 001 ) containing wells ( Fig 2G ) . The autologous sera also reduced the extracellular DNA release induced by PMA ( S1 Fig ) . Heat treatment of the autologous sera had no significant effect ( Fig 2G ) , suggesting that the complement system was not responsible for the inhibition of extracellular DNA release . We have previously shown that canine neutrophils isolated from uninfected dogs can attach to D . immitis Mf in vitro [14] . Therefore , we examined whether a similar phenomenon was observed when we incubated PMNs from uninfected humans with Mf of the human parasite , B . malayi . Video analysis of 96-well plate-based co-cultures allowed us to confirm PMN to Mf attachment in vitro and to count the number of individual Mf with at least one cell attached . After 24 hours , the addition of 5% autologous serum increased the number of worms with attached PMNs from 19 . 3% to 31 . 1% ( P = 0 . 035 ) . DNase I treatment virtually abolished PMN attachment in the absence of serum ( 0 . 4 ± 0 . 3% of Mf had ≥1 PMN attached at 24 hours post experimental set up ( p<0 . 001 ) , ) , suggesting that NETs were required for PMNs to attach to the worms ( Fig 3A ) . In contrast , DPI had no significant impact on PMN attachment under these conditions ( Fig 3A ) , suggesting that attachment took place via an NADPH oxidase-independent mechanism . Heat treatment ( 55°C for 30 min ) of the sera inhibited attachment , reducing it to levels less than in the zero serum controls , with only 7 . 9% of the Mf having at least one cell attached ( P = 0 . 015 ) ( Fig 3A ) . The effect of heat treatment suggested that heat-labile components of the autologous serum promoted PMN to worm attachment . The number of PMNs attached to individual Mf varied , and we often observed Mf that had large numbers of PMNs adhered to their surface ( Fig 3B ) ; however , only a single cell was required to be attached in these assays for the Mf to be scored . The previous experiments clearly showed that human PMNs can recognize and attach to B . malayi Mf . Bonne-Année and colleagues reported that PMNs and peripheral blood mononuclear cells ( PBMCs ) collaborate to kill S . stercoralis larvae when incubated in 25% human serum for 48 hours [29] , so we tested whether or not the formation of NETS and cell attachment we observed also resulted in Mf killing . Worm survival was monitored over 5 days in culture in the presence and absence of peripheral blood leukocytes ( PMNs and PBMCs ) isolated from uninfected people . Approximately 85% of Mf survived for 5 days in the absence of human leukocytes ( Fig 4C ) . Exposure to PBMCs alone did not significantly alter Mf survival in vitro ( Fig 4 ) , but when Mf were maintained in the presence of either PMNs alone or both PMNs and PBMCs , worm survival was significantly inhibited compared to the zero cell controls ( Fig 4 ) . Interestingly , when Mf were incubated with both PMNs and PBMCs , significantly fewer worms survived to day 5 ( 36 . 9 ± 14 . 7% , see Fig 4C; P<0 . 001 ) when compared to the Mf exposed to PMNs alone ( 60 . 6 ± 14 . 7% , see Fig 4C; P<0 . 001 ) . DNase I treatment of the cultures had no significant effect on Mf survival in the presence of PMNs , PBMCs or both , indicating that NET formation was not required for Mf killing , whereas heat treatment of the serum effectively blocked all leukocyte-mediated killing ( Fig 4A–4C ) . Over the course of these experiments we noticed that the levels of both cell to worm attachment and leukocyte mediated Mf killing varied greatly between individual experiments . In an attempt to better understand this variation we re-analyzed the videos of co-culture wells that contained both PMNs and PBMCs to score leukocyte to worm attachment and compare this to Mf survival . These data highlight an obvious negative correlation between the levels of leukocyte attachment at one hour post experimental set up and the percentage of worms that survived to day 5 ( Fig 4D; Spearman's rho = -0 . 85 , P<0 . 0001 ) . Each experiment was carried out using cells from a single donor and this donor was different for each experiment . The isolated cells appeared to split largely into two distinct phenotypes: “Mf killers” who displayed both rapid leukocyte attachment ( >90% Mf with ≥1 leukocyte attached at 1 hour ) and near complete Mf killing ( <10% of the Mf survived to day 5 ) and “non-killers” who displayed relatively low levels of leukocyte attachment ( <42% Mf with ≥1 leukocyte attached at 1 hour ) and little to no leukocyte mediated Mf killing compared to zero cell controls ( 60–80% Mf survival at 5 days post set up ) , though there were 2 donors whose cells had an intermediate phenotype . This explains the relatively high variation ( reflected in the error bars ) in attachment and killing seen between experiments . This analysis also reveals that when killing occurs the leukocytes recognize and attach to the Mf very rapidly and that if this attachment does not take place within one hour , the parasites survive quite well in these culture conditions . Serum heat inactivation has been shown to prevent the killing of S . stercoralis L3 larvae mediated by human PMNs and PBMCs [29] . These observations led to the conclusion that complement was required for larval killing . Given the effects of serum heat treatment on PMN attachment ( Fig 3 ) and subsequent Mf survival ( Fig 4 ) , we directly investigated the possibility that the complement system is involved in PMN to worm attachment and leukocyte-mediated Mf killing . We repeated our PMN attachment and Mf survival assays whilst blocking complement activation via the complement specific inhibitor , compstatin [30] . In these experiments we only analyzed data obtained from those experiments where substantial Mf killing was observed , since this is the phenomenon we were seeking to study . Pre-treatment of 25% serum with 100μM compstatin did not significantly affect either attachment ( Fig 5A , red panels ) or killing ( Fig 5B ) . These data suggest that the complement system does not significantly contribute to PMN plus PBMC mediated Mf killing in vitro . However , since the attachment experiments were conducted in 5% serum we also tested the effect of compstatin on attachment under these conditions . Although there was a small reduction between the control peptide ( ~90% attachment ) and the compstatin treated samples ( ~65% attachment ) ( Fig 5A , light blue panels , P = 0 . 045 ) , there was no significant difference in PMN to worm attachment between the no peptide control and compstatin treated wells . To further investigate the role of complement we also used blocking antibodies to two complement- and adhesion-associated proteins , Cd11b ( complement receptor 3 , CR3 ) and ICAM-1 . Neither blocking antibody had an effect on Mf survival ( Fig 5C ) . Complement therefore plays only a minor role , if any , in attachment and killing of Mf , though a heat-labile component of normal human serum is required . Addition of PBMCs to PMNs increased the amount of Mf killing , however , PBMCs alone were not sufficient to affect survival . Monocytes and neutrophils are both crucial in immune responses to infection [31] and interactions important for helminth clearance have recently been described [18] . Monocytes make up ~10% of the PBMC preparation isolated for our survival experiments ( see Materials and Methods ) so we hypothesized that monocytes may represent the microfilaricidal cells present within the PBMC population . To test this , we repeated our Mf survival assays but replaced the PBMC population with 1500 monocytes/Mf . In these experiments , incubation with either PMNs or monocytes reduced Mf survival after 5 days to about 40%; incubation with both cell types did not significantly reduce survival any further ( Fig 6A ) although attachment of both PMN and monocytes could be detected in co-cultures at 120 hours post incubation ( Fig 6B ) . These data highlight that both PMNs and human monocytes are independently capable of killing B . malayi Mf in vitro ( P = 0 . 005 ) . Immunofluorescence staining showed that both CD14+/CD16- monocytes and CD14+/16+ PMNs were attached to the Mf at 5 days post-infection by which time significant levels of killing had occurred ( Fig 6B ) . In contrast , purified lymphocytes were unable to kill Mf , nor did their addition to PMN reduce survival any further ( Fig 6C ) . Taken together , our data show that PMNs and monocytes , but not lymphocytes , isolated from North American human donors , who have presumably never been exposed to B . malayi , can recognize and kill Mf in vitro . There is some variation in the extent of parasite killing between experiments , which may represent differences between the cells isolated from individual human donors , or between batches of Mf . Killing is preceded by rapid attachment ( <1 hr ) of the leukocytes , but is rather slow , taking up to 5 days .
Filarial nematodes , including Mf , survive for months and years in their hosts without provoking an effective immune response . Nonetheless , in this paper we confirm that leukocytes taken from uninfected people can recognize , attach to and kill the Mf stage in vitro [15–17] . As a starting point we wished to determine if the results we previously obtained using the animal parasite , D . immitis , and canine leukocytes [14] , could be reproduced in vitro using human cells and a human filarial parasite . In particular , we wanted to extend these observations and determine if human PMN-derived DNA-based extracellular traps ( NETs ) [32] could ensnare and kill the Mf of B . malayi . NETosis remains poorly characterized with respect to immunity to parasitic nematodes . Recent studies have confirmed that NETs are released from human , bovine and mouse PMNs exposed to the L3 larval stages of S . stercoralis , H . contortus and L . sigmodontis respectively [19 , 24 , 25] . Both H . contortus and S . stercoralis L3 larvae were trapped but not killed by NETs alone [24 , 25] , though DNase I-mediated extracellular trap destruction prevented human PMN- , macrophage- and autologous serum-mediated killing of S . stercoralis larvae . Extracellular traps may therefore contribute to broader killing mechanisms that require multiple immune components . Mouse neutrophils release NETs when exposed to larvae of the human parasite , S . stercoralis [24] , though DNase I treatment did not block killing of the larvae by mouse leukocytes in vitro [24] . Our confocal microscopy-based assays confirmed the presence of classical NET markers when human PMNs were incubated with B . malayi Mf , demonstrating that Mf can induce NET release in vitro ( Fig 1 and Fig 2A–2D ) , and that these structures contain all of the components reported from other systems [20 , 21] . DNase I treatment effectively destroyed NET structure ( Fig 2E ) and blocked PMN to Mf attachment in the absence of serum ( Fig 3A ) , as predicted , but did not inhibit human leukocyte mediated Mf killing ( Fig 4C ) . DNA-containing NET formation is therefore not essential for human leukocytes to kill B . malayi Mf in vitro . This suggests that the importance of NETosis to nematode parasite killing varies with both host and parasite species . It is also possible that in vivo NETs do contribute to Mf killing but that additional components were missing from our in vitro survival assay . An increase in NET-like structures was correlated with reduced S . stercoralis L3 survival within cell impermeable diffusion chambers implanted into the mouse model [24] . DPI significantly inhibited B . malayi Mf-induced DNA release in the absence of autologous serum ( Fig 2E ) , mirroring the results associated with H . contortus [25] and indicating a role for NADPH oxidases in parasitic nematode driven NETosis . Despite this , DPI appeared to have no significant impact on PMN to Mf attachment ( Fig 3A ) and obvious NET aggregates could be seen within DPI treated wells . We were not able to examine the effect of DPI on Mf survival due to the negative effects of long term DPI exposure on worm health ( the worm appeared sluggish but not dead at day 5 ) ; these effects are presumably due to DPI inhibiting the nematodes’ NADPH oxidase . Human PMNs can attach to and kill B . malayi Mf in vitro , and Mf survival is further reduced if PBMCs are added to the cultures ( Fig 4 ) . The PBMC population failed to kill Mf in the absence of other cell types , yet significantly increased the level of parasite killing when co-cultured with PMNs ( Fig 4 ) , suggesting some cross-talk between PMNs and a component of the PBMC fraction , presumably monocytes since these cells are also able to kill Mf ( Fig 6 ) . We noted a large amount of variation in both leukocyte attachment and Mf survival between individual experiments , and the two measurements–attachment at 1 hour and survival at 5 days–are clearly negatively correlated ( Fig 4D ) . This could arise from differences between the cells and/or serum isolated from individual human donors , or between different batches of Mf . It is impossible to distinguish between the two using our current protocols as we do not know the identity of the human donors , and so cannot examine HLA genotypes for example , and since cells and sera were used immediately after isolation , we could not test them on different batches of Mf . In endemic regions the majority of infected individuals are tolerant of high parasite loads and microfilaremia . In contrast , individuals with pathological manifestations ( e . g . lymphedema and hydrocele ) show stronger immune reactions [33] . The genetic factors that regulate susceptibility to parasitic infections and the pathological heterogeneity associated with filarial nematode infection are not entirely understood [34] , but it is possible that these are reflected in the ability of the innate immune system to rapidly recognize and kill Mf , as shown here . In contrast , the nematodes used in this study are an inbred population that would not be expected to exhibit much genetic variation , but non-genetic factors may account for the differences observed between experiments . For example , differences in the amounts of immunomodulatory ES products present in the various batches of Mf may explain the variation in rapid attachment and subsequent killing that we observed . PMN attachment and Mf killing were both promoted by the addition of autologous serum , and heat treatment of the serum inhibited both PMN attachment and PMN plus PBMC-mediated Mf killing ( Fig 3 and Fig 4 ) . This was not due to complement as the complement specific inhibitor compstatin failed to have any biologically significant effects on either PMN attachment or leukocyte mediated Mf killing ( Fig 5 ) . Compstatin binds to C3 to prevent C3 cleavage and competitively inhibit all three complement activation pathways [30] . These data suggest that complement does not contribute much , if at all , to PMN attachment or Mf killing , but that another unidentified heat-labile component of human serum is involved . Antibodies against CD11b and ICAM-1 , two molecules previously implicated in the interactions between Mf and the immune system [18 , 33] , also failed to inhibit killing , suggesting that they are not required for this process to take place ( Fig 5C ) . In the presence of autologous serum both PMNs and monocytes are capable of killing Mf alone ( Fig 4 and Fig 6 ) . Bonne-Année and colleagues have shown that human PMNs can collaborate with either PBMCs or macrophages to kill S . stercoralis larvae [24 , 29] , however , despite using relatively high numbers of leukocytes , they did not observe reduced L3 larvae survival on exposure to individual leukocyte populations [24 , 29] . This could reflect differences in the nematode species or life-stages employed , particularly worm size which varies considerably between life-stages and perhaps influences the number of leukocytes required to kill the parasite . Perhaps surprisingly , we observed no increased killing when PMNs and monocytes were incubated together with the Mf . Neutrophils and monocytes are well known to communicate with each other and it is has been reported that neutrophils induce an anti-nematode immunity in monocytic cells [18 , 35] , however this was not reflected in our in vitro experimental system . These data describe the first example of Mf induced NETosis and contribute significantly to the growing body of evidence that suggest an important role for neutrophils in regulating parasitic nematode infections [18 , 36] . We show that human peripheral blood innate immune cell populations can recognize , trap and kill the blood circulating life-cycle stage of the human filarial parasitic nematode , B malayi . Variation in worm killing and leukocyte attachment between human blood donors suggest that the innate immune system could significantly contribute to the regulation of host tolerance and susceptibility to infection and more specifically the regulation of microfilaremia which is a key determinant of the transmission of lymphatic filariasis .
All experiments and informed consent procedures were approved by the Institutional Review Boards of the University of Georgia ( permit number 2012–10769 ) , and the studies were conducted in accordance with the ethical guidelines of Declaration of Helsinki . Human subjects recruited under the guidelines of IRB-approved protocols provided written informed consent for participation in the studies described below . Live B . malayi Mf isolated from the peritoneal cavity of infected Mongolian gerbils were provided by the Filarial Research Reagent Resource Center ( FR3: Athens , GA , USA ) . Mf were washed three times in phosphate buffered saline ( PBS; centrifuged at 1500 x g for 8 min ) and re-suspended in RPMI-1640 ( Gibco , Life Technologies , Grand Island , NY , USA ) . Note that all RPMI-1640 used in this study was supplemented with 100 U/ml penicillin-streptomycin ( Life Technologies , Grand Island , NY , USA ) and 0 . 1 mg/ml gentamicin ( Sigma , St . Louis , MO , USA ) . Re-suspended Mf samples were then filtered through a 5μm Isopore membrane ( Merck Millipore Ltd . , Carrigtwohill , Cork , Ireland ) to capture the Mf and exclude contaminating small particles . Membranes were socked in RPMI-1640 at 37°C and 5% CO2 for 20–30 min to facilitate the migration of viable Mf from the membrane . Viable Mf were incubated overnight in RPMI-1640 at 37°C and 5% CO2 . Mf samples were washed for a second time by 5μm Isopore membrane filtration just before use . Leukocytes were isolated from freshly donated peripheral blood drawn from healthy U . S . residents at the Health Center of the University of Georgia . 40ml of blood was anticoagulated by heparin . PMNs were isolated using the EasySep Direct Human Neutrophil Isolation Kit ( Stemcell Technologies , Vancouver , BC , Canada ) according to manufacturer’s instructions . PBMCs were isolated using SepMate-50 Tubes ( Stemcell Technologies , Vancouver , BC , Canada ) according to manufacturer’s instructions . The optional extended wash step ( 120 x g for 10 min ) of the SepMate protocol was included to remove contaminating platelets . Monocytes were isolated from the PBMC samples using the EasySep Human Monocyte Enrichment Kit ( Stemcell Technologies , Vancouver , BC , Canada ) according to manufacturer’s instructions . Isolated PMNs , PBMCs and monocytes were washed in PBS ( centrifuged at 300 x g for 5 min ) , re-suspended in a 1:1 mixture of RPMI-1640 and autologous serum , stored at room temperature and used within 6 hours post-isolation as previously described [37] . The concentration of cells per population was estimated in a sample of cells stained with 0 . 4% trypan blue ( Gibco , Life Technologies , Grand Island , NY , USA ) using a hemocytometer . All populations used were estimated at 95% or greater viability . To estimate the purity of isolated cell populations , cell slides were prepared using the Cytospin 3 CellPreparation System ( Shandon Scientific Limited , Astmoor , Runcorn , Cheshire , England ) and stained with Modified Wright’s stain ( Hema 3 Stat pack , Fisher Scientific , Kalamazoo , MI ) . The average purity of the PMN population was ~97% and the monocyte population ~92% . The PBMC population contained predominantly lymphocytes but included ~10% monocytes and ~10% PMNs . Leukocytes were washed once in Hanks-balanced salt solution ( HBSS; Gibco , Life Technologies , Grand Island , NY , USA; centrifuged at 300 x g for 5 min ) and re-suspended in RPMI-1640 before use . Lymphocytes were isolated using the EasySep Direct Human Total Lymphocyte Isolation Kit ( Stemcell Technologies , Vancouver , BC , Canada ) according to manufacturer’s instructions . Isolated lymphocytes were washed in PBS ( centrifuged at 300 x g for 5 min ) and re-suspended in a 1:1 mixture of RPMI-1640 and autologous serum . 10 ml of autologous blood was collected as described above but allowed to clot in the absence of heparin ( incubated for ~2 hours at room temperature ) [38] . Briefly , the liquid fraction of the blood sample was aspirated and centrifuged at 10 , 000 x g for 5 min . The supernatant was aspirated and filter sterilized . Where necessary the serum was heat treated ( 55°C for 30 min ) and/or diluted in RPMI-1640 to the desired concentration before use . For detection of NET-associated proteins , isolated PMNs , 1 . 5 x 105 cells in 100 μl RPMI-1640 , were seeded onto 12 mm #1 round coverslips in 24-well flat bottom dishes ( Costar , Corning , NY , USA ) and left to adhere for 1 hour at 37°C and 5% CO2 prior to adding 100 B . malayi Mf in 100 μl RPMI-1640 . The coverslips were incubated a further 18 hours at 37°C and 5% CO2 prior to immunostaining . Cells and Mf were fixed by adding 200 μl of 4% paraformaldehyde to the well and incubated for 20 min at room temperature . Supernatants were carefully removed and coverslips washed twice with PBS prior to blocking with PBS plus 5% FBS and 1% BSA for 20 min at room temperature . Primary antibodies were diluted in blocking solution as follows: ( 1 ) rabbit anti-histone H4 ( citrulline R3 ) antibody 1:500 ( ab81797; Abcam , Cambridge , MA , USA ) , ( 2 ) mouse anti-human myeloperoxidase 1:200 ( clone MPO455-8E6; eBioscience , San Diego , CA , USA ) and ( 3 ) goat anti-human neutrophil elastase 1:200 ( clone C-17 , sc-9520; Santa Cruz Biotechnology , Dallas , TX , USA ) . After 1 hour incubation at room temperature , coverslips were washed three times with PBS and incubated with a 1:200 dilution of donkey anti-rabbit IgG-Alexa488 , donkey anti-goat IgG-Alexa555 and donkey anti-mouse IgG-Alexa647 ( Life Technologies , Thermo Fisher Scientific ) in blocking buffer containing 0 . 1μg/ml DAPI for 30 min at room temperature . Coverslips were washed , mounted in Mowiol ( Calbiochem/EMD Biosciences , La Jolla , CA ) and imaged using a Nikon A1R confocal microscope ( Nikon Instruments Company , Melville , NY , USA ) . Figures were assembled using Photoshop software ( Adobe , San Jose , California , USA ) . Immuno-labelling of Mf-associated monocytes and PMNs was performed on samples of Mf incubated with purified PMNs and monocytes for 120 hours in 96-well round bottomed plates ( Corning Glass Works , NY ) . Samples were centrifuged onto microscope slides at 1000 rpm for 5 min in a Cytospin 3 cytocentrifuge ( Shandon Scientific Limited , Astmoor , Runcorn , Cheshire , England ) , fixed in methanol for 50 second blocked with 5% FBS , 1% BSA in PBS for 30 min . The slides were incubated with a 1:100 dilution of mouse anti-human CD16-Alexa Fluor 488 , clone 3GB ( Stemcell Technologies , Vancouver , CA ) and 1:100 dilution of mouse anti-human CD14 Alexa Fluor 594 clone HCD14 ( Biolegend , San Diego , CA ) in blocking solution for 2 hours RT in a humidified chamber in the dark . DAPI was added to the primary antibody solution for the last 25 min of incubation . Following rinsing for 2 x 5 min in PBS , the slides were coverslipped , mounted in Mowiol and set overnight prior to viewing . Z-stack images were collected using a Nikon A1R confocal microscope and NIS Elements software ( Nikon Instruments Company , Melville , NY , USA ) . Images were prepared using Adobe Photoshop software ( Adobe , San Jose , California , USA ) . Assays were set up in Nunc 384-Well optical bottom tissue culture plates ( THERMO Scientific , Rochester , NY , USA ) . There were four components added to each well in 12 . 5μl volumes , giving a total volume of 50μl . ~18 , 750 PMNs were added to each well . PMNs were suspended in RPMI-1640 containing 12 . 5μM SYTOX Orange Nucleic Acid Stain ( Life Technologies , Eugene , OR , USA ) to give a final concentration of 3 . 125μM of SYTOX Orange stain and enable the quantification of extracellular DNA [37 , 38] . The other three well components varied between treatment groups and included: 5% autologous serum , 5% autologous heat treated serum , ~25 B . malayi Mf , 10 μM diphenyleneiodonium ( DPI; Sigma , St . Louis , MO , USA ) , 30μg/ml DNase I ( Roche , Indianapolis , IN , USA ) and 25nM phorbol myristate acetate ( PMA; Sigma , St . Louis , MO , USA ) , which was employed as a positive control . The concentrations stated here represent the final concentrations obtained once all components had been added to the well . To create the negative controls , 12 . 5μl of RPMI-1640 was substituted for the relevant component . The tissue culture plates were incubated at 37°C and 5% CO2 throughout the experiment . Both transmitted light and fluorescence images were captured on a Nikon A1R confocal microscope system equipped with a 60X 1 . 4NA lens . A single field of view with was taken at a random position within each well . Images were taken every 30 min for 7 hours using automated capture software . The first images were taken 1 hour post-experimental set up to allow the worms and cells to settle to the bottom the wells . Mean SYTOX Orange intensities of the fluorescent images were quantified using the measure region of interest ( ROI ) feature of the Nikon A1 software . The entire image was highlighted as the ROI . These measurements were used to calculate changes in SYTOX Orange intensities over time ( ΔMean SYTOX Orange intensity ) . Each biological replicate represents the mean of three technical replicates . Assays were set up in Nunc 96-Well optical bottom tissue culture plates ( THERMO Scientific , Rochester , NY , USA ) . There were four components added to each well in 50μl volumes , giving a total volume of 200μl . ~100 B . malayi Mf , ~75 , 000 PMNs and 50μl of RPMI-1640 were added to each well . The worm to cell ratio selected ( 1:750 ) was sufficiently high so that the number of available cells did not limit attachment [14] . The final component added to each well varied between treatment groups and included: 5% autologous serum , 5% autologous heat treated serum , 10μM DPI and 30μg/ml DNase I . To create the respective controls , 50μl of RPMI-1640 was substituted for the relevant component . The tissue culture plates were incubated at 37°C and 5% CO2 throughout the experiment . Videos of each well were taken on an inverted microscope ( 40X magnification ) at 2 . 5 , 5 , 16 and 24 hours post-experimental set up . Videos were blinded and all Mf were scored for attachment . Mf that had at least 1 PMN adhered to their surface or indirectly fastened by extracellular DNA were scored as attached . Each biological replicate represents the mean of two or three technical replicates . PMN attachment to Mf was confirmed by cytocentrifugation and Wright stain as described above . Assays were set up in Nunc 96-Well optical bottom tissue culture plates . There were four components added to each well in 50μl volumes , giving a total volume of 200μl . ~100 B . malayi Mf were added to each well . The other three components varied between treatment groups and included: 25% autologous serum or 25% autologous heat treated serum , ~150 , 000 PMNs , ~150 , 000 PBMCs , ~150 , 000 monocytes , ~150 , 000 lymphocytes and 30μg/ml DNase I . To create the respective controls , 50μl of RPMI-1640 was substituted for the relevant component . The tissue culture plates were incubated at 37°C and 5% CO2 throughout the experiment . Videos of each well were taken on an inverted microscope ( 40X magnification ) at 1 , 24 , 48 and 120 hours post-experimental set up . Videos were blinded and the numbers of moving Mf present within each well were counted . Mf that were not moving were considered dead . The number of surviving Mf was normalized relative to the number of moving Mf scored 1 hour post-set up ( = 100% ) , and expressed as a relative percentage . Each biological replicate represents the mean of two or three technical replicates . For the compstatin inhibition studies , autologous serum was pretreated with either 100μM compstatin ( Tocris Bioscience , Avonmouth , Bristol , U . K . ) or 100μM compstatin control peptide ( Tocris Bioscience , Avonmouth , Bristol , U . K . ) for 30 min at 37°C . The compstatin treated serum and antibody treated leukocytes were added directly to the PMN attachment and Mf survival assays described above . Inhibition of Cd11b activity was implemented by incubating 2g of the monoclonal antibody clone M1/70 purified specifically for use with live cells ( BioLegend , San Digeo , CA , cat #101248 ) with PMNs for 2 hours at room temperature prior to adding B . malayi Mf and incubating for 5 days at 37°C prior to assessing survival as described above . Block of ICAM-1 function was carried out as for Cd11b using 2g of anti-ICAM1 antibody [MEM-11] ( Abcam , Cambridge , MA ) . Normality of the data was assessed based on examination of histograms and normal Q-Q plots of the residuals . Constant variance of the data was assessed by plotting residuals against predicted values . Data were analyzed using linear mixed effects modeling with independent experiment modeled as a random effect and treatment group and time ( when applicable ) modeled as fixed nominal effects . Two-way interactions were also included in the model when applicable . Model fit was assessed using Akaike information criterion values . When indicated , adjustments for multiple relevant comparisons were done using the method of Bonferroni . For all analyses , adjusted P<0 . 05 was considered significant . | Lymphatic filariasis is a disfiguring and debilitating neglected tropical disease caused by filarial parasitic nematodes including Brugia malayi . The immune mechanisms that drive host tolerance and host resistance to blood circulating microfilariae are not yet fully understood . In this study , we investigate the interaction between B . malayi microfilariae and human peripheral blood leukocytes . Polymorphonuclear leukocytes are the most abundant leukocyte population present within the human circulatory system and can release DNA-based extracellular traps ( NETs ) that capture and kill specific pathogens . We show that human neutrophils release NETs in response to B . malayi parasites . These NETs promote leukocyte-to-worm attachment but do not kill the microfilariae . Despite this , we highlight that neutrophils and monocytes can kill these parasites in vitro . Interestingly , the population of healthy human blood donors appeared to split into two distinct phenotypes: “Mf killers” and “non-killers” . Surprisingly , our findings suggest that activation of the complement system does not significantly contribute to leukocyte mediated parasite killing . Our study provides new insights into host-microfilariae interactions and the role that the human innate immune system plays in filarial parasite defense and the regulation of microfilaremia , a key determinant of disease transmission . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"blood",
"cells",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"engineering",
"and",
"technology",
"nucleases",
"deoxyribonucleases",
"enzymes",
"immunology",
"manufacturing",
"processes",
"dna-binding",
"proteins",
"enzymology",
"parasitic",
"diseases",
"animals",
"nematode",
"infections",
"brugia",
"heat",
"treatment",
"neutrophils",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"biochemistry",
"hydrolases",
"cell",
"biology",
"brugia",
"malayi",
"monocytes",
"nematoda",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
]
| 2017 | Human Leukocytes Kill Brugia malayi Microfilariae Independently of DNA-Based Extracellular Trap Release |
Many plants release airborne volatile compounds in response to wounding due to pathogenic assault . These compounds serve as plant defenses and are involved in plant signaling . Here , we study the effects of pectin methylesterase ( PME ) -generated methanol release from wounded plants ( “emitters” ) on the defensive reactions of neighboring “receiver” plants . Plant leaf wounding resulted in the synthesis of PME and a spike in methanol released into the air . Gaseous methanol or vapors from wounded PME-transgenic plants induced resistance to the bacterial pathogen Ralstonia solanacearum in the leaves of non-wounded neighboring “receiver” plants . In experiments with different volatile organic compounds , gaseous methanol was the only airborne factor that could induce antibacterial resistance in neighboring plants . In an effort to understand the mechanisms by which methanol stimulates the antibacterial resistance of “receiver” plants , we constructed forward and reverse suppression subtractive hybridization cDNA libraries from Nicotiana benthamiana plants exposed to methanol . We identified multiple methanol-inducible genes ( MIGs ) , most of which are involved in defense or cell-to-cell trafficking . We then isolated the most affected genes for further analysis: β-1 , 3-glucanase ( BG ) , a previously unidentified gene ( MIG-21 ) , and non-cell-autonomous pathway protein ( NCAPP ) . Experiments with Tobacco mosaic virus ( TMV ) and a vector encoding two tandem copies of green fluorescent protein as a tracer of cell-to-cell movement showed the increased gating capacity of plasmodesmata in the presence of BG , MIG-21 , and NCAPP . The increased gating capacity is accompanied by enhanced TMV reproduction in the “receivers” . Overall , our data indicate that methanol emitted by a wounded plant acts as a signal that enhances antibacterial resistance and facilitates viral spread in neighboring plants .
Plants are exposed to a diverse range of abiotic and biotic stresses [1]–[3] . Physical damage to a plant is a potential threat because it provides an opportunity for pathogen entry . Localized tissue damage elicits the expression of an array of antimicrobial phytochemicals [4] , proteins [5] , and systemic defense responses against microbial pathogens [6] , [7] and herbivore attack [1] , [8]–[14] . Systemic defense responses provide an attractive model for the study of cell-to-cell signal transduction pathways that operate over long distances [15] , [16] . The molecular mechanisms of systemic wound signaling are not yet fully understood , but several of the non-cell autonomous signals that are released from damaged cells have been studied . In response to pathogen attack or physical damage , several plant species emit volatile organic compounds ( VOCs ) , including ethylene [17] , methyl salicylate [18] , methyl jasmonate [19] , [20] , nitric oxide [21] , [22] and cis-3-hexen-1-ol [23] , which upregulate pathogen-related ( PR ) genes [14] , [23] , [24] . Pectin methylesterase ( PME , EC: 3 . 1 . 1 . 11 ) [25] is a PR protein [26] and is the first barrier of defense against invading pathogens [26]–[31] and herbivores [32]–[34] . In higher plants , PME is a ubiquitous multifunctional enzymatic component of the plant cell wall ( CW ) . The PME gene encodes a pro-PME precursor with an N-terminal extension of variable length [35]–[37] . The tobacco pro-PME contains a long N-terminal leader with a transmembrane domain , which is important for PME delivery into the CW [37] , [38] . PME participates in CW modulation during general plant growth [39]–[42] , nematode infection [43] and pollen tube growth [44]–[47] . PME interacts with the movement protein of the Tobacco mosaic virus ( TMV ) [48] , [49] , suggesting that PME may be involved in the cell-to-cell movement of plant viruses [50] . PME also efficiently enhances virus- and transgene-induced gene silencing ( VIGS and TIGS ) via the activation of siRNA and miRNA production [51] , [52] . In the case of bacterial and fungal phytopathogens , PMEs act as virulence factors that are necessary for pathogen invasion and spreading through plant tissues [53] . The general structure of plant PME is very similar to that of the enzymes produced by phytopathogens [54] . Due to this structural similarity , transgenic plants overexpressing PME can be used as a model of host responses to pathogenic attack . A transgenic tobacco plant ( Nicotiana tabacum L . ) expressing a fungal PME exhibited a dwarf phenotype , modified CW metabolism [40] and a two-fold increase in leaf sap methanol levels . The pectin demethylation directed by PME is likely to be the main source of methanol , which has long been assumed to be a metabolic waste product [55]–[57] . Methanol can accumulate in the intercellular air space at night after the stomata have closed [57] . Methanol emission peaks have been observed in the morning , when the stomata open [58] . Wounding and herbivore attack increase methanol emission levels [34] , [59]–[61] . Transgenic plants with a silenced PME gene had a 50% reduction of PME activity in their leaves and a 70% reduction of methanol emissions compared with wild type ( WT ) plants . This result demonstrates that herbivore-induced methanol emissions originate from pectin demethylation by PME [33] . However , there is no direct evidence that de novo synthesized PME participates in methanol synthesis . In a study of VOC emissions from Nicotiana attenuata plants attacked by Manduca sexta larvae [34] , [60] , methanol was detected in the headspace of leaves very quickly ( 10 min ) after leaf wounding . Therefore , it was concluded that the methanol detected was produced by PME that had been deposited in the CW before the leaf damage occurred . To investigate the metabolism of methanol in plants , Downie et al . [62] used foliar sprays to apply methanol stimulation to Arabidopsis thaliana and studied the resulting changes in gene expression in leaves harvested 1 , 24 , and 72 h after methanol treatment using a 26 , 090 element oligonucleotide microarray . A concentration of 10% ( v/v ) methanol containing Silwet surfactant was used , to expose plants to a methanol concentration in essential excess of endogenous levels . A total of 484 ( 1 . 9% ) transcripts were shown to be regulated in response to the methanol treatment . A group of genes encoding detoxification proteins , including cytochrome P450s , glucosyl transferases and members of the ABC transporter family , was the most strongly regulated group . Those authors concluded that a foliar spray of 10% methanol affects the expression of hundreds of genes , activating multiple detoxification and signaling pathways . Here , we show that wounding results in drastic de novo PME synthesis . The analysis of methanol in plant emissions presents serious technical challenges . To avoid the underestimation of methanol emissions , we developed a method of methanol registration based on the high solubility of methanol in water [59] . The usage of water traps in a hermetically sealed water-drop system and a flow-through system revealed a 20-fold increase in the emission of gaseous methanol 180 min after leaf injury . To clarify the role of methanol in antibacterial resistance , we examined plant susceptibility to infection with Ralstonia ( Pseudomonas ) solanacearum [63] , which causes wilt . Bacterial wilt is a devastating plant disease that affects several economically important hosts , including potatoes , tomatoes , bananas , and tobacco [64] . We showed that the methanol emitted by wounded and PME transgenic plants induced antibacterial resistance in non-wounded neighbor plants . We then identified more than three hundred methanol inducible genes ( MIGs ) that were upregulated in methanol-treated N . benthamiana plants . We further studied the function of three abundant MIGs: β-1 , 3-glucanase ( BG ) , non-cell-autonomous pathway protein ( NCAPP ) , and a previously unknown gene , designated MIG-21 . Quantitative real-time PCR ( qPCR ) analysis of mRNA from a plant treated with gaseous methanol confirmed changes in the expression of these MIGs and revealed a specific “wave” of MIG mRNA accumulation . The wave of MIG mRNA accumulation consisted of a peak followed by attenuation . We also showed that methanol and the selected MIGs ( NCAPP , MIG-21 , and β-1 , 3-glucanase ) induced an increase in the plasmodesmata ( Pd ) size exclusion limit ( SEL ) . This was demonstrated in experiments using two tandem copies of green fluorescent protein ( GFP ) ( 2×GFP ) as an indicator of Pd SEL . In addition to methanol-induced Pd gating , we also observed enhanced TMV reproduction in methanol-exposed plants and in neighbors of PME-transgenic and wounded plants . We hypothesize that methanol-mediated MIG upregulation and enhanced viral reproduction are unintended consequences of plant mobilization against bacterial pathogens .
Leaf wounding is often used as an experimental model of mechanical injuries sustained by a plant after wind , rain , hail , or herbivore feeding . However , serious leaf damage caused by , for example , crushing the leaf lamina with forceps [65] or puncturing leaves [34] has only a mild effect on PME gene expression . In nature , pathogen penetration of leaf tissue can occur via microdamage to the leaf cuticle , trichome or CW . Microdamage can be induced by wind-mediated leaf rubbing or insect attack . To test whether the expression of the endogenous PME gene is modulated by external mechanical stress , we rubbed N . benthamiana leaves with an abrasive water suspension of Celite . This approach is commonly used for plant virus inoculation . The 1 . 7-kb PME transcript was not detectable in intact leaves but was clearly induced after Celite rubbing at 1 hpi and was increased at 8 hpi ( Figure 1A ) . TMV inoculation increased the accumulation of PME transcripts , suggesting a role for viral infection in increased PME mRNA levels . We wanted to evaluate the effect of enhanced PME mRNA accumulation on methanol emission . We hypothesized that methanol from wounded leaves is produced by two forms of PME: pre-existing PME deposited in the CW before wounding , which allows rapid methanol release [34] , [60] , and PME synthesized de novo after wounding ( Figure 1A ) , which likely generates methanol for an extended period ( more than 8 h ) . Until now , the quantification of methanol emission by a plant leaf was conducted using methods based on the detection of gas-phase methanol [34] , [57]–[61] . Methanol is a polar , soluble compound that is easily lost due to condensation in sampling lines and traps . Methanol mixes readily with water , a property that we exploited by using water as a trap for methanol measurement . The methanol released by wounded leaves was measured in the headspace of either a hermetically sealed jar ( the water-drop system ) ( Figure 1B ) or a glass flow chamber ( Figure 1C ) . A drop of methanol added to the bottom of the jar will vaporize rapidly and dissolve in the water . The methanol content in the water phase may thus be used to estimate the methanol content of the leaf headspace . In the reconstruction experiment , we measured the methanol content in the water drop at different times following evaporation of various quantities of methanol that had been added to the jars . At 24°C , methanol was detected in the water drop 30 min after its addition . The water drop reached more than 80% of its saturation point after 3 h . Using calibration curves and the previously determined methanol recovery correction factors , we calculated the methanol emission of wounded leaves . Leaf wounding resulted in gaseous methanol emission , which was 20-fold higher than the methanol emission by the control intact leaf at 3 h of incubation ( Figure S1 ) . The water-drop and flow-through approaches yielded similar results for methanol emission after wounding ( Figure 1D ) . Analysis using the unpaired two-tailed Student's t-test confirmed a statistically significant difference in methanol emission between the control leaves and the wounded leaves . To determine whether methanol reabsorption might complicate our analysis , we measured the methanol content in the sap of control leaves and wounded leaves . No statistically significant methanol increase in leaf sap was detected ( Figure 1E ) . This result indicates that essentially all methanol generated by the wounded leaves was emitted into the air . Collectively , our data show that leaf wounding causes a rapid increase in the production of gaseous methanol . Biologically , wound-induced PME gene expression and the subsequent methanol emission should lead to increased resistance to pathogens , including pathogenic bacteria [26] . To determine whether the methanol emitted by wounded plants serves as a signal for antibacterial resistance , we developed an approach ( Figure 2A ) in which a wounded N . benthamiana plant ( an “emitter” ) was placed in a hermetically sealed 20-l desiccator along with an intact N . benthamiana “receiver” plant . The “receiver” plant , which had been stored adjacent to the “emitter” plant , was removed from the desiccator , and its leaves were injected with a suspension of R . solanacearum , which infects a wide range of host plants . Because both whole plants were confined within the sealed container , the available CO2 may have been depleted . Because CO2 depletion could cause several types of stress , we also tested for bacterial growth in the “receiver” plant stored together with an intact plant . Figure 2B shows that , as expected , incubation with the wounded “emitter” plants led to decreased R . solanacearum growth in the “receiver” plants ( diagram bar #3 ) compared with control plants ( diagram bar #1 ) . In control experiments , methanol evaporating from a piece of methanol-soaked filter paper also suppressed bacterial growth ( diagram bar #4 ) . We also tested N . tabacum as a “receiver” and confirmed bacterial growth suppression ( data not shown ) . We also examined whether green-leaf VOC ( GLV ) emission , which is known to be enhanced by plant wounding [34] , [66]–[68] , suppressed bacterial growth [69] . GLVs are lipoxygenase metabolic pathway products that include six-carbon aldehydes and alcohols . Unlike terpenoids , GLVs are rapidly , immediately and likely passively released from wounded leaves [70] , [71] . Our gas chromatography ( GC ) analysis confirmed the presence of methanol ( Figure 3A ) . In line with the data of von Dahl et al . [34] , our GC analysis revealed that cis-3-hexen-1-ol is emitted in the headspace of wounded leaves ( Figure 3C ) . We did not detect methyl salicylate or methyl jasmonate in the headspace of wounded leaves ( data not shown ) . Ethylene emission was detected , but there was no statistically significant difference in ethylene emission between the control and wounded leaves ( Figure 3B ) . Thus , the suppression of R . solanacearum growth observed in the “receiver” plants could be caused by gaseous methanol or by GLV . Indeed , cis-3-hexen-1-ol evaporated in the desiccator also resulted in decreased bacterial growth in target plants ( Figure 2B , diagram bar #5 ) . However , GLVs rapidly released from wounded leaves may stimulate PME-generated methanol production , and their influence on bacterial growth may thus be indirect . To examine the role of cis-3-hexen-1-ol in the emission of methanol from leaves , we measured the methanol content in a water trap system in which an N . benthamiana leaf was exposed to continuous airflow from an evaporator containing cis-3-hexen-1-ol for 3 h ( Figure 4 , upper ) . The diagram ( Figure 4 , bottom ) shows that the methanol content in the water trap increased after cis-3-hexen-1-ol treatment . We suggest that methanol emission induced by GLV may be responsible for the suppression of R . solanacearum growth in “receivers” . To further refine the role of methanol , GLV was excluded from the gaseous mixture emitted by wounded leaves . We used the previously engineered PME-transgenic tobacco line , pro1 , which has increased PME gene expression and resistance to TMV [52] . Transgenic plants produced higher levels of methanol in the leaf sap than did the control plants ( Figure S2 ) . Consistent with our expectations , the increased PME gene expression in the transgenic plants also resulted in a higher production of gaseous methanol , whereas cis-3-hexen-1-ol was not detected ( Figure 3C ) . To determine whether the methanol emitted by PME-transgenic plants serves as a signal for antibacterial resistance , we employed a hermetically sealed desiccator . Incubation with the PME-transgenic “emitter” plants slowed the growth of R . solanacearum compared to the control plants ( Figure 2B , diagram bar #2 ) . Although the retardation of R . solanacearum growth caused by a neighboring PME-transgenic plant was less than that caused by a wounded plant , the reduction in growth correlated with the level of methanol emission ( Figure 3A ) . The unpaired two-tailed Student's t-test confirmed the statistical significance of the differences in R . solanacearum growth retardation between the “receivers” of control and PME transgenic plant ( Figure 2B , diagram bar #2 ) . To further clarify the role of methanol as an airborne signal of antibacterial resistance , we again used a flow-through system that allows continuous airflow from PME-transgenic or wounded tobacco plants to intact target N . benthamiana plants ( Figure 5A ) . Control plants were exposed to air from a desiccator containing intact N . tabacum plants . After exposure , the target “receiver” plants were inoculated with R . solanacearum . “Receiver” plants exposed to air from the desiccator with evaporated methanol , PME-transgenic or wounded plants acquired antibacterial resistance ( Figure 5B ) . The evaporation from the wounded PME-transgenic plants had even greater effect on antibacterial activity . The unpaired two-tailed Student's t-test confirmed the statistical significance of the differences in decreased R . solanacearum growth . Collectively , these data indicate that gaseous methanol is an airborne factor that may induce antibacterial resistance in neighboring plants . In an effort to understand the mechanisms by which methanol can stimulate antibacterial resistance in “receiver” plants , we constructed forward and reverse suppression subtractive hybridization ( SSH ) cDNA libraries from N . benthamiana plants exposed to methanol . A total of 359 differentially expressed transcripts were identified; of these , 39 appeared to be more abundant in intact leaves , and 320 appeared to be upregulated after methanol treatment ( Table S1 ) . The cloned ESTs of genes that responded to the methanol treatment were considered for sequencing . The EST sequences of the upregulated genes were deposited in the NCBI dbEST database with accession numbers . Most of the ESTs identified ( Table S1 ) ( i . e . , 167 ) fell into the category of stress gene transcripts . We identified only one novel EST ( FN432041 ) , methanol-inducible gene-21 ( MIG-21 ) ( GenBank AC GU128961 ) , which was unrelated to all other nucleotide sequences in GenBank . MIG-21 contains an ORF encoding a protein with a repetitive amino acid sequence ( Figure S3 ) . The methanol-specific upregulation of the SSH-identified genes was validated by a Northern blot analysis hybridized with 32P-labeled probes , which were prepared from randomly selected differential clones that were found by differential screening . We selected and isolated the most abundant SSH-identified genes for further analysis ( Table 1 ) [72]–[76] . We validated the changes in gene expression observed by SSH by performing quantitative real-time PCR ( qRT-PCR ) to determine the mRNA levels from a plant treated with gaseous methanol . MIG mRNA accumulation depended on both the methanol concentration ( Table 2 ) and the length of treatment ( Figure S4 ) . PME is not likely to be a MIG because its mRNA accumulation was not significantly altered after methanol treatment . The level of BG mRNA accumulation increased with time , up to 400-fold after 18 h ( compared with the untreated control , Table 2 ) . The accumulation of the NCAPP mRNA ( GenBank AC FN432039 ) increased by almost 50-fold at 18 h and the level of NCAPP mRNA accumulation was the highest compared to BG and MIG-21 after 6 h of treatment ( Figure S4 ) . Our model proposes that a burst of methanol from wounded leaves should elicit an extended MIG induction in neighboring leaves . We exposed N . benthamiana plants to methanol vapors ( 160 mg ) applied to filter paper within a sealed 20-l desiccator for 3 h . RNA for qRT-PCR analysis was isolated from leaves at different times after the plant was withdrawn from the methanol atmosphere . Figure 6 shows the decaying wave of MIG mRNA accumulation after methanol treatment . MIGs mRNA accumulation reached a maximum at 24 h after methanol treatment and decreased slowly thereafter . Moreover , increased BG and NCAPP mRNA levels were observed as long as 5 days after methanol treatment . The suppression of R . solanacearum growth in “receiver” plants in a sealed desiccator ( Figure 2B ) suggests that MIGs may be involved in plant antibacterial resistance . We examined the accumulation of MIGs mRNA in N . benthamiana “receivers” that were kept together with wounded WT or PME-transgenic tobacco plants in a sealed desiccator ( Figures 7 A , B ) . The unpaired two-tailed Student's t-test confirmed the statistical significance of the differences in MIG induction between the “receivers” of control and wounded or PME-transgenic plants . Constant PME expression and increased methanol production in PME-transgenic tobacco was predicted to result in increased MIG mRNA accumulation . Indeed , RNA analysis of PME transgenic leaves ( Figure 8 ) confirmed this expectation , though the general profile was different from that of methanol-treated plants . The nearly 70-fold increase observed in PI-II mRNA accumulation is likely to be a response to long-term PME overproduction . It has been demonstrated previously [23] that several plant species emit VOCs , including ethylene , methyl salicylate , methyl jasmonate , and cis-3-hexen-1-ol , in response to pathogen attack and plant damage . In “receiver” plants , the emitted VOCs can upregulate PR genes , such as the basic type PR-3 ( chitinase ) , acidic type PR-4 ( thaumatin-like ) , lipoxygenase ( LOX ) , phenylalanine ammonia-lyase ( PAL ) , and farnesyl pyrophosphate synthetase ( FPS ) . We studied gene expression in plants treated with methanol and compared those results to the gene expression of plants treated with the VOCs listed above . As shown in Figure S5 , the expression of LOX , PR-3 , PR-4 , FPS and PAL genes increased slightly in methanol-treated plants . Treatment with cis-3-hexen-1-ol stimulated the accumulation of FPS mRNA , but ethylene , methyl salicylate , and methyl jasmonate treatment primarily upregulated the PAL and PR-4 mRNAs accumulation . Thus , the methanol emitted from a wounded plant most likely potentiates the antibacterial resistance of neighboring plants by increasing the MIG mRNA accumulation . Bacterial pathogens do not cross plant cell wall boundaries because they inhabit the intercellular spaces in plants . In contrast , viral pathogens require intercellular movement for local and systemic spread [16] . However , plasmodesmata ( Pd ) play an important role in both bacterial effector molecule spreading and host defense responses [77] . To evaluate cell-to-cell communication in leaves treated with methanol , a reporter macromolecule was used to test movement through Pd in different states of dilation . We chose a reporter containing two fused green fluorescent proteins ( 2×GFP ) to query the non-targeted Pd transport of macromolecules [78] . Mature source leaves have generally been considered closed to 2×GFP ( 54 kDa ) because their Pd size exclusion limit ( SEL ) does not permit proteins with a size of 47 kDa [79] . To establish a system to monitor cell-to-cell transit , we exploited an “agroinjection strategy” to deliver the 2×GFP plasmid into the cell nucleus [80] , [81] . To monitor single infection sites , N . benthamiana plants were agroinjected with a diluted ( 1∶1000 ) bacterial suspension . Plants were then exposed to methanol vapors and examined by fluorescent light microscopy 30 h after agroinjection . Counting the number of epidermal cells surrounding the initial Agrobacterium-transformed cell that display fluorescence provides a quantitative measure of 2×GFP movement . When the Pd were closed , 2×GFP was detected mainly in single cells ( Figure 9A , upper ) . However , fluorescent signals were distributed in 2- or 3-cell clusters ( Figure 9A , bottom ) when Pd were dilated . In the control plant , approximately 6% of the signal was distributed in 2- to 3-cell clusters ( Figure 9B ) . These observations were consistent with the known rate of 2×GFP movement through plant tissues [78] . When methanol-treated plants were examined , more than 20% of the signal was distributed in 2 - to 3-cell clusters , indicating that the ability to support cell-to-cell movement of 2×GFP was enhanced . Specifically , whereas only 1% of the signal was found in 3 cell-clusters in the control leaves , with methanol-treatment , this value was increased up to 7% . The unpaired two-tailed Student's t-test confirmed the statistical significance of the differences in the cell-to-cell movement of 2×GFP between the control plant and plants treated with methanol ( Figure 9B ) . Collectively , these data indicate that methanol acts as a signal that facilitates the movement of 2×GFP between cells . To examine the role of MIGs in Pd dilation ( gating ) , we monitored the relative cell-to-cell spreading of 2×GFP within the epidermis of N . benthamiana leaves co-agroinjected with binary plasmids encoding BG , NCAPP or MIG-21 directing the synthesis of the respective mRNAs , as tested by qRT-PCR ( data not shown ) . Figure 9C shows that in the control leaves , which were co-agroinjected with an empty Bin19 vector , approximately 7% of the signal was distributed in 2- to 3-cell clusters . When leaves were co-agroinjected with NCAPP , MIG-21 or BG , the movement of 2×GFP was enhanced: more than 23 , 26 or 22% of the 2×GFP signal was detected in cell clusters . The unpaired two-tailed Student's t-test confirmed the statistical significance of the differences in the cell-to-cell movement of 2×GFP between the vector-only control and leaves co-agroinjected with NCAPP , MIG-21 or BG ( Figure 9C ) . Collectively , these data imply that gaseous methanol may trigger leaf Pd dilation ( gating ) by inducing the mRNA accumulation of MIGs such as NCAPP , MIG-21 and BG . Our model suggests that methanol-triggered Pd dilation should enhance viral spread within the plant . To examine this possibility , we inoculated plants with a crTMV binary vector that carries an autofluorescent tag GFP ( crTMV:GFP ) in the place of its coat protein gene [82] and treated the transfected plants with methanol , as shown in Figure 10A . Figure 10B shows the quantification of GFP foci in leaves at 3 dpi . Methanol treatment reduced the number of GFP foci per cm2 presumably due to the induction of antibacterial resistance , which was consistent with our data showing that methanol exposure inhibited R . solanacearum growth ( see Figure 2B ) . Importantly , the stimulation of local viral movement by methanol was indicated by the appearance and spread of the GFP signal . Figure 10C shows that while viral foci became visible in all plants approximately at the same time ( 3 dpi ) after inoculation , viral reproduction representing viral RNA replication and RNA cell-to-cell movement occurred more rapidly in the methanol-treated plants than in the control plants . Figure 10D summarizes the results of the statistical analysis of the data , with the horizontal red lines across the boxes representing the median size of the GFP expression foci ( µm2×104 ) . ANOVA confirmed the statistical significance of the differences in focus size between the control and methanol-treated leaves ( P = 0 . 005 ) . Because BG , NCAPP and MIG-21 can enhance cell-to-cell movement , they may also increase viral RNA movement and/or replication . Therefore , BG , NCAPP and MIG-21 may increase TMV-directed GFP accumulation due to viral reproduction . We tested this hypothesis using crTMV:GFP and binary vectors encoding BG , NCAPP and MIG-21 through co-agroinjection of N . benthamiana leaves . At five days after co-agroinjection with vectors encoding BG , NCAPP and MIG-21 , the GFP accumulation in whole leaves increased by 13–23 fold ( Figure 10E ) . These results suggest that BG , NCAPP and MIG-21 enhance viral reproduction . A change in the accumulation of GFP expressed from the viral vector can be caused by a change in viral RNA movement and/or a change in viral replication . Under natural conditions , viral RNA directly enters the cytoplasm of a negligible number of cells following leaf wounding . Agrobacterium-delivered plant viral vectors exploit the host RNA polymerase II–mediated nuclear export system , which includes 5′-end capping , splicing and 3′-end formation [83] . To test whether methanol or vapors from wounded plants can enhance viral reproduction in TMV-inoculated leaves , we used an experimental design that mimicked the natural condition of viral entry , excluding Agrobacterial participation . In contrast to controls , plants incubated with wounded N . benthamiana in a hermetically sealed desiccator exhibited increased sensitivity to TMV , as reflected by TMV RNA accumulation ( Figure 11A ) . The same effect occurred when methanol was evaporated in the desiccator . The unpaired two-tailed Student's t-test confirmed the statistical significance of the differences in TMV RNA accumulation between the “receivers” of intact plants , wounded plants or methanol . The use of the flow-through system to provide continuous airflow from wounded N . benthamiana plants to intact target N . benthamiana plants ( Figure 11B , upper panel ) confirmed the results of experiments with the hermetically sealed desiccator . “Receiver” plants exposed to air from the desiccator containing wounded plants acquired increased sensitivity to TMV in comparison to control plants ( Figure 11B , bottom ) . The statistical significance of the differences in TMV RNA accumulation in the inoculated ( 48 and 72 h after TMV inoculation ) or systemically infected leaves ( 120 h after TMV inoculation ) between the “receivers” of intact or wounded plants were confirmed by the unpaired Student's t-test . These data indicate a role for methanol in triggering MIG expression , which leads to enhanced viral spread and/or reproduction .
The amazing capacity of plants to recognize pathogens through strategies that involve both conserved and variable pathogen elicitors has been previously reported [5] , [84] , [85] . However , the molecular mechanism by which plants protect themselves against bacterial pathogens remains obscure . This is mainly due to a lack of knowledge about the long-distance signals that trigger systemic reactions in plants . One recent study suggested that a long-range factor , GLV , may increase resistance to the bacterial pathogen Pseudomonas syringae [68] . Here , we characterized another VOC , methanol , which induces a protective reaction against R . solanacearum . Methanol is a natural plant product that accumulates in the leaf tissue and is emitted when the stomata open in the morning [57] , [58] . Our data reveal that leaf wounding stimulates additional methanol emission . Five aspects of wound-stimulated methanol production are especially interesting . First , there is a direct correlation between de novo PME synthesis and methanol emission ( Figure 1A , D ) . We observed a 20-fold increase in the emission of gaseous methanol at 3 h after leaf damage in comparison to the methanol emission by intact control leaves ( Figure 1D ) . Second , methanol generated by de novo synthesized PME is released into the air but does not accumulate in leaf tissue or sap ( Figure 1E ) . Third , gaseous methanol upregulates methanol-inducible genes ( MIGs ) in the leaves of neighboring plants ( Figures 6 , 7 ) . Fourth , methanol induces antibacterial resistance ( Figures 2 , 5 ) . Fifth , although virus entry per se induces PME mRNA accumulation ( Figure 1A ) , gaseous methanol drastically increases the TMV sensitivity of non-wounded leaves ( Figure 11 ) . We suggest the following model to explain the mechanism of the observed phenomenon ( Figure 12 ) . Microdamage ( Figure 12 , step 1 ) to the leaf caused by wind-induced leaf rubbing , human handling or insect attack , results in the upregulation of the PME gene ( Figure 12 , step 2 ) . Upregulation of the PME gene leads to at least three events . First , PME triggers defense reactions that provide resistance against bacteria and viruses; i . e . , wound mediated PME mRNA accumulation may promote the defense reactions described earlier [52] . It is worth to emphasize that a model for a mechanical damage – transgenic tobacco overexpressing PME – is resistant to R . solanacearum ( see Figure S6 and Table S2 ) . Second , PME enzymatic activity increased 2 . 5-fold in N . benthamiana leaves at 3 h after wounding ( 139±9 . 2 vs . 360±0 . 068 nkat/mg ) . Third , PME catalyzes the production of gaseous methanol ( Figure 12 , step 3 ) , which induces the MIG mRNA accumulation ( Figure 12 , step 4 ) . Gaseous methanol may provide a feedback loop and suppress PME transcription ( Figure 12 , step 5 ) such that the leaf returns to its pre-wounding methanol production state . PMEi is likely to take part in this process by suppressing PME enzymatic activity [75] . MIGs are responsible for TMV spreading/reproduction and resistance to R . solanacearum ( Figure 12 , step 6 ) . It was previously shown that transgenic tobacco with elevated PME synthesis is resistant to TMV [52] . This strain exhibits increased methanol emission levels and MIG expression but is not susceptible to TMV . We can consider the effects in PME-transgenic plants to be a consequence of long-term ( even “lifelong” ) MIG induction , which clearly differs from the effects of short-term methanol treatment . These cases are thus examples of ‘chronic’ and ‘acute’ situations , respectively . The patterns of MIG expression in these two cases are similar ( i . e . , activated ) but still very different ( compare Figures 6 and 8 ) . Methanol treatment elicits a MIG “wave” that eventually fades , while MIG expression in PME-transgenic plants is always slightly elevated , which might lead to some secondary effects . Moreover , the expression of PME is much higher in PME-transgenic plants than in methanol-treated plants . We believe that this increased PME expression , which is absent in methanol-treated plants , makes the PME-transgenic plant resistant to TMV . Methanol is not a plant poison . Treatment of plants with high-concentration methanol solutions ( 5–50% ) revealed that foliar sprays of aqueous methanol , even at a concentration of 50% , led to increased growth and development in C3 crop plants in arid environments [86] . This is likely to be the result of more effective utilization of light energy during photosynthesis [87] . Previously , foliar sprays of a 10% methanol solution were used to identify methanol-sensitive genes in Arabidopsis thaliana [62] . Methanol affected the expression of hundreds of genes , and multiple detoxification and signaling pathways were activated . We used gaseous methanol at physiological concentrations , which were likely 10 , 000 times lower than those used by Downie et al . [62] . This difference in methanol concentration may explain why we observed the upregulation of only a few previously identified genes ( see Table S1 ) . Most of the MIGs identified here ( 167 ESTs ) were classified as stress response genes . The majority of these ( 117 clones ) represented 6 of the most up-regulated SSH-identified genes: BG , PI-II , MIG-21 , PMEi , elicitor inducible protein and 1-aminocyclopropane-1-carboxylic acid oxidase , the latter of which is involved in ethylene biosynthesis [88] . The NCAPP transcript was represented by only three clones ( Table S1 ) ; however , qRT-PCR verification ( Table 2 ) showed that the NCAPP gene was highly inducible , the second most inducible after BG . The SSH approach did not identify the LOX , PR-3 , PR-4 , FPS or PAL genes , which are induced by other VOCs ( Figure S5 ) . Pathogen attack and plant damage accompanied by the emission of VOCs , including ethylene [17] , methyl salicylate [18] , methyl jasmonate [19] , [20] , nitric oxide [21] , [22] and cis-3-hexen-1-ol [23] , leads to the upregulation of different PR genes [14] , [23] , [24] . In addition to methanol , we detected the emission of ethylene and GLV . Ethylene is a simple gaseous hormone that integrates external signals with internal processes . Wound-induced ethylene production has been studied thoroughly [89] . The two-step ethylene biosynthesis , i . e . , the conversion of S-adenosyl-L-methionine to 1-aminocyclopropane-1-carboxylic acid ( ACC ) and its subsequent oxidation to ethylene , is regulated by ACC synthase ( ACS ) and ACC oxidase ( ACO ) , respectively . ACS and ACO are encoded by members of multi-gene families [90]–[92] . Ethylene production is regulated by different isoforms of ACO and ACS in response to different stresses [93] . For example , the accumulation of the transcripts of 3 out of 4 members of the ACO gene family has been examined in tomato , and only ACO1 was wound-responsive [90] . Our SSH approach revealed 6 ACO clones in leaves treated with methanol . We also showed that leaf wounding or PME overexpression ( Figure 3 ) did not increase ethylene emission as a secondary response to methanol . This contradiction might be explained by previous data indicating that ACC synthase , but not ACO , is rate-limiting in ethylene biosynthesis [94] . We have not detected ACC synthase gene upregulation ( Table S1 ) . Moreover , it has been shown that increased ACO activity does not always immediately lead to parallel changes in ethylene production [95] , e . g . , in stress response ( methylviologen , oxidative stress inductor , or methyl jasmonate ) . We hypothesize that methanol might be a similar stimulus , affecting ACO but not ethylene synthesis when applied at physiological concentrations . On the other hand , ethylene biosynthesis is regulated by different isoforms of ACO in response to particular stress cues [93] . Finally , the antibacterial effects of methanol were demonstrated not only in sealed desiccators but also in a flow-through system ( Figure 5 ) in which methanol was blown out . Therefore , the effects of virtual ethylene were excluded or at least significantly diminished . We also detected cis-3-hexen-1-ol as a representative of GLV emission after plant wounding ( Figure 3 ) . The suppression of R . solanacearum growth observed in the “receiver” plants could be caused by gaseous methanol and GLV . This was confirmed in experiments in which cis-3-hexen-1-ol evaporated in the desiccator resulted in decreased bacterial growth in target plants ( Figure 2B , diagram bar #5 ) . In an attempt to elucidate the mechanism underlying this phenomenon , we discovered that GLVs rapidly released from wounded leaves stimulate PME mRNA accumulation and therefore PME-generated methanol emission . In our experiments with detached N . benthamiana leaves incubated for 3 h in a 300-ml sealed container with cis-3-hexen-1-ol ( 0 . 36 µg ) , the level of PME mRNA increased by more than two times ( 2 . 41±0 . 37 ) in comparison to water control ( 1 . 00±0 . 25 ) . Taking into account the connection between cis-3-hexen-1-ol , PME and methanol emission , we believe that the effect of cis-3-hexen-1-ol on bacterial growth is indirect . Antibacterial resistance accompanied by MIG upregulation is likely to be related to PI-II gene transcription induction . Type I proteinase inhibitors are powerful inhibitors of serine endopeptidases in animals and microorganisms [96] . The PI-II gene is not expressed in the leaves of healthy plants , but it is induced in leaves that have been subjected to different types of stress , including wounding and bacterial infection [76] . R . solanacearum encodes several secreted proteases [97] , [98] , including a type III effector , PopP2 , which mimics a plant transcriptional activator and manipulates the plant transcriptome [99] , [100] . PME-transgenic tobacco with high levels of PI-II expression ( Figure 8 ) demonstrated high resistance to R . solanacearum ( Figure S6 and Table S2 ) . This finding supports the role of PI-II in the suppression of bacterial proteases . To determine whether BG , MIG-21 and NCAPP could enhance cell-to-cell communication , we used Agrobacterium to mediate the delivery of GFP- and MIG-expressing vectors . Although methanol treatment induced resistance against bacteria ( Figure 10B ) and therefore decreased the number of detected crTMV:GFP foci , we found that these foci increased in size ( Figure 10 C , D ) . Methanol changes Pd SEL and upregulates MIGs ( BG and NCAPP ) ; therefore , it is likely to promote cell-to-cell trafficking and TMV reproduction . The participation of MIG-21 in cell-to-cell trafficking is unconfirmed , but the role of BG and NCAPP in Pd dilation has been described previously [74] , [101] , [102] . However , there is no data explaining the correlation between methanol-mediated BG , NCAPP , MIG-21 upregulation and antibacterial resistance . A recently revealed link between nuclear transport and cell-to-cell movement [103] suggests that there may be competition between methanol-mediated cell-to-cell transport and R . solanacearum type III effector nuclear traffic . We cannot exclude the possibility that airborne signals from wounded leaves may also facilitate TMV spreading/reproduction in neighbors as an unintended consequence of the acquired antibacterial resistance of the plants . Interestingly , it has been suggested that the conditions generated by agriculture during the Holocene period may have promoted viral spreading in plants [104] . Further research is required to elucidate the mechanisms of the reactions triggered by methanol in plants . How methanol is regulated during wound stress conditions remains unclear , as do the identities of possible factors involved in this process . The involvement of MIGs in viral spreading has been clearly demonstrated . However , the underlying cellular mechanisms controlling the targeting of BG , NCAPP and MIG-21 to the Pd is still unknown . Finally , the factors that coordinate the spatiotemporal correlation of MIGs with bacterial resistance and viral cell-to-cell spreading and reproduction have yet to be determined .
N . benthamiana and N . tabacum plants were grown in soil in a controlled environment chamber in a 16 h/8 h day/night cycle . Full-length BG ( β-1 , 3-glucanase ) , MIG-21 and NCAPP cDNAs were obtained by PCR using the primer pairs 5′GAGCTCATGTCTACCTCACATAAACATAATAC3′/5′AAGCAGTGGTAACAACGCAGAGTACtttttttttttttttttttttttttttttt3′ , 5′GAGCTCATGGCATCACTTCAGTGCC3′/5′CTGCAGTCAGCAGCTCCCTCTATTC3′ and 5′GAGCTCATGTCTTCAAAGATTGGTCTG3′/5′CTGCAGCTATTTCTTGATAGAAAACGTG3′ , respectively , with total N . benthamiana cDNA as the template . The viral vector crTMV:GFP ( pICH4351 ) has been described previously [105] . To synthesize the 35S-based binary vectors pBG , pMIG-21 and pNCAPP , PCR-amplified cDNA was inserted into the XbaI , EcoRI ( pBG ) or SacI , PstI ( pMIG-21 and pNCAPP ) sites of pBin19 . The methanol , cis-3-hexen-1-ol and ethylene contents were determined by GC on a capillary FFAP column ( 50 m×0 . 32 mm; Varian Inc . , Lake Forest , CA , USA ) in a Kristall 2000 gas chromatograph ( Eridan , Russia ) . The methanol and cis-3-hexen-1-ol in water/decane samples were measured under the following operating conditions: carrier gas – nitrogen , nitrogen flow – 30 ml/min; air flow – 400 ml/min; hydrogen flow – 40 ml/min; injected volume – 1 µl; injector temperature – 160°C; column temperature – 75°C , increased to 150°C at a rate of 15°C/min; retention time – 6 . 5 min ( methanol ) or 17 min ( cis-3-hexen-1-ol ) ; and flame ionization detector temperature – 240°C . Ethylene content in air samples was analyzed under the following operating conditions: carrier gas – nitrogen; nitrogen flow – 30 ml/min; air flow - 400 ml/min; hydrogen flow – 40 ml/min; injected volume – 1 ml of vapor phase; injector temperature – 130°C; column temperature – 45°C; retention time – 4 . 5 min; and flame ionization detector temperature – 240°C . Methanol treatment was executed by exposing plants to methanol vapors on filter paper in a sealed desiccator . The effects of plant VOCs were measured in either a single hermetically sealed 20-l desiccator or a flow-through set-up involving two attached 20-l desiccators ( the first for the “emitter” plants and the second for the “receiver” plants ) supplied with filtered air at a rate of 0 . 15 l/min . Intact and wounded N . tabacum or PME-transgenic tobacco plants were used as “emitter” plants , whereas N . benthamiana plants were used as “receivers” . Pots ( width , 9 . 5 cm; depth , 9 . 5 cm ) containing plants ( 10 . 0±1 . 0 g ) and soil ( 198 . 0±20 . 0 g ) were placed into desiccators and maintained for 3 h or 18 h at a constant temperature of 24°C with a 16 h/8 h light/dark photoperiod . Then , “receiver” plants were withdrawn from the desiccator and tested for MIG RNA accumulation and bacterial and TMV resistance . In experiments assessing the decay of MIG mRNA accumulation , the plants withdrawn from the desiccator after methanol treatment were kept at 24°C with a 16 h/8 h light/dark photoperiod for leaf RNA isolation . The tobacco strain R . solanacearum was grown under routine conditions on yeast–peptone–glucose ( YPG ) agar containing the following ( per liter ) : 5 g yeast extract , 10 g peptone , 5 g glucose and 15 g agar . The incubation temperature was 28°C . Overnight cultures of R . solanacearum at the indicated concentrations in 10 mM MES ( pH 5 . 5 ) buffer supplemented with 10 mM MgCl2 were injected into fully developed leaves by syringe . At four days post inoculation ( dpi ) , bacterial growth was measured by macerating five leaf discs of 1 cm2 from the inoculated tissue of each sample in 10 mM MgCl2 , plating the serial dilutions on nutrient agar plates , and counting the colony-forming units ( cfu ) . Agrobacterium tumefaciens strain GV3101 was transformed with individual binary constructs and grown at 28°C in LB medium supplemented with 50 mg/l rifampicin , 25 mg/l gentamycin and 50 mg/l carbenicillin/kanamycin . Agrobacterium cells from an overnight culture ( 5 ml ) were collected by centrifugation ( 10 min , 4 , 500× g ) , resuspended in 10 mM MES ( pH 5 . 5 ) buffer supplemented with 10 mM MgSO4 and adjusted to a final OD600 of 0 . 2 for TMV-directed GFP accumulation or 0 . 001 for cell-to-cell movement assays . Agroinjection was performed on almost fully expanded N . benthamiana leaves that were still attached to the intact plant . A bacterial suspension was infiltrated into the leaf tissue using a 2 ml syringe , after which the plants were grown under greenhouse conditions at 24°C with a 16 h/8 h light/dark photoperiod . In the cell-to-cell-movement assay , N . benthamiana plant leaves were agroinjected with 2×GFP and were stored for 6 h in a plant growth chamber at 24°C with light; these plants were then loaded into the desiccator . Then , methanol ( 160 mg ) was added , and the desiccator was sealed . After a 3-h exposure to methanol vapors , the plants were withdrawn , and fluorescent cells were counted after 21 h of storage in a growth chamber . In the viral focal growth experiments assessing TMV cell-to-cell spreading , N . benthamiana plant leaves were agroinjected with crTMV:GFP , stored for 6 h in a plant growth chamber , and then loaded in the desiccator . Subsequently , methanol ( 160 mg ) was added , and the desiccator was sealed . After a 3-h exposure to the methanol vapors , the plants were withdrawn . Fluorescent cells were counted after 4 days of storage in a growth chamber at 24°C with a 16 h/8 h light/dark photoperiod . GFP fluorescence in the inoculated leaves was monitored by illumination with a handheld UV source ( DESAGA ) . At higher magnifications , GFP fluorescence was detected using a dissecting microscope ( Opton IIIRS ) equipped with an epifluorescence module . Unless otherwise indicated , the lower epidermal cells of injected leaves were observed at 24 or 72 h after agroinfiltration . 50 mg of leaf tissue from infiltrated areas were ground in the 1 . 5 ml tubes in 200 µl of GFP-extraction buffer ( 150 mM NaCl , 10 mM Tris-HCl , pH 8 . 0 ) . Then the samples were centrifuged 16 000× g 10 min and 1 ml of GFP-extraction buffer was added to the supernatant . The fluorescence was measured using Quantech fluorometer ( ThermoScientific , USA ) . Plant material was ground to a fine powder in liquid nitrogen using a mortar and pestle . Total RNA was extracted from leaves using TRIzol reagent ( Invitrogen ) . Approximately 5 µg of total nucleic acid isolated from mock-treated or virus infected leaves was denatured , separated in 1 . 5% agarose gels containing 10% formaldehyde in MOPS buffer , pH 7 . 0 , and transferred to a nylon membrane ( Hybond-N+ , Amersham ) . Membranes were incubated in a pre-hybridization solution containing 6× SSC , 0 . 5% SDS , 5× Denhardt's reagent and 200 µg/ml tRNA for 4 h at 68°C and probed with a denatured DNA fragment containing the PME coding sequence . Probes were labeled with [α32P]-dATP ( 3000 Ci/mmole ) in a PCR reaction . N . benthamiana plants withdrawn from the desiccator after exposure to methanol were mechanically inoculated with TMV virions ( 100 µg/ml ) in 50 mM sodium phosphate buffer , pH 7 . 0 , in the presence of Celite , as described previously [106] . Concentrations were determined using a Nanodrop ND-1000 spectrophotometer ( Isogen Life Sciences ) . All RNA samples had a 260∶280 absorbance ratio between 1 . 9 and 2 . 1 . After DNAse treatment ( Fermentas ) , 2 µg of denatured total RNA was annealed with 0 . 1 µg of random hexamers and 0 . 1 µg of Oligo-dT and incubated with 200 units of Superscript II reverse-transcriptase ( Invitrogen , USA ) for 50 min at 43°C to generate cDNA . Real-time qPCR was carried out using the iCycler iQ real-time PCR detection system ( Bio-Rad , Hercules , CA , USA ) . Target genes were detected using Eva Green master mix ( Syntol , Russia ) according to the manufacturer's instructions . The thermal profile for EVA Green real-time qPCR included an initial heat-denaturing step at 95°C for 3 min and 45 cycles with a denaturation step at 95°C for 15 s , an annealing step ( amplicon-specific temperatures provided in Table S4 ) for 30 s and an elongation step at 72°C for 30 s coupled with fluorescence measurement . Following amplification , the melting curves of the PCR products were monitored from 55 to 95°C to determine the specificity of amplification . Each sample was run in triplicate , and a non-template control was added to each run . Target gene mRNA levels were calculated according to the equation proposed by Pfaffl [111]: EtargetΔCt target ( sample-reference ) . PCR efficiency ( E ) was calculated according to the equation E = 10 ( −1/slope ) by performing the standard curves . Target gene mRNA levels were normalized to the corresponding reference genes ( 18S and ef-2ά for N . tabacum ) . Student's t-tests were performed using Excel ( Microsoft , Redmond , WA ) . ANOVA tests were performed using SPSS v . 18 ( IBM Corporation , Somers , NY ) . P-values<0 . 05 were considered significant . | The mechanical wounding of plant leaves , which is one of the first steps in pathogen infection and herbivore attack , activates signal transduction pathways and airborne signals to fend off harmful organisms . The mechanisms by which these signals promote plant immunity remain elusive . Here , we demonstrate that plant leaf wounding results in the synthesis of a cell wall enzyme , pectin methylesterase ( PME ) , causing the plant to release methanol into the air . Gaseous methanol or vapors from wounded PME-transgenic plants induced resistance to the bacterial pathogen Ralstonia solanacearum in the leaves of non-wounded neighboring “receiver” plants . To investigate the mechanism underlying this phenomenon , we identified the methanol inducible genes ( MIGs ) in Nicotiana benthamiana , most of which fell into the category of defense genes . We selected and isolated the following genes: non-cell-autonomous pathway protein ( NCAPP ) , β-1 , 3-glucanase ( BG ) , and the previously unidentified MIG-21 . We demonstrated that BG , MIG-21 and NCAPP could enhance cell-to-cell communication and Tobacco mosaic virus ( TMV ) RNA accumulation . Moreover , gaseous methanol or vapors from wounded plants increased TMV reproduction in “receivers” . Thus , methanol emitted by a wounded plant enhances antibacterial resistance as well as cell-to-cell communication that facilitate virus spreading in neighboring plants . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"agriculture",
"biology"
]
| 2012 | Airborne Signals from a Wounded Leaf Facilitate Viral Spreading and Induce Antibacterial Resistance in Neighboring Plants |
SIVmac239 infection of rhesus macaques ( RMs ) results in AIDS despite the generation of a strong antiviral cytotoxic T lymphocyte ( CTL ) response , possibly due to the emergence of viral escape mutants that prevent recognition of infected cells by CTLs . To determine the anatomic origin of these SIV mutants , we longitudinally assessed the presence of CTL escape variants in two MamuA*01-restricted immunodominant epitopes ( Tat-SL8 and Gag-CM9 ) in the plasma , PBMCs , lymph nodes ( LN ) , and rectal biopsies ( RB ) of fifteen SIVmac239-infected RMs . As expected , Gag-CM9 did not exhibit signs of escape before day 84 post infection . In contrast , Tat-SL8 escape mutants were apparent in all tissues by day 14 post infection . Interestingly LNs and plasma exhibited the highest level of escape at day 14 and day 28 post infection , respectively , with the rate of escape in the RB remaining lower throughout the acute infection . The possibility that CTL escape occurs in LNs before RBs is confirmed by the observation that the specific mutants found at high frequency in LNs at day 14 post infection became dominant at day 28 post infection in plasma , PBMC , and RB . Finally , the frequency of escape mutants in plasma at day 28 post infection correlated strongly with the level Tat-SL8-specific CD8 T cells in the LN and PBMC at day 14 post infection . These results indicate that LNs represent the primary source of CTL escape mutants during the acute phase of SIVmac239 infection , suggesting that LNs are the main anatomic sites of virus replication and/or the tissues in which CTL pressure is most effective in selecting SIV escape variants .
Human immunodeficiency virus ( HIV ) infection of humans and Simian Immunodeficiency Virus ( SIV ) infection of rhesus macaques ( Macaca mulatta , RM ) results in a progressive and irreversible decline of immune function characterized by depletion of CD4 T cells , chronic immune activation , and high susceptibility to opportunistic infections that is commonly referred to as AIDS . While the host immune system mounts strong cellular and humoral immune responses against HIV and SIV , these responses ultimately fail to control virus replication in the overwhelming majority of infected individuals . A key reason underlying this immune failure is the extreme genetic variability of these primate lentiviruses , which occurs as a result of a high mutation rate caused by the relative infidelity of the HIV and SIV reverse transcriptases [1] . This extreme genetic diversity combined with a large in vivo effective population size [2] practically ensures that the virus will always be able to evade or “escape” from recognition by the host immune system . In clinical terms , these biological features of HIV in the absence of “natural immunity” against the virus are key indicators of the complexity and difficulty that the scientific community faces when trying to design an effective AIDS vaccine . In the absence of immunogens that are able to predictably elicit the production of broadly reactive HIV- or SIV-specific neutralizing antibodies [3] , [4] , there has been significant interest in immunogens that elicit strong antiviral CD8+ T cell-mediated cytotoxic T lymphocyte ( CTL ) responses [5] . A large body of evidence indicates that CD8+ T cells do play a significant role in the control of HIV and SIV replication that involves both cytolytic and non-cytolytic mechanisms . First , CD8+ T cells can inhibit HIV and SIV replication in vitro [6] , [7] . Second , there is a temporal association between post-peak decline of acute viremia and emergence of CD8+ T cell responses [8] , [9] . Third , antibody-mediated in vivo depletion of CD8+ lymphocytes is consistently associated with increased virus replication in SIV-infected RMs [10] , [11] , [12] . Fourth , there is a strong association between specific major histocompatibility complex ( MHC ) alleles and ability to control virus replication during HIV and SIV infection [13] . In this context , the fact that CTL escape mutants consistently arise during both acute and chronic HIV/SIV infections [13] demonstrate the presence of selective CD8+ T cell-mediated immune pressure on the virus population . On the other hand , the fact that escape mutants are consistently observed is also an indicator of the overall inability of these cells to fully suppress virus replication . SIVmac239 infection of RMs bearing the MamuA*01 MHC class I allele elicits CD8+ T cell responses against two very well characterized immunodominant epitopes: Tat-SL8 [14] and Gag-CM9 [15] . While SL8- and CM9-specific CD8+ T cell responses are both generated during the acute phase of infection [14] , [16] , the emergence of CTL escape mutants occurs much more rapidly in the Tat-SL8 epitope than it does in Gag-CM9 [14] , [17] , [18] , presumably due to strong functional constraints imposed on the gag gene and the need for extra-epitopic , secondary compensatory mutations to allow effective virus replication [19] , [20] , [21] . Of note , in these studies the kinetics of the generation and fixation of CTL escape mutations occurring in the SL8 and CM9 regions of SIV was analyzed only in plasma virus , which is thought to provide an overall representation of the virus that are replicating within the host . However , to the best of our knowledge , a comprehensive and comparative longitudinal analysis of CTL escape mutants in the mucosal and lymphoid tissues of SIV-infected RMs has not been yet been conducted . Determining if CTL escape mutants emerge more rapidly in lymphoid or , alternatively , mucosal tissues would provide important information with respect to the predominant sites of CD8+ T cell-mediated immunological pressure in vivo . In this study , we characterized the appearance , dynamics , and dissemination of CD8+ T cell escape mutants in lymphoid vs mucosal tissues during SIVmac239 infection of RMs . To this end , we conducted an extensive longitudinal assessment of viral sequences derived from plasma viral RNA as well as cell-associated viral DNA in peripheral blood mononuclear cells ( PBMC ) , lymph node biopsies ( LN ) , and biopsies of the rectal mucosa ( RB ) of 15 SIV-infected RMs . These animals were included in a previously published study designed to investigate the immunogenicity and protection from SIVmac239 challenge conferred by two MVA-base candidate AIDS vaccines expressing SIVmac gag and tat [22] . As expected , we observed that CTL escape mutants occurred early in the Tat-SL8 epitope and much later in the Gag-CM9 epitope . Importantly , we found that in all RMs , Tat-SL8 escape mutants appeared earlier and in higher frequency in LNs then in RBs , with variants found at high frequency in LNs at day 14 post infection becoming dominant in the RBs at day 28 post infection . These results indicate that LNs represent the primary source of CTL escape mutants during the acute phase of SIVmac239 infection , suggesting that LNs are the main anatomic sites of virus replication and/or the tissues in which CTL pressure is most effective in selecting SIV escape variants .
A group of fifteen MamuA*01-positive rhesus macaques ( Macaca mulatta; RM ) were infected intravenously with 10 , 000 TCID50 of SIVmac239 as part of a previous study designed to assess the immunogenicity and potential protection from challenge conferred by MVA-based candidate AIDS vaccines [22] . In this study , ten RMs were immunized three times with either MVA “wild-type” ( 5 animals ) or a genetically engineered MVA in which the gene for the Uracil DNA Glycosidase ( UDG ) was deleted , i . e . ΔUDG ( 5 animals ) , with both vectors expressing the Gag and Tat proteins of SIVmac239 . Five additional unvaccinated RMs were used as controls . In all animals , SIV-specific CD8+ T cell responses were measured at various time points post-immunization and post-challenge in multiple tissues , including PBMC , RB , and LN by tetramer staining for the Gag-CM9 and Tat-SL8 epitopes of SIVmac239 . The main results of this study are summarized in Table S1 . Briefly , administration of MVA-SIV immunogens resulted in a partial ( ∼1 log ) and transient ( 60–120 days ) decline in plasma viral load that did not translate into protection from CD4+ T cell depletion and disease progression . Of note , in four RMs the level of virus replication decreased to near undetectable levels during the chronic phase of infection , suggesting that these “controller” animals were able to mount antiviral immune responses that could successfully suppress virus replication in vivo . The large amount of virologic and immunologic data collected in this study combined with an extensive archive of tissue samples presented an ideal opportunity to explore in detail the relationship between SIV-specific CD8+ T cell responses in various tissues , and the appearance and dissemination of CTL escape mutants in a relatively large cohort of SIVmac239-infected RMs . Due to relatively modest protective effect of the used immunization regimen [22] , we chose to conduct our analysis in the entire group of 15 SIV-infected animals without dividing them into vaccinated and controls . To characterize the emergence of CTL escape mutants in our cohort of SIVmac239-infected RMs , we amplified a 435 nucleotide region surrounding the Gag-CM9 epitope and a 390 nucleotide region surrounding the Tat-SL8 epitope from reverse transcribed plasma viral RNA and genomic DNA derived from PBMCs , RB , and LNs collected at multiple time points post infection . These amplicons were then sequenced from one end using Roche's 454 pyrosequencing technology . After eliminating all sequence reads that did not meet the minimum set of quality criteria ( see Materials and Methods ) , source animals for each read were identified via pre-determined barcodes , and the reads were aligned with the corresponding wildtype SIVmac239 sequence . High frequency insertions and deletions ( indels ) that resembled well-characterized artifacts of the 454 pyrosequencing procedure were repaired with reference to the wildtype SIVmac239 sequence . All resulting sequences that contained no indels and were at least 243 nucleotides long for Gag-CM9 and 220 nucleotides long for Tat-SL8 were included in all subsequent analyses . To ensure that the sequence selection process did not bias our results , all analyses were performed on the full set of reads that had been repaired indiscriminately using the SIVmac239 wildtype sequence . No bias was found in our results ( data not shown ) . HIV and SIV gag genes are highly conserved and thus escape mutations in CD8+ T cell epitopes in this region are typically associated with either highly unfit viruses or the appearance of compensatory mutations outside of the epitope itself [20] . In order to characterize the rate and mechanisms of escape in the highly conserved Mamu-A*01-restricted Gag-CM9 epitope in our SIVmac239 infected RMs , we amplified and sequenced this epitope from plasma virus and cell-associated viral DNA in PBMCs , LN , and RB . Escape in Gag-CM9 occurred via mutations at the second position ( threonine ) in the epitope ( Figure 1A ) . The substituted amino acids observed included serine , isoleucine , and cysteine . Interestingly , the threonine to cysteine amino acid substitution is achieved via a nucleotide substitution at the first and second positions of the threonine codon and was observed in the RB but not other tissues of a single animal , RDo8 ( data not shown ) . There was no evidence of a significant increase in the frequency of the intermediate codons , thus suggesting that viruses bearing the resultant amino acid substitutions are at a severe selective disadvantage . However , it is also possible that the frequency of sampling during the chronic phase of SIVmac239 infection was not sufficient to detect the presence of relatively transient intermediate amino acid substitutions . As expected based on previous studies [17] , [18] , [20] , escape in Gag-CM9 was observed in only two RMs ( RDo8 and RWi8 ) and not until day 84 post infection , when escape mutants appeared in plasma virus and , although at lower frequencies , in PBMC-derived cell-associated virus ( Figure 1B ) . In these two animals viruses isolated from RBs at day 168 post infection were almost entirely comprised of Gag-CM9 escape mutants ( Figure 1B ) . Previous studies of the kinetics of emergence of CTL escape in the Mamu-A*01-restricted Tat-SL8 immunodominant epitope have shown that in plasma virus , Tat-SL8 escape occurs during the early stages of SIV infection and via multiple amino acid substitutions [14] , [23] , [24] . For this reason , we focused our analysis of the emergence of Tat-SL8 escape mutants in our group of SIVmac239-infected RMs during the acute phase of infection . As described for the Gag-CM9 epitope , we PCR amplified the viral genomic region surrounding the Tat-SL8 epitope from reverse transcribed plasma viral RNA and sequenced it by 454 pyrosequencing . As expected based on previous studies [14] , [23] , [24] , the emergence of Tat-SL8 escape mutants consistently occurred at high frequency during acute SIV infection . In particular , we observed that CTL escape in Tat-SL8 occurred through one or more of several amino acid substitutions: serine to proline or phenylalanine at position 1 , threonine to isoleucine at position 2 , serine to leucine at position 5 , and alanine to aspartic acid at position 6 ( Figure 2A ) . These amino acid substitutions occur through single nucleotide mutations and have all been characterized as escape mutations in previous studies [14] , [23] , [24] . As shown in Figure 2B , Tat-SL8 escape mutants began to emerge by day 14 post infection in most RMs , with especially high levels of escape mutants evident in 4 animals ( RDo8 , ROu8 , RWi8 , RWu8 ) . By day 28 post infection , the majority of plasma virus in all SIV-infected RMs was comprised of Tat-SL8 escape mutants , with the median frequency of the wildtype Tat-SL8 epitope at 0 . 08 ( range , 0 . 008–0 . 261; Figure 2B ) . The frequency of the viruses bearing the wildtype Tat-SL8 epitope continued to decrease through day 84 post infection ( mean , 0 . 01; range , 0 . 004–0 . 027 ) . Overall , these data confirm the early emergence of numerous Tat-SL8 escape mutants , a finding that reflects both the selective pressure exerted by Tat-SL8-specific CD8+ T cell responses and the relative genetic flexibility of Tat in maintaining its function despite the presence of these mutations [19] . To investigate the kinetics of appearance and dissemination of escape mutants in SIVmac239 infection in different anatomic compartments , we next compared both the frequency and the character of Tat-SL8 CTL escape mutants in viral sequences derived from plasma virus and cell-associated viral DNA from peripheral blood mononuclear cells ( PBMCs ) , lymph nodes ( LNs ) and intestinal mucosa that was sampled by rectal biopsies ( RB ) . This analysis was performed on samples obtained at day 14 and day 28 post infection . While Tat-SL8 escape mutants were evident in all four examined tissues at day 14 post infection , LNs exhibited a significantly higher frequency of escape mutants than either plasma or RB ( Figure 3A; p = 0 . 0022 , Kruskal-Wallis with Dunn's multiple comparisons ) . Interestingly , the discrepancy in the level of escape between LNs and plasma virus may be seen as reflecting the fact that circulating virus at the peak of acute SIV infection originates primarily from other tissues . By day 28 post infection , plasma virus exhibited a greater level of Tat-SL8 escape than cell-associated virus sampled from either PBMCs or RBs ( Figure 3B; p = 0 . 0001 , Kruskal-Wallis with Dunn's test for multiple comparisons ) , but not from LNs . This finding suggests that plasma virus during the phase of post-peak decline in viremia is likely to originate primarily in lymphoid tissues . Escape in Tat-SL8 can occur through multiple amino acid substitutions [14] , [23] , [24] . In order to further delineate the sources of actively replicating viruses that contain Tat-SL8 escape mutants during acute SIVmac239 infection of RMs , we next characterized the frequency distribution of distinct intra-epitopic Tat-SL8 escape mutations . We found that , in the majority of RMs , Tat-SL8 escape mutants sampled from LNs at day 14 post infection were dominated by a serine to proline amino acid substitution at the first position of the epitope , although amino acid substitutions at other positions were observed at much lower frequency ( Figure S1A ) . This pattern of amino acid substitution observed in LNs appeared to be different from what we observed in the Tat-SL8 genotypes sampled from the other three examined tissues , in which there was no clearly dominant amino acid substitution ( Figure S1A ) . In particular , the distribution of Tat-SL8 escape mutants in the plasma virus was much more similar to that of cell-associated viral DNA from PBMCs , with small , equivalent frequencies of mutation at the first , second , fifth and eighth positions , than that of either RBs or LNs . Of note , by day 28 post infection , the distribution of Tat-SL8 escape mutants had become more similar across all tissues ( Figure S1B ) . However , while we observed an increase in the relative abundance of mutations at positions 2 through 8 in LNs , the serine to proline mutation at the first amino acid position of Tat-SL8 continued to be dominant ( Figure S1B ) . In fact , this particular escape mutation became the most frequently sampled mutation in all tissues at day 28 post infection ( Figure 3C; 2-way ANOVA with Bonferroni multiple comparison test , p<0 . 001 ) , and continued to dominate viral populations through the latest time points sampled from all tissues ( data not shown ) . The persistent high frequency of this escape mutation at the first position of Tat-SL8 suggests that anti-SIV immune responses during the acute infection in lymphoid tissues rather than at mucosal sites have a more lasting effect on the evolution of the SIV viral population . Interestingly , three of the four RMs exhibiting relatively large numbers of escape mutants at day 14 post infection had been vaccinated with the MVA-SIV vector ( Figure S1A ) , thus suggesting an early CTL-driven selection of these viral variants . Indeed , vaccinated RMs showed a non-significant trend towards increased frequencies of Tat-SL8 escape mutants at day 14 post infection compared to unvaccinated controls ( p>0 . 05 , Kruskal-Wallis with Dunn's test for multiple comparisons ) . However , vaccination did not preferentially increase the level of Tat-SL8 specific CD8+ T cells in LNs compared to PBMCs and RBs at days −28 , 0 , and 7 post infection ( Figure S2 ) . Taken together these observations argue against the possibility that the expansion of Tat-SL8 escape mutants in LNs was caused primarily by vaccine-induced prior expansion or redistribution of Tat-SL8-specific CD8 T cells to lymphoid tissues . The large amount of virological and immunological data collected during the course of SIVmac239 infection in the RMs included in this study allowed us to probe whether specific immune responses or virological outcomes ( see Table S1 ) were associated with the levels of CTL escape mutants observed at days 14 and 28 post infection . We did not find any significant correlations between the levels of escape mutants in any tissue at day 14 post infection and the levels of SIV-specific CD8+ T cell-mediated immune responses as measured by tetramer staining at the same time point . Furthermore , none of the virological parameters measured at days 14 and 28 post infection correlated with the level of escape mutants present in any tissue at the same time points . However , we observed a clear correlation between immune responses at day 14 post infection and the level of Tat-SL8 escape present in several tissues at day 28 post infection ( Figure 4 ) . First , the frequency of viruses bearing the wildtype Tat-SL8 epitope among plasma viruses at day 28 post infection was significantly inversely correlated with the abundance of anti-Tat-SL8 CD8+ T cells at day 14 post infection in LNs ( Figure 4A; Spearman's correlation , r = 0 . 6592 , p = 0 . 0075 ) and PBMCs ( Figure 4B; Spearman's correlation , r = 0 . 5836 , p = 0 . 0224 ) , but not RBs ( Figure 4C; Spearman's correlation , r = 0 . 4850 , p = 0 . 787 ) , suggesting that LNs are the major anatomic sites where the anti-SIV CD8 T cell response is exerting immunological pressure at the peak of acute SIVmac239 infection . Second , the frequency of wildtype Tat-SL8 epitope in RBs and PBMCs at day 28 post infection was strongly correlated with levels of Tat-SL8 CD8+ T cells in RBs at day 14 post infection ( Figure 4D–E; Spearman's correlation , RB: r = 0 . 7992 , p = 0 . 001; PBMC: r = 0 . 6485 , p = 0 . 0121 ) , perhaps suggesting a local effect of mucosal SIV-specific CTL responses in determining the emergence of escape mutants . Third , the levels of Gag-CM9-specific CD8+ T cell responses in RBs at day 14 post infection also correlated strongly with the frequency of tat-SL8 escape mutants in RBs ( Figure 4F; Spearman's correlation , r = 0 . 6731 , p = 0 . 0164 ) , but not PBMCs ( data not shown ) at day 28 post infection . Interestingly , despite the significant correlations between the frequency of Tat-SL8 escape and the magnitude of anti-SIV CD8 T cell responses in various tissues , the decline in CD4+ T cells in RB and PBMC did not correlate with the frequency of escape at 14 and 28 days post infection in any tissue ( data not shown ) . Taken together these findings indicate that SIV-specific CTL responses at the peak of acute infection have a strong impact on the resultant viral population , with the pattern of observed circulating virus variants at day 28 post infection being shaped predominantly by LN-based immune responses . Four SIV-infected RMs included in the current study ( three vaccinated and one control ) were able to spontaneously control virus replication to near undetectable levels shortly after the resolution of acute viremia ( Table S1 ) . These RMs exhibited a ∼1 log lower peak of acute viremia , yet similar levels of SIV-specific CD8+ T cell responses as the normal non-controller animals . To understand the contribution of CD8+ T cell escape to the control of SIV replication , we compared the prevalence of Tat-SL8 escape mutants and the diversity of virus sequences between the controller and non-controller groups of SIV-infected RMs . Control of SIVmac239 replication in RMs was associated with increased levels of viruses bearing the wildtype Tat-SL8 epitope among infected cells in PBMCs ( Figure 5A; 2-way ANOVA using the Bonferroni Correction for multiple comparisons ) while viruses circulating in the plasma were primarily escape mutants in both groups ( Figure 5B ) . Similarly , viruses sampled from RBs in controllers at day 168 post infection were also composed primarily of viruses bearing the wildtype Tat-SL8 epitope ( data not shown ) . It should be noted , however , that virus was amplifiable from day 168 RB samples of only two of the four controller . Finally , Tat-SL8 and the surrounding region more closely resembled wildtype SIVmac239 in viruses sampled at late time points from PBMC , but not plasma , of controllers than non-controllers ( Figure 5C–D; 2-way ANOVA using the Bonferroni Correction for multiple comparisons ) . In the setting of low virus replication , as in our controller SIV-infected RMs , virus isolated from cellular samples may include a large representation of archival sequences as compared to plasma . As such , it is possible that the large frequency of wildtype viruses in the controllers is a consequence , rather than a cause , of the low virus replication .
Despite the presence of a strong antiviral immune response , HIV infection of humans and SIV infection of RMs usually results in a chronic and progressive immunodeficiency . The inability of anti-SIV immune responses to effectively control virus replication is at least partially due to the singular ability of this primate lentivirus to evade immune responses via a high rate of genetic mutation [13] . The epidemic circulation of extremely diverse viral subtypes and the immune-mediated selection of viral escape variants are two of the main problems that must be overcome by a successful candidate HIV vaccine [25] . In the setting of experimental SIV infection of non-human primates , the comparison of virus sequences obtained in different tissues might be used to determine the anatomic origin of these escape variants . We postulate that this analysis will help us define the anatomic compartments where immune responses have their greatest impact on virus replication and fitness . In addition , as plasma virus includes all viral variants produced in the body at any given time , a comparative kinetic analysis of the emergence of viral escape mutants in plasma vs . tissues could provide an indirect but robust measure of the extent of virus replication in the examined tissues . Identifying these anatomic sites of high viral replication and dominant immune pressure during the acute phase of HIV/SIV infection might help the design of a vaccine that can elicit adaptive immune responses capable of durable control of HIV replication . In order to define these sites of virus replication and immune pressure during acute SIVmac239 infection of RMs , we investigated the kinetics and characteristics of escape in two well characterized Mamu-A*01-restricted CTL epitopes ( Gag-CM9 and Tat-SL8 ) . We sequenced the SIV genomic regions containing these epitopes in longitudinal archival samples of plasma virus and cell-associated viral DNA from PBMCs , LNs and RBs collected from a cohort of vaccinated and challenged RMs [22] . Ten of the rhesus macaques in this study had been vaccinated with MVA-derived vectors expressing the Gag-CM9 and Tat-SL8 epitopes , while five of the challenged RMs were unvaccinated . While our data do not rule out the possibility that other tissues ( i . e . , spleen , liver , etc ) are significant contributors to plasma viremia and/or immune pressure , the unique aspect of this study is the direct comparison between plasma , LNs , PBMCs , and RBs . As expected , Gag-CM9 escape mutants were observed in only two animals and after day 84 post infection , while the Tat-SL8 epitope acquired escape mutations very quickly during the acute phase infection . Interestingly , the comparative analysis of Tat-SL8 escape mutants revealed , at day 14 post infection , a higher frequency in LN as compared to other tissues . By day 28 post infection , Tat-SL8 escape mutants had become dominant in plasma virus , and were found at higher frequency in LNs as compared to RBs . Of note , at day 14 post infection virus escape mutants in LNs , but not other tissues , were overwhelmingly comprised of mutations at the first amino acid of the epitope . These mutants became dominant in all tissues by day 28 post infection . Taken together , these results suggest that , during acute SIVmac239 infection , lymph nodes are the main sites of virus replication and/or cellular immune pressure . The large amount of virological and immunological data collected during the original vaccination and challenge study of these SIV-infected RMs [22] allowed us to more explicitly test hypotheses regarding the immunological processes and anatomic compartments that ultimately produced these viral escape mutants . By correlating immune responses and changes in viral dynamics with the level of escape in the examined tissues , we can determine not only the origin of escape mutants , but the immune responses responsible for their generation . We found that the level of Tat-SL8-specific immune responses in LNs , and to a lesser extent PBMCs , at day 14 post infection correlated strongly with the level of escape in plasma virus at day 28 post infection . On the other hand , SIV-specific immune responses in RBs at day 14 post infection did not correlate with the level of Tat-SL8 escape at day 28 in plasma virus . Therefore , LNs rather than mucosal tissues appear to be the tissue in which the strongest immune pressure shapes the genetic composition of the currently replicating virus population during the phase of post-peak decline of viremia in SIV-infected RMs . In addition , we observed that , in controller SIV-infected RMs ( but not in non-controller RMs ) , a large proportion of PBMC-derived proviruses during the chronic phase of infection exhibited the wildtype Tat-SL8 rather than the escaped epitope . This high frequency of wildtype virus in controller RMs may be the result of the early establishment of a wildtype latent viral reservoir , which , due to their low level of replication , appears over-represented as compared with the non-controller animals . Since the actively replicating virus ( i . e . , in plasma ) contains very few , if any , viruses bearing the wildtype Tat-SL8 , these wildtype PBMC-associated viruses are likely to be either latent or non-replication competent . The observation that immune responses in LNs are more important for shaping post acute viral populations is somewhat surprising as the gut-associated lymphoid tissue has been implicated as the major source of virus replication during the acute phase of SIV infection [26] , [27] , [28] , [29] , [30] . One possibility is that , during the post-peak decline of viremia , the gut mucosa is a relatively isolated site of SIV replication , with slower migration of virions from the intestinal interstitial fluids into the systemic circulation as compared to LNs . Indeed , there is evidence for compartmentalization of HIV within the gut [31] suggesting that virus does not move freely within this tissue . An alternative possibility is that the level of immunological pressure , rather than the absolute level of virus production , is responsible for the directional appearance of CTL escape mutants from LNs to mucosal tissues . This concept fits with the fact that during acute SIV infection LNs are the site of antigen-specific CD8+ T cell expansion from the naïve pool , with virus-infected cells encountering anti-SIV CD8+ T cells more frequently than in the intestinal lamina propria . These data are also consistent with the possibility that the gut is not the main source of virus replication during acute SIV infection , as recently suggested in a modeling study [32] . However , we could not rule out the possibility that the mutations in Tat simply provide a selective advantage to viruses in lymphoid tissue ( or less of a disadvantage relative to viruses replicating in gut mucosal tissues ) independently of their effects on immune recognition . While our study agrees with the numerous published articles that have characterized immune escape in SIVmac239 infected RMs [14] , [15] , [17] , [23] , [33] , [34] , [35] , [36] , [37] , in almost all of these studies , CTL escape mutants were assessed only in plasma virus . Our study is unique in that we examined virus in multiple longitudinal tissue samples ( PBMC , LNs , and RBs ) in addition to the plasma virus . We are in fact aware of only a single study that examined SIVmac239 escape mutants in multiple tissues [38] , but this study focused only on Gag-CM9 escape measured at a single time point during the chronic infection . It will be important to determine whether the pattern of escape seen in Tat-SL8 during the acute infection is a general phenomenon for CTL epitopes that escape during the acute phase of infection with little or no measurable effects on viral fitness . In addition it will be informative to characterize the anatomic distribution of CTL epitopes that escape during the chronic phase of SIV infection ( e . g . Gag-CM9 ) by designing new studies in which tissue samples are collected accordingly . Furthermore , future studies should address whether the cell-associated viral sequences sampled from relatively small biopsies of the rectal mucosa are representative of the gut as a whole . While this caveat applies to most studies of gut-associated lymphoid tissue in the setting of SIV infection , it is possible that our current results are biased by the limited number of samples available for virus sequencing . It is unclear whether and to what extent the particular pattern of CTL escape observed in our cohort of SIV-infected RMs represents a generalized phenomenon or , alternatively , is the result of this specific experimental design ( i . e . , intravenous high titer infection ) , which does not mimic the typical route and circumstances surrounding HIV infection of humans . Furthermore , the route of infection may influence the anatomic pattern of emergence of CTL escape mutants . For example , a rectal challenge model might result in the early seeding of local draining lymph nodes with virus whereas an intra-venous challenge could possibly disseminate virus more efficiently to systemic lymphoid tissues throughout the body . Future studies in which RMs are infected with low dose intra-rectal or intra-vaginal challenge will help determine how the route and dose of infection influences the early kinetics of CTL escape . Additionally , while the MVA vaccination regime in this study did not protect from SIV disease progression , it would be interesting to determine whether more successful vaccines ( i . e . , rhesus CMV [39] and AdHu26 [40] ) alters the dissemination of viral escape variants and/or the pattern of immune responses during the acute phase of SIV infections .
These studies 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 were approved by the Emory University ( AWA# A3180-01 ) and University of Pennsylvania ( AWA# A3079-01 ) Institutional Animal Care and Use Committees . All animals were anesthetized prior to the performance of any procedure , and proper steps were taken to ensure the welfare and to minimize the suffering of all animals in these studies . The design of the SIV immunization and challenge study in RMs that has been used as a source of samples for the current manuscript are published elsewhere [22] . Briefly , ten RMs were divided in two groups of five animal and immunized three times with either one of two MVA vaccine vectors ( i . e . , wildtype and a genetically modified MVA in which the udg gene was deleted ) expressing the SIVmac239 gag and tat genes . One year after the initial immunization , all ten animals as well as five unvaccinated control RMs were inoculated i . v . with 10 , 000 TCID50 of SIVmac239 . Throughout the immunization and challenge phase plasma , peripheral blood mononuclear cells ( PBMC ) , lymph node biopsies ( LN ) and biopsies of the rectal mucosa ( RB ) were collected for virological and immunological analysis ( see Table S1 ) . The MHC-class I genotypes of all fifteen RMs were determined courtesy of Dr . David Watkins . Of note , exclusion of the only MamuB*08+ and MamuB*17+ animal did not change the result of our analyses . PBMC were collected by gradient centrifugation . LN and RB were processed fresh as previously described . The levels of Tat- and Gag- specific CD8 T cells were assessed via staining with Streptavidin-APC conjugated class I MHC tetramers folded with the Tat28-35-SL8 ( STPESANL ) or Gag181-189-CM9 ( CTPYDINQM ) peptide epitopes according to standard procedures . Genomic DNA was purified from thawed PBMC samples using the QIAamp DNA blood mini kit ( Qiagen; Valencia , CA ) , and from whole Streck-fixed , paraffin embedded RBs or from 50 µM slices of formalin-fixed , paraffin embedded LNs using the QIAamp DNA FFPE tissue kit ( Qiagen; Valencia , CA ) . Viral RNA was extracted from plasma samples using the Qiagen's Viral RNA mini kit ( Valencia , CA ) . Viral cDNA was reverse transcribed using Invitrogen's SuperScript III and primers specific for sequences upstream of the tat ( Tat-RT3: 5′-TGGGGATAATTTTACACAAGGC-3′ ) or gag ( Gag-RT2: 5′-AGCTTGCAATCTGGGTTAGC-3′ ) amplicons . Viral sequences were amplified from purified genomic DNA or viral cDNA in a nested two step PCR . The thermal cycler program for the first round was: 94°C for 2:00; cycle 94°C for 0:30 , 55°C for 0:30 , and 68°C for 1:00 , 35 times; and then a final 68°C for 7:00 before cooling to 4°C . After the tenth cycle , the extension step ( 68°C ) is extended by 5 seconds every cycle to account for the degrading polymerase . The thermal cycler program for the second round was: 94°C for 2:00; cycle 94°C for 0:30 , 53°C for 0:30 , and 68°C for 1:00 , 35 times; and then a final 68°C for 7:00 before cooling to 4°C . The first round primers for tat were Tat-F1 ( 5′-GATGAATGGGTAGTGGAGGTTCTGG-3′ ) and Tat-R2 ( 5′-CCCAAGTATCCCTATTCTTGGTTGCAC-3′ ) , and the first round primers for gag were Gag-F1 ( 5′-GAGACACCTAGTGGTGGAAACAGG-3′ ) and Gag-R2 ( 5′-GCTCTGAAATGGCTCTTTTGGCCC-3′ ) . The design of the second round primers involved the incorporation of Roche's 454 Adaptor sequences and an animal-specific , 4 nucleotide barcode that allows identification of individual animals in the pooled sequencing runs . The barcode key can be found in Table S1 , and their position is indicated by a lowercase ‘b’ in the following primer sequences . Second round primers for tat had sequences similar to those previously published [15] and were Tat-F3 ( 5′-GCCTTGCCAGCCCGCTCAGbbbbTGATCCTCGCTTGCTAACTG-3′ ) and Tat-R3 ( 5′-GCCTCCCTCGCGCCATCAGAGCAAGATGGCGATAAGCAG-3′ ) , and the second round primers for gag were Gag-F3 ( 5′-GCCTTGCCAGCCCGCTCAGbbbbCACCATCTAGCGGCAGAGGAGG-3′ ) and Gag-R3 ( 5′-GCCTCCCTCGCGCCATCAGACCCCAGTTGAATCCATCTCCTG-3′ ) . After amplification , each amplicon was gel purified using the QIAquick gel extraction kit ( Qiagen , Valencia , CA ) . The amplicons were then quantified on a NanoDrop ( Company ) , and mixed at equimolar concentrations within tissues and time points . Massively parallel pyrosequencing was performed by the University of Pennsylvania , Department of Genetics Sequencing Facility on a Roche 454 Genome Sequencer FLX ( Branford , CT ) . Raw 454 sequencing reads will be available in the NCBI Sequence Read Archive under accession number SRA027346 . 1 . Sequences were subjected to several steps of quality control [41] , [42] . First , sequences that contained ambiguous nucleotides and those that did not meet the minimum length requirements ( tat: 220 nucleotides; gag: 243 nucleotides ) were excluded . Next , each read was aligned individually to the SIVmac239 wildtype sequence using ClustalW and barcodes were identified . Reads where the barcode could not be identified were excluded from the analysis . Several common insertions and deletions ( indels ) that resembled common 454 sequencing artifacts were then identified and repaired using the wildtype SIVmac239 sequence as a template . Sequences containing multiple indels and low frequency sequencing artifacts were excluded . All sequence manipulations and analyses were performed and implemented in a suite of scripts written in Python . Statistical analyses were all performed in GraphPad Prism . | Strong antiviral CD8+ T lymphocytes limit SIV replication by recognizing short pathogen-derived peptide epitopes . The cytotoxic CD8+ T cell responses specific for this highly mutable virus can select for viruses bearing mutations that prevent CD8+ T cell recognition of cells infected with these escape mutants . To determine the anatomic origin of these escape mutants , we tracked a particular escape mutant in multiple tissues ( plasma virus , lymph nodes , rectal mucosa , and peripheral blood immune cells ) during the early , acute phase of SIVmac239 infection of rhesus macaques . We found that escape mutants first reach high frequency in lymph nodes 2 weeks after infection , and the particular mutants generated in lymph nodes disseminate to other tissues by week 4 . Furthermore , we found that epitope-specific CD8+ T lymphocyte responses in the lymph nodes and peripheral blood , but not the gut mucosa , are significantly correlated with the frequency of escape mutants in the plasma virus at week 4 . This suggests that lymph nodes , and not the gut , are the primary site of anti-SIV CD8+ T cell responses and/or SIV replication during the acute phase of infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"immune",
"cells",
"immunity",
"to",
"infections",
"immunology",
"microbiology",
"host-pathogen",
"interaction",
"immunodeficiency",
"viruses",
"adaptive",
"immunity",
"animal",
"models",
"of",
"infection",
"viral",
"immune",
"evasion",
"t",
"cells",
"biology",
"viral",
"evolution",
"immune",
"response",
"viral",
"persistence",
"and",
"latency",
"immunity",
"virology"
]
| 2011 | Viral CTL Escape Mutants Are Generated in Lymph Nodes and Subsequently Become Fixed in Plasma and Rectal Mucosa during Acute SIV Infection of Macaques |
Infection with Wuchereria bancrofti can cause severe disease characterized by subcutaneous fibrosis and extracellular matrix remodeling . Matrix metalloproteinases ( MMPs ) are a family of enzymes governing extracellular remodeling by regulating cellular homeostasis , inflammation , and tissue reorganization , while tissue-inhibitors of metalloproteinases ( TIMPs ) are endogenous regulators of MMPs . Homeostatic as well as inflammation-induced balance between MMPs and TIMPs is considered critical in mediating tissue pathology . To elucidate the role of MMPs and TIMPs in filarial pathology , we compared the plasma levels of a panel of MMPs , TIMPs , other pro-fibrotic factors , and cytokines in individuals with chronic filarial pathology with ( CP Ag+ ) or without ( CP Ag− ) active infection to those with clinically asymptomatic infections ( INF ) and in those without infection ( endemic normal [EN] ) . Markers of pathogenesis were delineated based on comparisons between the two actively infected groups ( CP Ag+ compared to INF ) and those without active infection ( CP Ag− compared to EN ) . Our data reveal that an increase in circulating levels of MMPs and TIMPs is characteristic of the filarial disease process per se and not of active infection; however , filarial disease with active infection is specifically associated with increased ratios of MMP1/TIMP4 and MMP8/TIMP4 as well as with pro-fibrotic cytokines ( IL-5 , IL-13 and TGF-β ) . Our data therefore suggest that while filarial lymphatic disease is characterized by a non-specific increase in plasma MMPs and TIMPs , the balance between MMPs and TIMPs is an important factor in regulating tissue pathology during active infection .
Lymphatic filariasis ( LF ) is characterized by dysfunction of lymphatics that can lead to severe and often irreversible lymphedema and elephantiasis [1] , [2] . The critical factor in the development of host pathology appears to reflect the parasite-mediated initiation of a cascade of events that leads to tissue fibrosis and scarring [2] , [3] . It is assumed that both parasite products and the host inflammatory response lead to lymphatic dysfunction and lymphangiogenesis that , in turn , predisposes infected individuals to secondary bacterial and fungal infection [2] , [4] , [5] . Host-parasite interactions as well as secondary infections then trigger inflammatory reactions in the skin and subcutaneous tissue with underlying fibrosis and cellular hyperplasia processes resulting in lymphedema and elephantiasis [2] , [4] . Typically in Wuchereria or Brugia infections , disease manifests years after exposure , while clinically asymptomatic infection is not only more common but can also occur at a relatively young age [1] . Although lymphatic dysfunction and localized/systemic immunologic and inflammatory responses are important features of lymphatic pathology [6] , perturbations in extracellular matrix ( ECM ) architecture and subsequent remodeling are also associated with filarial disease [7]–[9] . Chronic inflammation , as seen in LF disease , causes an excessive accumulation of ECM components ( such as collagen ) that can contribute to fibrotic scarring [2] . Patent filarial infections are typically associated with Type 2 and regulatory cytokine responses , but some of these are also pro-fibrotic , especially IL-5 , IL-13 , and TGF-β [10] , which are known to influence collagen deposition and ECM remodeling [11] . The turnover of collagen and other ECM proteins is controlled by a large family of proteolytic enzymes called matrix metalloproteinases ( MMPs ) and their inhibitors ( tissue inhibitors of metalloproteinases [TIMPs] ) , produced by a variety of cell types including macrophages , granulocytes , epidermal cells , and fibroblasts [12]–[14] . Tissue immunopathology is known to be associated with dysregulation of MMPs and TIMPs in several infections , including viral , bacterial , spirochetal , protozoan , fungal , and parasitic infections [15] . The MMP family consists of more than 26 different proteases that differ in their tissue expression/localization and target specificity , while the TIMP family consists of four ubiquitously expressed proteins ( TIMP1–4 ) [12]–[14] . In addition , other factors such as fibroblast growth factor ( FGF ) and platelet-derived growth factor ( PDGF ) are known mediators of tissue fibrosis [11] . Although the importance of tissue fibrosis in the pathology associated with LF is well known [3] , the molecular mechanisms underlying the fibrotic process in filariasis has not been well established . We therefore sought to delineate the role of the factors known to regulate fibrosis and tissue remodeling in filarial disease development . Our data suggest that while elevated circulating levels of MMPs and TIMPs are characteristic of filarial disease , it is the increased ratios of certain MMPs to TIMPs ( associated with elevated pro-fibrotic cytokines ) that is specifically associated with the pathogenesis of disease in LF .
We studied a group of 91 individuals with filarial lymphedema without active filarial infection ( hereafter CP Ag− ) , 28 individuals with filarial lymphedema with active filarial infection ( hereafter CP Ag+ ) , 90 asymptomatic or subclinical , infected individuals ( hereafter INF ) , and 80 uninfected , endemic normal individuals ( hereafter EN ) in an area endemic for LF in Tamil Nadu , South India ( Table 1 ) . Diagnosis of active filarial infection was performed by measuring circulating filarial antigen levels by both the ICT filarial antigen test ( Binax , Portland , ME , USA ) and the TropBio Og4C3 enzyme-linked immunosorbent assay ( ELISA ) ( Trop Bio Pty . Ltd , Townsville , Queensland , Australia ) and all the infected individuals were positive by both circulating antigen assays . All CP Ag− individuals had undergone treatment with repeated doses of diethylcarbamazine . All of the CP individuals had early stage lymphedema ( Grades 1 and 2 ) only , and individuals with concurrent overt and active bacterial infection were excluded from the study . Only 28 CP Ag+ individuals were detected after screening over 1000 individuals with chronic pathology reflecting the rarity of this group in an endemic area . Platelet-poor plasma samples collected using heparin tubes were used for the entire study . All samples were obtained after centrifugation of heparinized whole blood and were stored at −80°C . A subset of individuals in each group ( chosen consecutively ) was used to measure the various parameters by multiplex immunoassays , and the number of individuals used in each group is indicated in the figure legends . All individuals were examined as part of a clinical protocol approved by Institutional Review Boards of both the National Institute of Allergy and Infectious Diseases and the Tuberculosis Research Center ( NCT00375583 and NCT00001230 ) ; informed written consent was obtained from all participants . MMPs were measured using the Fluorokine MAP Multiplex Assay ( R&D Systems , Minneapolis , MN , USA ) , which is a bead based assay , run on a Luminex® ELISA platform . For preliminary experiments , plasma samples were assayed for MMP-1 , MMP-2 , MMP-3 , MMP-7 , MMP-8 , MMP-9 , MMP-11 , and MMP-12 . Because the levels of MMP-2 , -3 , -11 , and -12 were below the threshold for detection , subsequent studies were carried out using a panel of MMP-1 , -7 , -8 , and -9 . The MMP panel measures pro , mature and TIMP-1 complexed MMP-1 , 7 , 8 and 9 . TIMP-1 , TIMP-2 , TIMP-3 , and TIMP-4 levels were measured using the Fluorokine MAP Multiplex Assay ( R&D Systems ) , according to the manufacturer's instructions . Plasma levels of cytokines IL-5 and IL-13 were measured using the Bioplex® multiplex ELISA system . Plasma levels of fibroblast growth factor-2 ( FGF-2 ) and platelet-derived growth factor-AA ( PDGF-AA ) were measured using the Milliplex MAP kit system ( Millipore , Billerica , MA , USA ) . Plasma levels of active TGF-β were measured using an R&D ELISA kit . Data analyses were performed using GraphPad PRISM ( GraphPad Software , Inc . , San Diego , CA , USA ) . Geometric means ( GM ) were used for measurements of central tendency . Statistically significant differences between two groups were analyzed using the nonparametric Mann-Whitney U test . Correlations were calculated by the Spearman rank correlation test .
To determine the association of MMPs with filarial lymphedema , we measured the plasma levels of MMP-1 , -7 , -8 , and -9 in CP Ag+ , INF , CP Ag− , and EN . As shown in Figure 1A , compared with INF , CP Ag+ had significantly higher levels of MMP-8 ( GM of 38 . 2 ng/ml in CP Ag+ vs . 10 . 9 in INF; P<0 . 0001 ) and MMP-9 ( GM of 344 . 8 ng/ml in CP Ag+ vs . 113 . 2 in INF; P = 0 . 0098 ) and significantly decreased levels of MMP-7 ( GM of 2 . 7 ng/ml in CP Ag+ vs . 6 . 7 in INF; P<0 . 0001 ) . Similarly , as shown in Figure 1B , CP Ag− had significantly higher levels of MMP-1 ( GM of 1 . 8 ng/ml in CP Ag− vs . 0 . 7 in EN; P<0 . 0001 ) , MMP-7 ( GM of 8 . 2 ng/ml in CP Ag− vs . 3 . 7 in EN; P = 0 . 0001 ) , MMP-8 ( GM of 8 . 7 ng/ml in CP Ag− vs . 1 . 9 in EN; P<0 . 0001 ) and MMP-9 ( GM of 66 . 8 ng/ml in CP Ag− vs . 6 . 6 in EN; P<0 . 0001 ) in comparison to EN . Finally , INF had significantly higher levels of all the four MMPs compared to EN ( data not shown ) . Thus , filarial lymphedema with or without active infection is characterized by elevated levels of circulating MMPs . To determine the relationship of TIMPs to development of filarial lymphedema , we measured the plasma levels of TIMP-1 , -2 , -3 , and -4 in CP Ag+ , INF , CP Ag− , and EN . As shown in Figure 2A , CP Ag+ had significantly higher levels of TIMP-1 ( GM of 4 . 5 ng/ml in CP Ag+ vs . 3 . 4 in INF; P = 0 . 0045 ) and TIMP-2 ( GM of 17 . 6 ng/ml in CP Ag+ vs . 8 . 1 in INF; P = 0 . 0004 ) but significantly lower levels of TIMP-3 ( GM of 48 . 3 ng/ml in CP Ag+ vs . 553 . 7 in INF; P<0 . 0001 ) and TIMP-4 ( GM of 9 . 9 ng/ml in CP Ag+ vs . 16 . 1 in INF; P = 0 . 0005 ) in comparison to INF . Similarly , as shown in Figure 2B , CP Ag− had significantly higher levels of TIMP-1 ( GM of 3 . 3 ng/ml in CP Ag− vs . 1 . 4 in EN; P<0 . 0001 ) , TIMP-2 ( GM of 9 . 8 ng/ml in CP Ag− vs . 2 . 8 in EN; P<0 . 0001 ) and TIMP-4 ( GM of 15 . 1 ng/ml in CP Ag− vs . 4 . 5 in EN; P<0 . 0001 ) but significantly lower levels of TIMP-3 ( GM of 554 . 8 ng/ml in CP Ag− vs . 650 in EN; P = 0 . 0177 ) in comparison to EN . Finally , INF had significantly higher levels of TIMP-1 , 2 and 4 in comparison to EN ( data not shown ) . Thus , filarial lymphedema with or without active infection is characterized by elevated levels of circulating TIMP-1 and TIMP-2 and decreased levels of TIMP-3 . Because the increased levels of MMPs and TIMPs were not specific to actively infected CP individuals and as MMP/TIMP ratios are considered to be more reflective of the pro-fibrotic status [12] , we determined the ratios of the various MMPs to the four TIMPS . As shown in Figure 3A , we observed significantly increased ratios of MMP-1/TIMP-4 ( P = 0 . 0311 ) and MMP-8/TIMP-4 ( P<0 . 0001 ) in CP Ag+ compared to INF but not in CP Ag− compared to EN . Conversely , CP Ag+ exhibited significantly decreased ratios of MMP-1/TIMP-1 ( P = 0 . 0043 ) , MMP-7/TIMP-1 ( P = 0 . 0004 ) , and MMP-7/TIMP-2 ( P<0 . 0001 ) compared to INF , with no significant difference being observed between CP Ag- and EN . No significant differences were seen between the CP groups and INF or EN group in the ratio of other MMPs/TIMPs . Thus , an imbalance between the circulating levels of specific MMPs and TIMPs—especially increased MMP-1 and MMP-8 to TIMP-4 and decreased MMP-1 and MMP-7 to TIMP-1 and 2—is characteristic of filarial lymphedema in the presence of active infection . To determine the association of other pro-fibrotic factors with filarial lymphedema , we measured the plasma levels of FGF-2 and PDGF-AA in CP Ag+ , INF , CP Ag− , and EN . As shown in Figure 4A , CP Ag+ had no significant alterations in the levels of FGF-2 or PDGF-AA in comparison to INF . Similarly , as shown in Figure 4B , CP Ag− had no significant alterations in the levels of FGF-2 or PDGF in comparison to EN . Thus , filarial lymphedema with or without active infection is not associated with alterations in levels of pro-fibrotic factors FGF-2 or PDGF-AA . To determine the contribution of Type-2 and pro-fibrotic cytokines to the process of filarial lymphedema , we measured the plasma levels of IL-5 , IL-13 , and TGF-β in the four groups of subjects . As shown in Figure 5A , compared with INF , CP Ag+ had significantly higher levels of IL-5 ( GM of 1019 pg/ml in CP Ag+ vs . 37 . 9 in INF; P<0 . 0001 ) , IL-13 ( GM of 1022 pg/ml in CP Ag+ vs . 40 . 8 in INF; P<0 . 0001 ) , and TGF-β ( GM of 308 . 2 pg/ml in CP Ag+ vs . 231 . 6 in INF; P = 0 . 0160 ) . As shown in Figure 5B , no significant differences were observed in the levels of IL-13 or TGF-β between those without active infection , irrespective of clinical status . In addition , INF had significantly higher levels of IL-5 and TGF-β compared to EN ( data not shown ) . Thus , filarial lymphedema with active infection is characterized by elevated plasma levels of IL-5 , IL-13 , and TGF-β . The relationships among the Type-2 cytokines and MMP/TIMP ratios were assessed in those subjects with active infection ( CP Ag+ and INF ) . As shown in Figure 6 , plasma levels of IL-13 exhibited a significant positive correlation with the circulating levels of MMP-1/TIMP-4 ( r = 0 . 4477; P<0 . 0001 ) and MMP-8/TIMP-4 ( r = 0 . 4030 , P<0 . 0001 ) in actively infected individuals . Similarly , plasma levels of IL-5 exhibited a significant positive correlation with the MMP-8/TIMP-4 ( r = 0 . 4768 , P<0 . 0001 ) . Thus , the altered balance between MMPs and TIMPs in the circulation appears to be significantly associated with the Type-2 pro-fibrotic cytokine levels in filarial infection .
The dynamics of the ECM in tissues are orchestrated by the interplay among matrix breakdown , matrix deposition , and reorganization . MMPs are a family of zinc-metalloendopeptidases responsible for the turnover of ECM [12]–[14] . MMPs are tightly regulated at multiple levels including transcriptional and translational regulation as well as by TIMPs [12] . Therefore , tissue homeostasis is achieved by a tight balance of MMP proteolysis to TIMP production . When this balance is altered by inflammation and other mechanisms , dysregulated MMP activity ensues [12] . Altered MMP/TIMP expression ratios have been associated with many diseases including: those associated with tissue destruction , such as cancer invasion and metastasis [16] , rheumatoid arthritis [17] , osteoarthritis [18]; those associated with fibrosis , such as liver cirrhosis [19] , scleroderma [20] , systemic sclerosis [21]; and those associated with weakening of the extracellular matrix , such as dilated cardiomyopathy and epidermolysis bullosa [12] . Although some parasitic infections can induce fibrosis , few studies have actually examined the role of MMPs and TIMPs in human parasitic infections . MMPs and TIMPs have been shown to be associated with disease activity in human neurocysticercosis [22] , leishmaniasis [23] , schistosomiasis [24] , and eosinophilic meningitis [25] . In addition , the regulation of MMP activity has been demonstrated in animal models of Schistosoma mansoni , Toxoplasma gondii , Angiostrongylus cantonensis , Opistorchis viverrini , and Mesocestoides cortii [15] , [26] . Interestingly , although skin and subcutaneous fibrosis and scarring is a characteristic feature associated with longstanding LF and specifically elephantiasis , no study has examined the regulation of MMPs and TIMPs in LF infection . We utilized a cohort of clinically well defined individuals from an area endemic for LF to examine the role played by MMPs and TIMPs in filarial disease pathogenesis . We examined the expression pattern of the pro-fibrotic factors using a two-step comparison with the rationale that factors truly reflective of infection-driven pathogenesis ( as opposed to general factors ) would be significantly different between individuals with lymphedema/elephantiasis with active infection and clinically asymptomatic but actively infected individuals but not between those with disease ( CP ) without active infection and uninfected , endemic normal individuals . Our examination of the baseline levels of MMPs and TIMPs revealed that both groups of proteins were non-specifically elevated in individuals with chronic pathological consequences of LF infection irrespective of their circulating filarial antigen level . Therefore , this suggests that filarial disease per se ( and not the presence of viable parasites ) is associated with these elevations in ECM modulators; however , it also known that TIMPs mediate blocking of MMP-induced proteolytic activity by noncovalently binding to the MMP active site in a 1∶1 stoichiometric ratio . Therefore , we also examined the ratios of various MMPs to TIMPs in our cohort of patients [27] . Not surprisingly , we found evidence of an imbalance between MMPs and TIMPs . Our data would therefore imply that altered ratios of MMP/TIMP are an important underlying factor in the pathogenesis of tissue fibrosis in filarial lymphatic disease . This process is reflective of imbalances in the levels of MMP and TIMPs seen in other diseases with tissue fibrosis , including systemic sclerosis [21] , scleroderma [20] , amyotrophic lateral sclerosis [28] , coronary artery disease [29] , endometriosis [30] , rheumatoid arthritis [17] , viral hepatitis [31] , and a variety of cancers [16] . Our data also suggest that currently available synthetic MMP inhibitors could potentially be of benefit in ameliorating pathology or preventing the development of pathology in active infection [32] . In the context of a role for infection per se , we observed that the presence of active infection significantly enhanced the levels of most MMPs and TIMPs , suggesting that changes in ECM remodeling are occurring prior to the onset of clinically evident lymphedema . Interestingly , among the MMPs with enhanced baseline expression in filarial lymphedema are the two major collagenases in the MMP family , MMP-1 ( or collagenase-1 ) and MMP-8 ( or collagenase-2 ) [12] . While collagenases are known to degrade collagen , MMP-9 is also a gelatinase that can degrade collagen as well as gelatin [12] . As these proteolytic enzymes mainly target collagen and because collagen has been previously shown to be altered in filarial lymphatic pathology [7]–[9] , it is interesting to note that these enzymes are specifically elevated in chronic pathology individuals irrespective of their infection status . In addition to their role in ECM remodeling , TIMPs are also known to have MMP-independent functions [27] . TIMP-3 is an antagonist of vascular endothelial growth factor receptor-2 ( VEGFR-2 ) [33] , while TIMP-2 has been shown to antagonize VEGF signaling [34] , resulting in inhibition of angiogenesis . Because lymphangiogenesis is a well described feature of filarial lymphedema [2] , it is of interest to note the significantly deceased production of TIMP-3 as well as decreased levels of MMP/TIMP-2 ratios in filarial lymphedema . The decreased levels of TIMP-3 and the decreased MMP/TIMP-2 ratios therefore , could potentially signify an important role for these enzymes in promoting VEGF-mediated angiogenesis and/or lymphangiogenesis in response to inflammation , as has been described in neuroinflammatory disorders such as multiple sclerosis or experimental autoimmune encephalomyelitis [27] . We plan to explore the functional activity of MMPs and TIMPs in future experiments . While MMPs and TIMPs are major regulators of fibrosis , PDGF , FGF , epidermal growth factor ( EGF ) , VEGF , and bone morphogenic proteins ( BMPs ) are also known to influence tissue fibrosis [11] . By examining expression of PDGF-AA and FGF-2 in filaria-infected and filaria-uninfected individuals , we were able to show that neither PDGF nor FGF was associated with filarial lymphedema . These data suggest that while filarial lymphedema is characterized by alterations in certain pro-fibrotic factors such as MMPs/TIMPs , it is not associated with a generic increase in the circulating levels of other factors known to influence tissue fibrosis . Although persistent and progressive fibrosis is postulated to be a hallmark of LF disease , the data on the known pro-fibrotic cytokines in filarial disease are limited . By quantifying IL-5 , IL-13 , and TGF-β levels in the four well defined groups , our data indicate clearly that these particular cytokines are associated with development of overt pathology in actively infected individuals . Because pro-fibrotic cytokines are known to influence tissue fibrosis by regulating the levels of MMPs and TIMPs [35]–[38] , we also examined the association between these factors . In agreement with studies in animal models of parasitic infection showing association between MMP/TIMP levels and Type 2 cytokines [39]–[42] , our examination of filaria-infected individuals also reveals a significantly positive association between MMP-1/TIMP-4 and MMP-8/TIMP-4 ratios ( both of which were specifically elevated in CP Ag+ ) and Type 2 ( and pro-fibrotic ) cytokines . Our study clearly implicates a tissue-fibrosis promoting role for IL-5 and IL-13 in filaria-induced lymphatic pathology . Our study clearly identifies a novel role for MMPs and TIMPs as well as Type 2 cytokines in filarial infection-driven morbidity associated with a persistent and progressive tissue fibrosis . While requiring validation in future studies , these results point to potential therapeutic interventional targets in ameliorating filarial lymphedema and possibly even elephantiasis . | Lymphatic filariasis afflicts over 120 million people worldwide . While the infection is mostly clinically asymptomatic , approximately 40 million people suffer from overt , morbid clinical pathology characterized by swelling of the scrotal area and lower limbs ( hydrocele and lymphedema ) . Host immunologic factors that influence the pathogenesis of disease in these individuals are not completely understood . Matrix metalloproteinases are a family of circulating and tissue proteins that influence the development of tissue fibrosis . They are regulated by another family of proteins called tissue inhibitors of metalloproteinases . The interplay between these proteins governs tissue fibrosis in a variety of conditions . In addition , certain cytokines are known to promote pro-fibrotic events . We have attempted to elucidate the role of the above-mentioned factors in disease pathogenesis by comparing the plasma levels of the various markers in four groups of individuals: chronic pathology individuals with or without active filarial infection; asymptomatic , filaria-infected individuals; and uninfected , endemic normal individuals . We show that altered ratios of the metalloproteinases and their inhibitors—as well as elevated levels of pro-fibrotic cytokines—characterize filarial infection-induced lymphatic pathology . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"infectious",
"diseases",
"immunology",
"biology"
]
| 2012 | Altered Circulating Levels of Matrix Metalloproteinases and Inhibitors Associated with Elevated Type 2 Cytokines in Lymphatic Filarial Disease |
The role of RNA polymerase III ( Pol III ) in developing vertebrates has not been examined . Here , we identify a causative mutation of the second largest Pol III subunit , polr3b , that disrupts digestive organ development in zebrafish slim jim ( slj ) mutants . The slj mutation is a splice-site substitution that causes deletion of a conserved tract of 41 amino acids in the Polr3b protein . Structural considerations predict that the slj Pol3rb deletion might impair its interaction with Polr3k , the ortholog of an essential yeast Pol III subunit , Rpc11 , which promotes RNA cleavage and Pol III recycling . We engineered Schizosaccharomyces pombe to carry an Rpc2 deletion comparable to the slj mutation and found that the Pol III recovered from this rpc2-Δ yeast had markedly reduced levels of Rpc11p . Remarkably , overexpression of cDNA encoding the zebrafish rpc11 ortholog , polr3k , rescued the exocrine defects in slj mutants , indicating that the slj phenotype is due to deficiency of Rpc11 . These data show that functional interactions between Pol III subunits have been conserved during eukaryotic evolution and support the utility of zebrafish as a model vertebrate for analysis of Pol III function .
RNA Polymerase III ( Pol III ) is a 17-subunit complex that is responsible for the transcription of small noncoding RNAs such as transfer RNAs ( tRNAs ) , 5S ribosomal RNA ( rRNA ) , U6 small nuclear RNA ( snRNA ) , 7SL RNA , and others in eukaryotes [1 , 2] . The two largest subunits , Rpc1 ( ∼160 kDa ) and Rpc2 ( ∼130 kDa ) , are highly homologous to their counterparts in Pol I and Pol II , and together provide a large surface area for interaction with many of the other subunits [2] . Structural analyses of Pol III complexes [3 , 4] , together with two-hybrid analysis [5] , have identified multiple subunit interactions ( reviewed in [1] ) . These , together with biochemical and genetic analyses , have led to a model that attributes some of the unique functions of Pol III , including its high processivity , efficient transcription termination and recycling activity , RNA 3′ cleavage activity , and interaction with diverse promoters , to specific individual subunits . Mutational analyses in yeast clearly show that an intact Pol III system is essential for cell growth . The effects of reduced Pol III function are predicted to be broad , including protein synthesis necessary for cell-cycle progression ( tRNAs ) , ribosome biogenesis ( 5S rRNA ) , mRNA splicing ( U6 snRNA ) , and membrane targeting of newly translated proteins ( 7SL RNA ) . Pol III transcription is tightly regulated during the cell cycle [6] and in response to cellular stress [7] . Recent studies in human cells have also highlighted the roles of oncogenes and tumor suppressors such as Rb [8 , 9] , p53 [9–11] , and cMyc [9 , 12] in controlling the interactions between the transcription factors that bring the Pol III complex to the promoters of its target genes ( reviewed in [13 , 14] ) . Other proteins , such as Maf1 [15–18] and the oncogenic kinase CK2 [19–20] , can regulate Pol III function through direct interactions with the Pol III complex . Thus , eukaryotic cells have evolved multiple independent mechanisms for regulating Pol III activity . Given the importance of Pol III for cell growth and proliferation , it is not surprising that it is deregulated in cancers and in cells transformed by viral oncoproteins [13 , 14] . These findings suggest that it may be possible to disrupt transformed cells by inhibiting Pol III function . It is not known , however , whether Pol III inhibition has deleterious effects in nonproliferating cells of complex multicellular organisms . Here , we describe the positional cloning of a mutation , slim jim ( sljm74; hereafter , slj; [21–23] ) , that targets polr3b , the zebrafish ortholog of a yeast Pol III subunit gene rpc2 , which is highly conserved in eukaryotes . tRNAs and other Pol III transcripts are decreased in slj larvae . Accordingly , the slj mutation has a pronounced effect on the growth and proliferation of progenitor cells in the digestive tract of slj larvae , but surprisingly , no overt effect on the survival of cells in other nonproliferating mutant tissues at the same developmental stage . Also unexpectedly , the slj mutation does not interfere with differentiation of most epithelial lineages in the developing zebrafish intestine , but has a profound effect on epithelial cell morphology [22] . These data surprisingly indicate that specific cell types in the developing fish are differentially sensitive to the slj mutation and suggest that this may be due to different requirements for Pol III activity . Structural comparisons of yeast Pol II and Pol III suggest that the slj mutation might perturb interaction of Rpc2 with Rpc11 , an integral Pol III subunit that exhibits RNA 3′ cleavage activity and is required for efficient transcription recycling by Pol III [24–26] . Supporting this idea , we show that the Rpc11p protein is not present in the Pol III complex purified from S . pombe engineered to carry an Rpc2 deletion mimicking the polr3bslj allele . Microinjection of an overexpression construct encoding the zebrafish Rpc11 ortholog , Polr3k , suppresses the exocrine defects in slj larvae , indicating that the slj phenotype is due to deficiency of the Polr3b–Polr3k ( Rpc2–Rpc11 ) interaction , and supporting broad conservation of Pol III structure in eukaryotes .
The zebrafish slj mutation was recovered in a large-scale mutagenesis screen on the basis of altered intestinal morphology [21] . Compared to wild-type five-day post-fertilization ( dpf ) larvae , the intestine of 5-dpf slj larvae is small , thin walled , and lacks folds ( Figure 1A and 1B ) . In histological sections , the slj intestinal epithelium appears immature compared with wild-type siblings ( Figure 1C and 1D ) . Development of the exocrine pancreas is also severely affected by the slj mutation [23] . Little , if any , exocrine tissue is visible in histological sections of 5-dpf slj larvae ( Figure 1G and 1H ) . By contrast , the pancreatic islet , which undergoes little or no expansion beyond the first 48 hours post-fertilization ( hpf ) appears normal in slj ( Figure 1G and 1H ) . To investigate the underlying cause of this feature of the slj phenotype , we measured slj intestinal and pancreas cell proliferation using bromodeoxyuridine ( BrdU ) and phospho-Histone H3 ( PH3 ) immunohistochemistry . The BrdU assay revealed a nearly 2-fold reduction in the proportion of S-phase cells within the slj intestinal epithelium ( InE; Table 1 ) and a nearly 5-fold reduction of the proportion of S-phase cells in the developing exocrine pancreas ( ExP; Table 1 ) around the stage when the slj phenotype is first recognizable ( 72 hpf ) . By contrast , cell proliferation within the intestinal stroma was not affected by the slj mutation at this stage . Reduced proportion of intestinal epithelial and pancreatic S-phase cells , coupled with the normal percentage of M-phase cells at this stage ( as determined by anti-PH3 immunohistochemistry ) , suggest the slj mutation causes a delay in the G1–S transition within highly proliferative organ progenitor cells . Consistent with this idea , the size of other highly proliferative tissues , such as the liver , retina , and terminal branchial arches , was also reduced in slj larvae ( Figure 1E and 1F , and unpublished data ) . To gain a better understanding of the cause of the proliferative defects associated with the slj phenotype , a positional cloning strategy was used to identify the targeted gene . Using bulk segregant analyses , we first identified a marker on Chromosome 18 linked to the slj locus ( M . Mohideen , M . Fishman , and M . Pack; unpublished data ) . Mapping of subsequent markers identified a critical region surrounding the slj locus as described in Materials and Methods and Figure 2A . Ultimately , a bacterial artificial chromosome ( BAC , zk103i16 ) spanning two flanking simple-sequence repeat markers ( z15417 and z6098 ) was identified . Microinjection of the BAC DNA partially rescued slj exocrine pancreas defects ( Figure 2B; n = 6 of 17 injected slj larvae ) , thus confirming that the BAC spanned the slj locus . Sequence analysis from the zebrafish genome project identified three genes within BAC zk103i16 adjacent to the slj locus ( Figure 2A ) . Further meiotic mapping narrowed the critical interval to a region that contained the polr3b gene ( http://www . sanger . ac . uk/Projects/D_rerio ) . We then scanned pol3rb cDNA for mutations . Reverse-transcriptase PCR ( RT-PCR ) products and their sequencing identified a 123-bp deletion in polr3b cDNA amplified from slj mutant larvae , but not from homozygous wild-type larvae ( Figure 2C and 2D ) . Additional analyses indicated that the 123-bp deletion corresponded precisely to exon 10 of the polr3b gene [27] . A smaller deletion , comprising 66 nucleotides from exon 10 , was subsequently identified in a minority ( ∼10% ) of the polr3b cDNA fragments ( Figure 2E and 2F ) . Both deletions are predicted to occur in-frame , thus generating polr3b cDNAs encoding proteins truncated by internal deletion of either 22 or 41 amino acids . Exon 10 skipping induced by the slj mutation could arise from disruption of the polr3b intron 10 splice acceptor . Sequence analyses confirmed this prediction: a thymine-to-cytosine transition was present in the intron 10 splice acceptor of the polr3bslj , but not the wild-type polr3b allele ( Figure 2G ) . This single intronic substitution causes the exon 11 splice donor to utilize either the intron 9 splice acceptor , or a cryptic splice acceptor within exon 10 , thus deleting either 123 or 66 exon 10 nucleotides from the mature polr3b mRNA ( Figure 2C and 2E ) . Since the RT-PCR and sequencing analyses shown in Figure 2E and 2F were performed on whole embryos , the data indicate that the slj mutation leads to altered splicing with in frame codon deletions in the vast majority , if not all , of the polr3bslj transcripts . To confirm that the polr3b mutation was responsible for the slj phenotype we designed antisense Morpholinos that targeted the polr3b mRNA ( Figure 3 ) . Microinjection of a Morpholino targeting the translation initiation codon ( 5′ ATG ) led to early lethality ( prior to 24 hpf ) in the majority of injected embryos , with the remainder showing severe developmental delays ( unpublished data ) . Injection of a lower dose of this Morpholino produced an slj intestinal and pancreatic phenocopy in approximately 50% and 60% of surviving wild-type 5-dpf larvae ( n = 77 and 47 embryos analyzed , respectively , in two independent experiments; Figure 3A–3C and 3E–3G ) . We also designed a Morpholino spanning the intron 9 splice donor with the hope that targeting this site would induce deletion of exon 10 . Indeed , RT-PCR and DNA sequence analysis of the polr3b cDNA derived from embryos microinjected with this Morpholino , but not control-injected embryos , revealed in-frame deletion of exon 10 of the polr3b cDNA ( Figure 3I ) . Importantly , 52% and 43% of surviving embryos injected in two independent experiments showed an slj intestinal and pancreatic phenocopy ( n = 33 and 53 total embryos analyzed , respectively; Figure 3D and 3H ) . In summary , mutant phenocopy by two nonoverlapping Morpholinos confirm identification of polr3b as the gene targeted by the slj mutation . To define the location and levels of polr3b expression in developing zebrafish embryos and larvae , we performed RNA whole-mount in situ hybridization assays and quantitative real-time RT-PCR amplification of the polr3b cDNA ( Figure 4 ) . These data showed strong maternal polr3b expression ( Figure 4A ) and strong zygotic polr3b expression that peaked at 24 hpf and subsequently declined ( Figure 4B–4F ) . Beyond 3 dpf , we observed only low levels of polr3b expression in the digestive organs , that were only slightly above background and thus difficult to image ( unpublished data ) . This decline in polr3b expression coincides with 5-fold reduced cell proliferation in the digestive system and other tissues between 2 dpf and 4 dpf [22] . This relatively low level of Polr3b expression may sensitize cells to the slj mutation , and provide an explanation as to why the digestive system and other proliferative tissue may be selectively affected in slj mutants ( see Discussion ) . To determine whether the slj mutation disrupted Pol III function , we quantified the levels of Pol III target gene RNAs in wild-type and slj larvae . We used an Agilent Bioanalyzer for high-resolution analysis of total RNA ( see Materials and Methods ) . This revealed decreased levels of total tRNAs in 4-dpf and 5-dpf slj larvae , but normal 5S and 5 . 8S rRNA levels ( Figure 5A ) . Normal 5 . 8S RNA levels in slj larvae ( unpublished data ) were expected because this gene is transcribed by Pol I . Discordant effects of the slj mutation on 5S rRNA and tRNA gene expression , which are both transcribed by Pol III , were also not surprising , given similar findings with mutation of yeast Pol III subunit genes [28] . Such differences in the sensitivity of Pol III target genes has been attributed to variable specificity of the Pol III complex for its target gene promoters [28] . Such specificity may also explain the variable effects of the slj Pol III defect on tRNA levels ( discussed below ) . However , it is also possible that reduced 5S rRNA transcription and transcription of specific tRNAs in slj larvae is compensated by enhanced transcript stability . To confirm these data , we assessed the effect of the slj mutation on individual Pol III target genes . Quantitative RT-PCR experiments showed that there were reduced levels of a tRNA-leu in 3-dpf and 5-dpf slj larvae relative to the levels of the Pol II target gene hprt ( Figure 5B ) . Interestingly , tRNA-leu levels were normal in 4-dpf slj larvae , thus revealing a complex relationship between Pol III activity and target gene transcription . Expression of the 7SL RNA , a component of the signal recognition particle that is also transcribed by Pol III [29] , was also reduced in 4-dpf and 5-dpf slj larvae ( Figure 5C ) . Reduced levels of mature tRNA-leu and a tRNA-ile were also confirmed by northern analyses of 5-dpf slj larvae . Levels of these tRNAs were 48% and 36% , respectively , of their sibling wild-type larvae relative to the expression of the Pol II transcript U1 snRNA on the same blot ( Figure 5D ) . Reduced levels of mature tRNAs suggest reduced efficiency of the slj Pol III enzyme , but do not exclude post-transcriptional effects [25 , 30 , 31] . Because of the rapid rate at which pre-tRNAs are processed , they are considered to be reliable indicators of Pol III transcription rate [16 , 32] . Therefore , we used northern blot analysis and quantitative PCR to measure pre-tRNA levels in slj and sibling wild-type larvae , relative to the Pol II transcribed genes U1 snRNA and hprt mRNA ( Figure 5E and 5F , respectively ) . The levels of both of the pre-tRNAs examined were reduced in 5-dpf slj larvae compared with wild-type siblings , supporting the idea that Pol III transcription is reduced in slj mutants . Although high-resolution analysis of the 17 subunit Pol III structure has not been defined , a homology-based model of the nine-subunit core yeast Pol III complex and the structure of an associated subcomplex have been reported [3] . In addition , the 17-subunit complex has been visualized , with some of its subcomplexes localized by electron microscopy [4] . The Pol III structure derived from these studies and the original studies describing the Pol II structure [33 , 34] suggest a high degree of homology , especially in the largest subunits and their contacts with other subunits ( most of which are either shared by Pol II and III , or highly homologous [3 , 4] ) . Since Rpc2p is highly homologous to the second largest Pol II subunit , Rpb2p , the Pol II structure provided a guide to how the slj deletion might affect Pol III function . Examination of the yeast Pol II structure revealed that the region of subunit Rpb2p corresponding to the Rpc2 slj deletion makes contact with Rpb9p ( discussed in [4] ) , a subunit homologous to the Pol III subunit , Rpc11p [24] . This suggested the possibility that the region deleted in slj Polr3b might contact the zebrafish Rpc11p ortholog , Polr3k . This idea is supported by two hybrid studies of yeast Pol III that show interaction between Rpc11p and the N-terminal region of Rpc2p [5] . To further explore whether the slj Polr3b deletion might compromise interaction with Polr3k , we engineered S . pombe to express a mutant Rpc2p with a deletion ( Rpc2-Δp ) corresponding to the evolutionarily conserved region deleted by the polr3b slj mutation , and examined its subunit interactions . Hemagglutinin ( HA ) -tagged versions of wild-type or mutant S . pombe rpc2 were introduced into an S . pombe strain containing a His6-FLAG-tagged Rpc53p subunit , and Pol III complexes containing the tagged subunits were recovered by sequential affinity immunoprecipitation as described in Materials and Methods . Western blot analysis of multiple subunits revealed a markedly reduced amount of Rpc11p in the Pol III complex containing Rpc2-Δp ( Figure 6A , lane 8 ) relative to wild-type Rpc2p ( Figure 6A , lane 7 ) . This experiment supports the idea that exon 10 of zebrafish polr3b encodes amino acids crucial for the stable interaction of the zebrafish orthologs of Polr3b and Polr3k ( Rpc2 and Rpc11 ) , and suggests that the slj phenotype might be due , at least in part , to instability or failure of this interaction in vivo . Pol III subunit interactions in yeast have largely been defined via two-hybrid analyses or the more functional approach , overexpression–suppression experiments . The latter approach involves suppression of hypomorphic phenotypes by overexpression of a gene encoding an interacting subunit . To adapt this approach to the possibility that the slj mutation might affect the efficiency of the Polr3b–Polr3k interaction , we injected zebrafish polr3k cDNA driven by the heat-inducible hsp70 promoter , which drives high levels of gene expression throughout the embryo ( albeit in a mosaic fashion; [35] ) in transient expression assays . This led to partial but highly significant rescue of the slj exocrine pancreas defect ( Figure 6B; n = 8 of 13 and n = 9 of 9 injected slj larvae rescued from two independent experiments ) . This rescue , which in most larvae was more pronounced than that achieved with the BAC injection ( Figure 2B ) , is consistent with data on Rpc11p in yeast two-hybrid analyses of Rpc2p-Rpc11p , and the predicted interaction between these subunits in the structures noted above . Suppression of the slj phenotype by microinjection-overexpression of Polr3k provides strong evidence to indicate that the slj phenotype is due , in significant part , to the deficiency of a stable interaction between Polr3k ( Rpc11 ) and Polr3b ( Rpc2 ) in vivo . Polr3k-mediated rescue of the slj exocrine pancreas defects suggests that a decrease in the concentration of Polr3k may produce a phenocopy of the slj phenotype . To explore this hypothesis , we again used antisense Morpholino-mediated knockdown in developing zebrafish , this time targeting Polr3k . Consistent with this idea , injection of a Morpholino targeting the translation initiation codon ( 5′-ATG ) of the zebrafish polr3k mRNA generated an slj phenocopy in approximately 42% of injected embryos ( n = 31 and 43 embryos analyzed in two independent experiments; Figure 6C ) . By contrast , control Morpholinos did not generate a slj phenocopy in any of the injected embryos ( unpublished data ) .
Mutants targeting yeast Rpc2p had been isolated by a genetic screen that selected for impaired termination by Pol III , and were later found to affect elongation rate with reciprocal effects on termination [42–45] . Because the slj mutation was adjacent to some of these yeast rpc2 mutations , and because rpc11 was linked to Pol III termination and RNA 3′-end formation [24 , 25] , we wanted to know whether termination was affected in slj mutants . A minimal efficiency vertebrate Pol III terminator consists of a run of four thymidine ( T ) residues , with termination efficiency increasing as the number of T residues increases [46] . Transcription beyond a weak Pol III terminator in vivo can be visualized on northern blots as pre-tRNA transcripts that extend beyond the 3′ terminator into flanking DNA [47] . To examine this possibility , we identified a tRNAile gene with a minimal 4T terminator and examined its transcripts in wild-type and slj mutant fish . Using an intron probe as well as a 3′ flanking probe designed to detect read-through transcripts that extend beyond the terminator , we found no evidence of a difference in the Pol III termination efficiency of this gene in wild-type and slj larvae , since the pre-tRNA-Ile transcript size was identical in slj and wild-type larvae , with no evidence of longer transcripts on the blot ( Figure 5E and unpublished data ) . These data are consistent with the finding that function-altering mutations in rpc11 did not affect termination efficiency in fission yeast [25] and a revised role for Rpc11p in Pol III recycling rather than termination per se [26] . Recycling , or facilitated reinitiation , is a feature of the high efficiency of Pol III [26] . Given evidence of no impairment of termination by slj Pol III , we speculate that reduced reinitiation may be the cause of decreased Pol III transcription in slj mutants . Contrary to what might be expected from the study of a Saccharomyces cerevisiae rpc11 mutant that contained wild-type S . pombe Rpc2p [24] , Pol III purification from S . pombe revealed similar Rpc53p levels in the wild-type and slj Rpc2-Δp cells ( Figure 6A ) . This may be explained by the fact that the genetic approaches differed ( small deletion in Rpc2 in S . pombe , versus replacement of S . cerevisiae Rpc11 with S . pombe Rpc11 in S . cerevisiae ) and that very different Pol III purification schemes were used ( epitope tag affinity chromatography versus extensive ion exchange chromatography ) in the present and prior studies , respectively . In summary , the work described in this study reveals an unexpected importance of the Rpc2–Rpc11 interaction during development and demonstrates the utility of the zebrafish system . We have shown that in vivo analyses of Pol III function are feasible in zebrafish and that they complement analyses in other model systems . We have also shown that the zebrafish may be used to reveal the effects of disrupting Pol III in complex multicellular tissues . An idea suggested by these studies is that specific cell types within a multicellular organism may require different levels of Pol III activity and that this may reflect their rate of proliferation and/or growth . Thus , it is conceivable that selective Pol III inhibitors may be able to target metabolically active cells in proliferative or hypertrophic disease states .
The slj mutation bulk-segregant analyses identified two Chromosome 18 markers linked to the slj locus ( M . Mohideen , M . Fishman , and M . Pack; unpublished data ) . Subsequent analyses identified two closely linked markers , z15417 and z21330 , that were 0 . 08 cM and 0 . 02 cM from the slj mutation ( three recombinants out of 3 , 612 meioses and one recombinant out of 4 , 186 meioses ) . A BAC clone zK130I16 spanning the critical region bounded by these markers was identified . Within this BAC , two additional polymorphic markers , CA11 and GT14 , were identified . Meiotic mapping showed that one and zero of 1 , 806 slj larvae were recombinant for these markers . Sequence analyses indicated that the two markers , CA11 and GT14 , were located within intron 23 and intron 10 of the polr3b gene , respectively . To verify that BAC clone zk103i16 spanned the slj locus , phenotype rescue experiments were performed . BAC DNA was prepared using a commercially available kit ( PSI Clone BAC DNA kit; Princeton Separations ) . The BAC DNA ( 1 nl of 12 . 5 ng/μl in 0 . 1% phenol red ) was microinjected into the progeny of a slj/+ intercross at the one-cell embryonic stage . Equivalent numbers of embryos were injected with phenol red solution as control . The embryos were raised to 4 dpf and assayed by anti-carboxypeptidase A and anti-insulin immunohistochemistry as described [21–23] . Morpholinos targeting the polr3b and polr3k mRNAs were injected into newly fertilized one-cell to four-cell stage zebrafish embryos as previously described [48] . The sequences of the Morpholinos are: polr3k ATG: CAGGAGCATTTTCAAACAGTCATAG polr3k control: TTCAAGTTTCATTTGTTTACCTGCA polr3b ATG: TTTCCCCGAACTCCTCTTGCAGCAT polr3b splice: TCCACTCCCATAGCCTGACGAAAGA For the polr3k expression construct , PCR primers were designed that were complementary to the 5′ and 3′ regions of the zebrafish polr3k ortholog identified in the zebrafish database . The forward and reverse primer sequences are GATATCGTTTGAAAATGCTCCTGTTTTG and ACTAGTAAGTGAATGATCTGGTTATGC . The primers were used to amplify cDNA derived from wild-type 5-dpf zebrafish larvae . The predicted Polr3k protein consists of 108 amino acids , of which 102 are either identical or similar to the mouse Polr3k protein . The cDNA was cloned downstream of the zebrafish hsp70 heat shock promoter [48] . The construct was microinjected into the progeny of heterozygous slj/+ matings at the one-cell stage . Heat shock was performed approximately every 8 h beginning at 48 hpf through 84 hpf . Heat-shocked and control larvae were processed for carboxypeptidase A immunohistochemistry as described [22] . To assemble the zebrafish polr3b cDNA sequence containing the longest open reading frame , a comparative analysis of the predicted amino acid sequences of zebrafish , human , and mouse Polr3b proteins was performed . This showed that zebrafish Polr3b protein had 98% and 96% similarity to the human and mouse Polr3b proteins , respectively ( unpublished data ) . Primers ( RPC2-1FA and RPC2-2RB ) flanking the coding region of zRPC2 were designed from the conserved cDNA sequences and used for RT-PCR with total RNA derived from 5-dpf zebrafish larvae . A 3 , 514-bp fragment was amplified and sequenced . The comparison of the cDNA sequence of zebrafish polr3b to the corresponding genomic sequence indicated that this gene consists of 28 exons and 27 introns spanning a 38 . 5-kb genomic sequence . The longest open reading frame is 3 , 393 bp , and the predicted protein consists of 1 , 131 amino acids with a molecular weight of 127 . 8 kDa . Sequence analysis of the polr3b cDNA from separate pools of homozygous wild-type and slj larvae revealed complete or partial deletion ( 121 and 63 bp , respectively ) of sequence derived from exon 10 encoded by the slj and not the wild-type polr3b allele . Sequence analyses of exon 10 splice donor and intron 10 splice acceptor revealed a thymidine-to-cytosine transition of genomic fragments amplified from slj larvae , but not homozygous wild-type sibling larvae . BrdU and PH3 immunohistochemistry were performed as previously described [21 , 22] . Cell counts were performed using histological sections of whole-mount specimens . For BrdU counts , the number of intestinal epithelial or stromal cells analyzed per embryo ranged between 350 and 775 . For the exocrine pancreas , 50 cells per embryo were counted . Three wild-type and three slj larvae were analyzed for each time point . Histology and whole-mount RNA in situ hybridizations were performed as previously described [21 , 22] . Quantitative real-time PCR ( RT-PCR ) was performed as previously described [49] . RT-PCR was performed using total RNA derived from 30 or more pooled 2–5-dpf wild-type and sibling slj larvae that were identified either morphologically or molecularly . For measurements of total tRNA , 5S rRNA , and 5 . 8S rRNA levels , total RNA was separated electrophoretically using an Agilent Bioanalyzer . The chromatographic image was analyzed with image analysis software ( NIH Image 3 ) for transcript quantification . For northern analyses , total RNA from pooled wild-type and sibling slj larvae were used using standard techniques . Primers used in quantitative PCR experiments are: tRNAleu: GTAGGATGAACTGAGTTTTAA; AAAGGCAGAAGAGAACTGGTTTATT 7SL RNA: TTCGGTATCGATATGGTGCTC; GCTTTGACCTGCTCCGTCT pre-tRNAleu: AGAATGGCCGAGTGGTCTAA; CCAGCTGGAGACCAGAAATC pre-tRNAile: CGCGCGGTACTTATAAGACAAT; GAACTCACAACCTCGGCATT . The S . pombe FLAG-tagged version of rpc2+ was cloned by PCR using genomic DNA and two primers ( SPREPFOR1: 5′-GCTAGTCGACATGGATTACAAAGACGATGACGACAAGGGGGTAAATACTGC;5′-GCTAGTCGACATGGATTACAAAGACGATGACGACAAGGGGGTAAATACTGC; and SPREPREV: 5′-ATACCCGGGTCAATACTTAAATTCGT ) . The PCR products were digested with SalI and SmaI , and cloned between the XhoI and SmaI sites of pREP3X , yielding pREP3X-rpc2 . The region spanning amino acid ( aa ) 267 and aa308 of Rpc2p , which corresponds to aa240 and aa380 of the zebrafish Rpc2 ortholog , Polr3b , which was deleted in the slj mutant , was deleted by the QuikChange XL Site-Directed Mutagenesis ( Stratagene ) using pREP3X-rpc2 and two primers ( RPC2F: 5′-CAGTGTAGCAGATGATATTCCTATAGTGGTTGTTTTAAAAGCATTAGAATATATCGGTGCGCGTGTTAAGG; and RPC2R: 5′-CCTTAACACGCGCACCGATATATTCTAATGCTTTTAAAACAACCACTATAGGAATATCATCTGCTACACTG ) . The resulting Rpc2 mutant was called Rpc2-Δp . To replace a FLAG tag with an HA tag in Rpc2p and Rpc2-Δp , PCR was performed with primers retremfor3 ( 5′-ACGGTCGACATGTAC CCATACGACGTTCCAGACTACGCTGGGGTAAATACTGCC GGA ) and SPREPREV . The PCR products were cut with SalI and SmaI , and cloned between the XhoI and SmaI sites of pREP3X , producing pREP3X-HA-rpc2 and pREP3X-HA-rpc2-Δp , respectively . To purify wild-type Pol III and Pol III mutants containing Rpc2-Δp , pREP3X-HA-rpc2 , and pREP3X-HA-rpc2-Δp were transformed into yYH3282 ( h+ , his3-D1 , leu1–32 , ura4-D18 , ade6-M216 ret1D::[FH-rpc53 , ura4+] ) , in which rpc53 is tagged with FLAG and six histidine residues . Purification of wild-type and mutant Pol III was carried out as previously described [50] with the following modifications: the eluate from the Ni-NTA column was incubated with 20 μl of anti-HA beads for 4 h at 4 °C; the bound proteins were eluted by boiling the beads in 2× tris-glycine SDS gel loading buffer .
The GenBank ( http://www . ncbi . nlm . nih . gov/Genbank ) accession numbers for the proteins discussed in the paper are as follows: human Polr3b ( NM_018082 ) and mouse Polr3b ( NM_027423 ) . | The transmission of genetic information from DNA to messenger RNA to protein depends on the function of a large number of small noncoding RNA molecules . The genes encoding these RNAs are transcribed by RNA polymerase III ( Pol III ) , a 17-subunit protein complex whose structure is closely related to that of RNA polymerases I and II . Here , we report the effect of a mutation in a gene encoding one Pol III subunit , Polr3b , which disrupts proliferation and growth of tissue progenitor cells in the zebrafish digestive system . Analyses of a nearly identical mutation in the yeast S . pombe gene encoding Polr3b , also known as Rpc2 , suggested that the zebrafish mutation disrupted the mutant Polr3b protein's interaction with another Pol III subunit , Polr3k , also known as Rpc11 . Overexpression of the gene encoding Polr3k in the Polr3b mutants partially rescued ( reversed ) the mutant phenotype . These findings extend our knowledge of the mechanism of Pol III function , which appears to have been highly conserved during eukaryotic evolution . Furthermore , these data also suggest that assembly of the 17-subunit Pol III enzyme is a dynamic process , since Polr3k overexpression can partially rescue the mutant phenotype . Understanding how Pol III is assembled has implications for human disease , since Pol III activity is markedly increased in most cancers . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
]
| [
"molecular",
"biology",
"developmental",
"biology"
]
| 2007 | Mutation of RNA Pol III Subunit rpc2/polr3b Leads to Deficiency of Subunit Rpc11 and Disrupts Zebrafish Digestive Development |
The extent and strength of epistasis is commonly unresolved in genetic studies , and observed epistasis is often difficult to interpret in terms of biological consequences or overall genetic architecture . We investigated the prevalence and consequences of epistasis by analyzing four body composition phenotypes—body weight , body fat percentage , femoral density , and femoral circumference—in a large F2 intercross of B6-lit/lit and C3 . B6-lit/lit mice . We used Combined Analysis of Pleiotropy and Epistasis ( CAPE ) to examine interactions for the four phenotypes simultaneously , which revealed an extensive directed network of genetic loci interacting with each other , circulating IGF1 , and sex to influence these phenotypes . The majority of epistatic interactions had small effects relative to additive effects of individual loci , and tended to stabilize phenotypes towards the mean of the population rather than extremes . Interactive effects of two alleles inherited from one parental strain commonly resulted in phenotypes closer to the population mean than the additive effects from the two loci , and often much closer to the mean than either single-locus model . Alternatively , combinations of alleles inherited from different parent strains contribute to more extreme phenotypes not observed in either parental strain . This class of phenotype-stabilizing interactions has effects that are close to additive and are thus difficult to detect except in very large intercrosses . Nevertheless , we found these interactions to be useful in generating hypotheses for functional relationships between genetic loci . Our findings suggest that while epistasis is often weak and unlikely to account for a large proportion of heritable variance , even small-effect genetic interactions can facilitate hypotheses of underlying biology in well-powered studies .
The relevance of epistasis in genetic architecture is yet unresolved . In genetic screens of model systems , the evidence for genetic interaction is abundant [1–4] and has been proven biologically relevant [5–7] . However , the situation is less clear in human populations as epistasis is difficult to detect with confidence due to multiple testing across a high number of variants , underpowered samples , evolutionary history , imperfect model selection , or unmeasured confounding variables or noise [8 , 9] . While most studies detect only additive variance , recent studies have demonstrated a role of epistasis in the genetics of gene expression [10 , 11] and occasionally link genetic interactions to disease [12] . Thus the extent to which genetic interactions contribute to unexplained variance or provide biological insight in population-based studies is unclear . One strategy to address this problem is controlled experiments in mammalian model systems , in which genotypes are artificially determined and environmental variation is minimized [13–16] . In this work , we used a multi-trait strategy to investigate the role of epistasis in regulating complex traits in a large mouse intercross . Bone mineral density ( BMD ) is a complex trait regulated by the interaction of many genetic and environmental factors [17–19] , and is the best known surrogate measure of fracture risk in patients with osteoporosis [20–22] . Human and rodent studies have implicated many candidate quantitative trait loci ( QTL ) in influencing BMD [17–19 , 23–27] . Many of these loci are pleiotropic and have been found to influence body weight [28] , body fat [29–31] , and bone size [32] in addition to bone density . Epistatic interactions are also common among loci affecting BMD [26 , 33 , 34] . A deep understanding of the genetic regulation of BMD , as well as possible intervention points for therapeutics , requires addressing this complex genetic architecture . Assessing genetic interactions and their relation to multiple phenotypes provides an overall picture of the genetic network regulating BMD and related phenotypes . We used a recently developed method , Combined Analysis of Pleiotropy and Epistasis ( CAPE ) [35] , to integrate information across multiple phenotypes to infer directed genetic interactions between loci . We integrated genetic interactions influencing BMD , femoral circumference , body weight , and body fat percentage in a mouse intercross population [36–39] . In particular , we were interested in investigating the genetic architecture of these phenotypes in a population with reduced levels of circulating insulin-like growth factor I ( IGF1 ) . IGF1 is a major factor involved in bone development and mineralization [40–43] . Analysis in a population with reduced IGF1 can reveal aspects of bone density that vary at severely reduced levels of IGF1 [44] , thereby unmasking more subtle genetic loci involved in this phenotype [45 , 46] . To avoid major effects of the IGF1/growth hormone ( GH ) axis we used mice homozygous for the “little” or lit mutation [47] , a null mutation in the gene coding for growth hormone releasing hormone receptor ( Ghrhr ) . GH levels , and consequently circulating IGF1 levels , in mice homozygous for the lit mutation are reduced to about 10% of wild type levels [37 , 47 , 48] . These mice also exhibit reduced growth , increased fat mass , and decreased bone mass relative to heterozygotes and wild type mice [49] . Thus lit homozygotes offer the opportunity to study the genetics related to both bone growth and body fat composition in a population in which one of the major hormonal axes regulating these phenotypes is greatly reduced . A population of 2054 F2 male and female mice derived from a cross between B6-lit/lit and C3 . B6-lit/lit parental strains [36–39] were analyzed . Compared to B6 mice , C3H mice have 20-30% higher circulating IGF1 levels , higher volumetric bone density , higher rates of bone formation , lower rates of bone resorption , and greater breaking strength of bones [50–52] . These strain differences persist in the lit/lit homozygotes [52] . We investigated the genetic interactions influencing body weight , percent body fat , femoral circumference and femoral density in the near absence of one of the major contributors to these phenotypes .
All animal procedures followed Association for Assessment and Accreditation of Laboratory Animal Care guidelines and were approved by Institutional Animal Care and Use Committee ( The Jackson Laboratory , Protocol #99111 ) . Inbred mouse strains used in this study were obtained from our research colonies at The Jackson Laboratory , Bar Harbor , Maine . Mice were produced and housed as described in [76] . Briefly , the mice were housed in same-sex groups of 2-5 per cage in a 14:10 light:dark cycle . The mice had free access to acidified water ( pH 2 . 5 with HCl to retard bacterial growth ) and irradiated NIH 31 diet ( Purina Mills International , Brentwood , MO ) . To investigate heritable factors that control BMD in a model where circulating IGF1 levels are reduced , we used a spontaneous mouse mutation , lit , with a non-functional growth hormone releasing hormone receptor ( GHRHR ) . We generated a congenic strain by transferring the lit mutation from the low-BMD C57BL/6J ( B6 ) strain on which it arose to the high-BMD C3H/HeJ ( C3H ) strain by backcrossing for eighteen generations . In both C57BL/6J-Ghrhrlit/lit/J ( B6-lit/lit ) and C3H . B6-Ghrhrlit/lit/J ( C3 . B6-lit/lit ) mice , circulating GH is undetectable , serum IGF1 is low , and femoral volumetric BMD by pQCT , femur length , and body mass are reduced compared to heterozygous lit/+ mice [37 , 47 , 48] . Although C3 . B6-lit/lit mice are of the same body weight and femur length as B6-lit/lit mice , C3 . B6-lit/lit mice have higher BMD . Crosses between B6-lit/lit and C3 . B6-lit/lit F1 mice produced the 1008 male and 1062 female F2 GH/IGF1 deficient mice analyzed here . Mice were genotyped at 100 markers using PCR of oligonucleotide primer pairs ( MIT markers , www-genome . wi . mit . edu/cgi-bin/mouse/index ) from Research Genetics ( Birmingham , AL ) as described in [76 , 77] . The pairs amplified strain-specific sequence length polymorphisms , allowing identification of parental strain of origin . Genotypes at each locus were identified as B6/B6 , B6/C3H , or C3H/C3H . CAPE is a strategy that detects epistasis and interprets it in terms of directed enhancing and suppressing influences between genetic loci [81] . It has been released as an R package suitable for mouse intercross studies [35] . The method uses regression on pairs of loci to detect interaction effects from each locus pair on each phenotype . It then combines model parameters across phenotypes to infer directed , QTL-to-QTL influences that replace the interaction effects on each phenotype . The result is a pair of directed coefficients modeling how the two loci influence each other’s activity , rather than how each pair independently affects each phenotype . We began the analysis by using R/qtl [82] to impute psuedomarkers in between each measured marker , increasing the number of markers from 100 to 194 . We normalized all traits ( body weight , body fat percentage , femoral density , and femoral circumference ) using rank Z normalization . We then decomposed the normalized traits using singular value decomposition ( SVD ) to obtain orthogonal eigentraits ( ETs ) that combined common signals across all phenotypes . All markers were used in the pair-wise interaction scans . However , we filtered marker pairs tested by linkage disequilibrium ( LD ) to avoid false positive interactions . We excluded all pairs with genotype Pearson correlation r ≤ 0 . 5 . This reduced the number of pairs tested from 19 , 110 ( all pairs from 194 markers and 2 covariates ) to 18 , 576 pairs . Linear regression was then performed on all filtered pairs of markers 1 and 2: U i j = β 0 j + ∑ c = 1 2 x c , i β c j ︸ covariates + x 1 , i β 1 j + x 2 , i β 2 j ︸ main effects + x 1 , i x 2 , i β 12 j ︸ interaction + ϵ i j where U represents ETs , and ϵ is an error term . The index i runs from 1 to the number of individuals , and j runs from 1 to 3 ( the number of ETs . ) xi is the probability of the presence of the C3H allele for individual i at locus j . For each pair of markers , the regression coefficients are collected for all ETs and reparametrized to two new terms ( δ1 and δ2 ) . These terms represent the additional activity of each variant when the other is present . For example , δ1 is the additional effect that variant 1 has on each phenotype when variant 2 is present . It should be noted that the δ terms describe the interaction coefficient between the marker pair independent of phenotype . The δ are determined from the pairwise regression parameters via pseudoinversion as follows: δ 1 δ 2 = β 1 1 β 2 1 β 1 2 β 2 2 β 1 3 β 2 3 - 1 · β 12 1 β 12 2 β 12 3 To convert the δ terms to directed influence variables , the following transformation was applied: δ 1 = m 12 ( 1 + δ 2 ) , δ 2 = m 21 ( 1 + δ 1 ) The sign of the m12 and m21 terms indicates how each marker influences the other in terms of enhancement ( positive ) or suppression ( negative ) . A negative value for m21 indicates that variant 1 reduces the activity of variant 2 on all phenotypes , whereas a positive value increases the activity . To estimate variances of the new model parameters , error estimates were propagated via second-order Taylor expansion [81 , 83] . Permutation tests were used to calculate p-values for all model parameters . The pair of markers being tested was permuted together relative to the covariates [84] . By combining permutations across all marker pairs , we saved computation time while generating a single null distribution indistinguishable from the null distribution generated by repeatedly permuting each single pair . This null distribution was composed of 164 , 679 total permutations representing 3 permutations for each of 54 , 893 locus pairs . Main effect significance was also determined through permutation testing . The maximum main effect of each locus across all pairwise contexts was selected as the main effect for that locus . All p-values were corrected for multiple testing using the Holm step-down procedure [85] and all variant-to-ET main effects were translated to variant-to-phenotype effects through multiplication by the singular value matrices from the original SVD . To define distinct QTL regions for the interaction network , adjacent markers were combined into linkage blocks . We calculated the correlation matrix for all markers on each chromosome . We used this similarity matrix as an adjacency matrix to construct a weighted network depicting the similarity between all pairs of markers on a single chromosome . Using the fastgreedy community detection algorithm [86] in R/igraph [87] , we then calculated the community membership of the vertices in the network . We assigned adjacent markers in the same community to a single QTL region . This process ensured a robust grouping of markers based on their genotypic similarity . A block was considered to have a significant effect on a phenotype if one or more of the resident markers had a significant effect . After this point , we refer to linkage blocks with significant associations as QTL or loci . See S3 Table for markers included in each block . Some of the groups of linked markers interacted significantly with IGF1 , which is a sufficiently specific interaction to allow a candidate gene search . For each QTL that interacted significantly with IGF1 we generated a list of potential candidate genes by finding all genes in the QTL with annotations for bone density . To determine whether any of these genes interacted with IGF1 , we used the Integrative Multi-species Prediction tool [88] to generate a network between the query genes and IGF1 . We included the maximum of 50 additional genes and adjusted the confidence of the interactions until IGF1 was included in a subnetwork of greater than two genes . From this subnetwork , we extracted genes in the original QTL region , but which were not necessarily included the original query relating to bone density . Candidate polymorphisms in C3H alleles of these genes were identified using the Sanger SNP database [54] . Furthermore , we used previously published expression data from hepatic tissue of chow-fed B6 and C3H mice [56] to identify genes that were differentially expressed between the strains and pairs of genes that had correlated expression . Because most observed regulation of gene expression is in cis , it is reasonable to assume that the genetic variation that influences gene expression will influence expression levels similarly wherever the gene is expressed . Differential expression was determined with Student’s t-test , and significance of correlation was the significance of the Pearson correlation coefficient [89] . We examined the enrichment of three-node topological patterns , or network motifs [58] , in the set of all interactive genetic models . Although each interaction and main effect was independently derived , we grouped the significant effects into a network to detect general patterns . We combined interactions with main effects to generate three-node motifs , which included two interacting variants and a single phenotype . We counted enhancing and suppressing motifs either with two main effects of the same sign ( coherent ) , or two main effects of the opposite sign ( incoherent ) . To determine whether each type of motif was significantly enriched or depleted , we performed permutations by shuffling the signs of the significant interactions independently of the signs of the main effects . This permutation scheme preserved the topology of the network thereby acknowledging constraints caused by shared edges between motifs . By permuting the main effect edges independently of the interaction edges , we prevented spuriously enriching for enhancing or suppressing interactions simply because there were many negative or positive main effects on a given phenotype . We permuted the edge signs 100 , 000 times to generate a null distribution for each type of motif . The directions of the interactions were not taken into account for this analysis . We used linkage blocks as interacting units and included all interactions and main effects that were significant at a Holm-corrected p ≤ 0 . 01 .
We used linear regression to determine the association of each locus with each phenotype ( Fig 1 , S4 Table ) . Each of the phenotypes had multiple associated QTL , and these QTL often overlapped across multiple phenotypes . For example , femoral density and femoral circumference shared a large QTL on Chr 4 , and body weight , percent fat and femoral circumference showed overlapping QTL on Chr 17 . These overlapping QTL indicate the possibility of common genetic factors underlying multiple phenotypes , such that information can be combined across multiple phenotypes to gain information about individual loci . Unique QTL were also observed , providing non-redundant information to discern genetic factors with phenotypic specificity . We decomposed the normalized , mean-centered phenotypes into eigentraits ( ETs ) ( Fig 2A ) . The first ET represented an average of femoral circumference , percent body fat and body weight and captured 52% of the overall variance . The second ET represented the contrast between femoral circumference and femoral density and captured 25% of the total variance . The third ET captured 19% of the total variance and represented a contrast between femoral circumference and percent body fat . The fourth ET captured only 4% of the total variance , and described a contrast between body weight and percent fat . Because this ET captured a small amount of the total variance , did not include strong contributions from the bone phenotypes , and may add noise to the analysis , we excluded it from this analysis . Single-locus associations with each ET detected multiple QTL ( Fig 2B ) . Since ET are linear combinations of traits , each QTL indicates a potentially pleiotropic association with varying effect strengths on each trait . For example , data for body weight , percent fat and femoral circumference had overlapping QTL on Chr 17 . These phenotypes also contributed substantially to ET1 , and there was a corresponding significant QTL for ET1 on Chr 17 representing the common QTL . CAPE analysis of the first three ETs produced a large network of genetic interactions ( Fig 3 ) . The high-confidence network ( p ≤ 0 . 0005 ) consisted of a single connected component including 67 QTL linked by 102 directed interactions . Each interaction was directed from a source locus to a target locus and was either suppressing ( negative ) , or enhancing ( positive ) . In suppressing interactions the presence of the C3H allele at the source locus reduced the phenotypic effect of the C3H allele at the target locus regardless sign of the main effect . In enhancing interactions the presence of the C3H allele at the source locus increased the phenotypic effect of the C3H allele at the target locus regardless of the sign of the main effect . The QTL were distributed across the four phenotypes as follows: body weight had 13 QTL; percent fat had 11 QTL; femoral density had 19 QTL; and femoral circumference had 24 QTL . Among the interactions between QTL , 29 were suppressing and 73 were enhancing . We also used standard linear regression to assess the effect of each marker pair interaction on each normalized phenotype . We calculated the pairwise interaction coefficients for all marker pairs including the two covariates , sex and circulating IGF1 levels . We calculated empirical p-values from 50 , 000 permutations and corrected the p-values for multiple testing using the Holm step-down procedure . No marker-marker or marker-covariate interactions were significant using the standard epistasis analysis after correction for multiple testing . Sex had significant main effects on all phenotypes . Males had substantially higher body weight ( males: 22 . 0 g , females: 17 . 5 g; p ≤ 2 × 10−16 ) and percent body fat ( males: 42% , females: 36%; p ≤ 2 × 10−16 ) . Males also had slightly , but significantly higher femoral circumference ( males: 3 . 81 mm , females: 3 . 78 mm; p = 2 . 9 × 10−4 ) . Females had significantly higher BMD ( males: 0 . 49 mg/mm3 , females: 0 . 51 mg/mm3; p ≤ 2 × 10−16 ) . We found several significant sex-QTL interactions , all of which were enhancing ( Fig 3C ) . From the single-locus regressions , potential sex-interacting loci were seen on Chrs 1 , 6 , 7 , 10 , and 14 ( Fig 1 ) . We tested these conditional associations directly and confirmed genetic interactions with sex on Chrs 1 , 7 , 10 and 14 . For example , there was a larger sex difference in femoral density among animals homozygous for the C3H allele at the Chr 1 locus than among animals homozygous for the B6 allele ( Fig 4 ) . Thus the C3H status at this locus enhanced the negative effect of the male sex on femoral density , as well as the positive effect of the male sex on circumference , giving females with this genotype increased bone density and reduced circumference relative to males . An apparent locus on Chr 6 ( Fig 1 ) was not identified as interacting significantly with sex because a consistent directional model could not be fit across all phenotypes . Several markers on Chr 6 had relatively large interaction coefficients in the linear regression with sex , but none of these interactions were significant after correction for multiple testing . Like sex , IGF1 had significant main effects on all phenotypes . Higher levels of circulating IGF1 were associated with higher body weight , body fat percentage , femoral density , and femoral circumference . IGF1 was also found to interact significantly with several QTL ( Fig 3C ) . Interestingly , all interactions from IGF1 to loci were suppressing , i . e . IGF1 suppressed the effects of these loci . For example , one Chr 9 QTL had negative effects on both femoral density and femoral circumference when IGF1 levels were low , but at high levels of IGF1 this effect was suppressed , and the QTL had a positive effect on these phenotypes ( Fig 5 ) . Conversely , this locus also enhanced the effects of IGF1 . Looking at body weight as a function of IGF1 , for example , it can be seen that in the B6 homozygotes , IGF1 levels had a positive effect on body weight . There was a slightly larger effect in the heterozygotes , and the largest positive effect of IGF1 on body weight was in the C3H homozygotes ( Fig 5 ) . This general pattern is replicated across all phenotypes . The molecular specificity of an interaction with IGF1 offered the opportunity to search for candidate genes in QTL interacting with IGF1 , even though the regions were large ( Methods ) . The second QTL on Chr 10 yielded promising candidates . Using MouseMine , we found 19 genes in the region with annotations to bone density [53] . We used these genes as a query gene set in the IMP tool . The IMP tool determines the likelihood that pairs of genes in a query set interact . It uses databases of known physical interactions , genetic interactions , and correlated expression to pull in additional genes though which genes in the query set may interact . The result is a network of high-confidence interactions that relate the query genes to each other . IMP found a network of 45 genes that linked IGF1 with four bone density-related genes ( Esr1 , Nr1h4 , Kitl , Ctgf ) from the query gene set . The minimum confidence for interactions between gene pairs in this network was 91% . All four genes contain SNPs predicted to be functionally relevant between B6 and C3H [54] , including splice site variants , missense variants and variants in regulatory regions ( S5 Table ) . One of the genes , Kitl was differentially expressed in hepatic tissue between C3H and B6 , with C3H mice showing lower expression ( Student’s t-test , p = 0 . 012 ) [55 , 56] . Kitl expression also showed a strong negative correlation with Igf1 expression ( r = −0 . 67 , p = 0 . 016 ) [55 , 56] . These findings allow us to hypothesize that Kitl is a potentially causative gene in QTL 10 . 2 . Another potential candidate in this region , Bicc1 , was recently found to be related to bone density in mice [57] . However , it was not differentially expressed in the hepatic tissue of C3H and B6 mice ( Student’s t-test , p = 0 . 25 ) ( S5 Table ) , and thus we consider it a less likely candidate in the context of this study . Other candidate regions did not reveal promising causative candidates when analyzed using the same methods . In addition to individual interactions , we examined the enrichment of three-node patterns , or network motifs [58] ( Methods ) . Each motif consists of two interacting genetic loci and an affected phenotype . At significance p ≤ 0 . 01 , we identified a total of 116 motifs influencing body weight , 84 motifs influencing percent fat , 132 motifs influencing femoral density , and 274 motifs influencing femoral circumference ( Fig 6 ) . Motifs are classified as coherent , i . e . the main effects are of the same sign , or incoherent , i . e . the main effects are of opposite signs . In addition , motifs can be either suppressing , meaning that there is a suppressing interaction between the genetic loci , or enhancing , with an enhancing interaction between the loci ( Fig 6 ) . These different classes of motifs may speak to the general structure of the underlying biological interactions [59–63] . For example , a motif with coherent main effects and a suppressing interaction between them indicates genetic redundancy and may result from proteins operating in series in the same pathway or physical interaction between gene products [60 , 61] . Alternatively , a synergistic interaction exists in a coherent motif with an enhancing interaction between the variants . In this case , the variants and the interaction between them all push the phenotype in the same direction . Such an interaction may indicate regulatory interactions between parallel regulatory pathways that affect the same process [60] . These motifs can be divided into those that stabilize phenotypes and those that destabilize phenotypes . For example , incoherent enhancing motifs tended to have a stabilizing effect on phenotype because the main effects drove the phenotype in opposite directions . This was the largest class of motif represented in our network ( 42% of total motifs detected ) and was significantly enriched across all phenotypes except femoral density ( Fig 6 ) . In contrast , coherent enhancing motifs tended to be destabilizing . The main effects of the interacting loci both drove the phenotype in the same direction , and the enhancing interaction between these loci drove the phenotype further in the same direction , generating extreme phenotypes . These enhancing coherent motifs made up only 17% of the total motifs detected and were significantly depleted across all phenotypes in our network ( Fig 6 ) . Motifs with suppressing interactions were more evenly distributed . Coherent suppressing motifs , which tended to stabilize phenotypes , made up 33% of the total motifs and were enriched for all phenotypes except percent fat . Incoherent suppressing motifs , which tended to be destabilizing , were the smallest class of motif ( 8% ) and were enriched in body weight and femoral circumference ( Fig 6 ) . In summary , the network overall consisted of weak interactions ( mean β = 0 . 25±0 . 16 ) compared with the additive effects ( mean β = 0 . 52±0 . 36 ) , and the majority of the interactions ( 59% ) were enhancing . Further , the two largest motif classes , making up 75% of the total motifs , tended to be stabilizing .
Bone density and body composition phenotypes , such as percent body fat , are complex traits influenced by many genetic variants , both shared and distinct . Here we used combined analysis of pleiotropy and epistasis ( CAPE ) to investigate how genetic loci interact in a large population of mice to influence femoral density , femoral circumference , body weight , and percent body fat . Using this exceptionally well-powered mouse intercross we detected many main-effect and interacting QTL associated with these traits and found an extensive network of genetic loci influencing the four phenotypes . These genetic interactions were not detectable through standard regression-based epistasis analysis . We were also able to infer both the directionality and sign of the interactions , which improved our ability to identify candidate QTL genes and provided a uniquely broad view of the genetic architecture of bone and body composition phenotypes . One of the notable features of the network was an asymmetric distribution of enhancing and suppressing interactions , which was apparent for interactions between QTL , as well as for QTL interactions with sex and circulating IGF1 . Sex showed a strong bias in both the directionality and sign of interactions ( Fig 3B ) . Most ( 8 of 10 ) interactions were QTL that enhanced the phenotypic effects of the male sex on each phenotype . For example , the presence of the C3H alleles in QTL 1 . 1 increased the positive effect that the male sex had on weight , percent fat , and femoral circumference , as well as the negative effect the male sex had on femoral density . The remaining interactions ( 2 of 10 ) showed effects in the opposite direction . For example , being male enhanced the positive main effect that QTL 4 . 2 had on femoral circumference and density ( Fig 3C ) . Sex and sex hormones are known to influence bone growth and density [64–66] and genetic loci that interact with sex to influence bone phenotypes have also been previously identified in rodent models [23 , 27 , 33 , 67 , 68] . That the QTL were the sources and sex was the target of the majority of these interactions highlights how CAPE determines directionality and interpretation of interactions . Sex was widely pleiotropic , affecting all phenotypes significantly ( Fig 3A ) . An interaction in which sex is a target implies that the QTL influences all of these phenotypes to be identified by CAPE via its modifications on sex . Conversely , sex can target a non-pleiotropic QTL to influence individual phenotypes or a subset of phenotypes . It is possible that the sex-enhancing QTL contain variants in endocrine genes that globally affect sex effects . The two sex-enhanced QTL ( 4 . 2 and 13 . 1 ) , which reciprocally enhance sex effects , are loci for which the uniform enhancement of the sex effects was insufficient to fit all phenotypes simultaneously . These QTL are therefore more likely to be involved in processes that differentially affect the phenotypes , suggesting more specific roles in each phenotype that are responsive to sex difference . These findings imply that interactions between sex and QTL are commonly due to genes that lie “upstream” of processes that underlie sexual dimorphism , rather than “downstream” genes with functions that are altered by sex hormones . Circulating IGF1 , which is reduced to 10% of wild type levels in this lit/lit population , had both main effects and interaction effects . It suppressed the effects of four loci , whereas four loci enhanced the effects of circulating IGF1 ( Fig 3C ) . In contrast to sex , the interactions that IGF1 participates in are relatively balanced between incoming and outgoing interactions . This balance may reflect the fact that , unlike sex , IGF1 is a specific protein that physically interacts with other proteins . We interpret the role of the loci suppressed by IGF1 to be compensatory pathways influencing bone density when IGF1 levels are extremely low . At higher levels of circulating IGF1 the effects of these loci are diminished because the role of the causal genes becomes less relevant . For example , QTL 7 . 1 has a positive main effect on femoral density , which is suppressed by the presence of density-promoting IGF1 . Conversely , the loci that enhance the effects of IGF1 may be targets of IGF1 that act to increase bone density and other phenotypes when IGF1 is present . These QTL , e . g . QTL 10 . 2 discussed below , can therefore be interpreted to contain genes in pathways that regulate and/or respond to IGF1 signaling . We note that one QTL , Chr 9 . 2 , acts as an enhancer of IGF1 and is suppressed by IGF1 ( Fig 3C ) . This QTL may therefore contain multiple genes involved in both IGF1 pathways and compensatory pathways , or be the result of a gene with an IGF1 signaling role that also serves to trigger alternative pathways in the absence of circulating IGF1 . Although our QTL intervals were too large to decisively identify positional candidate genes , results for main effects and genetic interactions can be combined with prior data to reason about potential candidates . As an example , our model for QTL 10 . 2 interaction with IGF1 illustrates how hypotheses of causal QTL genes can be generated by requiring consistency in both main effects and interactions . Of the genes in QTL 10 . 2 , Kitl had the best evidence to suggest a role in interacting with IGF1 to influence bone density . In our study we found that the C3H allele at QTL 10 . 2 had a negative main effect on femoral density . Prior work determined that hepatic Kitl expression is lower in C3H mice than in B6 mice ( Student’s t-test , p = 0 . 012 ) [56] ( Fig 7A ) and that low expression is potentially associated with low femoral density [57] . Taken together these data suggest that the C3H allele of Kitl may be a loss of function variant that reduces Kitl transcript levels and consequently femoral density . Furthermore , this hypothesis can account for the directional interaction between QTL 10 . 2 and IGF1 , in which the C3H allele at QTL 10 . 2 enhanced the positive effects of IGF1 on femoral density . The hepatic expression study found that Kitl and Igf1 are negatively correlated ( r = −0 . 67 , p = 0 . 016 ) ( Fig 7 ) [56] ( Fig 7B ) . We can therefore hypothesize that a decrease in Kitl expression from the C3H allele corresponds to a rise in IGF1 activity and consequently its positive effects on femoral density . Combining the evidence for an interaction between Kitl and circulating IGF1 with the main effects of Kitl and QTL 10 . 2 suggests that increased Kitl transcript acts to reduce IGF1 activity in the reference B6 genotype ( Fig 7C ) . When the C3H allele is present Kitl expression decreases , allowing residual Igf1 transcript levels to remain relatively high and thereby enhancing the effect of IGF1 on femoral density in C3H mice relative to B6 mice . While speculative , this capacity to generate specific molecular hypotheses by combining genetic interactions with prior molecular results illustrates the importance of genetic interactions in hypothesis generation , even when they are a minor correction to additive effects . Such analysis is expected to be especially effective in a study with greater genetic mapping resolution and fewer candidate genes per QTL . Our interaction network was derived in a population homozygous for the lit mutation , a receptor variant that perturbs IGF1 levels . One possible consequence is that the measured phenotypes have been decanalized and some of the QTL we observed may correspond to cryptic genetic variation [45 , 46 , 69] that only influence traits in the presence of the lit/lit mutation . This release of cryptic variation may be due to IGF1 effects being reduced to the point that minor genetic effects become observable . Alternatively , low levels of IGF1 induce activity in pathways that are not used in non-lit mice for maintaining growth and bone density , thereby making variation in the genes in these pathways relevant [46] . Potentially cryptic variants could be identified as those which do not replicate in a similar analysis of standard B6 and C3H strains , although resolution of compensatory pathways would likely require a conditional Igf1 knockout . Overall , the QTL-QTL interaction network in this study had significant enrichment of enhancing interactions ( Fig 6 ) . Rather than instances of classic genetic synergy in which two variants combine to amplify a common effect , these enhancing interactions were mostly between variants with incoherent ( opposing ) main effects . We interpret these interactions as a signature of similar phenotypes between B6 and C3H strains that arise from different combinations of alleles . When alleles from two different strains are recombined in novel ways , unanticipated variation is introduced and extreme phenotypes result ( Fig 8 ) . Thus our enhancing interactions indicate a reduction of extreme phenotypes when both loci are homozygous for either parental allele . This moderating effect only occurs if the interaction between incoherent variants is enhancing . For any given phenotype an enhancing interaction from a positive main effect to a negative main effect is equivalent to a suppressing interaction in the reverse direction ( Fig 6 ) . However , when this model is applied across multiple phenotypes it is more likely to lead to high phenotype variability in the double homozygotes . This is because the suppressing QTL reduces an opposing main effect thereby favoring its own main effect . In contrast , in the enhancing model the enhancing QTL increases the opposing effect , thus balancing the effect of the overall interaction and bringing the phenotype toward the parental mean . Although additive genetic variance is usually sufficient to equalize opposing QTL effects , when the positive and negative main effects are of unequal magnitude , genetic interactions are required to stabilize the phenotypes at the levels seen in the founders . Our network outcomes suggest that genetic interactions in intercross populations will commonly have relatively weak effects , since they are a fractional correction to main effects . This can be contrasted with extreme phenotypes that arise from synergistic ( coherent enhancing ) interactions in which interaction effects are of greater magnitude than main effects , such as in the classic case of genetic buffering [70] . These interactions are significantly depleted in our network . Enrichment for motifs that drive phenotypes toward founder values is consistent with previous observations of phenotypic variation in recombinant populations . In populations of mice and Drosophila with introgressed genomic regions , simple additive contributions from all variants would result in phenotypic variance orders of magnitude greater than is observed between founders . That the founders have reduced phenotypic variance between them suggests that most interactions are less than additive [71–74] . Here , we see an enrichment of motifs leading to a reduction of extreme phenotypes , namely , suppressing coherent motifs , and enhancing incoherent motifs . While these effects are often weaker than main effects and therefore may not substantially improve the heritability accounted for , they nevertheless indicate genes acting together in a common pathway or process [60 , 61] Our analysis of a large intercross population has revealed a number of features that may be generalized to the genetic architecture of complex traits . First , we have found that a sufficiently powered study paired with a multi-trait analysis method can reveal a large network of genetic interactions between QTL . The systematic patterns in this network , including interactions with sex and a molecular marker , suggest that the interactions are signatures of the pathways and processes involved in the regulation of complex physiological traits . Second , we have found that the most common type of genetic interaction is a fairly subtle signal arising from allelic combinations that drive phenotypes towards median rather than extreme values . These interactions are either minor deviations from additivity or involve alleles with redundant effects , with the former being particularly difficult to detect in all but the largest study populations . These findings are consistent with recent work on a very large meta-analysis of twin studies [75] . Third , we note that the genetic interactions detected here form a connected network involving many interactions between the same subsets of QTL . We speculate that this is because the casual variants reside in groups of co-functional genes that compose specific pathways or processes , and that these pathways vary at multiple points between the B6 and C3H inbred strains . Since pair-wise combinations of C3H alleles have been shown to interactively drive the phenotypes towards the median , we speculate that as more genomic regions from C3H are inherited in a single individual , higher-order combinations of C3H alleles within these pathways will further canalize toward the C3H phenotype rather than cause large phenotypic variation . This may contribute to higher-order epistasis , since many of the variants will have strong effects in isolation that vanish in combinations . In sum , these findings suggest that the most common forms of epistasis may often be difficult to detect , and that the analysis of genetic interactions is nevertheless a powerful means to understand the genetics of the underlying biological pathways and processes . | The role of statistical epistasis in the genetic architecture of complex traits has been of great interest to the genetics community since Fisher introduced the concept in 1918 . However , assessing epistasis in human and model organism populations has been impeded by limited statistical power . To mitigate this limitation , we analyzed bone and body composition traits in an unusually large mouse intercross population of over 2000 mice , paired with a recently-developed computational approach that leverages information to detect interactions across multiple phenotypes . We discovered a large network of highly significant genetic interactions between variants that influence complex body composition traits . Although epistasis was abundant , the interaction network was dominated by epistasis that stabilizes phenotypes by reducing phenotypic deviation from the parent strains . Nevertheless , the observed network provides an overview of genetic architecture and specific hypotheses of how QTL combine to affect phenotypes . These findings suggest that epistatic effects are generally of lesser magnitude than main QTL effects , and therefore are unlikely to account for major components of variance , but also reinforce genetic interaction analysis as a potent tool for dissecting the biology of complex traits . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"and",
"health",
"sciences",
"quantitative",
"trait",
"loci",
"alleles",
"bone",
"network",
"analysis",
"genetic",
"interactions",
"computer",
"and",
"information",
"sciences",
"lipids",
"network",
"motifs",
"bone",
"density",
"fats",
"connective",
"tissue",
"biological",
"tissue",
"genetic",
"loci",
"biochemistry",
"anatomy",
"phenotypes",
"gene",
"identification",
"and",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences"
]
| 2016 | Weak Epistasis Generally Stabilizes Phenotypes in a Mouse Intercross |
The phytohormone abscisic acid ( ABA ) plays important roles during seed germination and early seedling development . Here , we characterized the function of the Arabidopsis WRKY6 transcription factor in ABA signaling . The transcript of WRKY6 was repressed during seed germination and early seedling development , and induced by exogenous ABA . The wrky6-1 and wrky6-2 mutants were ABA insensitive , whereas WRKY6-overexpressing lines showed ABA-hypersensitive phenotypes during seed germination and early seedling development . The expression of RAV1 was suppressed in the WRKY6-overexpressing lines and elevated in the wrky6 mutants , and the expression of ABI3 , ABI4 , and ABI5 , which was directly down-regulated by RAV1 , was enhanced in the WRKY6-overexpressing lines and repressed in the wrky6 mutants . Electrophoretic mobility shift and chromatin immunoprecipitation assays showed that WRKY6 could bind to the RAV1 promoter in vitro and in vivo . Overexpression of RAV1 in WRKY6-overexpressing lines abolished their ABA-hypersensitive phenotypes , and the rav1 wrky6-2 double mutant showed an ABA-hypersensitive phenotype , similar to rav1 mutant . Together , the results demonstrated that the Arabidopsis WRKY6 transcription factor played important roles in ABA signaling by directly down-regulating RAV1 expression .
Abscisic acid ( ABA ) is a key phytohormone that plays important roles in plant responses to stresses and plant development [1–2] . ABA is accumulated in the developing embryo , and modulates seed development and storage product accumulation [1] . In addition , ABA prevents premature seed germination and controls seed dormancy to ensure that seeds germinate under favorable conditions [1] . After germination , ABA content declines rapidly [3–4] , and exogenous ABA inhibits seed germination [5–6] . ABA functions through complex signaling networks , and some components of these networks are identified . The ABA receptors PYR/PYL/RCAR are identified in Arabidopsis thaliana [7–8] . Molecular genetics studies in Arabidopsis identify a number of genes involved in ABA signaling . The snrk2 . 2 srnk2 . 3 double mutant shows strong ABA-insensitive phenotypes in seed germination and root growth inhibition , and the two protein kinases SnRK2 . 2 and SnRK2 . 3 are demonstrated to mediate a major part of ABA signaling during seed germination [9] . The abi3 , abi4 , and abi5 mutants also show ABA-insensitive phenotypes during seed germination and early seedling development [10–12] , and the ABI3 , ABI4 , and ABI5 genes encode B3-type , APETALA2 domain and basic Leucine zipper ( bZIP ) -type transcription factors , respectively [10–13] . Three other bZIP-type transcription factors , AREB1/ABF2 , AREB2/ABF4 , and ABF3 , are also involved in ABA signaling . During seed germination , none of the areb1 , areb2 and abf3 mutants show ABA-sensitive phenotypes compared with wild-type plants , and during the vegetative growth stage , AREB1/ABF2 , AREB2/ABF4 , and ABF3 are key regulators of ABA signaling in response to osmotic stress [14–16] . The WRKY family is one of the largest transcription factor families in plants [17] . The WRKY proteins contain the conserved WRKY domain and zinc finger motif [18] . The conservation of the WRKY domain is mirrored by a remarkable conservation of the binding site , the W box ( T ) ( T ) TGAC ( C/T ) [18–19] . WRKY proteins act as repressors as well as activators by binding to their target genes’ promoters . Several WRKY transcription factors have been reported to be involved in the ABA signaling network . Three evolutionarily related WRKY transcription factors ( AtWRKY18 , AtWRKY40 and AtWRKY60 ) are negative regulators in ABA signaling , and AtWRKY40 directly represses the expression of ABI4 and ABI5 by binding to the promoters of ABI4 and ABI5 [20] . The knockout mutant of AtWRKY63 , the abo3 mutant , is hypersensitive to exogenous ABA during seed germination and the vegetative growth stage [21] , and the Arabidopsis wrky2 mutant has similar phenotypes to the abo3 mutant except that AtWRKY2 has no effect on stomatal closure [22] . Recently , the AtWRKY41 protein is reported to control seed dormancy via direct regulation of ABI3 expression [23] , and AtWRKY8 functions in the TMV-cg defense response by mediating ABA and ethylene signaling [24] . In this study , we find that the Arabidopsis WRKY6 is a positive regulator in ABA signaling during seed germination and early seedling development . The knockout of WRKY6 enhances plant ABA insensitivity during seed germination and early seeding growth , and WRKY6-overexpressing lines show ABA-hypersensitive phenotypes . The WRKY6 transcription factor represses RAV1 expression and enhances the expression of ABI3 , ABI4 and ABI5 , which are down-regulated by RAV1 . The WRKY6 protein can bind to the W-box motif within the RAV1 promoter , indicating that WRKY6 directly regulates RAV1 expression . Overexpression of RAV1 abolishes the ABA-sensitivity of WRKY6-overexpressing lines , and repression of RAV1 impairs the ABA-insensitivity of wrky6 mutants , demonstrating that RAV1 is genetically epistatic to WRKY6 .
Arabidopsis WRKY6 ( WRKY transcription factor 6 , At1g62300 ) is a WRKY transcription factor [25] and , from public microarray data , we found that WRKY6 expression was relatively high in dry seeds and reduced after imbibition . Then we examined the expression of WRKY6 during seed germination and early seedling development . The transcript level of WRKY6 was markedly repressed during seed germination ( Fig 1A ) , indicating that WRKY6 may be involved in seed germination and early seedling development . When germinated and grown on Murashige and Skoog ( MS ) medium containing 0 . 5 μM ABA ( MS+ABA ) , WRKY6 expression was significantly induced ( Fig 1B ) . The transcript level of WRKY6 was further tested in seedlings treated with exogenous ABA . The 7-d-old wild-type seedlings were transferred to MS solution with or without 100 μM ABA for 3 h , and then harvested for qRT-PCR assay . The qRT-PCR results showed that the transcript level of WRKY6 was significantly induced by exogenous ABA ( Fig 1C ) . WRKY6-overexpressing lines and wrky6 mutants were used to study the physiological function of WRKY6 in seed germination . The WRKY6-overexpressing lines ( 35S:WRKY6-5 and 35S:WRKY6-9 ) and the wrky6-1 mutant were provided by Dr . Somissich [26] . A WRKY6 T-DNA insertion line ( Salk_012997 ) , named wrky6-2 , was ordered from the ABRC ( Arabidopsis Biological Resource Center ) . The qRT-PCR results showed that WRKY6 expression was significantly repressed in the wrky6-1 and wrky6-2 mutants , and elevated in 35S:WRKY6-5 and 35S:WRKY6-9 , compared with wild-type plants ( Fig 1D ) . When germinated and grown on MS medium , all plants showed no obvious difference in their phenotypes ( Fig 1E , left panel ) . When grown on MS medium containing 0 . 5 μM ABA ( MS + 0 . 5 μM ABA ) , the wrky6-1 and wrky6-2 mutants were more insensitive to ABA than wild-type plants , whereas 35S:WRKY6-5 and 35S:WRKY6-9 showed ABA hyper-sensitive phenotypes ( Fig 1E , right panel ) . When grown on MS medium containing 0 . 5 μM ABA , the wrky6-1 and wrky6-2 mutants were less ABA insensitive than the abi4 and abi5 mutants ( S1 Fig ) . Seed germination was further tested , and in the absence of ABA ( MS ) , the seed germination percentages of different genotypes were similar ( Fig 1F ) . When germinated and grown on MS medium containing 0 . 5 μM ABA ( MS + 0 . 5 μM ABA ) , the two WRKY6-overexpressing lines ( 35S:WRKY6-5 and 35S:WRKY6-9 ) showed significantly reduced seed germination percentages , and the seed germination percentages of the two wrky6 mutants were similar to wild-type plants ( Fig 1G ) . When germinated and grown on MS medium containing 2 μM ABA ( MS + 2 μM ABA ) , the two WRKY6 mutants ( wrky6-1 and wrky6-2 ) showed significantly increased seed germination percentages compared with wild-type plants , and the WRKY6-overexpressing line ( 35S:WRKY6-9 ) showed reduced seed germination percentage relative to wild-type plants ( Fig 1H ) . The cotyledon-greening percentages were also measured , and in the absence of ABA ( MS ) , they were similar among different genotypes ( Fig 1I ) . When germinated and grown on MS medium containing 0 . 5 μM ABA ( MS + 0 . 5 μM ABA ) , the WRKY6-overexpressing lines ( 35S:WRKY6-5 and 35S:WRKY6-9 ) had lower , whereas the wrky6 mutants ( wrky6-1 and wrky6-2 ) had higher , cotyledon-greening percentages than wild-type plants ( Fig 1I ) . To further test whether WRKY6 was involved in ABA mediated root growth inhibition , the 4-d-old wrky6 mutants , WRKY6-overexpressing lines and wild-type seedlings were transferred to MS medium with or without 15 μM ABA for 7 d . When grown on MS medium , the primary root length was similar among different genotypes ( Fig 1J and 1K ) . When grown on MS medium containing 15 μM ABA , the WRKY6-overexpressing lines ( 35S:WRKY6-5 and 35S:WRKY6-9 ) showed shorter primary root compared with wild-type seedlings , and the primary root lengths of wrky6-1 and wrky6-2 were similar to that of wild-type plants ( Fig 1J and 1K ) . Together , these data indicate that WRKY6 plays important roles in ABA signaling during seed germination and seedling development . As WRKY6 is a WRKY transcription factor involved in ABA signaling ( Fig 1 ) , the expression of ABA inducible genes , such as RD29b , RAB18 and COR47 , was tested in the WRKY6-overexpressing lines and wrky6 mutants . The transcript levels of RD29b , RAB18 , and COR47 were elevated in the WRKY6-overexpressing lines and repressed in the wrky6 mutants ( Fig 2 ) . Then the expression of the following ABA-responsive genes was tested: ABFs ( ABF1 , ABF2/AREB1 and ABF3 ) [27] , SnRK2s [28–29] , ABI3 [10] , ABI4 [11] , ABI5 [12] , RAV1 [30] , Em1 [31] and Em6 [31] . The qRT-PCR results showed that the transcript levels of these ABA-responsive genes were elevated in the WRKY6-overexpressing lines , and the expression of most of these genes was repressed in the wrky6 mutants ( Fig 2 ) . It is notable that the expression of RAV1 was significantly repressed in the WRKY6-overexpressing lines and upregulated in the wrky6-1 and wrky6-2 mutants ( Fig 2 ) . It is also notable that the expression of ABI3 and ABI4 was elevated in the WRKY6-overexpressing lines and suppressed in the wrky6 mutants ( Fig 2 ) . The expression of these ABA-responsive genes was also tested in the wrky6 mutants and wild-type plants under exogenous ABA treatment . After the seedlings were treated with 100 μM ABA for 3 h , the expression of these genes , except RAV1 , was induced in the wild-type seedlings , and this inducement by exogenous ABA was obviously repressed in the wrky6-1 and wrky6-2 mutants ( Fig 3 ) . The RAV1 was repressed by exogenous ABA in the wild-type seedlings , and the transcript levels of RAV1 in wrky6-1 and wrky6-2 mutants were much higher than that in wild-type seedlings with or without ABA treatment ( Fig 3 ) . These data demonstrate that disruption and overexpression of WRKY6 alter the expression of the ABA-responsive genes . The expression of these genes was still ABA inducible in the wrky6 mutants , indicating that besides WRKY6 , there were other transcription factors regulating these genes expression . As our previous work showed that Arabidopsis RAV1 directly down-regulated the expression of ABI3 , ABI4 , and ABI5 [30]—and the RAV1 expression was lower , whereas the expression of ABI3 , ABI4 and ABI5 was elevated in WRKY6-overexpressing lines ( Figs 2 and 3 ) —we hypothesized that WRKY6 directly regulated RAV1 expression . Then the expression of ABI3 , ABI4 and ABI5 was further tested during the seed germination with or without exogenous ABA . During the seed germination , the transcript level of WRKY6 was obviously repressed , and the RAV1 expression was obviously induced ( Fig 4 ) . The transcript levels of ABI3 , ABI4 and ABI5 , which directly down-regulated by RAV1 , were obviously suppressed during seed germination ( Fig 4 ) . And the transcript levels of WRKY6 , ABI3 , ABI4 and ABI5 were obviously induced , and the RAV1 expression was repressed , by exogenous ABA ( Fig 4 ) . These data imply that WRKY6 may directly regulate the RAV1 expression . WRKY proteins act as regulators by binding to W-box ( es ) within their target genes promoters . First the RAV1 promoter sequence was analyzed and the results showed that there were two W-box motifs within the RAV1 promoter ( Fig 5A ) . To further test the function of WRKY6 on regulation of RAV1 expression , a transient expression experiment in tobacco leaves was performed—WRKY6 repressed RAV1 promoter activity ( Fig 5B ) . Then an electrophoretic mobility shift assay ( EMSA ) was conducted to test whether WRKY6 bound to the RAV1 promoter in vitro . The recombinant WRKY6-His protein and His protein alone were expressed in Escherichia coli and purified . The WRKY6-His fusion protein can bind to the P2 fragment of the RAV1 promoter , and this binding was effectively reduced by adding increasing amounts of unlabeled competitor with the same P2 sequence ( Fig 5C ) . When the W-box motif in the P2 fragment was mutated from TTGACC to TACGTC , the binding complex was not detected ( Fig 5C ) . No super-shifted WRKY6-P1 complexes were detected in EMSA ( Fig 5C ) . These data indicate that WRKY6 protein can bind to the P2 fragment of RAV1 promoter in vitro . Furthermore , a chromatin immunoprecipitation ( ChIP ) assay was conducted to determine whether WRKY6 bound to the RAV1 promoter in vivo . The anti-WRKY6 antibody ( AS111778; Agrisera ) was tested in the wrky6-2 mutant and wild-type seedlings , and the anti-WRKY6 antibody can specifically recognize the WRKY6 protein ( Fig 5D ) . For the WRKY6 expression was induced by exogenous ABA ( Fig 1C ) , the protein level of WRKY6 was also tested under ABA treatment . After treated with 100 μM ABA for 3 h , the WRKY6 protein was elevated in the wild-type seedlings and still not detected in the wrky6-2 mutant ( Fig 5D ) . Then the ChIP assay was conducted with anti-WRKY6 antibody . The chromatin immunoprecipitated with the anti-WRKY6 antibody was enriched in the P2 fragment of the RAV1 promoter , and the enrichment was enhanced under ABA treatment ( +ABA ) ( Fig 5E ) . In contrast , fragments from the P1 fragment of the RAV1 promoter and the exon region of the Actin gene ( ACTIN ) did not show any detectable binding by WRKY6 with or without ABA treatment ( Fig 5E ) . These results demonstrate that WRKY6 directly regulates RAV1 expression . The 35S:WRKY6-9 was crossed with RAV1-overexpressing line ( RAV1 OE2; [30] ) , and the 35S:WRKY6-9/RAV1 OE2 double overexpressing line was obtained ( Fig 6A ) . When germinated and grown on MS medium , there were no obvious phenotype differences among all genotypes ( Fig 6B , top panel ) , and their seed germination rates were similar ( Fig 6C ) . In the presence of 0 . 5 μM ABA ( MS + 0 . 5μM ABA ) , the 35S:WRKY6-9/RAV1 OE2 double overexpressing line displayed ABA-insensitive phenotypes , similar to RAV1 OE2 ( Fig 6B , bottom panel ) ; and both 35S:WRKY6-9/RAV1 OE2 and RAV1 OE2 had similar higher seed germination compared with wild-type plants , whereas the seed germination of 35S:WRKY6-9 was significantly reduced relative to wild-type ( Fig 6D ) . The cotyledon-greening percentage was also measured . In the absence of ABA , the different genotypes had similar cotyledon-greening percentages ( Fig 6E ) . In the presence of 0 . 5 μM ABA , both 35S:WRKY6-9/RAV1 OE2 and RAV1 OE2 had higher , whereas the 35S:WRKY6-9 had lower , cotyledon-greening percentages than wild-type plants ( Fig 6E ) . Expression of RAV1 target genes , ABI3 , ABI4 and ABI5 , was also tested by qRT-PCR and all were elevated in the WRKY6-overexpressing line ( 35S:WRKY6-9 ) , but repressed in the 35S:WRKY6-9/RAV1 OE2 lines , similar to RAV1 OE2 , compared with wild-type plants ( Fig 7 ) . These data together with phenotype tests indicated that RAV1 overexpression abolished the ABA-sensitivity of WRKY6-overexpressing line . We also introduced Super:RAV1 [30] to the wrky6-2 mutant , and got four wrky6-2 RAV1OE ( T1 ) transgenic lines ( Fig 8 ) . The four wrky6-2 RAV1OE lines had the higher RAV1 expression and much lower WRKY6 expression than wild-type plants ( Fig 8 ) . The transcript levels of ABI3 , ABI4 and ABI5 in wrky6-2 RAV1 OE lines were lower than those in wild-type plants , even lower than those in the wrky6-2 mutant , similar to those in RAV1 OE2 ( Fig 8 ) . The transcript levels of Em1 and Em6 in wrky6-2 RAV1OE lines were also lower than those in wild-type and wrky6-2 mutant , similar to those in RAV1 OE2 ( Fig 8 ) . These data indicate that overexpression of RAV1 represses the expression of ABI3 , ABI4 and ABI5 , and WRKY6 modulates the expression of ABI3 , ABI4 and ABI5 through down-regulating the RAV1 expression . The expression of ABFs and SnRK2s was also tested . The transcript levels of ABF1 and ABF2 in the wrky6-2 mutant , RAV1 OE2 and wrky6-2 RAV1OE lines were similar , and slightly lower than those in wild-type plants ( Fig 8 ) . And the transcript levels of SnRK2s were similar among RAV1 OE2 , wrky6-2 RAV1 OE lines and wild-type plants ( Fig 8 ) . The RAV1-underexpressing line ( RAV1-U ) is an antisense transgenic line , which has relatively lower RAV1 expression [30 , 32] . When grown on MS medium containing 0 . 5 μM ABA , RAV1-U shows ABA hyper-sensitive phenotypes [30] . The genetic relationship between WRKY6 and RAV1 was analyzed by crossing wrky6-2 with RAV1-U to produce the wrky6-2 RAV1-U double mutant ( Fig 9A ) . In the absence of ABA ( MS ) , all lines showed similar phenotypes ( Fig 9B , left panel ) . When germinated and grown on MS medium containing 0 . 5 μM ABA ( MS + 0 . 5 μM ABA ) , the wrky6-2 mutant displayed an ABA-insensitive phenotype , whereas the wrky6-2 RAV1-U double mutant showed an ABA-sensitive phenotype , similar to RAV1-U ( Fig 9B , right panel ) . The cotyledon-greening percentages were also tested and , in the absence of ABA ( MS ) , were similar for the different genotypes ( Fig 9C ) . When germinated and grown on MS medium containing 0 . 5 μM ABA ( MS + 0 . 5 μM ABA ) , the wrky6-2 RAV1-U double mutant ( similar to RAV1-U ) had a much lower , and the wrky6-2 mutant had a higher , cotyledon-greening percentage than wild-type plants ( Fig 9C ) . Expression of ABI3 , ABI4 and ABI5 was tested by qRT-PCR and showed clearly elevated transcript levels in the wrky6-2 RAV1-U double mutant , similar to that in RAV1-U , and repressed in the wrky6-2 mutant ( Fig 9D ) . Further , we used the CRISPR/Cas9 technology to generate rav1 mutant and rav1 wrky6-2 double mutant . A pair of closely located sgRNA targets in the RAV1 gene were selected ( Fig 10A ) . The CRISPR construct was transformed into wild-type Arabidopsis and the wrky6-2 mutant , and the homozygous rav1 mutant and rav1 wrky6-2 double mutant were obtained . The rav1 wrky6-2 double mutant contained a nucleotide insertion in C1 and C2 sites , separately ( Fig 10B ) , and the rav1 mutant had a nucleotide insertion in C1 site ( Fig 10B ) . These insertions lead to frameshift mutation . The qRT-PCR results showed that the transcript level of WRKY6 was significantly repressed in the rav1 wrky6-2 double mutant , similar to that in wrky6-2 mutant ( Fig 10C ) . These data indicate that we obtain the rav1 mutant and rav1 wrky6-2 double mutant . When grown on MS medium containing 0 . 5 μM ABA , the rav1 wrky6-2 double mutant showed ABA hyper-sensitive phenotypes , similar to the rav1 mutant ( Fig 10D ) , and both rav1 mutant and rav1 wrky6-2 double mutant had much lower cotyledon-greening percentages compared with wild-type plants ( Fig 10E ) . The expression of ABIs was also tested , and the qRT-PCR results showed that the transcript levels of ABI3 , ABI4 , and ABI5 were elevated in the rav1 wrky6-2 double mutant , similar to that in the rav1 mutant , typically under ABA treatment ( Fig 10F ) . Taken together , these results demonstrate that disruption of RAV1 abolishes the ABA-insensitivity of the wrky6-2 mutant . There were one or two W boxes within the promoters of ABI3 , ABI4 and ABI5 ( Fig 11A ) , and the expression of ABI3 , ABI4 and ABI5 was elevated in the WRKY6-overexpressing lines and repressed in the wrky6 mutants ( Fig 2 ) . It is hypothesized that WRKY6 directly regulates the expression of ABI3 , ABI4 and ABI5 . Then the EMSA experiment was conducted , and the results showed that WRKY6 can bind to the W-box within the ABI3 promoter in vitro ( Fig 11B ) . Although the super-shifted WRKY6-ProABI4 and WRKY6-ProABI5 complexes were detected , these bindings were not reduced by adding the unlabeled competitors , or not missing with the mutation probe with the mutated W-box ( TTGACC was changed to TACGTC ) ( Fig 11C and 11D ) , indicating that WRKY6 can not bind to the promoters of ABI4 and ABI5 in vitro . To further test the function of WRKY6 in regulation of ABI3 , ABI4 , and ABI5 expression , the transient expression experiment in tobacco leaves was performed . Although WRKY6 can bind to the ABI3 promoter in vitro , WRKY6 can not regulate ABI3 expression in tobacco leaves ( Fig 11E ) . And WRKY6 can not regulate the expression of ABI4 and ABI5 in tobacco leaves ( Fig 11E ) . All these data indicate that WRKY6 can not directly regulate the expression of ABI3 , ABI4 , and ABI5 .
Arabidopsis WRKY6 is a WRKY transcription factor [25] . In this study , we demonstrated that Arabidopsis WRKY6 played important roles in ABA signaling during seed germination and early seedling development . When germinated and grown on MS medium containing ABA , the wrky6 mutants were ABA-insensitive while WRKY6-overexpressing lines were ABA-hypersensitive compared with wild-type plants ( Fig 1 ) . As a WRKY transcription factor , WRKY6 is localized in the nucleus and has a DNA-binding domain ( WRKY domain ) [25] . One reason for the ABA-response phenotypes is that WRKY6 regulated the expression of ABA-response genes . The AREB/ABFs are bZIP-type transcription factors , which recognize the ABA-responsive elements ( ABRE ) in the promoters of ABA-inducible genes [27] , and the expression of AREB1/ABF2 , AREB2/ABF4 and ABF3 is induced by dehydration , high salinity and ABA treatment in vegetative tissues [33] . The expression of ABF2 and ABF3 was induced by exogenous ABA , and this inducement was obviously repressed in the wrky6 mutants ( Fig 3 ) , and the cotyledon-greening percentages of wrky6-1 and wrky6-2 were much higher than wild-type plants ( Fig 1I ) , indicating that WRKY6 played a role in response to ABA signaling during post-germination growth partially by regulating expression of ABF2 and ABF3 . Further promoter sequence analysis results showed that there was no W box ( TTGACC/T ) within the 2-kb promoters of ABF1 , ABF2 and ABF3 , indicating that WRKY6 can not directly regulate the expression of ABF1 , ABF2 and ABF3 . The transcription factors ABI3 , ABI4 , and ABI5 are well known positive regulators of ABA signaling during seed germination [10–12] . The qRT-PCR results showed that the transcript levels of ABI3 , ABI4 , and ABI5—typically the expression of ABI3 and ABI4—were repressed in wrky6 mutants ( wrky6-1 and wrky6-2 ) and elevated in WRKY6-overexpressing lines ( Figs 2 and 3 ) , suggesting that WRKY6 modulated the expression of ABI3 , ABI4 , and ABI5 . The qRT-PCR results also showed that RAV1 expression was significantly induced in wrky6 mutants ( wrky6-1 and wrky6-2 ) and repressed in WRKY6-overexpressing lines ( 35S:WRKY6-5 and 35S:WRKY6-9 ) ( Figs 2 and 3 ) . Our previous work showed that the Arabidopsis RAV1 transcription factor negatively regulated the expression of ABI3 , ABI4 , and ABI5 [30] , suggesting that WRKY6 modulates the expression of ABI3 , ABI4 , and ABI5 by negatively regulating RAV1 expression . The EMSA and ChIP analyses showed that WRKY6 could bind to the RAV1 promoter in vitro and in vivo ( Fig 5 ) , demonstrating that WRKY6 negatively regulated RAV1 expression by binding to the RAV1 promoter . Usually , WRKY transcription factors contain the conserved WRKY domain and bind to the W box ( es ) within their target genes’ promoters [18–19] . Interestingly , the genes ABI3 , ABI4 , and ABI5 contain several W boxes in their promoters [20 , 23] , and the expression of ABI3 , ABI4 , and ABI5 was enhanced in WRKY6-overexpressing lines and repressed in wrky6 mutants ( Figs 2 and 3 ) . Previous reports showed that WRKY40 directly represses ABI5 expression [20] , and WRKY41 directly regulates ABI3 expression [23] . Thus we investigated whether WRKY6 directly regulated the expression of ABI3 , ABI4 , and ABI5 , and whether the ABA-response phenotypes of 35S:WRKY6 and wrky6 mutants were due to the direct regulation of WRKY6 on ABI3 , ABI4 , and ABI5 . The phenotype of 35S:WRKY6-9/RAV1 OE2 was first tested . When grown on MS medium containing ABA , the WRKY6-overexpressing line ( 35S:WRKY6-9 ) showed an ABA-hypersensitive phenotype , whereas overexpression of RAV1 in 35S:WRKY6-9 ( 35S:WRKY6-9/RAV1 OE2 ) repressed the ABA-hypersensitivity of 35S:WRKY6-9 ( Fig 6 ) , indicating that its ABA-hypersensitivity was mainly due to repression of RAV1 by WRKY6 . And the transcript levels of ABI3 , ABI4 and ABI5 in wrky6-2 RAV1 OE were lower than those in wrky6-2 mutant , similar to those in RAV1 OE2 ( Fig 8 ) , indicating that the expression of ABI3 , ABI4 and ABI5 was regulated by RAV1 , not by WRKY6 . Then the wrky6 RAV1-U and rav1 wrky6-2 double mutant was generated . When grown on MS medium containing ABA , the wrky6-2 mutant showed an ABA-insensitive phenotype , and the repression or disruption of RAV1 in the wrky6-2 mutant ( i . e . wrky6-2 RAV1-U and rav1 wrky6-2 ) abolished the ABA-insensitivity of wrky6-2 ( Figs 9 and 10 ) , indicating that the ABA-insensitivity of wrky6-2 was mainly due to the disruption of the regulation by WRKY6 of RAV1 , and RAV1 was epistatic to WRKY6 . The EMSA results showed that WRKY6 could not bind to the promoters of ABI4 ( Fig 11C ) and ABI5 ( Fig 11D ) , indicating that WRKY6 could not directly regulate the expression of ABI4 and ABI5 . WRKY6 can bind to the ABI3 promoter in vitro ( Fig 11B ) , whereas WRKY6 can not modulate the ABI3 expression in plants ( Fig 11E ) , indicating that WRKY6 also can not directly regulate ABI3 expression . These data demonstrate that WRKY6 acted as a positive regulator mainly via direct regulation of RAV1 expression . The 2-kb promoter sequences of SnRK2s and Ems were also analyzed , and the results showed that there was no W box in SnRK2 . 2 promoter , one in SnRK2 . 6 and Em1 promoters , two in SnRK2 . 3 promoter and three in Em6 promoter . And the transcript levels of the SnRK2s and Ems were elevated in the WRKY6-overexpressing lines ( Fig 2 ) and repressed in the wrky6 mutants under ABA treatment ( Fig 3 ) , indicating that WRKY6 transcription factor may directly regulate the expression of SnRK2 . 3 , SnRK2 . 6 , Em1 and Em6 . In summary , our data show that the Arabidopsis WRKY6 transcription factor plays important roles in ABA signaling ( Fig 12 ) . The WRKY6 expression is repressed during seed germination and early seedling development , and induced by exogenous ABA . WRKY6 transcription factor acts in the ABA signal transduction pathway predominantly by directly down-regulating RAV1 expression; RAV1 mediates seed germination and early seedling development by directly down-regulating expression of ABI3 , ABI4 and ABI5 . The WRKY6 gene , encoding a WRKY transcription factor , is expressed in all tissues [25] , suggesting that WRKY6 plays widespread roles during different phases of plant development . The WRKY6 transcript is present in roots , shoots , flowers , siliques and senescent leaves , with the highest transcript level of WRKY6 in senescent leaves [25] . Overexpression of WRKY6 results in dwarfed Arabidopsis with partly necrotic leaves , early flowering and a reduction in their apical dominance [26] . Interestingly , overexpressing RAV1 caused a retardation of rosette leaf development , and underexpression of RAV1 caused an earlier flowering phenotype [32] . Recently , Arabidopsis RAV1 was reported to positively regulate leaf senescence , and overexpression of RAV1 caused premature leaf senescence [34] . The data in the present study showed that WRKY6 directly repressed RAV1 expression . The data suggested that the WRKY6-RAV1 regulatory pathway was involved in leaf senescence and flowering . In addition to modulating leaf senescence and flowering , the expression of WRKY6 was repressed during seed germination and early seedling development , and obviously induced by exogenous ABA ( Fig 1A–1C ) . When grown on MS medium with ABA , the wrky6 mutants showed ABA-insensitive phenotypes while the WRKY6-overexpressing lines were ABA-hypersensitive ( Fig 1E ) . WRKY6 could bind to the RAV1 promoter to repress RAV1 expression ( Fig 5 ) . Further genetic results showed that RAV1 was the main target gene of WRKY6 during seed germination and early seedling development ( Figs 6–10 ) . These data provide evidence of the major role of WRKY6 during seed germination and early seedling development . WRKY6 is also involved in controlling processes related to pathogen defense [25–26] . WRKY6 positively influences the promoter activity of the pathogen defense-associated PR1 gene , most likely involving NPR1 function [26] . In addition to abiotic stress , WRKY6 has also been reported to be involved in biotic stress responses . WRKY6 is a negative regulator in phosphate translocation [35] . When grown on inorganic phosphorus ( Pi ) -sufficient condition , WRKY6 represses PHO1 expression and reduces Pi translocation from roots to shoots . During Pi starvation , the WRKY6 protein is degraded and the repression of PHO1 by WRKY6 is abolished [35] . The expression of WRKY6 is induced by boron ( B ) deficiency , and wrky6 mutants showed growth defects compared with wild-type plants under B deficient condition [36] . Recently , WRKY6 was reported to be induced by arsenate stress , and the WRKY6 mediated the expression of a phosphate transporter gene and restricted arsenate-induced transposon activation [37] . Taken together , the WRKY6 transcription factor plays important roles in plant development and biotic and abiotic stress responses .
The wild-type plant used in this study was A . thaliana Col-0 . The WRKY6-overexpressing lines ( 35S:WRKY6-5 and 35S:WRKY6-9 ) , the wrky6-1 mutant , the RAV1-overexpressing line ( RAV1 OE2 ) and the RAV1-underexpressing line ( RAV1-U ) were described previously [26 , 30 , 32] . The WRKY6 T-DNA insertion line Salk_012997 , named wrky6-2 in the present study , was ordered from the ABRC . For the seed germination assay , seeds were surfaced sterilized and kept at 4°C for 72 h in darkness before germination . About 300 seeds of each genotype were sown on the same plate containing MS medium [with 3% ( w/v ) sucrose] with 0 , 0 . 5 and 2 μM ABA , and were kept at 22°C under constant illumination of 60 μmol·m−2·s−1 . Germination was defined as an obvious emergence of the radicle through the seed coat . The seed germination percentages were evaluated daily during the germination test . For qRT-PCR analysis , total RNA of seedlings and seeds was extracted with Trizol reagent ( Invitrogen ) and RNeasy Plant Mini kit ( Bioteke ) , separately . The total RNA ( 8 μg ) was treated with DNase I ( RNase Free ) ( Takara ) to eliminate genomic DNA contamination . Then the cDNA was synthesized from the treated total RNA ( 4 μg ) by SuperScript II Reverse Transcriptase ( Invitrogen ) using Radom Hexamer Primers ( Promega ) . 40 ng cDNA ( except RAB18 , with 80 ng cDNA ) and 50 nM each primer were used for each quantitative PCR reaction , which was performed by using the Power SYBR Green PCR Master Mix ( Life Technologies ) on a 7500 Real Time PCR System machine ( Life Technologies ) following the manufacturer’s protocols . The thermal treatment was 10 min at 95°C , then 40 cycles of 15 s at 95°C , 1 min at 60°C . Amplification was followed by a melt curve analysis . The 2-ΔΔCt method was used for relative quantification [38] . Actin2/8 expression was used as an internal control . The statistical significance was evaluated by Paired t-test analysis . The primers used are listed in S1 Table . The transient GUS expression assay was performed as described previously [35] . The ProRAV1:GUS and Super:WRKY6 constructs were described previously [30 , 35] . To construct RroABI3 , ProABI4 and ProABI5 , the ∼2kb promoters of ABI3 , ABI4 and ABI5 were cloned into the pCAMBIA1381 vector . The primer sequences used are listed in S1 Table . For each infiltration sample , Super:LUC was added as an internal control . The GUS and LUC activities of the infiltrated leaves were quantitatively determined , and the GUS/LUC ratio was used to quantify the promoter activity . The coding sequence of WRKY6 was amplified and cloned into the pET30a vector . The primer sequences used are listed in S1 Table . The recombinant plasmid was introduced to E . coli strain BL21 . E . coli cells were induced with 0 . 2 mM IPTG overnight at 18°C and collected by centrifugation . The WRKY6-His protein was purified using Ni-Sepharose 6 Fast Flow ( GE Healthcare ) , and the protein concentration was determined by Bio-Rad protein assay . The pET30a vector was also introduced into E . coli strain BL21 , and a protein with His tag was purified . This purified protein was named His protein , and used as a control in EMSA experiment . For EMSA assays , the fragment of the promoters were obtained by PCR using biotin-labeled or -unlabeled primers ( see S1 Table ) . Biotin-unlabeled fragments of the same sequences were used as competitors . The reaction mixture ( 20 μL ) for EMSA contained 0 . 5 μg purified protein , 1 μL 50 μg/mL biotin-labeled annealed oligonucleotide , 2 μL 10×binding buffer ( 100 mM Tris , 500 mM KCl , and 10 mM DTT , pH 7 . 5 ) , 1 μL 1% Nonidet P-40 , 0 . 5 μL 1 mg/mL poly ( dI-dC ) , and ultrapure water . The reactions were incubated at 22°C for 30 min . The reactions were fractionated on a 5% native polyacrylamide gel in 0 . 5 ×TBE buffer . The detection of biotin-labeled DNA by chemiluminescence was performed using a LightShift Chemiluminescent EMSA Kit ( Pierce ) following the manufacturer’s protocol . The ChIP experiment was performed as described previously [30 , 35] . For the ChIP assay , 1 g of 7-d-old seedlings grown on MS medium was transferred to MS solution with or without 100 μM ABA for 3 h , then harvested and cross-linked by 1% formaldehyde for 10 min , and then the purified cross-linked nuclei were resuspended in 4 mL lysis buffer . Following sonication , 1 mL lysis buffer with nuclei was used for each immunoprecipitation ( IP ) . The anti-WRKY6 antibody ( AS111778; Agrisera , http://www . agrisera . com/ ) was used to immunoprecipitate DNA/protein complexes from the chromatin preparation . IP DNA was dissolved in 25 μL TE buffer , and 1 μL IP DNA was analyzed by qPCR using the primers listed in S1 Table . As a control , ‘input’ DNA was isolated from 50 μL lysis buffer with nuclei without the IP step . The input DNA was suspended in 25 μL TE buffer and 1 μL input DNA was analyzed by qPCR . The ratio of IP DNA over the input was presented as the percentage of input ( IP % ) . An Actin fragment ( ACTIN ) was amplified as control . At least three independent experiments were performed with similar results . Data are mean values of three replicates ± standard error ( SE ) from one experiment . A pair of closely located sgRNA targets ( C1: GATGAGAGTACTACAAGTAC and C2: ACGGCGTAGAAGCTGAATCT ) in RAV1 gene was selected and cloned into the pHEE2A-TRI vector as described [39] . Then the CRISPR construct was transformed into wild-type Arabidopsis and the wrky6-2 mutant to obtain rav1 mutant and rav1 wrky6-2 double mutant , separately . The homozygous rav1 mutant and rav1 wrky6-2 double mutant were identified by sequencing . Sequence data for the Arabidopsis genes described in this study can be found in the Arabidopsis Genome Initiative or GenBank/EMBL databases under the following accession numbers , At1g62300 for WRKY6 , At1g49720 for ABF1 , At1g45249 for ABF2 , At4g34000 for ABF3 , At3g50500 for SnRK2 . 2 , At5g66880 for SnRK2 . 3 , At4g33950 for SnRK2 . 6 , At3g24650 for ABI3 , At2g40220 for ABI4 , At2g36270 for ABI5 , At1g13260 for RAV1 , At3g51810 for Em1 , and At2g40170 for Em6 . | The WRKY6 protein is a WRKY transcription factor which plays important roles in plant pathogen defense , phosphate translocation , and arsenate resistance . This study demonstrated that the expression of WRKY6 was obviously repressed during seed germination and significantly induced by exogenous ABA . In the presence of exogenous ABA , the two wrky6 mutants showed ABA-insensitive phenotypes , whereas the WRKY6-overexpressing lines were hypersensitive to ABA . The WRKY6 transcription factor repressed RAV1 expression and enhanced the expression of ABI3 , ABI4 and ABI5 , which was down-regulated by RAV1 . The WRKY6 protein could bind to the W-box motif within the RAV1 promoter , indicating that WRKY6 directly regulated RAV1 expression . Overexpression of RAV1 abolished the ABA-sensitivity of WRKY6-overexpressing lines , and repression of RAV1 impaired the ABA-insensitivity of wrky6 mutants . Our results reveal the important roles of WRKY6 in ABA signaling during seed germination and early seedling development . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"plant",
"anatomy",
"electrophoretic",
"mobility",
"shift",
"assay",
"gene",
"regulation",
"regulatory",
"proteins",
"dna-binding",
"proteins",
"plant",
"physiology",
"plant",
"science",
"transcription",
"factors",
"seedlings",
"molecular",
"biology",
"techniques",
"plants",
"research",
"and",
"analysis",
"methods",
"plant",
"reproduction",
"gene",
"mapping",
"proteins",
"hyperexpression",
"techniques",
"gene",
"expression",
"seeds",
"molecular",
"biology",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"gene",
"expression",
"and",
"vector",
"techniques",
"biochemistry",
"phenotypes",
"genetics",
"biology",
"and",
"life",
"sciences",
"restriction",
"fragment",
"mapping",
"seed",
"germination",
"organisms"
]
| 2016 | Arabidopsis WRKY6 Transcription Factor Acts as a Positive Regulator of Abscisic Acid Signaling during Seed Germination and Early Seedling Development |
While there is accumulating evidence for the importance of the metabolic cost of information in sensory systems , how these costs are traded-off with movement when sensing is closely linked to movement is poorly understood . For example , if an animal needs to search a given amount of space beyond the range of its vision system , is it better to evolve a higher acuity visual system , or evolve a body movement system that can more rapidly move the body over that space ? How is this trade-off dependent upon the three-dimensional shape of the field of sensory sensitivity ( hereafter , sensorium ) ? How is it dependent upon sensorium mobility , either through rotation of the sensorium via muscles at the base of the sense organ ( e . g . , eye or pinna muscles ) or neck rotation , or by whole body movement through space ? Here we show that in an aquatic model system , the electric fish , a choice to swim in a more inefficient manner during prey search results in a higher prey encounter rate due to better sensory performance . The increase in prey encounter rate more than counterbalances the additional energy expended in swimming inefficiently . The reduction of swimming efficiency for improved sensing arises because positioning the sensory receptor surface to scan more space per unit time results in an increase in the area of the body pushing through the fluid , increasing wasteful body drag forces . We show that the improvement in sensory performance that occurs with the costly repositioning of the body depends upon having an elongated sensorium shape . Finally , we show that if the fish was able to reorient their sensorium independent of body movement , as fish with movable eyes can , there would be significant energy savings . This provides insight into the ubiquity of sensory organ mobility in animal design . This study exposes important links between the morphology of the sensorium , sensorium mobility , and behavioral strategy for maximally extracting energy from the environment . An “infomechanical” approach to complex behavior helps to elucidate how animals distribute functions across sensory systems and movement systems with their diverse energy loads .
Animals must constantly negotiate trade-offs in sensory and motor performance . The most well known of these trade-offs occur within either movement or sensory systems , rather than between them . As an example within motor systems , fish body shapes and styles of movement that maximize cruising efficiency may suffer from poor maneuverability [1]–[3] . In sensory systems , converging signals from large numbers of photoreceptors for increased sensitivity results in reduced spatial resolution . What about trade-offs between movement and sensory systems ? For example , for a fixed amount of available energy from food sources , is it better to expend that energy on a larger visual sensing range ( via a larger eye and the brain tissue to process signals ) , or to move the body more so that the effective area that is scanned is similar ? One challenge in assessing such trade-offs is that it is difficult to compare measures of movement performance , such as energy efficiency , to sensory performance , such as acuity . Ultimately , however , these different subsystem performance measures translate into net energy gains and losses for the animal [4] . Consequently , examining energy provides a lens through which to look at how an animal can best trade off movement and sensing . Given that neuronal tissue requires about 20 times more energy than skeletal muscle per unit mass in mammals , where it has been measured ( [5] , after [6] ) , we already know that brains and sensory systems are metabolically expensive compared to movement systems . Recent studies have shown the important influence of the energetic costs of sensory systems , such as the role of these costs in the evolution of sensory systems ( review: [7] ) . Although looking at energetics enables comparison of the costs of movement and sensing in behaviors where these are closely interrelated , such an analysis has rarely been performed [7] . One simple source of trade-offs between movement and sensing can be easily understood . A key role of a sensory system is to support scanning the environment for food , threats , mates , competitors , or anything else which may affect the animal's continued existence . But the space where these items of interest exist will typically exceed the range of the sensory system . To scan a larger volume of space , an animal can move its body , or evolve increased sensory range . Either approach has its associated costs . In the case of body movement , it is the cost of locomotion . The amount of locomotion needed will depend on the range of the sensory system being used , with less movement needed by long-range systems , such as vision , and more movement needed for short-range systems , such as touch . If , for example , you need to detect the location of a split on a wood table , you can use your visual system and glance at the entire surface at once ( little dependence on movement ) , or you can move your hands across the surface and use your sense of touch to detect the split ( maximal dependence on movement ) . In the case of evolving increased sensory range , the associated costs include more neuronal tissue , development costs , maintenance , and the cost to carry the weight of the sensory system ( not insignificant for flying animals: the fly uses 3% of its energy simply to keep its visual system aloft [8] ) . The above type of trade-off between movement and sensing is indirect because the problem is how best to expend a fixed amount of energy ( more on movement , and less on sensing , or vice versa ) — but not a case where improvement in one domain comes at the expense of performance in the other domain . An example of a more direct trade-off like this is how moving the eye faster to increase the speed of visually inspecting an area of space can directly conflict with visual performance . The conflict arises when an image passes over more than one photoreceptor acceptance angle per response time , since this results in the visual percept being degraded by motion blur [9] . A thought experiment can help expose another direct way in which a trade-off between movement and sensing can occur , one similar to the kind at issue in this study . As the effective range of a given sensory epithelium approaches zero ( contact sensing ) , to increase the amount of space that is scanned while moving through space ( for example , in a straight line ) can require reorienting the sensory epithelium in a way that results in less efficient movement . For example , imagine your finger was an autonomous organism . Suppose this finger is feeling its way along a novel surface in a water current ( or a stiff wind ) , with the long axis of the finger parallel to the direction of movement so as to minimize drag effects . Now , the back portion of the finger is scanning the same surface as was already scanned by the front . To increase the amount of space being scanned per unit time , the sensory epithelium needs to be reoriented . Ideally , the finger would be oriented perpendicular to the line of travel . This way the rate of surface scanning is maximized; but now there is also maximal projected area in the direction of travel , and thus maximal drag . Contrast this situation with that involving a sensory epithelium whose range is far from zero , such as the retina of an eagle flying high and looking for prey on the ground . Now , suppose that the eagle is looking straight downward . The eagle's visual sensorium can be idealized as a cone whose apex is the eagle's head . The area scanned per unit time will be the width of the cone times the velocity of the eagle . If instead of looking straight down , the eagle sweeps its conical sensorium from side to side by moving its eyes , it will greatly increase the area scanned per unit time . In this case , however , to reorient the sensory epithelium through eye rotation comes at no change in the projected area of the body in the direction of travel , and thus no added costs due to increased drag . If the eyes were not movable , the eagle would have to turn its head , which could result in more drag; if the eye and head were not movable , the whole body would need to be reoriented , incurring even more costs . Note , however , that having the ability to reorient the sensory epithelium without changing body orientation can incur significant neuronal processing costs , since it may require coordinate transformations from a sensory organ-fixed coordinate frame ( e . g . , retinotopic coordinates ) to body-fixed coordinates . With sensors distributed over a sensory epithelium consisting of the entire body surface , as occurs in somatosensory and electrosensory systems , it becomes progressively less possible to reorient the sensory epithelium independently of full body reorientation . For example , it conflicts with the strategy of concentrating the sensors on one portion of the body which is moved with muscles , as with some eyes and pinnae . Full body reorientation , however , can be quite costly if the relative velocity between the body and the surrounding environment is sufficient to produce drag forces on the body — for example , if the animal is moving rapidly through the air . An example of this type of trade-off between sensing and movement can be found in chemosensory behavior of the blue crab [10] . Blue crabs move sideways up-current , with their body slightly rotated into the flow . The slight rotation into the flow is believed to result in improved sensing of the local gradient of odorant molecules , as this rotation causes their primary chemosensory appendages for this behavior—their legs [11] —to be sensing slightly across the flow . Without this slight rotation , the downstream legs receive fluid in which the odorant has been mixed and diluted from hitting the upstream legs , compromising the ability to detect and localize the odorant . With the body rotated into the flow , the crab avoids this dilution and can use bilateral comparisons between chemosensory input along the legs to help guide the body to the source . However , turning the body into the flow also increases drag . As Weissburg and coworkers increased flow speed in their experimental apparatus , they found a speed at which the crab chose not to rotate the body into the flow . The cost of movement at the increased drag appears to outweigh the gain in sensory performance at this critical flow speed . Here we present an analysis of a conflict between efficient movement and sensory performance using the model system of weakly electric fish ( Figure 1A ) , a leading system for the analysis of sensory function in vertebrates . These fish hunt for small insect prey at night in rivers of the Amazon Basin , through the use of an active electrosensory system . The fish generate an oscillating electric field ( near the body ) , that surrounds the whole animal . When prey enter the fish's electric field , a small change in voltage occurs across the skin ( ) [12] , [13] . This change in voltage is detected by electroreceptors covering the entire body surface . These voltage modulations are then transformed into changes in the firing rate of primary electrosensory afferents that terminate in the hind brain of the animal for further processing ( reviews: [14] , [15] ) . While searching for prey , these fish were previously shown to hold their body with the head down at a pitch while searching for prey [16] , as illustrated in Figure 1B . We show that this posture significantly increases the cost of movement . However , this increased cost is more than offset by the increase in sensory performance resulting from the posture . We observed that this increase in sensory performance is dependent upon the fish having an elongated sensorium . When we examined the effect of the fish having a non-elongated sensorium , such as a blunt-shaped sensorium or a forwardly-directed visual sensorium similar in aspect ratio to a visually-guided aquatic predator , we found that there was no benefit to increasing the pitch of the body . We show that if the black ghost could swivel its sensorium independently of body movement , as visually-guided animals can swivel their sensoria , the fish would obtain a significant benefit through reduced energy expenditure for prey search .
Body movement through any medium results in lost energy due to friction between the medium and the body . In air these effects are slight except for flying animals . In water , with 1 , 000 times the density of air , these effects are significant even at relatively low speeds . As mentioned above , black ghost knifefish tilt their body while searching for prey . To estimate the energetic consequences of tilting their body from neutral ( horizontal ) body pitch to the measured , the force needed to overcome the resistance to movement ( drag ) through water needs to be estimated at different body pitches and movement speeds . The energy needed to overcome this resistance is then simply this force times the distance moved . We estimated the drag in two ways . First , we performed high resolution computational fluid dynamic simulations of the black ghost as it was being virtually towed through water . The forces on the body are easily recovered from the simulations , as are the flow patterns , which give insight into the basis of the drag forces corresponding to each body pitch angle . The computed flow patterns are shown in Figure 2 . Second , we towed an accurate urethane cast of the knifefish through a large water tank at constant , behaviorally relevant velocities , measuring the steady-state resistance to movement with a force sensor that the cast was attached to . We highlight results for 15 cm/s , because our prior prey capture study with the black ghost knifefish found search velocities of 9 . 34 . 3 cm/s ( mean and std ) [16] . In that study , the tank in which we made our observations had to be small due to imaging constraints , making 15 cm/s a reasonable choice to focus on here . The drag force results are shown in Figure 3 . At 15 cm/s , the measured drag force was 2 . 00 . 4 mN ( ) , 5 . 20 . 4 mN ( ) , and 8 . 10 . 5 mN ( ) . The corresponding computed drag forces were 1 . 0 , 6 . 1 , and 12 . 2 mN . The measured drag was typically lower than the computationally estimated drag . As shown in the snapshots of the computed flow patterns around the fish being virtually towed at 15 cm/s in Figure 2 , at the flow separation is higher than in the other cases . Because of this degree of separation , computational fluid dynamic simulations that incorporate the effect of turbulence may be required to fully resolve the flows around the body . If turbulence is present in the empirical experiments with the fish cast , this could potentially reduce the drag . Given the disparities between measured and computed drag forces , we use the measured drag forces for the remainder of the study . Our key result , that observed pitch angles during search behavior are consistent with minimizing costs , are not affected by this choice . The fish has an omnidirectional field of prey sensitivity [12] ( Figure 4A ) because of the broad distribution of sensors and electric field described above . This volume is relatively uniform , although there are significant non-uniformities in electric field strength and sensory receptor density [12] . As shown in Figure 4A , as the fish increases its body pitch , the amount of space that it scans while moving increases . The volume the fish can sense prey within while moving is the product of the frontal area of the sensorium ( the area that results from projecting the volume to a plane at right angles to the direction of motion ) , and the distance traveled . For a cuboidal idealization of the complex natural shape of the sensory volume ( see Materials and Methods ) , we found that the projected frontal area increased with body pitch up to a maximum at a body pitch of ( Figure 5A ) . At neutral body pitch , the frontal area was , going up by 190% to at and up by 235% at . Our energetics model estimates the amount of energy needed to overcome drag forces for the fish to swim to a single prey of the kind used in quantifying the size and shape of the sensorium , Daphnia magna . These prey are typically found in stomach content analyses of Apteronotus albifrons [17]–[19] and have known energy content ( Table 1 ) . We assume that prey are uniformly distributed at the density shown in Table 1 . As derived below in Materials and Methods , the equation for estimating the energy in joules needed to overcome drag to reach a single prey is ( 1 ) is the power needed to overcome drag at the reference velocity ( during steady state swimming , thrust power must be equal to the power needed to overcome drag ) . We fixed to the power needed to overcome the experimentally measured tow drag at pitch and 15 cm/s , which was 0 . 3 mW ( 15 cm/s times the drag force at this velocity , 2 mN , Figure 3 ) . is the density of prey ( see Table 1 ) . is a function of body pitch angle which returns the area of the sensorium projected to a plane perpendicular to the path of motion . is a function of body pitch such that the drag force is equal to , where is the velocity of the fish . As shown in Figure 5B for the curve labeled “2 . 2 ( natural ) ” the energy needed to encounter one prey at neutral pitch was slightly over 25 J , going down by around 40% to near 15 J at the optimal pitch of just over , with a similar value at a pitch of . Changing the pitch of the body not only affects the drag on the body , and the search rate , it also affects propulsion . The black ghost knifefish generates force by undulating the extended ribbon fin along its underside ( Figure 1A ) while keeping its body semirigid except for bends to turn left or right [20] , [21] . The fin undulations are approximately sinusoidal and travel from one end of the fin to the other—from head to tail for forward movement . The fin generates two different forces: one along the length of the fin ( called surge ) , and one smaller force perpendicular to the fin , pushing the body up ( called heave ) [21] . As the fin tilts , the forward propulsive force reaches a maximum when the fin base is at an angle of approximately to the horizontal . This is its angle when the body axis is horizontal ( e . g . , when , then the fin base is at angle in Figure 1B , approximately ) . As the fin base tilts past ( body pitch ) , the sum of the surge and heave forces projected to the forward direction decreases . This effect is shown by Figure 6 , which depicts a family of curves relating forward propulsive force to body pitch ( Figure 6 ) . For the purposes of this illustration , we assume that the fish varies its frequency of undulation to vary propulsive force . This appears to be true [20] . We examined the influence of sensorium shape on the energy needed to encounter prey . We define the “elongation factor” as the ratio of the length to height of the sensorium . The effect of elongation factor on projected sensorium area and energy to encounter one prey is shown in Figure 5 . The naturally observed elongation factor is 2 . 2 . When the elongation factor was 1 . 0 ( sensorium length equal to height ) , the energy needed per prey decreased negligibly at low angles before increasing with body pitch angle; essentially , there was no improvement in performance with pitching the body . When the elongation factor was 4 . 0 ( sensorium length four times its height ) , the energy needed decreased with body pitch angle up to pitch angles of . With this elongation factor , the energy needed per prey encounter was typically less than half the energy per prey encounter for the 2 . 2 elongation factor at relevant body pitch angles . Sensorium elongation makes body pitching progressively more advantageous . The effect of blunt versus elongated sensoria was further explored through a scenario in which the black ghost has a frontally-directed visual sensorium ( see Figure 7A ) rather than its normal omnidirectional sensorium ( Figure 4A ) . A fish called the stone moroko ( Pseudorasbora parva ) is a visual predator whose vision-based sensorium for Daphnia has been measured ( [22] ) and is shown in Figure 4B . A cuboidal approximation of the stone moroko visual sensorium is 11 . 9 cm high ( vertical ) ×12 . 0 cm long ( distance of leading edge from the eyes ) ×18 . 7 cm wide ( left-right extent ) . The elongation factor , length over height , is therefore close to 1 . 0 . Given this aspect ratio , there is only a very slight increase in the swept volume of the sensorium with swiveling of the volume in pitch ( see the “1 . 0 ( blunt ) ” curve in Figure 5A ) . As shown in Figure 4B , this cuboidal approximation overestimates the effect of pitching the conical visual sensorium of the stone moroko . We will simplify the analysis slightly by 1 ) making the idealization that projected area does not change with pitch angle because of the aspect ratio of the visual sensorium , and by 2 ) allowing the projected area of the electrosensory sensorium at , ( Figure 5 ) , stand for the projected area of the visual sensorium at , which is 11 . 9 cm18 . 7 cm or . This facilitates comparison to the electrosensory case . The energetic consequence of this visual sensorium is then obtained by clamping the projected area ( ) term of the equation to its value when the body is pitched at , as shown in Figure 7C by the black ‘+’ curve . The energy to overcome drag monotonically increases; the benefit of holding the body at a pitch is lost . Long range sensing organs , such as eyes and pinna , are often clustered and invested with muscles that enable them to rotate , which in turn rotates their associated sensorium . What effect does sensorium mobility have on the amount of energy needed to encounter prey ? In another hypothetical scenario , we examined the consequences of the fish being able to pitch its sensorium around its head without moving its body , illustrated in Figure 7B . We do this by clamping the drag force ( ) term of the equation for above to its value at , with the result shown in Figure 7C by the red ‘x’ curve . There is a substantial decrease in energy needed per prey . Whereas this sensorium mobility is not biologically possible due to the near-field and broadly distributed nature of electrosense , this example serves to illustrate how sensorium mobility for a far-field sensory system can have beneficial consequences .
Given the limited availability of energy , all animals must balance the energy load of sensory and neuronal systems with motor and other body systems . However , active sensing animals such as bats , dolphins , and electric fish , have a particularly stringent constraint: they must generate the energy required to perceive their world . Both emitted energy and energy reflected from objects falls as [23] , so that the total power attenuation is inversely proportional to . By this , a doubling of sensory range takes sixteen times more energy . As an example of how constraining the physics of active sensing energy attenuation is , we consider the power the electric fish has to emit to detect prey . The electric fish's self-generated electric field allows them to detect prey at less than a body length away from the body [16] . The energetic cost of electric signal generation was recently measured at 3–22% ( depending on time of day and gender ) of the total metabolic rate [24] . For a 350 J/day total energy budget for the black ghost knifefish [25] , this amounts to a peak of up to about 80 J/day . This power level enables them to detect prey at up to 3 cm [16] . To detect prey at twice this distance , or 6 cm , would require , or 16 times more energy , or 1 , 280 J—four times the total energy budget of the fish . Although the signal generation power measurements used here are for a different species of South American weakly electric fish , the argument is hardly affected even at an order of magnitude lower power . Given these simple estimates , while all animals have to contend with trade-offs between more energy devoted to sensory systems versus other systems , we can expect these trade-offs to be especially clear in active sensing animals such as electric fish . Figure 8 shows one of our key results in summary form . We have found that as the body pitch increases from zero to , the drag force increases by a factor of between 2–4 times at a search swimming velocity of 15 cm/s . However , this increase in pitch angle also results in a near doubling in the search rate as quantified by projected area of the sensorium . In the simplified model , the balance of these two factors , which is quantified by the energy required to reach one prey ( Figure 5B ) , results in a best pitch angle of around . This results in a 40% energy saving over swimming at . Put another way , the number of prey encountered over a given distance of movement will be nearly doubled due to the near doubling of projected sensorium area . The measured fish pitch angle during search was [16] , significantly different from the optimum found here . There is an additional factor which will have the effect of reducing this disparity . This is a reduction in propulsive force from the ribbon fin with increased pitch angle , as shown in Figure 6 . In this figure , each solid curve shows how the thrust from the fin , with a traveling wave at the indicated frequency , decreases as the body pitches . Across the different undulation frequencies , the propulsive effectiveness of the fin drops around 25% at . If this effect were to be fully incorporated , the optimal swimming angle would clearly be less than . To illustrate this relationship , consider the dashed line of Figure 6 , which shows the drag force on the body when total power expended for swimming ( ) is clamped at a specific value , calculated given the drag force ( ) at a pitch angle of and a velocity ( ) of 15 cm/s ( 0 . 2 mN15 cm/s , which is 0 . 3 mW ) . Given that , an increase in drag requires a decrease in velocity to keep power fixed , resulting in a lower velocity as drag increases with pitch angle . Therefore , the dashed line indicates the thrust needed to overcome this drag when the power available is the same as when the fish swims horizontally . The intersection between the dashed line and the thrust curve indicates the approximate pitch angle required to move with a constant velocity , as propulsive thrust balances body drag resulting in zero acceleration . In particular , at the highest undulation frequency shown , 6 Hz , the fish would need to swim at , significantly below the pitch that would be best if loss of thrust with increased pitch were not a consideration . While we have found that the mechanical energy needed to find each prey is on the order of tens of microjoules , a small fraction of the energy gained per prey ( on the order of a joule; see Table 1 ) , the mechanical energy expended to overcome drag is only a fraction of the total energy the animal will use in finding each prey . This is because 1 ) not all the energy in food is converted to available energy [26]; 2 ) not all the available energy is used for swimming muscles ( e . g . , we estimate the mechanical power used for swimming at pitch and 15 cm/s is 0 . 3 mW ( the velocity times the drag at this velocity , 15 cm/s2 mN ) , while metabolic rate is on the order of 0 . 4 mW [25] ) ; and 3 ) not all the energy used for swimming muscles is converted into thrust . These factors combined are around a factor of ten . There is also significant uncertainty in the prey density numbers . The energy needed per prey will double if the prey density is half that used for these estimates ( 5 , 000 prey per cubic meter ) . The density appears to vary between 1 , 000–5 , 000 individuals per cubic meter for rivers typically inhabited by Amazonian electric fish [27] , [28] . However , this includes many different insect species and it is unclear what fraction of these are prey the fish would eat . Despite these uncertainties , the fish has few ways at its disposal for increasing search rate at a given velocity beyond changing pitch angle . For example , it cannot increase its sensorium size because it does not vary its electric field strength , although another species of weakly electric fish has recently been shown to vary its electric field strength [29] . Thus the increased mechanical load on the fish with increased body pitch is an appropriate variable to examine . A key factor in the beneficial effect of pitching the body is the shape of the sensorium . More specifically , how the projected area of the sensorium changes as a function of the sensorium position control variable , in this case body pitch ( ) , is crucial . As the sensorium becomes less elongated , the increase in projected area with increased pitch angle becomes negligible , and thus the benefit of body pitching disappears . This is shown in Figure 5A and B . As the sensorium becomes more elongated ( elongation factor 4 . 0 ) , the projected area increases more rapidly with pitch angle , and the net energy needed per prey decreases more rapidly . The opposite holds for the cube-like sensorium ( elongation factor 1 . 0 ) : there is nearly no increase in projected area with pitch angle , and thus the energy needed per prey only increases with pitch angle due to increased drag forces . As another way to examine this effect , we computed the energetic consequences of the black ghost using a visual sensorium , illustrated in Figure 7A . The visual sensorium for the detection of the same type of prey used in this study , Daphnia , in a visual predatory fish ( the stone moroko ) has been measured to be 11 . 9 cm high ( vertical ) ×12 . 0 cm long ( distance from eye to leading edge ) ×18 . 7 cm wide ( left-right ) ( Figure 4B ) . This sensorium has an elongation factor ( length to height ) of unity , so the projected area changes little with rotation in pitch . As a visually-guided animal with movable eyes , the stone moroko can choose to rotate its eyes with its oblique muscles to control the pitch angle of its sensorium [30] . For the purpose of this example , let's facilitate the comparison to the elongated body-fixed sensorium of the black ghost by supposing that this artificial visual sensorium is also body-fixed , as depicted in Figure 7A . Thus , the fish changes the pitch of its body to change the pitch of the sensorium . The effect of this faux visual sensorium on energy is shown in Figure 7C . There is no benefit to pitching when the effect of the elongated sensorium is removed , and only the cost of overcoming drag remains for the artificial case of a body-fixed visual sensorium . These results indicate that an elongated sensorium is beneficial . In this particular group of species , an elongated sensorium goes along with an elongated body that is characteristic of the knifefish body plan , common across some 180 different species ( Gymnotidae ) [31] , and the distributed nature of the electrosensory system of these fishes . For the stone moroko , a fish which swims by “tail-wagging” ( the carangiform mode ) , the instability in yaw induced by tail beating results in high yaw maneuverability [1] , and would facilitate prey capture lateral to the fish body . In addition , left-right eye movements will sweep the sensorium in azimuth . Therefore , this fish's vertically flattened sensorium , over one-and-a-half times wider than it is tall , seems likely to be beneficial . Further amplifying this point , Figure 4B shows that pitching the sensorium would in fact decrease the swept volume slightly . The relevant elongation factor for this fish will be length to width , since height will only have a constant factor effect on how projected area changes with azimuthal angle . Weakly electric fish have a body-fixed sensorium . If it were at all possible to change the position of the sensorium without changing body position , as animals that rotate their eyes or turn their heads can [9] , one possible scenario would allow the animal to have all of the sensory advantages of pitching the body , with none of the drag costs . In this scenario , imagine the fish could tilt the back of its sensorium up as illustrated in Figure 7B , but without tilting the body—analogous to how some animals can rotate their visual sensorium without moving their bodies . We can assess the energy implications of this scenario through the use of Eq . 2 , by fixing the drag force ( ) to its value when the body is at pitch , while allowing to vary . The result is shown by the ‘x’ curve of Figure 5B . Being able to dynamically reposition the sensorium without moving the body results in more than a factor of two decrease in energy per prey at , and even more at larger angles . Decoupling sensorium movement from whole body movement has been an ancient theme of vision , our most powerful teleceptive sensory modality . Independent eye movement and stabilization goes back to the very first vertebrates [32] . There are many benefits to eye movements , such as minimization of motion blur due to self movement and movement of the object of fixation [9] , but clearly not having to reposition the body to see something initially out of view can economize on energy [30] . Given that body mass is considerably larger than sensory organ mass , it also saves on time . One cost , however , is the need to translate the coordinates of perceptual information arriving in sensory-organ-fixed coordinates to the coordinates of the body , demanding significant neuronal processing . The tectum , or superior colliculus , is one structure where this occurs ( review: [33] ) . Whereas eye movement is quite ancient , the ability to turn the head is relatively recent in vertebrates . Our earliest evidence of this ability is from a 375 million year old fossil of an animal that appears to be a transitional form between fish and tetrapods , Tiktaalik roseae [34] . Some active sensing animals exploit head movements for sweeping their sensoria horizontally and vertically while keeping their body on a fixed course . For example , bats nearly double the angular range of their sonar-based sensorium by combining pinna and head movements [35]–[37] , and dolphins have also been shown to use head movements to manipulate their sonar-based sensorium to a similar end [38] . Rats also exploit this freedom , combining head movements with whisker movements to palpate objects [39] . Having relatively light and independently movable sensory appendages is a ubiquitous feature of animal body plans . It is particularly powerful for teleceptive systems such as vision and audition . The analysis here highlights how advantageous it can be to decouple sensorium movement from whole body movement from an energetics standpoint . It may also suggest that when developing assistive technologies for people with sensory challenges , a sensorium whose movement is independently controllable from body movement can be particularly helpful . Although there is a significant literature of how mechanical considerations enter into sensory performance in a large number of systems , and a growing literature on the metabolic cost of information , there has been little examination of how these two domains overlap and trade-off with one another . While measures of performance in these two areas typically are not commensurable , the impact of a change in sensing or movement on the net energy balance of an animal provides a basis of comparison . We have been able to quantify how this animal trades-off movement efficiency for sensory performance in prey search behavior . A simplified model illuminates why the animal searches with its body in a drag-inducing position , and suggests a possible basis for why this group of animals has evolved an unusual degree of elongation in their body plan . This model also illustrates the benefits of sensorium mobility that is decoupled from whole-body movement . In the traditional view , the nervous system performs the computational “heavy lifting” in an organism . This view neglects , however , the critical role of morphology , biomaterials , passive mechanical physics , and other pre-neuronal or non-neuronal systems . Given that neurons consume forty times more energy per unit mass than structural materials such as bone [40] , and twenty times as much as muscle ( [5] , after [6] ) , there are clearly advantages to distributing tasks between these tissues in a way that improves energetic efficiency . In this “bone-brain continuum” view [41] , animal intelligence and behavioral control systems can only be understood using integrative modeling approaches that expose the computational roles of both neural and non-neural substrates and their close coupling in behavioral output . The infomechanical approach taken here , in which information and mechanics are jointly examined with regard to energy consequences , is one such approach that can facilitate a more integrative understanding of animal system design .
An accurate urethane cast of a 190 mm long Apteronotus albifrons made for a prior study [42] was bolted to a rigid rod . This was suspended from a custom force balance that used three miniature beam load cells ( MB-5-89 , Interface Inc . , Scottsdale AZ USA ) . For force balance and calibration details , see [43] . The fish cast was towed through a large tank that was in length , width , and depth ( GALCIT towtank , Caltech ) using a gantry system driven by a speed-controlled DC servomotor above the tank [43] . Trials were conducted at three speeds: 10 , 12 , and 15 cm/s , and three angles to the flow: , , and . Only the data collected after the startup force transient had settled was analyzed , until just before the end of the towing distance ( 300 cm ) . The data was filtered with a digital Butterworth low pass filter ( cutoff at 5 Hz ) to remove transducer transients prior to further statistical analysis . We used a custom computational fluid dynamics solver to obtain the drag force on a fish model at different towing velocities . The fish model was derived from the same urethane cast as was used for the tow-tank measurements [42] . It is assumed to be rigid . In the numerical simulations , it is towed at 10 , 15 , and 20 cm/s , and three angles to the flow: , , and . All of the simulations were performed using the San Diego Supercomputer Center's IA-64 Linux Cluster , which has 262 compute nodes each consisting of two 1 . 5GHz Intel Itanium 2 processors running SuSE Linux . The computational fluid dynamics code was written in Fortran 90 and C ( for details , see [21] ) . In a prior study we used a combination of empirical measurements and computational models to determine the 3D volume around the fish body where a typical prey item , Daphnia magna , could be detected ( Figure 4A ) [12] . We idealized the resulting electrosensory sensorium as a cuboid ( Figure 9 ) whose width , height , and length is matched to the maximal dimensions of this volume , after scaling for body size ( the body length for the [12] study was 14 . 4 cm , while it is 19 . 0 cm in this study ) . The resulting dimensions are shown in Table 1 . As shown by Figure 9 , the projected area of this cuboidal sensorium in the direction of travel ( its silhouette if you were to look at it directly along the path of its approach ) is simply . We varied the ratio of the length to the height of the cuboidal sensorium . To assess the impact of elongation of this volume on projected area with pitch angle , we varied the ratio of the length to the height of the cuboidal sensorium ( the elongation factor ) . These two dimensions were chosen because by the above equation for , varying the width only results in a constant factor change in the projected area with respect to . The naturally observed elongation factor was 2 . 2 . We assume that prey are uniformly distributed at the density shown in Table 1 . As shown in Figure 9 , the projected search area is . Thus , the total water volume scanned for prey when the fish moves distance will be . The number of prey detected in that volume will be the volume times the prey density , or . The distance travelled to get one prey will then be . We fit our measured drag data to a function of the form , where is in degrees . The result is , where and , with an of . Thus thrust power . We can rearrange this to solve for . We rearrange to solve for and use the solution for from above to solve for , the time required to find one prey . Then we multiply this by to solve for the energy expended to overcome drag in obtaining one prey: ( 2 ) For , we used the power needed to overcome the experimentally measured tow drag at pitch and 15 cm/s , which was 0 . 3 mW ( 15 cm/s2 mN ) ( Figure 3 ) . For a previous study we used a computational model of a non-translating , non-rotating fin deforming in a sinusoidal pattern with time [21] . The instantaneous velocity of each point on the fin is specified as a function of time . The no-slip and no-penetration boundary conditions are imposed on the surface of the fin using an immersed boundary formulation , and the fluid flow around it is fully resolved using finite difference methods of order in space and order in time . The complete details of the computational algorithm and method are given in [21] , [44] . Mean forces on the fin were calculated as the time average of the hydrodynamic forces on the fin over at least one period of oscillation , after a quasi-steady state is reached . As shown in [21] , the force in newtons from the fin followed the correlation ( 3 ) where is a constant equal to 86 . 03 , is the density of water ( ) , is the frequency of the traveling wave on the fin ( Hz ) , is the maximal angular excursion of the traveling wave ( radians ) , is the fin length ( m ) , is the height of the fin ( m ) , is the wavelength of the traveling wave ( m ) , and is a function of the specific wavelength which can be approximated by: ( 4 ) This equation estimates the propulsive force parallel to the fin , or surge force . However , in addition to this force , the fin also generates a small force that is perpendicular to the fin base , pushing the body upward . This force , termed heave , has a magnitude of about 25% of the surge force for typical motion patterns [21] . Because of the relative magnitudes of the surge and heave forces , the angle of the fin that would maximize forward thrust is . This angle is nearly identical to the observed fin insertion angle on the body ( in Figure 1B ) when the fish is swimming straight . By knowing the surge force , and this angle , we can therefore compute the heave force as the tangent of the fin base angle times the surge force . As the body pitches , the contribution of the parallel surge force to thrust will vary with the cosine of the sum of the body pitch angle and fin base angle , whereas the contribution of the normal heave force will vary with the sine of the sum of these two angles . Thus the net force will be: ( 5 ) where is the body pitch angle , and is the angle of the fin base with respect to the body axis at body pitch ( ; shown in Figure 1B ) . For these force estimates , we used the length of the fin of the fish used for drag estimates ( 12 . 7 cm ) , a fin height of 1 cm , and typically observed kinematic values of an , and two waves along the fin ( cm ) [20] , [21] . To compare thrust to drag when the power expended on swimming is fixed , we derive the relationship between the drag function and swimming power . Based on and , . Thus . To assess the effect of sensorium shape , we examined elongation factors of 1 . 0 and 4 . 0 by changing the sensorium length to be equal to its normal height , and four times its normal height , respectively . We then examined the energetic consequences of these sensorium morphologies . This was done through the equation describing the energy needed per prey encounter described below ( Equation 2 ) through changing the function ( ) that returns projected sensorium area given the pitch of the body . For the artificial elongation factors of 1 . 0 and 4 . 0 , we make the following simplification . A change in elongation factor normally would be accompanied by a change in body elongation . This is because the electric organ and sensors , which together form the sensorium [12] , are along the full length of the fish; therefore a change in relative length of the sensorium would necessitate a change in body length . Any change in body length would in turn affect the drag force on the body and thus the energy needed per prey through the term of Equation 2 . Although this was not considered here due to the extensive computational demands of the drag study , the results of simple sensitivity analyses suggest that this simplification has negligible effect on the qualitative trends . We examined the energetic consequences of two “what if” scenarios: 1 ) There is no increase in projected sensorium area as the body pitches . To do this , we clamp to its value at . 2 ) There is no increase in body drag as the body pitches . This would be the case if the fish were able to independently control the pitch angle of its sensorium , analogous to how animals with movable eyes or heads can change the position of their visual sensory volume without changing body position . To do this , we clamp the drag term to its value at . | Animals thrive by sensing their environment and using the information they've gathered to guide their movement . But collecting better information can result in less efficient movement: Bicycling while standing up on the pedals may help you see over obstacles ahead of you , but it causes more air drag , forcing your legs to work harder . Nocturnal weakly electric fish search for prey with their body tilted . This tilting more than doubles the resistance to movement from the water , but because the fish's ability to sense prey improves when tilted , it is better to swim this way . Beyond a certain amount of tilt , the costs of movement become too great . Interestingly , the benefit of tilting is dependent on the shape of the volume around the fish where it detects prey . We also found that if the fish was able to swivel its region of prey sensitivity , like a vision-based animal can shift its gaze , it would save energy . This conclusion helps us understand why animals like us can move our eyes . A Polish folk saying succinctly captures the gist: “He who doesn't have it in the head has it in the legs” ( Ten kto nie ma w głowie ma w nogach ) . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"neuroscience/behavioral",
"neuroscience",
"neuroscience/sensory",
"systems",
"neuroscience/motor",
"systems",
"computational",
"biology/computational",
"neuroscience",
"evolutionary",
"biology/animal",
"behavior"
]
| 2010 | Energy-Information Trade-Offs between Movement and Sensing |
Temperate phages , the bacterial viruses able to enter in a dormant prophage state in bacterial genomes , are present in the majority of bacterial strains for which the genome sequence is available . Although these prophages are generally considered to increase their hosts’ fitness by bringing beneficial genes , studies demonstrating such effects in ecologically relevant environments are relatively limited to few bacterial species . Here , we investigated the impact of prophage carriage in the gastrointestinal tract of monoxenic mice . Combined with mathematical modelling , these experimental results provided a quantitative estimation of key parameters governing phage-bacteria interactions within this model ecosystem . We used wild-type and mutant strains of the best known host/phage pair , Escherichia coli and phage λ . Unexpectedly , λ prophage caused a significant fitness cost for its carrier , due to an induction rate 50-fold higher than in vitro , with 1 to 2% of the prophage being induced . However , when prophage carriers were in competition with isogenic phage susceptible bacteria , the prophage indirectly benefited its carrier by killing competitors: infection of susceptible bacteria led to phage lytic development in about 80% of cases . The remaining infected bacteria were lysogenized , resulting overall in the rapid lysogenization of the susceptible lineage . Moreover , our setup enabled to demonstrate that rare events of phage gene capture by homologous recombination occurred in the intestine of monoxenic mice . To our knowledge , this study constitutes the first quantitative characterization of temperate phage-bacteria interactions in a simplified gut environment . The high prophage induction rate detected reveals DNA damage-mediated SOS response in monoxenic mouse intestine . We propose that the mammalian gut , the most densely populated bacterial ecosystem on earth , might foster bacterial evolution through high temperate phage activity .
Bacterial viruses , called bacteriophages or phages , are present in all bacterial communities and have profound impact on bacteria either by killing them or by mediating horizontal gene transfer through lysogeny . Lysogeny refers to the ability of temperate phages , as opposed to virulent ones , to repress their lytic multiplication after infection and stably segregate with the bacteria . In most cases , the repressed phage , or prophage , is integrated into the bacterial chromosome , but it can also replicate as an extrachromosomal element in the bacterium . Nearly all bacterial genomes contain one or multiple prophages , which can constitute up to 14% of the genome for Escherichia coli strains [1] . Active prophages can be induced , i . e . switch back to lytic multiplication in response to a signal such as DNA damage and subsequent SOS response ( reviewed in [2] ) . Induction rates are usually too low to result in a cost to their host , and prophages were generally found to have positive impacts on lysogenic bacteria [3–6] . The benefits of lysogeny can result from three distinct mechanisms: ( i ) lysogenic conversion , by which phages bring useful bacterial accessory traits [4 , 7]; ( ii ) immunity , i . e . protection against other phages , as the prophage protects its carrier bacterium against the same , and sometimes other , phages [8]; and ( iii ) allelopathy , by releasing infectious virions that are able to kill susceptible bacterial competitors . While induction results in the death of the lysogen , it can provide a competitive advantage for the remaining lysogenic population . A large number of major bacterial toxins , such as the diphtheria , Panton-Valentine , cholera , Shiga- or scarlatin toxins are encoded on temperate phage genomes ( reviewed in [7] ) . However , pathogenicity does not always increase bacterial fitness in a human host , suggesting that some pathogenic traits can be coincidental ( reviewed in [9] ) . To our knowledge , except for Staphylococcus aureus , only a small proportion of prophages were demonstrated to carry beneficial traits for their bacterial host , such as improvement of the colonization of body surfaces—like intestine [10 , 11] , nasopharynx [12] , or skin [13]—or resistance to protozoa grazing [14 , 15] . The allelopathic character of temperate phages has been demonstrated by in vitro experiments and mathematical modelling [16 , 17] , but also recently during insect infection [18] . However , very few data exist concerning the impact of prophages on the fitness of their hosts in the most densely populated bacterial ecosystem , the intestine of mammals . Metagenomic studies have shown that gut bacteria harbor many temperate phages [19] , but whether carrying a prophage is generally costly or advantageous for its host has been rarely investigated in the intestinal environment [20 , 21] . A well documented case of beneficial interaction is the filamentous temperate phage of Vibrio cholerae VPIΦ , which encodes factors essential for bacterial adherence and intestine colonization [10 , 11] . E . coli prophages carrying Shiga toxin stx genes are known to be active in the intestine , but their excision and lysogenization rates were not quantified [22 , 23] . Another study demonstrated that a prophage of an Enteroccocus faecalis strain provided a 1 . 5-fold growth advantage after 24 hours of mouse gut colonization [24] , but the mechanisms involved were not entirely explored , nor the impact of prophage presence after the first day of colonization . The costs or benefits of lysogeny in the gastrointestinal tract cannot be inferred from in vitro studies , since the parameters that rule phage-bacteria interactions vary greatly with the environment , bacterial physiology and medium structure . For example , the lysogenization rate of phage λ , i . e . the proportion of infected E . coli bacteria that are lysogenized upon infection , varies from 10−3 when infecting cells in optimal growth conditions , to 0 . 5 when infecting starved cells [25] . This rate also varies with temperature and multiplicity of infection [26] . Three other main interaction parameters can be distinguished: ( i ) the induction rate , ( ii ) the adsorption rate onto the bacterial host , i . e . affinity of the phage for its receptors , a parameter that greatly depends on ionic conditions [27] , and ( iii ) the multiplication rate within the host . Up to now , none of these parameters has been determined for a temperate phage in the gut environment . Yet , characterizing temperate phage activity is essential to estimate their impact on lysogenic bacteria , and to evaluate the extent of the horizontal gene transfer they mediate in this environment . This point is of paramount importance because temperate phages are major actors of bacterial genome evolution , and as such they participate to the emergence of new pathogenic strains . Moreover they are suspected to be important disseminators of antibiotic resistance genes [28] . The extreme complexity of the gut microbiota prevents any exhaustive characterization of all the virus-host systems it hosts . It is thus necessary to first characterize specific virus–host systems in a controlled microbiota to bridge the existing gap between in vitro studies and the functional characterization of natural gut microbial communities . We used monoxenic mice , i . e . mice associated with a single bacterial species , to perform competition experiments between two isogenic E . coli strains , one carrying the λ prophage and the other devoid of it . These experiments , supported by a mathematical model consisting of five ordinary differential equations , allowed disentangling the different components of the impact of the prophage on bacterial reproductive fitness . We obtained quantitative estimations of the main parameters driving phage-bacteria interactions in monoxenic mouse intestine . Moreover , we demonstrate that efficient phage spreading enabled rare events of phage-mediated gene capture by homologous recombination , and transmission to new bacteria .
To characterize phage-bacteria interactions in the mouse digestive tract , we colonized germ-free mice with two isogenic E . coli MG1655 strains , except for antibiotic resistance markers and the presence of the λ prophage ( λble phage confering phleomycin resistance to the lysogen ) . Populations of free phage ( V ) , bacteria from the lysogenic lineage ( L ) , from the susceptible lineage ( S ) and newly lysogenized by λ ( SL ) were quantified in mouse feces for one week , based on their differential antibiotic resistance levels . During the first day of colonization , phage propagation , via free phage production and lysogenization of the susceptible bacteria , was highly efficient: after 24 hours of colonization , an average 73% of the bacteria from the initially susceptible lineage were either killed or lysogenized , and a transient increase in free phage had occurred ( Fig 1A ) . To determine what fraction of free phage was produced by multiplication on susceptible bacteria , as opposed to free phage produced by spontaneous induction in lysogens , the same experiment was repeated with lamB derivatives of the two strains , devoid of the phage receptor and resistant to phage infection ( Fig 1B ) . In such conditions all free phages result from the spontaneous induction of prophages in lysogenic bacteria . At the peak of free phage production , the free phage over lysogen ratio was 20-fold lower than in the experiment with wt strains ( Fig 1C ) , indicating that the transient increase of free phage observed with these strains resulted from multiplication on susceptible bacteria , and not from a transient increase in induction rate . By comparison , when the same S and L strains were co-cultured in vitro , in standard rich LB medium , phage propagation was almost undetectable ( Fig 1D ) , in line with previously published results [29] . This absence of propagation was due to low Mg2+ concentration in LB , drastically limiting λ adsorption ( [27] and Fig 1D and 1E ) . Addition of maltose did not improve phage propagation , suggesting that LamB expression in LB is sufficient for phage infection [30] . In mice , phage propagation stopped before the complete lysogenization of the S lineage . To test whether this resulted from changes in gut bacteria impairing infection , mice were monocolonized with the susceptible strain S only , and bacteria from feces were tested for affinity to λ ( Fig 2A ) . After one day , λ adsorption rate on bacteria from mouse feces was measured at 3 . 10−7 ml h-1 , which is similar to the value measured at day 0 in vitro . Therefore the phage receptor LamB is highly expressed in the mouse gut , and favorable ionic conditions allow for efficient binding . Later on however , the adsorption rate diminished continuously , suggesting a decrease in LamB expression ( Fig 2A ) . To investigate this phenomenon further , we determined the susceptibility to λ of S clones isolated from mouse feces two days after colonization . Nine out of the twelve clones tested turned out to be genetically resistant to λ , and were moreover unable to use maltose , as revealed by their inability to grow on minimal medium containing maltose as the unique energy source . In subsequent colonization experiments , we quantified the increase in maltose-deficient bacteria ( Mal- ) by using maltose agar plates containing a tetrazolium dye that turned red in Mal- colonies ( Fig 2B ) . Mal- bacteria were selected in the S and L lineages ( Fig 2C ) . A similar rise in Mal- bacteria occurred in mice monocolonized with the phage-free strain S , demonstrating unambiguously that their selection is not caused by λ ( Fig 2C ) . Mal- and λ resistance phenotypes , as well as previously published results [31] , guided our identification of mutations in the malT gene . MalT is the transcriptional activator of the maltose regulon . It notably controls expression of the λ receptor LamB . All six resistant clones studied carried a mutation in malT , among which three led to a truncated protein ( S1 Fig ) , which explains that the selected mutations prevent phage infection . The reason for the selection of these mutants might be linked to the LamB-induced envelope stress associated with osmoregulation [30] since bacteria in the gastrointestinal lumen are continuously exposed to osmotic stress ( reviewed in [32] ) . They are specific to monoxenic mice , as malT mutations are not selected for in the MG1655 E . coli strain when colonizing mice with a conventional microbiota [33] . The rise of malT mutants was nevertheless sufficiently delayed to permit observation phage infection of the majority of S bacteria during the first two days . In order to provide quantitative estimations of the parameters governing phage-bacteria interactions , we developed a mathematical model representing the dynamics of the different microbial populations in this model ecosystem ( Fig 3 ) . The model is based on the one in [17] and adapted to take into account our experimental settings . It consists of five coupled differential equations , representing time evolution of five population densities: S ( susceptibles ) , L ( lysogens ) , SL ( newly-lysogenized susceptibles ) , V ( free phage ) , as well as latent bacteria Q in which the phage undergoes lytic multiplication . Invasion of malT mutants is not included in the model . S1 Text gives a detailed description of the main modeling assumptions behind its construction , as well as a mathematical analysis of its dynamical behavior . A careful examination of the effect of the eight model parameters onto the dynamics enabled the quantitative estimation of six of them from our experimental datasets ( Table 1 ) . With these estimated values , numerical simulations of the model ( Fig 3C ) are in good agreement with experimental observations on the first two days , before invasion of malT mutants , suggesting it captures most of the relevant information contained in our data . The main discrepancy observed is in the initial velocity of the temporal evolutions , faster in the model than in experimental data . This might result either from incorrect estimation of some parameters , or from the neglect of a phenomenon not taken into account in the model , such as the binding of free phage on some intestinal components . Such binding would result in a “loss” of phage that would slow the dynamics , as exemplified by the effect of reduced burst size ( S2 Fig ) . Upon infection of a susceptible bacterium , λ goes to lysogenization with a high probability around 19% , leading to a very rapid rise of SL bacteria both in data and in numerical simulations ( Fig 4A ) . The remaining infected susceptibles were lysed , resulting in an increase of the L lineage relative to the S one , independently of the initial L/S ratios ( Fig 4B ) . LamB deletion abolished the competitive advantage of the L lineage ( Fig 4B , dotted line ) , confirming that the advantage of lysogens only stemmed from the lysis of susceptible competitors , and not from the presence of putative bacterial fitness genes in the λ genome that would improve growth in mice . However , the gain of the L lineage over the S one is limited by the lysogenization of susceptible , independently from the rise of λ resistant mutants . Its final value , as predicted by the model , seems to be directly proportional to the inverse of g at population equilibrium ( Fig 4C ) . Interestingly , other parameters governing phage-bacteria interaction have very modest impact on the final gain of the L lineage ( S3 Fig and S1 Text ) . Prophage induction rate in the mouse gastrointestinal tract was estimated to be 1 . 6% , several orders of magnitude higher than usually assumed . This high induction leads to a slight but systematic decrease of lysogens ( L and SL lineages ) in mice when bacteria are resistant to infection , either because of malT or lamB mutations ( Figs 1A and 1B and 5A ) . By contrast , in vitro , induction rate is 3 x 10−4 ( Table 2 ) , and competitions under conditions that did not permit phage infection resulted in a stable proportion over time of lysogenic ( L and SL lineages ) and non-lysogenic S bacteria ( Figs 1D and S3A ) . In mouse , model-based estimation of the induction rate was derived from latent Q cell counts in mouse colonized with lamB strains . Because of the small data set available ( one experiment with three mice ) , the confidence interval is relatively large ( 0 . 6%-3 . 6% , see Table 1 and S1 Text ) . In order to strengthen the estimation , we also computed the induction rate from the relative fitness of L compared to S lamB lineages ( Material & Methods ) . The value found ( 1 . 7% ± 0 . 5% ) was very close to that estimated by the model . To examine experimentally the impact of high induction rate , we used a non-inducible λ prophage , λcIind- , which has a mutation in the repressor of the lytic cycle , CI , preventing its RecA activated auto-cleavage upon DNA damage . As expected , in standard in vitro conditions , the cIind- mutation decreased the induction rate 1 , 000-fold ( Table 2 ) . In the mouse gastrointestinal tract , the mutation abolished the decrease in proportion of lysogens ( Fig 5B ) , demonstrating unambiguously that high prophage induction explains the disadvantage of lysogens . Moreover , in a lamB genetic background , S and λcIind- lysogenic strains presented no reproductive fitness differences over 9 days ( S4B Fig ) , validating the model hypothesis that in the absence of lysis and induction , the presence of the prophage makes no difference in growth rate . Interestingly , λcIind- experiments also validated the absence of a rarity threshold to phage multiplication in the mouse gut: since phage amplify on susceptible bacteria , even a very low initial number of phage can lead to killing of a significant part of S lineage ( Fig 5B and 5C ) . The switch from lysogenic to lytic cycle requires CI autocleavage , catalyzed by RecA nucleofilament formed by DNA damage [36 , 37] . In the λcIind- lysogens , RecA mediated CI autocleavage is prevented , and the few phage produced are CI low expression mutants [36] . Indeed , free phage isolated from feces of mice colonized with λcIind- lysogens formed clearer plaques than λ wild-type , which suggest they have a lower lysogenization rate . Sequencing of the cI gene from 9 phages isolated at day 2 either from free phage in feces or from SL bacteria revealed they all had a point mutation in the -35 box of CI promoter , PRM ( G->T , -33 relative to the cI start of transcription ) . Interestingly , this PRM mutation was previously shown to enable λ prophage induction in the absence of SOS activation , by decreasing by 80% intracellular CI levels , leading to much higher switching rates from the lysogenic to the lytic states [36] . Indeed , these PRM mutants , named λcI* , had an induction rate 50 , 000-fold higher than that of the ancestral λcIind- phage and 300-fold higher than the wild-type ( Table 2 ) . Measurement of induction rate from 12 other SL bacteria revealed they were all lysogenized by λcI* . The high induction of this virulent mutant enabled its propagation during the first days of colonization . However , in agreement with evolutionary epidemiology theory , that predicts that selection for virulence decreases with the pool of susceptible hosts [38] , the virulent λcI* mutant was counter selected later on in the prophage form , due to killing of its host through induction ( Fig 6 ) . We next investigated whether this high phage activity allowed for gene exchange between the phage and bacterial genomes . We have previously reported that λ captures bacterial genes by homologous recombination during the lytic cycle , at frequencies ranging between 10−4 and 10−6 depending on the extent of homology between the DNA segments ( Fig 7 and [20] ) . In our experimental system , recombination can lead to the incorporation of the chloramphenicol resistance gene ( cat ) of the L strain into the phage genome , since L bacteria have the cat gene in a chromosomal region of partial homology with λ ( 88% identity ) . We investigated the occurrence of this phenomenon in mice . Recombinant phages can be detected in their lysogenic form as they confer chloramphenicol resistance to the bacteria they are integrated in . On days 1 and 2 , recombinant prophages were detected in all mice , at frequencies around 5 . 10−8 relative to the number of new lysogens ( SL ) . PCR analysis confirmed that the cat gene was placed at the expected position in the λ prophage . No recombinants were detected with a λ phage deleted of its main recombination gene , bet ( or redβ ) , indicating the importance of phage recombination function for gene acquisition .
Complete genome sequencing of thousands of gut bacteria has shown that most harbor prophages , yet their impact on strain fitness in the gastrointestinal tract has rarely been investigated . Colonization experiments , supported by a mathematical model of phage/bacteria interactions , show that the advantage of λ lysogeny in monoxenic mice gut is valid only when susceptible bacteria are present; a situation that might be only occasional in the gut microbiota . Indeed , it is supposed that only one or two E . coli strains cohabit at the same time in the human gastrointestinal tract [39 , 40] , and moreover , to our knowledge λ phage infects only a small proportion of E . coli strains . In the absence of susceptible competitors , the prophage was costly for its host , due to frequent induction caused by DNA damage . Prophages were generally shown to positively impact their host fitness , and our study is , to our knowledge , the first demonstration that a prophage can be detrimental to bacteria in the gastrointestinal tract . The level of phage λ induction observed in monoxenic mice was remarkable: 1 to 2% of lysogenic bacteria were lysed per generation , which is almost two orders of magnitude higher than in standard laboratory conditions , in which induction was too low to constitute a measurable fitness cost ( Figs 1C and S4 ) . This result is reinforced by another study showing that the induction rate of 933W lambdoïd prophage is higher in the mouse gastrointestinal tract than in vitro , and constant over time [41] . However , the reporter assay used did not permit direct estimation of the induction rate and the associated cost for the bacteria [41] . We observed that the λ repressor mutation CIind- , which abolishes CI auto-cleavage , dramatically decreased prophage induction in the intestine . Many reports over a long period of time have proven unambiguously that for such cleavage to occur , a RecA nucleoprotein filament ( also called “activated RecA” ) must catalyze the reaction [36 , 42 , 43] , so lambda prophage induction reflects DNA damage . Since RecA nucleoprotein filament also triggers the general SOS response [44] , our results indicates that this response is activated in 1 to 2% of bacteria in the intestine of monoxenic mice . DNA damage sensing is not only responsible for the induction of most prophages [2] , but it also triggers activation of other mobile genetic elements such as integrative and conjugative elements or ICEs [45] , transposons [46] and integrons [47] . Moreover , the large number of defective prophages in E . coli genomes , i . e . prophages incapable of independent induction or particle formation [48–51] , suggests a regular selection in favor of bacteria having lost or inactivated these prophages , possibly in response to their frequent induction in the intestine . Interestingly , at least in the simplified gut environment used in this study , induction cost is not compensated by the putative adaptative genes lom , bor and rex described in the λ genome [52 , 53] . The low induction rate λcIind- phage used in this study was on the contrary beneficial to its host , since it confers no induction cost while still enabling efficient killing of competitors through amplification on susceptibles . However , phage mutants with higher induction are strongly selected for when susceptible bacteria are present [38] , as exemplified in our study by the selection of the λcI* virulent phage ( Fig 6 ) . Interestingly , all E . coli lambdoïd phages tested have comparable levels of induction [54] , close to that of lambda , suggesting they have all evolved toward the same optimal induction rate for propagation . Alternatively , it was proposed that optimal induction rates evolved to benefit the bacterial host–and thereby prophage vertical transmission- by a “bacterial altruism” mechanism [55] . Indeed , in the case of Shiga-toxin carrying prophages , prophage induction leads to the release of toxins killing bacterial protozoan predators , benefiting the bacterial host even in the absence of susceptible competitors . The level of phage λ lysogenization unraveled was also remarkable: we estimated that in monoxenic mice gastrointestinal tract the lysogenization rate is close to 20% . By comparison , in vitro , lysogenization rate is close to 0 . 1% in bacteria growing rapidly [54] , but almost 50% in starved bacteria [25 , 56] . This lysogenization rate estimated in mouse is therefore much higher than generally measured or assumed [16 , 17 , 57 , 58] . Since the final gain of the original phage carrier on the susceptible strain is inversely proportional to the lysogenization rate ( our results and [57] ) , in the gastrointestinal tract high lysogenization results in a smaller gain of the original phage carrier than previously described . Theory predicts that high lysogenization optimizes phage reproduction in an environment where the density of susceptible hosts is low or variable [58 , 59] . Although the gut microbiota is the densest bacterial community on earth , it includes hundreds of different species and thousands of bacterial strains , possibly making highly specific phage infection relatively rare . Low phage susceptibility seems to be the conclusion of a large-scale study of phage-bacteria interactions in a gnotobiotic mouse model [60]: in mice raised with a simplified microbiota composed of 15 strains belonging to dominant human species , only two were attacked by a cocktail of thousands of different phages isolated from a human gut microbiota . Moreover , a higher proportion of temperate phages was found in gut viromes than in other environments [19 , 61] . Altogether , these data support our results , suggesting that in the gastrointestinal tract the lysogenic life cycle of phages is favoured compared to lytic multiplication . Temperate phages being major actors of horizontal gene transfer in bacteria , a concern emerged recently regarding their role in the propagation of antibiotic resistance genes [62] . Indeed , some phage particles are vectors of antibiotic resistance genes [28] . Most of the time , gene transfer occurs by generalized transduction , the erroneous encapsidation of bacterial DNA . Such errors are rare: for instance , the proportion of E . coli phage P1 capsids leading to the production of an antibiotic-resistant clone is between 10−5 and 10−6 [63 , 64] . The incorporation of a bacterial gene into phage genome , and afterwards transfers by lysogenization is a much rarer event , that we could detect only when the gene was located in a defective prophage sharing homology with λ [20] . In the present study , we estimated the frequency of such gene capture ( cat gene , conferring chloramphenicol resistance ) by λ in mice to be 10−8 . Interestingly , up to now very few cases of phages encoding resistance genes have been reported [65–67] , suggesting that even if the lysogenization rate in the intestine is very high , the risk of antibiotic resistance spread mediated by temperate phage is low . Here we found that in monoxenic mouse gastrointestinal tract , lysogeny initially benefits its host during competitions with susceptible bacteria , in line with previous studies in other environments [16–18] . The mathematical model highlighted that the benefit of the original lysogenic strain depends critically on the lysogenization and induction rates: the lower these parameters , the higher the benefit . We also show that in monoxenic mice gastrointestinal tract , DNA damage leads to high prophage induction , which results in a significant cost for the lysogen . Provided that DNA damage observed in monoxenic mice gut also occurs in conventional animals , since all E . coli lambdoïd phages tested have comparable levels of induction [54] -and since most E . coli prophages are lambdoïd [68]- our results might prove to be general . Due to the highly specific phage-bacteria interactions , we hypothesize that the absence of bacteria susceptible to a particular phage in the gastrointestinal tract might regularly occur , and that on the long term , the parasitic aspect of at least some active prophages prevails .
All bacterial strains are described in S1 Table . All strains were constructed by modifying the MG1655 ΔfliC ΔompF strain . This strain was used because ompB mutations are rapidly and systematically selected in the MG1655 strain in the mouse gut as a result of their effects on flagellin ( FliC ) repression and of decreased membrane permeability via repression of the major porin OmpF [69] . As ompB mutants also display a reduced expression level of LamB [31] , a maltoporin used by phage λ for infection , we used a ΔompF ΔfliC strain in which no ompB mutations were selected [69] . The stfR::cat mutation was introduced in this strain by phage P1 transduction from the MD19 strain described in [20] . The ΔlamB strains were constructed by phage P1 transduction of the lamB::KanR cassette from the Keio collection strain JW3996 [70] . In the MD56 and MD74 strains , the KanR cassette was excised as described in [71] . The λ receptor being absent in ΔlamB strains , λ prophage was introduced by transformation with Urλble purified DNA . The λble phage strain used in this study was constructed by insertion of the phleomycin resistance gene ble into the Urλ strain of λ , as described in [20] . The λcIind- mutant contains a mutation ( A111T ) in the RecA cleavage site: the alanine in position 111 is replaced by a threonine [72] . This mutation was introduced by recombineering with the oligonucleotide AT111 ( GTAAAGGTTCTAAGCTCAGGTGAGAACATgCCgGttTGgACATGAGAAAAAACAGGGTACTCATACC ) . Small letters represent changes in the DNA sequence . Several neutral differences were added to the one necessary for the amino acid change in order to avoid recognition by MutS . Recombineering was performed in the HME57 strain [73] , which carried plasmid pKD46 [71] , and lysogenized with λble . The strain was co-transformed with two oligonucleotides , AT111 and Court’s lab oligonucleotide 144 , conferring it the ability to use galactose [74] . After transformation , colonies were isolated on M9 minimal galactose plates . 96 Gal+ clones were screened for the absence of spontaneous phage induction , by scoring the absence of infectious phage particles in culture supernatants . 1 out of 96 clones had the expected mutation , which was confirmed by sequencing of the cI gene . The λcIind- phage was next introduced into the MG1655 ΔfliC ΔompF stfR::cat strain by P1 transduction and selection on phleomycin plates . Germ-free C3H/HeN mice were bred at the germ-free animal facilities of the INRA Micalis Institute , Anaxem . Mice were reared in isolators and fed ad libitum on a commercial diet sterilized by gamma irradiation ( 40 kGy ) and supplied with autoclaved tap water . For colonization experiments , 8 week-old germ-free female mice were gavaged with 106 bacteria from the chosen strain , or the appropriate mixture of the two strains , in 0 . 1 mL of M9 minimal medium . The cassettes used to differentiate strains during competition confer resistance to chloramphenicol or to kanamycin . Their expression is known to have no significant cost during E . coli intestinal colonization , so no inversion of markers was performed [31 , 75] . Bacterial and phage populations in feces were monitored by colony forming unit ( CFU ) and plaque forming unit ( PFU ) counts in freshly harvested individual fecal samples , as described below . Feces were homogenized in a 10-fold volume of sterile water before dilution in LB and plating on LBA plates with the appropriate antibiotics . PFUs were enumerated in the supernatant of suspended feces centrifuged 3 minutes at 12 , 000 g . All procedures were carried out in accordance with the European guidelines for the care and use of laboratory animals . The project received the agreement of the local DDPP ( n° A48-195 ) and from the local ethic committee for animal experimentation , the Comethea ( n° 13–05 ) . After serial dilutions , bacterial populations were determined by plating on selective antibiotic-LB agar plates ( 1 . 5% agar ) . Antibiotics were used at the following concentrations: kanamycin ( 50 μg/mL ) , chloramphenicol ( 20 μg/mL ) , and phleomycin ( 5 μg/ml ) . PFUs were determined by spotting 10 μl of serial dilutions of diluted feces on a lawn of the indicator bacteria in top agar ( 0 . 4% agar , 10 mM MgSO4 ) . The indicator bacterial culture was fresh MD5 culture grown in LB containing 0 . 2% maltose . Latent bacteria were counted similarly but after elimination of free phage by centrifugation . CFUs and PFUs were counted after 12–16 hours of incubation at 37°C . The ability to use maltose was monitored in tetrazolium maltose ( TM ) indicator plates . Mal+ and Mal- clones respectively form white and red colonies on these plates . The TM medium was composed of tryptone ( 10 g/L ) , yeast extract ( 1 g/L ) , NaCl ( 5g/L ) , agar ( 16g/L ) , maltose ( 5 g/L ) and tetrazolium dye ( 50 mg/L , Sigma ) . The technique was essentially that of Hendrix [76] , with minor modifications . Adsorbing bacteria from feces were prepared as described for enumeration . A control culture was grown at 37°C with shaking in LB + 0 . 2% maltose + 10 mM MgSO4 up to an absorbance at 600 nm of 1 . 2 ( about 6 . 108 bacteria/ml ) . Aliquots of 200 μL of culture were added to 50-μL aliquots containing 500 phage particles in 400-μL PCR tube strips , and mixed at 37°C . Several strips were prepared , one for each time point . At the chosen time points , the bacteria were separated from the free phages by centrifugation . The PFU in the supernatants were enumerated as described above . PFUs at time zero were estimated by titering the phage suspension . The slope of a graph plotting the logarithm of the number of bacteria remaining unadsorbed as a function of time allowed us to calculate the adsorption rate a according to the equation Nt = N0 x e-Bat , where Nt and N0 are the numbers of phage particles unadsorbed at time t and at time zero respectively , B is the number of bacteria per millilitre , and t is time in hours . To calibrate the mathematical model , we used different datasets and versions of the model , aiming at the disentanglement of the eight parameters ( see S1 Text for details ) . We identified four groups of parameters , which were estimated sequentially . The first group consists of parameters r , k and d that directly control the general growth of bacterial populations . As a global parameter , independent of the bacterial strains , the dilution rate was fixed to d = 0 . 25 , in line with [77–79] . Subsequently , parameters r and k were estimated from single species experiments in mice , using a mono-dimensional logistic equation and nonlinear least square minimization . The second group consists of parameters x and l . A closer look at the equations points out the central role of these two parameters on the dynamics of latent cells Q . To further separate the estimation from phage-dependent parameters , we considered a LamB- version of the model ( i . e . with a = 0 ) implying only variables L , S and Q . This simplified model was then fitted to LamB- experimental data . The third group gathers phage-dependent parameters y and a . Based on the formula for the equilibrium population of free viruses , a linear relationship was obtained between the two . Using a value of y calculated from in vitro data , the corresponding value for the adsorption constant was deduced . Finally , the last parameter g ( probability of lysogenization ) was estimated by fitting the ratio L/ ( S+SL ) in L-S competition experiments . The best fit was obtained at 36 h . All estimated values are listed in Table 1 . When possible , bootstrap techniques were used to compute 95% confidence intervals ( see S1 Text ) . All numerical computations have been performed with Matlab ( The MathWorks , Inc . ) . Burst size was estimated in vitro in conditioned LB , i . e . the supernatant of a bacterial culture grown in LB ( +0 . 2% maltose , +10 mM MgSO4 ) at 37°C up to an absorbance at 600 nm of 1 . 5 . In such conditioned media we measured a growth rate of 1 . 1 h-1 , close to that measured in the mice gastrointestinal tract during the first 24 hours of colonization . For single burst experiments , 100 μl of a culture at OD600 of 1 . 5 is mixed with 1x105 λ phage . The mix of phage and bacteria is incubated 7 minutes at 37°C , centrifuged and washed twice to eliminate free phage , and then diluted 1 , 000 times in conditioned LB at 37°C . Samples were taken every 10 minutes , mixed with 200 μl of exponentially growing susceptible bacteria and 3ml of Top Agar ( 4 . 5 g/L , 0 . 2% maltose , 10 mM MgSO4 ) , and then poured on LB agar plates . Plaque-forming units ( PFU ) were counted . The burst size ( y ) is the factor between maximum and initial PFU counts . We determined that y = 12 . 1 ± 8 . Since induction rate is the sole parameter differentiating growth of S and L lamB lineages , the induction rate ( x ) was determined from the evolution of the L/S ratio between days 0 and 9 in function of the number of generations . The number of generations was estimated by assuming that growth rate is equal to r , 1 . 1 h-1 , during the first 24 hours , and after equal to excretion rate d , 0 . 25 h-1 . A linear regression on data from 6 mice ( lm function in R software ) gave x = 0 . 0170 ± 0 . 0046 . We adapted a method allowing for an measurement of the induction rate per bacterium [80] . Lysogenic bacteria were diluted 200 fold in LB and grown at 37°C with shaking . When specified , 0 . 8% w/v bile salts ( cholic acid-deoxycholic acid sodium salt mixture , sigma-aldrich ) were added when the absorbance at 600 nm of the culture was 0 . 2 or 0 . 1 . When the absorbance reached 0 . 4 , cultures were swirled on ice for 5 minutes and washed twice at 4°C to eliminate free phages . Washed lysogenic bacteria were then mixed at the appropriate dilution with 100 μl of a saturated culture of the indicator strain , 3 ml of top agar were added , and the mix was plated on LBA-ampicillin plates ( 50 μg/ml ) . The indicator bacteria were ampicillin-resistant ( strain JC10990 recF::Tn3 AmpR ) . Ampicillin prevents further lysogen growth but does not prevent the completion of the lytic cycle if already started , which results in an infective centre . The induction rate was calculated by dividing the number of infective centres by the number of plated lysogenic bacteria . | Dormant bacterial viruses , or prophages , are found in the genomes of almost all bacteria , but their impact on bacterial host fitness is largely unknown . Through experiments in mice , supported by a mathematical model , we quantified the activity of Escherichia coli prophage λ in monoxenic mouse gut , as well as its impact on its carrier bacteria . λ carriage negatively impacted its hosts due to frequent reactivation , but indirectly benefited its host by killing susceptible bacterial competitors . The high prophage activity unraveled in this study reflects a constant rate of SOS response , resulting from DNA damage in monoxenic mouse intestine . Our results should motivate researchers to take the presence of prophages into account when studying the action of specific bacteria in the gastrointestinal tract of mammals . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"bacteriology",
"organismal",
"evolution",
"medicine",
"and",
"health",
"sciences",
"gut",
"bacteria",
"bacteriophages",
"microbiology",
"vertebrates",
"mice",
"animals",
"mammals",
"viruses",
"animal",
"models",
"dna",
"damage",
"model",
"organisms",
"microbial",
"evolution",
"bacterial",
"genetics",
"dna",
"microbial",
"genetics",
"bacteria",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"mouse",
"models",
"gastrointestinal",
"tract",
"biochemistry",
"rodents",
"anatomy",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"evolutionary",
"biology",
"bacterial",
"evolution",
"organisms"
]
| 2016 | Carriage of λ Latent Virus Is Costly for Its Bacterial Host due to Frequent Reactivation in Monoxenic Mouse Intestine |
An important application of the RNA interference ( RNAi ) pathway is its use as a small RNA-based regulatory system commonly exploited to suppress expression of target genes to test their function in vivo . In several published experiments , RNAi has been used to inactivate components of the RNAi pathway itself , a procedure termed recursive RNAi in this report . The theoretical basis of recursive RNAi is unclear since the procedure could potentially be self-defeating , and in practice the effectiveness of recursive RNAi in published experiments is highly variable . A mathematical model for recursive RNAi was developed and used to investigate the range of conditions under which the procedure should be effective . The model predicts that the effectiveness of recursive RNAi is strongly dependent on the efficacy of RNAi at knocking down target gene expression . This efficacy is known to vary highly between different cell types , and comparison of the model predictions to published experimental data suggests that variation in RNAi efficacy may be the main cause of discrepancies between published recursive RNAi experiments in different organisms . The model suggests potential ways to optimize the effectiveness of recursive RNAi both for screening of RNAi components as well as for improved temporal control of gene expression in switch off–switch on experiments .
RNA interference ( RNAi ) is an RNA-mediated pathway of gene silencing mediated by small RNA molecules [1] , [2] . During RNAi , introduction of double-stranded RNA ( dsRNA ) encoding a sub-sequence of a gene leads to reduction in expression of the corresponding gene . The heart of the RNAi process involves two key steps . First , the dsRNA is cleaved into small RNA fragments by an enzyme called Dicer , and then these small fragments are used as a template by a complex called RISC which identifies matching sequences in target messages and leads to their degradation . RNAi technology has emerged as a powerful tool for artificially controlling gene expression , but it only works because cells have evolved small RNA based regulatory pathways in the first place . Natural regulatory pathways taking advantage of small RNAs include not only classical RNAi , which probably acts in host defense against viruses and transposons , but also microRNA-based ( miRNA ) regulatory pathways that regulate endogenous genes [3] . It is interesting to speculate that such pathways may have evolved in part because of unique aspects of regulation mediated by RNA . Compared to more classical regulatory networks based on transcription factors or kinases , the signal-processing properties of small RNA-based regulatory systems have not been extensively investigated at a theoretical level . One advantage of having a theoretical understanding of such pathways is that one could potentially predict the performance and response of systems that have been altered in defined ways , thus facilitating a “synthetic biology” of small RNA-mediated regulatory circuits [4] , [5] . For a more short-term application , one might hope that a predictive level of understanding of RNAi pathway behavior could allow improved design of experiments using RNAi as a tool . In this report the RNAi system is explored theoretically by considering its behavior following addition of an artificial negative feedback loop . It is well known in electronics that when the output of a circuit is fed back into one of its inputs , the resulting closed-loop circuit can have dramatically different behaviors than the open-loop circuit before the feedback loop was added . A key challenge for systems biology is to be able to predict the effect of feedback loops on biological circuits , either naturally occurring feedback or synthetic feedback produced by adding new linkages from output to input [6] . In the case of naturally occurring small RNA-mediated regulatory loops based on micro-RNAs , feedback loops are sometimes seen in which components of the RNAi/miRNA machinery such as Dicer or Argonaute are themselves targets of miRNA-mediated inhibition [7] , [8] . Being able to quantitatively or even qualitatively predict the effect of such feedback linkages would therefore seem crucial to developing a circuit theory for small RNA based signaling [9] . In the case of the RNAi pathway , synthetic feedback loops have been constructed by workers attempting to use RNAi to turn off the RNAi pathway . This is done simply by adding dsRNA molecules that target genes encoding components of the RNAi machinery . In such a situation , the feedback can be considered as arising from the output of the RNAi machinery ( that is , degradation of target message ) being applied as an input to the system in the form of message encoding RNAi components . This “recursive” RNAi has been used in genome-wide screens to discover new RNAi components [10]–[14] . In such screens , a reporter gene such as green fluorescent protein ( GFP ) or luciferase is silenced by RNAi , and then reporter activity is measured in the presence of a second dsRNA molecule targeting a candidate gene . Increased reporter expression indicates that the candidate gene is involved in the RNAi process . By using libraries of dsRNA molecules corresponding to all predicted genes in the genome , it is in principle possible to identify all components of the RNAi machinery . In order for screens of this type to be successful , the reporter activity must be significantly increased over the level seen when the reporter alone is targeted . Recursive RNAi has also been used as a way to reactivate genes previously silenced by RNAi . Such “switch-off/switch-on” experiments employ a procedure in which a dsRNA is introduced targeting a gene of interest , and then , following a period of inactivation , the RNAi is alleviated by adding a second dsRNA that targets the RNAi machinery itself [15] . This allows temporal control of gene expression during animal development , and has the advantage that it can be applied to any gene without having to engineer new inducible constructs for each experiment . In order for switch-off/switch-on experiments to work , the level of restoration of the targeted gene must be enough to restore approximately normal gene function . For strictly recessive genes this would probably require restoration to approximately half normal levels , while for haploinsufficient genes it would require a greater degree of restoration , to near wild-type levels . Recursive RNAi can thus potentially be a very powerful tool both for studying RNAi itself and also for controlling gene expression during development , provided a sufficient level of restoration can be achieved once the RNAi machinery is targeted . Despite the great potential of recursive RNAi , and the multiple published successes of the method , one cannot help but feel that the use of RNAi to inactive RNAi seems potentially self-defeating . Specifically , one might imagine that as the pathway is shut down , its ability to further shut itself down would be reduced , resulting potentially in a restoration of activity . Recursive RNAi presents the same difficulty as attempting to commit suicide by holding one's breath—even if one could hold one's breath to the point of passing out , the unconscious patient would at that point begin breathing again . The quantitative question thus arises as to whether introduction of recursive RNAi would provide a restoration of gene expression level that would be measurable or detectable relative to control levels . Indeed , in actual practice recursive RNAi doesn't always work . For instance , although some studies have reported that RNAi of genes encoding Dicer protein restores reporter gene expression [16] , other studies failed to observe significant restoration following RNAi of Dicer [11] . One possible explanation for the variability in results between different systems is the efficacy of RNAi at knocking down gene expression . Some cell types such as S2 cells can achieve extremely high levels of knockdown to a few percent of wild-type expression levels [11] while other systems such as C . elegans RNAi-by-feeding seem to produce a more moderate degree of knockdown . Might such variation make recursive RNAi possible in some systems and impossible in others ? This report investigates the conditions under which recursive RNAi can be effective , by constructing a mathematical model for recursive RNAi and predicting how its performance varies as a function of the efficacy of RNAi in a given system . The main prediction of the model is that increasing the efficacy of RNAi-mediated knockdown should make recursive RNAi less efficient and potentially impossible .
The RNAi pathway upon which the model is based is shown schematically in Figure 1 , and based on this diagram a model is presented in the Materials and Methods section below . Within the model , the steady-state behavior of the system is specified by a single parameter , γ , which determines the overall effectiveness of RNAi in a particular cell type . RNAi efficacy can be expressed in terms of the fold-knockdown achievable , that is , the ratio of expression level prior to RNAi relative to the expression level following RNAi . For instance , a gene whose expression is reduced to one half its normal level by RNAi would show a fold-knockdown of 2-fold . As derived in Materials and Methods , the fold knockdown predicted for a reporter gene such as GFP or luciferase , in the absence of any additional RNAi targeting Dicer or RISC , would be described in terms of an RNAi efficacy parameter γ according to the following equation: ( 1 ) Thus the parameter γ determines the efficacy of RNAi system , with larger γ indicating more extensive knockdown of gene expression . As described in Materials and Methods , and summarized in Table 1 , this parameter depends on all of the individual parameters of the detailed model , such as the catalytic rate constants of Dicer , the rate of mRNA degradation , etc . Many of the individual rate constants and parameters that contribute to γ may be extremely difficult to measure . In contrast , because of the simple relation between fold-knockdown and the value of the parameter γ this parameter is experimentally measurable simply by quantifying reporter level before and after RNAi . Typical values for γ are in the range 2–200 . Moreover , because the steady-state behavior of the system depends only on this one parameter γ , for many purposes it may not be critical to know the values of the detailed parameters given in Table 1 , as long as one knows the value of the aggregate RNAi efficacy parameter γ . In this paper the parameter γ is generally imagined to vary over the range 1–200 . The variations of the detailed parameters listed in Table 1 are not considered individually because their only effect on the model behavior is through their influence on the value of γ . A second model parameter β plays a role in determining the time-scale over which RNAi knocks down its targets , and is therefore also directly experimentally measurable . Because β has no effect on the steady-state level of knockdown , this parameter will not be considered except when the transient behavior of the system is analyzed . β and γ are the only two adjustable parameters of the model . Both parameters are phenomenological and easily measurable using standard methods of quantifying RNAi efficiency , but both parameters can also be defined in terms of detailed mechanistic parameters such as protein turnover rate , as described in Materials and Methods . When dsRNA is introduced to target a gene encoding a component of Dicer , the system stably attains a new steady state in which the level of the targeted Dicer-specific protein is partially reduced ( Figure 2 ) . As detailed in Materials and Methods , the model predicts that the inherent susceptibility of Dicer to knockdown by RNAi differs from that of a reporter gene , with the fold-knockdown for Dicer given by ( 2 ) The same equation is predicted to describe the susceptibility of RISC when it is targeted by recursive RNAi , indicating the two parts of the RNAi pathway have similar susceptibility to RNAi mediated knockdown . It is perhaps of interest to note that , for γ = 1 , corresponding to a two-fold knockdown of the reporter , the fold knockdown predicted for Dicer from Equation 2 is 2/ ( −1+√5 ) . This is the famous “Golden Ratio” , known since Greek antiquity to arise in situations involving self-similarity and recursion . The major biological significance of Equations 1 and 2 is that genes encoding components of the Dicer and RISC complexes are inherently less susceptible to RNAi knockdown compared to genes not involved in the RNAi pathway . This differential susceptibility raises questions about detectability of recursive RNAi . Would reporter gene expression be restored significantly if Dicer was simultaneously targeted ? As detailed in Materials and Methods , the model predicts RNAi-mediated reporter knockdown in the presence of RNAi targeting components of Dicer ( or of RISC—the equation ends up being the same ) to be: ( 3 ) Figure 3 graphs the predicted expression levels of a reporter gene targeted by RNAi in the presence ( Equation 3 ) or absence ( Equation 1 ) of recursive RNAi targeting Dicer , plotted as a function of the underlying RNAi efficacy in the system . Clearly , the level of reporter gene recovery depends on the efficiency of RNAi in the system , such that more effective RNAi predicts less recovery of reporter expression . As γ becomes large ( i . e . knockdown is very efficient ) , the reporter expression levels obtained with and without recursive RNAi gradually approach each other , making the effect potentially very hard to detect over measurement noise . These results can reconcile the apparent disagreement in the literature concerning the efficacy of recursive RNAi of Dicer , because the variation in RNAi efficacy ( as described by parameter γ ) between cell types and organisms should produce predictable variation in restoration ( Figure 3 ) . Comparison of the predicted restoration to published data reveals a remarkably good match . Bernstein et al [16] describe experiments in which a GFP reporter reduced to 15% of control levels by RNAi is restored to 40% of control levels when Dicer is simultaneously targeted . From Equation 1 , 15% knockdown implies γ = 5 . 5 , from which Equation 3 predicts restoration to 35% of control levels , consistent with the experiments . In a different cell type ( human HEK293 cells ) Schmitter et al [17] found that RNAi directed against a luciferase reporter knocked expression down to 45% of normal levels , and simultaneous targeting of Argonaute-2 restored expression to 60% of pre-RNAi levels . From Equation 1 , reporter knockdown to 45% implies γ = 1 . 2 , hence Equation 3 predicts restoration to 60% of control levels , exactly as observed . In these cases a moderately effective RNAi system yields substantial restoration during recursive RNAi , as predicted . In a contrasting example , Dorner et al . [11] describe a highly effective RNAi system in which the reporter was knocked down to 0 . 5% of control levels , corresponding to γ = 200 . Equation 3 predicts Dicer-specific RNAi should restore reporter expression only to 7% of controls , a relatively small recovery . Consistent with this prediction , Dorner et al . found that RNAi targeting a number of RNAi components such as Dicer-2 and R2D2 only increased reporter expression slightly to a few percent of control levels . A similar low level of restoration of reporter activity was reported in a separate study of RNAi of Dicer-2 in S2 cells [18] . In an even more extreme case , Hoa et al . [19] performed recursive RNAi in mosquito cells for which RNAi of luciferase knocks down the reporter 4000-fold . In this extremely efficient RNAi system , the authors found that targeting of Dicer only restored the luciferase reporter to 2% of control levels . A 4000-fold knockdown implies γ = 3999 , from which Equation 3 predicts a restoration of the reporter to 1 . 6% of control levels , again consistent with the observed level of restoration . These results suggest that poor restoration by recursive RNAi is likely to be a common feature of highly efficient RNAi systems . Dorner et al . [11] concluded in their study that most of the RNAi machinery genes tested in their experiments were not susceptible to RNAi . However , the model given here suggests the experiments were , in fact , effective , but due to the inherently self-limiting nature of recursive RNAi at high γ , the extent of recovery was simply not very large . The differences in performance between different systems are consistent with the predictions of the model for different values of gamma , but it is impossible to rule out that some of the differences could be due to differences in targeting sequences for the reporter versus for the RNAi machinery ( a point to be discussed further below ) . The predictions of this model regarding restoration achievable by recursive RNAi of Dicer only apply to experiments in which Dicer is targeted by addition of shRNA or other forms of dsRNA , and the limitation on knockdown is a result of the requirement for Dicer activity to generate siRNA against itself . If Dicer is targeted by directly by introduction of siRNA , then the model might predict a dramatically increased level of restoration since in this situation Dicer-mediated production of siRNA would no longer be required for its own knockdown . Consistent with this , experiments in which Dicer is targeted directly by exogenously introduced siRNA molecules show almost complete restoration of reporter activity [20] . On the other hand , dynamics of the system might be significantly different because while Dicer is not required to produce exogenously added siRNA , it may still be involved in loading these siRNA molecules into the RISC complex [21] . It is to be noted that different siRNA molecules can show extremely large differences in targeting efficiency [22]–[28] and unless the targeting efficiency of each construct is known , it is impossible to compare quantitative results between different constructs and systems , let alone compare a theoretical model with experimental data . Thus , the comparisons presented here should be viewed as showing a qualitative similarity in overall trends , with precise numerical equivalence being impossible to assess until targeting efficiencies are measured for each experiment . The foregoing results suggest that the effectiveness of recursive RNAi could be improved by reducing the effectiveness of RNAi , for example using mutant backgrounds with partial defects in one or more RNAi components . To optimize the design of recursive RNAi experiments , one approach is to define a figure of merit to describe restoration of reporter activity ( see Materials and Methods ) and then attempt to maximize its value . A figure of merit can be defined by the relative restoration ratio , R , which is the reporter-specific RNAi-mediated decrease in reporter level in the presence of Dicer RNAi divided by the decrease seen in the absence of Dicer RNAi . Figure 4 plots the value of R as a function of the RNAi efficacy parameter γ . It is easy to show that the restoration is maximal when γ equals 2 , which corresponds to a 3-fold reduction in reporter level . As overall RNAi efficacy increases past this point , the level of reporter gene restoration achievable by RNAi of RNAi decreases , in other words , the effect of recursive RNAi becomes more difficult to detect . An alternative figure of merit that may be more appropriate for certain types of screening experiments is the normalized absolute difference Δ between reporter levels with and without recursive RNAi of Dicer ( as described in Materials and Methods ) . As shown in Figure 4 , this figure of merit also predicts that the maximum restoration will occur for low values of γ . Thus , by either criterion , the success of recursive RNAi hinges on avoiding the use of highly efficient systems . This confirms the intuition that recursive RNAi can in fact be self-defeating . The analysis presented thus far treats only the steady-state behavior of the system . In many cases , however , experiments might be conducted before the system has achieved its final steady-state . Would the general conclusion presented above , namely that restoration decreases as RNAi efficacy increases , still hold in a transient condition ? Would restoration seen at a transient time-point be greater than that seen at steady state , or less ? To answer these questions numerical integration was used to simulate the transient response of the recursive RNAi system following induction of RNAi . Figure 5 illustrates the results of this analysis . First , as illustrated in Figure 5A , the restoration of reporter protein level is a monotonically increasing function of time , so that the restoration achievable at a transient time-point will always be less than that achievable at steady state . This plot shows that there are no unexpected transient dynamics or overshoots , and that rather the system smoothly approaches its steady state . Second , one can note in Figure 5A that the system always reaches its steady-state plateau at roughly the same time , with only a small variation in the time taken to plateau with respect to variation in gamma . This is confirmed in Figure 5B which shows that the time taken to reach a fixed percentage of final restoration depends only weakly on gamma . Indeed , the time to reach 50% or 90% of final restoration varies by less than two-fold when the RNAi efficacy parameter gamma varies by two orders of magnitude . Third , it can be seen in Figure 5A that at all time-points , systems with greater RNAi efficacy ( γ ) have lower restoration . This is confirmed in Figure 5C , which plots restoration versus gamma at a specific transient time-point defined as the time at which GFP would be knocked down to 50% of its steady-state knockdown level following induction of RNAi . At this transient time-point , the restoration clearly decreases as gamma increases , mirroring the results plotted in Figure 4 for the steady-state behavior . These results indicate that the general conclusions reached about the detectability and effectiveness of recursive RNAi obtained by analytic determination of the steady-state solution also apply to the transient case . The model described thus far assumes that the target gene ( GFP , for instance ) is targeted with the same efficiency as the RNAi component gene . It is well known that the efficacy of target degradation caused by a particular siRNA depends significantly on the precise sequence used for targeting [22]–[28] . The effects of unequal targeting of a reporter versus Dicer are derived in Materials and Methods and plotted in Figure 6 . The figure shows that as the relative targeting of the reporter is decreased compared to Dicer , the level of restoration can be increased significantly , as indicated by the difference in GFP expression levels with and without recursive RNAi . Figure 6 also shows that the effect becomes more pronounced as γ is increased . In particular , Figure 6 shows that for very efficient RNAi systems ( high γ ) , a more switch-like behavior could be obtained by recursive RNAi provided the targeting of the reporter gene is deliberately made inefficient . This is a prediction that could be tested experimentally by designing a series of dsRNA constructs targeting GFP chosen to span a range of targeting efficiencies , and then measuring the restoration achievable . Figure 6C shows that while restoration can be improved with targeting of the reporter is less efficient , when targeting of the reporter is made more efficient than targeting of the RNAi machinery restoration becomes progressively less efficient . It is thus clearly desirable to tune the relative targeting efficiencies of the two constructs using existing algorithms [22]–[28] in order to decrease the efficacy of reporter targeting relative to the RNAi component that is targeted in recursive RNAi experiments . A standard reason for employing feedback in electronic circuits is to reduce the sensitivity of the system performance to variations in the operating parameters of components . This is classically seen in operational amplifier circuits which , when connected in a negative feedback mode , produce an amplifier whose gain is almost completely insensitive to variations in the gain of the operational amplifier itself . Gene expression is an inherently noisy process [29] , leading to random variation in protein levels for any given gene product . Variation in levels of knockdown has been measured in RNAi experiments and is a significant problem for detectability in genome-wide screens [30] , [31] . Might recursive RNAi , by adding a feedback control to the RNAi system , make the system less sensitive to fluctuations in protein levels ? In order to investigate whether recursive RNAi might help make the operation of the RNAi system more tolerant to variations in its own components , the sensitivity of Dicer protein levels to variation in the rate of Dicer protein translation was analyzed . Translation of message into protein is often considered a major source of biological noise . Variation in Dicer was chosen for purely hypothetical reasons , there does not appear to be any published data on cell-to-cell variability in protein levels for RNAi components . Sensitivity is defined in this case as the change in Dicer protein level at steady-state caused by a given change in the translation rate of Dicer protein . As derived in Materials and Methods , the ratio of sensitivity in the recursive configuration to that in the open-loop ( i . e . , non-recursive ) configuration is a function of γ , given by the following equation: ( 4 ) This equation shows that any change to any parameter of the system that would increase γ will have the effect of making the system less sensitive to variation in the translation rate of Dicer . The same equation can easily be shown to hold for sensitivity to variation in the transcriptional rate of Dicer message . Feedback thus makes RNAi more robust to parameter variation , and the greater the efficacy of RNAi , the greater the improvement in robustness . This may explain why , in some cases , the Dicer gene appears to be under negative feedback control by the miRNA pathway [7] . In some systems , induction of RNAi leads to production of secondary siRNA using the targeted mRNA as a template for an RNA-directed RNA polymerase ( RdRP ) [32]–[34] . How would this amplification affect the behavior during a recursive RNAi experiment ? Figure 7 shows numerical simulation results plotting restoration of a reporter gene for different values of the efficacy of amplification ( as described by the parameter theta ) simulated at two different values of the RNAi knockdown efficiency parameter gamma . It is clear that increased amplification leads to reduced restoration . This is in keeping with the general conceptual idea that more efficient RNAi , which can be achieved either by higher knockdown efficacy or by increased amplification , leads to decreased restoration in recursive RNAi experiments . Comparing the two panels , it is clear that for any given value of the amplification parameter , lower gamma always leads to better restoration . Thus , the addition of the amplification pathway to the model has no effect on the overall qualitative conclusion that increased efficacy of RNAi leads to decreased restoration . The analysis presented thus far assumes that if a given RNAi pathway component was knocked down completely , it would result in complete loss of RNAi activity . This effect underlies the potentially self-defeating nature of recursive RNAi . However , only a few proteins of the RNAi pathway appear to be essential core components , with the rest making significant , but not essential , contributions to the process [35] . Even complete knockdown of the non-core components would thus allow some level of RNAi to continue . Would recursive RNAi of such non-core components produce restoration to a different degree than targeting a core component ? This question was addressed by modifying the model equations to add a new parameter rho that represents the degree of requirement of a given component for the process of RNAi . A value of ρ = 1 indicates the component is a core component essential for RNAi , while ρ = 0 indicates a component that is not involved in RNAi at all . Low values of rho would also apply for components encoded by multiple redundant gene copies . The expression of a reporter gene in the presence of recursive RNAi is plotted in Figure 8 ( based on equations derived in Materials and Methods ) as a function of the level of requirement ρ . The result is that recursive targeting of a non-essential component ( ρ<1 ) leads to less restoration than recursive targeting of an essential core component . This implies that variation in the degree of requirement of a given protein for RNAi could be an important source of variation in the level of restoration achievable by recursive RNAi inhibition of different components of the pathway . There are many ways to introduce dsRNA into cells to activate RNAi . In some cases , the dsRNA is added by soaking or feeding , in others it is expressed by stably integrated constructs . In other cases , however , the dsRNA is expressed as a short hairpin construct contained on a plasmid that is transiently transfected into cells . In this case , the rate of dsRNA production will not be uniform over time because the concentration of plasmid will decrease with first order decay kinetics as the plasmid becomes diluted during cell division . This situation was modeled as described in Materials and Methods , with results plotted in Figure 9 . The results show that introduction of a decay process for the dsRNA source leads to a transient knockdown that eventually returns to baseline expression of the reporter . For slow rates of decay , significant restoration can still be seen with recursive RNAi , but for very fast decay , the restoration becomes negligible . Transient transfection does not , however alter the basic conclusion that increased RNAi efficacy γ leads to decreased restoration . As plotted in Figure 9B , for all rates of decay that were modeled , after increasing γ past an optimum restoration value in the range 1–5 , further increasing γ decreases restoration . Thus the basic conclusion that increased RNAi efficacy leads to decreased effectiveness of recursive RNAi is predicted to still hold in transient transfection experiments , although the results also indicate that if the transfection is too transient , restoration might not be detectable in any case .
This report uses a mathematical model to predict the steady-state levels of reporter gene expression in recursive RNAi experiments . This model indicates that recursive RNAi is indeed possible , but that the level of restoration of a reporter gene , and therefore the ability to observe the effect of restoration , depends on the intrinsic efficacy of RNAi knockdown . Systems with more complete RNAi mediated knockdown are predicted to be less susceptible to RNAi . For screens in which the goal is simply to determine whether or not restoration has occurred in order to identify new RNAi components , the level of restoration only needs to be large enough relative to the measurement noise so that a reliable detection can be made . A much more stringent application is when recursive RNAi is used to restore expression of a gene previously inactivated by RNAi , as has been demonstrated in C . elegans [15] . For such switch-off/switch-on applications of recursive RNAi , the level of restoration needs to be sufficiently high to restore essentially wild-type levels of gene function . Restoration of the targeted gene to fully wild-type levels would correspond to a restoration ratio R = 1 , which according to Figure 4 is impossible to attain . In many cases , for example genes that are not haplo-insufficient , it may not be necessary to restore gene expression levels all the way to wild-type to rescue the phenotype . However , the results of the model suggest that in many cases , even a more moderate restoration , say to one half or one quarter normal expression levels , would also not be possible if the efficacy of RNAi-mediated knockdown in the organism is too high . One could , in such cases , conduct the experiment in a mutant background with a partial defect in one or more components of the RNAi machinery , so that the value of γ is reduced enough to allow a high level of restoration . Of course , this would entail a design tradeoff because decreased γ would lead to less repression during the switch-off phase of the experiment . In practice , the value of γ might need to be tuned quite carefully to achieve desired results . Moreover , genetic manipulation of the RNAi machinery may lead to undesirable side-effects due to alteration of endogenous small RNA mediated regulatory pathways . A preferable strategy , therefore , may be to carefully tune the relative targeting efficiency [22]–[28] of the reporter versus the RNAi component , so as to reduce the efficacy of targeting of the reporter , which as shown in Figure 6 can produce improved restoration . It is also worth pointing out that inducible systems for turning on and off production of siRNA have been demonstrated [36]–[38] . Recursive-RNAi based switch-off/switch-on has only been documented in nematodes where RNAi constructs can be easily introduced by soaking or feeding , and may be much harder in other types of animals , representing a distinct advantage for inducible systems . Overall , it remains to be seen whether switch-on experiments using recursive RNAi would have any advantages over these chemically inducible approaches . Switch-on by RNAi of RISC components might yield faster dynamics as it would not be limited by the degradation or dilution rate of the siRNA molecules . In comparing the predictions of the model to experimentally measured levels of reporter gene restoration ( see above ) , it was found that published values for the degree of restoration seen when Argonaute-2 is targeted are much higher than predicted by the model . This does not represent a discrepancy between the model and the data so much as a discrepancy between the experimentally observed behavior of Argonaute-2 and other RNAi components . Indeed , dramatically higher levels of reporter restoration have consistently been reported for Argonaute-2 compared to other RNAi components including Dcr-1 , Dcr-2 , R2D2 , Tudor-SN , FMRp , Drosha , Aubergine , and Piwi [11] , [39] , [40] . The fact that this protein seems to consistently show a distinctly different behavior in recursive RNAi experiments compared to all other known RNAi components [11] suggests that Argonaute-2 acts somehow differently from the RNAi components described within the model . Perhaps Argonaute-2 might be involved within additional control loops not included in the present model . Consistent with the notion that Ago-2 is somehow unique in its functions and interactions , it has recently been reported that Ago-2 depletion has a distinct and specific effect on RNAi competition that is not seen when other RNAi components are targeted [40] . These considerations suggest that the model used here , in its present form , must not fully represent the range of behavior of Argonaute-2 . The results of Equation 4 indicate that by some measures , the RNAi system may operate more reliably when operated in a closed-loop recursive mode . This result , together with the main result that the susceptibility of the RNAi machinery is to inhibition by RNAi , indicates that the RNAi pathway can demonstrate interesting properties when operated in a closed-loop “recursive” mode , even when represented by a fairly simple model . The favorable comparison with published levels of restoration versus efficacy suggests that the model may have predictive value . Other models of the RNAi pathway have previously been developed which model the system at varying levels of complexity [41]–[44] , and it would be interesting to see whether these different models give similar predictions when adapted to represent recursive RNAi experiments . It is also feasible to extend the approach described here to an analysis of the dynamic properties of other types of small RNA mediated control systems such as micro-RNA networks .
The RNAi pathway is represented using a model that is somewhat less complex than previous detailed but non-recursive RNAi models [41]–[44] but which encapsulates the main features of the system . The scheme of the model is given in Figure 1 and the parameters are defined in Table 1 . Both Dicer and RISC complexes are represented as single proteins even though in reality both are highly elaborate protein complexes . This representation , employed in most other RNAi models [41]–[44] is justified on the grounds that a typical recursive RNAi experiment would only target a single gene and its corresponding protein , and would not affect other proteins in the complex . Consequently , the protein levels of the other proteins can be simply treated by lumping their effect in with the other constants in the equations . In the following development only proteins specific to one complex or the other will be treated . In reality , some proteins are shared between the two but this analysis will not consider attempts to silence such shared factors by RNAi . The model will also not address the issue of partial redundancy , in which some RNAi machinery components may be present in multiple gene family members , such that complete inactivation of one member would only result in partial loss of RNAi function . Analysis of switching between different Dicer or Ago family members induced by recursive RNAi would be an interesting area for future study . To model transcription , it is assumed production of new mRNA at a constant rate rt which is approximately the same for all genes in the model . The model assumes that messenger RNA is degraded through a first-order decay with rate constant rdm . Translation of mRNA into protein is modeled assuming that protein is synthesized at a rate proportional to the concentration of message , with a rate constant rx , and is degraded with a first order decay rate constant rdp . Since the rates of mRNA production and degradation are significantly faster than the corresponding rates for proteins ( [45]–[47] and references cited in [42] ) , a quasi-steady state assumption may be invoked such that mRNA concentrations are set to their presumed steady-state value based on the rates of synthesis and degradation , ignoring the transient behavior while approaching this value . Production of siRNA by Dicer is represented by assuming that the siRNA is produced at a rate proportional to the concentration of Dicer , with an effective rate constant kcatD . The concentration of dsRNA is not explicitly represented , rather it is assumed to be lumped into kcatD , and it is taken as a constant thus assuming that dsRNA will not be degraded over time . The latter assumption is most appropriate for systems in which the dsRNA is expressed constitutively within the cell as a small hairpin construct . It is further assumed that siRNA is degraded by a first order decay with rate constant rds . In the simplest form of this model , to be described first , the production of additional dsRNA from targeted message by RNA-dependent RNA polymerase [32]–[34] , [42] is not modeled , but the effect of such an enzyme will be considered later in this report . It is assumed that an siRNA molecule is loaded onto a RISC complex according to a simple first-order binding process with an affinity described by the dissociation constant KDR . This assumption implies that the RISC complex is not saturated by siRNA during the modeled experiments . This assumption may not always hold true . It has been shown that when multiple siRNA species are added to a cell or in vitro RNAi system , they can compete with each other [48]–[50] , and this is thought to reflect a limited quantity of Ago2 that becomes saturated when too many siRNAs are present [40] . Whether or not RISC/Ago2 becomes saturated will depend on how much siRNA is used , for instance in one vitro study it was found that 100–200 fold more siRNA than normally used was required to show significant competition , suggesting that in the normal experimental regime employed by those workers , RISC was not saturated [48] . In the present model , saturation of RISC binding would imply an excess of siRNA thus rendering the system less sensitive to recursive RNAi targeting of Dicer . To model degradation of target messages by the RISC complex , it is assumed that a message targeted by an siRNA will be degraded by RISC at a rate equal to the product of the concentration of siRNA-loaded RISC and the concentration of the target message , with a rate constant kcatR . The linear dependence of RISC complex formation and activity on siRNA and RISC concentrations , including the assumptions of first order binding and lack of saturation , are in agreement with the prior modeling studies of RNAi [41]–[43] . The following analysis of the model will only keep track of proteins whose level will change during the course of an experiment . Proteins that are not affected by the addition of the dsRNAs , will be assumed to have attained their steady state value long before the beginning ( τ = 0 ) of the experiment . They will , therefore , be treated as constants of the model , just as the levels of basic transcriptional and translational machinery are assumed constant in the model . While the model explicitly treats only one protein component of the Dicer or RISC complexes at a time in the analysis , since in a typical recursive RNAi experiment only one protein would be targeted , in fact the model does not in any way place any limits on the number of proteins that may be present in the two complexes . However , the influence of the other proteins is subsumed within the other parameters of the model , and is taken as constant under the assumption that the other proteins in the Dicer and RISC complexes , apart from whichever protein might be targeted by recursive RNAi , do not vary in their expression levels . The following discussion will refer to the reporter gene as GFP , but would describe any target gene such as luciferase . The behaviors of components of the RNAi machinery , plus a reporter construct , can be represented as follows in three distinct cases: In order to simplify the equations representing the model , time , protein concentration , and siRNA concentration are rescaled as follows , representing the rescaled concentrations with capital letters:To simplify the resulting expressions , the following lumped parameters are defined as combinations of the detailed parameters of the model summarized in Table 1: First consider the relative susceptibility of Dicer and RISC proteins to downregulation by RNAi compared with a generic reporter protein that is not a component of the RNAi machinery . In other words , is recursive RNAi more or less effective compared with open-loop RNAi ? The relative susceptibilities SD and SR of Dicer and RISC , respectively , relative to the reporter gene , are defined as:it is obvious by inspection that the relative susceptibility of the two components of the RNAi machinery will decrease relative to the reporter gene as the efficacy of RNAi increases ( as judged by the parameter gamma ) . So as RNAi efficacy increases , RNAi genes become increasingly resistant to RNAi . In a typical recursive RNAi experiment , usually only the reporter protein level is measured , rather than the level of Dicer or RISC proteins . A candidate gene is scored in screens as being involved in RNAi if dsRNA directed against the gene results in a restoration of reporter gene activity back to control levels . In other words , if one monitors the reporter protein level , when it is targeted by RNAi the level will drop , and if a component of the RNAi machinery is also targeted , the level of the reporter will rise back up towards its level seen when no RNAi is performed . One way to quantify this restoration effect is to measure the ratio of recovery after recursive RNAi knockdown to the level of knockdown relative to control . This is expressed by the relative restoration ratios RD and RR which can be defined for the two cases RNAi of Dicer and RNAi of RISC , respectively , as follows: For a switch-off/switch-on experiment using Dicer , for example , one would want GTD≈G0 , which in turn would require that RD≈1 . In fact , RD is maximal when γ = 2 , and its maximum value is only 0 . 25 . It is thus not possible to restore gene expression back to fully normal levels , but only at most one quarter of the way back to normal levels from the level of maximum knockdown prior to “switch on” . As an alternative to these ratios , one may be more interested in the absolute difference in expression levels in the two conditions of knockdown versus knockdown in the presence of recursive RNAi . This difference ultimately determines the detectability of gene restoration when compared with the standard deviation of measurement of expression levels in the two states . The increase in expression levels , in units normalized to the control expression level of the reporter gene , is given by: Suppose that due to difference in targeting sequences , siRNA inhibition of GFP ( or whatever gene of interest is being knocked down ) is either more or less efficient than siRNA inhibition of Dicer in a recursive RNAi experiment . This effect can be represented in the model above as a difference in catalytic efficiency of siRNA-loaded RISC . This can be represented by a parameter ε such that if kcatR is the catalytic rate constant of RISC when acting on Dicer , the catalytic rate constant of RISC when acting on GFP would be ε*kcatR . In this case the only change to the systems described above will be to the differential equations representing the rate of change of GFP level , as follows:using this modified equation to solve for the steady-state GFP level yields:for the non-recursive case , andThese two expressions were used to plot the predicted expression during recursive RNAi with differential targeting in Figure 6 . The relative restoration ratio for the GFP target before and after recursive RNAi of Dicer is then given , as a function of the relative targeting efficiency of GFP , by the equation: | RNA interference is a gene regulatory system in which small RNA molecules turn off genes that have similar sequences to the small RNAs . This has become a powerful tool because a researcher can use RNA interference to turn off any gene of interest in order to test its function . There is great interest in identifying the genes required for the RNA interference pathway , and one approach to identifying such genes has been to use RNA interference to turn off potential RNA interference genes and to ask whether RNA interference still functions when these genes are turned off . The goal of our report is to ask how it is possible for RNA interference to turn itself off , using a mathematical model of the system . The results show that RNA interference cannot turn itself off if the RNA interference pathway is too effective to start with , so that experiments in which RNA interference acts on itself will only work in systems having a low efficiency . The results of our model suggest possible ways to improve the self-inactivation of RNA interference . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"mathematics",
"molecular",
"biology",
"computational",
"biology/systems",
"biology"
]
| 2008 | Modeling Recursive RNA Interference |
Tick-borne encephalitis virus ( TBEV ) is transferred to humans by ticks . The virus causes tick-borne encephalitis ( TBE ) with symptoms such as meningitis and meningoencephalitis . About one third of the patients suffer from long-lasting sequelae after clearance of the infection . Studies of the immune response during TBEV-infection are essential to the understanding of host responses to TBEV-infection and for the development of therapeutics . Here , we studied in detail the primary CD8 T cell response to TBEV in patients with acute TBE . Peripheral blood CD8 T cells mounted a considerable response to TBEV-infection as assessed by Ki67 and CD38 co-expression . These activated cells showed a CD45RA-CCR7-CD127- phenotype at day 7 after hospitalization , phenotypically defining them as effector cells . An immunodominant HLA-A2-restricted TBEV epitope was identified and utilized to study the characteristics and temporal dynamics of the antigen-specific response . The functional profile of TBEV-specific CD8 T cells was dominated by variants of mono-functional cells as the effector response matured . Antigen-specific CD8 T cells predominantly displayed a distinct Eomes+Ki67+T-bet+ effector phenotype at the peak of the response , which transitioned to an Eomes-Ki67-T-bet+ phenotype as the infection resolved and memory was established . These transcription factors thus characterize and discriminate stages of the antigen-specific T cell response during acute TBEV-infection . Altogether , CD8 T cells responded strongly to acute TBEV infection and passed through an effector phase , prior to gradual differentiation into memory cells with distinct transcription factor expression-patterns throughout the different phases .
Tick-borne encephalitis virus ( TBEV ) is a single-stranded flavivirus and the causative agent of tick-borne encephalitis ( TBE ) . TBEV is transferred to humans from infected Ixodes ticks . TBE is an increasing public health problem occuring throughout northern and central Europe and Asia , with thousands of encephalitis cases reported annually despite available TBE vaccines [1 , 2] . Epidemiological studies suggest that around 25% of all infected individuals develop clinical disease [3 , 4] . TBE has a characteristic biphasic course with influenza-like illness followed by a second neuroinvasive phase with neurological symptoms of variable severity , ranging from meningitis to severe meningoencephalitis . About a third of the patients eventually suffer from long-term sequelae including neuropsychiatric problems , headaches , and a substantial decrease in quality of life ( reviewed in [5] ) . The mechanisms behind TBE-pathogenesis are largely unknown . Direct infection of neurons has been suggested as the cause of neurological disease , and TBEV is present in brain tissue in most of the fatal cases [6] . More severe disease has been associated with low levels of neutralizing antibodies to TBEV , as well as a low early cerebrospinal fluid ( CSF ) IgM response [7] . The few studies that have addressed T cell responses in TBE have suggested that immunopathological effects caused by T cells may influence disease outcome , based on data showing CD8 T cell infiltration in brain tissue in fatal cases [8] . Murine models support this notion , demonstrating a prolonged survival of CD8-deficient and SCID mice , as compared to immunocompetent mice , following experimental viral infection [9] . In parallel , clonal T cell infiltration has been observed in the brains of mice dying from TBE [10] . Most current knowledge of antigen-specific CD8 T cell responses to acute primary viral infections still comes from murine models , where responses to viruses such as lymphocytic choriomeningitis virus ( LCMV ) or vaccinia virus have been studied [11 , 12] . In such models , activated T cells undergo a phase of rapid proliferation with an expansion of Ag-specific CD8 T cell clones . During the peak response , a majority of all CD8 T cells may be specific for the infecting virus . The response then contracts , forming a smaller memory population following clearance of the virus [11 , 13] . Human antiviral CD8 T cell responses have also been extensively analyzed in chronic infections , such as in infections with human immunodeficiency virus ( HIV ) , cytomegalovirus ( CMV ) and Epstein-Barr virus ( EBV ) [14–17] . Based on the results from such studies , distinct stages of CD8 T cell differentiation have been defined by the expression of specific surface markers , such as the isoforms of CD45 and of the expression of homing receptor CCR7 , defining CD45RA+CCR7+ as naïve , CD45RA−CCR7+ as central memory ( Tcm ) , CD45RA−CCR7− as effector memory ( Tem ) , and CD45RA+CCR7− as effector memory RA ( TemRA ) CD8 T cells [18 , 19] . Recent studies , adopting the live attenuated yellow fever virus ( YFV ) vaccine as a model to study acute viral infection in humans , have indicated that CD8 T cells against HLA-A2- and HLA-B7-restricted epitopes display a CD45RA−CCR7− phenotype during the peak effector response , which transits into a CD45RA+CCR7− phenotype at the memory stage [20 , 21] . Studies from the YFV and smallpox vaccine model systems have suggested that the magnitude of the total effector CD8 T cell response can be quantified with a set of four phenotypic markers , with the transient Ki67+Bcl-2lowHLA-DR+CD38+ phenotype defining the effector CD8 T cells during acute viral infection [22] . In recent years , the T-box transcription factors T-bet and Eomesodermin ( Eomes ) have been shown to play important roles in determining the fate of murine CD8 T cells during infection [23–25] . Their cooperative expression in chronic infection has been shown to be critical to sustain viral control since deletion of either one of them resulted in failure to control infection [26] . The expression patterns of T-bet and Eomes in CD8 T cells is not yet completely understood , and the analysis of these transcription factors during CD8 T cell differentiation may bring a novel molecular perspective to the phenotypic characterization of CD8 T cells and a deeper understanding of CD8 T cell differentiation during both acute and chronic viral infections . In this study , we have investigated the human CD8 T cell response to TBEV in a cohort of infected patients . The activated CD8 T cells were characterized with respect to perforin , granzyme B , HLA-DR , PD-1 , T-bet and Eomes and Bcl-2 expression . An immunodominant HLA-A2-restricted TBEV epitope was identified , which allowed for a detailed analysis of the antigen-specific CD8 T cell response . Studying these parameters with the tools available , the present study characterizes the kinetics and characteristics of the CD8 T cell response to TBEV . The implications of the findings are discussed in the context of host response to TBEV , possible role in immunopathogenesis , therapy and vaccination .
Peripheral blood mononuclear cells ( PBMCs ) , cerebrospinal fluid ( CSF ) , sera , and whole blood were collected from 20 patients with confirmed TBE ( IgM positive for TBEV in serum according to standard clinical diagnostic criteria ) ( S1 Table ) . Blood samples were obtained within 3 days after hospitalization ( day 0 ) and after 7 , 21 , and 90 days , during an interval ranging from the acute neuroinvasive phase to the convalescent phase of disease . Patients were evaluated at all sampling time points for symptoms such as fever , headaches , dizziness , impaired consciousness , nuchial rigidity , tremor , nystagmus , ataxia , emotional lability , difficulty in concentration , impaired memory , dysphasia , dysaesthesia , sleeping disturbances , irritability toward light and/or sound , generalized seizures , alterations of reflexes , impaired hearing , and disturbances of the cranial or spinal nerves . Patients were classified as having mild ( meningitis ) ( n = 16 ) or moderate ( meningoencephalitis ) ( n = 4 ) forms of TBE according to previous definitions [27] . Negative control samples were collected from 20 age-matched , healthy control subjects who were not previously vaccinated against TBE or had no symptoms of clinical TBE infection . Median time since symptom debut in the secondary neuroinvasive phase was 4 days ( range 1–15 ) , whereas the median time since symptom debut in the first phase were 15 days ( range 5–24 ) , at the time of inclusion . All patients had seroconverted at the time of inclusion . While all the patients were positive for TBEV IgM at day 0 and 7 after hospitalization ( S1 Table ) , all but one patient were positive for IgG at day 0 , and all donors had increasing levels of IgG over time ( S1 Fig . ) . TBEV RNA was not detected in plasma or CSF from any individual . To quantify the magnitude and expansion of CD8 and CD4 T cell responses in TBEV infected patients , we combined staining for CD38 , which is expressed by activated T cells during acute and chronic infections [28] , with staining for the intracellular marker Ki67 found exclusively in cycling or recently divided T cells [29] . CD4 and CD8 T cell gating strategy is found in S2 Fig . The patients’ levels of Ki67 and CD38 co-expressing activated CD8 T cells was 10-fold greater at 7 days than at 90 days after hospitalization and in comparison to levels in the healthy controls ( Fig . 1A and 1B ) . To evaluate general non-specific activation of CD8 T cells , known as bystander activation , in the course of infection we used an HLA-A2 tetramer refolded with a CMV-pp65 epitope peptide to identify CMV-specific cells . Activation was evaluated by the expression of Ki67 , which was low in the CMV-pp65 tetramer-defined population and remained so throughout the course of infection in all tested donors ( Fig . 1C and 1D ) . This finding suggests that bystander activation of CMV-specific CD8 T cells is not a major feature during acute TBEV infection . To further characterize the CD8 T cell effector response defined by Ki67 and CD38 expression at the peak of expansion at day 7 after hospitalization , staining for CD45RA , PD-1 , Bcl-2 , CD127 , CD27 and HLA-DR , together with perforin and granzyme B was performed ( Fig . 1E ) . Expression of HLA-DR , PD-1 , perforin and granzyme B was increased in activated CD8 T cells along with decreased expression of CD127 , Bcl-2 and CD27 ( Fig . 1F ) . We also measured the expression of CD127 ( IL-7Ra ) , which has been shown to be downregulated on effector cells and then to be re-expressed on precursors of the memory pool [30] , together with the homing marker CCR7 and CD45RA . Approximately 60% of the effector population consisted of a CD45RA−CCR7−CD127− phenotype ( Fig . 1G ) . These data showed that TBEV infection induced a robust CD8 T cell response that contracted to background levels over a period of 90 days post-hospitalization . As CD4 T cells are an integral component in effective antiviral responses , we extended our study to CD4 T cell subsets . Elevated levels of CD38 and Ki67 co-expressing CD4 T cells were detected in TBEV-infected patients at day 7 after hospitalization , as compared to healthy controls ( Fig . 1H and 1I ) . This CD38 and Ki67 co-expressing CD4 T cell population exhibited increased expression of HLA-DR , PD-1 and perforin together with low expression of Bcl-2 , suggesting that activated CD4 T cells have effector properties at day 7 after hospitalization ( S3A Fig . ) . The expression of CCR7 , CD45RA and CD127 in the CD4 population was more variable in comparison to the corresponding population of CD8 T cells ( S3B Fig . ) . To further characterize the phenotype of the activated CD8 T cells , we assessed expression of transcription factors including T-bet , Eomes and the Ikaros family transcription factor Helios . The expression of T-bet and Eomes have been suggested to impact exhaustion and terminal differentiation of CD8 T cells [31] , whereas Helios have been suggested as a marker of activation and proliferation in T cells [32] . The activated CD8 T cell population studied showed significantly increased expression of T-bet and Eomes , whereas Helios expression was lower , as compared to non-activated CD8 T cells in the same sample and to cells from healthy controls ( Fig . 2A and 2B ) . The activated cells remained Helios negative throughout the course of infection and CD8 T cells expressing Helios expressed low levels of Ki67 and PD-1 throughout the infection ( S4 Fig . ) . Cells with simultaneous expression of Eomes and T-bet constituted a dominant population of the activated CD8 T cells at day 7 after hospitalization ( Fig . 2C ) . These data indicate that effector CD8 T cells at day 7 after hospitalization have a distinct Eomes+T-bet+ profile . CD8 T cells have a spectrum of functions to control viral infections . Here , we assessed degranulation ( CD107a ) , cytokine expression ( IFN-γ and TNF ) and chemokine expression ( MIP-1β ) in TBEV-specific CD8 T cells over time in samples from five infected patients . We used a pool of potential TBEV peptide epitopes predicted by the NET-CTL algorithm ( Table 1 ) to stimulate PBMCs in vitro , and studied the kinetics and functional profile of the responding cells during the course of infection . CD8 T cell responses were very low or undetectable on the day of hospitalization . The frequency of CD8 T cells expressing IFN-γ and TNF in response to the peptide pool peaked at day 21 after hospitalization , comprising approximately 0 . 5% of the total CD8 T cell population ( Fig . 3A and 3B ) . CD107a together with MIP-1β peaked at day 90 with approximately 1 . 5% of the CD8 T cells ( Fig . 3B ) . The response pattern was primarily mono-functional ( >50% ) ; however , approximately 5–10% of the responding cells exhibited a two-functional profile , 15% exhibited a three-functional profile , and around 20% of the cells displayed a four-functional profile . This pattern was sustained over time in the infected subjects ( Fig . 3C ) . At the peak of the effector stage ( days 7 and 21 ) , CD107a mono-functional cells dominated the response , whereas MIP-1β-positive cells dominated the mono-functional response at day 90 after hospitalization ( Fig . 3D ) . A more diverse pattern could be observed in the bi-functional cells , with CD107a-expessing and MIP-1β-producing cells dominating at day 7 , IFN-γ and TNF producing cells dominating at day 21 , and CD107a expressing and TNF producing cells dominating at day 90 after hospitalization ( Fig . 3D ) . The triple functional cells produced mostly IFN-γ , MIP-1β and TNF at day 7 and 21 , while CD107a , MIP-1β and TNF producing cells dominated at day 90 after hospitalization ( Fig . 3D ) . These results indicate that the functional composition of TBEV specific CD8 T cells changes over time as they mature from an effector- to a memory-type response . To further characterize cells responding to the predicted TBEV epitopes , we assessed the expression of CD45RA , CD27 and the senescence marker CD57 on the surface of TBEV-specific CD8 T cells . CD8 T cells responding to the peptide-pool with CD107a , IFN-γ , TNF or MIP-1β expression consisted mainly of a CD27+CD45RA−CD57− phenotype ( Fig . 3E ) . The transcription factor expression pattern in CD8 T cells responding to the peptide pool was similar to that of the Ki67+CD38+ activated CD8 T cells ( Fig . 2C ) , with a dominant Eomes+T-bet+ profile comprising 50% of the responding cells at day 7 after hospitalization ( Fig . 3F ) . Responses to the TBEV peptide pool indicated that the pool contained at least one epitope targeted by the CD8 T cells during acute TBEV infection . Single peptide stimulations identified one HLA-A2-restricted peptide ( ILLDNITTL ) in the NS3 protein , which induced cytokine responses in all tested A2+ donors . The NS3 ILL-specific HLA-A2 tetramer identified detectable frequencies of cells in five donors with up to 1% positive cells at day 7 and 21 after hospitalization , whereas tetramer-positive cells were barely detectable at the day 0 time point ( Fig . 4A ) . Tetramer-positive CD8 T cells were further characterized for the expression of CD45RA , CCR7 , CD57 , granzyme B , perforin , PD-1 , and CD27 ( Fig . 4B and 4C ) . During the effector response at day 7 after hospitalization , the most prevalent phenotype of the TBEV-specific CD8 T cells was Tem , CD57− and PD-1+ , with this phenotype representing approximately 50% of the cells . This phenotype decreased and was less common at days 21 and 90 after hospitalization ( Fig . 4D ) . These data demonstrate how the TBEV-specific CD8 T cell response goes through a Tem , CD57−PD-1+ character that contracts with the establishment of T cell memory . CD8 T cell populations specific for human polyomavirus BK virus , influenza virus , CMV , and EBV display specific patterns of T-bet and Eomes expression in healthy blood donors [33] . Since the activated ( Ki67+CD38+ ) CD8 T cell population in TBE patients had increased expression of T-bet and Eomes along with low expression of Helios ( Fig . 2B ) , we measured the expression of these transcription factors in TBEV-specific CD8 T cells . NS3 ILL-specific CD8 T cells expressed no or very low levels of Helios , whereas T-bet was highly expressed at all time-points ( Fig . 5A and 5B ) . About 75% of the TBEV-specific CD8 T cells expressed Eomes at day 7 after hospitalization , declining over time to about 40% of the cells expressing Eomes at day 90 after hospitalization ( Fig . 5B ) . Ki67 expression in NS3 ILL-specific CD8 T cells was high at day 7 after hospitalization , and declined over time , to become almost undetectable at day 90 after hospitalization ( Fig . 5B ) . For total CD8 T cells , the expression levels of Helios was consistent around 10–15% at all time points ( Fig . 5C ) . The expression patterns of T-bet and Eomes also remained stable at all time points , whereas Ki67 expression in total CD8 T-cells was at its highest level at day 7 after hospitalization at around 10–15% , and declined by day 21 after hospitalization ( Fig . 5C ) . NS3 ILL-specific CD8 T cells thus have much less expression of Helios compared to the total CD8 T cell pool , along with a significant Eomes down regulation over time . The dominant phenotype of the TBEV-specific CD8 T cell population was Eomes+Ki67+T-bet+ at day 7 , as found in 50% of the epitope-specific cells ( Fig . 5D ) . This phenotype retracted to become almost undetectable by day 21 and 90 after hospitalization . Instead , an Eomes−Ki67−T-bet+ phenotype appeared by day 21 and 90 ( Fig . 5D ) . The dominant phenotype in the total CD8 T cell pool was Eomes+Ki67−T-bet+ and this did not change over time ( Fig . 5E ) . These results thus suggest that a set of three intracellular factors ( Ki67 , Eomes and T-bet ) characterize and discriminate stages of the antigen-specific T cell response during acute TBEV infection . We then also attempted to test the hypothesis that Ki67 , Eomes , and T-bet could identify the total CD8 T cell effector responses in acute TBEV-infection . To this end , we analyzed this population in total CD8 T cells over time in ten TBEV infected donors . Co-expression of Eomes , Ki67 and T-bet peaked at day 7 after hospitalization in the infected study subjects , contracting to healthy control levels at day 21 after hospitalization ( Fig . 5F ) . These levels were similar to Ki67 and CD38 positive CD8 T cells in Fig . 1B , and may represent an alternative way of describing activated CD8 T cells . Eomes has been shown to play a role in induction of cytolytic properties in murine CD8 T cells [34] . With our observation that Eomes expression declines from 75% of the ILL-specific cells at day 7 after hospitalization to a lower percentage at day 21 ( 50% ) and day 90 after hospitalization ( 35% ) ( Fig . 5B ) , we next aimed to further evaluate the role of Eomes and the global CD8 T cell cytotoxic potential at day 7 after hospitalization . We investigated the expression of granzyme B and perforin in Eomes+ and Eomes− CD8 T cells . Eomes+ CD8 T cells expressed higher levels of both granzyme B and perforin than did Eomes− cells ( Fig . 5G ) , suggesting that Eomes expression was associated with the expression of cytolytic effector proteins .
Understanding T cell responses during viral infections is necessary for the successful design of antiviral treatments and vaccines . We here pursued a detailed analysis of the temporal dynamics , specificity , as well as functional and phenotypical characteristics of the CD8 T cell response to acute human TBEV-infection . Peripheral blood CD8 T cells were activated ( determined by Ki67 and CD38 expression ) in response to infection and expressed perforin , granzyme B , HLA-DR , PD-1 , T-bet and Eomes together with low levels of Bcl-2 at day 7 after hospitalization , phenotypically defining these as effector cells . An immunodominant HLA-A2-restricted TBEV epitope was identified , and the corresponding HLA-tetramer defined TBEV-specific effector cells that predominantly displayed an Eomes+Ki67+T-bet+ effector phenotype at the peak of the response . The TBEV-specific CD8 T cells transitioned to an Eomes−Ki67−T-bet+ population as the infection resolved and memory was established . In summary , CD8 T cells responded to the virus and passed through an effector phase during acute TBEV-infection , prior to a gradual differentiation into memory cells with a distinct expression-pattern of transcription factors . The present results indicate that virus-specific effector CD8 T cells during acute TBE can be defined by the expression pattern of Eomes , Ki67 , and T-bet within the global CD8 T cell compartment . T-bet and Eomes are important in murine terminal effector and memory T cell development [23 , 35] and cooperate by inducing the expression of IFNγ , granzyme B and perforin early in the activation process of murine CD8 T cells [23 , 36 , 37] . Although T-bet and Eomes cooperate in many respects , their expression is to some extent reciprocal . T-bet expression has been reported to be highest in early effector CD8 T cells in mice , but its expression progressively declines as memory cells form [38] . In contrast , the expression of Eomes is upregulated in early effector cells , and is sustained and increased during the effector to memory cell transition [35 , 36] . Their expression pattern in TBEV-infection suggests that simultaneous upregulation is required to generate proper effector responses to control the infection in the acute stage . In human memory T cell subsets , the expression pattern of T-bet and Eomes may differ depending on the antigen . For instance , polyomavirus BK-specific cells display a T-bet intermediate and Eomes low phenotype . In the same donors , CMV-specific cells were high in both T-bet and Eomes , whereas influenza-specific cells were T-bet-high and Eomes-low [33] . The latter pattern is consistent with the phenotype of TBEV NS3 ILL-specific cells 90 days after hospitalization for TBEV infection ( T-bethigh , Eomeslow ) . Thus , long-term human memory T cells specific for cleared infections such as influenza or TBEV may have a shared Eomes-low profile , which is distinct from T cells specific for persisting infections such as CMV . Results from murine models indicate that Helios plays an important role in T cell development [39 , 40] . Furthermore , Helios has been suggested as a marker of activation and proliferation in T cells , since Helios positive CD8 T cells are enriched for mature cells in humans and mice , and Helios become upregulated under in vitro stimulations of murine CD4 T effector cells [32] . In contrast , to date , very little is known about Helios in human CD8 T cells . In the present study , at the peak of TBEV infection , around 10–15% of the total CD8 T cells were Helios-positive . These cells were CD45RA+/- , CCR7− , Ki67low , CD57int/- , PD-1− , CD27+/- , CD45RAhi and expressed perforin and granzyme B ( S4 Fig . ) . With regards to previous publications , we initially speculated that the activated antigen-specific cells would express Helios . However , most Ki67 positive NS3 ILL-specific cells were negative for Helios . The expression pattern of Eomes , Ki67 and T-bet was used to study the response longitudinally in CD8 T cells during TBEV infection . Approximately 5% of the CD8 T cells expressed a T-bet+Eomes+Ki67+ profile at day 7 after hospitalization . Cells with this phenotype contracted to healthy control levels at day 21 after hospitalization . Interestingly , this transcription factor co-expression-pattern was also observed in HLA-A2 tetramer-positive TBEV-specific cells . Therefore , Ki67 and CD38 co-expression describes the effector response in TBEV infection and the combination of T-bet , Eomes and Ki67 delineates the majority of antigen-specific cells in the acute stage of disease . Eomes has been suggested to play an important role in inducing lytic function in murine CD8 T cells [34] and consistent with this , we found that Eomes+ CD8 T cells express higher levels of both granzyme B and perforin than Eomes- CD8 T cells . In the case of T-bet , it has been shown that human CD8 T cells expressing high levels of T-bet rapidly can upregulate perforin upon stimulation with peptides [41] . Similar to the activation pattern of CD8 T cells , the HLA-A2-restricted NS3 ILL epitope-specific CD8 T cell response was absent at the day of hospitalization but appeared one week later . This indicates that the primary T cell response to TBEV infection occurs at this time and , hence , likely not in the first phase of infection; at least not in those patients which have a biphasic course of disease . In a YFV vaccine-based infectious model , the peak of viremia occurred at day 7 after immunization , and the peak of the CD8 T cell response was observed at day 15 after immunization [22] . However , TBEV RNA is usually not detected in plasma samples from patients during the neuroinvasive phase [42] , but has been detected in sera and whole blood from patients during early stages of infection before the appearance of antibodies [43] , and in brain tissue from patients dying from the disease [44] . We did not detect TBEV RNA at any time in plasma or cerebrospinal fluid ( CSF ) samples in our patient cohort; however , all the patients had seroconverted at the time of inclusion , and had entered the meningoencephalitic phase . Plausible reasons for the absence of detectable viral TBEV RNA would be that the primary viremia already had occurred and that the viral burden driving the CD8 T cell response was below the limit of our assay’s detection at the time of sampling . Alternatively , the virus may be located in other cells and tissues that produce and release viral antigen that stimulate the CD8 T cell response . Undetectable levels of viral RNA in blood with simultaneous presence of viral RNA in urine has previously been described in patients with Dengue fever [45] , and after vaccination against YFV [46] . West Nile Virus ( WNV ) RNA was detected in the urine of patients with symptomatic WNV infection ( neuroinvasive disease and fever ) at a higher rate and load and for a longer time than in the plasma of these patients , whereas the detection rate of WNV RNA in urine was lower than in plasma in asymptomatic donors [47] . The mean number of days after the first symptom debut in our patient cohort was 15 days , indicating that the TBEV-specific T-cell response in peripheral blood appeared at about day 21 in the course of infection . However , it is not known what the CD8 T cell response looks like at the site of pathogenesis , i . e . , the central nervous system , and if the TBEV-specific CD8 T cells that we observe in the periphery are representative of the population of CD8 T cells that is able to cross the blood-brain barrier . TBE pathogenesis is still largely unknown . TBEV has been shown to be present in brain tissue in most of the fatal cases [6] , so direct infection of neurons may cause the neurological disease; however , in the majority of patients it is not possible to detect TBEV RNA in CSF [42] . There are also studies suggesting that the CD8 T cell response in CNS contributes to the pathogenesis in humans and mice [8 , 9] . In this study , we were not able to draw any conclusions on relationships between disease severities and the phenotype or magnitude of TBEV-specific CD8 T cells , or virus replication in the CNS . Future studies with larger number of patients may help delineating the mechanisms and steps of the disease . Activated CD8 T cells ( Ki67+CD38+ ) were characterized by effector properties , such as increased expression of perforin and granzyme B , together with a Tem , CD127- profile . This phenotype has previously been studied in human yellow fever- and smallpox vaccine-models of acute viral infection [22] . Activated CD8 T cells were also observed in natural acute Hantavirus infection in humans , where up to 50% of the CD8 T cells have an activated profile one to two weeks after the symptom debut [48] . No or minimal bystander activation of CMV-specific memory CD8 T cells was detected during acute TBEV infection . This is in line with results by Miller et al . , who showed that CMV- , EBV- and influenza virus-specific CD8 T cells did not contribute to the effector T cell response to YFV and smallpox vaccines [22] . Together , these findings support the notion that the majority of activated CD8 T cells in the present patient cohort were specific for TBEV antigens . With regard to the NS3 ILL-specific cells , the dominant phenotype observed at the peak of activation was Tem PD-1+CD27high , which decreased significantly to approximately 15% of the peak in the convalescent phase . TBEV-specific cells show a Tem , PD-1+CD57− phenotype which is similar to what has been shown at the peak of the response to the YFV vaccine [20 , 21]; however , after YFV vaccination , a CD45RA+CCR7− late-stage effector cell ( TemRA ) phenotype was observed at the memory stage . It has been reported that TemRA cells may be involved in protective immunity against HIV , since HIV-specific T cells with this phenotype were associated with control of viremia [49 , 50] . Therefore , memory may not be fully formed three months after infection with TBEV since such a population is detectable , but not dominant , in NS3 ILL-specific cells . Given that TBEV is cleared after the acute phase of infection , the TBEV-specific CD8 T cell population may eventually obtain a central memory-like phenotype [51] . The quality of a T cell response is probably important for the level of protection it provides to the host . CD8 T cells can provide a range of effector functions , which rarely are co-expressed in the same cell with the same kinetic pattern . The heterogeneity in expression of CD8 T cell effector functions has been described [52 , 53] , but is not well understood . Our results show that cells responding to a pool of pre-selected TBEV peptides predominantly displayed monofunctional characteristics . In summary , the present results describe the phenotype and function of the CD8 T cell responses in acute TBEV-infection . In addition , based on the transcription factor expression profile in the TBEV-specific cells , Eomes , Ki67 and T-bet identifies cytolytic virus-specific CD8 T cells in the peak effector stage of acute TBEV infection . TBE is an emerging disease and is a growing health challenge in endemic parts , and with no antiviral drugs available . Indeed , the only effective protection against TBEV is vaccination . However , over the last few years vaccine failures have been reported , and it is also believed that a number of vaccine failures may have been overlooked due to difficulties in diagnosis , partly due to unusual antibody-kinetics in this patient group [54] . Taken together , these data may prove to be helpful for the future design of new therapeutic and immunotherapeutic treatment regimens as well as new options for vaccines to TBEV infection .
All included patients and healthy individuals gave written informed consent to participate in the study and the Kaunas Regional Research Ethics Committee , Lithuania and the Regional Ethical Review Board in Stockholm , Sweden approved of the study . Blood samples and cerebrospinal fluid used were collected after written informed consent and with approval from the Kaunas Regional Research Ethics Committee , Lithuania and the Regional Ethical Review Board in Stockholm , Sweden . Peripheral blood was collected from twenty confirmed ( IgM positive for TBEV in serum according to standard clinical diagnostic criteria ) TBEV infected patients hospitalized at the Clinic of Infectious Diseases at Lithuanian University of Health Sciences Kaunas in Lithuania . PBMC were isolated from CPT tubes ( BD Biosciences , San Jose , CA ) and cryopreserved in 90% FCS and 10% DMSO for later analysis . Whole blood and plasma were collected and cryopreserved for later analysis . T cell responses to TBEV were assessed using multi-color flow cytometry , and the monoclonal antibodies ( mAbs ) used were; anti-CD107a FITC , anti-CD4 Pacific Blue , anti-CD8 PerCP , anti-HLA-DR PerCP , anti-Ki67 FITC , anti-Ki67 Alexa Fluor 700 , anti-Bcl2 PE , anti-CCR7 PE-Cy7 , anti-MIP-1β PE , anti-CD14 BD horizon V500 , anti-CD19 BD horizon V500 , anti-perforin FITC and anti-granzyme B APC , anti-granzyme B PE-CF594 all from BD Biosciences ( San Jose , CA ) . Anti-CD45RA APC-Cy7 , anti TNF pacific blue , anti-IFN-γ Brilliant Violet 570 , anti-CD27 Brilliant Violet 785 , anti-CD27 Brilliant Violet 421 , anti-Helios Pacific Blue , anti-T-bet Alexa Fluor 488 , anti-T-bet PE-Cy7 , anti-CCR7 Brilliant Violet 785 , anti-CD279 Brilliant Violet 711 , anti-CD27 biotin and anti-CD127 Brilliant Violet 570 were all from Biolegend ( San Diego , CA ) . Anti-CD38 Alexa Fluor 700 , anti-CD38 eFluor 650 , anti-CD127 Alexa Fluor 780 , anti-PD-1 ( CD279 ) PE , anti-Eomes eFluor 660 and IgM eFluor 650 were all from eBioscience ( San Diego , CA ) . Anti-CD4 Qdot 605 , anti-CD8 Qdot705 , anti-CD8 Qdot 605 , Streptavidin-Qdot 585 , anti-CD57 pure and Aqua Live/Dead were all from Invitrogen ( Carlsbad , CA ) . Anti-CD3 ECD , anti-CD3 PE-Cy5 , HLA-A2 CMV pp65 tetramer in PE and anti-CD56 ECD were from Beckman Coulter ( Brea , CA ) . HLA-A2 ILLDNITTL tetramer in PE was kindly provided by the NIH Tetramer core facility . For phenotypic analysis of cells , PBMCs were incubated for 30 minutes 4°C in the dark , with surface mAbs , followed by washing with PBS . For the CD107a staining , the CD107a antibody was present during the 6 hours stimulation , and then additional CD107a antibody was added together with the surface mAbs for 30 minutes incubation at 4°C in the dark . Cells were fixed and permeabilized with fix/perm ( eBioscience ) for 30 minutes at 4°C in the dark . Cells were then washed and stained with intracellular mAbs . Samples were acquired on a BD LSRFortessa instrument ( BD Biosciences ) and analyzed using FlowJo software version 9 . 4 ( Tree Star , Ashland , OR ) , and SPICE 5 . 3 software provided by Dr . M . Roederer ( National Institutes of Health , Bethesda , MD ) [55] . Candidate epitopes of HLA-A2 , -A3 or-B7 super-types were predicted using the NetCTL search engine ( version 1 . 2 ) [56] . The HLA class I epitope predictions were performed on polyprotein consisting of a TBE EK-328 strain available in the Flavitrack database at http://carnot . utmb . edu/flavitrack . 15 peptides were predicted , and the 5 top scoring peptides in each HLA super-type were selected for synthesis . Peptides were synthesized by standard 9-fluorenyl-methyloxycarbonyl ( FMOC ) chemistry , purified to 70% purity by reverse-phase high-performance liquid chromatography and validated by mass spectrometry ( JPT Peptide Technologies , Berlin , Germany ) . PBMCs were rested in RPMI 1640 medium containing 10% FCS , 2 mM L-glutamine , 1% penicillin and streptomycin ( Invitrogen ) overnight at 37°C . Cells were stimulated with 10 μg peptides for 6 hours in 96-well round bottom plates in the presence of Brefeldin A ( Sigma-Aldrich , St . Louis , MO ) , monensin ( BD Biosciences ) and purified anti-CD28/CD49d ( 1 μl/ml ) ( BD Bioscienses ) . Staining , flow cytometry and analyses were performed as described above . Patient and control genomic DNA was isolated from whole blood using DNeasy kit ( QIAGEN ) . HLA typing was performed using a multiplexed reverse sequence-specific oligonucleotide probe method ( LABType SSO; One Lambda ) , according to the manufacturer’s instructions . For detection of TBEV-specific CD8 T cells were identified in peripheral blood by staining PBMCs with HLA-A2 NS3 ILLDNITTL tetramer for 15 minutes at 4°C in the dark , prior to the addition of surface mAbs . Cells were fixed and permeabilized with fix/perm ( eBioscience ) for 30 minutes at 4°C in the dark before the addition of intracellular mAbs . Samples were acquired on a BD LSRFortessa instrument ( BD Biosciences ) Detection of TBEV RNA was assessed as previously described [57] with these minor changes: RT-PCR was performed with TaqMan Fast Virus 1-step mastermix ( Applied Biosystem , Life Technologies ) using the StepOne RT-PCR system ( Life Technologies ) according to the manufacturer’s instruction . To ensure adequate RNA extraction from the samples , human B-actin ( Applied Biosystems , Life Technologies ) was assayed in parallel as an endogenous control . All sera were analyzed by using the Siemens Enzygnost TBE IgG assay ( Siemens Healthcare Diagnostics , Erlangen , Germany ) , and the sera isolated from blood draws at the first two time points of each patient were also analyzed by Immunozym FSM IgM assay ( Progen Biotechnik GmbH , Heidelberg , Germany ) . These analyses were performed according to the manufacturer’s instructions by using a combination of the Freedom EvoClinical pipetting platform ( Tecan Group Ltd , Männerdorf , Switzerland ) and the BEP III system ( Siemens Healthcare Diagnostics , Erlangen , Germany ) . Analyses were performed using GraphPad Prism software 5 . 0 for MacOSX ( GraphPad Software , La Jolla , CA ) . Data were analyzed by non-parametric repeated measures ANOVA test or Mann-Whitney test P values < 0 . 05 were considered statistically significant . | Tick-borne encephalitis virus ( TBEV ) belongs to the flavivirus family and causes tick-borne encephalitis . This is a severe meningoencephalitic disease with no available treatment . Detailed studies of the immune response during human TBEV infection are essential to understand host responses to TBE and for the development of therapeutics . Herein , we studied the primary T cell-mediated immune response in patients diagnosed with TBEV infection . We show that CD8 T cells mount a vigorous TBEV-specific response within one week of hospitalization . Moreover , TBEV-specific CD8 T cells displayed a distinctive phenotypic and functional profile , paired with a distinct transcription factor expression-pattern during the peak of activation . In summary , this is the first comprehensive study of the CD8 T cell response during acute human TBEV infection , and provides a framework for understanding of CD8 T cell-mediated immunity in this emerging viral disease . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | Specificity and Dynamics of Effector and Memory CD8 T Cell Responses in Human Tick-Borne Encephalitis Virus Infection |
Alveolar echinococcosis is a zoonotic disease caused by the metacestode of Echinococcus multilocularis . Many species of small mammals , including arvicolid rodents or Ochotona spp . , are natural intermediate hosts of the cestode . The main aim of this study was to identify natural intermediate hosts of E . multilocularis in Chenaran County , Razavi Khorasan Province , northeastern Iran , where the prevalence of infected wild and domestic carnivores is high . A program of trapping was carried out in five villages in which this cestode was reported in carnivores . The livers of 85 small mammals were investigated for the presence of E . multilocularis infection using multiplex PCR of mitochondrial genes . Infections were identified in 30 specimens: 23 Microtus transcaspicus , three Ochotona rufescens , two Mus musculus , one Crocidura gmelini , and one Apodemus witherbyi . A range of small mammals therefore act as natural intermediate hosts for the transmission of E . multilocularis in Chenaran County , and the prevalence suggested that E . multilocularis infection is endemic in this region . The existence of the life cycle of this potentially lethal cestode in the vicinity of human habitats provides a significant risk of human infection .
Echinococcosis is a near cosmopolitan parasitic disease caused by the cestode Echinococcus [1] , [2] , [3] . The potentially fatal zoonotic disease , alveolar echinococcosis , is caused by the metacestode of E . multilocularis , which has a sylvatic cycle , comprising wild carnivores as definitive hosts and more than 40 species of small mammals , including arvicolid rodents and the lagomorph Ochotona spp . , as intermediate hosts [2] , [4] , [5] . Humans are accidental and aberrant intermediate hosts infected by parasite eggs ingested in contaminated food or by direct contact with infected definitive hosts [5] , [6] . Echinococcus multilocularis distribution is restricted to the northern hemisphere , including Central Europe , the Near East , Russia , Central Asian republics , northern Japan , parts of North America [7] , [8] , [9] , [10] , [11] , and some countries of the Middle East [12] , [13] . In Iran , information about E . multilocularis infection is limited to a few studies restricted to the northwestern areas of the country [14] , [15] , [16] , [17] . The first study in Iran , conducted in 1971 on the Moghan Plain , reported E . multilocularis infection in 10% of red foxes ( Vulpes vulpes ) [14] , [15] , although its metacestode was not found in any of the 5000 rodents examined ( unpublished data ) . In 1992 , a further study of 130 wild carnivores and 1500 rodents showed 22 . 9% of red foxes and 16% of jackals ( Canis aureus ) infected with adult E . multilocularis but no metacestodes in the rodents [12] , [17] . Investigation of definitive and intermediate hosts of E . multilocularis in other parts of the country has been neglected . Recently , following a few reported cases of human alveolar echinococcosis [18] ( E . Razmjou , unpublished data ) , a morphological and molecular survey was carried out on wild and domestic carnivores from the Chenaran area in northeastern Iran [19] . Based on this study , the high prevalence of carnivores infected with E . multilocularis indicates that the life cycle of E . multilocularis is being maintained here , and Razavi Khorasan Province was shown to be an endemic area [19] . Therefore , the role of dogs , foxes , jackals , wolves , and hyenas was confirmed as a definitive host . However , no data were available on the intermediate hosts of E . multilocularis in the Chenaran area . To determine the E . multilocularis life cycle in a specific region , study of its potential intermediate hosts is imperative , since voles have a small home-range and infected voles are a good marker for the presence of E . multilocularis eggs [20] , thus indicating the risk of human infection at a local level [21] . This investigation was carried out to identify the natural intermediate hosts and determine the prevalence of infection in Chenaran County to confirm the life cycle of this pathogenic cestode in this region .
For investigating the presence of E . multilocularis infection , small mammals were trapped under license from the Iran Environment Protection Organization . Animals were handled according to the American Society of Mammalogists ( ASM ) guidelines for animal research , and the experimental protocols were reviewed and approved by the Ethics Committee of Tehran University of Medical Sciences ( Approval No 759-2008 ) . The inhabitants of Chenaran County villages , on whose land the specimens were collected , gave their informed consent for the trapping . Razavi Khorasan Province is located in northeastern Iran in the vicinity of Turkmenistan and Afghanistan ( Figure 1 ) . Chenaran ( 36°38′N , 59°7′E ) , one of 19 counties in the province , is northwest of the capital , Mashhad , and had a population of approximately126 , 000 in 2011 [22] . It is a region of highlands , located between Binalood Heights and the Hezar Masjed Mountains at an elevation of 1400–1600 m . Average temperature in winter is 4 . 1°C; colder at higher elevations . Mean summer temperature is 23 . 9°C . Annual precipitation averages 212 . 6 mm with the lowest rainfall in summer . Most villages of Chenaran County are located in valleys with natural rivers as a source of water for fruit gardens and household use . The grasslands and high soil moisture in these areas provides suitable habitat for small mammals that attract predators , and likely good conditions for taeniid egg survival [23] . Trapping was conducted in October 2010 and July 2011 in five villages that reported high rates of E . multilocularis in carnivores [19] . Thirty small mammals were collected specifically for this investigation , and 55 others were trapped by inhabitants of villages to reduce rodent damage to trees and gardens . Small Sherman live traps ( 25×10×10 cm ) were baited with cheese , muffins spread with butter , walnuts , or fruit . Trapping sites included gardens , river banks , storage rooms , and areas near burrows . All traps were checked twice daily and trapped animals were collected , labeled with date and place of sampling , and stored at −20°C . Small mammals were identified using standard morphological criteria [24] . They were dissected , and the thoracic and peritoneal cavity and visceral organs , particularly the liver , were examined macroscopically for cysts of E . multilocularis and other parasites . Distinguishable lesions and the liver of all specimens were excised and preserved in 80% ethanol for molecular examination . For molecular analysis , the ethanol was discarded and liver samples were hydrated with 0 . 9% NaCl overnight . The liver was forced through a 420 µm mesh sieve and washed with PBS buffer . The liver puree and PBS buffer were transferred to a 15 ml falcon tube , centrifuged at 800×g for 10 min , and 400 µl of the cell suspension in PBS buffer , equal to approximately 25 mg liver tissue , was transferred to 2 ml tubes . DNA was extracted using the QIAamp DNA Mini kit ( QIAGEN , Germany ) tissue protocol , according to manufacturer's instructions , and the Verweij et al . [25] protocol with slight modification as described [19] . DNA was stored at −20°C until molecular analysis . All DNA samples were amplified using primer pairs , and conditions in multiplex PCR as described for detection of E . multilocularis , E . granulosus , and Taenia spp . infections [26] . The primer pairs were arranged to amplify partial sequences of the mitochondrial genes for NADH dehydrogenase subunit 1 ( nad1 ) for detection of E . multilocularis , and the small subunit of ribosomal RNA ( rrnS ) for detection of E . granulosus and Taenia spp . Multiplex PCR was conducted on a final volume of 25 µl reaction mixture according to conditions and parameters previously described [19] . Amplification products were visualized by 2% ( W/V ) agarose gel electrophoresis , and the 395 , 117 , and 267 bp expected fragments were examined for presence of E . multilocularis , E . granulosus , and Taenia spp . , respectively . In all PCR reactions , distilled water was used as a negative control and standard DNA of E . multilocularis , E . granulosus , and Taenia hydatigena ( provided by Professor Deplazes , Institute of Parasitology , Zurich , Switzerland ) as positive controls , to validate the PCR reaction results . In order to decrease inhibition factors and increase likelihood of detecting positive samples , we diluted DNA samples with distilled water and conducted multiplex PCR on serial dilutions of DNA . The optimal volume of DNA for PCR reaction was in the range 0 . 25–2 µl in 25 µl of reaction mixture . For further confirmation , samples were examined using single PCR with primers Cest1/Cest2 [26] and EM-H15/EM-H17 [20] for E . multilocularis , Cest4/Cest5 for E . granulosus , and Cest3/Cest5 for Taenia spp . [26] and sequencing of E . multilocularis positive samples . Echinococcus multilocularis amplified fragments were extracted from agarose gels using the QIAquick Gel Extraction Kit ( QIAGEN , Germany ) , according to the manufacturer's instructions and were sequenced on both strands with primers Cest1/Cest2 ( Bioneer , Korea ) . The sequence results were compared with the Genbank database using the DNASIS MAX ( version 2 . 09; Hitachi , Yokohama , Japan ) software .
Base on morphological criteria , the 85 small mammals trapped in five villages of Chenaran County were classified into six species of the four families Cricetidae , Muridae , Soricidae , Ochotonidae . The majority of small mammals caught were Microtus transcaspicus ( 63 . 5% , 54/85 ) of the Cricetidae . Muridae was second , with 15 Mus musculus ( 17 . 6% ) , nine Apodemus witherbyi ( 10 . 6% ) , and one Nesokia indica ( 1 . 2% ) . Four were Ochotona rufescens ( 4 . 7% ) , and two were Crocidura gmelini ( 2 . 4% ) of the non-rodent families Ochotonidae and Soricidae ( Table 1 ) . Macroscopic examination of visceral organs showed liver cysts in nine of 85 ( 10 . 6% ) animals . Liver cysts were isolated from six of 54 ( 11 . 1% ) M . transcaspicus and two of 15 ( 13 . 3% ) M . musculus . Three cysts were observed , in liver of one C . gmelini . Multiplex PCR showed 30 of 85 captured specimens ( 35 . 3% ) to be infected with E . multilocularis and 14 ( 16 . 5% ) infected with Taenia spp . by amplification of 395 bp fragment of nad1 and 267 bp fragment of rrnS , respectively ( Figure 2 ) . Echinococcus multilocularis infection was identified in liver of 23 of 54 M . transcaspicus ( 42 . 6% ) , three O . rufescens ( 75 . 0%; 3/4 ) , two M . musculus ( 13 . 3%; 2/15 ) , one C . gmelini ( 50 . 0%; 1/2 ) , and one A . witherbyi ( 11 . 1%; 1/9 ) . Taenia spp . were found in liver of nine M . transcaspicus ( 16 . 7%; 9/54 ) , two M . musculus ( 13 . 3%; 2/15 ) , one C . gmelini ( 50 . 0%; 1/2 ) , and one A . witherbyi ( 11 . 1%; 1/9 ) . The only N . indica specimen captured was infected with Taenia spp . The single PCR amplifications confirmed the results of multiplex PCR . Echinococcus multilocularis and Taenia spp . co-infections were revealed in 11 of 33 PCR positive samples by amplification of two species-specific fragments ( Table 2 ) . A single amplicon detected 19 E . multilocularis ( 22 . 4% ) and three Taenia spp . ( 3 . 5% ) infected small mammals ( Table 2 ) . Fifty-two liver samples were negative for E . multilocularis and Taenia spp . by all methods used . Echinococcus granulosus infection was not found in any liver sample . All positive samples were confirmed as E . multilocularis using sequencing of the nad1 gene . The alignment of amplified nad1 sequences showed 100% identity with published reference sequences for E . multilocularis . The nucleotide sequence of five E . multilocularis amplified nad1 genes from five small mammal species were deposited in the DDBJ/EMBL/GenBank nucleotide sequence database under accession number AB720065–69 .
Although previous studies of definitive hosts have revealed that northwestern Iran is an endemic focus for E . multilocularis , its metacestode stages have not been found [12] , [17] . The most recent investigation , using morphological and molecular methods , indicated high endemicity in the newly surveyed Chenaran County , with 100% prevalence of infection in wild carnivores and 6 . 5% in domestic and stray dogs [19] . As the presence of both definitive and intermediate hosts is required for establishment and maintenance of the life cycle , finding a high level of E . multilocularis infection ( 35 . 3% ) among 85 small mammals belonging to two rodent families , Cricetidae and Muridae , and two non-rodent families , Ochotonidae and Soricidae , has confirmed the existence of the E . multilocularis life cycle in Chenaran County . Reports of the prevalence of E . multilocularis infection have ranged from less than 1% to more than 80% in small mammals of the Soricidae , Talpidae , Sciuridae , Cricetidae , Arvicolidae , Muridae , Dipodidae , and Ochotonidae [2] , [12] , [27] , [28] . This wide variation might be due to a wide spectrum of sensitive intermediate hosts [2] as well as to the number of investigated hosts and the diagnostic methods used [12] . The rate of infection in our study ( 35 . 3% ) was lower than that found in some regions [2] , [12] , [27] , but higher than reported in others [29] , [30] , [31] . The differing findings might be the result of identification based on gross and microscopic appearance of lesions found by histology [32] , [33] or conducting PCR only on visually unidentifiable lesions [29] , [30] , [31] . In our study , cysts were detected in only 10 . 6% of the 85 investigated liver samples by direct examination , but this increased to 38 . 8% positive E . multilocularis and Taenia spp . infection with multiplex PCR on liver of all sampled specimens . It may be assumed that this finding reflected the complexity of distinguishing small immature cysts [34] , especially in animals less than three months old [29] , or in atypical or calcified liver lesions less than 5 mm in diameter [27] , [35] by microscopic examination , while 14 pg of DNA can be detected by multiplex PCR [36] . An experimental infection of Microtus arvalis showed that PCR gives the only definitive diagnosis in lesions of less than two-weeks duration [37] . This may be the consequence of protoscoleces in the metacestode of E . multilocularis development extending over the course of 2–4 months in the liver of its natural intermediate host [12] . Stieger et al . [20] showed that E . multilocularis-specific PCR of 161 morphologically unidentifiable liver lesions of Arvicola amphibius ( formerly A . terrestris ) found 55 ( 34 . 2% ) positive for E . multilocularis infection , increasing the detected prevalence of E . multilocularis in A . amphibius from 2 . 9% ( 26/889 ) to 9 . 1% ( 81/889 ) [20] . In a study in Geneva , Switzerland , in which 658 non-commensal rodents were investigated using morphological and molecular methods , metacestodes of E . multilocularis were detected in 2 adult A . amphibius , while PCR identified E . multilocularis infection in 29/79 A . amphibius , 3/4 M . arvalis , and 6/9 Myodes glareolus which was not found using morphological methods [21] . Microtus transcaspicus was the most frequently captured species ( 63 . 5% ) and may be the dominant small mammal in the Chenaran area . It seems that this location , at an elevation of 1400–1600 m , having moist soil with trees and shrubs along river valleys is a suitable habitat for the Transcaspian vole ( M . transcaspicus ) . Factors such as high elevation , low temperatures , high precipitation , moist soil , and an abundance of green vegetation provide suitable conditions for survival of E . multilocularis eggs [29] , [30] , [38] in feces of infected wild carnivores [19] , in the studied habitats of small mammals . In central Europe , the main intermediate hosts are M . arvalis ( common vole ) , A . amphibius ( water vole ) , and Ondatra zibethicus ( muskrat ) [12] , while in our study area , the higher density of M . transcaspicus , along with a high prevalence of infection , suggested an important role for this rodent in the E . multilocularis life cycle . Although the previous study in this area showed high rates of E . granulosus infection in carnivores [19] , no infection was identified in the small mammals examined . The first molecular identification of natural E . granulosus infection was reported in a single ground squirrel ( Spermophilus dauricus ) , one of 500 small mammals trapped in northwest China , an endemic area for both E . granulosus and E . multilocularis [39] . While susceptible to E . multilocularis , E . granulosus infections have been seldom observed in rodents [40] . For detection of infected hosts , it may be necessary to investigate a greater number , and additional genera , of small mammals . The presence of Taenia spp . in 14 of 85 ( 16 . 5% ) specimens investigated , with 11 ( 12 . 6% ) found co-infected with E . multilocularis by multiplex PCR , is a good indicator of contamination of the environment with taeniid eggs . Voles are natural intermediate hosts of several zoonotic helminthes , including E . multilocularis , T . taeniaeformis , T . crassiceps , and Toxocara canis that can infect humans who ingest eggs excreted by the final hosts [41] . Under suitable conditions , taeniid eggs might survive up to eight months and can be spread by shoes , animal paws , flies , or other vectors , infecting small mammals , humans , and other intermediate hosts in the endemic area [42] . In conclusion , the presence of infection in small mammals suggests the active transmission of E . multilocularis in the selected area . The existence of the life cycle of this potentially lethal cestode in the vicinity of human habitats provides a significant risk of human infection . It is recommended that an extensive survey be conducted to investigate the prevalence of E . multilocularis in humans and domestic ungulates in Razavi Khorasan Province . In addition , there is a need to educate the local population about the infection , and programs for reducing the risk of transfer of infection to human and domestic animals should be initiated in Chenaran rural areas . As several human cases have been reported in other parts of Iran [43] , further studies to investigate the life cycle of E . multilocularis in other parts of the country is recommended . | Small mammals , especially rodents , coexist with humans , particularly in rural areas where they become infected with the larval stages of Echinococcus multilocularis via ingestion of eggs in feces from infected carnivores . As prey for carnivores , small mammals have an important role in the life cycle of E . multilocularis , the agent of a serious zoonotic disease , alveolar echinococcosis , infecting people in most northern hemisphere countries . We detected E . multilocularis infection in the liver of small mammals in villages of the Chenaran region of northeastern Iran , where we previously found adult E . multilocularis and/or eggs in all wild , and some domestic , carnivores examined . Several species of small mammals , especially the Transcaspian vole ( Microtus transcaspicus ) , were natural hosts of this cestode . Since infected voles are a good marker of soil contamination with E . multilocularis eggs , they are an indicator of the infection risk for inhabitants . There is a need to educate local populations about risk of infection to avoid transmission of this pathogenic parasite . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"echinococcosis",
"infectious",
"diseases",
"biology",
"microbiology",
"zoology",
"parasitic",
"diseases",
"parasitology"
]
| 2013 | Molecular Identification of Echinococcus multilocularis Infection in Small Mammals from Northeast, Iran |
While bacterial symbionts influence a variety of host cellular responses throughout development , there are no documented instances in which symbionts influence early embryogenesis . Here we demonstrate that Wolbachia , an obligate endosymbiont of the parasitic filarial nematodes , is required for proper anterior-posterior polarity establishment in the filarial nematode B . malayi . Characterization of pre- and post-fertilization events in B . malayi reveals that , unlike C . elegans , the centrosomes are maternally derived and produce a cortical-based microtubule organizing center prior to fertilization . We establish that Wolbachia rely on these cortical microtubules and dynein to concentrate at the posterior cortex . Wolbachia also rely on PAR-1 and PAR-3 polarity cues for normal concentration at the posterior cortex . Finally , we demonstrate that Wolbachia depletion results in distinct anterior-posterior polarity defects . These results provide a striking example of endosymbiont-host co-evolution operating on the core initial developmental event of axis determination .
The phylum Nematoda comprises up to 1 million species and is one of the most diverse and successful , with members colonizing all possible ecological niches on earth [1] , [2] . Nematodes have an extraordinary ability to adapt to the parasitic life style [3]–[6] and as a result exert profound impacts on agriculture and human health . The Spirurina clade contains only animal parasites , among them the Onchocercidae or filarial nematodes [5] . These thread-like worms are tissue-dwelling parasites , transmitted by arthropods , usually black flies or mosquitoes , to all classes of vertebrates except fish . It is estimated that 150 million people are infected with filarial nematodes , with 1 billion living at risk in tropical areas . Filarial nematodes lead to debilitating diseases such as onchocerciasis ( caused by Onchocerca volvulus ) and lymphatic filariasis ( Brugia malayi , Brugia timori , Wuchereria bancrofti ) [7] . A total of eight species of filarial nematodes are responsible for these neglected tropical diseases . With the exception of Loa and certain Mansonella sp . , all other human filariae harbor an alpha-proteobacterium of the genus Wolbachia . This symbiosis is restricted to the family of Onchocercidae among nematodes [7] , [8] . In addition , Wolbachia are also widespread among arthropods [9] and the bacteria of this genus have been classified into different supergroups , as defined by MultiLocus Sequence Typing [10] , [11] . The supergroups C and D represent the majority of Wolbachia in filarial species and are restricted to the Onchocercidae [8] . Wolbachia are required for filarial nematode fertility and survival [12] and we previously showed that removal of either supergroup C or D bacteria by antibiotic therapies against O . volvulus or B . malayi leads to extensive apoptosis [13] . Yet little is known about the actual basis of the mutualistic interaction . Genomic analysis and experimental studies suggest that Wolbachia may contribute to metabolic pathways absent or partially missing in the nematode host , including synthesis of riboflavin , nucleotides and hemes [14]–[16] . However , the recent publication of the Loa genome , a Wolbachia-free human filarial parasite , revealed no metabolic compensation for the lack of mutualistic endosymbionts , suggesting caution in drawing conclusions on the basis of the symbiosis from genomic studies [17] . In the vast majority of filarial species , Wolbachia are present in the hypodermal chords of both male and female adult specimens , and in the female germline [8] . This is achieved through both asymmetric segregation during the mitotic divisions and cell-to-cell migration [18] . Immediately following fertilization , Wolbachia concentrate at the posterior region of the embryo . Wolbachia first localize in the posterior germline precursor lineage by rounds of asymmetric segregation until the 12-cell stage . They then reach a hypodermal lineage , and from this subset of posterior hypodermal cells , the bacteria colonize the whole dorsal and ventral hypodermal syncytia during late larval development , spreading toward the anterior of the worm [18]–[20] . Here we focus on the rapid migration and concentration of Wolbachia at the posterior pole immediately during the oocyte-to-embryo transition in B . malayi as this is a key unexplored initial event determining the distribution of Wolbachia in adult tissues . We used C . elegans , the sole well-studied nematode , as a reference for the oocyte-to-embryo transition in B . malayi . Although phylogenetically distant , the free-living and parasitic species both belong to the secernentean nematodes , and share a very similar embryonic development [21] [22] [23] [1] . To identify host factors involved in Wolbachia asymmetric enrichment after fertilization , we first characterized the cytoskeleton of the B . malayi embryo . As described below , we discovered a posterior microtubule-organizing center ( MTOC ) in the unfertilized mature oocyte . This is in striking contrast to C . elegans , in which the MTOC originates from the sperm-derived basal body/centrosome and induces cytoskeletal asymmetries essential for proper anterior-posterior polarity establishment [24] . Thus centrosome inheritance and its role in anterior-posterior polarity determination are dramatically different in C . elegans and filarial B . malayi . This maternally-derived B . malayi posterior MTOC facilitates Wolbachia concentration in the posterior of the newly fertilized egg . Using immunofluorescence and recently developed RNA silencing techniques [25] , we show that host dynein is required for Wolbachia posterior enrichment in the egg . In addition , Wolbachia posterior localization requires B . m . PAR-1 and PAR-3 , the B . malayi orthologs of C . elegans polarity-determining proteins Ce PAR-1 and Ce PAR-3 . Finally , we demonstrate that Wolbachia removal results in Anterior-Posterior polarity defects , demonstrating for the first time that Wolbachia plays an essential role in these early embryonic developmental events .
Live specimens were obtained from the NIH/NIAID Filariasis Research Reagent Resource Center ( www . filariasiscenter . org ) . To obtain B . malayi adults devoid of Wolbachia , infected jirds were administered tetracycline at 2 . 5 mg/ml in drinking water ( water changed daily ) for a period of six weeks , followed by a one week clearance period . While this treatment is enough to deplete Wolbachia from filarial nematodes , the tetracyclin itself does not affect the host gene expression , including mitochondrial genes , as demonstrated by microarray after treatment of A . viteae , afilarial species devoid of Wolbachia [16] . Untreated infected jirds were maintained in a similar fashion as a control . After the clearance period , adult worms were recovered from the peritoneal cavities into preheated ( 37°C ) culture medium RPMI-1640 supplemented with 100 U/ml penicillin , 100 µg/ml streptomycin , 2 mM L-glutamine , 0 . 25 µg/ml amphotericin B , and 25 mM HEPES ( GIBCO ) . The Animal Research and Care Program at UWO follows regulations and guidelines established by the USDA Animal Welfare Act , Public Health Service Policy , and the Association for the Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) . The protocol followed has been approved by UWO IACUC . Protocol 0-03-0026-000252-11-22-11 , “Oral Tetracycline Treatment of Mongolian Gerbils ( Meriones unguiculatus ) ” . Approval Date: 11/22/11 . Expiration Date: 12/9/14 AAALAC #: 001268 . All the B . malayi genes have been identified as C . elegans orthologs by reciprocal BLAST using the NCBI protein BLAST tool ( http://blast . ncbi . nlm . nih . gov . gate1 . inist . fr/Blast . cgi ) Two peptides were designed for each of B . m . γ-tubulin and B . m . Zyg-9 and were used together to immunize rabbits: -B . m . gamma tubulin ( gene ID: 6105932 Bm1_55245 ) : ( VRETVQTYRNATKPDFIEIN ) and ( GSHALEKISDRFPKKLVQTY ) -B . m . zyg-9 ( gene ID: 6096160 Bm1_06160 ) : ( MHKSNPLKPPAP ) ( RSDRSSSRIGRNTHRSNSVSRDSS ) For Dhc-1 a single peptide was used ( gene ID: 6103168 Bm1_41435 ) : ( LGGSPFGPAGTGKTESVKAL ) Peptides were synthesized by the Organic Synthesis group of New England Biolabs with an additional N-terminal cysteine residue to facilitate conjugation to the carrier protein KLH using m-Maleimidobenzoyl-N-hydroxysuccinimide ester ( MBS; Pierce , Rockford , IL ) [21] . Sera were raised in rabbits by Covance Immunology Services , Denver , PA . Peptides were purified essentially according to a published procedure [26] . Antibodies raised against pericentriolar markers ( i . e . B . m . gamma tubulin and B . m . zyg-9 ) co-localize with MTOCs ( cf . Fig . 1 ) . In both Spirurida ( i . e . B . malayi ) and Rhabditina ( i . e . C . elegans ) , chromosomes are holocentric [27] . In B . malayi , the anti B . m . dhc-1 concentrates along the holocentric chromosomes during metaphase as dynein does in C . elegans [28] . All the silencing experiments were performed as already described [25] . Briefly , B . malayi females were soaked in 1 µM of heterogenous short interfering ( hsi ) RNA mixtures for 48 hours before egg and embryo collection and fixation . PCR primers used to generate the primary dsRNAs contained T7 promoter sequence followed by two guanine bases at their 5′ ends for transcription by T7 RNA polymerase and enhanced transcription yield . -Par-1 ( gene ID: 6100834 Bm1_29690 ) forward: 5′- TAA TAC GAC TCA CTA TAG GGG AGA GGA ATC TTG CCA ACG G -3′ reverse: 5′- TAA TAC GAC TCA CTA TAG GGA ACT GCT TGT GCA GAT GCG C -3′ -Par-3 ( gene ID: 6103110 Bm1_41135 ) forward: 5′- TAA TAC GAC TCA CTA TAG GGT TCT GGA TCC CGA TGA TCA G-3′ reverse: 5′- TAA TAC GAC TCA CTA TAG GGT AGA CGT GAT TTC CTA GCG G-3′ -Dhc-1 ( gene ID:6103168 Bm1_41435 ) forward: 5′- TAA TAC GAC TCA CTA TAG GGA GCA ACT GTC AAG GAA AAG -3′ reverse: 5′- TAA TAC GAC TCA CTA TAG GGA TGG AGA CAA GTC GAT ATC C -3′ Embryos were collected , fixed and stained as already described in detail [25] . Polyclonal anti B . m . Zyg-9 , anti B . m . gamma tubulin and anti B . m . Dhc-1 were used at a dilution of 1∶100 . Microtubule stainings were performed using the monoclonal DM1α antibody raised against α-tubulin ( Cell Signaling Technology , Danvers , MA , USA ) at a dilution of 1∶100 . Cy5 goat anti-rabbit IgG and Alexa Fluor 488 goat anti–mouse IgG antibodies were used at 1∶150 ( Invitrogen ) . Primary and secondary stainings were both performed overnight either at 4°C or room temperature . Actin stainings were performed using the fluorescent Atto 488 phalloidin ( Sigma ) at a dilution of 1∶100 , added with secondary antibodies . The Wolbachia were visualized with propidium iodide ( PI ) . We previously showed that the PI puncta only correspond to Wolbachia DNA by colocalization with Wolbachia-specific antibodies [19] . For propidium iodide ( Molecular Probes ) DNA staining , embryos were fixed then incubated overnight at room temperature in PBS+RNAse A ( 15 mg/mL , Sigma ) , in rotating tubes overnight . PI incubation itself was done after the secondary antibody wash ( 1 . 0 mg/mL solution ) by simply shaking the eppendorf for 10 seconds in PBS followed by a 5 minute wash . 30 second centrifugations at 4000 rpm in between steps are enough to pellet all embryos . They were then mounted into Vectashield ( Vector Laboratories , Burlingame , CA ) . Confocal microscope images were captured on an inverted photoscope ( DMIRB; Leica Microsystems , Wetzlar , Germany ) equipped with a laser confocal imaging system ( TCS SP2; Leica ) using an HCX PL APO 1 . 4 NA 63 oil objective ( Leica ) at room temperature . 3-D movies were generated using the Volocity 3D Image analysis software ( PerkinElmer ) .
Wolbachia have been shown to rely on host microtubules , kinesin and dynein in insects to properly segregate to the posterior germline pole plasm during oogenesis [29] , [30] . To establish whether or not Wolbachia transmission also depends on similar cytoskeletal interactions in filarial nematodes , the microtubule network was characterized during the oocyte-to-embryo transition . To follow the microtubules and pericentriolar material ( PCM ) , anti-B . m . γ-tubulin and anti-B . m . Zyg-9 antibodies were generated ( Cf . experimental procedures; Fig . 1 ) . In the free living nematode C . elegans , as in most animal species , centrosomes are degraded during oogenesis , prior to diakinesis [31] , [32] . In inseminated females , the cellularized oocyte follows a meiotic maturation phase , under the control of a sperm major protein ( MSP ) released from the sperm prior to fertilization [33] . During maturation , the germinal vesicle migrates away from the MSP source , with its associated acentriolar spindle , toward the unpolarized oocyte cortex . Centrosomes have a paternal origin and are inherited upon fertilization . The sperm-supplied centrosome participates to establishment of A-P polarity in the zygote , and the entry point defines the posterior pole of the egg [34] . In contrast to C . elegans , the presence of a microtubule-organizing center ( MTOC ) , located at the opposite pole of the germinal vesicle was detected in unfertilized mature meiosis I oocytes from B . malayi . This polar MTOC is defined by both the presence of PCM components γ-tubulin and Zyg-9 proteins , and its ability to nucleate microtubules ( Fig . 1 , see also Fig . 2A , Movie S1 ) . However , it disappears after fertilization , by the time pronuclei apposition occurs ( Fig . 2 ( A ) to ( B ) ) . Upon fertilization , no sperm-associated or paternal nucleus-associated MTOC was ever detected ( Fig . 3 ( A ) and ( B ) , Movie S1; n>100 ) . At this stage , microtubules do not nucleate at the surface of the paternal pronucleus , suggesting the absence of a paternally-derived MTOC . Rather , the anti-γ-tubulin antibody revealed numerous cytoplasmic foci ( Fig . 2 ( A ) ) . Some of these foci coalesce around the apposed pronuclei to form the MTOCs while the others are gradually degraded ( Fig . 2 ( B ) to ( D ) ) . This correlates with the microtubule dynamics at this stage ( Fig . 4 ( C ) to ( E ) ) . Together , these data demonstrate the presence in B . malayi of a MTOC-associated microtubule cytoskeleton in the mature cellularized oocyte , and suggest a maternal de novo origin of centrosomes in filarial nematodes , in contrast to C . elegans ( Fig . 2 ( E ) ) . We next examined Wolbachia dynamics in the mature oocyte and early embryo to better understand how they concentrate at the posterior blastomere during the two cell stage . We first characterized their dynamics in zygotes during the 1st cell cycle ( Fig . 4 , n>100 ) . Prior to , and soon after fertilization ( Fig . 4 ( A ) and ( B ) ) , Wolbachia are dispersed in the egg , sometimes showing a preference for the meiotic spindle and the opposite pole [19] ( see also Fig . 5 ) . The concentration in the posterior half of the egg starts during pronuclei migration and apposition ( Fig . 4C ) , and is achieved by the beginning of prophase ( Fig . 4D ) . This localization is maintained through mitosis ( Fig . 4 ( D ) to ( J ) ) and enables the vast majority of endosymbionts to segregate in the posterior blastomere P1 after cytokinesis ( Fig . 4K ) . This posterior segregation pattern is repeated in the dividing two-cell embryo ( Fig . 4L ) . Thus , Wolbachia are asymmetrically localized very early in the zygote , to become enriched at the posterior end before entry into mitosis . We established that Wolbachia asymmetrically localize in the egg prior to the first mitosis , and are maintained at the posterior pole during mitosis . To further investigate a possible role of the microtubule cytoskeleton in Wolbachia dynamics , we looked for close association between the endosymbionts and microtubule network ( Fig . 5 , n>100 ) . We found Wolbachia in the vicinity of microtubules emanating from the polar MTOC after fertilization ( Fig . 5 ( A ) and ( A′ ) ) . Later during mitosis , we found Wolbachia organized along the posterior astral microtubules ( Fig . 5 ( B ) and ( B′ ) ) . These data suggest that the microtubule cytoskeleton may be used by Wolbachia first for concentration , second for maintenance at the posterior pole of the egg . In Drosophila , Wolbachia rely on plus and minus end directed motor proteins for their concentration at the posterior pole of the Drosophila embryo [29] , [30] . Our finding that Wolbachia closely localize to microtubules suggests they may concentrate at the posterior pole through their association with microtubule based motor proteins . The polar MTOC projects microtubule plus-ends inward and it was of interest to ascertain whether or not the Wolbachia may use the host minus-end molecular motor Dynein to segregate to the future posterior pole of the egg . To achieve this , the B . m . Dynein heavy chain 1 ( B . m . Dhc-1 ) was silenced by soaking adult females in hsiRNA for 48 hrs [25] , ( Fig . 6 ) . We collected a vast majority of multinucleated 1-cell eggs , as a result of chromosome segregation and cytokinesis failure when Dynein was reduced or absent . These highly penetrant phenotypes indicates that the hsiRNA is efficiently knocking down the Dynein levels . In these eggs , Wolbachia were evenly distributed in the cytoplasm ( cf . Fig . S1 ) . To circumvent the lack of developmental timing information in these eggs , we focused on zygotes prior to entry into the first mitosis ( n = 10 ) . In wild-type eggs , the majority of bacteria are at the posterior pole ( n>100 ) . In contrast , upon B . m . dhc-1 hsiRNA treatment , they no longer distribute asymmetrically ( Fig . 6 ( A ) and ( B ) ) . To test a putative direct interaction between Wolbachia and Dynein , we raised an antibody against the B . m . dhc-1 . Similar to studies in C . elegans [35] , the anti-Dynein antibody decorates the condensed chromosomes in the zygote ( Fig . 6C arrowhead ) . Significantly Dynein also colocalizes with posterior localized Wolbachia ( Fig . 6 ( C ) and ( C′ ) , arrow ) . This strongly suggests that Wolbachia may use the host Dynein and the polar MTOC for their initial asymmetric enrichment . In B . malayi , after pronuclei apposition , the polar MTOC is no longer present in the egg . What then keeps Wolbachia in the posterior until the first division takes place ? We tested the influence of Anterior-Posterior ( A-P ) polarity establishment in Wolbachia localization and maintenance . Establishment of A-P polarity has been extensively studied in zygotes of the free living nematode C . elegans . In this species , symmetry breaking is triggered by sperm entry [34] . A remodeling of the cortical cytoskeleton is associated with a redistribution of the PARs polarity cues , as well as intense cytoplasmic streaming , to form an anterior and a posterior cortical domain by the beginning of mitosis . Subsequently , downstream polarity effectors are required to establish an asymmetric division [36] . To test whether PARs-induced symmetry breaking mechanisms dictate the bacteria asymmetric distribution , the B . malayi orthologs of C . elegans posterior PAR-1 and anterior PAR-3 were identified and silenced by hsiRNA . Due to the relatively low penetrance of the PAR-1 and PAR-3 hsiRNA phenotypes ( ∼30% , n>100 in both cases ) , we focused on dividing two cell embryos which showed classic PAR polarity-defect phenotypes: synchronous mitotic divisions and abnormal spindle orientation [37] . In wild-type B . malayi and C . elegans two-cell embryos , the anterior AB blastomere enters mitosis before the posterior P1 blastomere ( Fig . 7A ) . This asynchrony is even more pronounced in B . malayi , where three-cell embryos , composed of AB daughters and dividing P1 , are commonly observed . Also in B . malayi , like C . elegans , the posterior P1 spindle rotates by 90° to align along the A-P axis , while the AB spindle remains transverse ( Movie S2 ) . As in C . elegans , hsiRNA knockdown of either par1 or par3 disrupts the normal mitotic asynchrony between the two B . malayi blastomeres . In addition , upon B . malayi par-1 hsiRNA , the P1 spindle fails to rotate ( Fig . 7A , Movie S3 ) , while upon B . malayi . par-3 hsiRNA treatment , the AB spindle now rotates to align along the long ( A-P ) axis of the embryo ( Fig . 7A ) . These timing and spindle orientation defects are strikingly similar to those observed in C . elegans [37] and reveal at least partial evolutionary conservation of functions for B . malayi PAR-1 and PAR-3 . The presence of these polarity defects correlates with a loss of Wolbachia asymmetric segregation or maintenance at the posterior pole ( Fig . 7B ) . This indicates that the A-P polarity determinants are essential for the stable enrichment of Wolbachia in the posterior P1 blastomere . By the first mitotic division , Wolbachia are predominantly concentrated in the posterior half of the B . malayi egg . In the C . elegans zygote , the complete establishment of the anterior and posterior cortical domains is already achieved by the beginning of mitosis [38] . As it is likely that A-P polarity set up in B . malayi takes place no later than in C . elegans , it was of interest to determine whether Wolbachia might influence the A-P polarity in the zygote . To investigate this , we analyzed A-P polarity in normal and Wolbachia-depleted two-cell embryos ( cf . Experimental Procedures , [13] ) . This analysis yielded the following phenotypic classes ( Fig . 8A ) : Class I included those with normal division patterns exhibiting mitotic division asynchrony and proper spindle orientation . Class II included those with “posterior polarity” defects exhibiting a failure of P1 spindle rotation and division synchrony , and Class III included those with “anterior polarity” defects exhibiting inappropriate rotation of the AB spindle and division synchrony . The vast majority of wild-type embryos ( 97% , n = 75 ) showed class I normal division patterns ( Fig . 8 ( A ) and ( B ) ) . Embryos devoid of Wolbachia ( n = 27 ) displayed a dramatic loss of normal class I division patterns ( 48% ) . The remaining half of embryos lacking Wolbachia displayed either Class II posterior defects ( 40% , Fig . 8 ( B ) and Movie S4 ) or Class III anterior ( 11% ) defects . These results reveal that Wolbachia not only rely on A-P polarity cues for their posterior location but also are essential for proper establishment of AP polarity in its filarial nematode host .
Centrosome inheritance is asymmetric in metazoan sexual reproduction . Usually , but not always , centrosomes are degraded in the female germline and provided paternally through the transformation of the sperm-derived basal body . This mechanism of inheritance ensures a tight control of centrosome number and MTOCs in the zygote , [39] . A dramatic exception to the typical pattern of paternal centrosome inheritance occurs in parthenogenetic development of unfertilized eggs in Hymenoperta . In this case , centrosomes and their associated MTOCS are derived exclusively from maternally derived components [40]–[42] . Our studies demonstrate a third unique centrosome/MTOC inheritance pattern in B . malayi . First , the unfertilized mature oocyte contains a maternal-derived MTOC . Second , despite fertilization , centrosomes appear to be produced de novo and to be maternally supplied . Accordingly , no paternally-derived MTOC was observed associated with the paternal chromatin after sperm entry . Whether or not the maternal MTOC originates from a centrosome remains to be determined , since acentrosomal PCM has been shown to nucleate microtubules in vitro [43] . In any case , this maternal MTOC never interacts with the paternal chromatin and is degraded soon after fertilization . We find that during pronuclei apposition , the PCM component γ-tubulin accumulates around the nuclear envelopes as foci , and this correlates with microtubule enrichment at the nuclear surface . The presence of functional MTOCs capable of microtubule nucleation is only observed after entry of the pronuclei into mitosis . Together , these findings suggest centrosomes are derived exclusively from maternal components and perhaps form de novo in filarial nematodes . New centrosomal markers will be required to identify the origin and composition of the polar MTOC . These findings also raise important questions regarding the mechanism of symmetry breaking and polarity establishment in filarial nematode embryos . In C . elegans , the paternally supplied centrosome and its associated MTOC play a crucial role in polarity establishment . The sperm derived centrosome/MTOC elicits a dramatic reorganization in the actomyosin cortical network and asymmetric localization of polarity components such as PAR-1 [34] , [44] . It is currently unclear how much of a role the maternally-derived MTOC or fertilization plays in symmetry breaking and polarity establishment in the B . malayi embryo . The design of much needed new reagents suitable for filarial species will help us to understand the great variations on fundamental mechanisms between the free living C . elegans and filarial nematode species . This may help us to better understand peculiarities of the parasitic lifestyle , and sources of such evolutionary divergence . In insects , Wolbachia must navigate the constantly changing cytoskeletal environment of the oocyte in order to concentrate at the posterior pole where the germline will form . Wolbachia rely on host microtubules for their transport through the oocyte . Early in oogenesis they rely on the plus-end motor protein kinesin . Later , the microtubules reorganize and reverse orientation requiring Wolbachia to engage dynein to complete their poleward journey . The studies presented here indicate that Wolbachia in B . malayi are also very likely to rely on microtubules and motor proteins for their asymmetric concentration in the posterior pole of the embryo . Unlike in C . elegans , prior to fertilization B . malayi oocytes possess a robust posteriorly positioned MTOC with microtubules emanating towards the anterior positioned meiotic spindle . Upon fertilization , the Wolbachia associate with microtubules and concentrate at this unusual posteriorly positioned MTOC . We also observe a striking co-localization between Wolbachia and the host dynein heavy chain . Significantly functional RNAi analysis demonstrates that dynein is required for this posterior enrichment . Thus in both insects and filarial nematodes dynein mediated movement is required for the asymmetric posterior positioning of Wolbachia to ensure germline incorporation . Although the posterior MTOC is established prior to fertilization , fertilization is required for the posterior concentration of Wolbachia . We believe that the maternally supplied posterior MTOC contributes to the initial Wolbachia concentration at the posterior pole . However it appears that maintenance of Wolbachia at the posterior pole requires cytoplasmic rearrangements mediated by fertilization , such as the asymmetric cortical localization of PAR polarity cues , controlling an asymmetric dynein activity at the cortex . Unlike many intracellular bacteria , Wolbachia have no flagellum , and do not appear to rely on the actin cytoskeleton for intracellular transport . As with Drosophila , Wolbachia in B . malayi concentrate near , and perhaps associate , with microtubules [29] . Upon fertilization in C . elegans , the sperm brings a basal body giving rise to male pronucleus-associated MTOCs , establishing the posterior of the egg [34] . A similar mechanism in filarial nematodes would have explained an early , microtubule-based movement of Wolbachia toward the posterior of the embryo . However the fertilization mechanisms and remodeling of the cytoskeleton during this step appear dramatically different in B . malayi . Why is fertilization then needed to achieve the asymmetric enrichment , if the polar MTOC is already present in the oocyte ? A simple model taking into account the microtubule cytoskeletal peculiarities of the filarial zygote could be envisioned ( Fig . 9 ) . A maternal polar MTOC projects microtubules inward , while meiosis is resumed at the opposite pole during fertilization ( Fig . 9 I to II ) . In turn the polar MTOC is degraded and absent by the time pronuclei appose ( Fig . 9 III ) , followed by entry into mitosis and set up of the mitotic spindle ( Fig . 9 IV ) . After fertilization , when the meiotic spindle is no longer present , the bacteria concentration is preferentially displaced toward the MTOC . Cell cycle progression may also alter Wolbachia interaction with the Dynein complex , or its activation , resulting in more engagement on the microtubules [45] . At pronuclei apposition , Wolbachia are in the posterior compartment , most of them associated with the most posterior pronuclear envelope ( paternal ) , but also in the cytoplasm , and in contact with the posterior cortex ( Fig . 9 III , i . e . Fig . 2 ( B ) and ( C ) ) . The association with the nuclear membrane correlates with a perinuclear accumulation of γ-tubulin foci ( Fig . 2C ) . Dynein is known to anchor the MTOC to the paternal nuclear envelope in C . elegans [46][35] . This motor may play a role in centrosome biogenesis and recruitment to the nuclear envelope in B . malayi , and may also mediate this Wolbachia localization . Cortical dynein has also been shown to play a crucial role in spindle positioning in C . elegans [47] , and Wolbachia cortical posterior localization could be mediated by the dynein itself . Once mitosis is triggered , whether Wolbachia interact with the astral microtubules or the cortex through dynein and/or other host factors , they remain trapped in the posterior compartment until cytokinesis occurs , and eventually segregate into the posterior blastomere ( Fig . 9 IV , Fig . 4 ) . In the C . elegans two-cell stage embryo , symmetry breaking mechanisms similar to those observed in the zygote lead to a polarized P1 [24] . Wolbachia asymmetric pattern of segregation is perfectly repeated when P1 divides ( Fig . 4L ) , confirming the importance of host A-P polarity signals in Wolbachia distribution in the early embryo . No polar MTOC is however required in P1 to achieve the same segregation observed in the zygote P0 . It is interesting that Wolbachia has co-evolved to adapt to a microtubule dynamics and architecture unique to fertilization in filarial nematodes . This peculiar de novo centrosome inheritance raises many important questions regarding the filarial oocyte-to-embryo transition . There are now a number of examples in diverse phyla in which bacteria have a profound influence on metazoan development [48] . For example , mice raised in a germ-free environment , exhibit defects in the enteric nervous system regulating gastrointestinal function [49] . Another striking example of animal bacterial interactions occurs in the Squid- V . fischeri symbiosis . The V . fischeri bacteria are required for proper development and morphology of the light organ of the squid . The bacteria induce very specific changes in cell size , morphology and microvilli formation [50] . Our analysis of the Wolbachia-B . mayali symbiosis provides a unique example in which the bacteria are required for normal host axis formation and embryonic development . B . malayi and C . elegans share similar division patterns during early embryogenesis , with AB dividing first , while in the posterior germline precursor P1 , the spindle rotates to align along the long A-P axis . These traits are common among the nematode species so far examined [51] . Without Wolbachia , A-P polarity establishment is compromised in the filarial zygote , as revealed by division timing and spindle orientation defects at the two-cell stage , a hallmark of A-P polarity defects in nematode species . How do the endosymbionts influence A-P polarity ? Since Wolbachia concentrate to the posterior before mitosis in B . malayi , ( a stage prior to establishment of A-P cortical domains in C . elegans ) , it is possible that Wolbachia directly influence localization and/or activation of B . malayi posterior polarity cues ( i . e . PARs ) , or on downstream posterior polarity effectors . Conversely , our experiments silencing B . m par-1 and par-3 , result in a failure of Wolbachia to become posteriorly enriched indicating that the PAR proteins are required for proper Wolbachia localization . In Drosophila , Wolbachia also associate with polarity determinants . Wolbachia closely associates with the Gurken polarity complex in the Drosophila oocyte and its titer regulated by Gurken levels . Significantly an overabundance of Wolbachia disrupts Gurken function [52] . The pioneering work of Sander in the 1950's demonstrated that displacing the ball of endosymbionts present in the leaf hopper Euscelis plebejus embryo from the posterior to a more anterior position produced ectopic posterior structures . This demonstrated a close association with posterior patterning determinants [53] . In nematodes Wolbachia not only rely on key host polarity factors for their germline transmission , but have become essential for the proper functioning of these determinants . At this point , however , we cannot rule out a non cell-autonomous explanation for the effect of Wolbachia-depletion on host A-P polarity . Unlike in C . elegans , B . malayi embryogenesis takes place entirely in the female uterus , where the growth of the embryo is dependent on maternal nutrients acquired from the hypodermis [2] , [19] , [54] . In addition , the endosymbionts fill the hypodermal tissues , a major site for nutrient storage and metabolism in filarial nematodes , and this bacterial population is also cleared upon antibiotic treatment [13] . Thus , it is then possible that Wolbachia removal from the hypodermis leads to metabolic defects affecting a plethora of signaling pathways , including the embryonic polarity set up . A better understanding of symmetry breaking mechanisms in these parasitic nematodes will help us establish precisely how Wolbachia influence embryonic polarity . In conclusion , we have shed light on the symbiosis mechanisms underlying Wolbachia transmission in the filarial embryo . They suggest a reciprocal dependence between the host and the symbiont starting as early as in the egg , explaining the success of antifilarial antibiotic therapies targeting Wolbachia , leading to massive embryogenesis defects . | Filarial nematodes are responsible for a number of neglected tropical diseases . The vast majority of these human parasites harbor the bacterial endosymbiont Wolbachia . Wolbachia are essential for filarial nematode survival and reproduction , and thus are a promising anti-filarial drug target . Understanding the molecular and cellular basis of Wolbachia-nematode interactions will facilitate the development of a new class of drugs that specifically disrupt these interactions . Here we focus on Wolbachia segregation patterns and interactions with the host cytoskeleton during early embryogenesis . Our studies indicate that centrosomes are maternally inherited in filarial nematodes resulting in a posterior microtubule-organizing center of maternal origin , unique to filarial nematodes . This microtubule-organizing center facilitates the concentration of Wolbachia at the posterior pole . We find that the microtubule motor dynein is required for the proper posterior Wolbachia localization . In addition , we demonstrate that Wolbachia rely on polarity signals in the egg for their preferential localization at the posterior pole . Conversely , Wolbachia are required for normal embryonic axis determination and Wolbachia removal leads to distinct anterior-posterior embryonic polarity defects . To our knowledge , this is the first example of a bacterial endosymbiont required for normal host embryogenesis . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"developmental",
"biology",
"molecular",
"motors",
"cell",
"biology",
"fertilization",
"symbiosis",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"cellular",
"structures",
"and",
"organelles",
"mutualism",
"parasitology"
]
| 2014 | Co-evolution between an Endosymbiont and Its Nematode Host: Wolbachia Asymmetric Posterior Localization and AP Polarity Establishment |
During embryonic development , the positional information provided by concentration gradients of maternal factors directs pattern formation by providing spatially dependent cues for gene expression . In the fruit fly , Drosophila melanogaster , a classic example of this is the sharp on–off activation of the hunchback ( hb ) gene at midembryo , in response to local concentrations of the smooth anterior–posterior Bicoid ( Bcd ) gradient . The regulatory region for hb contains multiple binding sites for the Bcd protein as well as multiple binding sites for the Hb protein . Some previous studies have suggested that Bcd is sufficient for properly sharpened Hb expression , yet other evidence suggests a need for additional regulation . We experimentally quantified the dynamics of hb gene expression in flies that were wild-type , were mutant for hb self-regulation or Bcd binding , or contained an artificial promoter construct consisting of six Bcd and two Hb sites . In addition to these experiments , we developed a reaction–diffusion model of hb transcription , with Bcd cooperative binding and hb self-regulation , and used Zero Eigenvalue Analysis to look for multiple stationary states in the reaction network . Our model reproduces the hb developmental dynamics and correctly predicts the mutant patterns . Analysis of our model indicates that the Hb sharpness can be produced by spatial bistability , in which hb self-regulation produces two stable levels of expression . In the absence of self-regulation , the bistable behavior vanishes and Hb sharpness is disrupted . Bcd cooperative binding affects the position where bistability occurs but is not itself sufficient for a sharp Hb pattern . Our results show that the control of Hb sharpness and positioning , by hb self-regulation and Bcd cooperativity , respectively , are separate processes that can be altered independently . Our model , which matches the changes in Hb position and sharpness observed in different experiments , provides a theoretical framework for understanding the data and in particular indicates that spatial bistability can play a central role in threshold-dependent reading mechanisms of positional information .
How an embryo achieves pattern and form from an initially undifferentiated state has fascinated people at least since the time of Aristotle . Scientific advances on this began over a century ago , with , for example , the experiments of Hans Driesch on sea urchin embryos [1] , from which he proposed that the embryo has a coordinate system specifying cellular position; and from the experiments of Ethel Browne [2] , who showed that a piece of hydra mount induced a secondary axis when grafted into the body of another hydra . These and other subsequent results were synthesized by Lewis Wolpert in 1969 [3] into a definition of positional information . According to this concept , the spatial asymmetries of concentration gradients of chemical signals ( morphogens ) provide positional information during cellular differentiation; each cell ( or nucleus ) reads its position from the local morphogen concentration and differentiates accordingly . Wolpert's concept of morphogen gradients has become a central tenet of developmental biology [4]–[6] . Modern molecular techniques have demonstrated numerous cases of protein concentration patterns in embryogenesis , and many have been shown to act as morphogens . In the late 1980's , the Bicoid ( Bcd ) protein gradient was characterized and its concentration-dependent effect on downstream target genes in Drosophila was demonstrated [7]–[9] . This has since become one of the most studied examples of morphogen gradient signaling in developmental biology [10] , [11] . Reaction-network models have been successfully applied to describe a great variety of systems in physics , chemistry , and biology [12]–[14] . Along with this , many mathematical tools have been developed to support such applications . With these tools , one can show that certain reaction networks may exhibit multiple stationary states , for particular ranges of their rate constants . Bistability is a special case , in which the system can evolve to either of two asymptotically stable steady states ( concentration levels ) . Under certain conditions , spatial patterning or oscillations can arise [15]–[17] . In biology , bistability has long been established in control of the cell cycle and other oscillations [18] , [19] , and also recently reported in an artificial gene regulation network [20] . In Drosophila , spatial bistability has been proposed for dorso-ventral patterning [21] , [22] . In early embryogenesis , the diffusion of Bcd protein , translated from mRNA localized at the anterior end of the egg , forms an exponential concentration gradient , establishing the anterior–posterior ( AP ) axis ( Figure 1A and 1C ) [8] , [23] , [24] . Bcd is a transcriptional regulator , and through its asymmetric distribution acts as a morphogen , governing the positions at which the downstream gap genes will be activated . In combination with cross-regulation between these genes , the initial Bcd asymmetry is propagated and refined , establishing the first stage of embryo segmentation [9] , [25]–[32] . It is still not well characterized , however , what mechanisms interpret the smooth Bcd positional information into sharp and precisely positioned downstream target gene expression . hunchback ( hb ) is one of the first gap genes activated by Bcd , with strong anterior expression and a sharp on–off boundary at mid-embryo ( Figure 1B and 1C ) [9] , [33]–[35] . Anterior hb activation depends on Bcd , as shown by Struhl et al [9] and Driever et al [34] , and on its own self-regulation , as already reported by Treisman et al [35] and Margolis et al [36]; many Bcd and Hb binding sites have been identified in the hb promoter region , as reported by Treisman et al . , among others [35]–[37] . Hb has maternal ( hbmat ) and zygotic contributions , and provides positional information for other gap genes , such as Krüppel ( Kr ) , knirps ( kni ) , and giant ( gt ) , and for the homeotic gene Ultrabithorax ( Ubx ) [38]–[41] . Removal of both maternal and zygotic hb expression results in severe deletions and polarity reversals of the most anterior segments [42] . In normal development , Hb expression drops from highest to lowest over about 10% egg length ( Figure 1B and 1C ) ; Considerable attention has been focused on what molecular mechanism generates this Hb sharpness . Early on , it was shown that a hb enhancer element of 300 base pairs ( bp ) , containing 6 Bcd binding sites , is sufficient to reproduce the regulatory activity of Bcd on hb [9] , [34] . It was shown that Bcd binds to these sites cooperatively and it was hypothesized that , due to this cooperativity , a small change in Bcd concentration across some threshold could produce a large change in hb promoter occupancy , generating the on–off expression pattern [9] , [34] , [43]–[46] . However , these studies did not establish that cooperativity is sufficient to generate Hb border sharpness . To quantify the degree of Bcd's cooperativity , Ma et al . [44] used a six-Bcd site fragment of the hb promoter in a DNase I footprint assay , and found a Hill coefficient of about 3 . 6; Burz et al . [47] , using a gel-shift assay with a 230 bp hb enhancer , found a Hill coefficient of 3 . 0 . From quantified in vivo patterns of Bcd and Hb proteins , Gregor et al . [48] , estimated a higher value for this coefficient , of around 5 ( though the effects of hb self-regulation were neglected , addressed further in the Discussion ) ; and suggested that it could support the proposal of Crauk and Dostatni [49] that Hb expression is entirely determined by Bcd cooperative binding . However , systems with such high Hill coefficients would be expected to show temperature sensitivity . Houchmandzadeh et al . [50] showed that the Bcd gradient is strongly affected by temperature changes of 20°C , but that the Hb pattern is largely unaffected . Dependence on Bcd with Hill coefficients between 3 and 5 would be expected to show far greater effects on Hb than are observed , indicating that Bcd cannot be the only factor regulating the Hb border . The insufficiency of Bcd cooperativity to produce Hb sharpness is also supported by the findings of Simpson-Brose et al . [51] , who showed that synergy between Hb and Bcd is necessary to establish the expression patterns of the gap genes , including hb itself . To address these issues , we have taken a combined experimental and theoretical approach to understand how the hb gene converts the positional information of the smooth Bcd gradient into a sharp expression pattern . We used wild-type ( WT ) embryos to experimentally determine how Hb position and sharpness change in time; and we measured how these quantities are affected in embryos mutant for Bcd cooperative binding and for hb self-regulation , and by use of an artificial promoter with 6 Bcd and 2 Hb binding sites . We also developed a predictive reaction–diffusion model of hb transcription , incorporating both Bcd cooperative binding and hb self-regulation . By fitting this model to wild-type Bcd and Hb patterns , we determined kinetic parameters of the model , such as binding constants . With these parameters , our model successfully reproduces the dynamic development of the Hb pattern . By reducing Bcd binding constants or the number of Bcd binding sites , our model reproduces the same mutant phenotypes as our experiments , and predicts a loss of sharpness for a hb self-regulation mutant , which we experimentally verified . By applying dynamical systems theory to the model , we show that Hb sharpness is due to spatial bistability stemming from hb self-regulation . This means that Hb does not have a single steady-state concentration continuously dependent on Bcd , but that at a threshold Bcd concentration , two stable steady states become available to Hb , and a small change in Bcd concentration can create a dramatic shift in Hb concentration . This shift between steady states is responsible for the sharpness of the Hb boundary . The position of the Bcd threshold is controlled by Bcd cooperative binding , but this mechanism itself is not sufficient to generate hb's expression sharpness .
In order to investigate the relative contributions of self-regulation and Bcd cooperativity to Hb sharpness and position , we performed immunohistochemistry assays in wild-type and mutant embryos in nuclear cleavage cycle 14A ( Figure 1 and Figure S1; see also Materials and Methods ) . We used systematic image processing approaches to extract gene expression patterns from confocal microscopy images , to determine embryo ages and to quantify pattern position and sharpness . The effect of loss of hb self-regulation can be directly studied with the loss of function hb14F allele [52] , which codes for a mutant protein having no DNA-binding capacity ( i . e . , no self-regulation ) . We scanned 39 embryos expressing this allele and found that embryos homozygous for the hb14F allele ( Figure 1D ) have a strong decrease in sharpness ( 21 . 8% reduction , from 80 . 2° in WT to 62 . 8° ) , and small shift in position ( 9 . 3% decrease , from 47 . 0% EL in WT to 42 . 6% EL; Table 1 summarizes sharpness and position for all experiments ) . The hb14F homozygotes were easily identified by low signal intensities [53] ( see Figure S2 for non-normalized profiles ) . Heterozygote ( hb14F/+ ) and WT embryos were not easily distinguished , and are shown together in Figure 1E . The means for this group show little or no change from WT: sharpness shows a 1 . 3% change ( from 80 . 2° to 81 . 3° ) , and position shows a small shift , from 47 . 0% EL to 44 . 3% EL ( 5 . 7% change; Figure 1E ) . In order to compare Hb patterns between different assays , in which absolute intensity varies , all experimental and theoretical profiles in Figure 1 are normalized . With this , our measure of sharpness is determined by the AP projection of the profile ( see Figure S3 ) ; this is covered in more detail in the Discussion . Note that all hb14F homozygote embryos display lower sharpness than any WT embryo older than 8 minutes in cycle 14 ( the timing of sharpness maturation is discussed further in Figure 2 and in the Discussion ) . Despite Bcd's role in AP positioning during early Drosophila development , bcd heterozygotes ( bcdE1/+ ) are highly viable embryos , in which , for example , among 593 embryos only 4% were unhatched , and no larval head defects were found [46] . In these mutants , we found that Hb was shifted anteriorly ( 16 . 6% change , from 47 . 0% EL in WT to 39 . 2% EL ) , as previously reported [8] , [50] , but sharpness was not affected ( 0 . 5% change; from 80 . 2° in WT to 80 . 6°; Figure 1F ) . Mutant bcd genes encoding proteins specifically defective in cooperative DNA binding have been isolated by Hanes and collaborators [46] , [54] , using a genetic screen in yeast . It was shown that these mutations do not disrupt the DNA recognition or transcriptional activity of Bcd . We used one of these mutants , bcdK57R , to analyze the effect of Bcd cooperative binding on Hb sharpness and position . In embryos with one dosage of the bcdK57R allele ( in a null Bcd background ) , incomplete penetrance gives two outcomes: embryos with a weak response to the Bcd defect ( Figure 1G; showing a small anterior shift compared to bcdE1/+ , from 39 . 2% EL to 37 . 6% EL ) ; and embryos with a strong response , which have a large anterior shift compared to bcdE1/+ ( Figure 1H; from 39 . 2% EL to 24 . 8% EL ) . In both cases , sharpness is not reduced ( 0 . 5% and 2 . 5% change from WT , for weak and strong mutants , respectively ) . Driever et al . [43] used a series of lacZ constructs to describe the effect of high and low affinity binding sites for the establishment of the localized zygotic expression domains . These constructs have a lacZ coding sequence attached to different fragments of the hb promoter sequence . To make flies carrying these constructs , they are introduced into the fly genome , but the WT background is maintained . In one of these constructs , pThb5 , the promoter is a portion of the native sequence , having 6 Bcd and 2 Hb binding sites ( see Figure S4 ) . These constructs do not show self-regulation , since the protein encoded by them ( β-galactosidase ) has no transcriptional activity . pThb5 expression has significantly reduced sharpness ( from 80 . 2° in WT to 72 . 7° , Table 1; see Figure S5 and Figure S6 for lacZ expression patterns ) and a slight anterior shift in position ( from 47 . 0% in WT to 45 . 7% EL ) . In addition , we found that pThb5 expression is sharper than hb14F ( 72 . 7° vs . 63 . 9° ) , indicating that Hb protein , normally expressed in these flies , can increase the sharpness of pThb5 expression ( This effect of the Hb WT expression was predicted by our model , as shown below . ) To investigate what causes the changes in positioning and sharpness shown in Figure 1 and Table 1 , we developed a predictive reaction–diffusion model . This hunchback self-regulatory ( HSR ) model captures both Bcd cooperative binding and hb self-regulation , with six Bcd sites and two Hb sites driving hb expression . This model readily reproduced the phenotypes of the WT and mutant experiments , allowing us to make predictions and gain new understanding of the molecular mechanisms producing the measured macroscopic patterns . The model reactions are summarized in Figure 3 . hb expression requires two steps: Bcd and Hb protein binding to hb promoter ( reversible reactions ( 2n , n = 1 , … , 6 ) and ( 14 , 16 ) , respectively ) ; and Hb protein synthesis ( irreversible reactions ( 1+2n , n = 1 , … , 8 ) ) . Reaction ( 1 ) represents Bcd production; reactions ( 18 ) and ( 19 ) represent Bcd and Hb degradation , respectively . We used Fick's Law to describe Bcd and Hb diffusion , and the Law of Mass Action [55] for the reactions in Figure 3 , to derive a system of coupled partial differential equations ( PDE's; see Text S1 ) for the species B , H , b0–b6 , h0–h2 ( only species B and H are allowed to diffuse ) . The dynamics of the system are completely described by initial conditions , diffusion coefficients and the kinetic constants , found by fitting model output to experimentally measured expression patterns . We used a finite difference method to solve the model PDE's ( see Text S1 ) and a steepest descent method to determine the k parameters , by fitting the total Bcd ( [B]T ) and Hb ( [H]T ) concentrations ( Text S1 , Equations S1 and S2 ) to the respective patterns of an embryo in mid-nuclear cleavage cycle 14 ( about 36 . 4 minutes into the cycle , Figure 2A; see Materials and Methods for age determination method ) . The Bcd gradient was fit first , by using a zero initial concentration and fitting the model to Bcd experimental data ( Figure 2A ) without the Hb reactions ( Figure 3 , reactions 3 , 5 , 7 , 9 , 11 , and 13–18 ) , and with Bcd production ( Figure 3 , reaction 1 ) only at 9% EL ( the position of maximum Bcd level in the data ) . The Hb reactions were then taken into account , and the model fit to the Hb experimental data ( Figure 2A ) . With the k's determined ( Table S1 ) , we simulated the dynamics of the HSR model ( Figure 2B ) , using zero initial Hb concentration ( embryos lacking maternal Hb develop normally [42]; also , final concentrations are largely unaffected by low initial [HT]—see Discussion ) . The HSR model qualitatively reproduced the time development of the Hb pattern ( Figure 2B , 59 embryos of different ages ) , even though it was fit to only one pair of Bcd-Hb patterns ( Figure 2A; see Figure S7 for a direct comparison between data and model time evolution profiles ) . The match was best with respect to sharpness increase ( Figure 2D ) ; for position , the model shifted more than observed ( Figure 2C; see also Figure S7A and S7B ) , possibly reflecting the simplified aspect of the model , like the number of Hb or Bcd sites . Figure 2 indicates that developmental age , after the transient behavior in the first 10 minutes of cycle 14 , is not a significant factor in sharpness or position . Once we established that the HSR model accounts for wild-type expression , we could analyze its dynamics to determine what is responsible for converting the smooth Bcd spatial distribution into the sharp Hb pattern . We chose Zero Eigenvalue Analysis [56]–[59] as a technique for searching parameter values that produce bistability in our model . This method establishes a set of restrictions ( the sign compatible relations , see Text S2 ) which must be met in order for bistability to occur . Using this technique on a simplified HSR model [60] , with reduced number of Bcd binding sites ( reactions 4 to 13 removed , Figure 3 ) and normal Hb binding , we demonstrated that the model does have bistable solutions ( see Text S2 ) . Although this analysis was performed for a well-mixed system , where concentrations are assumed to be uniform , the bistable behavior is also found in numerical solutions of the full model , where spatial distributions of concentrations are considered . The bifurcation diagram ( Figure 4A ) , plotting [H]T steady-state solutions for various k0 , B and [B]T , shows that for Bcd concentrations within the bistable region ( green line ) , Hb concentration has two stable solutions ( blue lines ) , separated by an unstable solution ( red line ) . This bistability causes a very large change in Hb concentration ( ΔHb , from one stable branch to the other ) over a small change in Bcd concentration ( ΔBcd ) as it passes through a threshold ( at the anterior boundary of the bistable region ) . In the Drosophila embryo , the Bcd gradient provides different concentrations along the anterior–posterior axis , which work like different initial conditions in the well-mixed system . It creates a sharp Hb border ( Figure 1C ) at the position where Bcd crosses this threshold . Comparison between hb RNA and protein profiles shows that RNA patterns are sharper than protein ones ( 84 . 2° and 82 . 5° mean sharpness , respectively; Figure S8 and Table 1 ) . These findings combine with our results above , to indicate that Hb sharpness results from spatially bistable dynamics , which is due to hb self-regulation . In the HSR model , if the hb self-regulation reactions ( Figure 3 , reactions 14–17 ) are removed , the network loses bistability and has only a single steady state ( Figure 4B ) , in which Hb varies smoothly with Bcd . With Hb bistability characterized in the model , we can proceed to simulating the macroscopic behaviors ( i . e . , expression phenotypes ) shown in Figure 1 , by altering binding strengths and site numbers , to reproduce the corresponding mutant genotype . For example , to model hb expression in the bcdE1/+ heterozygotes we reduced the Bcd source term ( k0 , B ) to 66 . 6% of original ( see Table S2 ) , finding anteriorly shifted Hb pattern ( 16 . 5% change from WT simulation ) without disrupting sharpness ( 1 . 0% change from WT ) , in agreement with data ( Figure 1F , heavy blue curve; Table 1 ) . Hb sharpness was maintained because reduction of Bcd production by this amount does not change the model's bistable phase portrait , Figure 4A , but does shift the position of the Bcd threshold anteriorly . To simulate loss of self-regulation , we removed reactions 14–17 ( Figure 3 ) from the full model ( 6B2H , for 6 Bcd and 2 Hb sites ) , to give 6B0H sites . The model predicts a loss of sharpness ( 19 . 0% change from WT; from 84 . 3° to 68 . 3°; Figure 1D , heavy blue curve , Table 1 ) in qualitative agreement with hb14F homozygote experimental data ( which showed a 20 . 3% change from WT ) . The bifurcation diagram for 6B0H sites ( Figure 4B ) shows that this loss of Hb sharpness is due to loss of bistability , since Hb concentration becomes a smoothly decreasing function of Bcd concentration . The pThb5 construct contains an estimated six active Bcd sites and two Hb sites ( Figure S4 ) , but it does not exhibit self-regulation because the protein coded by it has no transcriptional activity . To reproduce the lacZ expression of pThb5 we derived an extra set of reactions by replacing H with L ( LacZ ) in reactions ( 1+2n , n = 1 , … , 8; and 18 ) , and added these reactions to the full model; with this model ( 6B2H_lacZ ) we have no self-activation for lacZ , but we still take into account both Hb sites in the lacZ promoter . We found no shift in position , but did find an 11 . 1% loss of sharpness from the WT simulation ( Table 1 , Figure S3 ) , in agreement with the experimental loss of sharpness from WT to lacZ ( 9 . 3% ) . This indicates that the loss of sharpness for pThb5 is caused by the lack of self-regulation . The lacZ experimental patterns are sharper than hb14F ( 72 . 7° and 63 . 9° , respectively ) , probably due to the Hb sites in the construct promoter region . Our model predicts this effect , with sharpness for 6B2H_lacZ ( 74 . 9° ) higher than sharpness for 6B0H ( 68 . 3° ) . The loss of sharpness for hb14F homozygotes , Figure 1D and Table 1 , demonstrates that cooperative Bcd binding is not sufficient to generate the sharp Hb border , since Bcd cooperativity is not affected in this hb allele . Bcd cooperative binding does play an important role in Hb pattern positioning , however , as demonstrated by the bcdK57R results ( Figure 1G and 1H; Table 1; see also [43] ) , in which pattern position is altered without affecting sharpness . We simulated the reduced cooperativity in bcdK57R embryos [54] by reducing the Bcd binding constants in reactions ( 2n , n = 1 , … , 6 ) by dropping factor to 67% and 95% of original , to simulate the strong and weak mutants , respectively ( Figure 3 caption , Table S2 ) . Figure 4C shows that this reduction in cooperativity shifts the bistable region towards regions of high Bcd concentration ( Figure 4C , red and black lines ) , anteriorly shifting the pattern without disrupting its sharpness , in agreement with the data ( Figure 1G , heavy blue line; Figure 1H , heavy blue line; Table 1 ) . These results show that Bcd cooperative binding controls the position at which the Hb bistable switch occurs . Though not sufficient for sharpness , Bcd cooperativity is necessary for Hb bistability to produce sharpness . We can demonstrate this by simulating a strong decrease in cooperativity in silico by a decrease in the number of Bcd binding sites ( removing reactions 10–13 , giving 4B2H sites ) without affecting self-regulation . These simulations show a strong reduction in hb activation , giving both a strong anterior shift and a drop in sharpness ( Figure 4D ) . The small box in Figure 4D shows that bistability was not disrupted , since self-regulation was not affected , but the Bcd threshold was shifted to a very high concentration not reached by the Bcd gradient ( indicated by red line ) . This result indicates that Bcd cooperative binding is necessary for hb activation to reach its bistable threshold , which in turn is necessary for sharpness to occur . The above results show that small disruptions of Bcd cooperative binding result in positional shifts , without loss of sharpness , while large enough disruptions of cooperative binding also disrupt Hb sharpness , since the bistable switch is not reached . However , the bistable switch itself can only be produced by hb self-regulation .
The role of Bcd cooperative binding on Hb positioning has been demonstrated since 1989 [9] , [34] , [43] . For example , Driever et al . [43] used a selection of artificial lacZ constructs , each containing some portion of the native hb regulatory sequence , with varying numbers of high and low affinity Bcd binding sites . They showed that reducing the number or strength of Bcd binding sites shifted the lacZ expression anteriorly , demonstrating the role of cooperativity for pattern positioning . Increasing the binding strength or number of sites gave posterior shifts and sharper borders , suggesting that cooperativity could also be responsible for Hb sharpness . However , even for the construct with the highest number of sites ( 12; 6 strong and 6 weak ) , which showed the strongest expression level , pattern was not as sharp as wild-type Hb . While a role for cooperative binding in sharpening was suggested by these results , the authors noted this could not be firmly concluded from their data [43] . Struhl at al . [9] also observed shallower than endogenous Hb borders with a series of similar lacZ constructs . Based on their results , Driever et al . [43] proposed a ‘gradient-affinity model’ , wherein target genes with high affinity binding sites , like hb , would be efficiently expressed even at low Bcd concentrations , and target genes containing low affinity binding sites would be positioned in more anterior positions . Our model reproduces these positioning effects of Bcd cooperative binding , as shown in the simulations in which binding site strength was varied ( e . g . , Figure 1F–H ) . It also explains that the border of lacZ expression patterns is not as sharp as wild-type Hb because of the loss of self-regulation in such constructs ( Figure S5 and Figure S6 ) . These results indicate that bistability can play a role in the gradient-affinity model , since they show that changing the cooperativity level shifts the Hb pattern ( Figure 1G and 1H ) but does not change its sharpness , allowing an on–off expression boundary to be placed at many positions in the embryo . Instead of changing cooperativity by changing the binding sites in artificial constructs , Burz and Hanes [54] generated several Bcd cooperativity mutants , such as the bcdK57R used in this study . They showed that this mutant is stable in vivo ( in yeast cells ) and is not affected in its DNA recognition , nuclear entry , or transcriptional activity characteristics [46] , [54] , and in situ hybridization showed that localization and expression of bcdK57R mRNA is normal [46] . Through analyzing the expression pattern of the gap genes hb , gt and Kr in this mutant , Lebrecht et al . [46] showed that cooperative Bcd binding is critical for embryonic patterning . That study also reported a reduction in Hb sharpness , contrary to what we report here ( see Figure 1G and 1H and Table 1 ) . In [46] , sharpness ( slope ) was calculated on non-normalized data , which makes their results susceptible to artificial variations in gene expression levels that can occur at many steps in a staining assay , such as embryo fixation and permeabilization . Using non-normalized data and measuring the slope by the quotient between profile intensity , Δy , by the distance along AP axis , Δx , introduces variability in slope due to variability in Δy , see Figure S3 . Here , we present new results for Hb sharpness , computing slope with normalized data ( Δy = 1 ) to reduce the contribution of these experimental errors and to compare mutants with different levels of expression . Our method , similar to that used by Crauk and Dostatni [49] , depends only on how far it takes Hb to drop from ‘on’ to ‘off’ expression ( Δx ) . Recently , Gregor et al . [61] have shown that immunofluorescent signals are proportional to protein concentration plus a nonspecific background . This indicates that normalizing immunofluorescent signals provides an equivalent result to normalizing real concentrations . With this approach , we do not find a significant difference in sharpness between bcdK57R , bcdWT , or bcdE1/+ . Gregor et al . [48] recently presented quantitative data comparing Bcd and Hb intensities from whole embryos , and analyzed the precision of this input/output relation . They fit the Hill equation ( Equation 1 ) to the Bcd/Hb input/output relation and estimated that Bcd binds to the hb promoter with a Hill coefficient of 5; somewhat higher than the in vitro values [44] , [47] , but within the known number of Bcd binding sites [9] , [34] . However , they neglected the contribution of hb self-regulation in establishing the levels of Hb protein , so the value reported for Bcd1/2 would not produce half-maximal Hb synthesis in the absence of self-regulation ( see Text S3 for more details ) . In other words , to reach maximum Hb production without self-regulation would require a higher Bcd concentration , and a higher value for Bcd1/2 . Calling this corrected value Bcd1/2Cor , Bcd1/2Cor>Bcd1/2 , which means that ln ( Bcd/Bcd1/2Cor ) <ln ( Bcd/Bcd1/2 ) , and using Equations 3 and 4 shows that nCor>n ( see Text S3 for derivation of these equations ) . This indicates that a corrected Hill coefficient ( nCor ) should be higher than that reported by Gregor at al . and likely higher than the six Bcd binding sites known for hb regulation [9] , [34] , making a claim that Bcd cooperativity determines Hb sharpness unlikely . In [48] it was argued that the effect of additional factors , uncorrelated with Bcd , would require hb readout of Bcd concentration to be even more reliable than reported . However , this argument does not justify neglecting hb self-regulation for establishing levels of Hb protein , because Hb precision ( measured by standard deviation ) is not correlated with Hb protein levels , as shown in [48] . In other words , hb self-regulation may not influence the precision of the readout process , but it does determine protein levels , and this role cannot be neglected in calculating the Hill coefficient . ( 1 ) ( 2 ) ( 3 ) ( 4 ) Our HSR model , though fit to WT data , predicts the loss of sharpness we found experimentally in the self-regulatory mutant , hb14F ( Figure 1D and Figure S1 ) . The experimental results directly support the need for hb self-regulation for sharp pattern development . hb14F is a lack of function mutant , generated by Lehmann et al . [42] , which forms normal mRNA [51] but has a truncated protein with no DNA binding capacity . The protein is stable , persisting into central nervous system development [62] , and has been visualized with Hb antibody [53] staining at lower intensity than WT ( Figure S2; we measured intensity at 10–20% WT , comparable to the 10% reported in [53] ) . Since the Bcd protein or its binding are not affected in these mutants , these embryos clearly show that Bcd cooperative binding is not sufficient for producing Hb sharpness . Our results with hb14F agree with the observations of Houchmandzadeh et al . [50] that expression in the hb6N allele ( also with non-functional protein ) suggested a role for self-regulation in sharpening . Our model shows that loss of self-regulation disrupts the bistable behavior in hb14F expression , resulting in the loss of sharpness . Similarly , expression of the pThb5 lacZ construct shows reduced sharpness in comparison to WT ( Figure S5 , Figure S6 , and Table 1 ) , since the protein coded by lacZ ( β-galactosidase ) is not self-regulatory . Construct sharpness is greater than hb14F sharpness , however . The model predicts this , by taking into account that native , patterned Hb protein can bind in the construct promoter ( compare Figure 1D to Figure S5 and Figure S6; see also Table 1 and Figure S4 ) . Crauk and Dostatni [49] recently reported sharp expression for a lacZ construct containing only three ( strong ) Bcd binding sites . We found lowered sharpness for the pThb5 construct ( Figure 1H ) , which contains six Bcd binding sites ( three strong , three weak; as well as two Hb and Kr sites ) . Since this is opposite to what any cooperative effect should be for increasing sites , the differences are likely to be methodological . We used both whole mount fluorescent in situ hybridization ( FISH; Figure S5 ) and traditional in situ hybridization ( Figure S6; as used in [49] ) to visualize lacZ expression . Both methods gave similar measures of sharpness ( Table 1 ) , but the enzymatic staining is more susceptible to signal saturation and tends not to be proportional to RNA concentration . Crauk and Dostatni [49] also reported reduced sharpness in embryos with truncated Bcd proteins , Bcd-ΔC and Bcd-ΔQC ( with specific defects in protein activity ) . In light of the hb transcriptional dynamics found in our analysis , we believe such alterations to Bcd could cause transcription to remain sub-threshold for bistable activation , similar to Figure 4D . All embryos used in Figure 1 and Table 1 were in nuclear cleavage cycle 14A ( precellular blastoderm ) , within time classes T5 to T6 ( 26 to 39 min into this cycle ) , during which Hb levels are at their highest ( Figure 1D shows two T7 embryos , to show the normal posterior patterning in these mutants ) . We staged each embryo by established methods [63] , following dorsal membrane invagination measured from images obtained by Differential Interference Contrast ( DIC ) optics . For comparing Hb dynamics and model simulation ( Figure 2 ) , we used embryos in the first 36 . 4 minutes of cycle 14A . While Hb expression dynamically amplifies over this period ( Figure 2B ) , the mature sharpness is reached within 5–10 minutes , after which it is stable ( Figure 2D ) . Note that Figure 2D shows that Hb pattern in all WT embryos older than 8 min in cycle 14 are sharper than any embryo in Figure 1D . The earliest Hb protein pattern in the embryo is of maternal origin . Before nuclear cleavage cycle 8 , maternal hb mRNA is distributed uniformly throughout the egg , but its translation is repressed by the posterior Nanos ( Nos ) protein gradient , resulting in a smooth anterior gradient of maternal Hb ( Hbmat ) protein . This is gradually substituted by the zygotically expressed Hb , starting in cycle 11 [27] , [33] , [64]–[69] . To see the effect of these early Hb distributions on cycle 14 dynamics , we ran HSR model simulations with an initial Hb pattern taken from cycle 13 data ( Figure S9; parameters and data from same assay as in Figures 1C and 2A ) . Simulation results were the same as in Figure 2 , indicating no effect from the Hb initial condition . This behavior is related to the bistable behavior of Hb: the diagram in Figure S10 shows that , inside the bistable region , only relatively high initial conditions ( above the unstable branch ) can produce high Hb concentrations . It indicates that the initial concentrations of Hb , determined by zygotic production , are low and not sufficient to carry the system through the transition from lower to upper stable branch . This agrees with previous results showing that embryos lacking maternal Hb develop normally [42] , [65] , [66] . Like the Drosophila embryos , the HSR model is robust to variability in Hb initial concentration . The method we have used to construct the HSR network , describing hb regulation by Bcd cooperative binding and hb self-regulation , can be readily applied to other genetic regulatory systems in Drosophila or other organisms , since the regulatory interactions are general . We avoided using a Hill kinetics approach to model cooperativity because this would require some questionable assumptions , such as all six sites being equal , which is counter to published Bcd binding data [43] , [47] , and bound simultaneously , which is highly improbable . One advantage of using the Hill equation could be its few number of parameters; however , using relations kb ( n-1 ) , bn = factorn . kb0 , b1 and kbn , b ( n-1 ) = kb1 , b0 for n = 2 , … , 6 in reactions ( 2+2n , n = 1 , … , 5 ) allowed us to describe cooperativity with just three parameters , kb0 , b1 , kb1 , b0 and factor . The effects of more or less binding sites and more or less transcriptional regulators can easily be built into our kind of model . Our method allows for a direct link between macroscopic pattern formation and its molecular basis . As well , such a model is amenable to mathematical analysis with modern nonlinear techniques , which have developed rapidly in recent years [17] , [55] , [70]–[73] . In the present example , using such techniques to search for multiple steady states allowed us to identify the bistability inherent in the self-regulatory reactions , and determine the model parameters necessary for triggering this . In reaction networks , bistability is frequently verified by changing the initial concentration of one species , in the well-mixed system , and checking the concentrations of all other species when the system reaches the stationary state . In a monostable regime , small variations in the initial concentration generally produce small variations in the stationary state . However , if the concentration is in the vicinity of a threshold , where the transition from the monostable to the bistable regimes occurs , small changes in the initial condition can produce large variation in the stationary state , because the concentrations of the species can follow a completely different trajectory ( i . e . , sequence of intermediate concentrations ) , ending up in a very different stationary state ( Figure 4A ) . In our spatially-patterned case , the anterior–posterior Bcd gradient provides many different concentrations that work like different initial conditions in the well-mixed system; in such a way that at the position where Bcd crosses the threshold ( Figure 4A ) , the Hb stationary concentration changes abruptly , producing the sharp Hb border ( Figure 2A ) . The origin of bistability in the Bcd-Hb system is a consequence of the positive feedback of hb self-regulation . If Hb production is not high enough , self-regulation can only produce a small change in Hb production , and the consequent increase in degradation counteracts almost all increase in Hb production; this regime occurs in the posterior half of the embryo . If Hb production is more effectively increased , by increasing Bcd concentration , the positive feedback can produce a certain additional amount of Hb protein , which can be sufficient to start increasing Hb production more efficiently . If this occurs , the additional amount of Hb will increase the feedback even more strongly , ending up in a completely different regime , having higher Hb concentrations; this regime occurs in the anterior half of the embryo . Our data and model show that positioning and sharpness of the Hb pattern are separable processes . With the hb14F allele and the pThb5 construct , we show that sharpness can be disrupted with self-regulation defects; and our theoretical analysis suggests this is due to loss of bistability . Earlier work has suggested many of the shifting and sharpening effects we find here . However , there has been debate about the relative roles of the transcriptional regulators: some studies have suggested a role for hb self-regulation in sharpening [50] , while others indicate that it could be completely controlled by Bcd [48] , [49] . It has also been known that the number of Bcd binding sites in the hb promoter affects pattern position [9] , [34] , [43] . Our data and model offer a synthesis: positioning is largely dependent on the Bcd occupation states of the hb promoter , but sharpening is a result of bistability in the hb activation dynamics , caused by hb self-regulation . Bcd cooperativity , through affecting hb transcription , determines the threshold at which bistability occurs , but is not itself sufficient for sharpening . In 1977 , Lewis at al . [74] used theoretical arguments to suggest that bistable control can account for the interpretation of gradients in positional information . More recently , bistability has been found in many complex biological processes [14]–[16] , [18]–[21] , [75] and spatial bistability has been proposed in dorso-ventral patterning in Drosophila [21] , [22] . Here , we have combined experiments , modeling and analysis to suggest that this dynamic feature may also be central to AP patterning , and that for hb transcription bistability arises from the convergence of two regulatory mechanisms ( Bcd cooperative binding and hb self-regulation ) . This provides a specific mechanism to the earlier indication that Bcd and Hb synergy is required for Drosophila gap patterning [51] . Moreover , in agreement with Lewis et al . [74] , our findings indicate that bistability may be central to threshold-dependent reading mechanisms of the positional information established by smooth maternal signals . Our approach , of developing a kinetic transcriptional model from molecular data such as binding sites and regulatory interactions ( repression or activation ) , using dynamical systems theory to determine the model dynamics , and confirming the model predictions against quantitative experiments , could be used for uncovering regulatory mechanisms in many other pattern formation systems , in fruit flies and in other organisms .
We stained for Bcd and Hb proteins in WT Oregon-R embryos , as well as in the hb mutant hb14F [52] , and two bcd mutants [46] ( bcdK57R , bcdE1/+ ) . lacZ expression for the pThb5 construct ( driven by a fragment of the hb promoter; [43] ) was visualized by two methods for staining β-galactosidase mRNA ( Table 2 ) . The simultaneous Hb protein and RNA visualization was also done in WT Oregon-R embryos . As outlined in Table 2 , three different staining procedures were used for obtaining expression patterns . For all procedures , embryos were dechorionated; heat fixed in NaCl 0 . 7%+Triton-X100 0 . 05% for 3 seconds and immediately chilled in ice; and devitellinized with methanol . For protein staining [76] , embryos were incubated with guinea pig and rat primary antibodies to Hb and Bcd , respectively , followed by secondary antibodies labeled with Alexa Fluor 647 ( to Hb ) and 488 ( to Bcd; Molecular Probes ) . All antibody incubations and washes were done in PBS+0 . 1% Tween-20 . Blocking was done in Western Blocking Reagent ( Roche ) , diluted 5 times . All secondary antibodies were preabsorbed by incubating them with 0- to 12-h-old WT embryos for at least 2 h at 4 C . For the lacZ embryos , we used simultaneous immunostaining to Hb and Bcd and in situ hybridization . With FISH , we followed the method of Janssens et al . [76]: a lacZ riboprobe was prepared with a 2 . 5-kb PvuII lacZ fragment blunt-cloned into the EcoRV site of pBluescriptIIKS+ ( gift from S . Small ) , labeled with fluorescein by transcription using T3 polymerase . After hybridization , lacZ mRNA was visualized by sequential incubation with rabbit antibody to fluorescein ( Molecular Probes ) , followed by antibody to rabbit labeled with Alexa Fluor 488 ( Molecular Probes ) . The embryos were simultaneously stained for Hb and Bcd proteins , as in the previous paragraph , using secondary antibody labeled with Alexa Fluor 555 to detected Bcd . Alternately ( Figure S6 ) , some lacZ embryos were measured via enzymatic staining ( whole mount in situ hybridization ) : β-galactosidase mRNA was hybridized in situ with a digoxygenin-labeled DNA probe , following standard protocols [67] . The hybridization products were detected with a phosphatase-coupled antibody against digoxygenin . For simultaneous determination of Hb protein and RNA , we used the same FISH procedure as above , sequentially using guinea pig and rabbit antibodies to Hb and fluorescein , respectively , and secondary antibodies to guinea pig and rabbit labeled with Alexa 647 and 488 , respectively . Following fixation and staining , embryos were mounted in 40 ml mounting medium ( Prolong Antifade by Invitrogen ) and covered with a 22×30 mm cover glass ( No . 1½ ) . Following the methods of Janssens et al . [76] , whole-embryo images were taken using a laser confocal scanning microscope ( Leica TCS SP2 ) . Images were collected using an HC PL APO 20× objective and variable digital zoom ( 1 . 2–1 . 5× ) . Fluorophores were excited by laser at different wavelengths ( 488 , 555 , and 647 nm ) , and detected via a filterless spectral separation system . Channels were scanned sequentially . To reduce image noise from the photomultiplier tubes , each embryo was scanned sequentially 16 times and the results averaged . The settings of the microscope were adjusted for each gene product such that pixels expressed at maximum intensity were 255 on the 8-bit scale . Initial image size before processing was 1024×1024 pixels . Raw images were averaged , cropped and rotated . This standardization allowed us to compare levels of gene expression at different times , or in different experiments performed on different days [77] . For embryos triply-stained for segmentation proteins , the extraction of AP intensity profiles is well established [76] , [77] . With such data , a nuclear mask can be created , and intensity data mapped to nuclei ( next section ) . Co-staining for Bcd and Hb proteins and β-galactosidase mRNA presents greater challenges: signal strength and quality are very different for proteins and RNA; and the anterior localization of Bcd and Hb make identification of posterior nuclei very difficult . We developed a non-mask method for profile extraction for these experiments ( section after next ) . For embryos stained for three segmentation proteins , the three images are used to generate a ‘pixel maximum’ image , of the brightest pixels among the images . On this image , pixels are then classified as belonging to a nucleus or not , by edge-detection of bright nuclei against dark background . An error-correction step repairs any ‘fused’ nuclei . With the resulting nuclear mask , dorso-ventral , AP coordinates , and average fluorescence level of the three gene products can be mapped to individual nuclei . Intensity profiles are extracted from a central 10% strip of nuclei along the AP axis [76] , [77] . For lacZ embryos , co-stained for Hb and Bcd proteins and β-galactosidase mRNA , nuclei cannot be reliably identified , especially in the posterior ( preliminary nuclear staining in a fourth channel shows much crosstalk ) . For these experiments , we directly extract the pixel intensities in a 10% strip ( corresponding area to above ) . For high-intensity protein staining the signal is strong , but for low-intensity RNA staining we must recover expression from a noisy signal ( next section ) . A one dimensional ( 1D ) AP profile was created from the strips , by averaging intensities in each DV pixel column from the central 10% strip along the AP axis . In addition to some between-pixel noise , the resulting profiles show noise in nuclear order and in the distribution of stained material between nuclei and cytoplasm . Minimization of these two sources of noise is described in the next section . To test the quality of our direct method , we manually made nuclear masks for several co-stained lacZ embryos using the multiple ROI feature in ImageJ software [78] . ROIs are circles with radii comparable to the nuclear radii in a given image . Each ROI was positioned manually to outline a given nucleus . Nuclear-resolution AP profiles from this method are of comparable quality to pixel-resolution profiles from our direct extraction method . Noise in intensity profiles can influence model-fitting and statistical analysis of expression patterns [77] , [78] . To obtain clear expression patterns , we used singular spectrum analysis ( SSA [79] ) , a non-parametric technique with an adaptive filter . This allowed us to remove experimental ( e . g . , photomultiplier tube ) noise and noise due to variability in nuclear order and in nuclear-cytoplasmic distribution of gene products . We used the methods of Golyandina et al [80] , and software developed by Nina Golyandina and Theodore Alexandrov [81] . Non-specific binding of antibodies to biological material results in background fluorescence in our images . For triple-stained protein images it has been shown [82] that this background is a paraboloid . For every image we calculated the parameters of this paraboloid from regions of the embryo in which a particular gene is not expressed , then transformed original fluorescence at or below this background to zero . For lacZ embryos simultaneously stained for protein and mRNA it is unclear whether background has a comparable shape; in these cases , we use a simple flat background , subtracting the minimum raw intensity off all values . An advantage of direct image processing is the large number of data points ( around 1000 ) and smoothness of each profile . This makes it possible to apply standard calculus techniques to characterize the profiles: we define the Hb domain border as the inflection point , and sharpness as the first derivative at that position . With normalized intensity data ( 0–100% scale ) , this slope can be expressed as an angle of inclination ( as in Figure 2 ) . These techniques can be applied to data , as well as HSR model results . In addition to confocal scanning , all embryos were observed along the dorsal edge with Differential Interference Contrast ( DIC ) optics . Distances were measured from the egg surface to the invaginating membrane , and from the surface to the cortex . The ratio of membrane depth to cortex depth was used to estimate embryo age in minutes , using a published standard curve [63] . The Zero Eigenvalue Analysis [56]–[58] is a very efficient method , because the search for bistability is reduced to the solution of a system of equalities and inequalities ( see Equation S2 . 22 in Text S2 ) that are easier to find than a direct solution of the polynomial equation describing the stationary states ( see Equations S1 . 3′–9′ , S1 . 15′ , and S1 . 16′ ) . This technique readily allows one to find the set of kinetic parameters that produce bistability , and gives two steady state solutions , which can be used to easily make the bifurcation diagram , like that shown in Figure 4 . Finding bistability with direct solution of a polynomial requires solutions that are different , real and positive . This is frequently not convenient for degree higher than 2 [83] , and not analytically solvable for degree higher than 4 . Zero Eigenvalue Analysis can be applied to such higher degree systems . For example , Li [58] has used this method to determine multiplicity of stationary states in the famous Goldbeter and Lefever allosteric model [84] , consisting of 14 species , 32 reactions , and 27 kinetic constants . | Pattern formation during embryonic development , or morphogenesis , is one of the most intriguing problems in biology , entailing the sequence of processes by which a relatively simple system , the fertilized egg , becomes a mature organism . In these processes , the genetic information , stored at the molecular scale in the DNA , is translated into the macroscopic spatial expression patterns that precede the tissue–organ scale of body organization . It can also be understood as a flux of information from the genetic to the organ–system level . In the fruit fly Drosophila melanogaster , one of the early processes during its embryonic development is the formation of the sharp Hunchback protein pattern . To generate this pattern , the hunchback gene interprets the position-dependent information in the shallow maternal Bicoid gradient and converts it into the sharp Hunchback protein pattern . We propose that bistability in the dynamics of hunchback gene regulation can account for this information reading process , and we show that this bistable mechanism can be produced by the ability of this gene to regulate its own expression . The solution of this problem offers new approaches to understand the phenomenon of morphogenesis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"physics/interdisciplinary",
"physics",
"computational",
"biology/transcriptional",
"regulation",
"cell",
"biology/morphogenesis",
"and",
"cell",
"biology",
"mathematics",
"biophysics/theory",
"and",
"simulation",
"developmental",
"biology/pattern",
"formation",
"biochemistry/theory",
"and",
"simulation",
"computational",
"biology/systems",
"biology"
]
| 2008 | Spatial Bistability Generates hunchback Expression Sharpness in the Drosophila Embryo |
Leishmaniasis is a debilitating disease caused by the parasite Leishmania . There is extensive clinical polymorphism , including variable responsiveness to treatment . We study Leishmania donovani parasites isolated from visceral leishmaniasis patients in Nepal that responded differently to antimonial treatment due to differing intrinsic drug sensitivity of the parasites . Here , we present a proof-of-principle study in which we applied a metabolomics pipeline specifically developed for L . donovani to characterize the global metabolic differences between antimonial-sensitive and antimonial-resistant L . donovani isolates . Clones of drug-sensitive and drug-resistant parasite isolates from clinical samples were cultured in vitro and harvested for metabolomics analysis . The relative abundance of 340 metabolites was determined by ZIC-HILIC chromatography coupled to LTQ-Orbitrap mass spectrometry . Our measurements cover approximately 20% of the predicted core metabolome of Leishmania and additionally detected a large number of lipids . Drug-sensitive and drug-resistant parasites showed distinct metabolic profiles , and unsupervised clustering and principal component analysis clearly distinguished the two phenotypes . For 100 metabolites , the detected intensity differed more than three-fold between the 2 phenotypes . Many of these were in specific areas of lipid metabolism , suggesting that the membrane composition of the drug-resistant parasites is extensively modified . Untargeted metabolomics has been applied on clinical Leishmania isolates to uncover major metabolic differences between drug-sensitive and drug-resistant isolates . The identified major differences provide novel insights into the mechanisms involved in resistance to antimonial drugs , and facilitate investigations using targeted approaches to unravel the key changes mediating drug resistance .
Health professionals are constantly challenged with the clinical polymorphism of infectious diseases . Pathogen diversity is known to play a major role in this clinically observed variability in disease manifestation , severity and drug response . However , to obtain a greater understanding of this relationship there is a need for in-depth characterisation of the diversity existing in endemic pathogen populations . We believe that metabolomics is a powerful tool for studying such phenotypic diversity at the molecular level [1] . The advent of ultra-high mass accuracy mass-spectrometers heralded a new era in the analyses of metabolomes . This technology permits identification with a high level of confidence of low molecular weight analytes present in complex metabolite extracts [2] and thus has great potential in the unveiling of the metabolic fingerprints marking various pathogen phenotypes [1] . In this study we put our hypothesis to the test and applied a metabolomic approach to characterise clinical isolates of the parasite Leishmania donovani with different sensitivity to the antileishmanial drug sodium stibogluconate . Leishmania donovani is the causative agent of the infectious disease visceral leishmaniasis ( also known as kala-azar ) , which is lethal if not treated [3] . Pentavalent antimonials such as sodium stibogluconate were for long used as the first-line treatment for leishmaniasis worldwide [4] . However , use of this drug was recently officially discontinued in the Indian subcontinent due to widespread resistance of the parasite to the antimonials , resulting in treatment failure in up to 60% of the patients [5] , [6] . Clinical use of replacement drugs like Miltefosine could be less successful than anticipated , as their mode of action may be hampered or challenged by some of the unknown molecular adaptations present in antimonial resistant Leishmania populations [7] . Furthermore , screening for resistance to antimonials in endemic regions has been hindered as no molecular detection tools could be developed and validated [4] , [8] . Hence there is an urgent need from a biological , clinical and epidemiological perspective to ( i ) characterise the molecular mechanisms underlying drug resistant phenotypes present in endemic parasite populations , and ( ii ) identify biomarkers of Leishmania drug-resistance . We explored in this study if metabolomics is an adequate approach to address these research needs . This paper presents a proof-of-principle untargeted metabolome comparison of clinical L . donovani isolates with different antimonial sensitivity analysed with LTQ-Orbitrap mass spectrometry coupled to ZIC-HILIC chromatography . The untargeted nature of the study guarantees that we get a general overview of metabolic variability , rather than focusing on a preselected set of target metabolites . The results show that there are indeed numerous metabolic differences between the drug-sensitive and resistant isolates and thus illustrate how metabolomic approaches offer a unique potential to characterise diversity in a natural population of a major pathogen .
Written informed consent was obtained from the patients and in case of children from the parents or guardians . Ethical clearance was obtained from the institutional review boards of the Nepal Health Research Council , Kathmandu , Nepal and the Institute of Tropical Medicine , Antwerp , Belgium . The L . donovani isolates MHOM/NP/02/BPK282/0 and MHOM/NP/03/BPK275/0 were obtained from bone marrow aspirates taken before treatment from confirmed visceral leishmaniasis patients recruited at the B . P . Koirala Institute of Health Sciences ( BPKIHS ) , Dharan , Nepal , as described by Rijal et al . [9] . The patients received a full supervised course of Sodium Antimony Gluconate ( SAG ) ( Albert David Ltd , Kolkata ) treatment of 20 mg/kg/day i . m . for 30 days in the BPKIHS hospital . The patients were followed up for clinical and parasitological evaluation at the end of the 1-month drug course , as well as 3 , 6 and 12 months after the start of treatment . Definite cure was defined as a patient with initial cure who showed no signs and symptoms of relapse at the 12-months follow-up visit . Non-responders were defined as patients with positive parasitology after a full 30-day SAG drug course . Two clinical isolates , one antimonial-sensitive BPK282/0 and one antimonial-resistant BPK275/0 , were selected for this study and were identified as L . donovani based on a CPB PCR-RFLP assay [10] . Both isolates belong to the same genomic subpopulation which is circulating in most leishmaniasis endemic regions in Nepal [11] . The two isolates were cloned using the micro-drop method [12] , in order to obtain homogenous working parasite populations . Two sensitive ( BPK282/0 ) and three resistant ( BPK275/0 ) cloned parasite populations ( further called clones ) were obtained and used for further analysis . The in vitro antimonial susceptibility of the two parasite isolates and the corresponding five clonal populations was tested as described in our previous studies [9] . Although the derived clonal populations were found to have very similar drug sensitivity as the respective original parasite isolates ( see Table 1 ) , that does not preclude that the different clones of each parasite isolate differ in other characteristics . Leishmania promastigotes were grown on modified Eagle's medium ( Invitrogen ) [13] supplemented with 20% ( v/v ) heat inactivated foetal calf serum ( PAA Laboratories GmbH , Linz , Austria ) pH 7 . 5 at 26°C . The cultures were initiated by inoculating day 3–4 stationary phase parasites in 20 mL culture medium to a final concentration of 5×105 parasites/mL; the resulting inoculated medium was equally distributed over 4 culture flasks . The four independently growing cultures of each parasite clone were further treated as biological replicates . The 5 different clones were grown synchronically with growth monitored by daily counting; the different clones were all harvested on day 3 of stationary growth phase for metabolite extraction . Day-3 stationary phase parasites were shown in pilot experiments to be the most reproducible source of metabolites , The differences in growth rate of the clones used in this study were relatively minor . The metabolite extraction protocol consists of ( a ) quenching ( <20 sec ) of L . donovani promastigotes in their culture flasks to 0°C in a bath containing a mixture of dry ice/ethanol , ( b ) aliquoting the necessary volume for harvesting 4×107 parasites , ( c ) triplicate washing of parasite cells in 1 ml of cold ( 0°C ) phosphate buffered saline ( PBS; pH 7 . 4 – Invitrogen ) by centrifugation ( 20 , 800× g , 0°C , 3 min ) and re-suspending cells using a vortex , ( d ) cell disruption and metabolite extraction of the washed cell pellet in 200 µl chloroform/methanol/water 20/60/20 ( v/v/v ) during one hour in a Thermomixer ( 1400 rpm , 4°C – Eppendorf AG , Hamburg , Germany ) , ( e ) separating the metabolite extract from cell debris by centrifugation ( 20 , 800× g , 0°C , 3 min ) and ( f ) deoxygenating the extracts with a gentle stream of nitrogen gas for 1 min prior to tube/vial closure . Vials were stored at −70°C and analysed within 48 hrs . Formic acid ( ULC grade ) , acetonitrile ( ULC grade ) , water ( ULC grade ) , methanol ( ULC grade ) and chloroform ( HPLC-S grade ) were purchased from Biosolve ( Valkenswaard , The Netherlands ) . The ZIC®-HILIC PEEK Fitting Guard column ( 15 mm×1 . 0 mm; 5 µm ) and ZIC®-HILIC PEEK HPLC column ( 150 mm×2 . 1 mm; 3 . 5 µm ) were obtained from HiChrom ( Reading , UK ) . Gradient elution was performed using a Surveyor HPLC pump ( Thermo Fisher Scientific Inc . , Hemel Hempstead , UK ) . Elution of the ZIC-HILIC columns was carried out with a gradient of ( A ) 0 . 1% formic acid in acetonitrile; ( B ) 0 . 1% formic acid in water . The flow rate was 100 µl/min , with an injection volume of 5 µl . Gradient elution chromatography was always performed starting with 80% solvent A . Within a 6 min time interval , solvent B was increased to 40% and maintained for 12 min , followed by an increase to 90% within 4 min . This composition was maintained for 2 min , after which the system returned to the initial solvent composition in 2 min . The whole system was allowed to re-equilibrate under these conditions for 14 min . High-resolution mass measurements were obtained with a Finnigan LTQ-Orbitrap mass spectrometer ( Thermo Fisher Scientific Inc . , Hemel Hempstead , UK ) . Optimal LTQ-Orbitrap parameters were based on previous results [14]–[16] . Briefly , the instrument was operated in both positive and negative ion electrospray mode . ESI source voltage was optimized to 4 . 0 kV and capillary voltage was set to 30 V . The source temperature was set to 250°C and the sheath and auxiliary gas flow rates were set respectively to 30 and 10 ( machine-specific units ) . Full-scan spectra were acquired over an m/z-range of 50–1000 Da , with the mass resolution set to 30 , 000 FWHM . All spectra were collected in continuous single MS mode . The LC-MS system was controlled by Xcalibur version 2 . 0 ( Thermo Fisher Scientific Inc . , Hemel Hempstead , UK ) . Raw data files acquired from analyzed samples were converted into the mzXML format by the readw . exe utility ( a tool of the Trans-Proteomic Pipeline software collection , downloaded from http://tools . proteomecenter . org/wiki/index . php ? title=Software:ReAdW ) . Further processing was handled by a flexible data processing pipeline mzMatch [17] ( http://mzmatch . sourceforge . net/ ) , performing signal detection [18] , retention time alignment [19] , blank removal , noise removal [20] , and signal matching . In order to minimize the effects of biological and technical variation , the normalization procedure of Vandesompele et al . [21] was applied . This approach detects the signals of housekeeping metabolites , such as amino acids , and scales the data according to the variation found for those metabolites . Masses whose abundance was not reproducible for all biological replicates , as indicated by a Relative Standard Deviation ( RSD ) larger than 35% , were discarded , as quantification is expected to be at least 20% accurate over multiple runs [22] . Derivative signals ( isotopes , adducts , dimers and fragments ) were automatically annotated by correlation analysis on both signal shape and intensity pattern [23] . The derivative signals were removed before further statistical tests , as they would give excessive weight to abundant analytes with many derivatives . The selected mass chromatograms were putatively identified by matching the masses ( mass accuracy <1 ppm ) progressively to those from metabolite-specific databases . In a first round of identification , LeishCyc [24] , LipidMAPS [25] , and a contaminant database were used [26] . The latter allows removal of typical impurities and buffer components often detected in metabolomics experiments . The putative identifications for the lipids were manually annotated with the total number of carbons and double bonds in the side-chains . Only the remaining unidentified peak went through a second round of matching to KEGG [27] and a peptide database; and finally a third round was done with the Human Metabolome Database for any remaining unidentified analytes [28] . This iterative process was used in order to restrict the number of potential matches to the most likely [29] . Metabolite identification was aided by MS fragment interpretation and retention time matching to metabolite standards [15] . Statistical analysis and graphical routines were handled in R ( http://www . R-project . org ) . Unsupervised hierarchical clustering analysis ( HCA ) and principal component analysis ( PCA ) are used to identify groups of samples that behave similarly or show similar characteristics . Hierarchical clustering algorithms build an entire tree of nested clusters out of objects in the dataset by an iterative clustering algorithm [30] . Principal component analysis ( PCA ) is an unsupervised multivariate analysis technique frequently used in metabolomics [31] . It implements a data dimensionality reduction of complex data matrices , so that clustering tendencies , trends and outliers can be visualized among samples . Rank products ( Bioconductor RankProd Package [32] ) is a non-parametric statistical method used to detect metabolites with significantly differential abundance in the two phenotypes studied [33] , [34] . The R code consisting of reading and writing routines of data from/to PeakML file format ( XML representation of processed data produced by the mzMatch pipeline ) is available from the authors upon request .
Two parasite isolates were selected for this study; we derived two clones from the drug-sensitive clinical isolate and three clones from the drug-resistant clinical isolate for metabolic analysis ( Table 1 ) . The documented genetic homogeneity of the L . donovani population in the Indian subcontinent [35] indicates that the isolates are genetically very similar , maximizing the chances that any observed metabolic differences are related to the relative sensitivity to the antimonial drugs . Mass spectrometry analysis of the metabolite extracts ( 4 biological replicates for each clone ) yielded 71 , 000–73 , 000 regions of interest ( mass spectrometry signals or potential peaks ) per extract for positive electrospray ionisation ( ESI ) mode and 56 , 000–61 , 000 for negative ESI mode . Automatic detection of irreproducible and/or noise regions , as described in Materials and Methods , removed between 91–95% of the regions ( i . e . non-reproducible and/or masses not producing a clear chromatographic peak ) , leaving a total of 4143 chromatographic peaks for positive mode and 4656 chromatographic peaks for negative mode as candidate biological analytes . Only 15–18% of these automatically extracted signals matched a compound of the selected metabolite databases ( 324 and 237 matches for positive and negative mode , respectively , using a mass accuracy <1 parts-per-million or ppm ) . The likelihood of the validity of the database hits was further assessed by manually verifying for each peak whether the retention time and mass spectrum fragment profile matched the chemical nature of the corresponding database hit . We accepted the metabolite identifications for 256 and 185 peaks from positive and negative mode respectively . Many of these metabolites ( 101 ) were present in both electrospray ionisation modes , in which case we selected the ionisation mode with the best quality signal ( according to peak shape and signal intensity ) . Finally , a list of 340 compounds for which we had strong confidence of the identification being correct , was created . Table S1 gives this list of all the metabolites putatively identified together with the detected abundance in each sample and the Rank Product statistical analysis used to identify significant differential abundance of metabolites between the two isolates with differing drug sensitivities . The largest class of metabolites identified is the lipids ( 116 glycerophospholipids , 18 sphingolipids , 9 glycerolipids , 9 sterol/prenol lipids ) , primarily eluting at an early chromatographic time-point as expected for HILIC chromatography . The next largest class is amino acids and their derivatives ( 40 amino acids , 49 amino acid derivatives subdivided in acylglycines , polypeptides and thiol compounds ) . Other metabolite classes detected include carbohydrates ( 21 ) , fatty acyls ( 26 ) , purines/pyrimidines and their conjugates ( 26 ) , polyamines ( 3 ) vitamins and cofactors ( 10 ) and organic acids ( 9 ) . Our total coverage is approximately 20% of the predicted core Leishmania metabolome ( about 600 metabolites , excluding lipids; [36] ) , thus exceeding the number reported in previous untargeted metabolomic studies [37] , [38] . The coverage over the various metabolic pathways is visualised on the L . donovani metabolic network in Figure 1 , which shows 163 of the 340 identified compounds . Unsupervised hierarchical clustering ( Figure 2 ) of the samples ( shown on x-axis ) revealed that the metabolite abundance profiles of the drug-resistant and -sensitive clones differ sufficiently that they can be distinguished clearly and robustly . The 4 biological replicates from the individual clones are also correctly clustered together . Clustering of the metabolites ( shown on the y-axis ) reveals several large groups of metabolites that are either significantly higher or lower in the drug-resistant compared with the drug-sensitive clones . The results of the hierarchical clustering are confirmed in a principal component analysis as shown in Figure 3 . Principal component analysis is a mathematical method to project a multidimensional dataset onto a smaller number of dimensions -principal components- which explain the maximum of variation in the data and thus enables the visualization of the major differences between samples . Clones of the drug-resistant and -sensitive isolate are clearly separated on the first principal component ( explaining 61 . 8% of the total variance ) , while the second principal component separates the different clonal populations ( explaining 8 . 9% of the total variance ) . We only considered a metabolite to have a significantly differential profile in drug-sensitive and resistant clones when ( i ) there was a statistically significant differential abundance in the samples from the two phenotypes ( Rank Product P-value <0 . 05 ) , ( ii ) there was at least a 3-fold difference in average signal intensity between the two groups of samples , and ( iii ) the metabolite was consistently detected in all replicate samples of either all the drug-sensitive or all the drug-resistant clones . Using these criteria , we identified 100 ( 29 . 6% of those detected ) compounds that differed between the samples of the two phenotypes . About half ( 51 ) of those compounds had a significant higher signal in drug-sensitive clones while the other half ( 49 ) had a higher signal in drug-resistant clones . The metabolites shown to differ in the two phenotypes participate in a variety of metabolic pathways , many related to sphingolipid , phospholipid , amino acid and purine/pyrimidine metabolism . Figure 4 shows the distribution of these 100 compounds; and 54 of those compounds have been mapped onto Figure 1 . Full details are provided in Table S1 . The detected compounds that are intermediates of the glycolytic pathway , the pentose phosphate pathway , and the TCA cycle , as well as growth factors and cofactors were found to be mostly similar between the two phenotypes ( Figure 1 , Table S1 ) . The most dramatic difference found between the two phenotypes is in phospholipid and sphingolipid metabolism . The heatmap in Figure 5 gives an overview of the full extent of the phospholipid/sphingolipid changes , the full details are given in Table S1 . The significantly different sphingolipids ( including 2 sphingomyelins ) are 3 . 5–13 fold ( median 4 . 1 fold ) more abundant in drug-sensitive clones compared with drug-resistant clones . For the phospholipids the pattern was more complex , with 19 phosphatidylcholines ( PC ) and 2 phosphatidylethanolamines ( PE ) being significantly more abundant ( 3–61 fold; median 5 . 3 fold ) in drug-sensitive clones and a different set of 10 PC and 12 PE being significantly more abundant ( 3–64 . 5 fold; median 5 . 7 fold ) in drug-resistant clones . Scrutinizing the structural properties of the fatty acyl side chains of PE and PC lipids further revealed that the changes are of a different nature in PC lipids compared with PE lipids . Figure 6 shows that only diacyl PC lipids with highly unsaturated fatty acyl chains are enriched in drug-resistant compared with drug-sensitive clones; while all the diacyl PE lipids are more abundant in drug-resistant clones . However , the total intensity of all phospholipids ( 110 ) detected was almost identical in the 2 phenotypes . A second major class of metabolites significantly modified in our drug-resistant parasites were the amino acids and amino acid derivatives . A total of 13 amino acids , including 9 proteinogenic amino acids ( Figure 1 ) , were 3–18 fold ( median 4 . 4 fold ) more abundant in the drug-resistant compared with the drug-sensitive clones ( Figure 4 ) . The remaining 11 proteinogenic amino acids were at similar abundance in the two phenotypes ( Figure 1 ) . In contrast to the amino acids , several purines ( hypoxanthine , guanine , xanthine and adenosine ) were more abundant ( 4–45 . 6 fold , median 8 . 7 fold ) in drug-sensitive clones compared with drug-resistant clones ( Figures 1 and 4 ) . However , the related nucleotides that could be detected all were at similar levels in the 2 phenotypes ( Figure 1 ) .
In this proof-of-principle study , we set out to explore whether metabolomics is applicable as a global approach to elucidate the various phenotypes present in a pathogen population . We here studied L . donovani and used clones of an antimonial-sensitive clinical isolate and an antimonial-resistant clinical isolate . The two isolates are known to be genetically very similar [11] , [35] . The molecular adaptations leading to antimonial resistance in natural Leishmania populations are still poorly understood; hypothesis-driven approaches have yielded fragmentary knowledge and suggest that antimonial resistance is multifactorial [39] . However , here we compared the global metabolomic profiles of the two phenotypes , and this has proved to be a method by which to clearly distinguish drug-sensitive and resistant isolates . Moreover , the data obtained highlights major metabolic differences between the two phenotypes which have not been reported before . The extraction procedure using chloroform/methanol/water 20/60/20 ( v/v/v ) leads to an enrichment of hydrophobic compounds in the metabolomic samples , which has revealed the notable differences in sphingolipid and phospholipid levels . However , other metabolites were also detected , with differences in amino acid and purine/pyrimidine metabolism also being observed ( Figure 1 and 4 ) . Leishmania primarily utilize salvaged and de novo synthesized sphingolipids/sphingomyelins as a source of phosphorylethanolamine for phospholipid biosynthesis , particularly phosphatidylethanolamine ( PE ) [40] , [41] ( Figure 1 ) . Our data on the steady-state lipid pools shows that there are clear differences in the metabolites of the pathways of both sphingolipid and phospholipid biosynthesis . Sphingolipids and sphingomyelins are less abundant in drug-resistant parasites , which could be consistent with their consumption at a higher rate to fuel PE biosynthesis which are more abundant in the resistant parasites ( Figure 1 ) . In contrast to PE , phosphatidylcholine ( PC ) profiles were changed in a more balanced manner; drug-sensitive clones had higher levels of PC with low fatty acyl unsaturation , while drug-resistant clones were enriched in PC with high fatty acyl unsaturation ( Figure 6 ) . This differential unsaturation profile in PC is unlikely to relate directly to the sphingolipid/PE pathway differences , but could point to another major metabolic difference between the 2 phenotypes . Although there are clear differences in the abundance of individual phospholipids , the total phospholipid content detected here appears to be similar in the 2 phenotypes . The total membranes ( plasma and internal ) of Leishmania contain 10–20% PE and approximately 40% PC [41] , [42] . PE and PC are major components of all membrane types ( e . g . plasma membrane comprises approximately 35% PE and 15% PC; mitochondrial membrane is approximately 10% PE , 25% PC ) [41]–[43] , hence it is not possible to know at present how the observed changes in phospholipid composition relate to functional changes in individual membranes . Nevertheless , the differences observed are strongly indicative that there are some functional differences too . High fatty acyl unsaturation , which is enhanced in the PC of drug-resistant parasites , is generally thought to decrease the ordered state of membranes and increase membrane fluidity [44] , [45] . Changes in membrane fluidity due to modified lipid composition have also been reported for Leishmania parasites resistant to several other drugs including miltefosine [42] , [46] , amphotericin B [45] , atovaquone [47] and pentamidine [48] . It was demonstrated that such changes in lipid metabolism affect ( i ) interaction between drug and plasma membrane and subsequent drug uptake [42] , [45] , [47] , [49] and/or ( ii ) the membrane potential of the mitochondria [48] . Thus the major phospholipid changes we have identified here in antimonial resistant clones may also have some impact upon the transport of antimonials . Modified uptake , export or sequestration of antimonials ( or a metabolite of it ) could underlie the modified antimonial susceptibility of these parasites . Leishmania are auxotrophic for many amino acids and must scavenge them from their environment . Additionally , they can also use amino acids , particularly proline , as a carbon source . Hence , free amino acids present in the environment are readily taken up by a large family of amino acid permeases [50] , [51] . Purine biosynthetic enzymes are absent in Leishmania , and the parasite depends entirely on nucleobase and nucleoside transporters to salvage from their environment [52] . The large changes in membrane-associated phospholipids observed here in drug-resistant clones could also affect uptake of both amino acids and purines , and account for the detected differences in the intracellular abundance of these metabolites between the 2 phenotypes . A large set of amino acids including several essential amino acids ( tryptophan , leucine , isoleucine , histidine ) and some atypical amino acids ( e . g . proline betaine and hydantoin-5-propionic acid , which are present in the culture medium and may simply be taken up by the parasite ) were detected at significantly different levels in drug-resistant and drug-sensitive clones . Similar differences were detected for several purines , especially nucleobases taken up by the Leishmania transporter NT3 [52] . It has been reported previously that modified lipid metabolism in other drug-resistant Leishmania resulted in significant modifications in transport of some amino acids and purines/pyrimidines which were structurally unrelated to the respective drug [49] , the changes being the indirect result of modifications in plasma membrane organisation [49] , [53] . Our findings also support this notion that modified membrane composition might indirectly alter transport of metabolites . The membrane changes we have identified in the antimonial-resistant parasites is concerning with regard to the newly installed drug policy in the Indian subcontinent . It is known that the two drugs in use , miltefosine and amphotericin B ( the second-line treatment ) , rely on their interaction with lipids in the membrane of the parasites [46] , [54] . Hence , a change in membrane composition of antimonial-resistant parasites may impact upon the efficacy of these drugs . Worryingly , there is a report of increased tolerance to all three drugs in some parasite isolates of the Indian subcontinent [7] . This demonstrates the importance of identifying the molecular mechanisms underpinning drug resistance in order to be prepared for using new drugs most effectively . Untargeted metabolomics has great potential to contribute to this much needed comprehensive characterisation of pathogens circulating in endemic regions . Our study has exemplified how the application of metabolomic approaches could play an important role in the characterisation of clinical pathogens by identifying a fingerprint of metabolic differences between various clinical phenotypes . Further experiments are currently underway to compare a much larger number of isolates representing the entire parasite population of the Indian subcontinent , in order to document the phenotypic diversity that currently exists in the L . donovani population of this kala-azar endemic region . In parallel , we are also assessing the nature and extent of genomic diversity of this parasite population by applying new sequencing technologies to characterise the whole genome of the isolates characterised by metabolomics . The integration of genomic and metabolomic approaches will result in an unparalleled source of data and promises to yield a holistic insight into the impact of endemic pathogen diversity on clinical polymorphic treatment outcome . Future application of such integrated genomic/metabolomic approaches holds great promise to address the many challenging research questions related to pathogen diversity encountered in the field of infectious diseases . | Visceral leishmaniasis is caused by a parasite called Leishmania donovani , which every year infects about half a million people and claims several thousand lives . Existing treatments are now becoming less effective due to the emergence of drug resistance . Improving our understanding of the mechanisms used by the parasite to adapt to drugs and achieve resistance is crucial for developing future treatment strategies . Unfortunately , the biological mechanism whereby Leishmania acquires drug resistance is poorly understood . Recent years have brought new technologies with the potential to increase greatly our understanding of drug resistance mechanisms . The latest mass spectrometry techniques allow the metabolome of parasites to be studied rapidly and in great detail . We have applied this approach to determine the metabolome of drug-sensitive and drug-resistant parasites isolated from patients with leishmaniasis . The data show that there are wholesale differences between the isolates and that the membrane composition has been drastically modified in drug-resistant parasites compared with drug-sensitive parasites . Our findings demonstrate that untargeted metabolomics has great potential to identify major metabolic differences between closely related parasite strains and thus should find many applications in distinguishing parasite phenotypes of clinical relevance . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"cell",
"biology/microbial",
"physiology",
"and",
"metabolism",
"computational",
"biology/synthetic",
"biology",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"microbiology/parasitology",
"infectious",
"diseases/protozoal",
"infections",
"infectious",
"diseases/antimicrobials",
"and",
"drug",
"resistance"
]
| 2010 | Metabolomics to Unveil and Understand Phenotypic Diversity between Pathogen Populations |
Trypanosoma brucei , the causative agent of African sleeping sickness , is transmitted to its mammalian host by the tsetse . In the fly , the parasite’s surface is covered with invariant procyclin , while in the mammal it resides extracellularly in its bloodstream form ( BF ) and is densely covered with highly immunogenic Variant Surface Glycoprotein ( VSG ) . In the BF , the parasite varies this highly immunogenic surface VSG using a repertoire of ~2500 distinct VSG genes . Recent reports in mammalian systems point to a role for histone acetyl-lysine recognizing bromodomain proteins in the maintenance of stem cell fate , leading us to hypothesize that bromodomain proteins may maintain the BF cell fate in trypanosomes . Using small-molecule inhibitors and genetic mutants for individual bromodomain proteins , we performed RNA-seq experiments that revealed changes in the transcriptome similar to those seen in cells differentiating from the BF to the insect stage . This was recapitulated at the protein level by the appearance of insect-stage proteins on the cell surface . Furthermore , bromodomain inhibition disrupts two major BF-specific immune evasion mechanisms that trypanosomes harness to evade mammalian host antibody responses . First , monoallelic expression of the antigenically varied VSG is disrupted . Second , rapid internalization of antibodies bound to VSG on the surface of the trypanosome is blocked . Thus , our studies reveal a role for trypanosome bromodomain proteins in maintaining bloodstream stage identity and immune evasion . Importantly , bromodomain inhibition leads to a decrease in virulence in a mouse model of infection , establishing these proteins as potential therapeutic drug targets for trypanosomiasis . Our 1 . 25Å resolution crystal structure of a trypanosome bromodomain in complex with I-BET151 reveals a novel binding mode of the inhibitor , which serves as a promising starting point for rational drug design .
Trypanosoma brucei is a unicellular , protozoan parasite and the causative agent of Human African Trypanosomiasis ( sleeping sickness ) . It also causes n'agana in cattle , a disease that imposes a severe economic burden in affected areas . The life cycle of Trypanosoma brucei requires adaptation to two distinct habitats: the fly ( tsetse ) and the bloodstream of its mammalian hosts . Within these habitats , the parasite assumes a succession of proliferative and quiescent developmental forms , which vary widely in metabolism , motility , and composition of the surface coat that covers the plasma membrane . In the fly , the trypanosome first resides in the midgut in its procyclic form ( PF ) , where its surface is coated with a group of proteins collectively termed procyclins , and then in the salivary glands in its metacyclic form , where surface procyclin is replaced with a dense coat of Variant Surface Glycoprotein ( VSG ) . The bite of the tsetse transmits the parasite to the mammalian host , where it resides extracellularly in its bloodstream form ( BF ) and continues to express VSG . The parasite relies on two strategies to evade the mammalian host antibody response . First , it varies ( “switches” ) its highly immunogenic surface antigen , using a repertoire of ~2 , 500 distinct VSG genes [1] . Only one VSG is expressed at a time ( monoallelic expression ) , and host antibodies mounted against the initially expressed VSG must be continually replaced by antibodies against antigenically distinct VSGs , resulting in waves of parasitemia in the infected host [2 , 3] . Second , antibodies bound to surface VSG are rapidly internalized by the parasite [4] , giving host effector cells less time to recognize and eliminate it . When trypanosomes enter the midgut of the tsetse following a bloodmeal , a temperature drop and an increase in acidity function cooperatively to induce differentiation from the BF to the PF . Together with remodeling of the parasite surface to replace VSG with procyclin , there are a number of cytoskeletal changes that occur , and the kinetoplast is repositioned . There is also a drastic change in metabolism as the trypanosomes leave the glucose-rich environment of the blood and transition to the fly midgut , where they are more reliant on mitochondrion Krebs cycle enzymes and respiratory chain and oxidative phosphorylation enzymes ( reviewed in [5] ) . Trypanosomes are unusual in that they are eukaryotes that transcribe their genome from polycistronic transcription units ( PTUs ) that often contain functionally unrelated genes . Thus , much regulation of gene expression occurs at the post-transcriptional level . With respect to differentiation , it is clear that the 3’UTRs of procyclin and VSG genes contribute to their developmental regulation [6–8] , and that RNA binding zinc-finger proteins also play a role in specifying a differentiation program [9] . Many recent studies have shown that there are large changes in gene expression at the mRNA level during differentiation [10–12] . While developmental changes in gene expression have been examined at the level of mRNA stability and association with the translation machinery , the mechanisms by which these changes are mediated at the DNA level are not well understood . The fact that trypanosomes have recognizable histones , histone modifications , and orthologues of many of the chromatin-interacting proteins found in other , more familiar model systems made us wonder whether differentiation from the BF to the PF may be epigenetically regulated . As epigenetic modifications are reversible , this would confer the plasticity necessary to maximize fitness according to environmental input [13] , allowing cells to be poised to move forward in development , but not until the receipt of the appropriate environmental signals . In mammals , bromodomain and extraterminal domain ( BET ) family members that recognize histone lysine acetylation through their bromodomains are necessary to maintain murine embryonic stem cell pluripotency [14 , 15] , and differentiation can be rapidly induced by subjecting cells to treatment with highly selective bromodomain inhibitors [16] . Chromatin-interacting proteins in T . brucei have thus far been studied primarily in the context of monoallelic VSG expression and switching [17–20] , but less is known about their potential role in lifecycle differentiation . Here , we report that acetyl-lysine ( AcK ) recognizing bromodomain proteins maintain the BF cell fate in trypanosomes . Using both small molecule inhibitors and genetic mutants of individual bromodomain proteins , we demonstrate that bromodomain inhibition results in gene expression changes that mimic those seen naturally in cells differentiating from the BF to the insect stage following a bloodmeal and transition to the fly midgut . Transcriptional changes are recapitulated at the protein level by the appearance of insect-stage procyclin on the cell surface . BF-specific processes , including monoallelic expression of VSG genes and rapid surface-bound antibody internalization are also compromised , establishing T . brucei bromodomain proteins as attractive therapeutic targets for trypanosomiasis . Indeed , bromodomain inhibition decreases virulence in an in vivo mouse model . Our crystal structure of a trypanosome bromodomain in complex with an inhibitor revealed marked differences in the AcK recognition site to its human counterparts , consistent with a novel binding mode of the inhibitor , which can now be exploited to develop trypanosome-specific , high-affinity drugs against trypanosomiasis .
Bromodomain proteins have recently been implicated in the maintenance of stem cell pluripotency [14 , 15] . The T . brucei genome contains five genes with predicted bromodomains [21] , and we hypothesized that these proteins may play similar roles in maintaining the parasites in their BF and preventing differentiation to the PF . To address this question , we took advantage of a highly selective small-molecule bromodomain inhibitor called I-BET151 , a well-characterized successor of I-BET762 , which inhibits BET bromodomain proteins in mammalian cells [22 , 23] . We performed RNA-seq experiments on I-BET151-treated cells and compared them to cells treated with a vehicle control ( DMSO; dimethyl sulfoxide ) . For the initial dataset , cells were subjected to either two days or three days of treatment with I-BET151 and compared to DMSO-treated cells . The RNA-seq libraries were prepared in triplicate from three independent cultures for each treatment using polyA+ selection . We used DESeq analysis [24] to identify the top up-regulated and down-regulated genes following I-BET151 treatment . A list of these genes along with their associated multiple testing corrected p-adjusted values are listed in Additional_File_1 and Additional_File_2 in S1 Data . We used the log-fold changes and p-adjusted values from the DESeq analysis on I-BET151 cells treated for 3 d to generate the volcano plot shown in Fig 1A . Strikingly , I-BET151 treatment resulted in a strong up-regulation of a number of genes expressed only in the insect-residing PF form , including procyclins ( EP1-3 and GPEET ) , procyclin-associated genes ( PAG1 , 2 , 4 , 5 ) , procyclin stage surface antigen ( PSSA ) , and protein associated with differentiation 2 ( PAD2 ) . We also observed a marked down-regulation of genes whose expression is normally higher in the mammalian BF , such as ISG64 , PYK1 , and GPI-PLC ( Fig 1A ) . To confirm whether the up-regulation in transcription of EP1 was reflected at the protein level , we performed flow cytometry experiments to ask whether EP1 was expressed on the surface of the cells . Indeed , a large proportion of I-BET151-treated trypanosomes expressed EP1 after 3 d of treatment , while this was not the case for vehicle-treated controls ( Fig 1B ) . We conclude that bromodomain proteins are important for maintaining a BF-specific cell surface landscape . Previous studies identified genome-wide changes in gene expression during differentiation from the BF to the PF and organized them both by hierarchical clustering of genes that varied with similar kinetics ( clusters ) and by functional group [12] . Each cluster contains a group of genes of both known and unknown function that were shown to change by roughly the same magnitude and with the same timing throughout differentiation . These genes are sometimes but not always functionally related . For example , about half the genes in cluster 30 are flagellar genes , but the remainder fall into different functional groups or have unknown function . Conversely , several flagellar genes are found in other clusters based on their pattern of expression . To analyze these same clusters of genes , we used a Gene Set Enrichment Analysis ( GSEA ) algorithm generated by the Broad Insitute at MIT [25] ( http://www . broad . mit . edu/gsea ) . This algorithm generates an Enrichment Score that estimates the degree to which a given set of genes is over-represented within the total set of the most highly up-regulated or down-regulated genes . It also provides an Estimation of Significance level and corrects for multiple hypothesis testing to generate a false discovery rate ( FDR ) . A list of the genes contained within each cluster is listed in Additional_File_3 in S1 Data and S12 Table . We analyzed our RNA-seq datasets comparing 2-d I-BET151-treated cells to control cells using GSEA enrichment with the clusters generated in [12] . S1 Table shows each cluster and its associated enrichment score and FDR for the 2-d I-BET151-treated dataset . To get a sense of how many of the genes within each cluster were moving coordinately in the same direction ( up or down with respect to untreated cells ) , we conducted an additional analysis that determined the percent of genes up-regulated or down-regulated by 1 . 1-fold within each cluster in I-BET151-treated cells . S1 Table includes these percentages . We generated a list of clusters that had a GSEA generated FDR of <0 . 1 and for which > 70% of the genes were up-regulated or down-regulated following treatment with I-BET151 . For the 31 clusters that met these criteria , we generated boxplots showing the expression level for all genes within that cluster in cells that had been treated for two days or three days with I-BET151 , and compared them to control cells ( Fig 1C ) . The red line on the boxplot indicates the median level of gene expression for all the genes within a cluster , while the whiskers denote genes falling within the inner quartile range . Outliers are shown as blue dots . In some cases , a progressive change was observed in the median level of gene expression for the cluster throughout the time course ( e . g . , 10 , 20 , 22 ) . In other clusters , there was a shift in the distribution of the data in untreated versus 2-d-treated cells , but not between 2-d and 3-d-treated cells ( e . g . , 11 , 46 , 56 ) . We conclude that gene expression changes that take place during differentiation for roughly half ( 31 ) of the clusters in the previous study were mirrored in I-BET151-treated cells . Queiroz et al . also organized groups of genes that change throughout differentiation by their annotated function . These groups and their abbreviations are listed in Additional_File_4 in S1 Data and S12 Table . We performed the same analysis described above on these groups of genes and each functional group’s GSEA scores are listed in S2 Table . We generated boxplots for the ten functional groups that met our criteria for differential expression following treatment with I-BET151 ( <0 . 1 GSEA FDR and >70% of genes up- or down-regulated ) ( Fig 1D , S2 Table ) . For example , during the transition from the BF to the PF , trypanosomes down-regulate genes required for high rates of glucose and glycerol metabolism ( denoted as glycBS in Fig 1D ) as they leave the glucose-rich environment of the blood and transition to the midgut of the fly . Other genes in the glycosomal pathway that are not involved in the pentose phosphate pathway are also down-regulated ( denoted as glyc in Fig 1D ) . Cells undergoing differentiation also up-regulate many of the genes involved in the citric acid cycle ( denoted as mit CAC in Fig 1D ) . We find these changes reflected in I-BET151-treated cells ( Fig 1D , S2 Table ) . Notably , genes found in the procyclin-specific transcription unit ( denoted as procyc in Fig 1D ) were also up-regulated upon treatment with I-BET151 ( Fig 1D , S2 Table ) . Other previously identified functional groups that met our criteria for differential expression in I-BET151-treated cells include the invariant surface glycoproteins ( ISGs ) , genes involved in cell cycle ( cyc ) , genes associated with glycosyl phosphatidylinositol metabolism ( GPI ) , cytoskeletal genes ( cyskel ) , flagellar genes , and genes associated with intracellular protein and vesicular transport pathways ( vestrans ) ( Fig 1D ) . We conclude that with respect to transcription , bromodomain inhibited cells share some similarities to differentiating cells . In order to compare the changes in gene expression that take place in I-BET151-treated cells with those that take place during differentiation , we performed a more extensive RNA-seq time course of I-BET151 treatment that included earlier time points than had been analyzed previously , at 3 , 6 , 12 , and 24 h . As above , RNA-seq libraries were generated in triplicate using poly-A+ selection . First , we performed GSEA analysis using the Pearson metric for ranking genes in a time series ( http://www . broadinstitute . org/gsea/doc/GSEAUserGuideFrame . html ? _Metrics_for_Ranking ) . We thus obtained a list of 17 functional groups and 35 clusters that were over-represented in the set of genes that are differentially expressed through the time course using a GSEA FDR cutoff of <0 . 1 ( S3 and S4 Tables ) . We used DESeq’s negative binomial test to generate a list of differentially expressed genes at every time point using a multiple testing corrected p-adjusted cutoff of < 0 . 1 . Differentially expressed genes and their associated expression and p-values for each time point are listed in Additional Files 5–9 in S1 Data . For those genes previously found to have altered gene expression during differentiation [12] , we binned each DESeq generated differentially expressed gene into its associated cluster or functional group . We then plotted the percentage of genes within each cluster or group that were differentially expressed over the time course ( Fig 1E and 1F , S1 and S2 Figs , S5 and S6 Tables , Additional Files 5–9 in S1 Data ) . Note that only the 17 functional groups and 35 clusters with a GSEA generated FDR of <0 . 1 ( S3 and S4 Tables ) are plotted in Fig 1E and 1F , S1 and S2 Figs . Note also that the percentages of differentially expressed genes used in the plots are in S5 and S6 Tables . For example , at 3 h after the initiation of I-BET151 treatment , 23% of genes in the glycosomal pathway ( glyc , GSEA FDR = . 018 ) are called by DESeq as being differentially expressed with a p-adjusted value of <0 . 1 , whereas at 48 h , 94% of the genes in this group are differentially expressed in our dataset ( gray line in Fig 1F , see also S5 Table ) . Other functional groups that showed early differential expression of their gene members include mitochondrial citric acid cycle enzyme genes ( mit CAC ) , flagellar genes ( flag ) , RNA binding protein genes ( RBP ) , chaperone or protein-folding-related genes ( chap ) , procyclic transcription unit genes ( procyc ) and genes associated with intracellular and vesicular transport pathways ( vestrans ) ( Fig 1F ) . The percentage of differentially expressed genes for each functional group rises sharply between 12 and 24 h for 15 of the groups analyzed ( e . g . , ISG and GPI , S1A Fig ) ( Fig 1F and S1A–S1C Fig ) and generally continues to rise between 24 and 48 h ( Fig 1F , S1B and S1C Fig ) . In addition to the ISG and GPI groups , genes involved in RNA degradation ( RNAdeg ) , glucose and glycerol metabolism ( glycBS ) , cytoskeleton ( cyskel ) , ubiquitination ( ubiq ) , DNA binding , replication , and nucleases ( DNA ) and protein kinase or kinase binding activity ( signal K ) were differentially expressed starting at 12 h after initiation of I-BET151 treatment ( S1B and S1C Fig ) . Finally , two groups of genes were differentially expressed late in the time course , between 24 and 48 h . These included ribosomal proteins ( ribprot ) and genes associated with the nucleolus ( nucleol ) ( S1D Fig ) . We performed the same analysis described above on the 35 clusters with GSEA FDRs of <0 . 1 . We found that six clusters had differentially expressed genes early in the time course ( Fig 1E ) . 27 clusters contained genes that began to be differentially expressed at 12 h postinduction , and the percentage of differentially expressed genes within the majority of clusters continued to rise until 48 h ( S2A–S2D Fig ) . Genes within Clusters 57 and 22 were not differentially expressed until 24 h postinduction ( S2E Fig ) . Overall , this analysis indicates that a small subset of genes ( between 21 and 29 ) are differentially expressed in the first 12 h of I-BET151 treatment , with a much larger number of genes ( ~700 ) showing differential expression between 12 and 24 h ( see Additional Files 5–9 in S1 Data ) . We wanted to compare the timing for gene expression changes that took place during our I-BET151 RNA-seq time course with those that took place in the Queiroz study . We compared only those clusters and functional groups with a GSEA FDR of <0 . 1 and >70% of genes up- or down-regulated at the 48 h time point . To get a global picture of the expression changes within each cluster or functional group , we plotted the median reads per kilobase per million mapped reads ( RPKM ) for each cluster or group over time ( Fig 2 and S3 Fig , solid lines ) . We artificially set the median for each group at time 0 to be 1 and normalized the other medians accordingly , in order to more easily compare changes between groups . We then calculated the median expression level for each cluster or group using data generated in the Queiroz study and plotted these medians alongside our data ( Fig 2 and S3 Fig dashed lines ) . Each solid and dashed line was given the same color for each functional group or cluster to ease comparison between the studies . However , we would like to emphasize that caution must be taken when comparing these two datasets , because they were not only generated using different cells and different induction techniques , but also the previous study used microarray technology rather than sequencing technology . Thus the level of expression is only grossly comparable between the two datasets . Nevertheless , we did observe that in general , for the 10 functional groups that met our criteria ( GSEA FDR of <0 . 1 and >70% of genes up- or down-regulated at the 48 h time point ) , the median gene expression between 12 and 76 h for each of the groups ( to the right of the black line on each plot ) roughly correlated with the previous study ( Fig 2 and S3 Fig ) . While the median gene expression at 48 and 76 h was quite similar for some functional groups ( e . g . , procyclic , glycolysis , GPI , ISG ) , other functional groups in our dataset changed to a lesser extent than those in the previous study ( e . g . , mit CAC and RNA degradation ) ( Fig 2A ) . Also of note is that the expression changes were more poorly correlated between 3 and 12 h ( to the left of the black reference line in Fig 2 ) . For contrast , three functional groups that did not meet our criteria for similarity to the previous study ( either > 0 . 1 GSEA FDR or <70% of group up- or down-regulated ) are plotted in Fig 2B . 31 clusters matched our criteria for having a GSEA FDR of <0 . 1 and greater than 70% of genes up- or down-regulated ( S3 Fig ) . In some cases , the median level of gene expression at 48 and 76 h matched quite well between the datasets ( e . g . , clusters 40 , 6 , 29 , 54 , 52 , 51 , 10 , 16 , 39 , and 31 ) , while in other cases the changes in our dataset were more muted ( e . g . , clusters 32 , 59 , 17 , 18 ) . As with the functional groups , changes in gene expression between 3 and 12 h were more poorly correlated between datasets . Examples of clusters that did not pass our criteria for similarity to the previous study ( either > 0 . 1 GSEA FDR or <70% of group up- or down-regulated ) are plotted in S3B Fig . For some clusters and functional groups analyzed ( e . g . , RNA deg , mit CAC , nucleolus , cluster 4 , 6 , 7 ) , changes in gene expression as a result of I-BET151 treatment were delayed when compared to the gene expression changes that take place upon initiation of differentiation ( Fig 2 and S3 Fig ) . We conclude that while I-BET151-treated cells show gross similarities to differentiating cells with respect to changes in the transcriptome , these changes are not identical for every group of genes , the changes are sometimes delayed , and in some cases the response is more muted . Genes with functions in the intracellular protein transport and vesicular transport pathways are necessary for VSG recycling and are down-regulated during differentiation [26] . Consistent with this observation , we found that genes within the vesicular and intracellular transport group were down-regulated in I-BET151-treated cells ( Figs 1D and 2A and Additional Files 1 , 2 , and 5–9 in S1 Data ) . Concurrently , expression of genes required for motility and for VSG-bound antibody internalization [4] are down-regulated during differentiation and were also reduced in I-BET151-treated cells ( denoted “flag” in Fig 1D and flagellar in Fig 2A ) . We therefore tested whether I-BET151-treated cells were able to internalize surface-bound antibody . We first coated the cells with a primary antibody against the expressed surface VSG . After washing unbound antibody away , we subjected the cells to a 5-min incubation at 4°C or 37°C . At 4°C , wildtype cells are unable to move and the antibody remains on the surface , whereas at 37°C they are able to move freely and internalize surface-bound antibody [4] . Following incubation , we immediately fixed the cells with formaldehyde . We then added a secondary antibody and measured how much primary antibody remained on the surface by flow cytometry and immunofluorescence . We observed a complete block in antibody internalization in I-BET151-treated cells ( Fig 3A and 3B ) , consistent with decreased motility observed in these cells ( compare S1 and S2 Movies ) . We also observed an increase in the amount of anti-VSG antibody bound to the surface in I-BET151-treated cells ( Fig 3C ) without a concomitant increase in total VSG protein ( Fig 3D ) . We attribute this accumulation in surface-bound antibody to the block in antibody internalization and possible defects in VSG recycling associated with down-regulation of many of the genes necessary for these processes ( Figs 1D and 2A ) . These data strongly suggest that bromodomain proteins are necessary for maintaining the BF cell fate , and that their inhibition has functional consequences on BF-specific immune evasion processes , such as rapid internalization of surface-bound antibodies . Recent work has demonstrated a mechanistic link between VSG expression and differentiation from BF to PF cells [27] . VSGs are located in metacyclic-specific expression sites ( metacyclic expression sites [ESs] ) , minichromosomes , and elsewhere in the megabase chromosomes ( S4A Fig ) . In the BF , they are transcribed from one of ~15 telomeric ESs , but only one ES is transcriptionally active at any one time . Each ES contains a promoter and a number of Expression Site Associated Genes ( ESAGs ) upstream of the VSG . Batram et al . discovered that ectopic expression of a VSG gene causes silencing to spread from the active ES telomere upstream toward the promoter , and that this silencing “primes” cells to transition from the BF to the PF . We used RNA-seq to ask whether a similar phenomenon occurs in I-BET151-treated cells by mapping reads that did not align to the megabase chromosomes to ES ESAGs and to VSGs [28] . We observed an increase in transcription of silent VSGs genome-wide ( S4B Fig , Additional File 10 in S10 Data ) . Q values for the VSGs were calculated by subjecting p-values generated from the replicate sets to Benjamimi and Hochberg correction using SeqMonk software from Babraham Bioinformatics ( see Materials and Methods ) and VSGs with a Q value of < 0 . 1 are shown in red in S4B Fig . Examples of cells expressing more than one VSG could be observed by staining I-BET151-treated cells with multiple VSG antibodies and using an Imagestream flow cytometer to visualize double-expressing cells ( S4C Fig , top ) . However , it should be noted that the proportion of cells with two different VSGs on the surface is <1% of the total population ( S4C Fig , bottom ) , and thus the loss of monoallelic expression appears to take place mainly at the transcriptional level . We hypothesized that the increase in VSG mRNA could result in active ES attenuation , as was observed by Batram et al . A decreasing transcription gradient at the active ES has also been demonstrated at the beginning of differentiation from the BF to the PF [29] . To unambiguously map the ESAG reads to each ES , we aligned the reads uniquely allowing for 0 mismatches . This analysis revealed an increase in transcription of ESAGs near the active ES promoter , while those near the active ES telomere were silenced , with silencing spreading upstream toward the promoter between 2 and 3 d of treatment ( S4D Fig , Additional File 11 in S10 Data ) . These results support the notion that I-BET151-treated cells are primed for differentiation . Another link between ES expression and differentiation was observed by Amiguet-Vercher et al . , who found an increase in transcription in regions near the ES promoter in silent ESs during the early stages of differentiation [29] . We also observed an increase in transcription of ESAGs at silent ESs upon treatment with I-BET151 ( S4E Fig ) . We plotted the median log2 ( RPKM ) for all ESAGs in silent ESs during the I-BET151 time course and found an initial increase in expression of these ESAGs at 3 h postinduction , with a much more dramatic increase in expression occurring between 24 and 48 h ( S5 Fig ) . To ascertain the timing for increased expression of silent VSGs , we plotted the median log2 ( RPKM ) for all inactive VSGs during our time course and observed that VSG expression rises later in the time course , primarily between 24 and 48 h ( S5 Fig ) . We conclude that bromodomain proteins are necessary for maintaining monoallelic expression of VSG genes , and confirm that transcription of ESAGs and VSGs appears intimately linked to differentiation processes . In mammalian cells , differentiation from a totipotent to a differentiated state is accompanied by a number of changes at the chromatin level , including deposition of repressive histone marks and increasing chromatin compaction . Cellular reprogramming to the undifferentiated state is characterized by a nucleus largely devoid of heterochromatin [30] , and is inhibited by repressive histone modifications [31] . While mammalian cells generally remain in their differentiated state , trypanosomes are required to cycle continuously between alternate differentiated states , and thus the epigenetic changes that accompany these transitions must necessarily retain some degree of plasticity . In agreement with this notion , the effects of I-BET151 treatment are reversible . Longer treatments ( 5 d ) with I-BET151 lead to very low levels of anti-VSG antibody bound to the cell surface ( Fig 4A ) , but memory of the active ES appears to be retained , as the VSG associated with this same ES is expressed on the surface after 3 d of washing out the drug ( Fig 4B ) . Similarly , levels of surface EP1 are high after 5 d of I-BET151 treatment , but expression is lost upon washout of the drug . These results imply that the transcription changes associated with differentiation are reversible , and supports the intriguing possibility that these changes may be epigenetically mediated , though this remains to be conclusively proven . We then set out to confirm that the observations associated with I-BET151-treated cells are mediated by the inhibition of trypanosome bromodomain proteins . First , we endogenously tagged Bdf1-4 with C-terminal HA ( Bdf5 function has already been studied [32 , 33] ) . We performed chromatin immunoprecipitation-sequencing ( ChIP-seq ) experiments and used the Model-based Analysis for ChIP-Seq ( MACS ) algorithm to determine peaks of localization for each tagged Bdf [34] . Each experiment was performed in duplicate , and only those peaks that were called by the MACS algorithm in both replicates were analyzed . Each peak was required to have a MACS algorithm-based FDR of <0 . 1 in at least one replicate to be included in the analysis presented in Fig 5B . Additionally , we called peaks in control samples that were subjected to anti-HA pull downs but that did not contain tagged proteins . Any peak that was called by MACS in these control lines was removed from the set of called peaks in each of the tagged lines . Genomic location , FDR values , and fold enrichment for each called peak are in Additional Files 12–14 in S5 Data . We confirmed binding of Bdf3 to sites of transcription initiation [35] and found that Bdf1 and 4 also localized to these sites , while Bdf2 did not ( Fig 5A ) . Specifically , 67% , 46% and 67% of Bdf1 , 3 , and 4 peaks , respectively , are within 100 bp of a divergent strand switch region ( with two PTUs going in opposite directions , presumed to be transcription start sites ) ( Fig 5B , left ) . 52% , 85% and 70% of all divergent strand switch regions ( Transcription Start Sites , TSSs ) are within 100 bp of a Bdf1 , 3 , or 4 peak , respectively ( Fig 5B , right ) . In contrast , although the MACS algorithm did call 71 Bdf2 peaks in the genome , none of them met the <0 . 1 FDR cutoff , and were thus not included in this analysis . Based on genomic localization , we divided bromodomain proteins into two classes: those that bind to transcription start sites ( Bdf1 , 3 , 4 ) and those that do not ( Bdf2 ) . We decided to further study Bdf3 and Bdf2 as representatives of each class . Using the recombinantly expressed and purified bromodomains from T . brucei Bdf2 and Bdf3 ( Tb927 . 10 . 7420 , Tb927 . 11 . 10070 , respectively ) , we performed isothermal titration calorimetry experiments and confirmed direct and specific binding of I-BET151 to their AcK recognition sites , as mutation of two conserved residues in the AcK sites abolished interaction with the inhibitor ( Fig 5C ) . By contrast , we could not detect binding of another well characterized bromodomain inhibitor , JQ1 [36] , to recombinant Bdf2 and Bdf3 bromodomains ( S6 Fig ) , which is consistent with the fact that JQ1 had no biological effects on trypanosomes in our hands ( S12 Data ) . Having confirmed Bdf2 and Bdf3 as I-BET151 targets , we next asked whether genetic ablation of these proteins would phenocopy the effects of treating trypanosomes with I-BET151 . To enable functional characterization of Bdf2 , we created a conditional knockout strain with a floxed , endogenously HA-tagged allele of Bdf2 ( Bdf2KO cells ) and replaced the remaining allele with a floxed drug resistance marker . Doxycyclin ( Dox ) treatment induces Cre expression to delete floxed alleles of Bdf2 in the Bdf2KO strain . To study the function of Bdf3 , we created a separate , inducible RNAi strain harboring an endogenously HA-tagged Bdf3 ( Bdf3kd cells ) . Here , Dox treatment induces expression of double-stranded RNA complementary to Bdf3 ( S7A Fig ) . Dox treatment of Bdf2KO and Bdf3kd cells resulted in efficient protein knockdown of Bdf2 and Bdf3 , respectively ( S7B and S7C Fig ) . We observed normal growth in the Bdf2KO cells until 48 h after induction , where a growth defect emerged , coinciding with perturbations in cell cycle ( S8A and S8C Fig ) . Bdf3kd cells were growth arrested between 12 and 24 h , and we observed cell cycle defects at 24 h ( S8B and S8C Fig ) . Cell cycle defects largely resolved between 1 and 2 d of treatment ( S8C Fig , right panel , coinciding with a restoration in Bdf3 protein level at 48 h ( S8D Fig , bottom ) , implying that the RNAi against Bdf3 appears to shut off after 24 h by an unknown mechanism . RNA-seq analysis was performed on Bdf2KO and Bdf3kd cells using similar methods as those described for I-BET151 treatment . Because the kinetics for deletion of Bdf2 were slightly slower than for knockdown of Bdf3 ( S8D Fig ) , we allowed a longer time period for induction of Bdf2 deletion and associated downstream transcriptional effects prior to harvesting RNA ( 48 h for Bdf2 versus 24 h for Bdf3 knockdown ) . polyA+ RNA was isolated from 3 independent biological samples to generate RNA-seq libraries for cells treated with and without Dox . DESeq analysis revealed that many of the insect-specific genes that were up-regulated in I-BET151-treated cells were also up-regulated in Bdf3kd , but not in Bdf2KO cells , including EP1-3 , GPEET2 and procyclin-associated genes PAG1 , 2 , 4 , 5 ( Fig 5D , Additional File 15 and Additional File 16 in S5 Data ) . We performed GSEA analysis in Bdf3kd and Bdf2KO cells using the same clusters and functional groups described earlier , which had previously been shown to be differentially expressed during the transition from the BF to the PF [12] . All results of GSEA analysis on Bdf3kd and Bdf2KO cells are listed in S7–S10 Tables . Of the functional groups analyzed , 10 groups met our criteria of having a GSEA FDR of <0 . 1 and 70% of genes moving in one direction in Bdf3kd cells ( S7 Table ) . Boxplots of these functional groups are shown in S9A Fig , left . By contrast , only one functional group passed our criteria in Bdf2KO cells ( S9B Fig , left ) . Of the clusters analyzed , six clusters passed our criteria in Bdf3kd cells ( S9A Fig , right ) , while four clusters passed in Bdf2KO cells ( S9B Fig , right ) . RNA-seq analysis revealed that transcripts for VSGs located at silent bloodstream expression sites ( BESs ) were significantly increased in both Bdf3kd and Bdf2KO cells . This was also true for metacyclic VSGs , while VSGs in minichromosomes were only slightly up-regulated in Bdf3kd cells ( S10A Fig ) . Consistent with its more widespread genomic localization , knockdown of Bdf3 increased transcription for more of the VSGs in the genome than was the case for Bdf2 ( S10B Fig ) . These results imply that while both proteins maintain monoallelic expression of VSG genes , Bdf2 has a more specialized role in maintaining silencing at ESs . We tested whether Bdf2 could be localized to the ES using chromatin IP followed by Q-PCR . We performed this test in a cell line with a Blasticidin resistance gene at the active ES promoter and a GFP marker at one of the silent ES promoters . We found that Bdf2 localized to the promoter region of both the active and silent ES , and we also detected weaker binding at each of the VSGs associated with these ESs ( S10C Fig , left ) . Binding to these regions was decreased in the presence of I-BET151 ( S10C Fig , right ) . We also tested binding of Bdf3 to the ES using a cell line that did not distinguish the promoters of active and silent ESs . We found that Bdf3 also binds to the ES promoter ( ESP1 in S10D Fig ) and to a pseudogene located further down the ES , but that it did not appear to bind directly to the VSG genes ( S10D Fig ) . Finally , transcription of ESAGs in inactive ESs is also higher in Bdf3kd and Bdf2KO cells ( S11 Fig ) . We further tested whether Bdf3 binding was altered in the presence of I-BET151 by collecting samples for ChIP at 6 h , 12 h , and 24 h following initiation of treatment . We then used Q-PCR to test whether Bdf3 binding was altered at the ES promoter ( ESP1 in S12A Fig ) or at several other sites identified as peaks of Bdf3 binding in our ChIP-seq experiment ( S12B–S12D Fig ) . We found that I-BET151 treatment decreased binding to both the ES promoter and to the other sites we tested in a progressive manner , whereas it did not decrease binding to the URA3 control region ( S12A Fig ) . Our RNA-seq experiments on Bdf2KO and Bdf3kd cells were conducted at 48 h and 24 h post-induction , respectively . We wanted to confirm that the changes in transcription at these later time points were reflected early upon depletion of each Bdf . To do this , we isolated RNA from Bdf2KO cells at 6 , 12 , 24 , 36 , and 48 h postinduction with Dox . For Bdf3kd cells , we isolated RNA at 6 , 12 , and 24 h postinduction . We confirmed that PAG2 , EP1 , and GPEET2 mRNA expression increased after 6 h of induction in Bdf3kd cells . This was also the case for a VSG gene in a silent site ( VSG3 ) ( S13A Fig ) . We were also able to confirm that the transcription of silent ES VSGs and one metacyclic VSG was increased in Bdf2KO cells as early as 6 h postinduction ( S13B Fig ) . Genetic depletion of Bdf3 did not result in the expression of EP1 on the surface of the cells ( Fig 5E ) , but previous results from Batram et al . indicated that the presence of an ectopically expressed VSG “primed” cells for differentiation from the BF to the PF and increased their sensitivity to cis-aconitate , an acidic small-molecule differentiation trigger that mimics the increased acidity of the fly midgut [27 , 37 , 38] . We wondered whether the increase in transcription of normally silenced VSGs might have a similar effect , so we subjected Bdf3kd and Bdf2KO cells to treatment with cis-aconitate or reduced temperature . Indeed , genetic depletion of bromodomain proteins primes BF trypanosomes for differentiation , and requires only one ( cis-aconitate ) , rather than two , environmental stimuli to induce EP1 expression ( Fig 5E ) . The effect was stronger in Bdf3kd cells , perhaps due to the significant up-regulation of EP1 mRNA . Unlike the Bdf2- and Bdf3-depleted trypanosomes , a large proportion of I-BET151-treated cells expressed EP1 at high levels in the absence of any environmental stimulus , and nearly all cells expressed EP1 following treatment with cis-aconitate or incubation at 27°C ( Fig 5E ) . Control cells with intact bromodomain proteins showed very low induction of EP1 upon exposure to differentiation triggers ( Fig 5E ) . These results confirm that bromodomain proteins are important for preventing remodeling of the parasite surface until all the appropriate environmental signals have been received . We were able to partially recapitulate the block in antibody internalization observed in I-BET151-treated cells in both Bdf3kd and Bdf2KO cells ( Fig 5F ) . The block in Bdf3kd cells is consistent with the decrease in transcription of flagellar genes necessary for motility ( S9 Fig ) . The partial block in antibody internalization observed in Bdf2KO cells was unexpected , but might be explained by a decrease in expression of the paraflagellar rod gene Pfr2 ( Tb927 . 8 . 5010 ) and in α-tubulin ( Tb927 . 1 . 2400 ) in the Bdf2KO RNA-seq dataset . We hypothesize that these might be indirect consequences of Bdf2 inhibition related to VSG derepression [27] . Notably , the effects of I-BET151 treatment were more pronounced than those in cells with depletion of individual bromodomains , which most likely reflects simultaneous inhibition of at least two bromodomains ( Bdf2 and Bdf3 ) by I-BET151 . We cannot rule out the possibility that I-BET151 may also bind Bdf1 , 4 , and/or 5 , and that these proteins may be able to partially compensate for the lack of Bdf3 in Bdf3kd cells . Overall , we show that most of the effects seen in I-BET151-treated cells are phenocopied in Bdf3kd cells , and that ES-specific effects are also observed in Bdf2KO cells . Our results imply that bromodomain proteins are required for maintenance of BF stage identity . Inhibition of bromodomain proteins compromises BF-specific immune evasion mechanisms and provides several advantages for the host immune system , including slower parasite growth and cell cycle defects ( Fig 6A and 6B and S8 Fig ) , expression of an invariant epitope on the surface of the cell ( EP1 ) ( Fig 1B ) , and defects in antibody internalization ( Figs 3A and 5F ) , making these proteins attractive drug targets . However , while repurposing known bromodomain inhibitors to combat trypanosomiasis is a promising strategy , in vivo treatment with I-BET151 itself is suboptimal because of its low affinity for trypanosome bromodomain proteins and its high affinity for mammalian BET proteins that affect the immune system [22] . We therefore asked whether the course of parasitemia is altered in mice infected with trypanosomes treated in vitro with I-BET151 prior to infection . Mice infected with parasites treated for two days with I-BET151 survived significantly longer than control mice . In addition , 80% of the mice infected with parasites treated for three days with I-BET151 did not develop detectable parasitemia ( Fig 6C ) . We wanted to make sure that the survival advantage in WT mice was not due to lack of viability of I-BET151-treated cells , and we also wanted to confirm that I-BET151-treated cells were able to establish infections in vivo . We reasoned that if the drug-treated trypanosomes were not viable or capable of establishing an infection , then RAG-/- mice , which lack B and T cells , would have a similar advantage to WT mice in surviving infection with I-BET151-treated trypanosomes . We thus performed the same experiment in RAG-/- mice and observed that survival was significantly lower in RAG-/- mice that had been treated with I-BET151-treated trypanosomes when compared to WT mice ( Fig 6C ) . We conclude that the survival advantage conferred by I-BET151-treated cells in WT mice cannot be solely due to lack of viability of drug-treated trypanosomes , since they are capable of killing immunocompromised mice . We performed two additional control experiments to ensure that drug-treated trypanosomes were viable in vitro . We treated trypanosomes with I-BET151 or DMSO for 3 d and then washed out the drug . We plated the treated cells in 3 separate 96-well dishes at a concentration of 50 cells/well and then counted the number of wells where trypanosomes were growing after 4 d . We chose this concentration because we infected mice with 50 trypanosomes . 100% of wells in all 3 plates had trypanosomes growing after 4 d for both DMSO and I-BET151-treated samples ( S14A Fig ) . This is not to say that there is no loss in viability; only 66% of wells with I-BET151-treated cells plated at a concentration of 5 cells/well grew into a colony after 4 d ( as opposed to 100% of WT cells ) ( S14A Fig ) . Finally , we used flow cytometry to measure the percent of propidium iodide ( PI ) + cells following 3 d of treatment with I-BET151 and again after washout and recovery . We found that while the percent of PI+ cells in the population was 3 . 5 times higher in I-BET151-treated cells compared to DMSO-treated cells , the percentage of PI+ cells in the I-BET151-treated population was only 1 . 1% overall ( S14B Fig ) . The percent of PI+ cells after 3 d of washout and recovery was 0 . 3% for both populations ( S14B Fig ) . We conclude that it is highly probable that there were live cells within the 50 cell starting population of I-BET151-treated cells used to infect the mice for the experiments shown in Fig 6 . Consistent with the data in Fig 6A , we did observe a lag in growth in I-BET151-treated cells , both prerecovery and postrecovery , which might contribute to the increased survival of WT mice infected with I-BET151 treated trypanosomes ( S14C Fig ) . We also tested the effects of genetically depleting bromodomain proteins in vivo . We infected mice with Bdf3kd cells and knocked down Bdf3 in vivo by feeding infected mice with Dox pellets either before or after parasitemia became detectable . We observed an increase in survival specifically in Dox-fed mice ( Fig 6D ) . Mice infected with Bdf2 conditional knockout parasites were able to clear the infections without Dox treatment , possibly due to leakiness of the Cre gene and lower levels of Bdf2 , compromising growth . Nonetheless , mice fed with Dox pellets had extremely low or undetectable parasitemia when compared with regular pellet-fed mice ( Fig 6E ) . To provide a blueprint towards designing high-affinity trypanosome-specific bromodomain inhibitors , we determined the crystal structure of the Bdf2 bromodomain in complex with I-BET151 at 1 . 25Å resolution . The overall structure adopts the canonical bromodomain fold comprising four α-helices linked by variable loop regions that form the ligand binding site [39 , 40] ( Fig 7A and 7B ) . Unexpectedly , but consistent with the reduced affinity , the ligand is flipped by roughly 180° with respect to the human BRD4-BD1-I-BET151 complex and cannot adopt its classical binding mode ( Fig 7C ) [23] . In human BRD4-BD1 , the dimethylisoxazole group mimics AcK , recognizing the conserved Asn ( Asn140 ) and the conserved water network . In Bdf2 , the dimethylisoxazole ring is solvent-exposed , while recognition of the conserved Asn takes place through a bidentate hydrogen bond to the imidazolone moiety . The new orientation is probably due to a clash of the pendent pyridine ring with the large trypanosome gatekeeper residue Trp92 , which dominates ligand recognition through parallel π–π stacking and which is not found in any of the 61 human bromodomains at this position [41] . The flipped binding mode is further enforced by the narrow AcK binding slot that cannot accommodate the energetically favorable twist of the dimethylisoxazole with respect to the quinoline ring ( Fig 7D ) . Notably , the pyridine moiety forces the imidazolone to sit high up in the AcK binding pocket , resulting in an imperfect fit with a gap at the bottom of the pocket that is filled by an additional , unconserved water molecule ( Wat1 , Fig 7C ) . By contrast , the smaller gatekeeper residue Ile146 in human BRD4-BD1 allows the twisted dimethylisoxazole to engage in the classical AcK recognition ( Fig 7E ) . BRD9 is one of the closest human structural homologs to Bdf2 and also possesses an aromatic gatekeeper residue ( Tyr106 ) that recapitulates the parallel π–π stacking interaction with a triazolophtalazine compound within a narrow slot ( Fig 7F ) . However , the negative electrostatic surface potential of Bdf2 markedly differs from the slightly positively charged BRD9 pocket , highlighting the distinct character of the Bdf2 AcK binding site , a prerequisite for developing a trypanosome-specific drug . Thus , the structure illuminates how the compound may be optimized towards selective treatments for trypanosomiasis .
T . brucei differentiation from the BF to the PF requires extensive reprogramming that results in drastic changes in morphology , metabolism , and protein expression . These changes are reflected at the level of gene expression , with recent studies indicating that 25%–40% of all genes are differentially expressed as the cells transition from one form to another [10 , 12 , 42] . A number of mechanisms have been shown to be involved in the developmental regulation of gene expression , including alternative splicing [42] and mRNA stability mediated by regulatory sequences within 5’ and 3’ UTRs [6–8 , 43] . These regulatory sequences are , in turn , bound by RNA binding proteins . For example , ZFP3 associates with EP1 and GPEET mRNA and can interact with translation factors [44 , 45] . Our studies indicate that there may be an additional layer of developmental regulation of gene expression that is mediated at the DNA level . We show that inhibition of bromodomain proteins cause some transcriptional changes that are similar to those observed during the physiological transition from the BF to the PF . These transcription changes have functional consequences , such as the appearance of EP1 on the cell surface and the gradual disappearance of surface VSG . An alternate explanation for the data is that treatment with I-BET151 and/or knockdown of Bdf genes results in a global transcriptional dysregulation , such that all formerly silenced genes show increased transcription and active genes show decreased transcription . To try and determine whether this was the case for our I-BET151 time course , we split genes into those that had previously been shown to change during differentiation as defined by the Queiroz study , and those that have not . If global dysregulation were occurring , we would expect that in the group of genes not associated with differentiation , there should be an increase in median RPKM for genes that are poorly expressed at time 0 , and a decrease in median RPKM for genes that are highly expressed at that time point . Genes in the “procyclic” group , as defined by Queiroz et al . are poorly expressed in BF cells . We examined the range of starting RPKMs for this group of genes and extracted genes of similar starting RPKM from the set of genes that have not previously been shown to be involved in differentiation . We then plotted the median level of gene expression over time for this group of poorly expressed genes . We did not observe a significant increase in the median RPKM for this set of genes over the course of I-BET151 treatment ( S15 Fig ) , indicating that global disregulation is not occurring . We obtained similar results for the set of genes with high starting RPKM , whose range was chosen based on the range of expression for the genes found in the “glycolysis 1” group of genes defined by Queiroz et al . ( S15 Fig ) . We did not observe a general decrease in median RPKM for the set of highly expressed genes that were not previously shown to be involved in differentiation over the course of I-BET151 treatment . While we have not formally disproved global disregulation , we believe these findings argue for a more specific effect of I-BET151 treatment on the set of genes known to be involved in differentiation . In T . brucei , there is a paucity of sequence-specific DNA regulatory sequences that can serve as transcription factor binding sites . Only one transcription factor has been well characterized [46] , and it appears to primarily mediate Pol I transcription . To our knowledge , enhancer-like sequences have not been identified . It is thus unlikely that transcription factors are the means by which developmental transcriptional programs are put in place . This leaves epigenetic marks and the proteins that read them as a potential mechanism for regulating developmental changes in gene expression at the DNA level . Our work implies that bromodomain protein localization could change during differentiation from the BF to the PF in the normal course of the trypanosome lifecycle , and that this may be part of the mechanism by which a new , PF-specific transcriptional program is initiated . While we have demonstrated that bromodomain proteins bind chromatin , and that binding is altered following their inhibition , technical considerations currently preclude a direct demonstration that epigenetic changes ( e . g . , changes in histone marks ) that fundamentally change the structure of the chromatin surrounding bromodomain-bound sites are occurring . Future studies may be able to address whether changes in the patterns of histone tail modification or DNA methylation result from altered localization of chromatin readers during differentiation . We were able to demonstrate that bromodomain protein-mediated changes in transcriptional program are reversible ( Fig 4 ) , and we theorize that this plasticity may be a significant advantage to an organism that must reprogram itself quickly to adapt to the changing environments encountered between the insect and the mammalian host . Our results are in good agreement with a very recent study in mice , where small-molecule bromodomain protein inhibition in embryonic stem cells resulted in loss of the undifferentiated state and in spontaneous onset of differentiation [16] . The surprising parallels between bromodomain-mediated maintenance of cell fate in mouse embryonic stem cells and BF trypanosomes have interesting implications for how these processes may have evolved across diverse biological systems . We found that inhibition of bromodomain proteins caused growth arrest , both in our genetic knockdowns of Bdf2 and Bdf3 and upon treatment with I-BET151 ( Fig 6 and S8 Fig ) , albeit with different timing in each case . This is interesting in light of the fact that long slender BF trypanosomes cease to divide as they progress to the stumpy forms that are eventually ingested by the tsetse . Transcriptional profiling has revealed that many of the gene expression changes that take place during the transition from the BF to the PF are already in motion during the stumpy stage [10 , 11 , 42] , and thus some of the transcription changes that we observe in I-BET151-treated cells might be shared with cells transitioning from the long slender to the stumpy form . That is to say that it is possible that the cell cycle arrest observed upon bromodomain protein inhibition is a result of the transcriptional program that is set forth following inhibition . The Lister427 strain used in this study is not pleomorphic and thus cannot be induced to form stumpies . However , it may be that bromodomain inhibition causes the cells to acquire some characteristics of stumpy forms . That said , we were unable to detect PAD1 on the surface of I-BET151-treated cells [47] , and thus do not think the drug is inducing formation of a true stumpy form . Conversely , it is possible that the growth arrest is causing transcriptional changes consistent with progression to the next lifecycle stage . However , the fact that we see an up-regulation of EP1 mRNA at 3 h following I-BET151 treatment , and that we do not observe a pronounced growth arrest in the first 24 h of treatment does not support this notion . In future studies , it will be interesting to see what effect I-BET151 has on pleomorphic strains , and whether it is able to instigate transition to the stumpy form . We initially chose to focus on the BF cell fate because its loss could have important therapeutic implications . For example , disrupting immune evasion mechanisms specific to BF cells ( e . g . , antigenic variation or antibody internalization ) could lend the host immune system a much-needed boost in combating infection . However , the question as to whether bromodomain proteins are important for maintaining cell fate in other lifecycle stages , such as in PF cells , is an important one to be addressed in future studies . Evidence has accumulated over the years that chromatin structure and chromatin-interacting proteins are critical for maintaining monoallelic expression of VSG genes [17 , 48 , 49] . Consistent with these results , inhibiting bromodomain proteins that read AcK marks on the histone tails affects monoallelic expression . One previous study demonstrated that deletion of a histone methyltransferase caused cell cycle arrest and cell death after induction of differentiation from the BF to the PF [50] . Because many PF-specific genes were up-regulated following induction of differentiation , the authors suggested that the phenotype could be a byproduct of a cell cycle checkpoint defect , rather than a problem with initiation of differentiation . To our knowledge , our study is the first report of a chromatin factor being required to prevent premature differentiation prior to environmental cues . This is particularly interesting in light of the fact that previous reports have linked gene expression within the ESs to differentiation processes [27 , 29] , and we hypothesize that bromodomain proteins may be one of the factors that mechanistically tie these two processes together . One scenario for how this might occur is that an unidentified signaling event could trigger removal of bromodomain proteins from the ESs , resulting in up-regulation of mRNA from many different ESs , which in turn could cause active ES attenuation and initiate a PF transcription program , as envisioned by Batram et al . Conversely , a global removal of bromodomain proteins from transcription initiation sites could be the event by which a PF-specific transcription program is initiated , with active ES attenuation and transient derepression being part of this more global differentiation program . Our observation that inhibiting the more ES-specific factor Bdf2 caused less of a change in the global transcriptome , but that Bdf2-depleted cells were still “primed” to undergo differentiation to a larger degree than wildtype cells ( Fig 5E ) supports the first scenario . Our insights into this differentiation process have fundamental implications for the design of new strategies to treat trypanosomiasis , as inhibiting bromodomain proteins decreases virulence in a mouse model ( Fig 6 ) . This finding is important because trypanosomiasis is fatal if untreated and there is a paucity of drugs available for this disease [33] . Novel drug targets are necessary; as late-stage treatments are highly toxic , few new drugs are in development , and drug resistance is increasing [22] . The notion that bromodomain inhibitors can “trick” a BF trypanosome into prematurely initiating differentiation to the insect stage while still in the bloodstream imparts this class of molecules with substantial therapeutic potential . In addition to the adverse effects of I-BET151 on the parasite’s fitness itself ( Fig 6A and 6B and S8 Fig ) , the host immune system would have specific advantages over T . brucei: ( 1 ) expression of invariant insect stage proteins on the T . brucei surface provides a “handle” for a single antibody type to recognize the parasite ( in contrast to the constantly antigenically varying VSG coat that is native to T . brucei in the bloodstream ) and ( 2 ) blockade of rapid antibody internalization , another hallmark of the T . brucei BF , would allow more time for host effector cells to recognize and eliminate the parasites . While I-BET151’s low affinity for trypanosome bromodomain proteins currently limits its use in therapeutic intervention , our crystal structure of Bdf2 in complex with I-BET151 provides critical information to guide the design of a highly specific trypanosome bromodomain inhibitor . The fact that the architecture of the AcK recognition site of the Bdf2 bromodomain markedly differs from its mammalian counterparts ( Fig 7 ) implies that a trypanosome-specific bromodomain inhibitor can in principle be developed . This is particularly relevant , because various bromodomain inhibitors such as I-BET151 affect the mammalian immune response [22] , which could pose problems during an infection . Our structure provides specific clues , for example deletion of the pendent pyridine ring or expansion of the tricyclic aromatic ring system to fill the gap at the bottom of the binding site , for augmenting the drug-Bdf2 interactions to yield a trypanosome-specific , high-affinity ligand , while concurrently decreasing affinity for mammalian bromodomain proteins ( Fig 7C ) . This would avoid compromising the immune response in the host , while inhibiting trypanosome growth , offering a promising new avenue toward therapeutic intervention for trypanosomiasis .
T . brucei BF cells ( strain Lister 427 antigenic type MITat1 . 2 clone 221a ) were cultured in HMI-9 at 37°C . We used either a “single marker” ( SM ) line expressing T7 RNA polymerase and Tet repressor ( TETR ) [51] , or a dual BES marked line ( L224 ) that expresses VSG3 from BES7 ( with NeoR downstream of the promoter ) and a PuroR gene downstream of the BES2 promoter ( expressing VSG2 ) [28] . Stable clones were maintained in HMI-9 medium containing necessary antibiotics . Trypanosomes were maintained at the following drug concentrations , unless otherwise stated: 2 . 5 μg/ml , G418 ( Sigma ) ; 5 μg/ml , blasticidin ( Invivogen ) ; 5 μg/ml , hygromycin ( Invivogen ) ; 0 . 1 μg/ml , puromycin ( Invivogen ) ; 1μg/ml , phleomycin ( Invivogen ) . The Bdf2KO strain was generated by cloning the coding sequence of Bdf2 ( Tb427 . 10 . 7420 ) upstream of an HA tag in the pMOTAG5H loxP plasmid . A portion of the 3’UTR was cloned downstream of a phleomycin resistance gene and cells were selected in phleomycin following transfection with an Amaxa nucleofector kit . The remaining endogenous allele of Tb427 . 10 . 7420 was replaced with a hygromycin resistance gene using 5’ and 3’ UTR homology arms cloned into the pyrFEKO-HYGRO loxP plasmid . The pLEW100cre-EP1-6G Phleo plasmid was integrated into the cell line by selecting for Phleomycin resistance following transfection to allow for inducible CRE expression as in [52] . The Bdf3kd strain was generated by cloning the coding sequence of Bdf3 ( Tb427 . 01 . 1830 ) upstream of an HA tag in the pMOTAG5H plasmid . A portion of the coding sequence of Tb427 . 01 . 1830 was cloned into the RNAi vector P2T7TABlue and integrated into the rDNA locus . All cell lines were checked for proper integration using PCR . The cell line used for ChIP ( HSTB-907 ) was generated by first randomly integrating a cassette containing puromycin-resistance gene , luciferase , and EmGFP ( pHJ1 , reference ) at a silent BES promoter , and then targeting a blasticidin resistance marker at the active BES promoter ( BES1-VSG2 ) . To determine which silent BES promoter was targeted with pHJ1 , in situ switched cells were selected at 100x higher concentration of puromycin ( 10 μg/ml ) and confirmed by GFP-FACS ( HSTB-717 ) . VSG RNA analysis confirmed that the switched cells express VSG11 in BES11 . Cells were treated with 20 μM I-BET151 in DMSO to inhibit bromodomains , or with 1 μg/ml doxycyclin to induce deletion of Bdf2 or RNAi against Bdf3 . 5–10 million cells were suspended in 2X Laemmli Sample Buffer ( Biorad 161–0737 ) and separated on a 4%–20% SDS gradient gel . After transfer to PVDF membrane , blots were blocked in 5% milk in PBS and probed with anti-HA HRP ( Miltenyi 130-091-972 ) , anti-H3 ( Abcam 1791 ) , or anti-tubulin ( gift from Keith Gull ) followed by anti-rabbit IgG-HRP or anti-mouse IgG-HRP . 1–2 million treated and control cells were suspended in HMI-9 and stained with fluorescence conjugated anti-VSG2 or anti-VSG3 for 10min with vortexing [53] . Cells were washed twice in HMI-9 before analysis on either a BD LSRII or a BD FACSCalibur . PI ( BD 51–66211 ) was added to the final suspension for dead cell exclusion at 0 . 75μg/ml . L224 , Bdf2KO , or Bdf3kd trypanosomes were treated for 23 d with I-BET151 and Dox , respectively . Cells were washed twice in media to eliminate I-BET151 or Dox , resuspended in HMI-9 and subjected to 24 h of treatment at 27°C , or in 6mM cis-aconitate ( Sigma A3412 ) . Control cells were kept at 37°C . Cells were stained with anti-EP1 ( Cedarlane CLP001A ) and an anti-mouse IgG FITC-labeled secondary antibody ( BD 349031 ) and analyzed by flow cytometry . L224 , Bdf2KO , or Bdf3kd trypanosomes were treated for 2–3 d with I-BET151 and Dox , respectively . Cells were stained with unconjugated anti-VSG3 ( L224 ) or anti-VSG2 ( Bdf2KO or Bdf3kd ) primary antibody and washed . Cells were subjected to a 5 min incubation at either 4°C ( controls ) or at 37°C to allow internalization to take place . Cells were immediately fixed in 1% formaldehyde and stained with a FITC-conjugated secondary antibody ( BD 349031 ) to measure the remaining amount of primary antibody on the surface . Cells were then analyzed by flow cytometry . For immunofluorescence , cells were adhered to poly-L-lysine coverslips following fixation , blocked in PBS with 0 . 2% cold water fish gelatin ( Sigma G7765 ) and 0 . 5% ( w/v ) BSA and stained with anti-mouse IgG FITC ( BD 349031 ) at 1:1 , 000 for 1 h . Cells were washed , counterstained with DAPI and imaged on a DeltaVision Image Restoration inverted Olympus IX-70 microscope ( Applied Precision ) . L224 trypanosomes were treated with I-BET151 or DMSO for 2–3 d and 1 . 5 million live cells in media were imaged in a MatTek glass-bottom dish ( P35G-0 . 17-14-C ) at 1 ms intervals on a DeltaVision Image Restoration inverted Olympus IX-70 microscope ( Applied Precision ) . Cells were fixed for 5 min in 1% formaldehyde in PBS . After 3 washes , cells were resuspended in solution containing 0 . 2 mg/ml RNAse and 0 . 05 mg/ml PI in PBS . Cells were incubated for 2 h at 37°C and analyzed by flow cytometry . Trypanosomes treated with either I-BET151 or Dox were diluted to 100 , 000 cells/ml in HMI-9 every 24 h , counted on a hemacytometer and compared to DMSO control-treated cells or untreated cells ( for Dox treatments ) . Three independent cultures of trypanosomes were treated with I-BET151 for 3 d or with DMSO as a control . Following treatment cells were washed , counted on a hemacytometer and then serially diluted to a concentration of 5 cells/100 μL HMI9 or 50 cells/10 0μL HMI9 with no drug . They were plated on three 96-well plates and cultured for 3–4 d at 37°C before quantifying the number of wells growing in each plate . An aliquot of each culture was subjected to PI staining ( BD 51–66211 ) at 0 . 75 μg/ml to assess viability and countbright beads ( Life technologies C36950 ) were added to quantify cell growth . Washed cells were allowed to recover for 3 d before being subjected to PI staining again . RNA was extracted from treated or control cells using RNA Stat-60 ( Tel-Test ) following the manufacturer’s protocol and quantified on a NanoDrop2000c . Following DNAse treatment on 5 μg of RNA , poly ( A ) + sequencing libraries were prepared using the NEBNext Poly ( A ) mRNA Magnetic Isolation Module ( E7490S ) followed by the NEBNext Ultra Directional RNA library prep kit for Illumina ( E7420L ) or by the Rockefeller University Genome Sequencing Center using the Illumina kit . Sequencing was performed on an Illumina HiSeq 2000 sequencer using 100 bp reads . Reads were trimmed for quality with TrimGalore from Babraham Bioinformatics ( http://www . bioinformatics . babraham . ac . uk/projects/trim_galore/ ) using stringency setting three and aligned first to the reference genome ( Tb927v5 . 1 ) and then to the Lister 427 VSGnome [1] using bowtie [54] to uniquely align reads and allowing for two mismatches . For the ESAG analysis , reads were uniquely aligned allowing for zero mismatches to unambiguously determine the BES . Example commands are given below . trim_galore—stringency 3 L224-1_CGATGT_L002_R1_001 . fastq . gz bowtie—best—strata -t -v 2 -a -m 1—sam—un UN_L224-iB_1_140804_tb927_v5 . 1 /Users/dschulz/To_Clean/bowtie-0 . 12 . 8/indexes/140804_tb927_v5 . 1 L224-1_CGATGT_L002_R1_001 . fq L224-iB_1_140804_tb927_v5 . 1 . sam bowtie—best—strata -t -v 0 -a -m 1—sam /Users/dschulz/To_Clean/bowtie-0 . 12 . 8/indexes/141107_VSGs_CDSs_ESAGs UN_L224-iB_1_140804_tb927_v5 . 1 L224-iB_1_141107_VSGs_CDSs_ESAGs_m1_v0 . sam RPKM values were quantified using SeqMonk from Babraham Bioinformatics ( http://www . bioinformatics . babraham . ac . uk/projects/seqmonk ) or , in the case of Vsg- and ESAG-aligned reads , custom python scripts . All sequencing reads were performed in biological duplicate or triplicate and RPKM values were averaged . For reads aligning to the genome , DESeq [24] was used to generate p-adjusted values using the negative binomial test for differences between the base means for two conditions with the following command: nbinomTest ( cds , condA , condB ) . For reads aligning to VSGs and ESAGs , Q values were calculated using SeqMonk’s statistical replicate set test , which adjusts the p-values using Benjamimi and Hochberg correction . Note that the Bdf2 dataset had to be sequenced more deeply for ESAG analysis because of the stringent parameters used ( unique alignments with no mismatches ) . This dataset was sequenced in duplicate rather than triplicate , so no p-values were generated for the Bdf2 ESAG analysis shown in S9 Fig . Plots from RNA-seq were generated using python’s matplotlib library . Boxplots were generated using the command plt . boxplot ( data ) . Scatterplots were generated using plt . scatter ( x , y ) . All other plots were generated using plt . plot ( data ) . GSEA analysis [25] was conducted using software available at http://www . broadinstitute . org/gsea/index . jsp . We used custom sets of functional groups and hierarchical clusters defined in [35] . These sets are provided as Additional File 3 and Additional File 4 in S1 Data . For time course data , GSEA was run using the Pearson metric for ranking genes in a time series with the following parameters: Permutations: 1000 Collapse dataset to gene symbols: False Permuation type: gene_set Enrichment statistic: weighted Metric for ranking genes: Pearson Gene list sorting mode: real Gene list ordering mode: descending Max size: 500 Min size: 1 An example command is given below java -Xmx512m xtools . gsea . Gsea -res /Users/danaeschulz/Science/Tryp_Science/15-7-7_L224_iBET_RNAseq_timecourse/15-7-19_L224_timecourse_Gsea_analysis/L224_timecourse_continuous_data_for_GSEA . txt -cls /Users/danaeschulz/Science/Tryp_Science/15-7-7_L224_iBET_RNAseq_timecourse/15-7-19_L224_timecourse_Gsea_analysis/L224_timecourse_continuous_profile . cls#IncreasingProfle -gmx /Users/danaeschulz/Science/Tryp_Science/15-7-19_GSEA_analysis_L224_iBET_Bdf3_Bdf2/mappable_clayton_clusters_GMT . gmt -collapse false -mode Max_probe -norm meandiv -nperm 1000 -permute gene_set -rnd_type no_balance -scoring_scheme weighted -rpt_label my_analysis -metric Pearson -sort real -order descending -include_only_symbols true -make_sets true -median false -num 100 -plot_top_x 20 -rnd_seed timestamp -save_rnd_lists false -set_max 500 -set_min 1 -zip_report false -out /Users/danaeschulz/gsea_home/output/aug06 -gui false For analysis at single time points , genes were ranked using Signal2Noise with the following parameters: Permutations: 1 , 000 Collapse dataset to gene symbols: False Permuation type: gene_set Enrichment statistic: weighted Metric for ranking genes: Signal2Noise Gene list sorting mode: real Gene list ordering mode: descending Max size: 500 Min size: 1 An example command is given below java -Xmx512m xtools . gsea . Gsea -res /Users/danaeschulz/Science/Tryp_Science/15-7-7_L224_iBET_RNAseq_timecourse/15-7-19_L224_timecourse_Gsea_analysis/L224_timecourse_continuous_data_for_GSEA . txt -cls /Users/danaeschulz/Science/Tryp_Science/15-7-7_L224_iBET_RNAseq_timecourse/15-7-19_L224_timecourse_Gsea_analysis/L224_timecourse_continuous_profile . cls#IncreasingProfle -gmx /Users/danaeschulz/Science/Tryp_Science/15-7-19_GSEA_analysis_L224_iBET_Bdf3_Bdf2/mappable_clayton_clusters_GMT . gmt -collapse false -mode Max_probe -norm meandiv -nperm 1000 -permute gene_set -rnd_type no_balance -scoring_scheme weighted -rpt_label my_analysis -metric Signal2Noise -sort real -order descending -include_only_symbols true -make_sets true -median false -num 100 -plot_top_x 20 -rnd_seed timestamp -save_rnd_lists false -set_max 500 -set_min 1 -zip_report false -out /Users/danaeschulz/gsea_home/output/aug06 -gui false Chromatin IPs were performed as in [35] except that a rabbit anti-HA antibody was used ( Sigma H6908 ) , and DNA was purified by phenol-chloroform extraction or with Ampure beads ( Agencourt A63880 ) . ChIP qPCR primers are listed in S13 Table . ChIP-Seq libraries were made as in [35] and run on an Illumina HiSeq 2000 sequencer using 50 bp reads . Reads were trimmed for quality with TrimGalore and aligned to the reference genome ( Tb927v5 ) using bowtie [54] to uniquely align reads and allowing for two mismatches . The MACS algorithm was used to identify peaks in the ChIP-seq data set [55] . An example command is as follows macs14 -t A5IP_tb927v5fix5with12 . bed -c A5In_tb927v5fix5with12 . bed -n A5IP_MACS -f BED -g 23650671 . Each replicate was compared to input to obtain an initial set of called peaks . Bedtools was then used to compare the two replicates for each Bdf , and return only those peaks that were called in both replicates . Finally , peaks called by MACS in the untagged control were eliminated from the final set of called peaks . For Fig 5B , only peaks with an FDR of <0 . 1 were used for the analysis . All sequencing reads were performed in biological duplicate or triplicate . For Q-PCR to quantify transcription , RNA was extracted from treated or control cells using RNA Stat-60 ( Tel-Test ) following the manufacturer’s protocol and quantified on a NanoDrop2000c . 5 μg of RNA was used to generate cDNA using random hexamer primed Superscipt III Reverse transcriptase ( Life Technologies 18080093 ) according to the manufacturer’s protocol . For Q-PCR to quantify ChIP , DNA was purified using Agencourt Ampure XP beads ( Agencourt A63880 ) . For amplification , cDNA or ChIP purified DNA was amplified using 2X Sybr green master mix ( Life Technologies4309155 ) and primers and quantified on an Applied Biosystems 7900HT Sequence Detection System . Primers used for Q-PCR are listed in S13 Table . I-BET151-treated trypanosomes and control cells were stained with antibodies against VSG in the transcriptionally active ES ( VSG3 ) and a VSG in a silent ES ( VSG2 or VSG13 ) as well as DAPI for dead cell exclusion . Cells were analyzed and photographed on an Amnis ImageStream-X flow cytometer . C57/B6 or RAG-/- ( Jackson Labs #002216 ) mice were fed with Dox feed ( Doxycycline 5053 200 ppm ) or regular feed ( Lab Diet 5053 ) for 24–48 h before infection or following the appearance of parasites . One million Bdf2KO or Bdf3kd trypanosomes suspended in HMI-9 were IP injected into each mouse . Mice were kept on either Dox or regular feed for the remainder of the experiment . Parasitemia was monitored daily starting 48 h postinfection by tail bleed , dilution , and counting on a hemacytometer . Mice were monitored for 7–20 d . Control experiments were performed where parasitemia was monitored in mice infected with wildtype parasites fed on either Dox feed or regular feed to ensure that Dox did not influence virulence ( control_mouse_experiment file in S6 Data ) . Infections proceeded similarly in these two groups . For the I-BET151 experiments , trypanosomes were pretreated in vitro for 2–3 d with 20 μM , and 50 trypanosomes were injected by IP into each mouse . Mice were infected between 4–8 wk of age . All studies were conducted in accordance with the GSK Policy on the Care , Welfare and Treatment of Laboratory Animals and were reviewed by the Institutional Animal Care and Use Committee at GSK and by the ethical review process at The Rockefeller University . DNA fragments of wild-type Bdf29-123 and wild-type Bdf332-150 were amplified by PCR from genomic DNA , cloned into a pET28a vector ( Novagen ) containing an N-terminal PreScission protease ( GE Healthcare ) cleavable His6-tag , and overexpressed in E . coli . Mutations in Bdf29-123 ( Y43A and N86A ) and Bdf332-150 ( Y80A and N124A ) were introduced by overlap extension PCR mutagenesis . For formation of the inhibitor complex , 10 mg/ml of purified Bdf29-123 was mixed in a 1:1 molar ratio with I-BET151 and incubated for 1 h on ice . The crystallization solution consisted of 2 . 2 M Na/K PO4 , pH 6 . 6 , 0 . 1 M NaOAc pH 4 . 1 , and 4% ( v/v ) 1-propanol . Crystals grew within two weeks . X-ray diffraction data were collected at the X29 beamline at the National Synchrotron Light Source ( NSLS ) at the Brookhaven National Laboratory ( BNL ) . Diffraction data were processed in HKL2000 [56] , the structure was solved by molecular replacement using individual α-helices one at a time as search templates with the program Phaser [57] , and an initial core model was built by ARP/warp [58] . Model building was performed in O [59] and Coot [60] . The model of Bdf2 was verified by peaks in an anomalous difference density map from data collected at 1 . 73Å that coincided with the sulfur positions of the cysteine and methionine residues . The final model spanning residues 9–114 was refined in Phenix [61] to an Rfree of 17 . 9% with excellent stereochemistry as assessed by MolProbity [62] . Details for data collection and refinement statistics are summarized in S11 Table . Atomic coordinates and structure factors have been deposited with the Protein Data Bank under PDB code 4PKL . ITC measurements were performed at 15°C using a MicroCal auto-iTC200 calorimeter ( MicroCal , LLC ) . Protein samples were extensively dialyzed against a buffer containing 20 mM Hepes , pH 7 . 5 , 150 mM NaCl , 0 . 5 mM TCEP , and 1% DMSO . Typically 5 μL of 2 . 0 mM protein was injected into 0 . 4 mL of 0 . 2 mM ligand in the chamber every 150 s . Baseline-corrected data were analyzed with ORIGIN software . Antibodies were generated at the Memorial Sloan Kettering Cancer Center Monoclonal Antibody Facility and conjugated with either A488 or APC fluorophores . All statistical tests were Student’s t tests except for those performed in Fig 6C and 6D . In Fig 6C and 6D , log-rank Mantel-Cox tests were performed on survival of the three groups . For Fig 6E , because parasites were below the lower level of quantification ( LLQ ) , we assigned values = ½ ( LLQ ) to those mice for which we could not get quantifiable parasite counts . A Mann-Whitney U test was then performed to compare Dox-feed and regular-feed groups of mice . The LLQ for these experiments is 5 . 56 x 105 parasites/ml . | Many parasites undergo major changes in lifestyle as they cycle between their vectors and their hosts . We use Trypanosoma brucei , the causative agent of human and animal African Trypanosomiasis , as a model to study how these changes are regulated and maintained at the level of gene expression . T . brucei lives in the bloodstream of the mammalian host before migrating to the insect through the bite of its insect vector , the tsetse fly . It escapes immune detection by varying the proteins on its surface and by rapidly internalizing host antibodies . We report here that a class of proteins , called bromodomain proteins , help maintain the identity of the parasite in its bloodstream form . When these proteins are inhibited , the parasites express an unvarying epitope that is usually expressed only at the insect stage and are compromised in their ability to internalize host antibodies . Bromodomain proteins bind to DNA that is wrapped around histone proteins , acting as mediators that interact with the transcription machinery to determine which genes are turned on and which are kept repressed . Our crystal structure of a bromodomain inhibitor in complex with a trypanosome bromodomain reveals a novel binding mode and demonstrates how these small molecule inhibitors could be optimized for therapeutic use in a parasite-specific manner . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | Bromodomain Proteins Contribute to Maintenance of Bloodstream Form Stage Identity in the African Trypanosome |
Staphylococcus aureus colonizes the nose , throat , skin , and gastrointestinal ( GI ) tract of humans . GI carriage of S . aureus is difficult to eradicate and has been shown to facilitate the transmission of the bacterium among individuals . Although staphylococcal colonization of the GI tract is asymptomatic , it increases the likelihood of infection , particularly skin and soft tissue infections caused by USA300 isolates . We established a mouse model of persistent S . aureus GI colonization and characterized the impact of selected surface antigens on colonization . In competition experiments , an acapsular mutant colonized better than the parental strain Newman , whereas mutants defective in sortase A and clumping factor A showed impaired ability to colonize the GI tract . Mutants lacking protein A , clumping factor B , poly-N-acetyl glucosamine , or SdrCDE showed no defect in colonization . An S . aureus wall teichoic acid ( WTA ) mutant ( ΔtagO ) failed to colonize the mouse nose or GI tract , and the tagO and clfA mutants showed reduced adherence in vitro to intestinal epithelial cells . The tagO mutant was recovered in lower numbers than the wild type strain in the murine stomach and duodenum 1 h after inoculation . This reduced fitness correlated with the in vitro susceptibility of the tagO mutant to bile salts , proteases , and a gut-associated defensin . Newman ΔtagO showed enhanced susceptibility to autolysis , and an autolysin ( atl ) tagO double mutant abrogated this phenotype . However , the atl tagO mutant did not survive better in the mouse GI tract than the tagO mutant . Our results indicate that the failure of the tagO mutant to colonize the GI tract correlates with its poor adherence and susceptibility to bactericidal factors within the mouse gut , but not to enhanced activity of its major autolysin .
Staphylococcus aureus is a bacterial pathogen that commonly colonizes the nose , skin , and mucosal surfaces of healthy individuals . However , S . aureus may also cause a variety of superficial and invasive infections in hospitalized patients , as well as in individuals within the community who lack the risk factors commonly associated with nosocomial infections [1 , 2] . Although the anterior nares are the most common anatomic site of S . aureus carriage , ~20% of adults are positive for intestinal carriage of S . aureus [3] . The gastrointestinal ( GI ) tract has been shown to be a potentially important reservoir for S . aureus in several clinical studies [4–6] . Although nasal carriage apparently predisposes the host to intestinal carriage , ~37% of intestinal carriers are not positive for S . aureus nasal colonization [3] . Compared to nasal colonization only , simultaneous nasal and intestinal colonization was associated with a significant increase in the frequency of positive skin cultures [7] . Squier et al . [8] observed that critically ill patients who had both rectal and nasal carriage were significantly more likely to develop staphylococcal infection ( 40% infection rate ) than those with nasal carriage only ( 18% infection rate ) . Patients positive for staphylococcal GI colonization often contaminate their environment with S . aureus [3 , 9] . Thus , intestinal carriage may serve as an important reservoir for S . aureus transmission , contributing to bacterial dissemination and subsequent risk of infection [3] . Faden et al . compared methicillin-resistant S . aureus ( MRSA ) nasal and rectal colonization rates in children with staphylococcal skin abscesses and a control group of children without staphylococcal disease [10] . Whereas rates of nasal colonization were equivalent for both groups of children , MRSA was detected significantly more often in the rectum of children with skin abscesses ( 47% ) compared with controls ( 1% ) . Moreover , S . aureus recovered from the abscesses and rectum were identical in 88% of cases , compared with 75% of nasal isolates . Almost all abscess isolates ( 57/60 ) were USA300 strains , whereas only 2 of 22 isolates from the control groups were USA300 . In a study of HIV-infected men who have sex with men , Szumowski et al . reported that perianal ( but not nasal ) colonization by MRSA was significantly associated with skin abscess formation [11] . These studies suggest that rectal colonization by S . aureus , probably reflecting gastrointestinal carriage , is an important reservoir from which person to person transmission occurs . Host factors that facilitate staphylococcal colonization of the GI tract are poorly understood . Intestinal carriage occurs at a high frequency within the first six months of life , after which the incidence drops [3] . Additional factors , such as decreased stomach acidity , antibiotics that disrupt the indigenous microbiota , or the administration of cyclophosphamide or prednisone , may also influence acquisition of S . aureus in the human GI tract [4–6] . Whereas staphylococcal factors that promote GI colonization have not been reported , several independent investigations have identified surface antigens that impact S . aureus nasal colonization in rodent models . Mutants deficient in either clumping factor B or cell wall teichoic acid ( WTA ) showed reduced nasal carriage in experimentally inoculated rats or mice [12 , 13] , and clumping factor B promoted nasal colonization of humans [14] . In this investigation , we sought to develop and characterize a reliable murine model of S . aureus GI carriage to better understand its relevance as a risk factor for subsequent infection and its potential for transmission and spread of this pathogen . S . aureus antigens important for nasal colonization were assessed to determine whether they might also play a role in colonization of the GI tract . We identified several surface-associated S . aureus antigens that modulated colonization of the GI tract and identified WTA as critical for the early steps in colonization . The failure of the WTA mutant ( ΔtagO ) to colonize the GI tract correlated with its defects in bacterial adherence and greater susceptibility to antimicrobial factors within the mouse gut .
The mouse intestinal tract is comprised of diverse commensal bacteria , including Bacteroides , Clostridia , segmented filamentous bacteria , members of the Enterobacteriaceae , Lactobacilli , and Enterococci [15 , 16] . These normal flora provide some protection against invading pathogens , as evidenced by the fact that we were unable to establish stable ( ≥ 1 week ) GI colonization by S . aureus in conventional mice in the absence of selective antibiotic pressure ( S1 Fig ) . In previous studies , we maintained stable nasal colonization of mice with S . aureus by supplementing their drinking water with streptomycin ( Sm ) and inoculating with Sm-resistant ( Smr ) staphylococcal isolates [12 , 17] , and a similar approach was taken here . Fecal cultures performed on mice prior to inoculation were negative for Smr S . aureus ( lower limit of detection ~3 log CFU S . aureus/g stool ) . Awake mice maintained on Sm water and administered intranasal inocula of Smr S . aureus Newman ranging from 8 x 108 to 2 x 105 CFU showed stable colonization of the GI tract for at least 3 wks ( Table 1 ) . Recovery of Smr S . aureus Newman from the stool cultures consistently averaged ~105 CFU/g stool . Differences in GI colonization of the WT strain were not observed when we compared inoculation by the intranasal route with inoculation by oral gavage . A subset of animals were given Sm water during week 1 and then given regular drinking water thereafter . These mice maintained GI colonization for at least 3 weeks at levels ( ~105 CFU/g stool ) similar to those of mice maintained for 3 wks on Sm water ( S1 Fig ) . To determine where S . aureus resided in the GI tract of conventional mice , we inoculated animals with ~109 CFU of S . aureus , and euthanized the animals on days 1 , 2 , 4 and 7 . Segments of the GI tract were removed and cultured quantitatively . The colonized staphylococcal density in the conventional mice was quite variable in the small intestine with the highest being observed in the distal segment of the small intestine ( 1 . 3 x 105 CFU/g tissue ) , cecum ( 3 . 9 x 105 CFU/g tissue ) and colon ( 3 . 2 x 105 CFU/g ) . To measure the impact of the commensal GI flora on S . aureus colonization , germfree Swiss Webster mice ( n = 8 ) were inoculated with ~109 CFU of S . aureus , and quantitative stool cultures were performed on days 1 , 5 , 7 , 12 , and 14 . In contrast to conventional mice , germfree mice were readily colonized by S . aureus as early as day 1 without the need for Sm drinking water . Quantitative cultures yielded concentrations ranging from 1 x 107 to 9 . 7 x 108 CFU/g stool over the course of the two-week time period . Four to five mice were euthanized on day 7 or 14 , and segments of the GI tract were removed , homogenized , and cultured quantitatively . Whereas S . aureus was detected throughout the GI tract at concentrations >104 cfu/g , significantly higher staphylococcal densities were achieved in the distal portions of the GI tract ( Table 2 ) . These findings corroborated those that we found in conventional mice , i . e . , the highest bacterial burdens were localized to the mouse cecum and colon . Previous studies revealed that S . aureus mutants deficient in sortase ( ΔsrtA; affects all cell wall anchored proteins ) , clumping factor B , the serotype 5 capsule , and WTA were defective in their ability to colonize the nasal cavity of rodents [12 , 13 , 17 , 18] . Evaluation of microbial fitness for colonization of the GI tract is often performed in the context of competition experiments [19–21] . When we evaluated the relative fitness of S . aureus mutants vs . the parental strain , we observed that the acapsular Newman cap5G mutant showed enhanced fitness , colonizing the mouse GI tract in numbers greater than that of the parental type 5 capsule positive strain ( Fig 1 ) . The sortase A mutant and a clumping factor A ( clfA ) mutant showed impaired fitness in vivo , since both were out-competed by the wild type ( WT ) strain Newman ( Fig 1 ) . To validate the latter finding , we repeated the competition experiment between strain Newman and its clfA mutant with inoculation by oral gavage . Similar to our initial findings , the clfA mutant again showed a significant colonization defect between 7 and 21 days after inoculation ( S2 Fig ) . Newman Δica , Newman ΔclfB , Newman Δspa , and Newman ΔsdrCDE showed no colonization defect ( S3 Fig ) . When we inoculated separate groups of 5 to 10 mice with the parental strain Newman or isogenic mutants lacking the serotype 5 capsule , sortase , or ClfA , we observed no differences between the colonization levels achieved by the parental or mutant strains ( S4 Fig ) . Thus , the colonization defects of these three mutants were only evident as measured by in vivo competition experiments . Likewise , a mutant defective in beta hemolysin ( Hlb ) colonized the mouse gut at levels consistent with the respective parental strains ( S4 Fig ) . Weidenmaier et al . reported that a tagO mutant failed to colonize the nasal cavity of cotton rats [13] . To elucidate the role of WTA in GI colonization , separate groups of mice were inoculated with S . aureus SA113 or Newman or their isogenic tagO or dltA mutants . Both SA113 ΔtagO and Newman ΔtagO showed a significant ( P <0 . 01 ) reduction in GI colonization as early as one week after intranasal inoculation compared to the parental strain ( Fig 2A ) , despite the fact that the growth of the mutant in vitro was comparable to that of the parental strain [22] . To confirm these findings , additional groups of mice were inoculated by oral gavage with strain Newman or its tagO mutant , and GI colonization was monitored up to 72 h . By 24 h after inoculation , there were significantly fewer Newman ΔtagO recovered from the stools of mice inoculated with Newman ΔtagO compared to the WT strain ( S5 Fig ) . These data suggest that the WTA polymer is a critical determinant for colonization of the GI tract by S . aureus . To determine whether modifications of the WTA backbone affect GI colonization , we inoculated mice with dltA mutants of strain SA113 or Newman , which lack the ester-linked alanine substituent on staphylococcal teichoic acid . The dltA mutants showed no colonization defect ( Fig 2A ) . Similarly , a ΔtarM ΔtarS double mutant of S . aureus RN4220 , which lacks alpha- and beta-O-linked GlcNAc modifications of WTA , colonized mice at levels similar to the WT strain ( S6 Fig ) . WTA has been reported to promote staphylococcal adherence to human nasal epithelial cells and endothelial cells [13 , 23] . As a result , we compared the relative in vitro adherence of strains Newman and Newman ΔtagO to T84 human intestinal epithelial cells . As shown in Fig 2B and 2C , strain Newman showed significantly ( P = 0 . 0009 ) greater adherence to the T84 cells than did its ΔtagO mutant . Likewise , the clfA mutant , which showed an in vivo fitness defect ( Fig 1C ) , also showed impaired adherence to T84 cells ( Fig 2B and 2C; P = 0031 ) . The acapsular cap5G mutant , on the other hand , showed a modest increase in overall adherence to the epithelial cells , consistent with previous studies showing that the capsule can mask adhesins that mediate attachment to mammalian cells [24 , 25] . To determine whether a WTA-deficient mutant was competent for transit through the stomach and duodenum to establish a colonization niche in the intestines , we performed several short-term in vivo experiments . Mice were inoculated with either strain Newman or the Newman tagO mutant and euthanized after 1 h . Recovery of S . aureus by quantitative cultures of the stomach and four different segments of the GI tract were quite variable among different mice . For this reason , we performed subsequent experiments in a competition format wherein mice were inoculated with a 1:1 mixture of Newman and Newman ΔtagO . A control in vitro competition experiment whereby Newman and its tagO mutant were cultured together for 1 h showed a CI of 0 . Strain Newman and its tagO isogenic mutant were co-inoculated into mice , and after 1 h segments of the gut were homogenized and cultured quantitatively . The Newman ΔtagO mutant exhibited a substantial fitness defect in vivo with median CIs of 0 . 33 and 0 . 92 in the murine stomach and duodenum , respectively ( P < 0 . 05 ) ( Fig 3A ) . Few S . aureus colonies were recovered from the fourth segment of the small intestine , the cecum , or the colon , and consequently these data were not analyzed further . Our findings indicate that the ΔtagO mutant is significantly impaired in surviving transit through the mouse stomach and duodenum , and that its failure to colonize the GI tract may reflect its inherent susceptibility to antimicrobial factors in the gut . Because the tagO mutant survived poorly in vivo after inoculation , we hypothesized that Newman ΔtagO might be more susceptible than the WT strain to bile salts or low pH . As shown in Fig 3B , the tagO mutant was more susceptible than strain Newman to killing by bile salts ( 0 . 075% sodium deoxycholate ) , and the killing was rapid ( ~1 h ) . Because the concentration of bile salts in hepatic bile can reach concentrations as high as 1 . 66% [26] , the enhanced susceptibility of the tagO mutant may contribute to its poor recovery from the gut after only 1 h in vivo . In contrast , both the parental and mutant strains were killed under low pH conditions ( glycine buffer , pH 3 . 0–3 . 6 ) , and no differences in their relative susceptibilities to acidic conditions were observed ( S7 Fig ) . The antimicrobial peptide α-defensin 5 , secreted by mammalian Paneth cells , plays a role in the intestinal host defense against bacterial pathogens . Furthermore , α-defensins demonstrate homeostatic control of the host commensal microbiota [27] . Our in vitro studies indicated that the Newman ΔtagO mutant was killed to a significantly greater extent than the WT strain after a 2 h exposure to α-defensin 5 concentrations ranging from 5–10 μM ( Fig 3C ) , which are far below the 14–70 μM concentrations found in the human intestinal lumen [28] . Bacteria transiting the GI tract are also exposed to proteolytic enzymes . Although we could not demonstrate consistent differences in susceptibility during short-term exposures to proteases ( S8 Fig ) , the tagO mutant was more sensitive than the parental strain Newman to overnight treatments with proteinase K ( Fig 3D ) , pepsin ( at low pH ) and trypsin ( S8 Fig ) . As noted above , when 109 CFU of Newman and its tagO mutant were incubated together in vitro for 1 h at 37°C in tryptic soy broth ( TSB; Difo , Sparks , MD ) , the competitive index was 0 , i . e . , both strains were recovered in equivalent numbers . However , if the two strains were incubated together overnight in TSB , the tagO mutant could not be recovered from the culture ( >107 reduction in CFU compared to strain Newman ) . Similar experiments performed with WT strain SA113 and its tagO mutant yielded comparable results ( ~105 reduction in the SA113 ΔtagO CFU compared with SA113 ) . Filter-sterilized culture supernatants of S . aureus Newman ( but not culture supernatants from the tagO mutant ) also showed bactericidal activity toward Newman ΔtagO within 1 to 3 h of incubation ( Fig 4A and 4B ) . The bactericidal activity of the Newman culture supernatant was lost if it was boiled for 5 min prior to inoculation with Newman ΔtagO ( Fig 4A ) . In contrast , the addition of a protease inhibitor cocktail to the supernatant had no effect on its killing activity ( Fig 4A ) . Consistent with this finding , we observed no differences in the protease activity of culture supernatants derived from the WT or the ΔtagO mutant strain when measured on casein agar plates , azocasein assays , or the PDQ protease assay [Athena Environmental Sciences , Inc . , Baltimore , MD] ( S9 Fig ) . Peptidoglycan fragments added to filter-sterilized culture supernatants of strain Newman inhibited killing of the ΔtagO mutant in a dose dependent manner ( Fig 4B ) . This finding suggested that autolysins present in the Newman culture supernatant might be lysing the tagO mutant . Indeed , zymogram analysis revealed that the Newman culture supernatant showed greater autolytic activity than that of the tagO mutant ( Fig 4C ) . That the bactericidal effect on the WTA mutant was due to exogenous autolytic activity was supported by the observation that both the tagO and an autolysin ( atl ) tagO double mutant showed a >2-log reduction in CFU/ml following a 3-h exposure to the Newman culture supernatant ( Fig 4D ) . In contrast , neither strain that produced WTA ( the atl mutant nor Newman ΔtagO complemented with a plasmid carrying the cloned tagO gene [ΔtagO C’] ) were killed after incubation with the Newman supernatant . The susceptibility of the tagO mutant to the bactericidal activity of culture supernatants was not limited to strain Newman , since Newman ΔtagO was also killed by supernatants harvested from S . aureus strains USA300 LAC and ST80 , but not from Sanger 252 ( Fig 4E ) . None of the culture supernatants were bactericidal against the parental strain Newman . Koprivnjak et al . reported that S . aureus SA113 ΔtagO was more sensitive than the WT strain to Triton X-100 at concentrations >0 . 1% [29] . We observed that Newman ΔtagO also was hypersusceptible to Triton X-100 mediated bacterial lysis at concentrations as low as 0 . 05% ( Fig 5A ) . As expected , a Newman atl mutant and an atl tagO double mutant were resistant to Triton X-100 induced autolysis ( Fig 5A ) . Because factors that stimulate autolytic activity include detergents , proteolytic enzymes , and cationic peptides [30] , we hypothesized that the Newman ΔtagO might be more sensitive than the WT strain to factors that promote staphylococcal autolysis . This hypothesis is consistent with the observation made by Atilano et al [31] that peptidoglycan prepared from a ΔtagO mutant shows reduced cross-linking compared to the WT strain . Autolysin extracts from strain Newman and Newman ΔtagO showed similar lytic activity toward peptidoglycan prepared from the strain Lafferty ( S10 Fig ) . However , peptidoglycan prepared from the Newman ΔtagO mutant was more susceptible to Newman autolysin extracts than a peptidoglycan preparation from the parental strain Newman ( Fig 5B ) .
In addition to its niche within the nares , in the throat , and on the skin of humans , S . aureus may also colonize the GI tract of ~20% of otherwise healthy individuals [3] . While no disease is clearly associated with GI colonization by staphylococci , it serves as a source of recolonization after eradication of nasal colonization . In a hospital setting or in the community , fecal contamination associated with diarrhea or incontinence could contaminate the environment and facilitate transmission among individuals . As such , the GI tract is an under-appreciated reservoir for methicillin-resistant and-sensitive strains of S . aureus . Among patients in an intensive care unit or transplant unit , those positive for both rectal and nasal carriage of S . aureus were more likely to develop a staphylococcal infection than those with nasal carriage alone [8] . Other reports have similarly suggested that GI carriage is an important risk factor for S . aureus infections [3 , 4 , 9 , 10] . Importantly , staphylococcal skin and soft tissue infections are linked to rectal , but not nasal , colonization by MRSA strains in children [10] . Molecular typing experiments indicated that rectal , nares , and infecting isolates ( including blood isolates ) were clonally identical in 82% of the patients with S . aureus infections [8] . A comprehensive understanding of the factors involved in asymptomatic carriage by S . aureus in humans is critical to our ability to control the incidence of infection . Kernbauer et al . [32] recently reported that mice challenged intravenously with S . aureus develop staphylococcal colonization of the GI tract , and that fecal shedding resulted in S . aureus transmission to cohoused naïve mice . They noted that GI colonization resulted in no obvious signs of infectious abscesses or inflammation . In our studies , healthy mice inoculated intranasally developed persistent S . aureus GI colonization only when they were administered Sm in their drinking water for at least one week . This suggests that transient suppression of the indigenous intestinal flora may reduce bacterial interference and created an environment amenable to stable S . aureus colonization . The concentration of S . aureus in the cecum , colon , and stool was not high in conventional mice , achieving levels ~105 CFU/g , very similar to those reported by Kernbauer et al . [32] . When germ-free mice that lacked a competing indigenous flora were inoculated with S . aureus , the concentrations in the intestinal contents and stool rose to ~108 CFU/g . To investigate bacterial factors that impact GI colonization , we assessed colonization of mice inoculated with a single bacterial strain , and we also performed competition experiments between WT and mutant S . aureus strains . A Newman type 5 capsule-negative mutant was able to persist in the gut in higher numbers than the WT encapsulated parental strain in competition experiments , consistent with previous studies showing that the capsule can impede factors critical for colonization [24 , 25] . The Newman sortase and clfA mutants each showed impaired fitness in the mouse gut . In contrast , strains with mutations in clfB , spa , sdrCDE , or the ica locus showed no fitness defect in the GI colonization model . This suggests that ClfA and perhaps other wall-anchored protein adhesins ( possibly masked by the capsule ) play a role in promoting S . aureus colonization of the GI tract . Competition experiments are commonly used to assess microbial fitness for colonization of the GI tract [19–21] , probably because they are more sensitive and may reveal more subtle competitive advantages demonstrated by a particular wild type or mutant strain . More striking , however , was our observation that tagO mutants of strains SA113 and Newman that fail to produce WTA were unable to colonize the nose or the GI tract of mice . This colonization defect was evident in groups of conventional mice inoculated with pure cultures of the mutant strain; competition assays were not necessary to demonstrate this impaired colonization phenotype in vivo . S . aureus mutants that lacked the D-ala or GlcNAc substituents of WTA , on the other hand , showed no defects in GI colonization . These findings suggest that the glycopolymer itself , and not its modifying groups , are critical for interactions with the host . In vitro assays demonstrated that both the tagO and clfA mutants of strain Newman were less adherent to T84 cells in vitro than the parental strain Newman . These data suggest a critical role for these surface antigens in GI colonization . Baur et al . [33] recently reported that S . aureus WTA interacts with a scavenger receptor ( SREC-I; scavenger receptor expressed by endothelial cell-I ) detected on nasal epithelial cells to promote bacterial adherence . Whether such a receptor is found on intestinal epithelial cells remains to be determined . In addition to its adherence defects , the tagO mutant showed fitness defects in transit through the mouse stomach and intestines . In an in vivo competition study , the tagO mutant was outcompeted by the WT strain within 1 h after inoculation . This finding is consistent with our in vitro observations that the tagO mutant was more susceptible than the parental strain to bile acids , proteases , and alpha defensin 5 , which would be encountered in the GI tract . Like humans , mice have Paneth cells in the small intestinal crypts , and these cells secrete alpha defensins that show bactericidal activity against a number of microbes , including S . aureus [34] . S . aureus autolysins are activated by detergents ( Triton X-100 and bile acids such as deoxycholate ) , defensins , and proteases [30] . Because these factors are present in the GI tract , we postulated that the Newman tagO mutant did not survive well because it is more susceptible to stress-induced autolysis than the wild-type strain . To address this , we created an atl deficient mutant of strain Newman . Atl is the major S . aureus autolytic enzyme , and it binds preferentially at the septum site ( where WTA is less abundant ) to facilitate cell division . However , Atl loses its selective localization to division sites in a tagO mutant . In the absence of WTA , Atl binding was reported to be evenly distributed on the cell surface , which explains the increased fragility and susceptibility of the ΔtagO mutant to autolysis [31 , 35] . As expected , the atl and the atl tagO double mutant were both resistant to Triton X-100-induced lysis ( activation of endogenous autolysins ) , whereas both the tagO and atl tagO double mutant were highly susceptible to exogenous autolysins ( in Newman culture supernatants ) . Our final in vivo colonization studies showed that the WT strain Newman and its isogenic atl mutant were both proficient for GI colonization . However , neither the tagO nor the atl tagO double mutant colonized the mouse gut . These results indicate that the enhanced sensitivity of the tagO mutant to antimicrobial factors in the GI tract , as well as its poor adherence characteristics , likely contribute to its impaired fitness in vivo . However , the enhanced vulnerability of the tagO mutant to clearance from the GI tract is apparently not due to activation of its endogenous autolysins by host-induced bacterial stress . Previous studies have characterized the in vitro susceptibility of S . aureus ΔtagO to a variety of host immune defenses . Compared to the parental strains , a ΔtagO mutant showed increased susceptibility to antimicrobial fatty acids on human skin [36] , but it was not more susceptible to killing by neutrophils , lysozyme [37] , lactoferrin , or cationic antimicrobial peptides such as hNP1-3 , LL37 , or magainin II amide . Mutants lacking WTA were more resistant to mammalian group IIA phospholipase A2 and human β-defensin 3 ( HBD-3 ) than the WT strain [29] . Additional studies will be necessary to further delineate the essential role of WTA in S . aureus colonization of the mammalian GI tract . The murine colonization model may be useful in further characterization of factors that impact S . aureus carriage in the GI tract , with the acknowledged limitation that the human intestinal tract and its flora have characteristics distinct from those of the mouse [38] . Nonetheless , our results show that S . aureus WTA contributes to staphylococcal fitness within the GI tract , providing resistance to host bactericidal factors and promoting bacterial adherence to epithelial cells . A better understanding of mechanisms that lead to asymptomatic colonization by S . aureus may lead to preventive therapies that impact transmission among humans . Recent efforts to develop antimicrobials that target WTA [39–41] may lead to effective agents that diminish both nasal and GI colonization by MRSA and may impact invasive disease [23] .
The S . aureus strains tested in the murine gastrointestinal colonization model are listed in Table 3; each was resistant to Sm . Mutations in the S . aureus ica locus , spa , and atl were moved into Sm-resistant Newman by transduction with phage 80α or phage 85 from the original antibiotic marked mutant strains [42–44] . The tagO::tet mutation was moved from S . aureus RN4220 into Newman Δatl as described previously [22] . Mutants were confirmed by PCR or Southern blot analysis and were phenotypically identical to the parental strains in terms of the growth rate , hemolysis on sheep blood agar plates , and the metabolic profile on API Staph test strips ( Biomerieux , Inc . , Durham , NC ) . S . aureus strains were cultivated in TSB to the logarithmic growth phase , unless otherwise noted . Sm ( 0 . 5 mg/ml; Sigma Chemical Co . , St . Louis , Mo . ) , erythromycin ( Em; 5 μg/ml; Sigma ) , tetracycline ( Tc; 2 . 5 μg/ml; Sigma ) , or spectinomycin ( Spc; 100 μg/ml; MP Biomedicals , Solon , OH ) was added to culture medium for selection where appropriate . Animal experiments were approved by the Longwood Medical Area's Institutional Animal Care and Use Committee under protocol 86–02131 . All studies were performed in strict accordance with the National Institutes of Health standards as set forth in "Guide for the Care and Use of Laboratory Animals" ( DHSS Publication No . ( NIH ) 85–23 ) . Conventional female ICR mice aged 4–6 wks old were purchased from Harlan Sprague Dawley , Inc . ( Indianapolis , IN ) or Charles River Laboratories ( Wilmington , MA ) . The mice were given drinking water containing Sm ( 0 . 5 g/L ) one day prior to inoculation and for the course of experiment ( unless otherwise noted ) , and the drinking water and cages were changed twice a week . Conventional mice were inoculated without prior anesthesia by the intranasal route with 10 μl of an S . aureus suspension as described [12] . Gastrointestinal colonization by S . aureus was evaluated by quantitative cultures of representative mouse stool pellets that were collected prior to inoculation and weekly thereafter . The samples were weighed , suspended in 2–5 ml of TSB , diluted , and plated quantitatively on mannitol salt agar ( MSA ) containing 0 . 5 mg/ml Sm . The plates were incubated aerobically for 48 h at 37°C , the colonies were enumerated , and the data were expressed as CFU/g stool . Significant ( P < 0 . 05 ) differences between quantitative culture results were determined by the two-tailed Student t test , and the Welch correction was applied to pairs with unequal variances . P-values for experiments having three or more groups were determined by the Kruskal-Wallis test with Dunn’s multiple comparison analysis ( Prism; GraphPad Software , La Jolla , CA ) . Germfree Swiss-Webster female mice ( 3–4 weeks old ) were purchased from Taconic Farms ( Hudson , NY ) and maintained in a negative-pressure BL2 isolator by personnel at the gnotobiotic core facility of the Harvard Digestive Diseases Center . S . aureus suspensions were drawn into a sterile syringe with a neonatal feeding tube , and ~ 200 μl ( ~109 CFU ) was orally fed to the mice . Gnotobiotic mice were given sterile ( no Sm ) drinking water , and their stools were cultured on nonselective medium ( tryptic soy agar [TSA] plates ) . For competition experiments , the mutant and parental strains were mixed in equal numbers ( total 109 CFU/mouse ) prior to inoculation of conventional mice . The input CFU ratio was calculated by dividing the wild type inoculum CFU by the mutant inoculum CFU ( WT/mutant ) . At various time points , stool samples were collected and plated quantitatively on MSA + Sm plates . Colonies of the mutant strains were distinguished from those of the parental strain by transferring ( via sterile toothpicks ) ~100 colonies to TSA and TSA + antibiotic ( Tc , Em , or Spc ) plates . The output CFU ratio was determined by dividing the WT CFU by the mutant CFU for each fecal sample . A competitive index ( CI ) was calculated as described [45] by dividing the output ratio by the input ratio , and the CI was expressed as log10 . A CI = 0 indicates similar numbers of CFU of the two competing strains recovered in vivo , suggesting comparable fitness of the strains in vivo . A CI >1 indicates the fitness advantage of the parental strain over the mutant , whereas a CI <1 indicates that the mutant is more fit than the parental strain . A Wilcoxon signed-rank test was used to determine whether the CI values were significantly different from the hypothetical value of zero . For in vitro competition experiments , the parental and mutant strains were inoculated together into 5 ml of TSB , grown with aeration for either 2 or 24 h at 37°C , and subsequently plated quantitatively . Colonies of the ΔtagO mutant were distinguished from those of the parental strain by their antibiotic resistance as described above , and the CI was calculated . To determine whether the tagO mutant survived as well as the parental strain in its transit through the mouse stomach , rodent chow was withheld for 3 h prior to inoculation of mice with either S . aureus Newman , Newman ΔtagO , or Newman ΔtagO Δatl . The mice were euthanized by CO2 asphyxiation after 1 h , and quantitative cultures were performed on homogenates made from different segments of the GI tract . However , variations in the recovery of S . aureus among individual mice were substantial , and we were unable to draw meaningful conclusions from these data . To circumvent this problem , competition experiments were performed as described above in which equal numbers ( ~5 x 108 CFU ) of the two strains were mixed before inoculation . One hour after bacterial challenge , the mice were euthanized , the GI tract was excised , and the stomach , small intestine ( 4 segments ) , cecum , and colon were separated . Each segment was weighed , homogenized in 1 ml TSB , and cultured quantitatively on MSA + Sm plates . Colonies of the ΔtagO mutant were distinguished from those of the parental strain by replica plating on TSA + Tc or TSA + Em , and the CI was calculated . Polarized monolayers of human intestinal T84 cells were cultured on 0 . 33 cm2 polyester Transwell inserts ( Corning , Acton , MA ) as previously described [46] . Log-phase TSB cultures of S . aureus Newman or its isogenic mutants were washed in Hanks Balanced Salt Solution ( HBSS ) and incubated at a multiplicity of infection of 100 on the apical surfaces of polarized T84 monolayers for 1 h at 37°C . Inserts were washed five times with HBSS and processed for confocal microscopy as previously described [47] . Briefly , the T84 cell monolayers were fixed in 4% paraformaldehyde for 20 min . The cells were washed 3x in HBSS , permeabilized with 0 . 2% Saponin/HBSS solution , washed again , and blocked with 10% normal goat serum ( NGS ) /HBSS solution . Immunostaining was performed with mouse anti-ZO-1 ( 1:200; Invitrogen ) and S . aureus rabbit antiserum ( raised to whole killed bacteria and diluted 1:400 ) in 10% NGS/HBSS . The cells were incubated with AlexaFluor 568 labeled goat anti-mouse conjugate and AlexaFluor 488 labeled goat anti-rabbit conjugate ( both 1:400 in 10% NGS/HBSS solution; Invitrogen ) for 1 h at room temperature . Samples were imaged by confocal microscopy on a Nikon TE2000 inverted microscope ( Nikon Instruments , Melville , NY ) coupled to a Perkin-Elmer spinning disk confocal unit ( Boston , MA ) , using a Nikon PlanFluor 40× ( 1 . 3 NA ) and Nikon PlanApo 60x ( 1 . 4 NA ) oil immersion objective lens and an Orca AG scientific-grade cooled CCD camera ( Hamamatsu Photonics K . K . , Japan ) . Confocal images , collected en face to the apical membrane , were 3D stacks collapsed into a single projection . Adherence was quantified by counting the total bacteria in 20 random fields of view ( FOV ) among triplicate samples with a 63X objective and a 10X ocular lens . Newman WT and ΔtagO were cultivated in TSB to log phase at 37°C , washed in 10 mM phosphate buffer ( pH 7 . 4 ) , and suspended to 1–3 x 108 CFU/ml in either buffer , bile salts ( sodium deoxycholate [DOC]; Sigma ) , or human α-defensin 5 ( HD5 ) ( Peptide Institute , Inc . , Osaka , Japan ) . Susceptibility to low pH was assessed by suspending S . aureus strains in glycine buffer ( pH 3 . 0 , 3 . 3 , or 3 . 6 ) or in phosphate buffer at neutral pH . Bacterial suspensions were incubated at 37°C for 1 to 2 h , and viable counts were determined . Susceptibility to proteases was assessed by preparing serial dilutions in optimal buffers ( proteinase K in 10 mM phosphate buffer , pH 7 . 4; pepsin in glycine buffer , pH 3 . 6; trypsin in 10 mM phosphate , pH 8 . 0 buffer ) . The protease solutions were then inoculated with logarithmic-phase S . aureus to a final concentration of 1 x 106 CFU/ml , and viability counts were performed after incubation for 24 h at 37°C . The bactericidal activity of S . aureus culture supernatants was assessed by culturing the S . aureus strains in TSB at 37°C for ~21 h . Filter-sterilized culture supernatants were inoculated to a final 105 CFU/ml with strain Newman or ΔtagO , ΔtagO ( pRBtagO ) [22] , Δatl , or Δatl ΔtagO mutants . CFU/ml determinations were performed after incubation for 1 to 5 h at 37°C . For some assays , the Newman supernatant was boiled for 5 min , treated with a protease inhibitor cocktail ( Sigma P-8465 ) , or supplemented with 0 to 4 mg peptidoglycan purified from S . aureus Lafferty [48] as described [22] before inoculation with strain Newman or its tagO mutant . Triton X-100-induced autolysin assays were performed as described by Shaw et al . [49] . Briefly , Newman WT and ΔtagO strains were grown to logarithmic phase , and the bacterial cells were washed twice and suspended to OD650 nm of 1 . 0 in 50 mM Tris-HCl ( pH 7 . 6 ) , 2 mM CaCl2 , 0 . 05% Triton X-100 . The suspensions were incubated in a 37°C shaking water bath , and the OD580 nm was monitored for 3 h . Zymography was used to detect the lytic activities in culture supernatants from Newman and Newman ΔtagO , as described by Groicher et al . [50] . Overnight culture supernatants , concentrated with Centricon-5 centrifugal filter units ( Millipore ) to a protein concentration of 2 . 65 mg/ml , were electrophoresed in a 10% polyacrylamide SDS gel supplemented with 1 mg/ml S . aureus cell walls . After washing the gel in 1% Triton X-100 and 25 mM Tris-HCl buffer overnight to allow cell wall hydrolysis , the gel was stained with 1% methylene blue and destained in water . Lytic zones in the gel indicated proteins with peptidoglycan hydrolase activity . S . aureus autolysin extracts were prepared by 3 M LiCl treatment as described [51] . Peptidoglycan samples , prepared from strains Newman or Newman ΔtagO as described previously [22] , were suspended in 0 . 01 M KHPO4 buffer ( pH 7 . 1 ) to OD580 nm of 0 . 5 to 0 . 6 . The suspensions were mixed with an S . aureus Newman autolysin extract ( 25 μg/ml protein ) , incubated in a 30°C water bath , and OD580 nm readings were taken at 1 h intervals . Lytic activity was expressed as a percentage of the OD580 nm at time zero . | Staphylococcus aureus persistently colonizes ~20% of the human population , and 40–60% of humans are intermittently colonized by this bacterium . The most common reservoir for S . aureus is the anterior nares , and the incidence of staphylococcal disease in higher in individuals who are colonized . Rectal colonization by S . aureus isolates , reflecting gastrointestinal ( GI ) carriage , has recently been recognized as an important reservoir from which person to person transmission occurs . We developed a murine model of S . aureus GI colonization to investigate bacterial factors that promote staphylococcal colonization of the gut . We identified several surface-associated S . aureus antigens that modulate colonization of the GI tract and identified a surface glycopolymer ( cell wall teichoic acid ) as critical for the early steps in colonization . The failure of the teichoic acid mutant to colonize the GI tract can be attributed to its defects in bacterial adherence and to its enhanced susceptibility to mammalian host defenses unique to the gastrointestinal tract . Efforts to develop antimicrobials that target WTA may lead to an overall reduction in asymptomatic colonization by antibiotic-resistant S . aureus and may impact the incidence of invasive disease . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | Staphylococcus aureus Colonization of the Mouse Gastrointestinal Tract Is Modulated by Wall Teichoic Acid, Capsule, and Surface Proteins |
Malaria control efforts have a significant impact on the epidemiology and parasite population dynamics . In countries aiming for malaria elimination , malaria transmission may be restricted to limited transmission hot spots , where parasite populations may be isolated from each other and experience different selection forces . Here we aim to examine the Plasmodium vivax population divergence in geographically isolated transmission zones in Thailand . We employed the P . vivax merozoite surface protein 3β ( PvMSP3β ) as a molecular marker for characterizing P . vivax populations based on the extensive diversity of this gene in Southeast Asian parasite populations . To examine two parasite populations with different transmission levels in Thailand , we obtained 45 P . vivax isolates from Tak Province , northwestern Thailand , where the annual parasite incidence ( API ) was more than 2% , and 28 isolates from Yala and Narathiwat Provinces , southern Thailand , where the API was less than 0 . 02% . We sequenced the PvMSP3β gene and examined its genetic diversity and molecular evolution between the parasite populations . Of 58 isolates containing single PvMSP3β alleles , 31 sequence types were identified . The overall haplotype diversity was 0 . 77±0 . 06 and nucleotide diversity 0 . 0877±0 . 0054 . The northwestern vivax malaria population exhibited extensive haplotype diversity ( HD ) of PvMSP3β ( HD = 1 . 0 ) . In contrast , the southern parasite population displayed a single PvMSP3β allele ( HD = 0 ) , suggesting a clonal population expansion . This result revealed that the extent of allelic diversity in P . vivax populations in Thailand varies among endemic areas . Malaria parasite populations in a given region may vary significantly in genetic diversity , which may be the result of control and influenced by the magnitude of malaria transmission intensity . This is an issue that should be taken into account for the implementation of P . vivax control measures such as drug policy and vaccine development .
Of the four species of human malaria parasites , Plasmodium vivax is the second most prevalent and the most geographically widespread parasite . Each year , P . vivax infects an estimated 130–391 million people , of which a large majority was in Central and Southeast Asia [1]–[3] . Recent data demonstrate that the traditionally called “benign tertian malaria” is certainly a misnomer since P . vivax infection brings enormous morbidity and mortality in affected populations [4] , [5] . In addition , the development of resistance to chloroquine and possibly primaquine in P . vivax has raised a great concern for the control of the disease [6] . Outside sub-Saharan Africa , the proportions of malaria cases caused by P . vivax are arising , a clear indication of the resilience of this parasite to control measures [7] . Especially in areas of P . falciparum and P . vivax co-existence , their intricate interspecies interactions suggest that control measures against one species may inevitably lead to increased prevalence of the other [8] , [9] . This has resulted in renewed interests in developing P . vivax vaccines . Vaccine development against such complicated eukaryotes like malaria parasites is not straightforward . Multivalent and multistage vaccines are proposed because the malaria parasite's life cycle involves multiple stages with each stage expressing different antigens . Merozoites as the invasive stage of the erythrocytic cycle are exposed to host immunity , and therefore are important vaccine targets [10] . Some merozoite antigens such as merozoite surface protein 1 ( MSP1 ) and apical membrane antigen 1 ( AMA1 ) have been extensively studied . Meanwhile , these antigens are subject to the selection forces imposed by the host immunity and exhibit extensive diversity [11] . As such , antigenic variation is an important consideration when identifying and prioritizing antigens for vaccine development . A number of MSPs have been identified as the coat constituents of the P . vivax merozoites . These include PvMSP1 [12] , the PvMSP3 family members [13] , [14] , PvMSP4 , PvMSP5 [15] , PvMSP7 [16] , PvMSP8 [17] , PvMSP9 [18] , and PvMSP10 [19] . The first member of the PvMSP3 family identified in 1999 was named PvMSP3 ( α ) due to its similarity to PfMSP3 [14] . Two paralogs were identified later and named PvMSP3β and 3γ , respectively [13] . In spite of limited sequence identify , the PvMSP3 protein family members share characteristics such as the central alanine-rich domain , which is predicted to form a coiled-coil structure that may involve in protein-protein interactions [13] , [14] . Genome sequencing of the P . vivax Salvador I strain revealed 12 PvMSP3 paralogs clustered in a ∼60 kb locus on chromosome 10 , which led the authors to speculate that this gene family might have undergone species-specific expansion [20] . Although a number of studies suggested relatedness of msp3 genes in P . vivax and P . falciparum , a closer comparison between the syntenic loci on chromosome 10 and domain organizations of pvmsp3 and pfmsp3 did not suggest that these are homologs [21] . Though these putative msp3 family proteins share an N-terminal NLRNG peptide motif , comparison of the msp3 loci in several Plasmodium species of Asian primates and the African monkey parasite Plasmodium gonderi further established that the expanded pvmsp3 family had a much earlier origin . Furthermore , analysis of several additional P . vivax genomes found expansion and contraction of the pvmsp3 loci , with pvmsp3 family members ranging from 9 to 14 in each parasite genome [21] . Based on this finding , the authors hypothesized that the pvmsp3 gene family might be under multi-allelic diversifying selection to increase antigenic diversity [21] . Recent studies detected protein expression for 10 members of the PvMSP3 family with eight proteins being visualized surrounding merozoites and one being localized at the apical end of merozoites [22] . Since PvMSP3 family members do not appear to have transmembrane domains or the GPI-anchor site , their association with the merozoite surface is predicted to be through protein-protein interactions via the central coiled-coil domain [13] , [14] . Yet , this region in both PvMSP3α and PvMSP3β harbors large deletions in worldwide collections of P . vivax strains [23]–[26] . Sequencing analysis revealed that both genes are highly polymorphic and the highest nucleotide diversity is clustered towards the N-terminal region [26]–[34] , making them very useful genotyping markers for differentiating P . vivax field isolates . PCR-RFLP methods developed for genotyping pvmsp3α and 3β have been used widely for studying P . vivax genetic diversity in various endemic settings [29] , [30] , [33] , [35]–[39] , albeit caution needs to be exercised when interpreting the RFLP patterns since frequent insertion-deletion mutations and recurrent recombination events may obscure the distinctions between RFLP haplotypes [21] . Despite their polymorphic nature , the potential of PvMSP3α and 3β as vaccine candidates has been evaluated . Both proteins are found immunogenic and naturally acquired antibodies are associated with exposure to P . vivax parasites in vivax-endemic regions [40]–[43] . Antibodies against PvMSP3α in Papua New Guinea children were associated with protection from clinical P . vivax malaria [44] . Therefore , further evaluation of the polymorphism of these proteins in endemic countries is needed . In this study , we have obtained full-length PvMSP3β gene sequences from 58 P . vivax clinical samples collected mostly from two endemic regions in Thailand . This has allowed us to further evaluate the genetic diversity and dissect the domain structure of PvMSP3β . We demonstrate drastic geographical differentiation of P . vivax populations , with its genetic structure being correlated with the endemicity of P . vivax malaria . The monomorphic pvmsp3β gene in a P . vivax population from the hypoendemic southern Thailand suggests a clonal expansion of the parasite strain .
Plasmodium vivax isolates were collected from two study sites in northwestern and southern Thailand . Forty-five patients were recruited at the malaria clinic located in Mae Sot district , Tak province , northwestern Thailand [31] , while 28 isolates were from Yala and Narathiwat Provinces , southern Thailand [45] . Collection of finger-prick filter paper samples from malaria patients was approved by the institutional review board of Chulalongkorn University under the auspice of the Thai Ministry of Health . After obtaining written informed consent , blood samples were collected on filter papers and dried . Parasite DNA was extracted from the filter papers using a QIAamp DNA Mini kit ( Qiagen , Germany ) and DNA was eluted in 100 µl of water . The complete pvmsp3β gene of 2 . 0–2 . 5 kb was amplified using primers Pv3BF ( 5′ AAATGGTATTCTTCGCAACAC 3′ ) and Pv3BR ( 5′ TTCGTCACCAATTTGTTTAGC 3′ ) . All primers were designed based on the PvMSP3β gene ( AF099662 ) of the Belem strain [13] . PCR was done in 25 µl consisting parasite DNA , 1× reaction buffer , 200 µM of dNTPs , 0 . 05 µg each of the outer primers , and 0 . 5 µl of KlenTaq ( BD Biosciences ) using a program of 1 . 5 min of initial denaturing at 94°C followed by 35 cycles of 94°C for 30 sec , 55°C for 30 sec and 68°C for 3 min . The PCR products were sequenced directly using the BigDye terminator kit ( Applied Biosystems ) . For accuracy , two different amplification products of each isolate were sequenced . Complete sequences were assembled using the DNASTAR program ( Lasergene ) . Both DNA and predicted protein sequences were aligned using Clustal X version 1 . 83 . The internal region of the sequences was further edited manually due to the presence of insertion/deletion in several isolates . Sequences available in the GenBank that were included in this analysis are from isolates/strains from Bangladesh ( AY454084 ) , Brazil ( PVBG05499 , AF099662 , AY454080 , AY454081 , AY454082 , AY454085 , AY454086 , AY454087 , AY454088 , AY454089 ) , El Salvador ( XM001613146 ) , Ecuador ( AY454091 ) , India ( PVIIG04181 , AY454092 ) , Mauritania ( PVMG01384 ) , Papua New Guinea ( AY454083 ) , North Korea ( PVNG01493 ) , Sri Lanka ( AY454096 and AY454097 ) , Thailand ( AY454098 ) and Vietnam ( AY454094 ) . New pvmsp3β sequences in this study were submitted to GenBank ( KM041050 to KM041113 ) . Molecular evolutionary analysis was performed using MEGA version 6 . 0 [46] and DnaSP version 5 . 10 . 1 [47] . The genetic diversity of pvmsp3β gene was estimated using the parameter π , which calculates the average number of substitutions per site between two sequences . The rates of synonymous ( dS ) and nonsynonymous ( dN ) substitutions were estimated using the method of Nei and Gojobori with the Jukes and Cantor correction . The standard error was calculated using the bootstrap method with 1000 pseudoreplications . To test whether pvmsp3β is under positive selection , Z-test was performed by comparing the nonsynonymous and synonymous substitutions using MEGA . Different regions of the gene were further evaluated by a sliding window analysis and their significance was determined using the Tajima's D test [48] and the D* and F* statistics of Fu and Li [49] . The minimum number of recombination events ( Rm ) was estimated using DnaSP . The degree of linkage disequilibrium ( LD ) between distance of parsimony informative variant nucleotide sites was estimated by the r2 values [47] and significance levels ( more than 95% ) were determined by the two-tailed Fisher's exact test . Evidence of intragenic recombination was also determined by using the RDP4 package [50] and the Genetic Algorithm Recombination Detection ( GARD ) method implemented in the HyPhy package [51] . The evolutionary relationships of the P . vivax isolates were inferred from phylogenetic analysis of the complete or available pvmsp3β gene using the Maximum Likelihood method based on the General Time Reversible model . The Neighbor-Joining method was applied to generate initial trees for the heuristic search using the Maximum Composite Likelihood approach with a discrete Gamma distribution to model evolutionary rate differences among sites . The reliability of the tree was assessed by the bootstrap method with 500 pseudoreplications . The analysis was performed using the MEGA 6 . 0 [52] . The population genetic structure was evaluated by the analysis of molecular variance ( AMOVA ) approach implemented in the Arlequin 3 . 5 software [53] using the Weir and Cockerham's method while taking into account the number of mutations between haplotypes [54] . All domains were included in the analysis excluding sites with alignment gaps . Populations were classified as South America ( n = 13 ) , Asia excluding Thailand ( n = 13 ) and Thailand ( n = 59 ) . The fixation index FST identical to the weighted average F-statistic over loci , θw [54] and the significance levels of the fixation indices were estimated by non-parametric permutation as implemented in Arlequin 3 . 5 [53] .
We have amplified and sequenced a total of 73 P . vivax clinical samples from two study sites in Thailand . The PCR products were sequenced directly . Mixed infections were evidenced by the presence of multiple PCR bands and superimposed signals on electropherogram from DNA sequencing . All 28 samples from southern Thailand were single clone infections . However , 15 of the 45 samples ( 33% ) from Mae Sot in northwestern Thailand had multiple clone infections and were excluded from sequencing analysis . Four isolates from China and one isolate from India were sequenced and included in this study . The size of the pvmsp3β sequences ranges from 2 , 109 to 2 , 478 bp , encoding a predicted protein of 703–826 amino acids ( aa ) . Our data revealed drastic difference in haplotype diversity of pvmsp3β gene between the two study sites . The 30 pvmsp3β sequences from northwestern Thailand are all unique ( haplotype diversity = 1 ) . Surprisingly , 28 sequences from southern Thai P . vivax population were all identical , suggesting a potential result of clonal expansion . Previous studies showed that pvmsp3β gene has large variations in gene size among world parasite populations [26] , [33] , [34] , which are due to insertions or deletions occurring in the central domain of the gene . Alignment of the predicted aa sequences from Thailand and those determined earlier [26] showed that PvMSP3β is highly polymorphic , containing numerous substitutions , insertions and deletions of variable sizes ( Figure S1 ) . The distribution of substitutions is not random , and the C-terminal block is more conserved . Most notable is the two blocks of large insertions in the central Ala-rich domain . Previous report by Rayner et al . ( 2004 ) divided the gene into four regions: the N-terminal part , insert A , insert B and the C-terminal part . Based on more detailed analysis of available sequences , we divided the sequences into seven blocks . Figure 1 is a schematic representation of the PvMSP3β protein using the Salvador I ( Sal-1 ) strain as the reference . Block 1 ( 1–157 aa ) is the conserved N-terminus without obvious insertions or deletions . Block 2 ( 158–336 aa ) is less conserved with multiple interspersed short insertions or deletions , which is followed by insertion A ( 337–456 aa ) referred here as block 3 . Block 4 is conserved , spanning a short region of 24 nucleotides ( 457–464 aa ) . Insertion B or block 5 ( 465–664 aa ) is highly polymorphic , and the Thai isolate-105 has a novel sequence in this block . Block 6 ( 665–777 aa ) was originally included in a portion of insertion B , which is a dimorphic region represented by the Sal-1 type and the Bangl type . Block 7 ( 778–967 aa ) is the conserved C-terminus . BR67B is the only isolate that has an unusually long deletion in this region [26] . Worldwide distribution of pvmsp3β alleles based on blocks 3 , 5 and 6 is shown in Figure 2 . Most of the sequences ( 89% ) did not harbor insertion A . Samples with insertion B of block 5 was less prevalent . Most of the sequences in the dimorphic C-terminal block 6 had the Sal-1 type , whereas the Bangl type was absent in the American samples . The overall nucleotide diversity of pvmsp3β was similar to the estimates for other P . vivax merozoite antigens such as pvmsp1 , pvama1 and pvmsp3α [23] , [55] . The conserved C-terminal block was the least variable , whereas the N-terminal regions had more than two times higher nucleotide diversity ( Table 1 ) . The PvMSP proteins harbor extensive sequence diversity in the N-terminal region , suggesting of selection . To identify the signature of selection in pvmsp3β in the parasite populations , we performed neutrality tests on individual blocks of pvmsp3β where nucleotide sequences could be unambiguously aligned . Block-wise analysis showed that dN is significantly greater than dS in blocks 1 and 2 , suggesting positive selection in these blocks . In contrast , dS significantly out-numbers dN in block 7 , suggesting purifying selection in the C-terminal-coding region . Although dS was greater than dN in most of blocks 4 and 6 , they were not significantly different . Tajima's D statistics , however , did not detect signatures of selection in block-wise analysis ( Table 1 ) . On the other hand , significant departure from neutrality was observed in Fu and Li's D* and F* tests for conserved block 4 and isolates harboring Bangl type of the dimorphic block 6 . Negative departure from neutrality of these tests could be resulted from purifying selection or population expansion after a bottleneck effect . The minimum number of recombination events ( Rm ) estimated using DnaSP showed that for the Thai isolates 72 recombination events were detected in blocks 1 , 2 and 4 . The high number of predicted Rm value might be accountable for the observed slight decay in significant loci over molecular distance within blocks 1 , 2 and 4 as revealed from LD analysis ( Figure 3A ) . Likewise , decay in significant loci over molecular distance was also observed in block 7 although relatively lower Rm value of 17 was identified in this block ( Figure 3B ) . Analysis of recombination break points inferred 40 recombination events by one or more methods implemented in the RDP4 program package ( RDP , GENECONV , BOOTSCAN , MAXCHI , CHIMERA , SISCAN , 3SEQ , PHYLPRO and LARD ) using the default parameters ( Figure S2 ) . Additional analysis using the GARD program implemented in the HyPhy package revealed that at least 3 break points ( nucleotides 406 , 1500 and 2608 , positions based on the alignment shown in Figure S1 ) gave significant topological incongruence ( p<0 . 01 ) between AICc ( Akaike Information Criterion derived from a maximum likelihood model fit to each segment ) score of the best fitting GARD model . We used the coding region of PvMSP3β gene to infer the phylogenetic relationship among the isolates . Phylogenetic analysis revealed no geographic clustering among isolates from diverse origins and that Thai alleles were placed throughout the tree ( Figure 4 ) . This phenomenon has also been observed for the PvMSP3α gene among the Thai isolates [23] . However , phylogenetic tree has defined two distinct clusters of isolates with 100% bootstrap values , corresponding to the dimorphic Sal-1 type and the Bangl type ( Figure 4 ) . It is noteworthy that recombination events inferred from the RDP4 algorithm could be commonly observed between isolates within and between diverse geographic origins ( Figure 4 ) akin to those occurred at the pvmsp3α locus [23] . Despite the small number of samples used in this study , we observed significant genetic structure among isolates from Thailand ( n = 59 ) , Asia excluding Thailand ( n = 13 ) and the Americas ( n = 13 ) . The highest FST value was noted between P . vivax populations from Thailand and the Americas ( Table 2 ) .
The contemporary malaria parasite populations have been shaped by many selective forces such as human malaria control activities . As the malaria control and elimination course progresses , it is expected that the range of malaria endemicity will shrink , which should influence the effective population size . In many countries of the Greater Mekong Subregion ( GMS ) of Southeast Asia , malaria is restricted to the international border regions and some even exists in isolated pockets , further restricting gene flow among parasite populations . In this study , we investigated the genetic diversity and evolution of the pvmsp3β gene in parasites from two endemic regions of Thailand . We have further confirmed that sequencing of pvmsp3β offers significantly increased power for determining parasite genetic diversity compared with the simple PCR-RFLP method [33] . This is analogous to using pvmsp3α as a molecular marker , where PCR-RFLP is not as informative of population genetic diversity [32] . In the northwestern Thai provinces bordering Myanmar , the progress of malaria control has been slow , and cross-border human population movements and malaria introduction from Myanmar have been partially blamed for the continued malaria transmission [56] . Our analysis of the PvMSP3β polymorphism revealed that P . vivax parasites originated from the northwestern border region were extremely diverse , a result that is consistent with the analysis of PvMSP3α gene from the same parasite population [23] , [31] . The P . vivax parasite population has attained a haplotype diversity approaching 1 at both loci . Our finding of extensive diversity of the vivax parasite population in the western border of Thailand is in line with the results from genotyping other genetic loci such as PvMSP1 and PvAMA1 [57] . Despite the low level of malaria transmission in this region , genotyping PvMSP3α and PvMSP3β loci consistently revealed that more than 30% of vivax patients still harbored mixed strain infections [58] . Mixed parasite infection may have resulted from activation of latent heterologous hypnozoites in the liver by new parasite infections [59] , . Inoculation by the same mosquito carrying mixed parasite strains [61] or by different mosquitoes carrying different parasite strains may play an additional role . Mixed-strain infections favor genetic recombination , generation of new genetic alleles , and maintenance of genetic diversity . In agreement , analysis of intragenic recombination revealed numerous number of recombination events within the PvMSP3β gene . Ultimately , high genetic recombination rates correspond to the high genetic diversity observed in western Thailand . Since the levels of genetic diversity and endemicity are often correlated [62] , [63] , the high genetic diversity observed with the PvMSP3β gene may indicate higher malaria endemicity at the Thai-Myanmar border area . Using the PvMSP3β gene as a molecular marker , we have detected significant parasite population differentiation . While it is easy to understand that the Thai and American parasite populations differ drastically because of geographical separation ( Fst = 0 . 28 ) , we also observed significant population structure of the parasites from the GMS . While this result supports those derived from analysis of other polymorphic markers , such studies will need to be bolstered by using larger sample sizes and multiple molecular markers . Our block-wise analysis has shown significant negative departure from neutrality of Tajima's D , Fu and Li's D* and F* values for the dimorphic Bangl type of block 5 , which mostly include isolates from southern Thailand . This seems to be due to population expansion after bottleneck effect rather than strong purifying selection because the rates of nucleotide substitutions at synonymous and nonsynonymous sites did not differ significantly ( Table 1 ) . In the extreme south that borders Malaysia , control activities since the early 1990s have led to a sharp reduction of the annual malaria incidence . As a result , the parasite genetic diversity has also been curtailed . Genotyping multiple merozoite antigens from the southern parasite population revealed drastically reduced haplotype diversity [57] . While reduced genetic diversity of the malaria parasite populations due to intensified control efforts has been reported for P . falciparum parasites [64]–[66] , P . vivax populations are more resilient to control measures . For example , in the Solomon Islands which is progressing towards malaria elimination , despite that the P . falciparum population exhibited low diversity , P . vivax populations remained highly diverse as revealed by microsatellite genotyping [63] . Similarly , in the temperate zone region of central China , P . vivax parasites also maintained relatively high genetic diversity as shown for the pvmsp3β diversity in spite of drastic reduction in annual incidence [34] . In this regard , our finding of a single haplotype from 28 southern Thai parasite samples suggestive of clonal expansion is truly surprising . Without a potential cause of selective sweeps such as drugs , mosquitoes or host immunity that might have led to the sharp decline of genetic diversity in this region , a bottleneck hypothesis is very plausible [57] . Under this scenario , extensive malaria control efforts and lack of re-introduction from the neighboring country dramatically reduced the parasite population size , resulting in the survival of only a small number of parasite strains . In recent years , political unrest and the consequential inability to deploy effective control measures in the southern region have led to an increase in malaria incidence during the past decade . Since the majority of the cases in the southern region were indigenous , they were highly likely resulting from clonal expansion of the remaining parasite strains . This suggestion based on the analysis of pvmsp3β is supported from the analysis of another highly polymorphic marker PvAMA1 [57] . This analysis also highlights the need of more genetic markers to infer population structure of the parasites . These findings suggest that initial non-vaccine control measures leading to severe parasite population bottlenecks may circumvent the problem associated with antigen polymorphism in vaccine development . The high level of genetic polymorphism of vaccine candidates poses a major challenge for malaria vaccine development . Since effective immunity against these vaccine candidates are often strain-transcending , meaning that antibodies against one strain may not be effective against another , the genetic diversity of a vaccine candidate needs to be evaluated in regions of potential deployment . Many of the MSP proteins contain highly polymorphic and conserved domains , signifying the results of diversifying selection and functional constraints , respectively . The most studied PvMSP1 has a mosaic structure composed of seven conserved and six variable blocks . The potentially high frequencies of recombination suggested from the presence of numerous recombination sites within the locus may be responsible for the observed linkage equilibrium at the distance of>3 kb [55] . In the case of pvmsp3β , we identified a remarkably high level of nucleotide diversity , especially in the 5′ half of the gene . This genetic diversity might have been maintained by balancing selection and reinforced by frequent genetic recombination , which is supported by the high Rm values , and identification of decay of LD with the increase of molecular distance in the N-terminal region . This finding together with our earlier study of PvMSP3α suggests that PvMSP3 family proteins share a similar domain structure with N-terminal regions exposed on the merozoite surface , which are subject to strong balancing selection by host immunity . This has potential implications for further considering PvMSP3β's vaccine potential , as antibodies against such domains are often allele-specific [67] . The high level of genetic diversity in the N-terminal region argues against consideration for vaccine development , whereas the relatively conserved C-terminal region may be more favorable . In endemic area of Brazil , a large proportion of vivax-infected individuals developed antibodies against PvMSP3α and PvMSP3β , indicating immunogenicity of both proteins [42] . As speculated , significantly more people contained antibodies recognizing the C-terminal portion of the PvMSP3β protein [42] . Since the PvMSP3 gene family in a parasite strain can contain more than 12 paralogs and they potentially have redundant roles [21] , [22] , evaluation of the vaccine potential for conserved members is needed . | With intensified malaria control in endemic countries , there have been dramatic changes of malaria epidemiology . One of such changes is the increased proportion of Plasmodium vivax malaria , a demonstration of resilience of this parasite to control efforts . In Thailand , malaria has been largely eliminated from the central plain , and transmission is concentrated in isolated international border regions . This study aimed to examine whether the changing malaria epidemiology was reflected in the population dynamics and genetic diversity of the isolated parasite populations . We collected parasite samples from two regions in Thailand with drastically different endemicity settings and used a polymorphic genetic marker ( Plasmodium vivax merozoite surface protein 3β – pvmsp3β ) as an indicator of genetic diversity of the populations . Analysis of the pvmsp3β sequences revealed high genetic diversity of parasites from western Thailand , and suggested the suitability of this gene as a molecular marker to infer parasite genetic diversity . Comparing the pvmsp3β sequences , we further discovered extreme divergence in genetic diversity between the southern and northwestern Thai P . vivax populations . Our study offers important insights into malaria epidemiology and provides the needed knowledge for designing novel control tools in the malaria elimination campaigns . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"evolutionary",
"biology",
"biology",
"and",
"life",
"sciences",
"parasitology"
]
| 2014 | The Plasmodium vivax Merozoite Surface Protein 3β Sequence Reveals Contrasting Parasite Populations in Southern and Northwestern Thailand |
Neuronal activity is associated with transmembrane ionic redistribution , which can lead to an osmotic imbalance . Accordingly , activity-dependent changes of the membrane potential are sometimes accompanied by changes in intracellular and/or extracellular volume . Experimental data that include distributions of ions and volume during neuronal activity are rare and rather inconsistent partly due to the technical difficulty of performing such measurements . However , progress in understanding the interrelations among ions , voltage and volume has been achieved recently by computational modelling , particularly “charge-difference” modelling . In this work a charge-difference computational model was used for further understanding of the specific roles for cations and anions . Our simulations show that without anion conductances the transmembrane movements of cations are always osmotically balanced , regardless of the stoichiometry of the pump or the ratio of Na+ and K+ conductances . Yet any changes in cation conductance or pump activity are associated with changes of the membrane potential , even when a hypothetically electroneutral pump is used in calculations and K+ and Na+ conductances are equal . On the other hand , when a Cl- conductance is present , the only way to keep the Cl-equilibrium potential in accordance with the changed membrane potential is to adjust cell volume . Importantly , this voltage-evoked Cl--dependent volume change does not affect intracellular cation concentrations or the amount of energy that is necessary to support the system . Taking other factors into consideration ( i . e . the presence of internal impermeant poly-anions , the activity of cation-Cl- cotransporters , and the buildup of intra- and extracellular osmolytes , both charged and electroneutral ) adds complexity , but does not change the main principles .
The transmembrane movements of ions during neuronal activity are inevitably associated with changes in ionic concentrations , both intracellularly and extracellularly . This activity-dependent redistribution of ions can be osmotically imbalanced and consequently can lead to changes in volume of the cells and of the extracellular space ( ECS ) . It is usually assumed that three principal ions–Na+ , K+ , and Cl-—are responsible for the link between transmembrane conductance , voltage and volume alterations associated with neuronal activity . Among them , extracellular K+ experiences the largest relative activity-dependent changes . The increase of the extracellular K+ concentration ( [K+]o ) during neural activity is the easiest to detect , and was recorded first [1 , 2] , using ion-selective microelectrodes [3] . Soon it was discovered that the increased [K+]o induced by electrical stimulation was associated with a 50% reduction of ECS volume ( [4] in the cortex of cat ) . However , it also was shown that the relationship between ions , voltage and volume was not simple , since in some other layers of the cortex , shrinkage of ECS was not detected in spite of considerable elevation of [K+]o [4] . The correlation between the increase of [K+]o and decrease of ECS volume was found in different parts of the nervous system under various conditions ( honey bee eye , light stimulation—[5]; optic nerve , electrical stimulation—[6]; spinal cord , electrical stimulation—[7]; cortex during spreading depression—[8] ) . However , when extracellular concentrations of Na+ and Cl- ( [Na+]o and [Cl-]o ) were measured to obtain a full picture of the anticipated osmotic imbalance of ions , the results were sometimes confusing . For instance , it was shown that the decrease of ECS volume evoked by electrical stimulation was indeed accompanied by a decrease of [Na+]o , since Na+ enters cells during stimulation , but this decrease had the same amplitude as the [K+]o increase , and [Cl-]o started to change only after the stimulation , during so called “self-sustained neuronal afterdischarges” [9] . During spreading depression , when ECS was reduced to one fourth of its original volume [8] , a large increase of [K+]o was indeed exceeded by even larger decrease of [Na+]o [10 , 11] . But the decrease of [Cl-]o was markedly larger ( by 12–31 mM ) than needed for electrical compensation of extracellular cation deficiency . As discussed further below , these changes still have not been fully explained with a mechanistic model . The vertebrate retina presents a case of special interest in this respect because it consists of cells that respond to light differently–with predominantly depolarization in proximal layers ( ganglion cells , amacrine cells; see S1 Text ) , but with hyperpolarization in distal layers ( photoreceptors , horizontal cells ) . Accordingly , during illumination the ECS volume decreases in proximal retina , but increases in distal retina [12 , 13] . In proximal retina , similarly to brain , ECS shrinkage is associated with a [K+]o increase and also with a larger [Na+]o decrease and with a compensating [Cl-]o decrease [14] . In distal retina the changes are reversed: ECS expansion is associated with a [K+]o decrease and with a larger [Na+]o increase [14] . However , [Cl-]o still decreases in the outer retina when it is expected to increase in order to compensate for total ECS cation excess . It should be noted that although measuring extracellular ion concentrations with ion-selective microelectrodes is the best available method to obtain data on ionic redistribution during neuronal activity , these measurements are technically difficult , and results must be interpreted with caution . In the case of [Na+]o and [Cl-]o changes , the measured voltage changes of the ion-selective electrode are small ( except in spreading depression ) and can be partly compromised by possible electrical artifact arising from a combination of changes in field potential and the huge resistance of the ion-selective microelectrode [14 , 15] . Also , the sensors are not absolutely selective; for instance , some Na+-selective sensors are influenced by changes in extracellular Ca2+ [9] , and some Cl--selective sensors respond to pH fluctuations [14] . Most importantly , the changes of extracellular ions and volume evoked by neuronal activity often stimulate reactions of ever present nearby glial elements , which in turn can alter those ionic changes and affect ECS volume [16 , 17] . Nevertheless , it is possible to conclude that measurements with ion-selective microelectrodes revealed a certain pattern of related changes in voltage , volume and distribution of the principal ions ( K+ , Na+ , and Cl- ) that more or less consistently ( within the limits of the method ) repeats itself in various parts of the nervous system . Neuronal excitation , which is in most cases associated with depolarization due to an increase of Na+ conductance , results in Na+ influx responsible for the [Na+]o decrease . This Na+ influx is electrically compensated partly by outward K+ flux and partly by inward Cl- flux , which leads to an increase of [K+]o and a decrease of [Cl-]o . The consequent transmembrane transfer of NaCl into the cells evokes osmotically obliged movement of water , decreases ECS volume , and increases the cell volume . When the neuronal activity is associated with hyperpolarization ( like light-induced responses of photoreceptors and horizontal cells in the distal part of the vertebrate retina ) the changes of ion concentrations and volumes have opposite signs . However , the reasons for the voltage-dependent , osmotically imbalanced redistribution of the main ions that lead to volume changes are not entirely clear . Movement of a very small amount of ions is enough for changes of the membrane potential , and that amount has no practical effect on ionic concentration , so ion and volume changes must be more complex than expected from membrane potential considerations alone . The ionic concentrations are changed when the precise balance between influxes and effluxes through membrane passive and active transport systems that existed at rest is temporarily disturbed during activity . For instance , the light-induced decrease of [K+]o in the distal retina is a result of a temporal inequity between the passive K+ leak out of the photoreceptors , which is quickly reduced by the hyperpolarization , and the active K+ pumping into the cell by Na+/K+-ATPase , which needs time for adjustment [18] . It is natural to assume that the osmotically imbalanced ionic changes that lead to changes in volume could be a result of the unequal exchange of Na+ for K+ by the Na+/K+-ATPase ( 3 Na+ out of a cell for 2 K+ in ) . Because of this , the decrease of [Na+]o during neuronal depolarization could be expected to be larger than the increase of [K+]o . That cation imbalance must be electrically compensated by the extracellular decrease in an anion ( most probably by Cl- ) concentration , further diminishing the extracellular osmolarity compared to the intracellular . On the other hand , the passive transmembrane Cl- flux itself is directly affected by changes in membrane potential–attracted into the cell by depolarization and repelled out of the cell by hyperpolarization . In this case , appropriate Na+ and/or K+ flux is needed for electroneutrality , creating osmotically active NaCl/KCl transfer . Thus , there is currently no simple and single explanation for the link between voltage , volume and ions in the nervous system . Is unequal exchange of Na+ for K+ by the Na+/K+-ATPase responsible for osmotically imbalanced redistribution of ions ? Or is ion redistribution the consequence of the direct influence of the changing membrane potential on Cl- flux ? Maybe both factors play their roles; in this case , what is the contribution of each ? Here we will use computational modeling to provide answers for those questions . Numerous computational models that aim to understand the interrelations among ions , voltage and volume have been developed recently [19–25] , ( for review see [26 , 27] ) . But most of them are based on modifications of the Goldman equation , and the limitations of this approach ( particularly for a dynamic , changing system ) have been well described [27] . Alternatively , a much more attractive “charge-difference” method for calculations of simultaneously changing ionic concentrations , membrane potential , and cell volume was introduced [28] . In this work a charge-difference computational model was used for further understanding of the link between ions , voltage , and volume . The list of new and improved features that make our program different from that published earlier is presented in the Methods . The focus was on the specific roles for cations and anions , which is probably the most important feature of the work distinguishing it from current literature , where Na+ , K+ , and Cl- were usually treated together . Special attention was directed to the energy requirement to support voltage and volume changes . It also was demonstrated that , contrary to intuitive assumptions , changes of the membrane potential do not necessarily lead to changes in volume , and changes of volume can have no effect on the cation concentration and the membrane potential . Additionally , both Donnan and Double Donnan equilibrium were reexamined . Since the water permeability of the membrane is critically important for Donnan equilibrium , some simulations with various values of this parameter were performed . Our program is the only one that is capable of such calculations . It should be noted that bicarbonate ion , which is probably the second most important anion after Cl- , was omitted here . Although HCO3- can move across cell membranes through numerous GABA and glycine channels and by certain cotransporters and exchangers , it is involved in the fundamentally important CO2/HCO3- buffering system , and its concentration is mostly determined by highly diffusible CO2 . Accordingly , it is tightly linked to energy metabolism , generation and evacuation of acidic metabolic wastes and other processes that require separate study , and it is beyond the scope of this paper . The software is offered to share ( https://sites . northwestern . edu/ralcomputational/ ) and significant efforts were directed toward making it flexible and user friendly .
The model calculates membrane potential ( Em ) and cell volume ( Vol ) depending on extra- and intracellular concentrations of K+ , Na+ , and Cl- , their transmembrane conductances , and the activities of the Na+/K+ ATPase , Na+ , K+ , 2Cl--cotransporter and K+ , Cl--cotransporter . These two cation-Cl- cotransporters were included in the model because they directly link Cl- with cations and are prevalent in the nervous system . We also take into account the concentration ( [An-]i ) and mean charge valence ( z ) of intracellular membrane-impermeant osmolytes , which comprise a substantial amount of the intracellular anions . In this respect our model is similar to the most advanced and recent models [23–25 , 28] . Additionally , our model has some capabilities which previous models do not have . First , the calculations can be performed with different values of transmembrane water permeability . In existing models on volume regulation very high water permeability is accepted by default , so water instantly follows the ions and other osmolytes , changing the cell volume , but preventing any osmotic difference between the internal and external solutions . In most of our simulations the water permeability was also assumed to be large in order to focus on other aspects of ion-dependent volume-voltage regulation . But in one series of simulations we used the widest possible range of water permeability ( from infinity to zero ) to investigate its consequences for the cell volume and internal osmolarity . In the extreme case of zero water permeability we can simulate the development of Donnan equilibrium . Second , electrically neutral impermeant osmolytes ( like glucose ) can be added to extracellular fluid to simulate Double Donnan effects . Third , to simulate certain experimental environments , the model can perform calculations in conditions where osmolyte concentrations change with time . In this part of the work , we modeled the buildup of four different substances: a ) external electrically neutral impermeant osmolytes , b ) extracellular NaCl , c ) internal electrically neutral impermeant osmolytes , and d ) internal electrically charged impermeant osmolytes with the addition of an appropriate amount of Na+ to maintain electroneutrality . Besides , the model permits various stoichiometries of the Na+/K+-ATPase , which has been done [24] or can be done [23] in previous work . In most simulations a normal stoichiometry of 3 Na+ to 2 K+ was used , but we also examined a hypothetical case when an electroneutral pump ( 3 Na+:3 K+ ) was responsible for redistribution of the cations , and some other more exotic stoichiometries were also tested . With the exception of mentioned above buildups of external NaCl and electrically neutral impermeant osmolytes , all extracellular concentrations during all other simulations were assumed to be constant ( as for an isolated cell in a Petri dish ) . The parameters are expressed in an easily appreciable physical form and the values are biophysically realistic as far as possible . The conductance of ions is expressed in ions/ ( sec*V ) and can be converted to the usual electrophysiological measure of conductance in Siemens . For instance , the value of 2*1010 ions/ ( sec*V ) for total ionic conductance , which was often used in this work , is equal to 3 . 2*10−9 coulombs/ ( sec*V ) , i . e . 3 . 2*10−9 Siemens ( or 3 . 2 nS ) . This corresponds to an input resistance of 312 . 5 MΩ , a reasonable value for a small sized neuron ( for instance , the input resistance of a starburst amacrine cell in the rabbit retina is in the range of 200–250 MΩ [29] . Similarly , the basis for many parameters that were chosen here is experimental work performed mostly on vertebrate retina . For example , a relatively low ratio of gK/gNa is characteristic of the photoreceptors [30 , 31] , which respond to light with a decrease of gNa [32] , as in the Results: “Conductance of Cl- and cell volume . ” . The transporter activities are presented in cycles/sec for better comparison to fluxes through the conductances , since the number of cycles/sec is proportional ( and in some cases–equal ) to the number of ions transferred per second . The same principles for calculations were used here as in the “charge-difference” method of Fraser and Huang [27 , 28] . There was no attempt to derive an equation that describes Em and Vol from ionic concentrations and conductances and activities of transporters . Instead , our program 1 ) counts transmembrane fluxes through the channels and transporters for each ion during a short time period , when ( important ! ) the conditions are assumed to be unchanged , 2 ) calculates the resulting changes in intracellular ionic concentrations , osmolarity and electrical charge at the end of this time period , and 3 ) makes appropriate adjustments of the intracellular concentrations , Em and Vol . These three steps in the calculation are described in the sections below ( e . g . “The 1st set of calculations” ) . This cycle repeats over and over again . If a balanced ( resting ) state was reached , the combined fluxes through all appropriate channels and transporters for each ion ( K+ , Na+ , and Cl- ) would equal 0 and ionic concentrations , Em and Vol would not change . When the conditions ( a conductance or activity of a transporter ) change , the balance will be disturbed , and a new balanced state will be found with potentially new values for Em and Vol , as well as for ionic concentrations . Consequently , at the end of the calculation the activities of the transporters that depend on ionic concentrations can be different from their initial values . The discretization of a continuous process , which is the core of this method , is an apparent idealization , but it does not affect the precision of results when the resting state is found . And during the dynamic phase , any desired level of accuracy can be achieved by choosing an appropriately short discrete time step . In this work the duration of the time step was 0 . 1 , 0 . 5 , or 1 . 0 millisecond . After choosing an initial set of concentrations and pump rates , transmembrane fluxes for each ion ( in ions/sec ) through each transport system are calculated . Inward fluxes are assumed to have a positive sign , and outward fluxes are considered to be negative , in accordance with the way they affect intracellular ionic concentrations . First , fluxes are added separately for each ion and the sums are multiplied by the time step to produce the amount of ions that were moved in or out of the cell . Then the amounts are converted into concentrations . Buildups of extra- or intracellular osmolytes defined by the investigator are also taken into account . Buildups are distinguished from fluxes because these are exogenous substances that are added de novo to one side of the membrane at a specified rate . They then can affect the concentrations of substances , but never affect electrical charge , since they are either electrically neutral substances or an electrically balanced combination of cations and anions . The concentrations at the beginning of the current time step are marked with index b and the concentrations at the end of the time step ( yet before possible adjustments for volume changes ) are marked with the index e . To obtain the changes in total intracellular electrical charge , the change in the amount of Cl- is subtracted from the change in the amount of cations and the result is multiplied by the charge of one cation . dNa= ( NaFc+NaFp+NaFnkc ) *st; ( 12 ) dK= ( KFc+KFp+KFnkc+KFkc ) *st; ( 13 ) dCl= ( ClFc+ClFpc+ClFnkc+ClFkc ) *st; ( 14 ) [osm]o , e=[osm]o , b+bOso*st; ( 15 ) [Na+]o , e=[Na+]o , b+bNaCl*st; ( 16 ) [Cl‑]o , e=[Cl‑]o , b+bNaCl*st; ( 17 ) [osm]i , e=[osm]i , b+bOsi*st; ( 18 ) [An‑]i , e=[An‑]i , b+bAn*st; ( 19 ) [Na+]i , e=[Na+]i , b+dNa/ ( vole*L ) +bAn*st* ( ‑z ) ; ( 20 ) [K+]i , e=[K+]i , b+dK/ ( vole*L ) ; ( 21 ) [Cl‑]i , e=[Cl‑]i , b+dCl/ ( vole*L ) ; ( 22 ) dQ= ( dNa+dK–dCl ) *e; ( 23 ) where dNa , dK , and dCl are changes in the intracellular amount of respective ions; st is the duration of the discrete time step , [osm]o and [osm]i are extra- and intracellular concentrations of impermeant neutral osmolytes; [An-]i is the concentration of internal impermeant anion; z is the mean valence of An; bOso , bOsi , bNaCl , and bAn are “buildups” of extra- and intracellular impermeant neutral osmolytes , external NaCl and of internal impermeant anion , respectively; vole is the cell volume at the beginning of the time step; L is Avogadro’s number ( 6 . 02*1023 mol-1 ) ; dQ is the change in internal electrical charge ( in coulombs ) ; and e is the electrical charge of one cation ( 1 . 6*10−19 coulomb ) . “Buildups” are inputs to the model to allow gradual changes in applied concentrations over some period of time . Notes: All fluxes are added to each other algebraically . For instance , KFp and KFnkc are positive and increase the amount of intracellular K+ , but KFc and KFkc are negative and decrease it ( Eq 13 ) . Addition of Cl- ( dCl ) increases the intracellular osmolarity , but it decreases the total intracellular electrical charge; so dCl is subtracted from total charge ( Eq 23 ) . The buildup of internal impermeant anion ( bAn ) is associated for electroneutrality with buildup of internal Na+ , which must be multiplied by the mean valence of An- ( -z , Eq 20 ) . The changes in the ionic concentrations obtained in the 2nd set of calculations can affect the intracellular osmolarity . Buildup of extra- or intracellular osmolytes , if defined by the investigator , will result in an additional imbalance in osmolarity . In the 3rd set of calculations , water is allowed to move across the cell membrane to restore osmotic equilibrium by changing the cell volume . The calculations can be performed with different transmembrane water permeability , i . e . with different rates of osmotically driven adjustment of the cell volume . This is an important difference of our program from previous models concerning volume regulation . The change in Em is found from the change in intracellular charge ( dQ ) divided by the cell membrane capacitance ( c ) . The concentrations , Em and vol at the end of the current time step , but before the final volume adjustments , are marked with index e and after it with the index f . Em , f=Em , e+dQ/c; ( 24 ) osV= ( [Na+]i , e+[K+]i , e+[Cl‑]ie+[An‑]i , e+[osm]i ) / ( [Na+]o , e+[K+]o , e+[Cl‑]o , e+[osm]o ) ; ( 25 ) VoR=1/tau; ( 26 ) chV=1‑ ( 1‑osV ) *VoR*st; ( 27 ) volf=vole*chV; ( 28 ) [K+]i , f=[K+]i , e/chV; ( 29 ) [Na+]i , f=[Na+]i , e/chV; ( 30 ) [Cl‑]i , f=[Cl‑]i , e/chV; ( 31 ) [An‑]i , f=[An‑]i , e/chV; ( 32 ) [osm]i , f=[osm]i , e/chV; ( 33 ) where tau is the time constant of exponential changes of the cell volume in response to a sudden change in osmolarity , VoR is a constant inversely related to tau , but VoR = 1/st if tau < st and VoR = 0 if tau > 108 sec ( more than 3 years ) , osV is a coefficient of osmosis driven volume changes assuming instant transmembrane water movement , chV is a coefficient of osmosis driven volume changes corrected for limited water permeability of the membrane . Notes: The calculations in this part are straightforward , but the section concerning the water permeability and the time constant of volume changes ( Eqs 26 and 27 ) should be explained . Imbalance between intra- and extracellular osmolarity will induce transmembrane movement of water that changes the cell volume and restores osmotic equilibrium . If membrane water permeability is assumed to be infinite , the water shift and volume changes would happen instantly . In reality , the water permeability is high , but not infinite . It is also different in different cell types . The higher the water permeability , the faster the volume change . The water flux , and consequently , the speed of volume changes also depends on the osmotic gradient , and both the flux and the speed decrease with time as the gradient diminishes . This is similar to the well described discharge of a capacitor in a simple RC circuit . In our case the cell volume is analogous to the voltage of the RC circuit , and membrane water resistance ( which is the inverse of water permeability ) –to the electrical resistance . Thus , the dynamics of cell volume changes evoked by a sudden shift in osmolarity can be described , by analogy with an RC circuit , with the following equation of exponential decay: vol ( t ) =vol1+ ( vol2–vol1 ) * ( 1–e−t/tau ) ( 34 ) where vol ( t ) is the cell volume in time t , vol1 is the initial cell volume , vol2 is the final cell volume with osmolarity equilibrated , and tau is a time constant , which is inversely proportional to the membrane water permeability . Thus , tau is a convenient measure of water permeability , and if water permeability is such that the time constant of the volume change is 1 minute , it means that after 1 minute the cell experiences 63 . 2% of the expected volume change . Twofold smaller water permeability will correspond to twofold larger time constant of the volume changes , and the cell will need 2 minutes for 63 . 2% of the volume change . Eq 25 determines the coefficient osV , by which the cell volume and consequently intracellular concentrations must be corrected in order to return to osmotic equilibrium . But due to limited water permeability another coefficient ( chV ) , which is a fraction of osV and determined by Eq 27 , is used later for correction of the cell volume and the intracellular concentrations . The smaller tau , the larger chV ( and closer to osV ) . For practical purposes , in Eq 27 we used VoR , which is the inverse of tau . It permits a pair of convenient conditions . If tau determined by the investigator is 0 or just shorter than the time step ( st ) , VoR is assumed to be equal to 1/st; as a result osV = chV , i . e . water instantly corrects the cell volume and osmolarity . If extremely low water permeability with tau > 108 sec is chosen , VoR is assumed to be equal to 0; as a result osV = 1 , i . e . water does not move across the membrane at all as in simulation of Donnan equilibrium . After completing the 3rd set of calculations , the program moves forward in time by one time step and repeats all the cycles . Our modeled cell is assumed to have an effective volume of 7 . 5*10−13 L , equivalent to a cube 10x10x10 μm , with 25% of the volume occupied by organelles ( like in rat rod photoreceptors [38] ) . All calculations are done with respect to this effective cell volume of water containing ions and other osmolytes . Since the surface area of a cube 10x10x10 μm is equal to 6*10−6 cm2 , the total membrane capacitance of the cell is 1 . 2*10−11 F ( given a specific capacitance for a neuronal membrane of 2 μF/cm2 ) , and remains the same in all calculations . During volume changes cells usually alter their shape but keep the same surface area ( for a recent reference see [39] ) . We specifically pointed out that our model cell is a cube , which permits an increase in cell volume by almost 40% by turning it into a sphere without any changes to surface area , and consequentially with no change in capacitance . During our simulations the volume increases were usually within this range , except in the catastrophic occasion of a swelling cell whose membrane was permeable to Na+ , Cl- and water , but lacked Na+/K+-ATPase ( Figs 1D and 2C ) . Nothing prevents the cell from keeping the same surface area when the volume decreases . The initial conditions and manipulations of the examined system ( concentrations , conductances , transporter activities ) varied from simulation to simulation , and their exact values are listed in the figure legends . It should be noted that initial conditions that describe the starting point of the system before manipulations can refer to equilibrium states ( Figs 1–5 ) or nonequilibrium but steady states with stable concentrations and Em ( Figs 6–10 ) . In the second case ionic conductances and the rate of the Na+/K+-ATPase were chosen for reasons explained in the text , but concentrations and Em that are given as initial conditions in the figure legends were the results of preparatory calculations that led to the state of the system at t = 0 . Preparatory calculations were also necessary to determine initial conditions in some equilibrium states . One particular set of conditions includes 6 mOsm external impermeant osmolyte and a valence of -1 . 5 for internal impermeant anions , which exemplifies an osmolarity-charge asymmetry , i . e . conditions when the quantity of equivalent charge is not the same as the quantity of osmotically active molecules . Since such asymmetry is expected to be common in real cells , this set is called the “realistic” conditions and often will be compared to “simplified” conditions , which are symmetrical in osmolarity-charge respect , when the valence of internal impermeant anions is equal to -1 and there are no other external osmolytes besides NaCl and KCl .
The classical example of a system that generates a considerable transmembrane potential without spending energy is the one based on Donnan equilibrium . The conditions which can lead to Donnan equilibrium—unequal distribution of Cl- across the cell membrane and selective permeability of the membrane to cation ( s ) and to Cl- , but not to other intracellular anions–are typical for living cells , including neurons . Those impermeant “other intracellular anions” consist of a diverse group of large and small molecules which contribute noticeably to the voltage-volume regulation , and they are addressed specifically later . For the current simulation of Donnan equilibrium let us just assume that the impermeant anions ( [An-]i ) account for most of the internal anion concentration ( 135 mM ) and have a mean valence = -1 . At the beginning ( before time = 0 ) the membrane is permeable to nothing , the membrane potential is 0 mV , and for the osmotic and electrical balance the extracellular Cl- concentration ( [Cl-]o ) is 150 mM , the intracellular Cl- concentration ( [Cl-]i ) is 15 mM , and cation concentrations ( in this case Na+ ) are 150 mM , both inside and outside . The membrane remains impermeable to water , so the cell volume does not change . The results of the simulations are presented in Fig 1A and 1B . At time = 0 the cell begins to be permeable to Cl- and to Na+ . In the simulation , gNa and gCl are set to be equal . Cl- immediately starts to move into the cell because [Cl-]o is 10 times larger than [Cl-]i . Rapidly the influx of Cl- brings negative charge into the cell and hyperpolarizes the membrane . This creates an electrical driving force for Na+ influx , and the entering Na+ partly neutralizes intracellular negativity . As a result , Cl- continues to enter because its strong inward concentration-dependent driving force exceeds its outward electrical driving force . Na+ also continues to enter because its inward electrical driving force exceeds its weak outward concentration-dependent driving force . In 20 minutes both [Na+]i and [Cl-]i are increased by 81 . 8 mM and practically stabilized . At these concentrations ( 231 . 8 mM for [Na+]i and 96 . 8 mM for [Cl-]i ) their Nernst potentials are both equal to the membrane potential Em = - 11 . 6 mV ( Fig 1B ) , which means that the concentration-driven fluxes are counter-balanced by electrically-driven fluxes for both ions and the system has reached a stable equilibrium . This is Donnan equilibrium , and as expected [Na+]i * [Cl-]i = [Na+]o * [Cl-]o . Indeed , 231 . 8 * 96 . 8 = 22438 . 24 ≈ 22500 = 150 * 150 . Additional time is required to achieve a more accurate fit between the results of the calculation and the Donnan expectations . With a chosen accuracy of 15 significant digits , the concentrations are finally stabilized after about 90 min at 231 . 986477689649 mM for [Na+]i and 96 . 9884116601443 mM for [Cl-]i; their product is 22499 . 9999977506 . Special calculations were not performed for validating the model but results like these ( as well as some others that will be presented later ) demonstrate the model’s validity . It is also worth mentioning that the initial change of Em , which looks instantaneous in Fig 1B , in reality decreases exponentially with a time constant of 3 . 7 ms ( see S1 Fig ) , and this is exactly what is expected in our cell , which was set to have an input resistance of 312 . 5 MΩ and a membrane capacitance of 12 pF . Increasing both [Na+]i and [Cl-]i , of course , increases the total intracellular osmolarity . In this example the intracellular osmolarity increased from 300 to 463 . 6 mOsm ( [Na+]i = 231 . 8 mM , [Cl-]i = 96 . 8 mM , and [An-]i = 135 mM ) , creating a strong osmotic gradient . Thus , another fundamental condition that is necessary for Donnan equilibrium is the prevention of transmembrane movement of osmotically obliged water . Otherwise the water enters the cell following ions , increasing the cell volume and diluting ionic concentrations . Exactly that happened when the same simulations as for Fig 1A and 1B were repeated , but under the assumption that the membrane was highly water-permeable , and water instantly compensated for the potential osmotic imbalance associated with ionic transfer ( Fig 1C and 1D ) . As in the previous simulation , opening of Cl- and Na+ conductances permits both ions to enter the cell ( Cl- due to the concentration gradient , and Na+ because of the intracellular negativity created by the influx of Cl- ) . But water also enters , increasing cell volume and diluting intracellular concentrations . One of the effects of the dilution is a decrease of the impermeant anion concentration [An-]i . The other is that the [Na+]i remains the same and equal to [Na+]o because the increase in the amount of Na+ is precisely compensated by the increase in volume . As a result , the equilibrium potential for Na+ is 0 , and Na+ will continue to enter the cell as long as the membrane potential stays negative . Cl- also will continue to enter ( and [Cl-]i will continue to increase in spite of the dilution ) because its equilibrium potential is twice as negative as the membrane potential . Speaking of the membrane potential , in a system of two unevenly distributed ions with different equilibrium potentials , the Em will obviously be somewhere in between those potentials . In this case , when the membrane is equally permeable to both Cl- and Na+ , Em is the arithmetic mean of their Nernst potentials: Em = ( ECl + ENa ) / 2 . Since ENa = 0 , Em should be equal to half of ECl . Again , the simulations show exactly what is theoretically expected . Fig 1C and 1D show that the addition of water permeability to a system which consists of permeable Na+ and Cl- and impermeant An- makes equilibrium unattainable . The equilibrium requires that Em = ENa = ECl , and since ENa = 0 it is only possible if [Cl-]i = [Cl-]o and accordingly [An-]i = 0 . It should be remembered that in the simulation of Fig 1C and 1D the water permeability was assumed to be extremely large , permitting no osmotic imbalance . Although this idealization is not too far from reality and is usually accepted as true in computational models of volume regulation , it is still unrealistic . So Fig 2 shows a series of simulations with the same initial conditions as for Fig 1 but using various values of water permeability expressed for simplicity as a time constant of osmotic volume adjustment ( see Methods ) . At a reasonable water permeability ( time constant = 1 min , Fig 2A ) the changes of ionic concentrations were similar to the data presented in Fig 1C ( instant transmembrane water transfer ) , and at very low water permeability ( time constant = 1 hour , Fig 2B ) the concentrations were similar to the data presented in Fig 1A ( no transmembrane water transfer at all ) . The larger the water permeability ( shorter time constant ) , the faster the volume increase ( Fig 2C ) . At a time constant of 1 sec the volume increase is indistinguishable from that under the assumption of instant water movement , and it is not much different at a realistic time constant of 1 min . Importantly , even with unrealistically low water permeability ( time constant = 1 hour ) the cell volume will increase slowly but steadily , theoretically , to infinity , and practically until the cell blows up . Also , the smaller the water permeability ( longer time constant ) , the larger the transmembrane osmotic gradient ( Fig 2D ) . Bacterial and plant cells can counteract the osmotic pressure with hydrostatic pressure because they have rigid cell walls to preserve the cell volume . Animal cells have no such walls , and their ability to withstand osmotic pressure is limited . Thus , in animals , osmotically imbalanced transmembrane transfer of ions is inevitably associated with changes in cell volume . The time constant of volume adjustment of neuronal and glial cells is in the range of tens of seconds to a minute . For instance , the osmotically evoked volume increase of retinal Muller cells has a time constant of 0 . 5–1 . 0 minute [40] . The light-induced volume changes of vertebrate photoreceptors have approximately the same dynamics , judging by changes of ECS in the retina [13 , 14] . For all other calculations in this work , simulations were done with a highly water permeable membrane ( volume time constant = 1 sec ) in order to focus on other aspects of voltage-volume regulation . This will affect the dynamics ( although slightly , as indicated by the similarity of the 1 min and 1 sec curves in Fig 2C ) , but not the overall conclusions . As we can see , Donnan equilibrium is not applicable for animal cells , including neurons . However , it is theoretically possible to achieve equilibrium in conditions described above if an impermeant osmolyte were added to the external solution to create a so-called Double Donnan system [41] . Let us assume that the extracellular solution contains 135 mM of an impermeant neutral osmolyte , the same concentration as the internal impermeant anion ( An- ) . To keep the same external osmolarity , [Na+]o and [Cl-]o have to be reduced from 150 mM to 82 . 5 mM . Once again , after opening the gNa and gCl , Na+ , Cl- , and water will enter the cell and volume will increase . Again , [Na+]i does not change , and [Cl-]i increases when [An-]i decreases , creating the illusion that An- is being replaced by Cl- ( Fig 3A ) . And again , for equilibrium , both ENa and ECl should be equal to Em . But in this case it is possible because [Na+]o ≠ [Na+]i , and ENa = -15 . 96 mV . After 30 min [Cl-]i increases to 45 . 35 mM which corresponds to ECl = -15 . 98 mV , and the system is approaching equilibrium ( Fig 3B ) . This is the Donnan equilibrium and when [Cl-]i reaches 45 . 375 mM , the equation [Na+]i * [Cl-]i = [Na+]o * [Cl-]o will be true . Of course , entering Na+ , Cl- , and water will increase the cell volume , but it stabilizes at 129% of the initial value . Thus , in a Double Donnan system equilibrium can be achieved without compromising osmotic balance if an impermeant external neutral osmolyte is present in a sufficient concentration . The problem is that in order to keep [Cl-]i low ( and make osmotic room for important internal anions like proteins and nucleic acids ) the concentration of the external neutral osmolyte must be high . However , in reality the total concentration of all external impermeant neutral osmolytes is quite low . Glucose is by far the most concentrated neutral osmolyte in ECS ( 5–6 mM ) . Numerous others have concentrations of small fractions of mM , and the total concentration of all external impermeant neutral osmolytes in normal conditions hardly ever exceeds 10 mOsm . But with only 10 mOsm of the neutral osmolyte , [Na+]o and [Cl-]o will be 145 mM , so according to the Donnan equation [Cl-]i will be 140 . 17 mM , leaving almost no osmotic room ( less than 10 mM ) for other internal anions including vitally important proteins and nucleic acids . The most concentrated substance outside of the cell is Na+ . If the membrane were impermeant to Na+ , the Double Donnan system could be created by utilizing Cl- and another permeant cation , K+ ( Fig 3C and 3D ) . Initial concentrations for these simulations are: [Na+]o = [Na+]i = 145 mM , [K+]o = [K+]i = 5 mM , [Cl-]o = 150 mM , [Cl-]i = 15 mM , and [An-]i = 135 mM . When gCl and gK are open , both ions will enter the cell as in previous simulations . Water will follow and increase the cell volume and decrease the concentration of impermeant substances , which are [Na+]i and [An-]i in this case ( Fig 3C ) . After approximately 5 minutes , the equilibrium potential for K+ decreases and the equilibrium potential for Cl- increases to the same level as membrane potential , EK = ECl = Em , ( Fig 3D ) , and equilibrium is reached . The equilibrium potential for Na+ is different from the membrane potential , but it has no consequence since we assumed that the membrane is not permeable to Na+ . Needless to say , this is purely theoretical and an absolutely unrealistic case , because every neuron has a significant Na+ conductance , and the presence of even the smallest Na+ conductance makes this equilibrium unachievable ( see S2 Fig ) . Concluding this part , it should be noted that the equilibrium conditions described above were determined completely by concentrations of the ions involved . The values of their conductances influence the time to achieve equilibrium but have no effects on the equilibrium potentials . Accordingly , when equilibrium is reached it cannot be changed by alterations of conductances . The simulations in this section deal with two cations , Na+ and K+ , and consequences of their nonequilibrium distribution across the cell membrane due to activity of the Na+/K+-ATPase . It has been recognized for more than a half century that “the big triad” of Na+ conductance , K+ conductance and Na+/K+-ATPase not only determines the membrane potential but forms the basis of the whole system of ionic homeostasis . Accordingly , the previous models concerning cell volume regulation starting from the early works [41–43] and up to the most recent [23–25] paid considerable attention to the cation conductances and Na+/K+-ATPase . Nevertheless , some features of the big triad were overlooked , and some others were misinterpreted . The transmembrane redistribution of Na+ and K+ by the Na+/K+-ATPase is illustrated in Fig 4A . Before time = 0 , [Na+]o = [Na+]i = 145 mM and [K+]o = [K+]i = 5 mM . Also 150 mM of an impermeant monovalent anion was present both inside and outside of the cell for electrical and osmotic balance . Since in these calculations the membrane is permeable to nothing but Na+ and K+ , it does not matter what kind of anion it is ( Cl- or something else ) , but it is important that this anion is monovalent to preserve osmolarity-charge symmetry . In an attempt to make the cation transfer electrically neutral , an imaginary electroneutral Na+/K+-ATPase , which exchanges 3 Na+ for 3 K+ was used in this simulation . Also , gNa was equal to gK . At time = 0 the pump starts to transfer Na+ out of the cell and K+ into the cell against growing concentration gradients for both cations . Na+ starts to leak back to the cell and K+ - out of it , and with changes of the cation intracellular concentrations these leaks increase . At the same time the active transport of Na+ and K+ by the pump slows down because it strongly depends on [Na+]i ( see Methods ) , which is decreasing . The pump rate was set high ( 24 billion transfers/sec ) and the pump activity with initially high [Na+]i was 20 . 4 billion transfers/sec . ( For explanation of our definitions of the pump rate and the pump activity see Methods ) . The pump activity decreased to 3 billion transfers/sec in about 1 . 5 seconds when [Na+]i decreased to 8 mM and after about 5 seconds it stabilized at 331 . 4 million transfers/sec when [Na+]i decreased to 2 . 52 mM . At this time the active transport of the pump and passive leaks of Na+ and K+ were equal , and steady state ( when the concentrations , and consequently Em , remain the same ) was achieved . It is important to emphasize that this is not an equilibrium state like Donnan equilibrium , because both ENa and EK are different from Em , and this stable state must be supported by constant energy expenditure . Interestingly , in spite of the effort to make transport electrically neutral , Em also changed , first decreasing to -27 . 35 mV and then increasing and stabilizing at +8 . 90 mV . The reason for Em changes in a system with seemingly equal exchange of Na+ for K+ is the asymmetrical effect of the same absolute changes in [Na+]i and [K+]i on their respective equilibrium potentials . Indeed , when [K+]i increases by 5 mM ( from 5 to 10 mM ) , [K+]i is doubled and ΔEK = -18 . 5 mV; the simultaneous decrease of [Na+]i by 5 mM ( from 145 to 140 mM ) means a relative change of only 3 . 4% and ΔENa < 1mV . The different time courses of EK and ENa ( Fig 4A ) illustrate this asymmetry , contrasting with symmetrical changes of [Na+]i and [K+]i . In the model , the activity of the pump is conveniently expressed in cycles per second and , accordingly , in ATP spent per second . So , we can directly connect the energy spent for the active transport of Na+ and K+ with the final intracellular concentrations of these ions in steady state and the resulting resting Em . The steady state values of concentrations and Em that are reached after a few seconds in the simulation of Fig 4A were the result of a particular rate of steady state ATP utilization ( 331 . 4 million ATPs/sec ) . Additional simulations with larger or smaller initial pumping rates were done and yielded different sets of values of steady state concentrations and Em . This allowed the creation of the graph ( Fig 4B ) on which steady state values of [Na+]i , [K+]i and Em were plotted against an energy cost expressed in ATP spent per second . In subsequent figures lines through symbols will be used for graphs of this type; each set of points associated with a certain abscissa value represents a separate simulation . For instance , the concentrations and Em from simulations presented in Fig 4A are included in Fig 4B and marked by the arrows . As more energy is spent , stronger electro-chemical gradients for Na+ and K+ are created . But as shown above , the stronger electro-chemical cation gradients do not necessary mean more negative Em . With the pump activity at 149 . 8 million ATP/sec , [K+]i = [Na+]i = 75 mM and Em reached its most negative value ( -27 . 35 mV ) for this condition . With the pump activity at 299 . 7 million ATP/sec , the cation concentrations are exactly reversed from the extracellular values ( [K+]i = 145 mM and [Na+]i = 5 mM ) ; accordingly , ENa = -EK , and Em = 0 mV . We regard the expenditure of ATP on the rising side of the U-shaped Em curve to be wasteful , because the same Em could be achieved at a lower ATP cost at a point on the falling side of the Em curve . Thus , we call the range of ATP utilization above the point of minimum Em “overpumping . ” In reality , the Na+/K+-ATPase is of course electrogenic , transferring 3 Na+ for 2 K+ , and that directly influences Em ( Fig 4C ) . Simulations with all the same conditions as previously , but with an electrogenic 3Na+/ 2K+ pump , show that Em is equal not to 0 , but to -17 . 98 mV at the point of the reversal of the cation concentrations . To reach this point , activity of the pump must be 359 . 6 million ATP/sec , i . e . the pump generates a current of 57 . 5 pA ( 3 . 596*108 multiplied by the charge of one cation , which is 1 . 6*10−19 coulomb ) . Multiplication of this current by the input resistance of the modeled cell ( 312 . 5 MΩ ) gives us the same voltage ( -17 . 98 mV ) that was calculated by the model using only the transmembrane movements of mass and charge . ENa and EK calculated from [Na+]i and [K+]i that were determined in simulations for Fig 4C , the voltage generated by the pump ( Epump ) , and the final Em , are plotted against the pump activity in Fig 4D . To calculate Em ( under the modeled condition of equal gNa and gK ) we used the equation: Em= ( ENa+EK ) /2+Epump This analytically calculated Em is equal to the Em calculated by the model with precision better than 3*10−5 mV . Since the program calculates Em from the total intracellular electrical charge and the membrane capacitance ( see Methods , Eq24 ) this fit provides another validation of the program . Obviously , the cost of Na+ and K+ electro-chemical gradients and the resulting Em depends on the intensity of cation leakage . The cell with half the gNa and gK ( input resistance = 625 MΩ instead of our usual 312 . 5 MΩ ) will spend half the energy for the same gradients and voltage , and this is true for both an electroneutral and electrogenic pump ( see S3 Fig ) . The next step toward a more realistic system is considering the fact that the mean valence of impermeant anions is probably never equal to -1 . The value of the mean valence of impermeant anions is difficult to determine experimentally; it probably varies in different cell types and possibly also in the same cells in different conditions . It also can be defined differently ( more on this in Discussion ) . In this work we define the mean valence of the impermeant anions as the total charge of all impermeant intracellular anions divided by their total osmolarity . An exception will be made for Cl- , which is only temporarily , in this set of simulations , assumed to be impermeant; so , Cl- is counted separately from the impermeant anions . For the simulations presented in Fig 4E we assumed that the mean valence = -3 to show what is probably the largest effect . When the mean valence of impermeant anions is larger than -1 ( here and later , when we say “larger” in respect to mean valence of an anion we refer to the absolute value , ignoring the sign ) , fewer anion molecules are necessary to electrically compensate the cations . For instance , 150 mM of monovalent cations from the previous simulation can be neutralized by 45 mM of anions with valence = -3 in addition to 15 mM of Cl- . However , this would result in an intracellular osmolarity of only 210 mOsm/L . Then , to keep internal and external osmolarity equal , all intracellular concentrations should be proportionally increased by a ratio of 300/210 , so the conditions for Fig 4E when the pump rate is zero are [Na+]i = 207 . 14 mM , [K+]i = 7 . 14 mM , [Cl-]i = 21 . 43 mM and [An-]i = 64 . 29 mM . Since intracellular concentrations of Na+ and K+ are higher than their respective extracellular concentrations , the membrane is hyperpolarized to -9 . 5 mV; this is an equilibrium state because Em = ENa = EK , and support of those unequal transmembrane distributions of Na+ and K+ as well as negativity of Em does not cost any energy . It is notable , comparing the results of calculations presented in Fig 4E ( when the mean valence of the impermeant anions was = -3 ) with that in Fig 4C ( when the mean valence of impermeant anions was = -1 ) , that increasing the mean valence of impermeant anions shifts down the whole curve of Em by the same value ( -9 . 5 mV in this case ) regardless of the pump activity . In both cases the most negative Em was achieved at the pump activity of 201 . 5 million ATP per second . Also , in both cases the same amount of energy ( 359 . 6 million ATP per second ) is necessary to reverse intracellular Na+ and K+ concentrations , although those reversed concentrations were different ( see Fig 4C and 4E ) . Fig 4F shows that similar effects were found in a condition when 15 mM of neutral impermeant osmolyte was added externally ( the mean valence of intracellular impermeant anions was = -1 in this simulation ) . This increase of external osmolarity by 5% requires a proportional increase of all intracellular concentrations ( including Na+ and K+ ) , which in turn leads to generation of a small ENa = EK = Em = -1 . 3 mV in the equilibrium state with no pump activity , as well as a downward shift of the Em curve by -1 . 3 mV throughout the whole range of pump activity . Again , the same amount of energy as in previous simulations was needed to reach the critical point of the most negative Em and reversed [Na+]i and [K+]i . ( see Fig 4C , 4E and 4F ) . In all simulations of this part so far , Na+ and K+ have had the same conductance of 1*1010 ions/ ( sec*V ) , which is equal to 1 . 6 nanosiemens , resulting in a realistic input resistance of our modeled cell ( 312 . 5 MΩ ) . But it is typical for neurons that their gK is several times larger than gNa . For next set of simulations gK = 1 . 8*1010 and gNa = 2*109 ions/ ( sec*V ) , so the input resistance remains the same , but the ratio gK : gNa is 9 : 1 . The results of calculations performed with “simplified” conditions ( mean valence of internal impermeant anions = -1 , concentration of external impermeant osmolyte = 0 ) are presented in Fig 5A . They are clearly different from the results of calculations performed with the same conditions , but with gK = gNa ( Fig 4C ) . First , as expected , Em is significantly more negative . According to the chord conductance equation [44] Em= ( gK*EK+gNa*ENa ) / ( gK+gNa ) an increase of the K+ contribution makes Em more negative . Interestingly , even in this case Em can be still “overpumped” i . e . the negativity of Em diminishes when the pump activity grows too large . Presumably there is little or no advantage in spending this energy for the pump when it leads to a smaller value of Em . Second , a high gK : gNa ratio enables the pump to spend less energy for creating ionic gradients . For instance , 179 . 8 million ATP per second is needed to equilibrate [Na+]i and [K+]i ( both equal to 75 mM ) when gK = gNa ( Fig 4C ) , but only about 55 million ATP per second is sufficient when gK : gNa = 9:1 ( Fig 5A ) . Repeating these calculations with “realistic” conditions ( mean valence of internal impermeant anions = -1 . 5 , concentration of external electrically neutral impermeant osmolyte = 6 mM ) gives results presented in Fig 5B . After necessary osmotic adjustments , [Na+]i and [K+]i in the equilibrium stage ( pump activity = 0 ) increased to 174 and 6 mM , respectively , and the membrane hyperpolarized to Em = -4 . 87 mV . Accordingly , compared to “simplified” conditions ( Fig 5A ) , the range of cation concentration changes is wider and the whole Em curve is shifted down by -4 . 87 mV . The terms “simplified” and “realistic” appear in quotation marks as a reminder that we use them only with respect to different concentration configurations that lead to symmetrical ( convenient in calculations ) or asymmetrical ( usual in nature ) osmolarity-charge configurations , respectively . Here both conditions are tested in a purely theoretical case where there is no Cl- conductance , and later they will be applied to much more real situations with Cl- conductance present and intracellular Cl- concentration affected by cation-Cl- cotransporters . With many differences described above , all graphs of this part have one thing in common–the changes of [Na+]i were always mirrored by the changes of [K+]i , i . e . all removed Na+ was replaced with an equal quantity of K+ . This is true even if the only job of ATPase was to remove Na+ ( 3Na+ per ATP in this simulation ) without transferring any K+ . As soon as both [Na+]i and Em decrease due to the electrogenic 3Na+-ATPase , Na+ begins to leak back to the cell and K+ also enters the cell attracted by the negativity . After some time a steady state will be established when K+ will be in equilibrium ( EK = Em ) and the passive leak of Na+ into the cell will be equal to the active Na+ pumping . The larger the activity of the pump , the stronger the electro-chemical Na+ gradient , and the more Na+ is replaced with K+ in the cell , as illustrated by Fig 5C ( “simplified” conditions ) and 5D ( “realistic” conditions ) . Here all energy was spent for the Na+ gradient; K+ was distributed passively . But still K+ plays key role - replacing Na+ it makes possible creating the Na+ gradient . If gK = 0 , i . e . K+ cannot enter the cell , [Na+]i would remain practically the same regardless of the activity of a pump moving only Na+ . In this case all energy of the pump will be spent on generating a negative Em , which drives back all Na+ that was actively removed . For better comparison of the effects that different parameters of the modelled cell have on the relation between Na+/K+-ATPase activity and Em , the results of several calculations are presented together in Fig 5E . A pair of simulation conditions—“simplified” and “realistic”–was used in each of four general settings . In Setting 1 an electrogenic 3Na+/2K+ pump was used and gK = gNa with the input resistance = 312 . 5 MΩ ( as in Fig 4C ) . Other settings were different from Setting 1 in one of the following respects - imaginary electroneutral 3Na+/3K+ pump in Setting 2 ( as in Fig 4B ) , ratio gK : gNa = 9:1 in Setting 3 ( as in Fig 5A and 5B ) , and imaginary only Na+ pump ( 3Na+ per ATP ) in Setting 4 ( as in Fig 5C and 5D ) . Thus , it is proper to compare Setting 1 with the three others . As expected , an electroneutral 3Na+/3K+ pump ( Setting 2 ) had a smaller effect on Em than an electrogenic 3Na+/2K+ pump ( Setting 1 ) ; what was interesting is that the difference was not large before the pump activity reached the level of overpumping . From the energy point of view the most interesting result comes from comparing Setting 1 with Setting 3 - high gK relative to gNa enabled the pump to create stronger ionic gradients and promote a more negative Em while spending less ATP . A strongly electrogenic 3Na+ pump ( Setting 4 ) significantly hyperpolarized the membrane at high pump activities , but it came with a high ATP price if gK = gNa . It is also notable that under every setting and with every pump activity the difference between the “simplified” and “realistic” conditions was always the same and equal to what was it was in the passive , no pump condition , i . e . -4 . 87 mV . As we already pointed out , in all simulations in this section the internal Na+ was replaced with an almost exactly equal quantity of the external K+ regardless of relative conductance of these ions and the stoichiometry of the pump . Here we would like to emphasize the word “almost , ” since the exchange of Na+ for K+ was not exact . The negative Em means that there is some deficiency of internal cations . Similarly , when ions were overpumped sufficiently for Em to become positive , it is due to some surplus of internal cations . Subtraction or addition of an intracellular substance leads to osmotic imbalance , compensatory water movement , and consequently , appropriate cell volume changes . The volume changes associated with Em changes from Fig 5E are presented in Fig 5F . As expected , dependence of the cell volume on the pump activity closely followed the dependence of Em on the pump activity in every setting . Since all volumes were normalized to the initial cell volume in the passive state with no pumping ( defined as 100% ) , the pre-existing cation deficiency associated with initial negativity of Em in the “realistic” conditions ( i . e . the -4 . 87 mV ) was already counted; so , the curves for the volume changes in the “simplified” and “realistic” conditions are identical in all four settings . The main result related to the cell volume , however , is that its changes are extremely , unnoticeably small . A tiny quantity of ions is needed to recharge the cell membrane significantly . A simple calculation shows that 9 . 6*10−13 coulomb of charge will hyperpolarize our modelled cell , with membrane capacitance 1 . 2*10−11 farad , to -80 mV . This charge is carried by 6*106 ions , which equals 13 . 3 μM in our cell with volume 7 . 5*10−13 L , i . e . only 0 . 0044% of the total internal osmolarity . The results of simulations presented in Fig 5C and 5D demonstrate precisely that: when the cell was hyperpolarized to about -80 mV ( setting 3 ) , the cell volume change was a bit more than 0 . 004% . Thus , we can conclude that the Na+/K+ pump replaces internal Na+ with practically the same quantity of K+ . Accordingly , the pump has practically no measurable effect on the cell volume , and this is true with any stoichiometry of the pump and conductance of Na+ and K+ . To summarize this part , we can conclude that in the “only cation” system tested above three properties - Na+ conductance , K+ conductance and Na+/K+-pump activity - determine with certainty the values of three features - [Na+]i , [K+]i and Em . When the properties are constant , the determined features also stay unchanged in steady or resting state . But it is not an equilibrium state , and a constant expenditure of energy is required to keep it . Changes in the pump activity lead to changes in [Na+]i , [K+]i and Em until a new steady state is reached . The same is true for changes of Na+ or K+ conductances . It is also important to note that changes in the pump activity as well as changes in cation conductances have no practical effects on the cell volume . In this part , the cation system described above will be enriched by addition of Cl- conductance and later by Cl--cation cotransporters . Cl- is by far the most concentrated external anion and numerous Cl- permeable channels and Cl- transferring transporters make this ion unavoidably important for the nervous system . Here we will show that Cl- conductance is the reason for voltage-related cell volume changes . The results presented in Fig 6 illustrate the changes in concentrations , voltage and cell volume when Cl- conductance was opened at time = 0 to disturb the resting state achieved with open cation channels and active Na+/K+-ATPase ( gNa = 8*109 ions/ ( sec*V ) , gK = 1 . 2*1010 ions/ ( sec*V ) , pump activity 2 . 64*108 ATP/sec , Em = -43 . 3 mV ) . This resting state , with a relatively high gNa and a moderate Em , was chosen so that changes in gCl could potentially cause either depolarization or hyperpolarization . As a result of this disturbance [Cl-]i changed significantly , but [Na+]i , and [K+]i , changes were barely noticeable ( Fig 6A ) and only transient ( Fig 6B ) . Regardless of the initial [Cl-]i , ( 15 mM for solid and dashed lines , and 45 mM for the dotted line ) or the value of gCl ( 1011 ions/ ( sec*V ) for the solid line and 1010 ions/ ( sec*V ) for the dashed and dotted lines ) , [Cl-]i in the new resting state was 29 . 6 mM . At this concentration ECl = -43 . 3 mV , i . e . ECl is equal to the Em established by the cations . To electrically compensate transmembrane movement of Cl- , Na+ and K+ move together with it , entering the cell when [Cl-]i increases from 15 to 29 . 6 mM and leaving the cell when [Cl-]i decreases from 45 to 29 . 6 mM . These ionic movements cause the cell volume to increase in the former case and decrease in the latter case ( Fig 6A ) keeping cation concentrations constant . The intracellular concentration of impermeant anion An- experiences opposite changes to Cl- as a result of the volume change , keeping [Cl-]i + [An-]i almost constant . The small deviations of total anion concentration ( ~0 . 25 mM and ~0 . 55 mM with gCl of 1010 ions/ ( sec*V ) and 1011 ions/ ( sec*V ) , respectively ) from the initial value of 150 mM that peaked in the first 2–3 seconds after increasing gCl was a result of restricted transmembrane water movements assumed in this modeling . There was no such deviation if we assumed instant water transfer . Some small changes of [Na+]i and [K+]i , as well as of Em , also happened after opening of Cl- conductance , but , in contrast to anion concentrations and volume , both cation concentrations and Em recovered to their original values in 5–6 minutes when the system reached the new steady state ( Fig 6B ) . In the case of low initial [Cl-]i ( 15 mM , dashed lines ) Em was temporary hyperpolarized , Cl- and both cations entered the cell , and the cell volume increased . Initially Na+ entered the cell faster than the volume increase and [Na+]i slightly increased; at the same time [K+]i decreased ( in spite of the influx of K+ ) and the sum of [Na+]i and [K+]i remained practically equal to 150 mM , with small deviations in the first seconds due to limited water transfer , as in the case of the anions . The transient increase of [Na+]i and the decrease of [K+]i are due to the stronger diluting effect of increasing volume on the ion in higher concentration , which is K+ . Opposite transient changes in Em , [Na+]i and [K+]i occurred if [Cl-]i started at 45 mM ( Fig 6B , dotted lines ) . If the steady state condition before opening gCl had been set to achieve [Na+]i = [K+]i = 75 mM ( at a pump activity 1 . 66*108 ATP/sec ) , opening of gCl would have led to a decrease of [Na+]i and an increase of [K+]i , because K+ conductance in these simulations was set to be larger than Na+ conductance . If the initial steady state condition had been [Na+]i = [K+]i = 75 mM and gNa = gK ( at the pump activity 1 . 80*108 ATP/sec ) , opening of gCl would have had no effect on the cation concentrations , besides small and short-lived deviations ( both increases ) related to delayed water movements , and it would have been the same for electrogenic and neutral pumps . gCl affected nothing except the time necessary for equilibration; it took 6–7 minutes with gCl = 1010 ions/ ( sec*V ) ( dashed and dotted lines in Fig 6 ) and 3–4 minutes with gCl = 1011 ions/ ( sec*V ) ( solid lines in Fig 6 ) . Calculations presented in Fig 6 , parts A and B were done in “simplified” conditions , i . e . assuming that the mean valence of internal impermeant anions = -1 and that there were no external osmolytes besides Na+ , K+ and Cl- . Largely similar results were obtained in “realistic” conditions ( the mean valence of internal impermeant anions = -1 . 5 and the concentration of external electrically neutral impermeant osmolytes = 6 mM ) with the most notable difference being in the cation concentrations ( Fig 6C and 6D ) . Smaller amounts of polyvalent intracellular anions were needed to electrically compensate intracellular cations and that , together with the addition of impermeant extracellular osmolyte , demanded certain osmotic adjustments that affected the initial ionic concentrations . As a result , the resting [Na+]i and [K+]i were different from those in “simplified” conditions described above with the same rate of the Na+/K+-ATPase . Moreover , [Na+]i and especially [K+]i were different depending on how much of the initial anion concentration was due to [Cl-]i ( 10% and 30% of the total charge of intracellular anions for the dashed and dotted lines , respectively ) . Accordingly , Em also was different under low or high initial [Cl-]i , although the difference was only ~1 mV ( Fig 6D ) . In this “realistic” condition , just as in the previously described “simplified” condition , opening of gCl led to changes of [Cl-]i toward a new value that was the same regardless of initial [Cl-]i . Again , the transmembrane movement of Cl- was accompanied by co-directed movements of Na+ and K+ that led to appropriate changes in the cell volume and consequently [An-]i ( Fig 6C ) . However , the absolute change of [An-]i was 1 . 5 times smaller than the change of [Cl-]i , since each An- was carrying 1 . 5 times more charge . In this condition , [Na+]i and [K+]i , and accordingly Em , were also shifted to new levels ( Fig 6D ) . When a new resting state was established , Em and all four cation and anion concentrations stabilized at new values which were not dependent on initial [Cl-]i . And the new [Cl-]i was again exactly what was required to make ECl = Em ( -46 . 15 mV ) . Since Cl- is in equilibrium in this new resting state , alterations in gCl cannot change anything in the system . But alterations in conductances of nonequilibrated cations can , and some of the results are different depending on gCl . Fig 7 illustrates how temporal changes of gNa affect the ionic concentrations , cell volume and voltage under various gCl . The cell modelled here is permeable to Na+ , K+ , and Cl- and is at rest under “realistic” conditions . So , it is similar to the one presented in Fig 6C and 6D , but with one difference in Fig 7A and 7B: it uses the imaginary electroneutral 3Na+/3K+-ATPase that does not transfer any net charge or mass and consequentially cannot directly influence the cell volume or voltage . At time = 0 gNa was reduced by a factor of 4 ( from 8*109 to 2*109 ions/ ( sec*V ) ) and 10 seconds later gNa returned to its original value . The temporary decrease of gNa leads to a hyperpolarization of Em , moving it closer to EK . This reduces the driving force for K+ , and passive K+ efflux decreases . Passive Na+ influx also decreases ( in spite of the increased driving force for Na+ ) due to the reduction in gNa . But the Na+/K+-ATPase continues to pump K+ in and Na+ out of the cell initially with the same activity . As a result , [K+]i increases and [Na+]i decreases during the temporary decrease of gNa . These cation changes are almost identical when gCl is negligible ( 108 ions/ ( sec*V ) , i . e . less than 0 . 5% of the total transmembrane conductance , dashed lines ) or considerable ( 1010 ions/ ( sec*V ) , i . e . more than 30% of the total transmembrane conductance , solid lines ) . Em is more sensitive to gCl ( Fig 7B ) , demonstrating the well-known “shunting inhibition” ( see S2 Text ) , supported in many neurons by the Cl- permeable GABA- and glycine-gated channels . But it is the cell volume that is affected most by the value of Cl- conductance . The hyperpolarization evokes efflux of Cl- , which was in equilibrium before the time = 0 . In order to electrically compensate it , an efflux of cations is required . The model shows that under these conditions Na+ influx decreases more than K+ efflux , causing a net efflux of cations that is equal to Cl- efflux , and when the ions leave the cell , the cell volume decreases . Depending on the value of gCl , the ionic fluxes could be large or small , determining the size of volume changes . Thus , during electrical activity associated with changes of cation concentrations the cell may or may not experience detectable changes of its volume , depending on the value of gCl . It should be noted that the decrease of [Na+]i slows down the Na+/K+-ATPase . The simulations in Fig 7A and 7B used an electroneutral 3Na+/3K+-ATPase , and alterations in the pump activity had no consequences for the cell volume and voltage . But the real electrogenic 3Na+/2K+-ATPase does transfer both charge and mass , and it is intuitively expected that the electrogenic pump should contribute to changes of Em and the cell volume . The results of modeling with an electrogenic Na+/K+ pump are presented in Fig 7C and 7D . As expected , decrease of the pump activity associated with the decrease in [Na+]i clearly manifested itself in a slow reduction of the hyperpolarization ( Fig 7D ) . However , the volume changes were again completely under control of the gCl . Independently of the stoichiometry of the pump , changes in its activity cannot change the cell volume if the membrane is not permeable to Cl- , and on the other hand , when Cl- conductance is considerable the volume changes happen irrespective of whether the pump is electroneutral or electrogenic . One more point should be made concerning the relation between the Na+/K+-ATPase and the cell volume . Activity of the pump , and consequently the expenditure of ATP , follows [Na+]i . Since the changes of [Na+]i were practically identical with high and low gCl ( Fig 7A and 7C ) , the expenditure of ATP was the same and irrelevant to the volume changes . This also can be seen in the simulations of Fig 6 . Changes of [Na+]i tell us that some extra energy was spent during the transition when the cell volume increased ( Fig 6 dashed lines ) and some energy was saved when the cell volume decreased ( Fig 6 dotted lines ) , but after reaching the resting state , the cell spent exactly the same amount of energy to keep the larger volume as to keep the smaller volume . And it also was equal to the amount of energy the cell spent before gCl opening , precisely in “simplified” conditions and with precision of a fraction of 1% in “realistic” conditions . Certain molecular mechanisms can influence [Cl-]i , shifting it away from equilibrium , so that ECl ≠ Em . The most important of these for the nervous system are two cation-Cl- cotransporters - the Na+ , K+ , 2Cl- cotransporter and the K+ , Cl- cotransporter ( NKCC and KCC , respectively ) . Fig 8A shows steady state values of intracellular ion concentrations , Em , and cell volume as a function of the rate of the NKCC , in the condition when Cl- conductance is very small ( 108 ions/ ( sec*V ) ) . The lower row of numbers under the x-axis represents the corresponding activity of NKCC , which gives information on the actual quantity of ions transferred across the membrane . As for the Na+/K+-ATPase , the activity , in distinction to the rate , is dependent on ionic concentrations and will change together with them . The activity of the transporter is k1 times its rate ( see Methods ) , where k1=log10 ( ( [Na+]o*[K+]o*[Cl‑]o2 ) / ( [Na+]i*[K+]i*[Cl‑]i2 ) ) . ( 35 ) The NKCC pumps Na+ , K+ , and Cl- into the cell , and the direct result of that is an increase in cell volume . The higher the cotransporter rate ( and activity ) , the larger the cell volume ( Fig 8A ) . [Cl-]i also increases with the cotransporter rate , but , interestingly , [Na+]i and [K+]i remain almost exactly the same ( note: there is a 3:2 Na+/K+-ATPase in these simulations ) . [Na+]i increased by only 0 . 006 mM and [K+]i actually decreased by 0 . 005 mM . Em also changed very little–from -43 . 3 mV to -43 . 2 mV , reflecting a subtle depolarizing influence of Cl- , which is not in equilibrium in this case . At the highest activity of the cotransporter , [Cl-]i reached 82 . 98 mM; at this concentration ECl = -15 . 8 mV . But the conductance of Cl- in this simulation was very low , so the Cl- contribution to Em is negligible . Fig 8A also clearly demonstrates that the capability of NKCC to elevate [Cl-]i ( and the cell volume ) is limited . When [Cl-]i is increasing , kl approaches 0 . Because the cation concentrations are constant ( Fig 8A ) and so is [Cl-]o , this limiting value of [Cl-]i can be obtained by setting kl = 0 and solving the Eq 35 for [Cl-]i . In our calculations [Na+]o = 145 mM , [K+]o = 5 mM , [Na+]i = 17 . 9 mM , [K+]i = 132 . 1 mM , and [Cl-]o = 150 mM , so the largest [Cl-]i that can be achieved is 83 . 06 mM . [Cl-]i approaches this level when the cotransporter rate is 108 cycles/sec and activity = 1 . 35 million cycles/sec . At that point the driving force of the transporter is almost exhausted , and increasing its rate by 100 times means increasing activity only to 1 . 37 million cycles/sec , i . e . only by 1 . 5% . Fig 8B shows how the activity of NKCC influences concentrations , voltage and volume when Cl- conductance is high ( 1010 ions/ ( sec*V ) ) . In such conditions the cotransporter is similarly capable of elevating [Cl-]i and increasing the cell volume to about the same values as in the case of low gCl , but the activity of the cotransporter has to be roughly 100 times higher because it has to overcome a leakage of Cl- that is 100 times larger through the high conductance . More importantly , the high gCl makes Cl- a noticeable contributor to Em . Thus , the cotransporter-generated increase of [Cl-]i is accompanied not only by an increase in cell volume , but also by a depolarization from -43 . 3 mV to -34 . 6 mV . [Na+]i and [K+]i were again almost unaffected , although their small changes were larger than in the case of low gCl: +0 . 42 mM for [Na+]i and -0 . 42 mM for [K+]i . The KCC uses the strong K+ outward concentration gradient to extract Cl- from the cell . The coefficient k2 that links the activity of the cotransporter to its rate is expressed as follow: k2=log ( ( [K+]o*[Cl‑]o ) / ( [K+]i*[Cl‑]i ) ) . ( 36 ) According to this equation , when k2 = 0 the lowest [Cl-]i which can possibly be achieved in our conditions ( [K+]o = 5 mM , [Cl-]o = 150 mM , [K+]i = 132 . 1 mM ) is 5 . 68 mM , and the calculations show that our modeled cell approaches this limit with a cotransporter activity of 4 . 13 million cycles/sec when gCl is low ( Fig 8C ) . Together with lowering of [Cl-]i , KCC decreased the cell volume , but the cation concentrations remained remarkably similar ( only +0 . 040 mM for [Na+]i and -0 . 037 mM for [K+]i ) , in spite of the fact that the cotransporter removed exactly the same amount of K+ as Cl- . Em also was very little affected - the cotransporter at its maximal activity produced only -0 . 27 mV of additional hyperpolarization . As expected , increasing gCl 100 times demanded much higher activity of the cotransporter for lowering of [Cl-]i toward the limit ( Fig 8D ) . Also , as expected for a high gCl , the decrease of [Cl-]i caused by KCC was accompanied by significant hyperpolarization ( from -43 . 3 mV to -59 . 7 mV ) . The cation concentrations were affected as well , although not as much as [Cl-]i: [K+]i decreased by 2 . 6 mM and [Na+]i increased by the same 2 . 6 mM . It is also noteworthy that activity of both cation-Cl- cotransporters is associated with increased expenditure of energy . Moving one more element of the system ( Cl- ) out of the equilibrium state obviously should cost some extra energy , regardless of the direction of this movement - an increase or decrease of [Cl-]i and , consequently , an increase or decrease of the cell volume and depolarization or hyperpolarization of the cell membrane . In this respect , it is surprising how little extra energy was needed in the case of the NKCC . When [Cl-]i increased by 177% and cell volume increased by 77% with a very active cotransporter , ATP consumption increased only by 2 . 3% . And this was under the high gCl condition . When gCl was low , even larger increases of [Cl-]i and the cell volume were achieved with a tiny cost of 86 , 000 extra ATPs per second , which is 0 . 03% of the total energy . The KCC is more demanding . A decrease of [Cl-]i to 19 . 7% of its initial concentration with a cell volume reduction to 83 . 5% required an additional 12% of ATP when gCl was high , and an extra 2% of ATP did comparable work when gCl was low . All simulations of cation-Cl- cotransporters above were done in the “simplified” condition . We performed the same series of calculations in the “realistic” condition and high gCl for NKCC ( Fig 8E ) and for KCC ( Fig 8F ) . Changes of [Cl-]i in an asymmetric concentration-charge system , like the “realistic” condition , is associated with some additional complications in cation concentrations . The [Na+]i + [K+]i is not constant anymore; contrarily , the sum must change due to changes in the ratio “total anion charge”/”total anion concentration” resulting from opposite changes in the concentrations of monovalent Cl- and polyvalent An- ( more on this in Discussion ) . As a result , the logic of cation concentration behavior induced by cation-Cl- cotransporters is not obvious . Specifically , NKCC noticeably decreased [K+]i , but did not change [Na+]i , in spite of pumping both cations into the cell . In its turn , KCC increased [Na+]i , although it did not transfer this ion; the cotransporter also had a biphasic increase-decrease effect on [K+]i . The movement of water and changes in Em are important in understanding these unintuitive changes . Effects of cation-Cl- cotransporters on [Na+]i and [K+]i are intriguing and deserve a more detailed analysis in the future , but for the purpose of this paper it should be stressed that those effects were small . In most cases there were almost no cation changes compared with changes of Cl- , which was transferred simultaneously with cations and in equal amount . And [Cl-]i changes were always accompanied by changes of the same sign in the cell volume . Low gCl allowed to the system to achieve large effects on [Cl-]i and the cell volume at a small activity of the cotransporters , but high gCl was needed to influence Em . It seems that those cotransporter-evoked Cl--dependent alterations of Em are responsible for the disturbances in cation concentrations . In the last part of this work we will examine how changes of external and internal osmolarity affect the cell volume , [Na+]i , [K+]i , [Cl-]i , [An-]i , and Em . The reason-consequence chain in this section will be different from the previous sections . Up to this point changes in concentration of permeant ions evoked changes in Em and possibly in cell volume , if the redistribution of ions was not osmotically balanced . Here , the initial event was an alteration of osmolarity that directly and predictably influences the cell volume . When external osmolarity increases , the cell shrinks; when internal osmolarity increases , the cell swells . These changes of cell volume may or ( surprise ! ) may not lead to changes in intracellular ionic concentrations , as will be shown . Finally , changes of ion concentration , if they occur , will inevitably change Em . Simulations in this part resemble what happens or may happen during a regulatory volume increase ( RVI ) . The initial set of calculations simulates the first phase of RVI in which extracellular osmolarity is increased by adding some impermeant neutral osmolyte . The set includes simulations with high gCl ( 1010 ions/ ( sec*V ) , solid lines in Fig 9 ) and with low gCl ( 108 ions/ ( sec*V ) , dashed lines in Fig 9 ) , both in “realistic” conditions . The first phase of RVI lasts a few seconds to minutes [45] , so 30 mM of an external osmolyte was added at the rate of 0 . 5 mM/sec for 1 minute ( gray area in Fig 9A and 9B ) . Qualitatively all changes of the ion concentrations , the volume and the voltage were in accord with expectations . The cell shrank , and [Na+]i , [K+]i , [Cl-]i , and [An-]i increased in proportion to their initial level , at least at first glance . Em hyperpolarized , which was anticipated because the increase of [K+]i enhanced the K+ transmembrane gradient and its hyperpolarizing effect , and the increase of [Na+]i diminished the Na+ transmembrane gradient and its depolarizing effect . But the quantitative picture is more complicated . First , the cation increases were not proportionally equal . At the end of the first minute , when external osmolarity reached its maximum ( 336 mOsm . i . e . 9 . 8% more than the initial value of 306 mOsm ) , [K+]i increased by more than 10% and [Na+]i increased by a little more than 2% . This is an apparent deviation from the simplistic volume-induced increases of concentrations that were expected to be proportionally equal , and it points toward a redistribution of ions during osmosis-related changes . Redistributions of Cl- are the most interesting because Cl- is tightly connected with cell volume . There were no cation-Cl--cotransporters in this simulation , so Cl- was in equilibrium before the increase of external osmolarity . Osmosis-related shrinkage of the cell increased [Cl-]i and diminished the concentration-dependent inward-directed component of the driving force for Cl- . At the same time a cation-induced hyperpolarization enhanced the voltage-dependent outward-directed component of the driving force for Cl- . Thus , Cl- left the cell , and the increase of [Cl-]i was smaller than expected from the volume decrease itself . The difference , of course , depended on the value of gCl . When gCl was low , the increase of [Cl-]i was close to the expected change from volume alone ( 9 . 5% vs 9 . 7% ) , but when Cl- conductance was high , those two numbers were very different ( 3 . 1% vs 11% ) . After 1 minute of increased external osmolarity [Cl]i was noticeably out of equilibrium . In the case of low gCl , [Cl-]i after 1 minute was 28 . 10 mM; accordingly , ECl = -44 . 7 mV , i . e . 4 . 6 mV more positive than Em ( -49 . 3 mV ) . High gCl shunted the membrane , so the hyperpolarization was smaller ( to -48 . 3 mV ) . The rise of [Cl-]i also was smaller due to Cl- leakage ( to 26 . 47 mM ) , so ECl ( -46 . 3 mV ) was 2 mV more positive than Em . When external osmolarity stabilized , [Cl-]i began to decrease , leading to both further hyperpolarization and a further volume decrease . If Cl- conductance was high , Cl- quickly equilibrated and after 10 minutes [Cl-]i = 23 . 70 mM , i . e . about 2 mM less than the initial concentration . That new [Cl-]i was at equilibrium and fit with new electrical conditions ( ECl = Em = -49 . 3 mV ) . If gCl was low , no changes in [Cl-]i , volume or voltage were visible from 1 to 10 minutes , except a small and quick depolarization that reflected cation adjustment after the disturbance . However , the calculations showed that after several hours the cell would come to the same equilibrium state as in the case of high gCl . Knowing that the ions were redistributed during the osmotic shock , it is not surprising that the cell volume changes were themselves different from expectations . An increase of external osmolarity by 8 . 9% should decrease volume in a cell that is impermeant to anything but water by 8 . 2% , and that was close to the volume reduction at the end of osmotic shock when gCl was low ( 8 . 8% ) . But when Cl- conductance was high , the cell volume decreased by 9 . 8% at 1 minute and by 11 . 8% in the eventual resting state . Fig 9C and 9D demonstrates that fundamentally similar changes happen when NaCl , i . e . a substance that easily can cross the cell membrane , was used instead of a neutral impermeant osmolyte . As in the previous case , the external osmolarity was slowly elevated by 30 mOsm by the end of 1 minute; to do this NaCl was added at a rate of 0 . 25 mM/sec . Addition of extracellular NaCl directly influenced not only external osmolarity , but also the transmembrane gradients of Na+ and Cl- . Increasing the Na+ gradient enhanced the depolarizing effect of this cation on Em , which resulted in a smaller osmotic-dependent hyperpolarization ( if compared to the case of an increased external neutral osmolyte described above ) , but only when gCl was low ( compare dashed lines in Fig 9B and 9D ) . When gCl was high the hyperpolarization during the osmotic shock was even slightly larger . The increased transmembrane Cl- gradient had more recognizable effects . Now Cl- did not move very far from equilibrium , as in the case of a neutral osmolyte . The difference between ECl and Em never exceed 0 . 8 mV with high gCl , and 1 . 5 mV with low gCl . As a result , [Cl-]i experienced much smaller changes after the osmotic disturbance on its way to equilibrium . Accordingly , smaller further hyperpolarization and volume decreases happened after adding NaCl than a neutral osmolyte ( Fig 9B and 9D ) . Again , a long time ( many hours ) is needed to reach equilibrium if gCl is low , which leads to the illusion that nothing changed in this case after the disturbance . It also should be noted that some extra energy is needed to support steady state in a smaller cell volume after the external osmotic increase , although the cost is not high . ATP expenditure increased by about 1 . 5% when the system stabilized after the neutral osmolyte-induced disturbance , and about 3% extra ATP was needed in case of NaCl . In the next two sets of simulations the internal osmolarity was increased . This was similar to the second , active phase of RVI . First , a neutral impermeant osmolyte was added to intracellular space ( Fig 10A and 10B ) . Such an increase could happen when some macromolecules were broken down to many smaller molecules ( like glycogen to glucose ) or some osmolyte ( for instance , taurine ) was transported into the cell from the extracellular space with an appropriate transporter . For the simulation , we assume that a neutral impermeant osmolyte increases with a rate of 0 . 05 mM/sec for 10 minutes until its concentration reaches 30 mM . This might be too fast to be real , but slowing down the process does not change the results ( except diminishing the difference between simulations with different gCl ) . Most ( but not all ) changes of concentrations , voltage and volume associated with elevation of internal osmolarity are just opposite to those observed with a simulated increase of external osmolarity ( Fig 10A and 10B ) . The cell swelled , and intracellular concentrations decreased , with the important exception of [Cl-]i , which increased when gCl was high . Again , decreases of ionic concentrations were not proportionally equal , in spite of the proportionality that would be expected as the direct effect of the increasing volume . [K+]i decreased by more than 10% , while [Na+]i decreased by less than 2% . Em depolarized due to decreases in both cation concentrations . The depolarization forced Cl- to enter the cell , but it had little effect , and [Cl-]i still decreased when gCl was low . However , when gCl was high , this depolarization-driven Cl- influx was substantial and [Cl-]i increased . Accordingly , the difference between ECl and Em was small ( maximum 0 . 44 mV ) , and likewise the effect of Cl- on Em was also small . Still , Cl- was out of equilibrium , and when the buildup was complete , [Cl-]i continued to increase to equilibrate with the new Em , initiating a further depolarization and swelling . The new resting state was reached much more slowly with low Cl- conductance . Finally , although it might cost energy to build up an intracellular osmolyte , the cell actually saved about 1 . 5% of the ATP required to support ionic balance with the new larger volume . The last set of simulations dealt with the curious case of the buildup of an intracellular impermeant anion with a mean valence equal to -1 . 5 . It should be noted that synthesis of a new organic anion must be accompanied by a cation for electroneutrality . The most probable cation in such a case is H+ , so the addition of an anion would also cause the addition of an acid . The regulation of intracellular pH is an undoubtedly important , but complicated , problem that goes beyond the scope of this paper . So , we assume that our modeled cell is capable of resolving the problem of stabilization of pH . For instance , the cell could exchange each new internal H+ for external Na+ using a Na+/H+ exchanger . Thus , in our simulations the buildup of impermeant anion will be supplemented by appropriate addition of Na+ . In the “realistic” conditions of our simulation 3 Na+ were needed to electrically compensate 2 An- , which have a mean valence = -1 . 5 . Accordingly , buildup of the anion with rate of 0 . 02 mM/sec for 10 minutes was accompanied by addition of Na+ at 0 . 03 mM/sec , to produce osmotically the same increase as the electrically neutral osmolyte in the previous set of simulations . The results of this buildup were a bit surprising ( Fig 10C and 10D ) . Besides the inevitable increase of the cell volume , there were no other changes of significance . Addition of 12 mM of An- with 18 mM of Na+ was largely compensated by the cell swelling , so [An-]i and [Na+]i increased only by 0 . 54 mM ( 0 . 5% of initial ) and by 0 . 3 mM ( 1 . 6% of initial ) , respectively . [K+]i decreased as expected , but only by 0 . 13 mM ( less than 0 . 1% of initial ) . [Cl-]i experienced the largest relative changes ( 2 . 5% ) , which still was less than 1 mM . The small changes in ion concentrations produced small changes in Em ( -0 . 46 mV ) . And most importantly , all concentrations , including [An-]i and [Na+]i , returned to their initial values in a few minutes after the end of the buildup . Together with ions , Em also returned to its initial value . Thus , the lone result of addition of AnNax was increase of the cell volume . Some extra energy was spent during swelling , but when ionic gradients were restored , exactly the same amount of ATP could support the resting state at a larger cell volume .
The first part of this statement brings no news , but the second does . Long ago the existence of a pump that actively extruded Na+ against its concentration gradient was postulated to explain cell volume [42 , 43] and until now the key role of the Na+/K+-ATPase in volume regulation was not questioned [19 , 23–25] ( for review see [26] ) . Our simulation , however , demonstrated that in a system when only cations were concerned ( and it is obvious that Na+/K+-ATPase deals only with cations ) the pump is responsible for the voltage , but not for the volume . Any changes in the principal cation triumvirate - Na+ conductance , K+ conductance , Na+/K+-pump activity - always and inevitably lead to changes in Em ( Figs 4 and 5 ) , even in theoretical conditions specifically designed to make an equal exchange of Na+ for K+ ( electrically neutral 3Na+/3K+-pump , gNa = gK ) . But , as was well known , the cell voltage is much more sensitive to transmembrane movement of ions than the cell volume . The same amount of ions that is sufficient to charge the membrane capacitance and create a considerable change in Em is negligibly small for cell volume and is associated with practically undetectable volume changes ( Fig 5F ) . As a result , the slightest imbalance in total cation transfer across the membrane , which is irrelevant for the cell volume , can be important for Em and will stimulate strong negative feedback to prevent further imbalance . In this respect Em ensures osmotically balanced changes in Na+ and K+ , and this is true in all conditions with any combination of the stoichiometry of the Na+/K+-ATPase , its activity , and cation conductances , including the case when the pump only removes Na+ from the cell ( Fig 5C and 5D ) . Thus , the cations are not directly involved in cell volume regulation . It would be reasonable to say that Na+ and K+ are not for volume , but for voltage . Importantly , they have to pay for this privilege with ATP . The ability of our program to show the definite ATP cost of the cation nonequilibrium appears to be useful for better understanding of relationships between ions , voltage and volume . For instance , the electrogenic pump needs more energy than our hypothetical electroneutral one to create the same concentration gradients . But the reason for the increased energy requirement is the different quantity of transferred ions per one cycle of the pump , not electrogenicity as is intuitively expected . To equilibrate [Na+]i and [K+]i ( both equal to 75 mM , under the condition where gNa = gK ) , a 3Na+/2K+-pump which transfers 5 ions per cycle needs to spend 20% more ATP ( 179 . 8 million/sec ) than a 3Na+/3K+-pump ( 149 . 8 million/sec ) which transfer 6 ions per cycle . In both cases an equal number of cations is transferred per second ( 899 million ) by the ATPase , and because these are resting states the same amount of ions passively leak back ( Na+ into the cell , and K+ out of it ) . Of course , Em is more negative with an electrogenic pump than with an electroneutral one ( -36 . 34 vs . -27 . 35 mV ) , but since a very small amount of ions produces this voltage shift , it is practically not reflected in the energy expenditure . The stoichiometry of the pump has a great influence on Em , and simulations , which are not shown , revealed that pumps that all transferred the same amount of charge per ATP , with Na+:K+ ratios of 6:0 , 4:2 , 3:3 , 2:4 and 0:6 will generate -72 . 30 , -42 . 34 , -27 . 35 , -12 . 37 , and +17 . 60 mV of Em , respectively , in order to achieve [Na+]i = [K+]i , but they all spend the same amount of energy with a precision of less than 0 . 002% . Thus , the energy is spent for cation electro-chemical gradients , which of course influence Em , but not for voltage itself via electrogenicity . Half the energy would be enough to reach this resting stage if the cell had half the cation conductance ( S3 Fig ) . Interestingly , a cell can save a lot of energy supporting the cation electrochemical gradients and negative Em if its membrane is preferentially permeable to K+ ( Fig 5C ) . The same electrogenic 3Na+/2K+-pump in the cell with the same total cation conductance needs 3 . 2 times less energy to equilibrate [Na+]i and [K+]i if gK/gNa = 9 compared to gK/gNa = 1 ( 55 . 8 million/sec instead of 179 . 8 million/sec ) . This is because the dominance of gK over gNa results in a smaller leak of the cations in spite of a much more negative Em ( -66 . 11 vs . -36 . 34 mV ) . It should be remembered that the importance of creating the transmembrane cation electro-chemical gradients by the Na+/K+-ATPase goes far beyond of generation of Em . These gradients ( particularly the strong Na+ gradient ) are heavily utilized by a cell for transport of metabolites , supporting Ca2+ homeostasis , controlling pH , and clearing neurotransmitters from the extracellular space , among other functions . These gradients can be used to cause non-equilibrium transmembrane distribution of Cl- , with all the following consequences . Changes in Em associated with changes in activity of the Na+/K+-ATPase ( as well as changes in Na+ and K+ conductances ) are also a prerequisite for possible cell volume changes . But in an “only cation system” all ionic transmembrane movements are osmotically balanced to satisfy macro electroneutrality . It is the addition of a membrane permeant anion ( Cl- ) what makes possible electrically neutral and osmotically significant ionic fluxes that lead to cell volume changes . In the absence of specialized cotransporters ( mainly the cation-Cl- transporters that were modeled here and to some extent the bicarbonate-Cl- transporter , which was beyond the scope of this work ) Cl- is distributed passively across the cell membrane . This means that Cl- has to adjust its intracellular concentration to be in equilibrium with the cation-controlled Em ( Fig 6A ) . When Em changes due to changes in cation conductance or Na+/K+-ATPase activity , [Cl-]i is forced to change also in order to fit the new Em . How fast these changes in [Cl-]i occur depends on the magnitude of the Cl- conductance ( Fig 7 ) . If gCl is low , [Cl-]i changes will develop slowly and will not be noticeable in a short time . But if gCl is high , [Cl-]i changes will be comparable to those of [Na+]i and [K+]i . Since Cl- “shares the room” with impermeant intracellular anions ( An- ) , all changes in [Cl-]i must be associated with opposite sign changes in [An-]i which is only possible if the cell volume is changed . Thus , alterations of Em induced by changes in cation ( mostly Na+ ) conductance during normal neuronal activity will inevitably be accompanied by volume changes if gCl is substantial , or will have no visible volume effects , if gCl is low . The presence of cation-Cl- cotransporters complicates the matter . NKCC elevates [Cl-]i above equilibrium and KCC lowers it below equilibrium . A nonequilibrium distribution of Cl- enables this anion to contribute to Em ( Fig 8 ) . Regulation of Na+ conductance is still by far the more common ( and more effective ) way to manipulate Em of neurons , because Na+ is much further from equilibrium than Cl- , but the contribution of unequally distributed Cl- to Em should not be ignored ( for review see [37] ) . For instance , gCl in combination with a nonequilibrium distribution of Cl- plays an important role in such complex neuronal process as direction selectivity in the retina [29] . Still , the cation-Cl- cotransporters do not disrupt the link between Cl- and the cell volume . The cotransporters determine not the absolute value of [Cl-]i , but the Cl- electro-chemical driving force , i . e . the difference from the concentration that would be equilibrium at current Em . When Em changes due to changes in gNa , gK , or the pump activities , [Cl-]i is forced to adjust accordingly , leading to cell volume changes , just as in the case with no cation-Cl- cotransporters . To conclude this part , Cl- may or may not influence Em , depending on its transmembrane distribution ( nonequilibrium or equilibrium ) . But the presence of substantial gCl is absolutely necessary for activity-dependent cell volume changes . The importance of Cl- in cell volume regulation was discussed theoretically and demonstrated experimentally in the literature ( recently [46 , 47] ) . What is stressed in this paper is the fact that gCl , not Na+/K+-ATPase , is responsible for volume changes . The apparent dependence of cell volume on the activity of the Na+/K+- ATPase can be explained by the following chain of events: changes in the activity of Na+/K+-ATPase lead to osmotically balanced ( and therefore volume-irrelevant ) changes in [Na+]i and [K+]i , that in turn affect Em . With the presence of significant gCl , the changes in Em force [Cl-]i to adjust accordingly . The transmembrane flux of Cl- is electrically neutralized by a co-directed flux of cations and the resulting transfer of NaCl and KCl is osmotically noteworthy , leading to changes of the cell volume . So , swelling of the cell following suppression of the Na+/K+-ATPase could be avoided if it would be possible to completely block gCl . Cl- is “sharing room” with impermeant anion , An- , and the sum of [Cl-]i and [An-]i must be constant if extracellular concentrations remain unchanged . This undisputable fact has led to the suggestion that changes in [An-]i should induce compensatory changes in [Cl-]i and , consequently , that [An-]i can be the key factor to determine [Cl-]i , making possible a nonequilibrium distribution of Cl- [48] . This work was criticized from both theoretical [49] and experimental [50] points of view . And actually it also was shown computationally ten years earlier that a slow leak of An- out of the cell diminishes the cell volume , but eventually does not change [Cl-]i nor does it change [Na+]i , [K+]i , Em , and [An-]i itself [28] . The same results were obtained during influx of An- [25] as well as a buildup of An- in this work ( Fig 10C and 10D ) ; only cell volume in those two cases increased because of An- addition . Also , when [Cl-]i was changed due to , for instance , the activity of cation-Cl- cotransporters , the problem of keeping the sum of [Cl-]i and [An-]i constant was resolved by appropriate adjustment of “the room , ” i . e . the cell volume ( Fig 8 ) . However , when the mean valence of impermeant anions was altered , not only [Cl-]i , but also cation concentrations and Em are changed [25] . [Na+]i , [K+]i , [An-]i , [Cl-]i and Em were also changed in our simulation of intracellular buildup of the neutral osmolyte ( Fig 10A and 10B ) , which could be viewed as analogous to a decrease of mean valance of the impermeant anion ( if the impermeant anion were defined as everything inside the cell except Na+ , K+ , and Cl- ) . In this work neutral osmolyte is treated separately from other internal impermeants , and they all are considered to be parts of a broader concept—the osmolarity-charge asymmetry . The osmolarity-charge asymmetry inevitably arises when the internal impermeant anion ( An- ) has a mean valence ( z ) different from -1 . The equation for internal macro electroneutrality is: [K+]i+[Na+]i=‑z*[An‑]i+[Cl‑]i , ( 37 ) The complication here is that An- is not a certain anion or even a set of anions of the same kind , but rather a collection of very different small and large molecules . In a cell with a membrane that is permeable only to Na+ , K+ and Cl- , An- can be defined as “everything which is internal , charged and impermeable” and it definitely must be an “anion” to compensate for the deficiency of negative charge of the main inorganic ions inside the cell . In this case the mean valence of An- ( z ) is the ratio of all electrical charges that belong to An- to the concentration of An- . Osmotically active proteins that carry multiple negative charges per molecule support the case for z < -1 ( i . e . larger negative charge ) , but they are responsible for less than 10% of total cytoplasmic osmolarity [51] . Immobile proteins incorporated in the cell membranes , which can represent half of all proteins [52] provide some more negative charge without any osmotic contribution . However , the majority of [An-] is made of small organic molecules [53] that are mostly monovalent under physiological pH , such as creatine phosphate ( about 40 mM in frog muscle [54] ) and other phosphates and sulfates . Also , a significant part of the internal osmolytes is comprised of amino acids ( up to 37 mM in rat brain , [55] ) , but among them only glutamate and aspartate are negatively charged , and some others are positively charged . Taking into account the wide diversity of components from which An- is comprised , it is not surprising that the mean valence of An- could be very different from cell to cell . In the case of myocytes a reasonable value of z is -1 . 65 [28] , while for lymphoid cells it can be as large as -2 . 5 [23 , 56] . Dusterwald and coworkers have used z = -0 . 85 [25] . Whatever z is , as long as z ≠ -1 , it creates an asymmetrical osmolarity-charge setting , when the quantity of internal anions is not equal to the quantity of internal cations in an osmotic sense . The presence of internal neutral osmolytes , such as the just mentioned neutral amino acids , contributes to the asymmetry . A special place among them is occupied by sulfonic amino acid taurine , a zwitterion which is neutral at physiological pH . The concentration of taurine can be as high as 60 mM or even more , but varies significantly from species to species and from cell type to cell type , with higher concentration in mammals than in amphibians or reptiles , in retina than in brain or muscle , and in photoreceptors than in other retinal cells [57–59] . Addition of an external neutral osmolyte ( although in the case of neural systems it is only a few mM , mostly from glucose ) completes the equation of osmotic balance: [K+]o+[Na+]o+[Cl‑]o+[osm]o=[K+]i+[Na+]i+[An‑]i+[Cl‑]i+[osm]i ( 38 ) where [osm] is the concentration of uncharged osmolytes . To quantify the extent of the osmolarity-charge asymmetry a new parameter - the coefficient of asymmetry ( ka ) - is introduced as follows: ka= ( ‑z*[An‑]i+[Cl‑]i ) / ( [An‑]i+[Cl‑]i+[osm]i–[osm]o ) ( 39 ) Because charge of monovalent Na+ and K+ is equal to their osmolarity , both intracellular and extracellular , the charge-osmolarity imbalance , quantified by ka , results from anions and uncharged species . Thus , ka is the ratio of all intracellular negative electrical charges to all osmotically active intracellular molecules except cations . The equation also includes external osmolyte , because addition of [osm]o is the same for net osmolarity as subtraction of an equal amount of [osm]i . It is convenient to replace [osm]i−[osm]o with d[osm] , which is the difference in concentrations of internal and external neutral osmolytes and can be positive or negative . Accordingly , the equation for internal macro electroneutrality ( Eq 37 ) can be rewritten in terms of ka: [K+]i+[Na+]i=ka* ( [An‑]i+[Cl‑]i+d[osm] ) ( 40 ) Since [K+]o + [Na+]o = [Cl-]o , Eq 38 for osmotic balance can be rewritten with regard to cations as: 2*[cat]o=[cat]i+[cat]i/ka ( 41 ) where [cat]i and [cat]o are total intracellular and extracellular concentrations of the cations . From here one can derive the equation of osmolarity-charge asymmetry that describes the uneven , yet equilibrium cation distribution: [cat]i/[cat]o=2*ka/ ( ka+1 ) ; ( 42 ) Accordingly , Em in this equilibrium state with no Na+/K+-ATPase activity will be determine by ka: Em= ( RT/F ) *ln ( 2*ka/ ( ka+1 ) ) ( 43 ) In this condition , both ENa and EK must be equal to Em , and both cations separately follow Eq 42: [Na+]i/[Na+]o=2*ka/ ( ka+1 ) ; ( 44 ) [K+]i/[K+]o=2*ka/ ( ka+1 ) ; ( 45 ) This is true equilibrium when the concentration driving forces for both Na+ and K+ are countered by the electrical driving force , and osmotic balance also holds . In this respect it is similar to Double Donnan equilibrium , but with one important difference - the equilibrium based on the osmolarity-charge asymmetry is possible only if the cell membrane is permeable to cations , but not to Cl- . Opening of gCl will lead to inevitable and unlimited swelling . Of course , the system can be stabilized if the cation gradients are supported by the Na+/K+-ATPase . Thus , there are two factors of different nature that determine cation concentrations: one active , dependent on the cation conductances and the pump activity , and the other passive , dependent on ka . Accordingly , there are two parts of cation-dependent Em . The active part of Em is much larger than the passive ( Fig 5E ) , and also the active part can be quickly , in a small fraction of a second , changed by manipulating cation conductances . There is no doubt that this active component , which is determined by gNa , gK and the activity of the Na+/K+-ATPase , dominates Em . On other hand , although the passive osmolarity-charge asymmetry dependent component is probably present in Em of every cell ( since it is very unlikely that ka = 1 in any cell ) , its contribution will be insignificant in most conditions . Changes in ka , are probably common since they can be a result of changes in z , [An-]i , [osm]i , or [osm]o , but in most cases with small effect on Em . As illustrated in Fig 11B , a decrease of ka from 2 to 1 . 5 ( which was achieved by an increase of d[osm] by 32 mM with constant z ) depolarized Em by only 2 . 8 mV , since it was associated with relatively insignificant changes in [K+]i and [Na+]i ( from 6 . 9 to 6 . 0 mM and from 193 . 3 to174 mM , respectively , Fig 11A ) . Also accompanying this was a modest 8 . 3% increase in cell volume ( normalized to the volume in osmolarity-charge symmetric conditions when ka = 1 ) . The small changes in Em ( about 2 mV ) demonstrated in simulations of osmotic disturbances ( Fig 9A and 9B , Fig 10A and 10B ) are a consequence of changes in osmolarity-charge asymmetry , when the cell was slightly hyperpolarized due to an increase of ka resulting from the buildup of external osmolyte and slightly depolarized due to a decrease of ka resulting from the buildup of internal osmolyte . But when Em was changed , [Cl-]i had to change too , assuming the presence of substantial gCl . In this respect Glykys and coauthors [48] were right in claiming that impermeant anion An- can influence [Cl-]i , although it happened not because of a direct link , but due to a chain of events including changes in osmolarity-charge asymmetry , cation concentrations and Em . However , if Cl- was in equilibrium with Em , it will continue to be in equilibrium . Importantly , osmolarity-charge asymmetry is also affected by changes in [Cl-]i . if the distribution of ions was already asymmetric . As was mentioned in the results concerning Fig 8E , the elevation of [Cl-]i by NKCC in “realistic” ( i . e . osmolarity-charge asymmetric ) conditions led to a decrease of the total internal anion charge because An- with z = -1 . 5 was replaced by the monovalent Cl- . The inevitable result of that were changes of cation concentrations and Em . Effects of cation-Cl- cotransporter-induced changes in [Cl-]i on [Na+]i , [K+]i , and Em can be revealed by comparing “realistic” conditions with “simplified” ( i . e . osmolarity-charge symmetric ) conditions , especially when gCl was low ( 108 ions/ ( sec*V ) ) and Cl- practically had no direct contribution to Em . When Cl- was pumped by cation-Cl- cotransporters in symmetric “simplified” conditions , it replaced ( or was replaced by ) an equal quantity of monovalent An- . As a result , ka continues to be 1 , [Na+]i and [K+]i remain almost the same , and slight changes of Em did not exceed 0 . 3 mV ( Fig 8A for NKCC and Fig 8C for KCC ) . But in asymmetric “realistic” conditions an increase of [Cl-]i by NKCC resulted in a reduction of ka from 1 . 47 to 1 . 28 , a decrease of [Na+]i and [K+]i and a depolarization by 1 . 42 mV ( see S4A Fig ) ; a decrease of [Cl-]i by KCC led to smaller changes in ka ( from 1 . 47 to 1 . 56 ) and a subsequent increase in both cation concentrations and hyperpolarization by -0 . 78 mV ( S4B Fig ) . Finally , it should be noted that this asymmetry-dependent voltage is completely determined by ka and is independent of the internal ionic and osmotic compositions as long as they result in the same ka . The data for Fig 11 were obtained during manipulation of the neutral osmolytes ( see explanation in the figure legend ) , but the stars with numbers were from our previous simulations with different internal compositions ( star 1: Fig 4E , star 3: Fig 4F , and star 2: Fig 5B , “realistic” conditions; all for the points at the left of the graphs where there is no Na/K pumping ) . The results of computational simulations were exactly the same as predictions from Eqs 43 , 44 , and 45 . To summarize , [An-]i and its mean valence play an important role in determination of cell volume . It was shown earlier , and it was confirmed here . [An-]i and its mean valence also , together with other factors ( [Cl-]i , internal and external neutral osmolytes ) , contribute to creation of osmolarity-charge asymmetry , which passively influence cation distribution and Em , although the effect is small compared to the active Na+/K+-ATPase dependent cation voltage . | We have developed software that calculates membrane potential and cell volume that result from redistribution of principal ions ( K+ , Na+ , and Cl- ) during normal cellular activity and experimental manipulations . Calculations in the model are done by an iterative charge-difference method that makes few assumptions about governing equations . Most of the features that were considered to be important for volume and voltage regulation were incorporated in the model , including the unique capability to perform calculations with different values of transmembrane water permeability . We have used the program to reexamine interactions between ionic fluxes , membrane potential , and cell volume and found that there was a previously unappreciated difference in the way that the distribution of cations and anions affect the cell . Na+ and K+ , which are distributed unevenly across the membrane by the Na+/K+-ATPase , are primarily responsible for the membrane potential , but , contrary to popular belief , do not directly participate in volume regulation . On the other hand , the Cl- conductance determines the extent of volume changes , because Cl- has to follow the changes of membrane potential , which inevitably leads to changes in cell volume . The software is available to download and use for other investigations . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"and",
"health",
"sciences",
"membrane",
"potential",
"ions",
"anions",
"permeability",
"electrophysiology",
"materials",
"science",
"cations",
"cellular",
"structures",
"and",
"organelles",
"physical",
"chemistry",
"chemistry",
"cell",
"membranes",
"hyperpolarization",
"intracellular",
"membranes",
"cell",
"biology",
"physiology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"material",
"properties"
]
| 2019 | The logic of ionic homeostasis: Cations are for voltage, but not for volume |
The viral population of HIV-1 , like many pathogens that cause systemic infection , is structured and differentiated within the body . The dynamics of cellular immune trafficking through the blood and within compartments of the body has also received wide attention . Despite these advances , mathematical models , which are widely used to interpret and predict viral and immune dynamics in infection , typically treat the infected host as a well-mixed homogeneous environment . Here , we present mathematical , analytical , and computational results that demonstrate that consideration of the spatial structure of the viral population within the host radically alters predictions of previous models . We study the dynamics of virus replication and cytotoxic T lymphocytes ( CTLs ) within a metapopulation of spatially segregated patches , representing T cell areas connected by circulating blood and lymph . The dynamics of the system depend critically on the interaction between CTLs and infected cells at the within-patch level . We show that for a wide range of parameters , the system admits an unexpected outcome called the shifting-mosaic steady state . In this state , the whole body’s viral population is stable over time , but the equilibrium results from an underlying , highly dynamic process of local infection and clearance within T-cell centers . Notably , and in contrast to previous models , this new model can explain the large differences in set-point viral load ( SPVL ) observed between patients and their distribution , as well as the relatively low proportion of cells infected at any one time , and alters the predicted determinants of viral load variation .
In 1979 , Bormann and Likens introduced the concept of the shifting-mosaic steady state ( SMSS ) to describe biomass in forested ecosystems . This concept was based on the intuition that although the patches comprising the forested ecosystem might each be in different phases of ecological succession due to past disturbance events , the biomass of the whole forest will be at an equilibrium [1 , 2] . We suggest that for pathogens that cause systemic infection , such as HIV or hepatitis C virus , the viral population , host cells , and the immune system form a complex ecosystem within the host , with localized succession dynamics . We focus on HIV , characterized by rapid dynamics and trafficking between localized sites of replication in the body . The hypothesis that HIV is at SMSS in some individuals explains why viral loads vary so dramatically among patients , why only a small proportion of patients are natural controllers , and why a relatively low proportion of cells are infected during chronic infection . Set-point viral load ( SPVL ) is the approximately constant viral load observed during early chronic asymptomatic infection . It varies by four orders of magnitude between patients [3] and is the most commonly used and robust predictor of the severity of infection [4 , 5] . Factors that have been implicated in determining SPVL include how rapidly the virus replicates and infects new cells [6–8] , the efficacy of the cytotoxic T lymphocyte ( CTL ) immune response [9] , and the activation rate of susceptible cells [10 , 11] , all of which , in vivo , are probably influenced by a combination of viral and host factors [12 , 13] . However , using standard models of HIV within-host dynamics , in which the virus , susceptible and infected cells , and CTLs are assumed to be well mixed , these factors only mildly affect the SPVL unless the virus is close to extinction [14–17] . Introducing more complicated functions to describe the rate at which CTLs accumulate in response to the number of infected cells can help to explain more of the variation in SPVL [18–22] , as can small differences in a large number of parameters [20] . Even with these models , though , it is still hard to explain orders of magnitude differences in SPVLs without fine-tuning parameters , and especially to reproduce the left tail of the distribution composed of patients with low viral loads . As a further refinement to these models , incorporating activation of cells from a viral reservoir can explain very low viral loads ( below the level of detection by conventional assays ) for parameters in which otherwise the virus is expected to go extinct , e . g . , [22 , 23] , although not the large number of patients with low viral loads but above the level of detection . HIV replication is focused in the secondary lymphoid organs , such as the lymph nodes ( LNs ) , the spleen , and the gut [24] . At an even finer scale , replication is likely centered within the T cell areas of these organs , such as the Malpighian bodies found in the spleen [25] and Peyer’s patches in the gut [24] . Genetic analysis has revealed that the genetic composition of the viral population differs among these sites of replication , even at very fine scales [25 , 26] , and that this differentiation might well be transient [27] . This has led to the speculation that the within-host structure of HIV is best thought of as a metapopulation , in which T cell areas can be considered patches of HIV replication , and with long-range migration of the virus among these patches via the blood [28–30] . This differentiation is formally supported by calculations of FST ( a measure of the genetic differentiation among groups or patches ) from viruses in the spleen , which reveals high levels of differentiation in most patients and is suggestive of a within-host metapopulation structure [30] . Here , we develop a metapopulation model of within-host HIV dynamics during chronic infection in which within-patch dynamics are explicitly incorporated , and which we investigate in the form of population-based stochastic simulations . To gain further analytical insight , we also derive a mathematically tractable analytical approximation in which within-patch dynamics are nested within a model of between-patch dynamics , using a time-since-infection framework to link the two scales [31 , 32] . We show that for a wide range of parameters , the system as a whole reaches a steady state even though the patches themselves are not necessarily at equilibrium . The dynamics of the system at the whole patient ( between-patch ) level depend critically on the interaction between CTLs and infected cells at the within-patch level . If the CTL response is able to clear infected cells from patches before a within-patch steady state is established , the metapopulation can reach a dynamic SMSS , at which the whole body’s viral population is stable over time but the overall equilibrium results from an underlying , highly dynamic process of local infection and clearance within T-cell centers . If the CTL response is not able to clear infected cells from patches , the system will reach a full equilibrium ( FE ) , at which all of the patches , as well as the system as a whole , are at equilibrium . In contrast to the corresponding single-patch model , the metapopulation model can explain the distribution of SPVLs observed among patients due to modest differences in viral infectivity and the strength of the CTL response without model fine-tuning .
To investigate the behavior of the model , we first developed a simulation based on a stochastic counterpart of the set of ordinary differential equations that describe the metapopulation model ( Eq 1 . 1–1 . 5 ) . To accommodate heterogeneity among patches , for each simulation , we sampled viral infectivity for each patch from a uniform distribution with mean β¯ in the range 0 . 5β¯–1 . 5β¯ , thus implicitly assuming some patches are more densely packed with T cells than others ( see Methods , S1 Simulation Code , and S1 Data ) . Depending on β¯ , and the maximum strength of the immune response , k , the dynamics of the metapopulation typically converges to one of three stationary states: a disease-free trivial stationary state , a “full equilibrium” ( FE ) at which the patches and the metapopulation as a whole are at a stationary state ( Fig 2A ) , and an SMSS in which the number of infected cells and HIV-specific CTLs in some , or all , of the patches fluctuates , but the total number of infected cells is relatively constant ( Fig 2B ) . To gain a sense of the dynamics of the model when at SMSS , we have provided an animation ( S1 Animation ) . Although analytical expressions for the disease-free equilibrium and FE of the metapopulation model can be found directly from the differential equations , this is not possible for the SMSS . We derived a deterministic , mathematically tractable approximation to the full metapopulation model , assuming all patches are identical , using a time-since-infection framework ( see S1 Text ) . Which of the three metapopulation steady states is established , disease-free , FE , or SMSS can be understood by considering the within-patch dynamics ( Fig 3 ) . When an infected cell entering a patch fails to establish a local burst of infection , the metapopulation is at the disease-free equilibrium . When an infected cell establishes a non-zero endemic equilibrium in a patch , the metapopulation is at FE . When an infected cell establishes a local burst of infection , which is then cleared by the CTL response , the metapopulation will be at SMSS if these local bursts can propagate from patch to patch and sustain the infection . In mathematical terms , the “patch-to-patch” reproduction number ( RP ) , the number of patches that are colonized as a result of a typical single burst in an otherwise fully susceptible host , must be greater than one ( otherwise disease-free equilibrium occurs ) . This is analogous to the concept of R0 in epidemiology , where an epidemic will not be sustained if the number of people that a typical person infects , in a totally susceptible population , is less than one [47] . Which of these equilibria occurs depends critically on the infectivity of the virus , β , and the maximum strength of the CTL response , k ( see S1 Text for details ) . As an additional observation , the proportion of susceptible cells that are infected within a patch is low for most of the parameter values we tested ( often less than 5% , Fig 3A ) , in agreement with what is observed at sites of HIV replication during chronic infection [48] . This is in contrast to most well-mixed models , in which almost all susceptible cells are expected to be infected ( although see [19] for an exception ) . This disparity occurs because of the high rates at which CD4+ T cells traffic through the patch ( with a mean residence time of about 10 hours ) , and because there is insufficient time to infect a large number of susceptible cells when within-patch bursts of infection are short lived . During acute infection , a much larger proportion of CD4+ T cells tend to be infected than in chronic infection ( up to 60% of memory CD4+ T cells [49 , 50] ) , probably because the full force of the host immune system is yet to kick in [49] . In agreement with this observation , if we set the immune response in our model to be very weak ( k = 1 per day; see Fig 3B ) , the CTL response fails to control the viral infection and about half of the activated cells in a patch will be infected ( unless the infectivity of the virus is extremely low ) . Because in this low immunity scenario the viral population is target-cell limited , once the within-patch dynamics have reached equilibrium we see little variation in the proportion of cells that are infected for large differences in viral infectivity ( β ) . It is also interesting to note , from Fig 3 , that although the maximum strength of the CTL response in a patch , k , might be relatively high , the actual rate at which CTLs kill infected cells ( measured as k * z ( τ ) /zmax , where z ( τ ) is the number of HIV-specific CTLs in the patch at time τ since the patch was infected , and zmax is the maximum possible number of HIV-specific CTLs the patch can hold ) is often much lower . This might , in part , explain the large discrepancy between different estimates of the rate at which CTLs kill infected cells ( see Methods ) . We next assessed the impact that a metapopulation structure has on the number of infected cells within an infected individual during chronic infection , which is assumed to be the stationary state number of infected cells in our models . Specifically , we compared the metapopulation simulation model with the analytical approximation of this model and the equivalent single-patch differential equation model in which lymphocytes are assumed to traffic between one well-mixed patch and the blood ( see Methods ) . We first consider the total number of infected cells for different values of viral infectivity , β , and the maximum strength of the CTL response , k , in the absence of a reservoir of latently-infected resting CD4+ T cells ( Fig 4 , columns 1 and 2 ) . Significantly , for a large range of parameters , spatial structure enables the viral population to persist at an SMSS in the metapopulation model , whereas extinction is predicted in the single-patch model for the same parameters ( Fig 4 ) . Moreover , when the metapopulation is at SMSS , a broad range in the number of infected cells is predicted for relatively small changes in β and k . In contrast , the single-patch model exhibits strong threshold-like behavior , where the number of infected cells is typically either zero ( low β , high k ) or very high ( high β , low k ) , with intermediate viral loads only possible when the parameters are close to the virus “extinction threshold” [15] . Note that we have presented results for both low and high effective migration rates ( Me ) among patches . The low Me scenario is used to check the analytical approximation , because it most closely reflects the assumption of no patch super-infection required for the analytical approximation ( achieved by increasing the death rate in the blood from a realistic δB = 1 per day to δB = 432 per day; see Methods ) . The high Me scenario reflects estimates of lymphocyte trafficking made from experimental data ( see Methods and Table 1 ) . Wide variations in the number of infected cells occur at SMSS because colonized patches are , for most of the time , in a near-exponentially growing phase ( Fig 2 ) . Thus , modest differences in viral infectivity and the strength of the CTL response translate into very large differences in the total number of cells infected during a burst of infection . In other words , when the system is at FE , the virus is close to carrying capacity within each of the patches , and is therefore sensitive to parameter values that affect this carrying capacity ( e . g . , the rate at which susceptible cells enter patches ) . However , at SMSS , the system becomes much more sensitive to parameter values that affect the rate at which the number of infected cells and , by proxy , the viral population grows ( e . g . , β and k ) . If a reservoir is present ( Fig 4 , columns 3 and 4 ) , the same qualitative arguments apply , although the virus does not go extinct ( at least in the short term ) because it is maintained at very low levels by the reactivation of cells from the reservoir . In the parameter range in which local bursts of infection are possible but RP is less than one , more infected cells are maintained in the metapopulation compared to the single-patch model , in which localized bursts of infection are not expected . This has similarities to the verbal argument proposed by Grossman et al . [28] , who suggested that a reservoir is needed for HIV to persist within hosts at low SPVLs . The wide variations in the number of infected cells observed in the metapopulation stationary state are robust to the key unknowns in the model , specifically the number of patches ( S1 and S2 Figs ) and the effective migration rate ( Me; S3 and S4 Figs ) , although , as expected , the threshold-like behavior of the model increases if Me is substantially increased or if the number of patches decreases . Due to uncertainties in how HIV-specific CTLs accumulate in patches , we also analysed an alternative to the “CTL immigration” metapopulation model described here , which we call the “CTL proliferation” metapopulation model . In this alternative version of the model , CTLs accumulate within patches due to local proliferation rather than the immigration of CTLs from outside of the patch ( see S1 Text and S5–S9 Figs ) . Although this also has the effect of increasing the threshold-like behavior of the model , a broad range in viral loads is still observed , making our conclusions robust to this fundamental change in the modeling assumptions . It is noticeable from the simulation dynamics ( Fig 2B ) that at SMSS the number of infected cells at the population level can oscillate , suggesting the dynamics among patches are synchronized . However , the level of synchrony among patches is generally quite modest , as is the amplitude of the oscillations ( S1–S4 , S6–S9 Figs; see S1 Text for further discussion ) . Although with increasing patch heterogeneity these oscillations are likely to be eroded , it is tempting to speculate that differences in viral load measurements taken during untreated chronic infection for individual patients might partly reflect these dynamics . We have presented a model of the within-host metapopulation structure of HIV during untreated chronic infection . Conceptual models such as this should aid understanding , but also make qualitative and/or quantitative testable predictions that enable discrimination from other plausible models; in this case , the single patch model . Here , we outline four testable predictions of the metapopulation model: two of them are supported by available data , while the other two require further data collection and analysis to be tested . ( i ) Distribution of set-point viral loads among individuals in a population Because this is a model of chronic infection , the most pertinent clinical data to test the model is the distribution of SPVLs observed among individuals . We fitted the metapopulation simulation and the corresponding single patch deterministic model to the distribution of patient SPVL measured amongst seroconverters in the Netherlands ( S1 Table ) . Using the quasi-steady state assumption that the number of virions is proportional to the number of infected cells [16] , and using the same reasoning as [45] , we calculated the SPVL predicted by our models as approximately yB* ( pr*VolB ) , where yB* is the mean number of infected cells in the blood at stationary state , p is the production rate of virions from infected cells ( 5x104 per day [45] ) , r is the removal rate of virions from the blood ( 23 per day [46] ) , and VolB is the volume of blood in a typical human adult ( 5 litres ) . Since data on the distribution of β and k among patients is lacking , we assumed truncated normal distributions ( see S1 Text for full details ) . For each set of parameters describing the distribution of β and k , the likelihood was computed as the product , over all SPVL values in the data , of the probability this SPVL value occurred under the model ( S2 Data ) . The distribution of the Netherlands data and the maximum likelihood distributions for the single patch and metapopulation models are shown in Fig 5 , and the bivariate marginal likelihood profiles are shown in S10 Fig . For all distributions of β and k that we examined , the likelihood of the metapopulation model is considerably higher than that of the single patch model ( S10 Fig ) . The poor fit of the single patch model is largely due to the large number of natural controllers it predicts ( which we have defined as patients with SPVLs <1 , 000 per ml ) . Among the Netherlands seroconverters , 5 . 8% of patients are natural controllers . The maximum likelihood metapopulation model predicts 3 . 3% of patients should be natural controllers , whereas the maximum likelihood single patch model predicts 27% should be natural controllers . Additionally , the bivariate marginal likelihood surfaces are much flatter for the metapopulation compared to the single patch model , and therefore the metapopulation is much more robust to uncertainties in the true distributions of β and k ( since high likelihood values are observed for a broad range of distributions ) . Although the metapopulation model is a much better fit to the data and captures well the left-tail in the distribution of SPVLs , it predicts a bimodal distribution that is not evident in the data , suggesting there are aspects of the biology that our model is not yet capturing , including heterogeneity among patches within a patient [40] , non-normal and/or non-independent distributions of β and k among patients , and the role of other host and viral factors . Interestingly , the alternative CTL proliferation version of the model has the largest maximum likelihood value of all the models considered and does not predict a high proportion of natural controllers or a strong bimodal distribution , although the bivariate marginal likelihood surfaces are less flat than for the CTL immigration model ( S10 Fig ) . Given these differences , future work should focus on which mechanism ( s ) best describes how CTLs accumulate within patches . ( ii ) Distribution of infected cells within individuals A prediction of the metapopulation model is that the number of infected cells should be unevenly distributed among potential sites of HIV replication at any one moment in time when at SMSS . The distribution of HIV in human spleens shows this pattern [25 , 51] . Additionally , observation of the spleens of simian immunodeficiency virus ( SIV ) -infected macaques [52] shows that virus with high replicative capacity is found in most , if not all , potential sites of replication ( i . e . , consistent with a FE ) , but that virus of low replicative capacity is not ( i . e . , consistent with SMSS ) . More direct evidence might come from real-time observation of individual sites of potential infection in vivo , from which we would predict the number of infected cells to increase and decrease in a succession of local epidemics intermittently controlled by the immune system , but this data is so far unavailable . ( iii ) Genetic structure of the viral population within individuals Previous studies have shown that the effective population size , Ne , of HIV is much lower than consensus population size [53 , 54] , and that Fst values ( a measure of population subdivision ) are high in the spleen , with values ranging from 0 . 08 to 0 . 6 in different patients [30] . These data provide strong evidence that HIV exhibits metapopulation dynamics that are characterised by colonisation and extinction of virus in potential sites of replication [30]; the same dynamics seen in the metapopulation model at SMSS . To formally test whether the metapopulation model can explain observed Ne and FST values , it will be necessary to extend the model to include explicit evolutionary processes , an analysis that will be a focus of future work . Another feature of HIV infections are the ladder-like time-resolved phylogenies observed during the course of infection [55 , 56] . The metapopulation model could be further extended to test whether phylogenies predicted by the model are different from those predicted from a single patch model , and whether these more closely resemble observed phylogenies . ( iv ) The dynamics of chronic infection A prediction of the metapopulation model is that we might expect to see oscillations in viral load over short time-scales in patients with intermediate SPVLs . Testing this prediction would require obtaining viral load data from very frequently sampled individuals during chronic infection ( every 1 or 2 weeks ) and testing statistically for periodic oscillations . Unfortunately , suitable data is not yet publicly available . If these oscillations were found , it would be strong evidence for an SMSS type process , but we note that the absence of oscillations would not be strong evidence against SMSS; the within-host environment is likely to be much more heterogeneous than we have modelled , which would dampen out oscillations , and stochastic effects and errors in viral load estimates would mask any underlying oscillations . Note that our model does not generate testable predictions regarding early infection dynamics , as it does not include biological features that are likely prominent in that phase , such as HIV-specific CTL accumulation , CD4+ T cell loss at the systemic level , and evolution of CTL escape . Similarly , our model does not generate testable predictions on the impact of antiretroviral therapy ( ART ) , since both in the metapopulation and single patch models the viral load decline is rapid and driven by the death of infected cells , which is unaffected by patch structure , and by CTL killing ( simulation not shown ) . Furthermore , for both models , the rate of viral load decline is much faster than observed in patients unless very low rates of CTL killing ( k << 1 per day ) are assumed , although it has been shown that adding an eclipse phase to single patch models can reconcile high values of k with the data [57] .
We have developed a within-host metapopulation model of chronic HIV infection in which viral replication is focused within T cell centers , derived a mathematically tractable approximation , and ran stochastic simulations . For a broad range of parameters , the virus can be maintained at equilibrium even though the patches themselves are not at equilibrium . This quasi-equilibrium represents an SMSS , a concept first described 35 years ago by Bormann and Likens [1] to describe the biomass in forested ecosystems . Importantly , this model can easily account for the broad distribution of SPVLs observed among patients and the low proportion of susceptible cells that are infected during chronic infection due to differences in the replicative capacity of the virus ( because of either host or viral factors ) and the strength of the CTL response . Moreover , the model can explain why only a small number of patients ( about 5% ) go on to control the virus in the absence of therapy ( <1000 virions per ml ) after the initial burst of viremia during acute infection . As well as explaining the wide range of viral loads observed among patients during chronic infection , our model also gives insight into the marked changes in viral load seen between acute , chronic , and late-stage infection . Around the peak of acute infection , when host immune responses against the virus are still weak [49] , or during late stage infection , when the immune system has essentially collapsed , we suggest that within-patient dynamics are probably close to FE . The small impact of viral replication rate on FE viral load might explain why peak viremia tends to be similar among patients and why a large proportion of CD4+ T cells become infected with virus [49 , 50] . In contrast , during chronic infection , immune responses against the virus are much more effective; thus , we propose a large proportion of patients will be at an SMSS , with small differences in the strength of the host immune response and the infectivity of the virus explaining the broad ranges in viral loads . It is important to recognize that the metapopulation model we have presented here has been used to describe the chronic phase of HIV infection , during which , unlike during acute and late stage infection , the system can be assumed to be at a steady state . To fully test how a metapopulation structure might affect the predicted dynamics of the model over the entire course of the infection , and to fit it against early and late infection clinical data , would require further processes , such as HIV-specific CTL accumulation , CD4+ T cell loss at the systemic level , or evolution of CTL escape , to be modeled mechanistically or , at least , as external driving factors carefully parameterized from data . In both cases , incomplete biological knowledge would likely require strong assumptions to be made in parameter values and model structure . In terms of caveats , we should also recognize that the proliferation and trafficking of lymphocytes around the body via the blood , lymph , secondary lymphoid tissues , and nonlymphoid tissues is complex and incompletely understood [36 , 58] . This matters because the dynamics of multitrophic systems can depend crucially on the precise assumptions used to formulate the model [19] , and the within-patch processes modelled here are no different . Despite these gaps in our knowledge , the observations that not all T cell zones contain infected cells [51] , that HIV is spatially and genetically structured in the body over very small spatial scales [25 , 26 , 52] , that this structure might not be stable over time [27] , that higher fitness clones tend to co-localise but lower fitness clones do not [52] , the small effective population size of HIV compared to the actual population size [53 , 54] , and high values of FST [30] together provide compelling evidence that the HIV within-host landscape is highly dynamic and characterised by local bursts of viral replication . Although we have focused here on HIV , it is likely that consideration of within-patient spatial structure will also have implications for other chronic viruses , such as Hepatitis C , in which infection of hepatocytes in the liver is spatially structured [59] , and where the number of hepatocytes infected correlates with viral load [60] . More generally , we think this model is useful in explaining immunopathogenesis , because it naturally explains large variations in viral load . It is also likely to be useful for studying viral evolution , and because it alters the predicted determinants of viral load to viral and host factors ( β and k ) , it will lead to different evolutionary predictions . In a metapopulation that is at SMSS , viral evolution is expected to proceed quite differently from viral evolution in an admixed system at equilibrium . Mathematical modeling has provided many clinically useful insights into HIV dynamics; improved models are likely to give rise to improved insights .
The full metapopulation model consists of a set of ordinary differential equations describing the number of susceptible cells xi ( t ) , infected cells , yi ( t ) , and HIV-specific CTLs zi ( t ) in each patch i at time t , as well as the number of infected cells in the blood , yB ( t ) , and the number of latently infected long-lived resting CD4+ T cells , L ( t ) , which we refer to as the reservoir . We do not model the number of free virions and instead use the steady-state assumption that the number of free virions is proportional to the number of infected cells [16] . Eq 1 . 1 describes the dynamics of susceptible cells in patch i . Although CD4+ T cell count falls during the course of chronic infection , this decline is slow , particularly compared to the short durations of 100 days that we consider in our simulations , and at a good approximation the loss of CD4+ T cells is compensated for by their production through thymus-dependent and -independent pathways [61 , 62] . In addition , CTL and CD4+ T cell counts vary by a relatively small amounts among HIV-infected patients compared to the orders of magnitude differences in viral load [63] . We therefore assume that susceptible cells enter patches at a constant rate , γiM xB , where xB is the constant total number of susceptible cells in the blood , M is the rate at which cells leave the blood and enter patches , and γi is the probability that these cells enter patch i ( ∑i γi = 1 ) . Viral infectivity in patch i is given by βi , ximax=γiMxB/ ( d+ε ) is the number of susceptible cells expected in patch i in the absence of infection , d is the death rate of susceptible cells , and ε is the rate at which lymphocytes exit patches . In deriving the transmission term , we have assumed that patches have a fixed volume ( proportional to ximax ) and used the reasoning that virions released from an infected cell are more likely to reach other cells if the target cell population is more dense . Furthermore , once cells are infected by HIV , the CD4 is down-regulated on the cell surface , thus substantially reducing the probability of cell superinfection [64] and suggesting that free virions are only mildly “wasted” on already infected cells . Therefore , we assumed the number of susceptible cells infected by a single infected cell is βixi ( t ) /ximax , which increases with the number ( and density ) of susceptible cells . The denominator ximax appears explicitly , rather than being subsumed into βi , so that the value of βi gives the maximum number of infections per day generated by a single infected cell . Eqs 1 . 2 and 1 . 3 describe the dynamics of infected cells in patch i and in the blood , respectively . Here , λ is the proportion of cells that become long-lived upon infection and thus enter the reservoir , a is the activation rate of latently infected resting CT4+ T cells , ω is the proportion of activated cells that produce infectious virions , δ and δB are the death rates of infected cells within patches and in the blood , respectively , k is the maximum rate at which infected cells can be killed by CTLs ( achieved when CTLs are at a maximum density within a patch ) , and zimax is the maximum number of CTLs that patch i can accommodate . We assume that all patches are well connected , so that an infected cell egressing from a patch is as likely to enter a patch a long distance away as it is to enter an adjacent patch . Although this is clearly a simplification of the true trafficking pattern of infected cells [40] , this is a reasonable first approximation , because significant migration of activated CD4+ T cells occurs over long ranges via the blood [36] . In addition , we also define the effective migration rate , Me , which is the rate at which infected cells leave a patch multiplied by the probability that an infected cell leaving a patch successfully enters a new patch , such that Me = ε M/ ( M + δB ) . Note that we assume that latently infected resting CD4+ T cells enter patches at random immediately upon reactivation . Although little is known about how resting CD4+ T cells circulate , their trafficking patterns are probably similar to other CD4+ T cells [36 , 65 , 66] , and because these cells are long-lived , we can assume to a good approximation that they are randomly distributed among patches . From experimental data , we know that HIV-specific CTLs accumulate at sites containing HIV-producing cells [48] , although it is not clear whether CTL accumulation is due to increased rates of entry , e . g . , [40] , or lower rates of egress , e . g . , [67–69] . In Eq 1 . 4 , we assume that the maximum rate at which HIV-specific CTLs enter patches is c zmax , with this rate gradually falling to zero as the patch reaches its maximum carrying capacity , zmax; we impose a maximum carrying capacity on the patches , because otherwise the number of CTLs would grow without bound . The indicator function 1yi ( t ) =0 takes the value 1 when yi ( t ) = 0 and 0 otherwise . Thus , HIV-specific CTLs are assumed to egress from patches at the same rate as CD4+ T cells , ε , if the patch contains no infected cells , but egress is prevented if infected cells are present [69] . We do not specifically model non-HIV-specific CTLs , because evidence suggests that only the egress of pathogen-specific CTLs is prevented in the presence of localized infection [69] , and , thus , in the absence of co-infections , the number of non HIV-specific CTLs is expected to remain constant in all patches . Although we do not model the proliferation of HIV-specific CTLs within patches here , in the supplemental text we present a “CTL proliferation” version of the model in which this process is included . Since most CTL escape mutations probably sweep through the viral population during early infection [70–72] , we also ignore the appearance of CTL escape mutations during chronic infection . Although such mutations do occur and sweep through within-host populations , they do so only slowly , suggesting their selective advantage is relatively weak [73] . Finally , Eq 1 . 5 describes the dynamics of the reservoir compartment , containing the set of latently infected resting CD4+ T cells , which die at rate δL . A list of all of the parameters and their values is given in Table 1 . This metapopulation model has some similarities to a recently published model of HIV infection in a network of lymphoid tissues [74] , with the crucial difference that in our model local colonisation and extinction dynamics are observed due to the localised killing of infected cells as a consequence of the accumulation of CTLs within patches . We wrote a fully stochastic population-based simulation in C++ based on Eqs 1 . 1–1 . 5 and with a time step of 0 . 0001 to 0 . 001 days , depending on parameters ( S1 Simulation Code ) . The number of events of each type ( infection , death , and migration ) occurring during a time step was drawn from either binomial or multinomial distributions as appropriate . We chose this simulation method rather than using Gillespie-type simulation to increase computational efficiency . Because this is intended to be a model of the chronic , asymptomatic phase of infection , all simulations were initiated with 108 infected cells distributed evenly across the patches , so as to represent the large viral load associated with the end of the acute phase of infection [75] . We also assume that the number of HIV-specific CTLs has reached equilibrium at the systemic level ( reflected in the constant parameter c ) and that any CTL escape mutations have already swept through the viral population ( enabling us to assume k is also constant ) , although the patches themselves are initiated with no HIV-specific CTL present . Because a large reservoir of long-lived latently infected cells is established during the early phases of infection [76] , the simulations were initiated with a reservoir size of 107 , if included [44] . We introduced heterogeneity among the patches by sampling the values of βi in patch i from the uniform distribution on [0 . 5 β¯ , 1 . 5 β¯] . For all other parameters , the patches are assumed to be identical . Because for almost all parameter values the system reaches a steady state well before 40 days , the stationary number of infected cells was calculated by averaging the total number of infected cells in the metapopulation between days 60 and 100 from the start of the simulation . See S1 Data for the output from all the simulations . We want to compare the results of the metapopulation model with the equivalent model , in which we have a single well-mixed tissue . The equations describing the single patch model are obtained by setting N = 1 and dropping index i from Eqs 1 . 1–1 . 5 . At equilibrium , the total number of infected cells is therefore given by: Ytot*=max{Bβ[β ( 1−λ+ωλ ) −CC] ( 1+εM+δB ) , 0} ( Eq 2 ) where C = δ+ε ( 1−MM+δB ) +k . If the reservoir size is fixed , the equilibrium number of infected cells is instead given by: Ytot*= ( 1+εM+δB ) 12βC[B[β ( 1−λ ) −C]+ωaβL¯+ ( B[β ( 1−λ ) −C]+ωaβL¯ ) 2+4ωaBβCL¯] ( Eq 3 ) Viral infectivity within patches , β , can be estimated from published values of R0 ( the within-host basic reproduction number of HIV in the absence of a CTL response ) . Specifically , assuming infected cells spend the vast majority of the time within patches , β ≈ δ R0 , where δ is the CTL-independent death rate of infected cells within patches . Since δ ≈ 1 per day [42 , 43] , β ≈ R0 . For HIV , estimates of R0 vary between 2 and 26 [77 , 78] , giving us estimates of β between 2 and 26 per day . Noting that only two of the 51 patients included in these studies had an R0 higher than 20 , and because in vitro measures of viral infectivity vary by more than a factor of 10 among patients [79] , as does replicative capacity [80] , taking a range of β values between 1 and 20 per day is likely to reasonably capture the heterogeneity among patients . Estimating the maximum rate at which CTLs kill infected cells , k , is more problematical , with estimates in the literature ranging between 0 . 1 per day to 500 per day , depending on which methods and which viruses are used [81 , 82] . Estimates for HIV-1 range between 0 . 1 and 10 per day [81] , but there is still considerable uncertainty as to how best to analyse available data , and particularly how the inclusion of high rates of CTL killing during the eclipse phase of the cell infection cycle ( i . e . , before viral production has started ) alters the interpretation of the data [57] . In addition , we are interested in the maximum rate of CTL killing in very localized areas of the body , not the average rate of killing at the systemic level that is estimated from in vivo data , and therefore even the higher estimates for HIV-1 might underestimate k . We therefore take a range of k between 1 and 20 per day to reflect heterogeneity among individuals , with the caveat that it is important to recognize the uncertainties surrounding the choice of this range of values . We interpret patches as the T cell areas in secondary lymphoid organs , such as lymph nodes , the spleen , and Peyer’s patches . Lymph nodes often contain several physically separated T cell areas [83] , the number of which will depend on the size of the lymph node . In mice , there are about 1 . 6 T cell areas per lymph node on average [84] . Because the lymph nodes of humans are about 100 times heavier than those of mice , and there are 550 or so lymph nodes in adult humans , there are likely to be many thousands of patches in the lymph nodes of humans . In the spleen of a mouse there are about 10 Malpighian bodies , which are nodules of white pulp that contain many lymphocytes . Because a human spleen weighs about 1 , 500 times more than a mouse spleen , it also could easily contain thousands of patches . Finally , there are about 100–200 Peyer’s patches in a human gut [85] . Although by these rough calculations there could be , as an upper estimate , 100 , 000 patches in a human adult , in the main text we have used an estimate of N = 10 , 000 , which is probably correct within an order of magnitude and makes the simulations computationally feasible . Because we do not yet have a full understanding of lymphocyte trafficking , patches might also be interpreted as larger areas within which there could be rapid movement of lymphocytes ( a whole lymph node , for example ) . Therefore , we also investigate how the system behaves for N = 1 , 000 and N = 100 ( see S1 Text ) . To estimate the number of susceptible cells in the blood , xB , we first assume that the CD4+ T cell count in the blood of an HIV infected individual has fallen to a third of those in a healthy adult , giving a count of 333 cells per mm3 . Although HIV is capable of infecting unactivated CD4+ T cells [86] , we only consider replication in activated CD4+ T cells , since this is much more efficient . If we assume that 1% of CD4+ T cells are activated and a typical blood volume is 5 liters , this gives us xB = 1 . 67 x 107 . The residence time of CD4+ T cells in the blood ( including the vasculature of non-lymphoid tissue ) is about 30 min [36] , giving us a per capita migration rate , M = 48 per day . Estimates from experimental data suggest that the residence time of lymphocytes in secondary lymphoid tissue is in the order of about 10 hours [39 , 40] , and therefore we use a value of ε of 2 . 5 per day in our core parameters . In the spleen , residence time has been estimated to be shorter than this [39 , 40] giving a faster rate of egress from patches in the spleen . However , by our calculations , there are about 10 times more patches in the lymph nodes than in the spleen , and , in addition , a faster rate of egress will result in fewer infected cells; thus , the overall number of infected cells in the metapopulation will be driven by the longer duration patches . To estimate the maximum number of HIV-specific CTLs per patch , we first note there are approximately 1x1011 CTLs in a typical human adult , many of which reside in secondary lymphoid tissue [24] . Since HIV-infected individuals have , on average , twice as many CTLs in the blood as uninfected individuals [87] , and about 10% of these are HIV specific [88] , then the average patch will have about 2x1010/N CTLs , where N is the number of patches . If the maximum carrying capacity of any individual patch is 5 times this , then a ballpark figure for zmax is 1x1011/N . Note that the within-patch dynamics depend on the CTL immigration parameter , c , and the density of CTLs in a patch ( zi ( t ) /zimax ) , not the absolute value of zmax . We use a value of c = 0 . 5 , as this means the number of CTLs within a patch typically reaches a maximum in between 1 and 4 days , in line with empirical observations [89] . Using these values , we can estimate the number of HIV-specific CTLs expected in an uninfected patch , z0=czimax/ ( c+ε ) = 1 . 67 x 106 , if N = 10 , 000 and all patches are identical . All other parameters are estimated directly from the literature ( see Table 1 ) . | When a person is infected with HIV , the initial peak level of virus in the blood is usually very high before a lower , relatively stable level is reached and maintained for the duration of the chronic infection . This stable level is known as the set-point viral load ( SPVL ) and is associated with severity of infection . SPVL is also highly variable among patients , ranging from 100 to a million copies of the virus per mL of blood . The replicative capacity of the infecting virus and the strength of the immune response both influence SPVL . However , standard mathematical models show that variation in these two factors cannot easily reproduce the observed distribution of SPVL among patients . Standard models typically treat infected individuals as well-mixed systems , but in reality viral replication is localised in T-cell centres , or patches , found in secondary lymphoid tissue . To account for this population structure , we developed a carefully parameterised metapopulation model . We find the system can reach a steady state at which the viral load in the blood is relatively stable , representing SPVL , but surprisingly , the patches are highly dynamic , characterised by bursts of infection followed by elimination of virus due to localised host immune responses . Significantly , this model can reproduce the wide distribution of SPVLs found among infected individuals for realistic distributions of viral replicative capacity and strength of immune response . Our model can also be used in the future to understand other aspects of chronic HIV infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"blood",
"cells",
"cell",
"death",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"body",
"fluids",
"viral",
"transmission",
"and",
"infection",
"population",
"dynamics",
"immunology",
"cell",
"processes",
"microbiology",
"simulation",
"and",
"modeling",
"population",
"biology",
"viral",
"load",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"white",
"blood",
"cells",
"metapopulation",
"dynamics",
"animal",
"cells",
"t",
"cells",
"viral",
"replication",
"hematology",
"blood",
"cell",
"biology",
"anatomy",
"virology",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"viral",
"diseases"
]
| 2016 | Large Variations in HIV-1 Viral Load Explained by Shifting-Mosaic Metapopulation Dynamics |
Long non-coding RNAs contribute to dosage compensation in both mammals and Drosophila by inducing changes in the chromatin structure of the X-chromosome . In Drosophila melanogaster , roX1 and roX2 are long non-coding RNAs that together with proteins form the male-specific lethal ( MSL ) complex , which coats the entire male X-chromosome and mediates dosage compensation by increasing its transcriptional output . Studies on polytene chromosomes have demonstrated that when both roX1 and roX2 are absent , the MSL-complex becomes less abundant on the male X-chromosome and is relocated to the chromocenter and the 4th chromosome . Here we address the role of roX RNAs in MSL-complex targeting and the evolution of dosage compensation in Drosophila . We performed ChIP-seq experiments which showed that MSL-complex recruitment to high affinity sites ( HAS ) on the X-chromosome is independent of roX and that the HAS sequence motif is conserved in D . simulans . Additionally , a complete and enzymatically active MSL-complex is recruited to six specific genes on the 4th chromosome . Interestingly , our sequence analysis showed that in the absence of roX RNAs , the MSL-complex has an affinity for regions enriched in Hoppel transposable elements and repeats in general . We hypothesize that roX mutants reveal the ancient targeting of the MSL-complex and propose that the role of roX RNAs is to prevent the binding of the MSL-complex to heterochromatin .
In many animal species with distinct sexes , sex-chromosomes contribute to genetic sex determination . In species with male heterogamety such as humans and fruit flies , the male sex-chromosome pair consists of two morphologically and genetically different chromosomes ( one X and one Y ) whereas females are homogametic , having two X chromosomes . The heteromorphic sex-chromosomes are believed to have evolved from a pair of autosomes in which a male-determining locus was acquired on one homolog to form a proto-Y chromosome that subsequently underwent a series of mutation and selection events that conferred male advantage and suppressed recombination with the proto-X , eventually leading to the degeneration of the Y-chromosome . Gene expression imbalances arise because X-chromosomal genes in male genomes are only present in one copy whereas autosomal genes and X-chromosomal genes in females are present in two copies . Dosage compensation mechanisms evolved in order to balance the relative expression levels of X-chromosomal genes between the sexes and in relation to autosomal genes [1]–[3] . Dosage compensation in D . melanogaster involves a combination of general buffering effects that act on all monosomic regions [4]–[6] and the specific targeting and stimulation of the male X-chromosome by the male-specific lethal ( MSL ) complex . Together , these processes increase X-chromosomal gene expression by approximately a factor of two [1] , [7] . The MSL-complex consists of five proteins ( MSL1 , MSL2 , MSL3 , MLE , and MOF ) and two redundant long non-coding RNAs ( roX1 and roX2 ) [7]–[9] . It is believed that the hypertranscription of the male X-chromosome is partly due to the enrichment of histone 4 lysine 16 acetylation ( H4K16ac ) . This acetylation is catalyzed by the acetyltransferase MOF and opens the chromatin's structure [10] , [11] . The complete MSL-complex only forms in males due to the male-specific expression of MSL2 and the roX RNAs [12]–[16] . Notably , even though most genes on the X-chromosome appears dosage compensated in the 2-fold range [17] , [18] the MSL-complex only contributes to part of this increase [6] , [19] , [20] . In addition , many genes are compensated without any significant recruitment of the MSL-complex [17] . An alternative model for the role of the MSL-complex in dosage compensation has been proposed by Birchler and colleagues [21]–[23] . According to their inverse dosage effect model the compensation in males is caused by the stoichiometric change of regulator ( s ) on the X-chromosome relative to the remainder of the genome . The main role of the MSL-complex is to sequester MOF from the autosomes to avoid autosome up-regulation and to limit the activation potential of MOF when targeted as part of the MSL-complex [21]–[24] . It is still not clear when , where and how the MSL-complex is assembled or which features of the X-chromosome allow its recognition . Several lines of evidence indicate that MSL1 and MSL2 are the core components of the MSL-complex . Notably , the absence of either one abolishes the binding of the remaining components of the complex to the X-chromosome [25] . The RING domain of MSL2 allows it to interact with MSL1 , and the cystein-rich ( CXC ) domain of MSL2 allows the MSL1-MSL2 complex to recognize and bind DNA [26] , [27] . The incorporation of roX RNAs into the MSL-complex is hypothesized to occur co-transcriptionally [28] and depends on their interaction with MSL2 and the RNA helicase MLE , which binds to stem-loop structures on roX RNAs in an ATP-dependent manner [29]–[32] . The roX1 and roX2 gene loci have been identified as two of the strongest high affinity sites ( HAS ) for MSL-complex targeting , out of the roughly 250 HAS on the X-chromosome [33] , [34] . HAS are defined as sites targeted by MSL1 and MSL2 in the absence of msl3 , mle or mof [25] , [34] , [35] and sites that are sufficient to recruit MSL even when inserted on an autosome [36] . HAS are enriched in a conserved consensus GA-rich DNA sequence motif [34] , [37] , [38] . The prevailing model is that the MSL-complex initially binds at the HAS and that its presence at these sites facilitates the more transient binding of additional MSL-complexes to neighboring active genes [8] , [9] , [39] . The transcriptional statuses of X-chromosomal genes influence the distribution of MSL binding because the complex is biased to exons and the 3′ ends of actively expressed genes; its binding correlates with enrichment in histone 3 lysine 36 trimethylation ( H3K36me3 ) [34] , [37] , [40]–[43] . Other features such as the local chromatin context , H3 depletion , MSL-complex concentration , levels of affinity and sequence composition also contribute to the recognition and spreading of the MSL-complex over the male X-chromosome [35] , [38] , [44]–[46] . The roles of the two redundant long non-coding RNAs , roX1 and roX2 , in the targeting of the entire male X-chromosome by the MSL-complex are not fully understood at present . Studies on polytene chromosomes have shown that in the absence of both roX1 and roX2 , MSL2 and H4K16ac become less abundant on the male X-chromosome , with the MSL-complexes being relocated to the chromocenter , the 4th chromosome and a few other autosomal sites [20] , [47] , [48] . In this work , we analyzed MSL-complex targeting in roX1 roX2 mutants in order to unravel the specific roles of roX RNAs in MSL targeting and to learn more about the evolution of chromosome-specific targeting and dosage compensation . We performed ChIP-seq and cytological analyses of MSL proteins in roX1 roX2 double mutants , and analyzed their genome-wide binding profiles . It was found that in the absence of roX RNAs , the MSL-complex binds to the previously identified HAS on the X chromosome , the pericentromeric regions of all chromosomes , and specifically to six genes on the 4th chromosome . Analysis of the autosomal sequences bound by MSL in roX mutants showed that MSL has an affinity for regions enriched in Hoppel transposable elements , NTS ( non-transcribed spacers ) and repeats . Our results suggest that one role of the roX RNAs is preventing the MSL-complex from binding to heterochromatic repeats , suggesting that targeting heterochromatin is an intrinsic and ancient property of the MSL-complex .
To test the role of roX RNAs in MSL targeting , we performed immunostaining experiments on polytene chromosomes of roX1 roX2 double mutants ( hereafter called roX mutants ) . In the absence of roX RNAs , the extent of MSL-complex targeting to the X-chromosome was dramatically reduced and the complex was relocalized to the chromocenter and to three distinct regions on the 4th chromosome ( Fig . 1A ) . The disruption of MSL targeting seen in roX mutants is clearly different from the disturbance that occurs when the protein components of the complex are removed: in msl1 or msl2 mutants , no MSL-complexes are formed on the X-chromosome at all [25] . Conversely , as shown in Fig . 1A , in mle or mof mutants , the MSL-complex is exclusively targeted to a limited number of bands on the X-chromosome . This shows that the roX RNAs and the protein components of the complex have different functional roles in MSL chromatin targeting . To exclude the possibility that the binding of the MSL-complex to pericentromeric heterochromatin in roX mutants is unique to polytene chromosomes , we analyzed MSL binding in relation to the pericentromeric repeat 1 . 686 ( which is known to be enriched in the pericentromeric regions of chromosomes 2 and 3 [49] ) in interphase nuclei from brain cells of wild type samples and roX mutants . In the wild type and mof mutants , the MSL3-bound X-chromosome occupies a part of the nucleus that is clearly separated from the pericentromeric regions ( Fig . 1B and 1C ) . In roX mutants , the normal binding of MSL3 is altered and the complex is observed in spots that colocalize with the centromeric repeats . This colocalization is three times more frequent in roX mutants than in wild type or mof mutants ( Fig . 1C ) . We therefore conclude that the relocalization of MSL in the absence of roX RNAs observed in salivary gland nuclei also occurs in diploid interphase nuclei . Interestingly , MSL binding in metaphase chromosomes of wild type and roX mutants is similar and is restricted to the euchromatic part of the X-chromosome ( Fig . 1D ) . It remains to be determined why MSL doesn't target centromeres in the highly compacted mitotic chromatin in roX mutants . To better understand how the genome-wide targeting of MSL depends on roX RNAs , we performed MSL1 , MSL2 and MOF ChIP-seq experiments on salivary glands from wild type individuals and roX mutants . These experiments confirmed the results of the immunostaining studies , showing that there is a pronounced decrease in MSL binding along the X-chromosome in roX mutants although binding persists at specific locations . Visual inspection demonstrates that the MSL enrichment peaks along the X-chromosome in roX mutants coincide with the previously defined HAS ( Fig . 2A ) [34] , [37] , [38] . Notably , although all previously defined HAS are not recognized by MSL enrichment in roX mutants , all enrichment peaks coincide with HAS . To verify that the roX RNA-independent MSL enrichment peaks on the X-chromosome correspond to the previously mapped HAS , we calculated the shortest distance between the coordinates of the HAS and those of the MSL1 binding sites on the X-chromosome in roX mutants identified in our ChIP-seq experiments . The fractions of sites bound by MSL1 in roX mutants were plotted against distance to nearest HAS , and the distances between the coordinates of HAS and random positions on the X-chromosome were used as controls . As seen in Fig . 2B , the largest fraction of the MSL binding sites in roX mutants overlap with HAS . This is in clear contrast to the control , in which the largest fraction of random X-chromosome sites are>35 kb away from HAS . Taken together this means that although all of the 263 previously defined HAS are not bound by MSL in roX mutants , in principle all of our 208 defined MSL binding sites in roX mutants target HAS . These results demonstrate that MSL binding to HAS on the X-chromosome occurs independently of roX RNAs . It has been shown that roX RNAs evolve rapidly , only sharing about 90% and 80% sequence homology in such closely related species as D . simulans and D . yakuba , respectively [30] . We therefore sought to determine whether HAS , previously shown to be enriched in a GA-rich motif [37] , [38] , are under high evolutionary pressure . To facilitate comparison with other Drosophila species , we generated ChIP-seq data for MSL1 binding in wild type Drosophila simulans and performed a motif analysis in the MSL1-bound regions on the X-chromosome of this species ( Fig . 2C ) . We found highly similar GA-rich motifs to be enriched within MSL targets on the X-chromosome in roX mutants as well as on the X-chromosome of wild type D . simulans ( Fig . 2D ) . These results show that the roX RNAs are not involved in MSL targeting to HAS and that the HAS motif is evolutionarily conserved . The binding of MSL to the 4th chromosome in the absence of roX RNAs is intriguing because there are several lines of evidence suggesting an evolutionary relationship between the 4th chromosome and the X-chromosome [1] , [50]–[52] . Our ChIP-seq profiles show that the MSL-complex binds specifically to six genes on the 4th chromosome in roX mutants: Ankyrin , Rad23 , CG2177 , PMCA , Mitf and Dyrk3 . The locations of these genes correspond to those of the MSL-stained bands seen on polytene chromosomes ( Fig . 3A ) . One important question when considering the binding of MSL outside the X-chromosome is whether a complete and functional MSL-complex is formed at these locations . Our immunostaining experiments in roX mutants showed that all of the complex's protein components ( MSL1 , MSL2 , MSL3 , MLE and MOF ) colocalize perfectly at the chromocenter and at the three bands on the 4th chromosome ( Fig . 3B ) . In addition H4K16ac is also enriched at these three bands in roX mutants , which indicates that the MSL-complex is complete and active ( Fig . 3B and S1 Figure ) . Note that H4K16ac on the 4th chromosome shows a broader enrichment pattern compared to the MSL proteins in similarity to what previously have been observed for H4K16ac in relation to MSL on the male X-chromosome in wild type [10] . Next we tested the H3S10 kinase JIL1 , previously shown to be enriched on the male X-chromosome and dependent on a functional MSL-complex for its targeting [53]–[55] . JIL1 has previously been shown to co-immunoprecipitate with the MSL-complex under low stringency conditions or after formaldehyde cross-linking [54] , [56] . Interestingly , like the MSL-complex , JIL1 is also relocalized to the chromocenter and the three regions on the 4th in the absence of roX RNAs ( Fig . 3C ) . Since H4K16ac overlaps with all the other proteins from the MSL-complex in roX mutants we asked if the six identified genes on the 4th bound by MSL in roX mutants have higher transcriptional output than in wild type . The relative expression of the six 4th chromosome MSL-bound genes was not found to differ significantly between wild type and roX mutants ( Fig . 3D ) . One tempting hypothesis based on the targeting of the 4th chromosome is that the MSL-complex in D . melanogaster still has an affinity for ancestral X-chromosomal sequences , now present on the 4th . We performed BLAST searches for the sequences of the six 4th chromosome genes targeted by MSL in D . melanogaster roX mutants in the distantly related species D . virilis and D . willistoni and found that in both species , these six genes are assigned to the sequence scaffold on which all of the other 4th chromosome-linked genes are located rather than to the X chromosome . In the even more distantly related species D . busckii , the whole correspondent to the 4th chromosome of D . melanogaster is fused to the X-chromosome [57] , [58] . Interestingly , it has recently been shown [50] that in D . busckii the sequences corresponding to the D . melanogaster 4th chromosome are present in more copies in females than males . However , the female-to-male ratio is less than 2 meaning that the corresponding homologs on the Y-chromosome are not fully degenerated or that some but not all genes on the corresponding 4th chromosome have degenerated homologs on the D . busckii Y-chromosome . We hypothesized that the six genes targeted by MSL in roX mutants actually skew the ratio and therefore calculated the female-to-male ratio of these D . busckii orthologs relative to the other chromosome 4 genes . By using previously reported data [50] we found that there is no significant difference between the 6 genes and the other 4th chromosome genes ( S2 Figure ) . We conclude that a complete and active MSL-complex binds with high specificity to six genes on the 4th chromosome , although the reason for this specificity remains elusive . The pericentromeric regions and the 4th chromosome are both heterochromatic regions of the D . melanogaster genome that are targeted by MSL in roX mutants and are enriched in satellite repeat sequences , transposable elements , and the heterochromatic proteins HP1a and HP2 , among others . Our ChIP-seq results show that in the absence of roX RNAs , the MSL-complex targets the pericentromeric regions of all chromosomes and that its abundance increases gradually on moving towards the centromere . This tendency is illustrated for chromosome 3L in Fig . 4B . One possible mechanism underlying this binding is that MSL recognizes a specific recruitment element but that its binding to this element is blocked by the presence of roX RNAs . Another possibility is that the MSL-complex has an intrinsic affinity for heterochromatic sequences or repeats in general . Using the MEME software , we analyzed the identified MSL1-bound heterochromatic sequences of each chromosome ( 2LHet , 2RHet , 3LHet , 3RHet , 4Het , XHet ) . A specific motif corresponding to repeats in the Hoppel ( 1360 ) transposable element was found to be significantly enriched ( Fig . 4A and B ) . In situ DNA hybridization experiments using the identified motif as a probe in conjunction with MSL staining revealed a high degree of colocalization between the MSL-complex and the motif in both the pericentromeric heterochromatin and the three specific bands on the 4th chromosome ( Fig . 4C ) . To determine whether this motif can act as an MSL-recruitment element , we generated a construct containing three tandemly repeated copies of a 108-nucleotide Hoppel element featuring the motif in question . This repeat segment was placed upstream of a cDNA copy of ankyrin ( a gene on the 4th targeted by MSL in roX mutants ) under an endogenous promoter . The construct was inserted into the 3L:65B2 PhiC landing platform and tested for MSL binding in a roX mutant background . The transgene was visualized using mini-white DNA-FISH and MSL-complex was not detected on the target ( S3 Figure ) . These results suggest that the repeat motif from the Hoppel transposable element , which is enriched at MSL-targeted regions ( 4th and pericentromeric ) in roX mutants , is not by itself sufficient to recruit the MSL-complex . However , we cannot exclude the possibility that recruitment might be achieved with a greater number of motif copies . DNA sequences from centromeres , telomeres , the Y-chromosome and other heterochromatic regions are not assembled to any region of the D . melanogaster genome due to their highly repeated nature , and the mapping of sequences recovered in the ChIP-seq normally discards the large number of repeated sequences in the genome . We suspected that the non-mapped reads recovered by ChIP-seq might hold information about other transposable elements targeted by MSL in roX mutants in the above-mentioned heterochromatic regions . To test this hypothesis , we aligned all of the ChIP-seq reads to the repeat class sequences from the Repbase Update database and calculated RPKM values for each repeat class . Using this approach we found that in roX mutants there were strong enrichments of three repeat classes: PROTOP_B , PROTOP_A and NTS ( Non-transcribed Spacer ) ( Fig . 5 ) . Interestingly , the PROTOP is a family of autonomous DNA transposons that have been suggested to be ancient ancestors of the P-element and Hoppel element transposon families [59] and PROTOP_A and PROTOP_B are listed as synonyms of Hoppel [60] . Our results confirm that MSL has an affinity for regions enriched in repeats from Hoppel and PROTOP transposable elements and for NTS . All of these are highly repeated elements that are present in heterochromatic regions of the genome . In addition to the Hoppel transposable element repeats , the analysis of the mapped and unmapped sequences bound by MSL in roX mutants recovered NTS sequences , which occur between ribosomal DNA genes which are organized in tandem repeats . Because MSL targets some autosomal sites across the genome in the absence of roX , we wondered whether these sites were also enriched in repeats and whether repeats in general were enough to recruit MSL . To test this hypothesis , we analyzed the enrichment of repeat masked sequences around MSL targeted regions . Since the MSL-complex mainly targets expressed genes we calculated the enrichment of repeats surrounding the TSS ( transcription start site ) of genes that are expressed in salivary glands ( the tissue of our binding data ) and are located in either MSL-bound or MSL-unbound regions of the genome . We examined both the X-chromosome and the autosomes ( excluding chromosome 4 and the mapped pericentromeric regions ) in this analysis , taking into account the MSL-bound regions in wild type and roX mutants that were identified based on our ChIP-seq data . The density of satellite repeats on the X-chromosome is reportedly greater than on the 2nd and 3rd chromosomes [61]–[63] . Our results are consistent with this finding and show that the regions surrounding expressed genes on the X-chromosome have a somewhat higher repeat content than those surrounding autosomal expressed genes ( Fig . 6A ) . Strikingly , autosomal expressed genes bound by MSL in roX mutants are enriched in surrounding repeats whereas the regions surrounding unbound autosomal expressed genes have a low repeat content . The repeat content of regions surrounding X-chromosomal genes bound by MSL was also higher than that of unbound regions , but the difference was less pronounced than for autosomal genes . Our results show that MSL targeting of autosomal sites in roX mutants correlates with high repeat content . We therefore sought to determine whether any repeat sequence would recruit MSL in the absence of roX . Two repeat types were tested . First we analyzed clusters of tandemly repeated P[lacW] transgenes on the 2nd chromosome in roX mutants [64] . The P[lacW] transgene contains sequences of the P-element flanking the E . coli lacZ gene and the mini-white gene . In the absence of roX , MSL2 did indeed bind to a cluster of 7 tandemly repeated copies of P[lacW] ( C-2 , BX-2 , T-1 ) but not to a cluster with only 2 copies ( 1A-6 ) ( Fig . 6B ) . Although mini-white originates from an X-linked gene , it does not contain HAS and is not an ectopic MSL target in the wild type . These results show that genes in tandem repeats are enough to recruit the MSL-complex when roX RNAs are absent . The second repeat cluster tested in a roX mutant background was the 256 lac repeats of E . coli upstream of a white reporter gene [65] . In this system , the DNA binding domain of the lac repressor ( lacI BD ) fused to HP1a is tethered to a reporter transgene that contains repetitive binding sites for the lacI BD ( lacO repeats ) . The transgene was targeted by HP1a in about 50% of nuclei but never by MSL3 ( S4 Figure ) . Overall , these results suggest that in absence of the roX RNAs , MSL has an intrinsic general affinity for repeated sequences . However , the influence of the repeat length and number of copies as well as the sequence specificity of the complex all remain to be elucidated .
We and others have previously shown that in the absence of roX the MSL-complex is redistributed to targets corresponding to pericentric heterochromatin and the 4th chromosome , or “green chromatin” according to recent chromatin structure-based definitions [20] , [47] , [48] , [67] . Since large parts of these “heterochromatic” regions targeted by MSL in roX mutants are under-replicated in polytene chromosomes it was important to determine whether this redistribution also occurs in diploid cells that have very different ratios of the relevant DNA motifs . Notably , although the reduction in MSL-complex abundance on the X-chromosome is much more dramatic in roX mutants than in mle or mof mutants and the MSL-complex is relocalized to heterochromatic regions in roX mutants , escaping males are recovered in roX mutants in contrast to the complete male lethality observed in mle , msl3 or mof mutants [47] , [68] , [69] . The results of our studies on interphase nuclei from brain tissue showing colocalization between MSL3 and centromeric regions further support the interpretation that in absence of roX , the MSL-complex targets heterochromatin . It has previously been shown that the fraction of escaper males in roX mutants is significantly higher in roX1 roX2/0 males , i . e . males lacking a Y-chromosome , than in roX1 roX2/Y males . Importantly , the Y-chromosome is predicted to be 40 Mb in length and thus accounts for>10% of all genomic DNA in male cells [70] . It seems likely that the reduced abundance of heterochromatic target DNA and/or the greater compaction of the remaining heterochromatin caused by the loss of the Y-chromosome increases the X-chromosomal targeting of MSL in roX1 roX2/0 males , explaining their increased survival . It is tempting to speculate that in roX mutant interphase nuclei , the centromeric regions of autosomes have a tendency to colocalize with the X-chromosome within the nucleus , in a region where the local concentration of the MSL-complex is expected to be high . In fact , previous studies have shown that HAS are closer in the nuclear space in males than in females , suggesting that long-range associations between MSL-complex target sites shape nuclear organization [71] . On mitotic metaphase chromosomes , the MSL-complex is only seen on the distal X-chromosome in the wild type . Surprisingly , the same pattern is seen in roX mutants , although the specificity of the MSL targeting is somewhat lower in these cases . We speculate that the HAS present on the X-chromosome provide superior targets for the complex when transcription is suppressed ( as is the case in metaphase ) compared to the centromeric regions . Overall our results suggest that the MSL-complex has a greater affinity for roX RNAs than its heterochromatic targets and so roX RNAs restrict the targeting of the complex to the X-chromosome . In keeping with this hypothesis , a previous study showed that when MSL1 and MSL2 are overexpressed , MSL2 targets not only the X-chromosome but also some autosomal sites , the 4th chromosome and the chromocenter [72] . roX1 RNA was only detected on the X-chromosome and on some autosomal sites but not on the 4th or the chromocenter . This suggests that the MSL-complex has an intrinsic affinity for heterochromatin and a balanced amount of roX RNAs are required to restrict the complex from these targets . An affinity for heterochromatin components may in fact be part of the mechanism to limit the activating potential of MOF when sequestered to the male X-chromosome . Although not detected in ChIP experiments or chromosome immunostainings [73]–[76] , a low amount of HP1a along the entire male X-chromosome has been found in genome-wide mapping of HP1a using the DamID technique [77] . In addition , a knock-down of HP1a indicated more lethality in males than in females [78] . It is possible that when the strong binding to the X-chromosome is decreased in the roX mutants , the affinity for HP1a relocates the complex to canonical HP1a binding sites: heterochromatin , 4th chromosome , repeat arrays . In the absence of roX , the MSL-complex still targets a reduced number of sites on the X-chromosome . Based on previous cytological analysis it has been argued that these sites are similar but not identical to the sites targeted in msl3 mutants , i . e . HAS [47] , [68] . Our ChIP-seq results show an almost perfect overlap between MSL targets in roX mutants and the 263 previously identified HAS . In addition , the enriched sequence motif revealed by our bioinformatics analysis is nearly identical in wild type and roX mutants and also when comparing D . melanogaster and D . simulans . Since previous studies showed that MSL targeting to HAS is independent of MSL3 , MOF and MLE , and we found HAS targeting to be independent of roX RNAs , we propose that MSL1 and MSL2 are the only components required for the correct targeting of HAS . Interestingly , we detected a very strong MSL signal and a high enrichment of the complex in the regions surrounding six specific genes on the 4th chromosome in roX mutants . Several lines of evidence suggest an evolutionary relationship between the 4th chromosome and the X-chromosome [1] , [50]–[52] . However , we cannot presently explain why these genes are specifically targeted in this way . Our findings indicate that the MSL-complexes formed at these non-X chromosome locations are complete and active , and even include associated factors such as JIL1 . Importantly we did not observe any obvious change in the expression of the six 4th chromosome genes targeted by the intact MSL-complex ( although this may be partly due to the limited sensitivity of qPCR ) . This again suggests that the activation potential of MOF within the MSL-complex is limited [17] , [24] , [79] . It has been shown that the targeting of MOF alone to reporter transgenes result in a strong increased expression . In contrast , when MOF was targeted as part of the MSL-complex no increased expression was observed [24] . In addition , it is important to note that the expression of all genes from the 4th chromosome is fine-tuned by a balance between HP1a , which represses gene expression , and POF , the chromosome 4-specific protein that stimulates gene expression [80]–[82] . The predicted effect on gene expression due to MSL targeting to the 4th chromosome in roX mutants might be counteracted by HP1a and/or POF . We did not observe any clear difference between roX mutants and the wild type with respect to the binding of POF to the 4th chromosome or the binding of HP1a to either the chromocenter or the 4th . This demonstrates that MSL binding does not interfere with that of HP1a or POF . A link between dosage compensation and transposable elements has previously been suggested for both mammals and Drosophila . The mammalian X-chromosome is enriched in LINE elements ( particularly in its pericentromeric region ) . It has been suggested that these elements boost the X-inactivation signal by acting as anchoring stations for the spread of the Xist RNA [83] . We identified a strong connection between MSL targeting and the Hoppel transposable element using two different approaches . The first involved identifying sequence motifs in the mapped regions , typically pericentric regions on chromosome arms and targets on the 4th chromosome; this revealed a recurring sequence motif from the Hoppel element . The second involved calculating the enrichments of all reads , both mapped and unmapped . Strong enrichment was observed for PROTOP_A and PROTOP_B ( both corresponding to Hoppel [59] ) as well as NTS Dm . Notably , NTS is enriched by MOF also in wild type suggesting that this is an intrinsic target of MOF which is stabilized by the complete MSL-complex in roX mutants . The importance of Hoppel in MSL targeting is supported by its high degree of colocalization with MSL staining in roX mutant males . Because the ChIP technique relies on the analysis of fractionated DNA , we cannot currently say whether it is Hoppel itself or transcribed regions in its vicinity that are targeted by MSL . The Hoppel elements are non-autonomous DNA transposons and among the most abundant transposable elements in the D . melanogaster genome , being enriched in the pericentric heterochromatin and on the fourth chromosome [84] . Since we also observed MSL enrichment in NTS ( Non-Transcribed Spacer ) regions , active autosomal genes surrounded by DNA repeats , and tandemly repeated gene constructs ( P[lacW] ) , it seems that MSL is recruited to repeats in general . A recent study demonstrated that the neo-X of D . miranda has newly evolved chromatin entry sites ( CES ) , also known as high affinity sites-HAS , that recruit MSL and are enriched in the ISX helitron transposable element ( TE ) [85] . The authors suggest that the evolutionary acquisition of the MSL-complex by X-chromosomes involved the acquisition of GA-rich sequence motifs by transposable elements that were capable of functioning as HAS for the MSL-complex , followed by amplification of the TEs across the genome . This may then have been followed by positive selection for these elements on the X-chromosome followed by a refinement process that eroded TEs in non-functional regions and increased their affinity for MSL . The authors further suggest that the heterochromatization of the neo-Y occurs in parallel with the acquisition of dosage compensation on the neo-X . We speculate that the MSL targeting seen in the absence of roX RNAs represents an ancient but still intrinsic property of the MSL-complex . This model suggests that roX RNAs are younger in evolutionary terms than the protein components of the MSL-complex and evolved in parallel with the degeneration of the Y-chromosome , redistributing the MSL-complex to the male X-chromosome and restricting its intrinsic heterochromatic targeting . In fact , a human MSL-complex ( hMSL ) has been identified that contains the homologs of the Drosophila proteins MSL1 , MSL2 , MSL3 and MOF , indicating that the MSL-complex protein components are highly conserved in evolution . Conversely , the roX RNAs evolve rapidly [30] . The hMOF is responsible for the majority of H4K16 acetylation as well as being involved in the acetylation of other substrates such as the p53 protein , and in the regulation of various cellular processes ( reviewed in [86] , [87] ) . The ancient function of MSL in the heterochromatin of a Drosophila melanogaster ancestor may have been to activate the expression of active genes present in repressive environments . The binding of MSL to NTS in roX mutants supports the hypothesis that MSL may have had a role in protecting active genes that are present in multiple copies in the genome , like the ribosomal genes , against repeat-induced gene silencing [83] , [88] . This is also supported by our finding that active genes on autosomes bound by MSL in roX mutants are in repeat-enriched regions . It has been shown that sequences containing P[lacW] in tandem repeats become heterochromatic and the repeated mini-white gene becomes partially repressed [64] . MSL is recruited to this transgene in a roX mutant background suggesting that it is recruited to repeat-induced gene silencing regions .
Flies were cultivated and crossed in vials containing potato mash-yeast-agar medium at 25 C . The wild type strains used were D . melanogaster ( Oregon R ) and D . simulans/w501 ( UC San Diego Drosophila Stock Center ) . The D . melanogaster roX1 roX2 double-mutant males were selected as non-GFP males from a y w roX1ex6 Df ( 1 ) roX252 P[w+ 4Δ4 . 3]/FM7i , P[w+mC ActGFP]JMR3 stock obtained from Yongkyu Park ( New Jersey Medical School , Newark , NJ ) . The mof mutants were obtained by crossing virgin females from the stock mof2; P[w+ mof+]/CyO GFP , obtained from Peter Becker ( Ludwig Maximilians Universität Munchen ) , to wild type males and selecting green-fluorescent males in the progeny . The mle mutants were obtained by selecting the non-green fluorescent males from the cross: mle9 cn1 bw1/CyO , P[w+mC ActGFP]JMR1 ×FM7i , P[w+mC = ActGFP]JMR3/Y; mle1/CyO , P[w+mC ActGFP]JMR1 . The strains carrying the P[lacW] transgene in repeats C-2 , BX-2 , T-1 , and 1A-6 were kindly provided by Stephane Ronsseray ( CNRS-Université Pierre et Marie Curie ) and are described elsewhere [64] , [89] , [90] . Insertions on the second chromosome were rebalanced with CyO , P[w+mC ActGFP]JMR1 , the rebalanced males were crossed to y w roX1ex6 Df ( 1 ) roX252 P[w+ 4Δ4 . 3]/FM7i , P[w+mC ActGFP]JMR3 females , and salivary glands were dissected from non-GFP male larvae . To study the targeting of MSL to lac repeats we used the strains P[hs-HP1 . lacI . BD] and P[Ecol\lacO . 256x . w]157 . 4 . 112 [65] , kindly provided by Lori Wallrath ( University of Iowa ) . roX1ex6 Df ( 1 ) roX252 P[w+4Δ4 . 3]/Y;P[Ecol\lacO . 256x . w]157 . 4 . 112/+; P[hs-HP1 . lacI . BD]/+ males were obtained by crossing w; P[Ecol\lacO . 256x . w]157 . 4 . 112 males with roX1ex6 Df ( 1 ) roX252 P[w+4Δ4 . 3]/FM7i; P[hs-HP1 . lacI . BD] females . Third instar larvae were heat-shocked for 45 minute at 37 C and recovered at room temperature for 2–3 h prior to dissection . To generate transgenic flies carrying a transgene with repeats of the ankyrin gene together with the motif found to be enriched at heterochromatic sites bound by MSL in roX mutants , a DNA fragment containing the attB integration site was excised from pTA-attB [91] with EcoRI and cloned into the CaSpeR-4 vector . The resulting pCas-attB plasmid was used as a cassette for integrating ankyrin cDNA downstream of a 108 nucleotide-long DNA fragment identical to the Hoppel element 1360{}6073 ( FBti0064134 ) , repeated three times . A plasmid containing this Hoppel repeat was produced by GenScript USA Inc . The repeat was excised with KpnI and cloned into pCas-attB ( pCas-attB-1360 ) . A cDNA clone of ank-RB ( LD10053 ) was purchased from the Drosophila Genomic Research Center . The desired DNA fragment was excised with NotI and XhoI and cloned into pCas-attB-1360 digested with the same nucleases . Finally , the promoter region of ank was amplified with the primers 5′-atagcggccgcttaggtatgtaaaattcacgcaa-3′ and 5′-cgagcggccgcaaggcaggctcaggtatttg-3′ , digested with NotI and cloned upstream the ank-RB fragment . Embryo microinjection into the Bl9750 strain ( 3L:65B2 PhiC landing platform ) was performed by BestGene ( Inc ) . Males homozygous for the transgene were crossed to y w roX1ex6 Df ( 1 ) roX252 P[w+ 4Δ4 . 3]/FM7i , P[w+mC ActGFP]JMR3 females and salivary glands were dissected from non-GFP male larvae . Third instar larvae polytene chromosomes from salivary glands were prepared as described previously [92] . Larval brain squashes were performed according to protocol 1 . 9 , method 3 in [93] . Immunostainings were performed as described previously [94] with the following antibodies ( dilutions in parentheses ) : rabbit anti-MSL1 ( 1∶400 ) , MSL2 ( 1∶200 ) , MOF ( 1∶400 ) and MLE ( 1∶2000 ) , and goat anti-MSL3 ( 1∶2000 ) from Mitzi Kuroda ( Harvard Medical School ) ; rabbit anti-JIL1 ( 1∶1000 ) from Peter Becker ( Ludwig Maximilians Universität Munchen ) ; and rabbit anti-H4K16ac ( 1∶300 , sc-8662-R , Santa Cruz ) . The secondary antibodies used were donkey anti-goat or donkey anti-rabbit conjugated with AlexaFluor555 or AlexaFluor488 , respectively ( 1300 dilution , Molecular Probes ) together with DAPI ( 1 µg/ml ) . DNA-FISH combined with immunostaining on polytene chromosomes and brain squashes was performed according to a standard protocol [95] . The probe against mini-white was excised from CaSpeR-4 plasmid using the EcoRI restriction endonuclease and biotin labelled with the BioNick DNA Labeling System ( Life Technologies ) . A FAM-labelled probe against 1 . 686 g/cm3 satellite was purchased from Exiqon . The sequence of the 33 nucleotide-long biotin-labelled probe that was used against the heterochromatic motif found to be enriched at MSL-bound regions in roX mutants was 5′-TAACAAGATGCGTAACGGCCATACATTGGTTTG-3′ . Antibodies for the detection of DNA probes were mouse anti-FITC and mouse anti-biotin ( 1∶500 , Jackson ImmunoResearch ) with goat anti-mouse labelled with AlexaFluor488 as secondary antibody . HP1a was detected with rabbit PRB291C antibody ( 1∶400 , Covance ) and with donkey anti-rabbit AlexaFluor555 . Preparations were analyzed using a Zeiss Axiophot microscope equipped with a KAPPA DX20C CCD camera . For comparisons between strains or proteins stained , the protocol was run in parallel . Nuclei with clear cytology were chosen on the basis of DAPI staining and photographed . At least 20 nuclei for each genotype were used in these comparisons , and at least four slides of each genotype were analyzed . For the colocalization analysis of the 1686 probe DNA/FISH combined with MSL3 immunostaining , 8 biological replicates ( 8 slides with one brain per slide ) from each of the wild type , mof mutants , and roX1 roX2 mutants were analyzed with 30–50 nuclei scored per replicate . Nuclei were chosen on the basis of DAPI staining and colocalization was scored . Total RNA was extracted from third instar larvae using TRI reagent ( Ambion ) according to the manufacturer's protocol . Three biological replicates from the wild type and roX1 roX2 mutants were produced , consisting of 10 male larvae each . The RNA was reverse-transcribed using the iScript cDNA Synthesis kit ( Bio-Rad ) and amplified by real-time PCR using iQSYBR Green Supermix ( Bio-Rad ) according to the manufacturer's instructions . Primer pairs used are listed in Supplementary Table S1 . The expression levels were normalized to the amount of RpL32 mRNA in each replicate . The ChIP experiments were performed in salivary glands from third instar larvae as previously described [80] , [81] using 3 µl of anti-MSL1 , 3 µl of anti-MOF and 2 µl of anti-MSL2 ( provided by Mitzi Kuroda , Harvard Medical School ) . To verify the quality of the input and ChIP samples before sequencing , we analyzed the ChIP DNA/input DNA ratio , using real time PCR as described previously [81] . We generated one replicate of MSL1 , MOF and MSL2 for each genotype ( D . simulans wild type , D . melanogaster wild type and D . melanogaster roX1 roX2 homozygous mutant ) . Library preparation and AB SOLiD 5500xl sequencing were performed by Uppsala Genome Centre . The MSL1 sample from the D . melanogaster wild type was unfortunately lost . The complete dataset is available at http://www . ncbi . nlm . nih . gov/geo/ ( Accession: GSE58768 ) . Uniquely mapped reads from all samples were aligned against the D . melanogaster ( Dm ) reference sequence ( release 5 ) and D . simulans ( Ds ) reference sequence ( release 1 ) using the Applied Biosystems Bioscope software v1 . 2 . 1 . Enrichment ratios for the MSL1 , MSL2 and MOF proteins in Dm and Ds wild type and Dm roX1 roX2 mutant samples were calculated as described previously [96] . Ratio values every 10 bp were extracted across the genome and median smoothed using a window size of 500 bp or 2000 bp; windows with fewer than 25 and 100 data points , respectively , were discarded . Since the MSL1 enrichment ratios for the roX1 roX2 mutant samples and the MSL2 enrichment ratios for the wild type samples were the most distinct , these samples were selected as the representative ones in the wild type and roX1 roX2 mutant groups . To define the MSL-bound regions , the highest 1 . 5 percent of the ratio values were extracted . Data units that crossed this cutoff and that are spaced no more than 200 bp from each other were then combined into MSL-bound regions . Regions of less than 200 bp or containing fewer than five data units were discarded . Each bound region was assigned a value equal to the average of the top five consecutive ratio values . The MSL peak centre for each MSL bound region was set to the centre position of the top five consecutive ratio values . The 263 high affinity sites ( HAS ) as defined in [34] , [37] , [38] were used to calculate the closest distance to the MSL1-bound regions ( defined as above ) on the X-chromosome in roX mutants . The distances were divided into 8 bins and the fraction of MSL1-bound sites in each bin was calculated . As a control we used the distance between the HAS and random locations on the X-chromosome . In order to search for HAS motif in our ChIP-seq data , 200 bp regions around the centres of peak MSL1 abundance on the X-chromosome in roX mutants were analysed with the MEME program [97] using default parameters . Similar analyses were also performed for the top 200 MSL1-bound regions on the X-chromosome in the Ds wild type . In order to search for DNA motifs enriched in heterochromatic regions bound by MSL in roX mutants , 200 bp regions around the centres of peak MSL1 abundance on the heterochromatin scaffolds of each chromosome ( 2LHet , 2RHet , 3LHet , 3RHet , 4Het , XHet ) in roX mutants were analysed with the MEME program [97] together with scrambled sequences of binding sites as negative sequences , using default parameters . In order to analyse the repeat content in regions surrounding the expressed genes ( defined in [76] ) overlapping with MSL1 bound/unbound regions , repeat masked sequences were downloaded from UCSC [98] , [99] and the fractions of repeat masked nucleotides in 200 bp windows at 10 bp intervals across the genome were calculated . The percentages of these repeats in 20 bins of 1 kb around the Transcription Start Site ( TSS ) of MSL1 bound/unbound genes in roX mutants , on autosomes and on the X-chromosome were then calculated . Repeat percentages were also calculated around the TSS of all expressed genes of the X-chromosome and autosomes and around MSL2 bound/unbound expressed genes in the wild type , filtered using a 5 percent highest ratio cutoff on MSL2 enrichment ratio values . To search for repeat classes enriched in MSL-bound ChIP-seq reads , repeat classes in Dm available from the Repbase Update database ( release 19 . 01 ) [100] were used . Reads from wild type and roX mutant as well as the corresponding inputs were mapped to different repeat classes using the Bowtie software parameters –a ( to map all reads ) –v 2 ( with two mismatches ) [101] . For each repeat class , an RPKM value ( Reads Per Kilobase per Million mapped reads ) [102] was calculated which was used further to calculate a ratio between ChIP/input in wild type and roX mutants , respectively . The number of reads that mapped to the genome in the original ChIP-seq analysis was used as the number of mapped reads . In each repeat class , read counts per nucleotide was also calculated from wild type and roX mutant as well as input , and normalized to the number of mapped reads ( in millions ) from each sample . | In both fruit flies and humans , males and females have different sets of sex chromosomes . This generates differences in gene dosage that must be compensated for by adjusting the transcriptional output of most genes located on the X-chromosome . The specific recognition and targeting of the X-chromosome is essential for such dosage compensation . In fruit flies , dosage compensation is mediated by the male-specific lethal ( MSL ) complex , which upregulates gene transcription on the male X-chromosome . The MSL-complex consists of five proteins and two non-coding RNAs , roX1 and roX2 . While non-coding RNAs are known to be critical for dosage compensation in both flies and mammals , their precise functions remain elusive . Here we present a study on the targeting and function of the MSL-complex in the absence of roX RNAs . The results obtained suggest that the dosage compensating MSL-complex has an intrinsic tendency to target repeat-rich regions and that the function of roX RNAs is to prevent its binding to such targets . Our findings reveal an ancient targeting and regulatory function of the MSL-complex that has been adapted for use in dosage compensation and modified by the rapidly evolving noncoding roX RNAs . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"x",
"chromosome",
"inactivation",
"gene",
"regulation",
"histone",
"acetylation",
"euchromatin",
"histone",
"modification",
"epigenetics",
"molecular",
"genetics",
"chromatin",
"heterochromatin",
"chromosome",
"biology",
"proteins",
"gene",
"expression",
"biochemistry",
"cell",
"biology",
"post-translational",
"modification",
"acetylation",
"gene",
"regulatory",
"networks",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"dosage",
"compensation"
]
| 2014 | Non-coding roX RNAs Prevent the Binding of the MSL-complex to Heterochromatic Regions |
To determine the effect of helminth infection during pregnancy on the cognitive and motor functions of one-year-old children . Six hundred and thirty five singletons born to pregnant women enrolled before 29 weeks of gestation in a trial comparing two intermittent preventive treatments for malaria were assessed for cognitive and motor functions using the Mullen Scales of Early Learning , in the TOVI study , at twelve months of age in the district of Allada in Benin . Stool samples of pregnant women were collected at recruitment , second antenatal care ( ANC ) visit ( at least one month after recruitment ) and just before delivery , and were tested for helminths using the Kato-Katz technique . All pregnant women were administered a total of 600 mg of mebendazole ( 100 mg two times daily for 3 days ) to be taken after the first ANC visit . The intake was not directly observed . Prevalence of helminth infection was 11 . 5% , 7 . 5% and 3 . 0% at first ANC visit , second ANC visit and at delivery , respectively . Children of mothers who were infected with hookworms at the first ANC visit had 4 . 9 ( 95% CI: 1 . 3–8 . 6 ) lower mean gross motor scores compared to those whose mothers were not infected with hookworms at the first ANC visit , in the adjusted model . Helminth infection at least once during pregnancy was associated with infant cognitive and gross motor functions after adjusting for maternal education , gravidity , child sex , family possessions , and quality of the home stimulation . Helminth infection during pregnancy is associated with poor cognitive and gross motor outcomes in infants . Measures to prevent helminth infection during pregnancy should be reinforced .
Intestinal helminths infect more than two billion of the world’s population , with the highest prevalence in Asia and sub-Saharan Africa . [1] The burden of intestinal helminth infection is estimated to be five million disability-adjusted life years ( DALYs ) . [2] Helminth infections are rarely directly associated with increased mortality but are related to increased morbidity arising from the chronicity and consequences of infection . [3] Although the World Health Organization ( WHO ) highly recommends anthelmintic therapy for pregnant women in their second trimester[4] , the benefits on anemia , congenital anomalies and perinatal mortality remains unequivocal[5] . In sub-Saharan Africa , it is estimated that one-third of pregnant women are infected with soil-transmitted helminths[6] although several studies have shown wide variation in prevalence across different countries , 11 . 1% in Benin[7] , 25 . 7% in Ghana[8] and 49% in Gabon[9] . In Benin , anthelminthics are a component of the routine antenatal care ( ANC ) package given to pregnant women after their first trimester . [10] A recent systematic review found little evidence that deworming in children is associated with better cognitive function , though most trials included were of poor quality . [11] A cross-sectional study revealed that compared to 7 to 18 year-old-children who were not infected with Ascaris lumbricoides and Trichuris trichiura , children who were infected with either of these species of helminth performed poorly on tests of memory and verbal fluency , respectively . [12] Over the past decades , many studies have confirmed helminth infection during pregnancy as a risk factor for maternal iron deficiency ( ID ) and anemia[3 , 13 , 14] . However , evidence remains limited on the effects on adverse birth outcomes such as low birth weight ( LBW ) [15] which is known to be associated with poorer cognitive function in children . [16] Additionally , ID and anemia during pregnancy may be associated with poor cognitive function of infants as shown in a study in rural China which revealed that children of iron deficiency anemic ( IDA ) women performed significantly lower than those of non-IDA women in cognitive assessment tests . [17] The rapid rate of development of fetal organs makes them particularly susceptible to prenatal insults that are injurious to fetal development , and which could influence their development persisting even after birth . The early onset of delayed cognitive development could negatively influence several aspects of child development including preparedness for school . [18] Notwithstanding the evidence that helminths are associated with these indirect threats , very little is known about the impact of helminth infection during pregnancy on actual infant cognitive development . A study in Uganda concluded that Mansonella perstans and Strongyloides stercoralis infection during pregnancy may be associated with impaired executive function in children . [19] The objective of this study was to determine whether maternal infection with helminths , both in general and with specific helminth species , during pregnancy , is associated with cognitive and gross motor functions of one-year-old children in Benin .
Our prospective cohort included singletons born to pregnant women who were enrolled before 29 weeks of gestation in the Malaria in Pregnancy Preventive Alternative Drugs ( MiPPAD ) clinical trial ( NCT00811421 ) comparing sulfadoxine-pyrimethamine and mefloquine as intermittent preventive treatment of malaria in pregnancy ( IPTp ) . The study was conducted in the district of Allada in Benin . One thousand and five HIV-negative pregnant women attending their first ANC visit in the health centers in each of the three sub districts of Allada ( Sekou , Allada and Attogon ) were recruited . Detailed inclusion and exclusion criteria in the MiPPAD trial are explained elsewhere . [7] All live born children of recruited pregnant women who survived to 12 months were invited for neurocognitive assessment in the TOVI study ( Fon language: Tovi means Child from the country ) . We first described and compared the baseline characteristics of women with singleton live births whose children were assessed and those whose children were not assessed for cognitive function . Secondly , we performed univariate analyses to assess crude associations between the ELC and the gross motor scores with helminth infection , helminth species , helminth density , co-infection with malaria , and covariates [maternal prepregnancy body mass index ( BMI ) , family possessions , maternal occupation , education , the RPM and HOME scores] . These covariates were considered as potential confounding factors as they are known risk factors for poor cognitive development and may share common causes with helminth infection . Next , we conducted a multiple linear regression adjusting for covariates whose p-values were less than 0 . 20 in the univariate analysis . Finally , we performed stepwise removal of covariates from the model if they were found not be statistically significant . From the final model , we evaluated the adjusted mean difference in ELC and gross motor scores . Infant characteristics at birth or age one-year including birth weight , preterm birth and infant helminth infection were hypothesized to be within the causal pathway ( as mediators ) . All multivariate models were adjusted for infant sex . Although infant characteristics ( preterm births , low birth weight , and weight-for-age at MSEL assessment ) were hypothesized to be mediators in the association between prenatal helminth infection and infant cognitive function , we adjusted for these variables in a sensitivity analyses . Statistical analyses were conducted using Stata IC/11 . 2 for Windows ( StataCorp Lp , College station , TX ) . We used Pearson’s correlation to assess the associations between the dependent variables and other continuous variables . The student t-test , Wilcoxon rank sum test and chi-squared test were used to compare means , medians and proportions , respectively . Statistical significance was defined as p-value less than 0 . 05 . The study was approved by the institutional review boards of the University of Abomey-Calavi in Benin , New York University and Michigan State University in USA and the Research Institute for Development’s ( IRD ) Consultative Ethics Committee in France . At recruitment , we obtained written informed consent from all pregnant women and guardians of children who participated in this study in the presence of a witness . Women who could not read and write provided thumbprints to confirm their agreement to participate in the study after a nurse had explained the study .
As shown in Fig . 1 , 863 live born singletons were enrolled into the birth cohort but 35 died before the age of one year leaving 828 eligible children . Of these , 635 ( 76 . 7% ) were assessed for cognition using MSEL at approximately one year of age . The median age during MSEL assessments was 12 . 1 months ( range: 11 . 3–15 . 3 months ) . Two children were not able to complete all of the MSEL subtests , leaving 633 children who were fully assessed . Maternal baseline characteristics were similar between women whose children were fully assessed for cognitive function and those whose children were not , as shown in Table 1 . Also there was no significant difference between infant characteristics between children assessed and those not assessed . At first ANC visit , the prevalence of helminth infection was 11 . 5% of which hookworm infections were the most prevalent ( 9 . 5% ) . Of the 52 women with hookworm infections at the second ANC visit , 12 were infected with the same species at first ANC ( see Table 2 for prevalence and density of helminths ) . The prevalence ( 95% CI ) of helminth infection among children by age one was 32 . 8% ( 26 . 0%-39 . 6% ) . Maternal education , occupation , family possession , RPM and HOME scores and infant weight-for-age were associated with ELC and gross motor scores . Of note , maternal malarial infection was not statistically significantly associated with ELC and gross motor scores ( see Table 3 ) . Infant ELC and gross motor scores increased with increasing prepregnancy BMI class . As shown in Table 3 , children born preterm and those with low birth weight had lower ELC and gross motor scores , respectively . As shown in Table 4 , maternal occupation and educational status were associated with helminth infection at second ANC visits . Family possessions score was associated with helminth infection at both ANC visits . The difference in mean ELC scores between children whose mothers were infected with helminths at first ANC visit and those whose mothers were not infected with any helminth remained significant after adjusting for maternal education , child sex and HOME score ( p = 0 . 013 ) . Pregnant women who were infected with helminths at least , once during pregnancy had children with poorer ELC scores , thus-4 . 4 ( 95% CI: -7 . 2 to-1 . 5 ) compared to those of mothers who were never infected during pregnancy after adjustment ( see Table 5 ) . After adjusting for gravidity , maternal education , family possession , child sex and HOME score , helminth infection at first ANC visit was negatively associated with infant gross motor function ( p = 0 . 028 ) . We observed that mothers who were infected with hookworms during the first ANC visit had children who scored less in the gross motor scale , -4 . 9 ( 95% CI: -8 . 6 to-1 . 1 ) , compared to those whose mothers were never infected with hookworms at first ANC visit . With the exception of the association between gross motor scores and the occurrence of helminth infection over the course of pregnancy , sensitivity analyses performed by further adjusting for infant preterm status and weight-for-age , yielded similar results in the association between infants gross motor function and prenatal helminth infection . Helminth infection at second ANC was no longer statistically significantly associated with infant ELC scores after sensitivity analyses , p = 0 . 074 ( see Table 5 ) . Further adjustment for LBW ( not preterm birth ) and weight-for-age showed similar conclusions in the sensitivity analyses . We performed multiple regression analysis further adjusting for research nurses and found little difference in the results .
This study provides evidence of an association between intestinal helminths and hookworms among pregnant women and poor cognitive and gross motor functions in their children at approximately 12 months of age . In view of these findings and as recommended by the WHO , measures to prevent helminth infections should be reinforced . Further studies are needed to corroborate our findings and explain the pathophysiological mechanisms of this relationship . | The WHO recommends anthelmintics for pregnant women after their first trimester but the benefits remain unequivocal . Although the consequences of helminth infection during pregnancy on the health of pregnant women have been well studied , the impact on the early development of offspring has been understudied . Studies suggest that helminth infection in children may be associated with poor cognitive development , but very little is known whether a similar consequence exists for offspring of women infected with helminths during pregnancy . From our study , we found that women who had intestinal worm infection at least once during pregnancy , had children with lower cognitive and motor scores . Moreover , hookworm infection in pregnant women prior to receiving anthelminthic treatment was associated with poor gross motor functions of children at one-year of age . Based on the findings of this study , measures to prevent helminth infections during pregnancy should be reinforced . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| []
| 2015 | Impact of Helminth Infection during Pregnancy on Cognitive and Motor Functions of One-Year-Old Children |
Human heading perception based on optic flow is not only accurate , it is also remarkably robust and stable . These qualities are especially apparent when observers move through environments containing other moving objects , which introduce optic flow that is inconsistent with observer self-motion and therefore uninformative about heading direction . Moving objects may also occupy large portions of the visual field and occlude regions of the background optic flow that are most informative about heading perception . The fact that heading perception is biased by no more than a few degrees under such conditions attests to the robustness of the visual system and warrants further investigation . The aim of the present study was to investigate whether recurrent , competitive dynamics among MSTd neurons that serve to reduce uncertainty about heading over time offer a plausible mechanism for capturing the robustness of human heading perception . Simulations of existing heading models that do not contain competitive dynamics yield heading estimates that are far more erratic and unstable than human judgments . We present a dynamical model of primate visual areas V1 , MT , and MSTd based on that of Layton , Mingolla , and Browning that is similar to the other models , except that the model includes recurrent interactions among model MSTd neurons . Competitive dynamics stabilize the model’s heading estimate over time , even when a moving object crosses the future path . Soft winner-take-all dynamics enhance units that code a heading direction consistent with the time history and suppress responses to transient changes to the optic flow field . Our findings support recurrent competitive temporal dynamics as a crucial mechanism underlying the robustness and stability of perception of heading .
Heading perception is not only accurate , it is also remarkably robust and stable . These qualities warrant further investigation and are the focus of the present study . The robustness and stability of heading perception are especially evident in dynamic environments containing independently moving objects . Regions of the optic array corresponding to moving objects generally contain optic flow that is inconsistent with the background optic flow and uninformative about heading . Nonetheless , heading perception is biased by moving objects by no more than a few degrees . Objects that approach the observer in depth ( approaching objects; Fig 1A ) induce a bias in the direction opposite the object motion of ~3° [20] . When objects maintain a fixed depth with respect to the observer as they move laterally ( fixed-depth objects; Fig 1B ) , heading perception is biased by ~1° in the direction of object motion [21] . Objects that recede in depth from the observer as they move across the observer’s future path ( retreating objects; Fig 1C ) yield a heading bias in the direction of object motion of less than 3° ( Layton & Fajen , in preparation ) . These biases are surprisingly small when one considers the conditions in which they are induced . The moving objects in the aforementioned experiments were generally large and moved near the observer , such that a sizeable proportion of the visual field contained discrepant optical motion . They often crossed the observer’s future path , thereby occluding the region of the optic array near the background FoE ( as in Fig 1 ) , which is known to be the most informative region for heading perception [16 , 22] . Moving objects that approach the observer in depth generate a radial pattern of optic flow with a FoE of their own that may be offset from the background FoE by much more than a few degrees . In some circumstances , all of these potential complications may occur at the same time . In Fig 1D , for example , the moving object occupies approximately half of the visual field , occludes the background FoE ( indicated by the blue dot ) , and generates radial motion with a FoE ( red dot ) that is offset from the background FoE by 20° . The fact that heading perception under these conditions is biased by no more than a few degrees attests to the robustness of the visual system . Furthermore , such biases are induced only when objects cross or move near the observer’s future path [20 , 23] . Objects that move far away from the future path do not influence heading perception . Nonetheless , our experience when an object approaches and eventually crosses our future path is not that heading abruptly shifts; that is , humans do not perceive themselves as moving in one direction at one instant and then in a different direction at the next instant when the object begins to cross the path . Heading perception is more stable and less susceptible to fluctuations . Previous research on heading perception in the presence of moving objects [20 , 21 , 24] has focused on the sources of bias . The fact that heading perception is as reliable and stable as it is under such conditions has been largely overlooked . Nonetheless , these qualities of heading perception are worthy of investigation . Understanding the mechanisms that underlie the robustness and stability of heading perception was the primary aim of the present study . We test the hypothesis that the robustness and stability of heading perception is rooted in a particular form of temporal dynamics within the visual system—specifically , recurrent competitive interactions that unfold over time among units in area MSTd . As a simple demonstration that heading perception has an important temporal component , consider a scenario in which a moving object approaches and crosses the observer’s future path from the left or right . Recall that heading perception is biased under these conditions but that the bias is surprisingly small . One possible reason why the bias is not larger than it is , is that heading perception is based on the temporal evolution of the optic flow field , including not only the period of time when the object crossed the path but also prior to this point , before the more informative regions of the flow field were occluded by the object . Although this may seem obvious , existing models of heading perception [20 , 25] have no features for capturing heading perception as a process that evolves over time . As we explain below , these models estimate heading based on the instantaneous flow field and generate a new estimate that is independent of the previous one at each successive instant . We tested the role of temporal dynamics in an experiment in which human subjects made heading judgments in the presence of an object that approached from the side and crossed the observer’s path at the end of the trial [23] . Stimulus duration was varied between 150 and 1500 ms . Importantly , within each object trajectory condition , the last 150 ms was the same across stimulus durations , but conditions with longer durations also included the earlier part of the event leading up to the last 150 ms prior to occlusion of the background FoE . If heading perception is based on the temporal evolution of the optic flow field , the heading bias should be weaker in the longer duration conditions because the visual system should be able to use the information from the earlier part of the trial—before the object occluded the path—to improve the accuracy of the estimate . Indeed , when stimulus duration was short ( i . e . , when subjects only saw the last part of the trial ) , they exhibited a very large heading bias ( ~6° ) . However , the bias was dramatically reduced when stimulus duration was longer–that is , when the earlier part of the trial before the object crossed the path was also included in the stimulus . The findings indicate that heading perception is based on the evolution of the optic flow field and the ability to integrate information over time underlies the surprising accuracy and stability of heading perception . In other words , temporal dynamics offers a candidate solution to the problem posed above about why heading perception is not more biased by moving objects than it is , and why we do not experience abrupt transitions in perceived heading . These findings provide compelling evidence that heading perception is based on the evolution of the optic flow field , but are not especially informative about the nature of the underlying neural mechanisms . In the present study , we used modeling and simulation to test the sufficiency of a particular type of mechanism involving on-center/off-surround recurrent interactions among MSTd neurons . Neurons in a recurrent network send inhibitory feedback to other neurons to balance the excitatory feedback they send themselves to potentiate their activity . Recurrent dynamics among neurons in MSTd are compatible with the neuronal decay time constant in MSTd ( 81 msec ) , which is as much as five times slower than that of areas from which MSTd receives input and indicates a persistence in the heading signal long after the visual optic flow signal ceases [26] . Competition among heading-sensitive neurons in MSTd over time through recurrent interactions may exert nonlinear effects on the heading signal—the contrast of the heading signal could be enhanced and the uncertainty about heading reduced . In the language of Bayesian inference , MSTd neurons may update the network’s belief about heading over time [27] . As we demonstrate below , these interactions that unfold over time may serve as a mechanism to stabilize heading perception , even when the visual signal is temporarily disrupted . Our approach is to compare the performance of a model with recurrent temporal dynamics in MSTd against two models without this property , focusing on whether these models capture the spatio-temporal robustness of human heading perception in dynamic environments . The model with recurrent temporal dynamics is an updated version of the model introduced by Layton , Browning , & Mingolla [28] . The other two models are the motion pooling model developed by Warren & Saunders [20] and the differential motion model developed by Royden [25] . These models were chosen because they are representative of existing biological modeling approaches and because they were designed to estimate heading in the presence of independently moving objects . In all three models , the similarity is computed between the optic flow field ( or a transformation thereof ) and a number of vector field templates containing radial expansion with different FoE positions . Each template resembles the canonical receptive field organization of a MSTd cell selective to a particular FoE location in the visual field and is center-weighted—motion vectors nearby the preferred FoE position are weighted greater than those located further away . Such templates have characteristics that are consistent with those that develop in models using supervised and unsupervised learning [29–32] . Motion pooling models have demonstrated that matching these biologically inspired global motion templates with the patterns of optic flow that arise during self-motion provides a plausible means for cells in MSTd to extract heading [33–36] ( but see [37] ) . The template match feeds MSTd units with their inputs and the preferred FoE position of the most active unit reflects the heading estimate of the model . The differential motion model [25] is distinct from the others in that the template match is performed on a field of difference vectors rather than the optic flow field . Differential motion models were originally proposed to account for the ability to perceive heading while making eye movements , which introduce rotation into the flow field . Any instantaneous optic flow field can be decomposed into translational and rotational components [38] . A vector’s translational component depends on the corresponding point’s depth in the environment , whereas the rotational component does not . Therefore , subtracting nearby motion vectors that correspond to points at different depths within the environment eliminates the rotational component and results in a scaled version of the translational component . Because the translational component is informative about the observer’s heading and the rotational component is not , certain difference vectors may be used to recover a heading estimate . Rieger & Lawton [39] developed the first differential motion algorithm to compute heading based on a decomposition of translational and rotational flow when differential motion parallax is present . Hildreth [40] later extended the approach with a voting procedure to account for the presence of moving objects . A number of neural models that decompose flow into translational and rotational components have successfully simulated properties of MT and MST [41 , 42] ( but see [43] ) . The differential motion model developed by Royden and simulated in the present study [25] is a refinement of an earlier version [44] to situate the differential motion algorithm in a biological framework . The field of difference vectors in the Royden model is obtained by processing the optic flow field with motion sensitive units with antagonistic surrounds whose properties resemble those of cells in primate MT— . These operators respond optimally when a sharp change in speed occurs within the receptive field along the preferred motion direction , which may coincide with a sudden change in depth and result in motion parallax: near background motion results in a faster optical speeds than far background motion . The heading estimate in the differential motion model is the direction that corresponds to the preferred FoE of the most active center-weighted template . Both the motion pooling model [20] and the Layton et al . model ( as well as its successor introduced here ) compute the template match directly on the optic flow field . We refer to the latter as the competitive dynamics model to highlight its unique feature—that it is a dynamical system that continuously integrates optic flow within a competitive network of MSTd neurons [28] . This differs from the differential motion and motion pooling models , which only process vector fields at independent points of time . Two other models , neither of which can be classified as motion pooling or differential motion models , warrant mentioning . First , the analytic model of Raudies & Neumann [45] relies on neither motion differences nor templates to account for the pattern of human heading biases in the presence of moving objects , but rather a weighted combination of segmentation cues derived from the flow field . Heading bias arises in the model even without segmentation cues because the moving object induces a discrepancy compared to analytic parameters that describe the observer’s self-motion in a static environment . The pattern of bias produced by the model does not resemble that of humans , but segmenting the optic flow field by accretion/deletion , expansion/contraction , and acceleration/deceleration improves the correspondence . Second , Saunders & Niehorster [46] cast the problem of estimating heading in the presence of moving objects into a Bayesian context whereby the objective is to estimate the translational and rotational components of an ideal observer from optic flow along with the depth of points in the scene . The model estimates the posterior probability that an observer moves along a particular heading by multiplying the likelihoods that each motion vector in the optic flow pattern was independently generated by a particular combination of observer translation and rotation parameters . The model accounts for human heading bias in the presence of approaching and fixed-distance objects . We will not give further consideration to either model in our simulations below because both process vector fields at independent points of time and because our focus in the present study is on neural models . We simulated the differential motion , spatial pooling , and competitive dynamics models under a variety of conditions to test for robustness and stability in heading estimates ( see Methods section for details about the models and simulations ) . To anticipate the results , we found that the differential motion and spatial pooling models yield erratic , sometimes wildly fluctuating heading estimates over time . Furthermore , simply adding temporal smoothing of optic flow signals to these models does not capture the spatio-temporal characteristics of human heading perception . In contrast , the estimates from the competitive dynamics model are less biased by moving objects , less variable , more stable , and more similar to human heading estimates in the presence of moving objects . Taken together , the findings imply that competitive interactions within MSTd are a plausible mechanism to account for the robustness and stability of human heading perception .
Previous efforts to evaluate the differential motion and motion pooling models have focused on how accurately they reproduce patterns of human heading judgments [20 , 25 , 28] . Both models succeed in capturing the heading bias in humans for approaching objects , but only the differential motion model has been shown to match human judgments for fixed-depth objects . It has been argued that the motion pooling model fails to capture the human heading bias for fixed-depth objects [25 , 28] , but this has not actually been formally tested . Neither model has been evaluated in the retreating object scenario . More importantly , the robustness and stability of these models has never been systematically explored . In this section , we examine model estimates of heading during self-motion in the presence of moving objects that cross the observer’s future path while approaching ( Fig 2A and 2B ) , maintaining a fixed-depth ( Fig 2C ) , or retreating ( Fig 2D ) . The blue and gold curves in each plot show the mean heading error over time for the differential motion and motion pooling models , respectively , with lighter shaded regions indicating ±1 SE . The color of the horizontal bar at the top of each subplot in Fig 2 indicates when the moving object is crossing the observer’s future path ( red ) , as well as the portions of the trial before ( orange ) and after ( green ) crossing . First , we consider simulations with an approaching object that crosses the observer’s future path at a 15° angle ( 7 . 5° object FoE offset ) ( Fig 2A ) . The typical bias in human judgments in the presence of objects that approach at comparable angles is about 2 . 5° in the direction opposite object motion [23] . Note that subjects in the human experiments made judgments after viewing the entire stimulus , so the existing data are not informative about how perceived heading evolves over time as the object changes position in the visual field . As such , we represent typical human performance in Fig 2 using a single dot positioned at the far right of the figure with a dashed line of the same color for reference . The motion pooling model initially yields unbiased heading estimates but quickly exhibits a ~6° bias in the direction opposite of object motion as the object approaches and crosses the future path . This is consistent with the human data in direction but greater in magnitude . The heading bias remains relatively stable while the object occludes the background FoE . The differential motion model also yields a bias ( ~7° ) in the direction opposite object motion during object crossing , but the bias arises later . That is , the differential motion model yields accurate heading estimates for a longer period of time while the object occludes the heading direction compared to the motion pooling model . Although the heading biases from both models are only slightly greater than those exhibited by humans , model performance deviates from human judgments much more dramatically at larger object trajectory angles . When the object approaches along a 70° angle ( 35° object FoE offset ) , the differential motion model exhibits a bias that exceeds that of humans by a factor of 10 ( Fig 2B ) . The motion pooling model yields a large bias in the direction of object motion , followed by a dramatic reversal in the opposite direction . The large initial bias in the direction of object motion ( positive in Fig 2B ) was unexpected because the motion pooling model is known to exhibit heading bias in the opposite direction for approaching objects ( i . e . , toward the object FoE or negative in Fig 2A ) [20 , 24] . The positive bias arises from the strong rightward radial motion of the background flow ahead of the leading edge of the moving object , which activates templates weighted to the far right of actual heading . The level of activation is only moderate because the rightward flow is not a perfect match for templates in that direction . Nonetheless , the activation level is higher than for other templates , including those more closely aligned with the object FoE . This is because as the object draws closer in depth , the spatial distribution of dots nearby the object FoE becomes sparser . Radial templates weighted nearby the object FoE are only weakly activated because of the limited amount of motion . Eventually , as the object continues to cross the observer’s path , it occludes enough of the background flow to diminish activation of templates weighted to the far right . At this point , the most active templates are those closely aligned with the object FoE , causing the heading estimate to abruptly reverse and exhibit a bias in the direction opposite object motion . Although the direction of the bias generated by the motion pooling model was initially unexpected , the magnitude of bias in both models is not surprising given that both models estimate heading based on the instantaneous flow field , which is dominated by discrepant object motion toward the end of the trial . Nonetheless , human heading judgments are far more robust even when objects approach at larger angles [24] . Fig 2C shows the simulation results with a fixed-depth object . Both the differential motion model and motion pooling model yield unbiased or weakly biased heading estimate before and after the objects crosses the observer’s path . During the crossing period , the differential motion model produces a 1–2° heading bias in the direction of object motion , consistent with human heading judgments . On the other hand , the motion pooling model produces a weak bias in the direction opposite object motion , which is not consistent with human heading judgments . Variability is slightly larger in the differential motion model , except for the spike that occurs in the motion pooling model estimates when the object begins to cross the future path . Fig 2D depicts the simulation results for the retreating object scenario . The differential motion model yields accurate heading estimates until shortly before the object crosses the path , at which point there is a sharp rise in the bias . The 1–2° bias in the direction of object motion is consistent with the human data ( Layton & Fajen , in preparation ) . Model variability is comparable to that obtained for the fixed-depth object . The heading error generated by the motion pooling model gradually ramps up while the object is approaching the observer’s path , and then sharply reverses to a bias in the direction opposite object motion . The reversal and subsequent gradual bias reduction occur because templates in the model respond to the radial-like motion pattern created by the trailing edge of the object and the background . The most active template in the model tracks the position of trailing edge , which progressively moves toward the heading direction , resulting in a weakening of the bias . Although the mean estimates from the differential and motion pooling models match those of human observers in some conditions , the model and human estimates differ dramatically in direction and/or magnitude under other conditions . Furthermore , these models do not exhibit the stability that is characteristic of human heading perception . Heading estimates from both models often changed very abruptly , increasing or decreasing by many degrees of visual angle in less than 100 ms . If human heading perception was subject to such wild fluctuations , moving objects would induce easily noticeably shifts in perceived heading as they approach and cross the future path . Yet both psychophysical studies and introspection while driving or walking in busy environments suggest that heading perception is far more stable and that any changes in perceived heading are small and too gradual to be noticed . Because the differential motion and motion pooling models were not designed to integrate optic flow over time , we explored whether smoothing the activation of model MSTd units over time with a moving average would address some of the issues with the stability of heading estimates . This was implemented by applying a 3 ( ‘Low’ ) , 6 ( ‘Med’ ) , or 9 ( ‘Hi’ ) frame moving average to the activation produced by each unit in the 2D MSTd array . As illustrated in Fig 3 , temporal smoothing was not effective . In both models , the large heading biases , abrupt changes , and reversals remained even with a high degree of temporal smoothing . Of course , the degree of smoothing could be further increased , but that would not qualitatively change the estimates and would introduce significant lag into the signal , making the model sluggish in response to actual changes in heading . We now introduce the competitive dynamics model , which is based on the model of Layton et al . [28] , and explore whether it better captures the robustness and stability of human heading perception . The model contains areas that correspond to the primate retina , lateral geniculate nucleus ( LGN ) , primary visual cortex ( V1 ) , medial temporal area ( MT+ ) , and the dorsal medial superior temporal area ( MSTd ) ( see Methods for details ) . These areas are organized into three functionally distinct stages: sensitivity to change in the retina; motion detection in LGN , V1 , and MT+; and self-motion estimation in MSTd . The self-motion estimation mechanisms are the same as those in the model of Layton et al . [28] , but the stages for sensitivity to change and motion detection are new . As a large dynamical system , populations of neural units in each area obey systems of Hodgkin-Huxley-like ordinary differential equations . In other words , model cells in each area temporally integrate the network response to the optic flow time history and the bottom-up signal derived from the presently available optic flow . The main feature that differentiates the competitive dynamics model from the differential motion and motion pooling models is the use of recurrent competitive dynamics among model MSTd cells that unfold over time . Units in model MSTd obey on-center/off-surround recurrent competitive dynamics , which refine the heading estimate over time . Each unit competes for its heading representation: a unit enhances the signal for its preferred heading through self-excitation and suppresses other heading signals generated through the pattern of activity produced by other units in the network . As a dynamical system , the network takes time to develop a reliable heading estimate , which persists for some time—even if the optic flow signal is interrupted . The competitive dynamics that refine the heading estimate and the persistence of activity over time in the competitive dynamics model may hold the key to robust heading perception . To test the performance of the model , we ran simulations under the same conditions used to test the previous models . In Fig 4 , we plot the mean heading error produced by the competitive dynamics model for the approaching , fixed-depth , and retreating object conditions . The mean heading bias reaches ~2 . 5° in the direction opposite object motion for the object approaching along a 15° angle , ~4° in the same direction for an object approaching along a 70° angle , ~2 . 5° in the direction of object motion for the fixed-depth object , and ~4° in the direction of object motion for the retreating object . Like the differential motion model , the competitive dynamics model yields heading biases that consistently match those of human observers in direction . Unlike the differential motion and motion pooling models , however , heading error builds up in each condition over several hundred milliseconds and does not exhibit large , sudden excursions or reversals in the direction of bias . Furthermore , the competitive dynamics model yields far more accurate , stable , and human-like heading estimates when objects approach at extreme angles ( e . g . , 70° ) compared to the differential motion and motion pooling models ( compare Figs 4B and 2B , noting the difference in the scale of the y-axis ) . In summary , the competitive dynamics model better replicates the complex pattern of human heading biases across variations in object trajectories , and yields heading estimates that are relatively stable and change gradually . Before we can conclude that the improvement in performance is due to recurrent competition , it is necessary to rule out other differences between the competitive dynamics model and the differential motion and motion pooling models . As such , we simulated a version of the competitive dynamics model without connectivity between model MSTd units . The thick fuchsia curves in Fig 5 depict the model performance on the approaching , fixed-depth , and retreating objects simulations when we lesioned connections within model MSTd . Without recurrent competition , heading estimates change more abruptly , are strongly influenced by the object FoE in the presence of the objects that approach along 15° and 70° trajectories , and are in the incorrect direction in the presence of the fixed-depth and retreating objects . Together , these findings support the hypothesis that competitive interactions in MSTd play a crucial role in the robustness of human heading perception . Our second additional test of robustness and stability uses video rather than analytic optic flow as the input signal . Flow detected from the video is inherently noisier and sparser than that that is analytically specified . Since decreases in the optic flow density do not weaken the influence of moving objects on heading judgments , this serves as an important test of model robustness . For this set of simulations , the object approached from a 35° angle and the motion gave rise to a pseudo FoE when the object occluded the future path at the end of the trial . This differs from the stimuli used in the previous pseudo FoE simulations wherein the object crossed the future path earlier in the trial . The object was cylindrical and traveled along a ground plane , which resembled the condition from Experiment 3 of Layton & Fajen [24] when the object was laterally shifted by 0 . 75 m . The estimates from the differential motion model fluctuate about zero before the moving object crosses the observer’s path , and then turn sharply negative ( Fig 8 ) . The motion pooling model yields smaller biases , but also shows fluctuations throughout the trial . In contrast , mean heading error for the competitive dynamics model is close to zero until the object crosses the path , at which point a small negative error ( as in human performance ) emerges . These results highlight the robustness of recurrent competition that unfolds over time . Even though the density and position of the dot motion may vary considerably frame-to-frame in the optic flow detected from video ( yielding changing global motion patterns ) , on-center/off-surround competition in the competitive dynamics model stabilizes the heading estimates . Human heading perception is remarkably stable even when vision is temporarily interrupted , such as during eye blinks , or in the more extreme scenarios when the entire optic flow field suddenly changes , such as when large moving objects occupy most of the visual field ( e . g . a train crossing a driver’s future path ) . To test how extreme perturbations to the optic flow field affect model heading estimates , we replaced optic flow of simulated self-motion through a static environment with full-field laminar flow for 1 , 2 , 5 , or 10 contiguous frames ( Fig 9 ) . We did not plot the results for the motion pooling model because heading estimates were accurate , except for during the period of laminar flow when the maximal activation shifted instantaneously from the central to most laminar template . We could not simulate the differential motion model , because the laminar flow did not contain depth variation . Fig 9 shows how full-field laminar flow perturbations affect the absolute heading error produced by the competitive dynamics model . For this set of simulations , we show individual trials ( one per condition ) rather than averages across multiple trials . The duration of the laminar flow perturbation had a graded effect on the maximal absolute heading error–in general , longer laminar flow perturbations yielded larger heading errors . Heading estimates restabilized in each case , but the model required longer periods of time to recover from the longer perturbations . This occurred because the recurrent mechanisms in the competitive dynamics model not only integrate the presently available optic flow , but also the response to the optic flow time history . This leads to the prediction that misperceptions in heading that may result from prolonged extreme disruptions to the optic flow field endure for a longer period of time than those caused by shorter disruptions . In general , however , the mechanisms in competitive dynamics model tolerated even the most extreme optic flow disruptions , greatly mitigating heading errors as compared to the motion pooling model . The simulations presented thus far have focused on self-motion in the presence of moving objects . However , the stability of heading also improves over time during self-motion through static environments . This was demonstrated by Layton & Fajen [23] , who showed that the variability in human heading judgments was affected by the duration of the stimulus , with the greatest variability occurring when trial duration was short ( 150 ms ) and variability decreasing and eventually reaching a plateau at 500 ms . Fig 10 depicts the variance of the model MSTd population across the template-space , which indicates the quality of the heading estimate , over 1 sec of simulated self-motion through a static environment . Similar to the variability in human heading judgments , the population variance is high early in the trial , indicating uncertainty in the heading estimate due to the broadness of the MSTd activity distribution . Also like human judgments , model variance decreases over time and plateaus at approximately 500 msec . These simulations suggest that even in static environments , recurrent mechanisms within MSTd , such as those in the competitive dynamics model , may play an important role in refining and stabilizing heading estimates over time .
The primary aim of this study was not merely to introduce a new model that generates more accurate heading estimates , but rather to examine the general principles that underlie robust and stable heading perception . The most obvious difference between the competitive dynamics model and the motion pooling and differential motion models is the introduction of temporal dynamics . However , this feature of the competitive dynamics model does not merely smooth the heading estimate over time . As demonstrated above , simply adding temporal smoothing to the motion pooling and differential motion models barely reduces the fluctuations in heading estimates . Rather , the competitive dynamics model integrates a collection of neural mechanisms ( i . e . , divisive normalization , on-center/off-surround interactions , recurrent connectivity , and thresholding ) that are fundamental , ubiquitous operations performed by populations of neurons all throughout cortex [47–50] . In this section , we explain how these mechanisms offer a more principled account of the robustness and stability of human heading perception . The MSTd network in the competitive dynamics model implements a recurrent competitive field , wherein neural interactions lead to the divisive normalization of the population activity [51] . Divisive normalization represents an important property whereby the total energy in the network is conserved and the dynamic range automatically adjusts to process the input pattern , irrespective of gain fluctuations [52] . Automatic gain control plays a particularly important role in stabilizing network dynamics in a number of scenarios , such as in the pseudo FoE video simulation , wherein the reliability of the optic flow signal may vary considerably over time . In model MSTd , units compete via on-center/off-surround interactions for their preferred heading direction , determined by the FoE position of their receptive field template . Each unit enhances its own heading signal through recurrent self-excitation and inhibits competing heading signals through recurrent inhibition ( Eq 21 ) . Crucially , the estimated heading arises through divisive or shunting interactions that unfold over time to balance the MSTd population activity and the continuously evolving bottom-up optic flow signals . These recurrent interactions result in soft winner-take-all dynamics within model MSTd and the simultaneous representation of multiple possible heading estimates at any point in time . In both the 70° approaching object ( Fig 4B ) and pseudo FoE simulations ( Fig 7A ) , soft winner-take-all dynamics led to candidate heading estimates in multiple locations . This allowed the competitive dynamics model to balance , rather than abruptly switch between , estimates based on their salience over time . A similar mechanism allowed Layton & Browning [53] to capture the effects of spatial attention on the tuning of MSTd neurons . While others have proposed models that depend on recurrent mechanisms in MSTd [54–56] , there are important differences between these models and the competitive dynamics model . First , activity normalization and feedforward/feedback signal integration occur within the competitive dynamics model as a single process , tightly linked to the network dynamics . By contrast , the other models decompose these operations into three separate stages within each model area ( e . g . MSTd ) . Second , the continuous integration of optic flow over time represents a distinctive property only captured by the competitive dynamics model . Computations performed by other models of MSTd with recurrent mechanisms occur at equilibrium , when units reach a steady state at each point in time . The competitive dynamics model is a dynamical system that is agnostic as to whether a steady state is ever reached: units respond to both the changes in the optic flow input and the evolving state of interacting V1 , MT+ and MSTd networks . This exclusive property of the competitive dynamics model is consistent with how MT relies on a temporal solution to overcome the aperture problem [57] and how human heading perception depends on the temporal evolution of the optic flow field [23] . The fact that optic flow acceleration/deceleration played an important role in accounting for human judgments in the analytical model of Raudies & Neumann [45] strengthens the evidence that the temporal evolution of the optic flow field plays an important role in heading perception . Returning to the question of why heading perception does not abruptly shift when an object crosses the observer’s future path , the explanation provided by the competitive dynamics model is that the competitive dynamics among MSTd neurons take time to unfold . Optic flow signals arriving at the present time interact with the present state of the network that reflects the information about the heading direction detected over the course of the recent time history . For example , when many MSTd units are simultaneously active , there is greater uncertainty about the heading direction across the population and a reliable optic flow signal may quickly influence the heading estimate ( Fig 10 ) . On the other hand , if few MSTd units are active , meaning there is a high degree of confidence in the heading estimate across the population , it may take some time for even a strong distinct optic flow signal to influence the heading estimate . This property contrasts with the behavior of other models whereby the activation of model MSTd always reflects an instantaneous transformation of the optic flow field . The time course of competitive dynamics likely interacts with the relatively slow response latencies ( median: ~190 msec ) of MSTd neurons to radial expansion [58 , 59] . The fact that the competitive dynamics model captures human heading perception so well highlights the importance of these competitive mechanisms . The motion pooling and differential motion models implement algorithms that are sufficiently generic that they could in principle be carried out on vector field representations of the optic flow field without regard to neural systems . For example , the motion differencing operations performed by the differential motion model could just as well be carried out on raw optic flow vectors without the interpretation that MT units perform the operation . On the other hand , the competitive dynamics model explicitly models individual neurons and their interactions in dynamically evolving networks—the computations and neural interpretation are inextricably linked . We emphasize that the upshot is not that the competitive dynamics model is superior to the other models , but that the competitive mechanisms that implement more realistic neural dynamics play a central role in the robustness and stability of heading perception . To account for human heading judgments in the presence of moving objects , the differential motion model segments the optic flow field on the basis of local speed differences [25] and the model of Raudies & Neumann [45] segments on the basis of accretion/deletion , expansion/contraction , and acceleration/deceleration . This raises the question of whether the visual system requires segmentation to perceive heading in the presence of moving objects . The fact that the competitive dynamics model captures patterns of known human heading judgments without any explicit segmentation of the optic flow field , shows that , at least in principle , segmentation may not be necessary and recurrent mechanisms in MSTd are sufficient . While segmentation likely plays a fundamental role in object motion perception [60] , several lines of evidence do not support a role in heading perception . First , neurons in MSTd do not appear to extract the translational component when the optic flow field contains both translation and rotation [43] , which is a core prediction of differential motion models [25 , 39 , 40 , 44] . Second , the MT cells that possess antagonistic surrounds that are proposed by Royden to perform the differential motion computations and that could be used to extract the accretion/deletion segmentation cue used by Raudies & Neumann do not appear to project to heading sensitive cells in MSTd [61–64] . Third , sensitivity to several of the “segmentation cues” of Raudies & Neumann may be achieved without explicit segmentation stages . Radial and spiral templates that realize the properties of MSTd receptive fields extract information from optic flow about expansion/contraction and spatial curvature , respectively . In addition , the competitive dynamics model extracts information about acceleration/deceleration and temporal curvature by integrating the optic flow field over time . Although our focus has been on visual mechanisms that underlie robust heading perception , self-motion perception is inherently multi-sensory . The majority of heading sensitive cells in primate MSTd are primarily driven by visual input but modulated by vestibular signals [65] . That is , the heading tuning curve of MSTd neurons tends to be sharper and more selective when self-motion occurs in the presence of optic flow compared to in darkness [66] . The multi-sensory tuning of MSTd neurons may contribute to the robustness of heading perception during ordinary locomotion in a manner that goes beyond the visual mechanisms explored in existing models . For example , as many as half of MSTd neurons demonstrate an enhanced response when the visual and vestibular signals are consistent with one another ( “congruent cells” ) and others demonstrate a diminished response when the multi-sensory signals are in conflict ( “opposite cells” ) [67] . The multimodal response differences between these populations of MSTd neurons may increase the robustness of heading perception by discounting optic flow , such as that produced by a moving object , that does not agree with non-visual self-motion directions [68] . Despite the coarseness of vestibular tuning in MSTd , it may be sufficient to resolve whether a region of optic flow within the visual field arises due to self- or object motion [10 , 69] and allow heading perception to persist when optic flow is intermittently unavailable or unreliable . The availability of proprioception during active self-motion likely provides another redundant signal to facilitate heading perception [4] . More neurophysiological and modeling work needs to be performed to clarify how multisensory signals interact to contribute to the robustness of heading perception . In the present article , we simulated biological models of heading perception to investigate why perceived heading in humans is only biased by several degrees in the presence of moving objects and why the perceived heading does not abruptly shift when the object crosses the observer’s future path . We found that passive temporal smoothing alone was not sufficient in accounting for the characteristic robustness of human heading perception . However , recurrent competitive interactions that unfold over time among model units in area MSTd resulted in stable heading estimates .
We generated 1 . 5 sec ( 45 frame ) sequences of optic flow that simulated self-motion toward two frontoparallel planes initially positioned 800 cm and 1000 cm away in depth from the observer . This environment was selected to accommodate the differential motion model , which performs best when the scene contains depth discontinuities . Each plane consisted of 3000 dots and occupied the entire visual field at the outset . The observer moved along a straight-ahead heading at 200 cm/sec . In sequences that contained a moving object , the object was rectangular ( 150 cm x 150 cm ) and consisted of 320 dots . Its trajectory was parameterized in terms of a starting lateral offset from the observer’s path , starting relative depth to the observer , speed , and heading-relative trajectory angle . Table 1 specifies the parameters of object trajectories . In the video simulation , we detected optic flow in the video ( 320 px x 240 px ) using the Horn-Schunk algorithm built into MATLAB’s computer vision toolbox , which served as the input to the differential motion and motion pooling models [70] . The competitive dynamics model detected motion from the video directly . Note that the Horn-Schunk optimization includes a regularization term in the objective function that smoothes the estimated motion field . Compared to correlational [71] or motion energy [72] approaches to motion detection , which do not require smoothing of the motion vector field , the Horn-Schunk algorithm should generate motion and heading estimates in motion pooling and differential motion models that are no less stable and accurate . To ensure that the variability is not due to an insufficient sample size , we ran each of the models on a number ( usually 25 ) of the optic flow sequences that were identical except for the initial random placement of the dots until the variance in the heading estimates plateaued . We also simulated all three models in a static environment ( no moving object ) to confirm that they were operating as expected and that there were no systematic biases . These simulations revealed a mean heading error close to zero with low variability for all three models . The differential motion [25] and motion pooling [20] models were implemented in MATLAB according to their published specifications . We changed several parameters to ensure that the models performed as expected on our visual displays . For example , we set the model MSTd Gaussian pooling variance parameter ( σ ) to 19 px in the model of Royden ( 2002 ) to ensure that heading estimates were in the direction of object motion for the fixed-depth object and in the direction opposite object motion for the approaching object . We changed the same parameter in the model of Warren & Saunders [20] to 25 px to achieve the best performance across the visual displays that we tested . Model MT in both models consisted of units that tiled the input dimensions of the visual displays . To generate the opponent operators with different differencing axes in the model MT of Royden [25] , we filtered the input with 7x7 rectified Gabor kernels . MSTd units had overlapping receptive fields and were centered every two pixels along each of the input dimensions . Parameter values remained the same in all models and simulations . Analytical optic flow computed by a pinhole camera projection served as input to the differential motion and motion pooling models [38] . We derived model heading estimates by considering the location of preferred FoE of the maximally active MSTd unit along the horizontal cross-section that contained the observer’s heading . Heading error was computed by subtracting the location of the preferred FoE of the maximally active MSTd unit from that which coincides with the observer’s heading direction . The model presented here encompasses the V1-MT+-MSTd processing stages of the Layton et al . model [28] . Updates have been performed to the front-end so that the model detects optic flow from video input using stages that correspond to those along the primate magnocellular pathway . Moreover , algorithmic simplifications used in the Layton et al . model have been replaced so that each stage consists of networks of coupled Hodgkin Huxley type ordinary dynamical equations . Our model builds on the STARS and ViSTARS models [73–75] . Fig 11 schematically depicts an overview of the model . The model consists of three main stages: detecting changes in luminance ( model Retina and LGN ) , detecting motion ( model V1 and MT+ ) , and estimating self-motion ( model MSTd ) . The details of these stages are described in the following sections . Fig 12 shows the response of each model area to simulated self-motion in a static environment toward two frontoparallel planes . The ordinary differential equations described in the following sections model the dynamics of cells across multiple brain areas along the magnocellular pathway of primate cortex . Equations often assume the form of a recurrent competitive field [51]: dxidt=−xi+ ( 1−xi ) ( f ( xi ) +Ii+ ) −xi ( ∑k≠if ( xk ) +Ii− ) ( 1 ) The firing rate x of unit i in the network layer described by Eq 1 obeys shunting dynamics , which implement a number of important dynamical properties , such as divisive normalization and boundedness [51 , 52 , 76] . Eq 1 contains a passive decay term −xi , excitatory input term ( f ( xi ) +Ii+ ) that is shunted by ( 1 − xi ) to ensure the firing rate remains bounded above by 1 , and inhibitory input term ( ∑k≠if ( xi ) +Ii− ) that is shunted by xi to bound the firing rate from below by 0 . The variables Ii+ and Ii− denote excitatory and inhibitory inputs to unit i , respectively . In Eq 1 , the function f ( xi ) regulates the recurrent self-excitation or inhibition that the unit receives from others in the same network . In model networks , f is a sigmoid function that gives rise of soft winner-take-all dynamics [51 , 77] . We use the sigmoid f ( w;f0 ) =w2w2+f02 ( 2 ) that achieves its half-maximum value of 0 . 5 at f0 . The output of network layers may be thresholded by Γ according to the following function g g ( w;Γ ) =[w−Γ]+ ( 3 ) where the notation [∙]+ indicates the half-wave rectification max ( ∙ , 0 ) . Network equations apply to all units in the layer and as such we use matrix bold notation . For example , x denotes the array of cells at each spatial location ( i , j ) . Connectivity between units in different network layers are either connected 1-to-1 or through a Gaussian kernel that defines the convergence of feedforward or feedback signals . When units are connected 1-to-1 , the unit with a receptive field centered at position ( i , j ) projects to a unit in another layer at position ( i , j ) . The Gaussian kernel Gσ , s defines how connections converge from one layer onto the next when the spatial extent of the receptive field is larger than that of its input . In Eq 4 , the operator ∙ defines the dot product , σ indicates the standard deviation , and s defines the radius of the kernel . Convolutions in the following sections are always centered at the position of each unit ( i , j ) . In some cases , the Gaussian kernel that has radius s and is elongated in the direction d corresponding to the angle θ , which we define as Wσ¯ , s , d . In Eq 5 , ∇ is the rotation matrix ∇=[cosθsinθ−sinθcosθ] , θ=d4 , Σ=[σx00σy] , and σ¯=σx/σy . The kernel Wσ¯ , s , d is normalized to sum to unity . Units in areas LGN and V1 uniformly tiled 1-to-1 the spatial dimensions of the visual display . The overlapping receptive fields of MT+ and MSTd units were spaced according to their speed sensitivity . Units in MT+ tuned to speeds of 1 , 2 , and 3 px/frame had receptive fields uniformly distributed throughout the visual input array at every single , second , and third pixel , respectively . MSTd units had receptive fields centered every 6 px , doubling the maximum offset of MT+ . The pattern of luminance at time t in the visual signal I ( t ) is transformed into signals of increments J+ ( t ) and decrements J− ( t ) , which represent the change in the input across successive frames . These signals correspond to the coding of luminance increases and decrease by ON and OFF retinal ganglion cells . The notation ⌈∙⌉ refers to taking the ceiling of the operand . Model LGN units L± ( t ) respond to transient changes in the visual signal , but are not direction selective [78] . ON and OFF LGN units remain sensitive to the luminance increments and decrements in their retinal inputs , respectively [79] . The following equation describes the activity of LGN units: L± ( t ) =[R± ( t ) Z± ( t ) ]+ ( 8 ) where R± indicates the population of units that perform a leaky integration of its retinal inputs , which is gated by the habituative transmitter Z± . Habituative gates , sometimes called dynamic or depressing synapses , prevent responses to persistent inputs by modeling the slow-term deletion of a neuron’s neurotransmitter stores [80 , 81] . Tonic inputs depress the habituative gates , which when multiplicatively combined with R± , lead to rapid suppression of the LGN activity L± . Habituative gates slowly recover to their full capacity in the absence of input . In effect , L± responds well to motion and weakly to stationary inputs . In Eqs 9 and 10 , ϵLGN , R corresponds to the inverse time constant of each cell R± , ϵLGN , Z corresponds to the inverse time constant of each gate Z± , and λ indicates the transmitter depletion/repletion rate . For our simulations , we fixed ϵLGN , R = 2 sec−1 , ϵLGN , Z = 0 . 01 sec−1 , and λ = 10 . The detection of motion direction occurs through a three stage process that corresponds to simple and complex cells in V1 , and cells in area MT+ with excitatory surrounds ( Fig 13 ) . First , motion is detected by simple cells using a Reichardt or correlation-based mechanism based on the arrival of signals from LGN with different conduction delays and receptive field locations [71] ( but see [73 , 74 , 82–84] for an alternative biological mechanism that relies on nulling inhibition ) . The motion signal is refined through short-range feedforward on-center/off-surround pooling of simple cell activity by complex cells ( Fig 13 , bottom two panels ) . Finally , a feedback loop between V1 complex cells and MT+ cells disambiguates local motion signals ( i . e . solves the aperture problem ) through the spatial pooling of complex cells by units in MT+ tuned to the same motion direction and the suppression of complex cells tuned to dissimilar motion directions ( Fig 13 , top two panels ) . An asterisk appears adjacent to the inhibitory feedback connections from MT to V1 in Fig 13 because the model proposes that the net effect on V1 complex cells is inhibitory , not that the individual feedback projections are inhibitory . In fact , feedback projections to V1 are likely mostly excitatory and target excitatory neurons [85] . MT feedback projections target several V1 layers , including layer 6 [86] , and neurons therein project to inhibitory interneurons [87] and those involved in feedforward processing [88] in layer 4C . Given that layer 4 neurons project to layer 2/3 [89] , which contains complex cells , the circuit may serve a modularity role on complex cells . Even though the MT-V1 feedback projections may be excitatory , they may exert an inhibitory effect on complex cells , consistent with predictions from the competitive dynamics model . Modeling the laminar microcircuitry of V1 extends beyond the aims of the present paper , so we use inhibitory MT feedback as a simplification ( see [90] for a laminar model of V1 ) . The feedback loop between MT and V1 proposed by Bayerl & Neumann [91] bears some similarity to the one described here , but , contrary to the predictions of the competitive dynamics model , the feedback exerts an excitatory rather than an inhibitory influence on V1 units . Feedback signals from MT that suppress complex cells sensitive to dissimilar motion directions rather than enhance complex cells sensitive to similar motion directions prevent the biologically implausible scenario of a runaway positive inter-areal feedback loop between V1 and MT and are consistent with the reduced suppression in V1 neurons following MT inactivation [92 , 93] . In the following sections , we describe the details of model areas V1 , MT and MSTd . Signals from model LGN with spatially displaced receptive field centers ( Δ→ ) that possess a spectrum of conduction delays ( δ→ ) converge onto V1 [94] . In other words , each V1 unit acquires its speed and direction tuning based on the spatiotemporal correlation present in its convergent afferent LGN signals . For simplicity , we consider the two conduction delays δ0 and δ1 that implicate motion detection across successive frames: δ0 = ⌊t⌋ and δ1 = ⌊t − 1⌋ , where ⌊∙⌋ indicates taking the floor of the operand . V1 units were tuned to speeds ( s ) of 1 , 2 , or 3 px/frame in the 8 cardinal directions ( d ) : up , down , left , right , and the four diagonals . The speed-direction tuning was determined by summing the LGN signal centered at position ( i , j ) derived from the present time with the delayed signal centered at position ( i − Δx , j − Δy ) derived from the optic flow on the previous frame . We use the set Φ ( s , d ) to refer to the nonzero spatial offsets from position ( i , j ) that implicate motion with speed s in the direction d . For example , a V1 simple cell tuned to unit speed motion in the rightward direction would have an offset Δ→∈Φ ( 1 , 1 ) , where Δ→= ( −1 , 0 ) . In other words , comparing the delayed signal displaced one unit to the left to the present signal yields sensitivity to rightward motion . To account for the increased distance along the diagonals , we scaled these signals by a factor of s . To ensure that the V1 unit receives input from the most correlated input ( Eq 10 ) , the signal is thresholded with ΓLGN = 0 . 45 and squared [48 , 72] . The set of spatial offsets |Δ→∈Φ ( s , d ) | grows with speed , but not all offsets factor into the output signal , so we normalized by the number of active contributions ( 1|As , d±>0| ) ( Eq 11 ) . This adaptive normalization approximates the homeostatic plasticity process known as synaptic scaling [95] . The bottom-up input B± to V1 is computed based on the afferent LGN signals L± according to the following equations: As , d , i , j± ( t , Δ→ ) =[Li , j± ( δ0 ) +Li−Δx , j−Δy± ( δ1 ) −ΓLGN]+2 ( 11 ) Bs , d , i , j± ( t ) =1|As , d±>0|∑Δ→∈Φ ( s , d ) As , d , i , j± ( t , Δ→ ) ( 12 ) Simple cells in model V1 perform a leaky integration of their adaptive spatiotemporal inputs dSs , d±dt=ϵS ( −Ss , d±+ ( 1−Ss , d± ) Bs , d± ) ( 13 ) where the units have an inverse time constant ϵS = 5 sec−1 . Fig 12B shows the pattern of activity of direction selective simple cells tuned to unit speed at the end of a self-motion simulation through a static environment . Model complex cells are tuned to a direction and speed , but unlike simple cells , are insensitive to contrast polarity . Complex cell units refine and enhance their directional selectivity by locally pooling over simple cells with the same direction and speed tuning parallel to the preferred motion direction ( Fig 13 ) . The units also receive shunting inhibition from nearby simple cells whose receptive fields are positioned in the orthogonal direction . This feedforward connectivity between model simple and complex cells enhances responses to uniformly moving dots or surfaces and implements the collinear facilitation and orthogonal suppression properties of V1 complex cells [96 , 97] . V1 complex cell units compete across preferred motion direction in the following contrast enhancing network: Qs , d=g ( Ss , d++Ss , d− , ΓC ) 2 ( 14 ) dCs , ddt=−Cs , d+ ( 1−Cs , d ) ( Cs , d2+∑ ( I , J ) ( Wσ¯MT , s , θ∥d , i−I , j−JQs , d , I , J ) ) −Cs , d ( ∑k≠d ( Cs , k2+Ys , k ) +∑ ( I , J ) ( Wσ¯MT , s , θ⊥d , i−I , j−JQs , d , I , J ) ) . ( 15 ) In Eq 15 , Wσ¯MT , s , d is the anisotropic Gaussian kernel elongated in the direction d ( σ¯MT=15 ) , the notation θ ∥ d means the angle θ of the preferred motion direction , the notation θ ⊥ d means the angle θ+π2 orthogonal to the preferred motion direction , and Y defines the feedback a complex cell receives from area MT+ ( see Eq 18 ) : Ys , d=∑ ( I , J ) ( GσV1 , 3s , i−I , j−JMs , d , I , J ) ( 16 ) where σV1 = 0 . 5 . Complex cells receive inhibitory feedback from model area MT+ that suppresses the activity of units with preferred directions k that differ from the complex cell preferred direction d . Note that each matrix Ms , d has been scaled back into the coordinate system of the visual display before performing the convolution . Fig 12C depicts the pattern of activity of complex cells tuned to unit speed at the end of the self-motion simulation . The complex cell output from V1 Os , d is thresholded by Γv1mt = 0 . 01 , squared , and transformed by the sigmoid function f ( ∙;γv1mt ) , with an inflection point γv1mt = 0 . 01 . Units in model MT+ perform a long-range spatial pooling over complex cell signals . These cells inherit their directional tuning from V1 complex cells , but integrate motion within their larger receptive fields . Analogous to V1 complex cells , the receptive field of units in MT+ is elongated in the direction parallel to preferred motion direction [97] . Units in MT+ respond well when complex cells with a consistent directional tuning are active within the receptive field ( Fig 13 , top panel ) . Over time , the suppressive feedback loop between V1 and MT+ resolves ambiguity in the detected motion direction ( Fig 13 , top two panels ) . Fig 12D depicts the MT+ cells tuned to unit smallest scale at the end of the self-motion simulation . The output signal Nd from MT+ collapses across speed , after scaling each matrix Ms , d into the coordinates of MSTd . Units are collinearly pooled , parallel to their preferred motion direction , thresholded with ΓMT = 10−3 , and subsequently squared: Nd=g ( ∑ ( I , J ) ( Wσ¯MT , 2 , θ∥d , i−I , j−J∑sMs , d , I , J;ΓMT ) ) 2 ( 19 ) Fig 12E shows the output signal from MT+ . Although a full characterization of MSTd neuron receptive fields remains an ongoing challenge [48] , the set of spiral templates spanned by linear combinations of radial and circular motion patterns has proven successful in capturing the pattern sensitivity across the population [98] . Analyses performed with a prior version of the competitive dynamics model revealed that linear self-motion did not activate spiral templates; such templates may play an important role in perceiving self-motion when traveling along a curved path [77] . Even in the presence of objects moving along a linear trajectory , the optic flow does not contain the spatial or temporal curvature required to activate spiral templates [45] . Because the range of conditions considered in the present study focus on observer and object motion along linear trajectories , we restricted the model MSTd template space to radial patterns . Our model MSTd does not include cells with sensitivity to laminar or “planar” flow [14] , but the role of such cells should be investigated in future research . Whereas objects that approach the observer along nearly parallel trajectories generate radial motion patterns within the object contours ( Fig 1A ) , fixed-depth objects ( e . g . Fig 1B ) and objects that approach with large path angles generate more laminar flow patterns . Through potential interactions with radial cells , activation of cells in MSTd tuned to laminar flow by these moving objects may stabilize heading signals and reduce bias . In this regard , cells tuned to laminar flow may participate in a redundant mechanism that complements or factors into the recurrent competition used in the competitive dynamics model . Future physiological and modeling work should clarify the role of cells tuned to laminar flow for heading perception , particularly in the presence of moving objects . Model MSTd matches an array of templates T± tuned to radial expansion ( + ) and contraction ( - ) with the bottom-up signal from model MT+ Nd . The templates Ti , j± have FoE/FoC positions centered on every position ( i , j ) within the spatial grid tiling of MSTd . The specification of expansive and contractive templates has been described previously [77] . In the following equation , the match is normalized by the energy of each template: Vi , j±=1∑ ( I , J ) ( Ti , j , I , J± ) ∑d∑ ( I , J ) ( Ti , j , I , J±Nd , I , J ) ( 20 ) MSTd cells compete in a soft winner-take-all network across the polarity of radial motion ( expansion versus contraction ) and over 2D space: Each cell has a firing threshold ΓMST = 0 . 3 and sigmoid inflection point fMST = 0 . 001 . The kernel GσMST , 7 specifies how units with neighboring singularity selectivities compete with one another , with σMST = 10 . The dynamics of the MSTd cell tuned to radial expansion or contraction with FoE/FoC selectivity at position ( i , j ) is defined as follows: dPi , j±dt=−Pi , j±+ ( 1−Pi , j± ) ( f ( g ( Pi , j±;ΓMST ) ;fMST ) +Vi , j± ) −Pi , j± ( ∑w≠±∑ ( n , m ) ≠ ( i , j ) GσMST , 7 , n , mf ( g ( Pn , mw;ΓMST ) ;fMST ) ) ( 21 ) Fig 12F shows the activity of expansion-sensitive MSTd cells at two different points in time . The heading estimate h* from model MSTd is determined by considering the FoE selectivity of the most active cell tuned to expansion along the horizontal cross-section x that contained the heading direction . While other expansion-selection units were often active , the most active unit had a centrally positioned FoE selectivity due to the vertical radial symmetry of the optic flow displays simulated . | Humans have little difficulty moving around in dynamic environments containing other moving objects . Previous research has demonstrated that moving objects may induce biases in perceived heading in some circumstances . Nevertheless , heading perception is surprisingly robust and stable . Even when large moving objects occupy much of the visual field and block our view of the future path , errors in heading judgments are surprisingly small—usually less than several degrees of visual angle . Furthermore , perceived heading does not abruptly shift or fluctuate as moving objects sweep across the observer’s future path . The aim of the present study is to investigate the qualities of our visual system that lead to such robust heading perception . We simulated two existing models that specify different heading mechanisms within the visual system and found that they could not capture the robustness and stability of human heading perception in dynamic environments . We then introduced the competitive dynamics model that succeeds due to its reliance on recurrent , competitive interactions among neurons that unfold over time that stabilize heading estimates . Our results suggest that competitive interactions within the visual system underlie the robustness and stability of human heading perception . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"classical",
"mechanics",
"fluid",
"mechanics",
"social",
"sciences",
"neuroscience",
"simulation",
"and",
"modeling",
"systems",
"science",
"mathematics",
"vision",
"neuronal",
"tuning",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"animal",
"cells",
"flow",
"field",
"fluid",
"dynamics",
"dynamical",
"systems",
"continuum",
"mechanics",
"physics",
"psychology",
"cellular",
"neuroscience",
"cell",
"biology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"sensory",
"perception",
"physical",
"sciences",
"motion"
]
| 2016 | Competitive Dynamics in MSTd: A Mechanism for Robust Heading Perception Based on Optic Flow |
HoxA genes exhibit central roles during development and causal mutations have been found in several human syndromes including limb malformation . Despite their importance , information on how these genes are regulated is lacking . Here , we report on the first identification of bona fide transcriptional enhancers controlling HoxA genes in developing limbs and show that these enhancers are grouped into distinct topological domains at the sub-megabase scale ( sub-TADs ) . We provide evidence that target genes and regulatory elements physically interact with each other through contacts between sub-TADs rather than by the formation of discreet “DNA loops” . Interestingly , there is no obvious relationship between the functional domains of the enhancers within the limb and how they are partitioned among the topological domains , suggesting that sub-TAD formation does not rely on enhancer activity . Moreover , we show that suppressing the transcriptional activity of enhancers does not abrogate their contacts with HoxA genes . Based on these data , we propose a model whereby chromatin architecture defines the functional landscapes of enhancers . From an evolutionary standpoint , our data points to the convergent evolution of HoxA and HoxD regulation in the fin-to-limb transition , one of the major morphological innovations in vertebrates .
The Hox gene family encodes transcription factors with central roles in patterning of the body plan and organogenesis . Hox genes are grouped into clusters in most animal species , and mammals possess 39 genes divided into four clusters named HoxA to HoxD . In mice , deletion of the HoxA cluster is embryonic lethal [1]–[2] whereas mutants lacking HoxB , HoxC , or HoxD are viable at least until birth [3]–[5] . Inactivation of individual Hox genes identified Hoxa13 as a gene required for proper placenta function and thus embryonic survival [2] , [6]–[7] . Mutations in HoxA genes have been found in various human syndromes ( e . g . HFGS-OMIM140000 , Guttmacher syndrome-OMIM176305 , MRKH-OMIM277000 ) including limb malformations . Studies of gene inactivation in mice demonstrated that genes located at the 5′ end of the HoxA cluster ( Hoxa9–13 ) are required for proper patterning of the three limb segments: the upper arm ( humerus; Hoxa9 , 10 ) , lower arm ( radius and ulna; Hoxa10 , 11 ) , and the hand/foot ( autopod; Hoxa13 ) [6] , [8]–[11] . Despite their pivotal roles during embryogenesis , little is known about the regulation of HoxA genes . This is in contrast to HoxD , which transcriptional control has been more thoroughly studied , especially in the limb where the HoxD genes play partially overlapping functions with HoxA [12] . Expression at the HoxA and D clusters follows similar dynamics during limb development , and occurs in two phases [12] . In the first phase , which starts at embryonic day 9 . 5 of development ( E9 . 5 ) , expression at both clusters is comparable suggesting that the control mechanisms are likely similar . During this phase , gene expression generally follows the collinear strategy observed in the trunk , characterized by sequential gene activation from one end of the cluster ( Hox1 ) to the other ( Hox13 ) , with early activated genes expressed throughout the limb bud and those activated later ( Hox10-13 ) gradually restricted to posterior cells [13] . In contrast , the expression domains of HoxA and HoxD genes partly differ in the second phase ( from E11 . 5 onwards ) , suggesting some differences in the regulatory mechanisms controlling the clusters in this later phase . Previous studies show that transcription at the HoxD cluster is regulated long-distance by enhancers in several tissues ( reviewed in [14] ) . Notably , expression of Hoxd10 to Hoxd13 in the distal part of the limb bud ( presumptive hand/foot ) is controlled by several remote cis-regulatory sequences located in the gene desert upstream of the cluster [15] . Hands/feet , in particular digits , are evolutionary novel structures and the hallmark of Vertebrate adaptation to terrestrial habitats . The fact that Hoxa10 and Hoxa13 are also expressed in the presumptive hand/foot domain therefore raised the possibility that specific recruitment of HoxA and HoxD gene functions in developing digits stem from the implementation of similar cis-regulatory elements during the fin-to-limb transition . Whereas sequence conservation analysis of the region upstream of these clusters did not identify cognate cis-regulatory elements driving HoxA expression in limbs [16] , BAC transgenesis revealed the existence of a “digit” enhancer activity located between 250 and 500 kb upstream of the Hoxa13 gene , in the neighborhood of the 3-hydroxyisobutyrate dehydrogenase ( Hibadh ) gene [17] . As Hibadh is also expressed in distal limb buds [16] , this study could not resolve whether the “digit” enhancer activity detected within that region controls Hibadh , Hoxa10/13 , or both . Thus , the enhancer sequence ( s ) and whether HoxA expression in limbs is regulated by long-range control mechanisms has remained unknown . It was previously shown that control DNA elements could regulate the expression of remote genes by physically interacting with them [18] . Physical contacts between chromatin segments can be measured using the chromosome conformation capture ( 3C ) methods , a series of assays that use proximity-based ligation to infer the three-dimensional organization of genomes [19] . 3C assays were used to show that regulation of HoxD genes in the presumptive digit domain is mediated by physical contacts with remote enhancers , and led to a model whereby expression of Hoxd10 to d13 associates with the formation of DNA loops between the genes and regulatory islands [15] . This was further supported by Fluorescence In Situ Hybridization data showing the co-localization of HoxD genes and one of its regulatory islands , specifically in digit progenitor cells [20] . Whether the proximity between target genes and regulatory DNA elements requires transcription appears to be loci-dependent and it remains unknown whether a given mechanism prevails over others . Indeed , while such “loops” were sometimes reported in absence of transcription [21]–[22] , transcription factors requirement for DNA looping was uncovered at the β-globin locus [23]–[24] and Igh gene [25] . Here , we show that during limb development , expression of HoxA genes is controlled by multiple remote enhancers located upstream of the cluster . In limb cells , these enhancers are grouped into distinct sub-megabase topological domains ( sub-TADs ) that contact each other and the sub-TADs containing target genes . In head tissues , the topology is drastically different , modifying both gene-enhancer and enhancer-enhancer interactions . Interestingly , enhancers located in the same sub-TAD are active in distinct subset of limb cells suggesting that spatial clustering of enhancers does not simply reflect enhancer co-activity . We also present evidence that enhancer-HoxA contacts are maintained even when enhancer activity is suppressed , suggesting that the HoxA regulatory region acquires a permissive conformation prior to gene activation . We suggest a model whereby sub-TAD formation and/or contacts between sub-TADs define the cis-regulatory network controlling gene expression . From an evolutionary perspective , this first extensive characterization of HoxA regulation in developing limbs provides new insights into the evolution of Hox regulation in the emergence of hand/foot . Our study suggests that while the DNA sequences of the distal limb enhancers for HoxA and HoxD genes are different and have likely emerged independently , chromosome partitioning into topological domains has similarly constrained the evolution of HoxA and HoxD cis-regulatory landscapes underlying the emergence of digits , one of the major morphological innovations in Vertebrates .
To identify enhancer sequences regulating HoxA expression during limb development , we used a combination of genetic and genomics approaches that probe enhancer features in mouse embryos . We first tested whether HoxA transcription in developing limbs involves cis-regulatory sequences outside of the gene cluster itself . To this end , we used two mutant lines with targeted genomic rearrangements at the HoxA cluster [1]–[2] to monitor activation of reporter transgenes by surrounding enhancer activity ( Figure 1A , B ) . Whole mount in situ hybridization shows that a neomycin reporter transgene located downstream of Hoxa1 is not expressed in limbs at E11 . 5 ( Figure 1A , left ) . In contrast , we found that a hygromycin transgene inserted at the opposite end of the cluster , 3 . 5 kb upstream of Hoxa13 is robustly transcribed in distal limbs at this stage ( Figure 1A , right ) . These expression patterns correlate well with the expression profile of the endogenous HoxA genes adjacent to the reporter transgenes . Upon deletion of the entire HoxA cluster , the neomycin transgene becomes activated in distal limbs suggesting that sequences upstream of the cluster are sufficient to trigger distal expression ( Figure 1B ) . These results support the previously proposed hypothesis that a 250 kb region in the neighborhood of Hibadh contains an enhancer activity controlling HoxA expression in developing limbs [17] . Given the results described above , we focused our analysis on the genomic region upstream of the cluster . To identify active enhancers in distal limbs , we used dissected distal forelimbs ( Figure 1C ) , which are composed of cells expressing mainly Hoxa10 and a13 , but also a small amount of Hoxa9 and a11 from the presumptive wrist domain ( mesopod ) . Active enhancers are characterized by the binding of several proteins including RNA polymerase II ( RNAP2 ) , and subunits of Mediator like Med12 [26] . We therefore mapped candidate enhancer sequences by identifying genomic sites enriched in these proteins using chromatin immunoprecipitation combined with deep sequencing ( ChIP-seq ) in cells isolated from E12 . 5 distal limb buds ( Figure 1D ) . This data was considered together with previously published datasets derived from whole limb buds for the transcriptional co-activator p300 [27] and acetylated histone H3 lysine 27 ( H3K27Ac ) , which also mark active enhancers [28] . Sequences distinct from proximal promoters ( RefSeq ) that were bound by RNAP2 and at least one other mark , or by both p300 and H3K27Ac were retained as candidate enhancers . Using these criteria , 19 putative enhancers were identified within 850 kb upstream of Hoxa13 ( Figure 1D , top ) . The number of candidate sequences identified upstream of HoxA was rather large , particularly compared to HoxD for which seven enhancers have been identified [15] . Also , in contrast to the gene desert surrounding HoxD , the region upstream of HoxA encompasses several genes ( Figure 1D ) . Candidate HoxA enhancers therefore reside amidst other genes including Hibadh , Tax1bp1 , and Jazf1 , for which expression has been reported in the limb [16] . As ChIP-seq datasets cannot resolve the targets of enhancers , we used a structural approach to assess the potential interactions of the candidate enhancers with HoxA genes . We profiled the interaction pattern of the HoxA cluster with the upstream 850 Kb region in distal limb buds and head tissues using 5C technology combined with deep sequencing [29]–[30] , which provides insights into chromatin architecture at high resolution ( down to 4–6 kbs on average ) . We found that the 5′ part of HoxA , containing Hoxa9 to Hoxa13 , frequently interacts with several regions upstream of the cluster ( Figure 2 , top , Figure S1 ) , and that most of these regions contain the candidate limb enhancers ( Figure 2 , bottom ) . In contrast , none of the enhancers interact with the 3′ part of the cluster containing Hoxa1 to Hoxa7 ( Figure 2 ) , which have no detectable expression in limb buds at this developmental stage . This result is reminiscent of the distal enhancers controlling the 5′ HoxD genes , which are also located upstream of the cluster and specifically interact with genes located in the 5′ part [15] . Interestingly , previous studies based on Hi-C analysis revealed that the HoxA and HoxD clusters each span a junction between two so-called “topologically associated domains” ( TADs ) , with 3′ genes residing into one TAD and the 5′ part extending into the other [31] . TADs are thought to represent a basic unit of chromatin organization at the megabase-scale that is largely conserved between cell types [32] . Our data therefore points to a common relationship between chromatin topology and the limb regulatory landscapes of the HoxA and HoxD clusters . Interestingly , sequences with the highest interaction frequencies with 5′HoxA genes ( e10 , 13 , 14 and e15 , 16 , 18 ) locate farther from the cluster , within the Jazf1 gene , and correspond to those loci most enriched in marks typical of active enhancers ( Figure 1D ) . High interaction frequencies being associated either with stronger , more abundant and/or stable spatial contacts , these data likely reflect a prevalent activity of these enhancers in distal limbs . In contrast to the other enhancers , e1 and e3 do not show enriched interactions with the 5′ part of the HoxA cluster in distal limbs compared to head tissue ( Figure 2 , bottom ) . e1 is located close to Evx1 , within a region of high interaction frequencies with HoxA both in limb and head tissues . This is not the case for e3 so we further tested interaction frequencies between Hoxa13 and e2 to e5 using 3C ( Figure S2 ) . This analysis shows higher frequency of interactions between these enhancers and Hoxa13 specifically in the limb . Yet , based on our 5C data , these interactions are modest compared to those observed for the other enhancers ( Figure 2 ) . Interestingly , contacts such as those with e5 , e13 and e15 were also observed in the head at low frequencies ( Figure 2 ) . As there is no evidence of HoxA expression in the head at the stage analyzed , the contacts observed might be evidence of default chromatin architecture in this tissue . Alternatively , these enhancers may drive HoxA expression in head tissues at levels below detection by whole-mount in situ hybridization . Finally , our 5C data also reveals high interaction frequencies with at least two loci that have no apparent characteristics of transcriptional enhancers ( Figure 2 bottom , blue stars ) . These may reflect additional structural anchors that stabilize the chromatin architecture , such as those mediated by CTCF and Cohesin [33]–[34] . Interestingly , loci bound either by CTCF or cohesin in limb buds have been recently identified [35] and comparison with our data shows that almost all loci interacting with 5′ HoxA genes overlap with either CTCF or cohesin binding ( Figure 2 , bottom ) . Having confirmed the spatial proximity between 5′ HoxA genes and most of our candidate enhancers , we proceeded to test their in vivo activity in the mouse by transgenesis . Putative enhancer sequences were subcloned into vectors carrying the β-globin minimal promoter and lacZ reporter . Except for e1 and e2 , X-Gal staining in transgenic embryos shows that all candidate enhancers tested activates transcription in developing limbs ( Figure 3 , Table S1 ) . Interestingly , the confirmed enhancers exhibit distinct but overlapping activity domains in limb buds , and all trigger expression in the presumptive hand/foot ( Figure 3 ) . While some are active mostly in the distal part of the limb ( e3 , 4 , 5 , 10 , 12 , 13 ) , others are functional also in the proximal domain ( e5 , 16 , 18 ) . The only candidate enhancers that fail to trigger reporter expression in our transgenic assays are e1 and e2 ( Table S1 ) . The absence of activity for these two candidates indicates either that these sequences are not limb enhancers or that the transgenes did not include all the necessary sequences to reflect their transcriptional activities . For e1 , our 5C data ( Figure 2 ) neither supports nor disagrees with it being a HoxA enhancer since it lies within a large region of high interaction frequency . Interestingly , e1 is located within a 50 kb DNA fragment that was previously shown to trigger gene expression in distal limbs [2] , suggesting that it is possibly a bona fide limb enhancer but that some sequences required for its activity are absent from the 2 . 9 kb fragment tested in our transgenic assay . Similarly , absence of X-Gal staining in e2 transgenic embryos does not prove that e2 is not an enhancer . Yet , the fact that it does not strongly interact with 5′ HoxA genes in our 3C and 5C assays suggest that e2 may not be tightly linked to the regulation of HoxA genes . Nonetheless , analysis of the other identified enhancers shows that multiple enhancers with overlapping domain-specific activities regulate transcription at the HoxA cluster in the limb . While “DNA looping” is associated with long-range transcriptional control , the extent to which spatial structure exists prior to or as a consequence of enhancer activation remains elusive . This issue partly originates from the fact that most studies have compared the spatial distance of enhancers and target gene ( s ) in tissues expressing the genes with others where they are never expressed . To gain insight into the causative relationship between spatial proximity and long-distance enhancer regulation , we examined the outcome of enhancer silencing on long-range enhancer-gene interactions in developing limbs . During limb development , the transcriptional repressor Gli3R negatively regulates the expression of HoxA genes [36]–[37] . While Gli3 is expressed almost throughout the limb in wild type ( wt ) embryos , the Gli3R domain is restricted anteriorly as a consequence of posterior Sonic hedgehog ( Shh ) signaling emanating from the Zone of Polarizing Activity ( ZPA ) , which blocks processing of the full length Gli3 protein into its truncated repressor form [38] . In Shh−/− limbs , the Gli3R functional domain extends posteriorly leading to the down-regulation of HoxA as well as HoxD genes [36]–[37] . Amongst the HoxA-associated limb enhancers identified , we found several that overlap with loci bound by Gli3R in the limb ( e3 , e5 , e9 and e16; [39] ) . The activity of these enhancers should thus be suppressed in Shh−/− mutant . Of these , e5 is of particular interest because it triggers robust gene expression ( Figure 3 ) , and there is no other limb enhancer in its genomic neighborhood allowing us to assess its interaction frequency with HoxA without potential interference from surrounding enhancers . We first verified the activity of e5 in Shh−/− by generating mutant embryos homozygous for Shh inactivation and carrying the e5 transgene . X-Gal staining shows that e5 activity is suppressed in limbs upon inactivation of Shh ( Figure 4Aa–d , compare a to b ) while still functional in the developing genitalia ( Figure 4A , panel d ) . In contrast , a transgene containing the e1 enhancer , which does not overlap with a Gli-bound locus , remained expressed in a Shh−/− background ( Figure 4Af–h ) although in a smaller domain consistent with Shh−/−embryos having reduced limb size ( [40]; Figure 4A , compare e to f ) . To assess whether HoxA-enhancer proximity requires enhancer activity , we measured contacts between Hoxa13 and e5 in wt and Shh−/− distal limb buds from E11 . 5 embryos . As e5 activity is suppressed in the absence of Shh , the enhancer should no longer interact with Hoxa13 if enhancer activity is required for the contact . 3C analysis shows that e5 interacts with Hoxa13 even in the absence of Shh ( Figure 4B ) . Although interaction frequencies are lower than in wt limbs , the interaction pattern is similar and contacts are much stronger than in the head , which was used as control . These data show that even though e5 silencing may affect the robustness of the interactions , the spatial proximity between e5 and Hoxa13 does not require the transcriptional activity of the enhancer . As Hoxa13 expression is severely reduced in the absence of Shh [37] , [41] , we next wondered whether the contact pattern of HoxA genes with the distal limb enhancers was similarly preserved in Shh−/− limbs . To address this question , we compared the interaction profile of the HoxA cluster with its upstream regulatory region in wt and Shh−/− limbs , and in the head . For these 5C experiments , we used a modified 3C library protocol optimized for the production of libraries from a small number of cells ( see Materials and Methods ) . This protocol largely recapitulated the contact pattern detected in wt limbs and the head with our standard approach ( compare corresponding panels in Figures 2 and 4C ) . Consistent with our 3C data , this 5C analysis revealed a similar contact pattern between the 5′ HoxA genes and upstream regulatory region in the Shh−/− mutant and wt limbs ( Figure 4C , Figure S3 ) . These include contacts with e5 and e16 enhancers , which overlap with Gli3R sites and others like e10 and e13 that are not regulated by Shh . As observed in our preliminary 3C analysis , the contacts were weaker in the Shh−/− suggesting that strengthening a given enhancer-promoter contact upon enhancer activation may impact on the stabilization/strength of other interactions . Together , these data indicate that enhancer activity strengthens , but is not mandatory for spatial proximity between enhancers and their target genes . The observation that different enhancers drive transcription in the same areas of the limb suggests a possible physical link between some of them . To test this possibility , we extended our 5C analysis to the whole regulatory region . The HoxA cluster was previously found to span the junction between two adjacent TADs in human IMR90 and mouse embryonic stem cells ( ES ) analyzed with Hi-C at the mega-base scale [31] . We observed a similar megabase scale organization in our samples , where 5′ HoxA genes and distal limb enhancers are located in the same TAD ( Figure 5A , B , and Figure S4 ) . We found that this TAD is subdivided into domains of interactions that differ between the limb and head tissues at E12 . 5 ( Figure 5A , B ) . In addition , contacts between sub-TADs are different in the two tissues . For example , the HoxA sub-TAD containing Hoxa9 to Hoxa13 preferentially forms long-range contacts with the enhancers in the limb ( e . g . e10–14 , e15–18; Figure 5A , Figure S5 ) , while it interacts strongly with the 3′ HoxA genes in the head ( Figure 5C , bottom , Figure S4 ) . Similarly , Evx1 , which spatially localizes within its own domain , interacts long-distance with a subset of the identified enhancers in limbs , consistent with its expression pattern being similar to Hoxa13 . This is different in the head where Evx1 and HoxA are mostly inactive and the genes form a large interacting domain ( Figure 5B , C ) , which likely reflects chromatin compaction at transcriptionally silent loci . This result raises the possibility that chromatin conformation within TADs could vary in a tissue-specific manner . In support of this , a recent 5C analysis in mouse ES and neural progenitor cells identified tissue-specific topological domains at the sub-megabase scale , termed “sub-TADs” [33] . Our 5C analysis therefore revealed the existence of tissue-specific sub-TAD interactions underlying the regulation of HoxA genes in developing limbs . The chromatin architecture resulting in the spatial proximity between 5′ HoxA genes , the enhancers , and the promoter of Hibadh suggests that Hibadh itself interacts with HoxA-associated limb enhancers . Indeed , Hibadh shows enriched interaction with e5 , e13 and e16 in limb compared to head tissue ( Figure S5 ) . Interactions between Evx1 and Hibadh are also enriched in limb compared to head tissue . Our experimental design unfortunately did not retain the promoter region of Tax1bp1 and thus we could not profile its connectivity with the region . As for the promoter of Jazf1 , it contacts neither the genes nor the enhancers consistent with the absence of RNAP2 and Med12 at its promoter ( Figure 1D ) , and in agreement with previous work showing that Jazf1 is expressed in distal limbs only at later developmental stages [16] . Together , these results show that a subset of HoxA-associated enhancers likely regulate also Evx1 and Hibadh . Interestingly , there is an extensive connectivity between the enhancers themselves in the limb but not in head tissues ( Figure 5A , B ) . Similarly to the genes , the enhancers partition among different sub-TADs that interact together . This is particularly visible for the most distal ones where e10–14 localized within one sub-TAD , and e15–18 into another ( Figure 5A , D ) . This organization suggests that enhancers are spatially grouped into regulatory modules , which can interact with each other , eventually triggering specific expression patterns . Such interaction between genomic domains is reminiscent of the contacts identified in Drosophila [42] . It is thus likely that long-range gene regulation relies on sub-TAD interactions rather than discrete looping between specific DNA elements . Moreover , interactions between gene and enhancer sub-TADs in the limb strengthened and better defined the position of the corresponding TAD as compared to head tissues ( Figure 5 , compare A and B ) . This result suggests that although largely invariant , the partitioning of chromosomes into TADs can be affected by the tissue-specific folding of the chromatin at the sub-megabase level .
In this study , we identified the very first set of bona fide limb enhancers controlling 5′ HoxA gene expression . We show that these enhancers , like the HoxA genes , are grouped into distinct topological domains at the sub-megabase scale , and that long-range contacts between the sub-TADs underlie the expression of 5′ HoxA genes in the developing limb . This result suggests that long-distance regulation of HoxA genes is based on sub-TAD interactions rather than discrete looping between enhancers and target genes . In the head , sub-TAD interactions are barely detectable thus indicating that the chromatin architecture of the region upstream HoxA varies in a tissue-specific manner at the sub-megabase scale . The apparent lack of sub-TAD interactions in the head could also be the consequence of the greater cellular complexity of this tissue , which would equally imply that sub-TAD interactions are cell type/tissue-specific ( Figure 5B ) . A similar conclusion was recently reached based on the comparison between 5C data in mouse ES cells and neural progenitor cells {Nora , 2012 #165} [32]–[33] . The cell-type/tissue specificity of sub-TADs contrasts with the mostly invariant nature of TADs , which partition the genome into topological domains at the megabase scale [31]–[32] . While it was proposed that TADs could represent the structural basis of regulatory landscapes [43] , the actual chromatin folding associated with transcriptional activity likely relies mostly on sub-TAD interactions ( Figure 6 ) . The diametrically opposed invariant nature of TADs and the tissue-specificity of sub-TADs also imply that distinct structural parameters define them . Accordingly , while arrays of CTCF sites characterize TAD boundaries [31] , there is no obvious correlation between CTCF binding and the sub-TAD boundaries observed in limb buds ( Figure S5 ) . Our results also indicate that at least some of the gene-enhancer contacts form independently of enhancer transcriptional activity ( Figure 4 ) , and we suggest that this structure largely exists before the gene transcription begins . This view is supported by the existence of interactions with loci for which there is no evidence of transcriptional activity ( Blue stars in Figure 2 ) . Moreover , our data shows that enhancers triggering distinct expression patterns in the limb ( i . e . active in different cells ) actually belong to the same sub-TAD , which further supports the notion that organization of the chromatin into sub-TADs does not simply reflect physical clustering of active enhancers . Chromatin interactions nonetheless appear strengthened by enhancer activity consistent with the recent concept of self-enforcing structure-function feedbacks , considered as a mechanism propagating cell-fate memory [44] . Our data also reveal better-defined boundaries of the 5′ HoxA-containing TAD in limbs , where sub-TADs robustly interact ( Figure 5 ) . This result raises the possibility that upon enhancer activation , the robustness of sub-TAD interactions within two adjacent TADs may change and consequently re-define the position of the TAD boundary . This potential TAD/sub-TAD interplay may actually provide a mechanistic explanation for Hoxd9 to Hoxd11 switching from one TAD to the neighboring one in proximal limb compared to distal limb cells [45] . The identification of multiple enhancers controlling 5′HoxA genes in distal limbs raises questions about the potential role and benefits for this apparently complex control mechanism . The evidence that the various enhancers identified have distinct spatial specificities , together with the eventual morphological diversity of the hand/foot , points to the existence of an early molecular heterogeneity among the mesenchymal progenitors of the hand/foot . Accordingly , Shh signaling regulates a subset of enhancers identified here while others are not ( Figure 4 ) . Nonetheless , most enhancers also appear to share overlapping functional domains . Interactions between some enhancers may reflect their co-function in some cells , which could correspond to specific cell populations in which a higher HoxA dosage is required . Alternatively , enhancer interactions could be the consequence of a “pre-set” chromatin architecture whereby a series of enhancers is brought in the vicinity of the same target genes , without having necessarily a combined transcriptional input in the same cell . Finally , it should also be mentioned that multiple enhancers with overlapping function can be beneficial , as exemplified with the discovery of shadow enhancers , which compensate for each another in sub-optimal conditions [46]–[47] . The hand/foot ( autopod ) is one of the major morphological novelties that accompanied Vertebrates adaptation to terrestrial habitats . As autopod development requires the function of HoxA and HoxD genes , the mechanism that led to the emergence of this new Hox function appears as a key molecular event associated with the fin-to-limb transition . By mapping active enhancers in the presumptive hand/foot and testing their interaction with HoxA genes , we provide evidence that HoxA expression in this tissue relies on long-range regulation by multiple enhancers . Previous studies on the regulation of HoxD genes led to the same conclusion [15] , [48]–[49] suggesting that HoxA and HoxD genes have been recruited in this evolutionary novel structure through the implementation of a similar regulatory strategy . Yet , sequence comparison between HoxA and HoxD specific enhancers failed to identify obvious conservation thereby favoring a model whereby the recruitment of HoxA and HoxD genes in the presumptive hand/foot was likely achieved through independent implementations of novel cis-regulatory elements . Since these enhancers were identified with a resolution varying between 0 . 5 and 2 kb , it is possible that they are bound by the same transcription factors but with a distinct layout of their binding sites , as it would be expected from the independent evolution of the HoxA and HoxD regulatory landscapes . It is also possible that some ‘HoxA’ and ‘HoxD’ enhancers are bound by distinct combinations of transcription factors , in agreement with a subset of ‘HoxA’ enhancers having domains of activity within the developing limb distinct from the ‘HoxD’ enhancers ( Figure 3; [15] ) . Notably , the differences in enhancer functional domains are consistent with the specificities of HoxA expression as illustrated in the presumptive digit one domain: while HoxD expression in digit one is restricted to Hoxd13 as a result of the quantitative collinearity [50]–[52] , no such phenomenon is observed for HoxA genes , the regulation of which involves a digit one-specific enhancer not identified for HoxD genes [15] . The independent evolution of HoxA and HoxD regulatory landscapes suggested by the absence of obvious sequence conservation of their respective enhancers is further supported by several findings . First , the recent evidence that HoxA and HoxB clusters most likely stem from the duplication of an ancestral HoxA/B cluster [53] implies that putative ancestral regulatory modules common only to HoxA and HoxD should have been lost at HoxB and HoxC . This scenario however appears unlikely to account for the specific HoxA and HoxD regulation associated with hand/feet development as the tandem duplications of the ancestral Hox cluster that led to the four Vertebrate Hox clusters occurred prior to the fin-to-limb transition . Second , there is a major difference in the layout of the HoxA and HoxD regulatory landscapes controlling their expression in the developing hand/feet . While HoxD-associated enhancers are part of a gene desert [15] , [48]–[49] , a large number of HoxA-associated enhancers are embedded in genes . Notably , HoxA enhancers with the highest enrichment of RNAP2 , Med12 and p300 , which also show the highest frequencies of interaction with HoxA genes , are located within Hibadh and Jazf1 . Moreover , the genomic domain between HoxA and Jazf1 , has undergone significant expansion from fish to mice ( about 50 kb in fish and 800 kb in mice ) , indicating that an extensive genomic reshuffling at the HoxA regulatory landscape occurred during the fin-to-limb transition , which further support an independent evolution of the HoxA and HoxD regulatory landscapes . Interestingly , this genome expansion affected both the size of Hibadh , Jazf1 and the intergenic regions . The absence of preferential localization of HoxA-associated enhancers in gene-free regions thus suggests that introns are equally amenable to sequence evolution and emergence of new regulatory elements . Although enhancers controlling HoxA and HoxD expression in distal limb most likely emerged independently , in both cases the distal limb regulatory landscape is located within the TAD containing the 5′ genes ( [45] and this work ) . As long-range physical contacts between DNA sequences preferentially occur within TADs , it is conceivable that topological constraints have influenced the evolution of HoxA and HoxD regulatory landscapes associated with their distal limb expression . Interestingly , the early/proximal limb regulatory landscape of HoxD was identified on the opposite side of the gene cluster , within a TAD containing the 3′ HoxD genes , and not contacting Hoxd13 [45] . Whether the existence of a TAD boundary within the HoxA and HoxD clusters has favored the differential expression of Hox genes in proximal and distal limb bud or spatially constrained the emergence of proximal and distal limb enhancers remains unclear . Nonetheless , the deleterious modification of proximal limb development upon expression of Hoxa13 or Hoxd13 in early/proximal limb bud [54] raises the possibility that the TAD boundary embedded in both the HoxA and HoxD clusters has influenced the evolution of the tetrapod limb morphology . Although , chromosome partitioning into TADs remains to be characterized in most animal species , the presence of a TAD boundary embedded in each Hox cluster both in mice and humans [31] suggests a possible widespread impact of genome topology on the evolution of Hox regulation , and perhaps more generally on the evolution of regulatory landscapes . In summary , our study reveals that extensive three-dimensional chromatin interactions control the expression of HoxA genes in developing limbs by forming distinct topological domains containing limb enhancers , which interact with each other and with the topological domains containing their target genes . Although this chromatin architecture is tissue-specific , our data suggests that it forms independently of enhancer activity , and is strengthened upon enhancer activation . Importantly , our data provide evidence that target genes and regulatory elements physically interact with each other through contacts between sub-TADs rather than by the formation of discreet “DNA loops” . From an evolutionary standpoint , the identification of HoxA-associated enhancers in limbs reveals major differences with the HoxD regulatory landscape suggesting that the changes in HoxA and HoxD regulation associated with the emergence of the hand/foot likely occurred through the independent emergence of regulatory sequences but common topological constraints .
The HoxAFlox , HoxADelNeo , IR50 , and Shh−/− mice lines were described previously [1]–[2] , [40] . Candidate enhancer sequences identified from ChIP-seq data were amplified by PCR using the primer sequences reported in Table S1 and PCR products were verified by sequencing . Enhancer sequences were cloned upstream of the chicken β-globin minimal promoter and the LacZΔCpG NLS reporter . Transgenic embryos were generated by pronuclear injections , and at least three transgenic embryos per construct were analyzed . The stable mouse line for e5 was generated using the same protocol . X-Gal staining was performed on E12 . 5 embryos following standard procedures . In situ hybridization was conducted using a standard procedure [55] . Hygromycin and Neomycin probes were generated as described previously [2] . Distal limb and head tissues were dissected at E12 . 5 for wt and at E11 . 5 for Shh−/− mice . Tissues were collected in 1×PBS containing 10% FBS ( 100 µl per embryo ) , and the samples were incubated 20 min at 37°C with collagenase ( 0 . 025% final concentration ) to obtain single-cell suspensions . The number of cells in suspension was then counted under the microscope , and each sample was diluted in 9 ml of 1×PBS containing 10% FBS ( 5 ml for Shh−/− embryos ) . The cells were then fixed with 1% formaldehyde for 10 min at room temperature . Crosslinking was stopped with glycine ( 125 mM final concentration ) , and incubated 5 min at room temperature followed by 15 min on ice . Cells were centrifuged at 400 g for 10 min at 4°C . Supernatants were removed and the cell pellets were flash frozen on dry ice . Chromatin immunoprecipitation was performed as previously described with some modifications [56]–[57] . Briefly , chromatin from 5 million cells was sonicated using a Branson Sonicator 450D to obtain fragments with average sizes ranging between 100–600 bp . Cell debris was removed by centrifugation at 20 , 000 g for 15 min at 4°C and aliquots of the supernatant were taken for quantification and to confirm proper sonication . Remaining samples were stored at −80°C until use . Chromatin from 5 million cells was used for each immunoprecipitation . Protein G Dynal Beads ( Invitrogen ) were incubated 8 hours at 4°C with either 5 or 10 µg of antibodies . The chromatin was then incubated with the beads overnight . Immunoprecipitated complexes were sequentially washed in low salt ( 150 mM NaCl , 0 . 1% SDS , 1% Triton X-100 , 2 mM EDTA , 20 mM Tris-HCl ( pH 8 . 0 ) ) , medium salt ( 250 mM NaCl , 0 . 1% SDS , 1% Triton X-100 , 2 mM EDTA , 20 mM Tris-HCl ( pH 8 . 0 ) ) , LiCl ( 0 . 25 M LiCl , 0 . 5% NP40 , 0 . 5% Na-Deoxycholate , 1 mM EDTA , 10 mM Tris-HCl ( pH 8 . 0 ) ) , and 1×TE buffers . The protein/DNA complexes were eluted in an SDS buffer ( 1% SDS , 50 mM Tris ( pH 8 . 0 ) , 10 mM EDTA ) by incubation at 65°C for 15 min on a rotating platform . Crosslinks were reversed by incubating the complexes at 65°C overnight . Samples were treated one hour at 55°C with RNAseA ( 0 . 2 µg/ml final concentration ) and then with Proteinase K for two hours . Finally , the DNA was purified on QIAquick columns ( Qiagen ) . Specific antibodies for Med12 and RNAP2 were purchased from Bethyl ( A300-774A ) and Abcam ( ab5131 ) , respectively . The ChIPed material was sequenced on a Hi-Seq 2000 high-throughput DNA sequencer . Sequencing libraries were prepared from 31 ng ( RNAP2 ) , 5 ng ( Med12 ) , and 345 ng ( input ) of ChIPed DNA . The libraries and flow cells were prepared by the IRCM Molecular Biology and Functional Genomics platform . The libraries were multiplexed and sequenced on one lane . The sequencing was performed by the McGill University and Génome Québec Innovation Centre following recommendations by the manufacturer ( Illumina , San Diego , CA ) . For RNAP2 , Med12 , and the input , we obtained a total of 151 , 045 , 509 , 110 , 507 , 927 , and 98 , 043 , 425 sequence reads , respectively . The first base of each sequence read was trimmed to ensure high base calling quality . The trimmed reads were mapped to the mouse mm9 genome assembly with Bowtie using the –best parameters [58] . To identify the highly significant RNAP2 and Med12 peaks , we used the MACS 1 . 4 . 1 peak finder with the following parameters: --format SAM --wig --bw 250 --mfold 7 , 30 -pvalue 1e–5 -g mm [59] . Redundant reads were filtered out for peak finding and wiggle file generation . Thus the wiggle files enclose the total number of uniquely mapped and non-redundant reads . After processing the data , the number of sequence reads we obtained was 129 , 222 , 085 for RNAP2 , 91 , 816 , 355 for Med12 , and 88 , 141 , 136 for the input . The position of RNAP2 and Med12 peaks genome-wide identified in our study is provided in Tables S21 and S22 , respectively . We also provide the wig files for the data on chromosome 6 ( Dataset S1 , S2 and S3 ) . Limb and head cell pellets were treated as previously described [29] , [60] . Briefly , 10 million fixed cells ( 2 . 87 million for Shh−/− library used for the 3C experiments ) were incubated for 15 min on ice in 200 µl of lysis buffer ( 10 mM Tris ( pH 8 . 0 ) , 10 mM NaCl , 0 . 2% NP40 , supplemented with fresh protease inhibitor cocktail ) . Cells were then disrupted on ice with a dounce homogenizer ( pestle B; 2×20 strokes ) . Cell suspensions were transferred to eppendorf tubes and centrifuged 5 min at 2000 g . Supernatants were removed and the cell pellets were washed twice with 100 µl of 1×EcoRI buffer ( NEB ) . After the second wash , the cell pellet was resuspended in 100 µl of 1×EcoRI buffer , and divided into two eppendorf tubes containing 50 µl of cell suspension . 1×EcoRI buffer ( 337 µl ) was added to each tube , and the mixture was incubated 10 min at 65°C with 0 . 1% SDS final ( 38 µl ) . Triton X-100 ( 44 µl of 10% Triton X-100 ) was added before overnight digestion with EcoRI ( 400 Units ) . The restriction enzyme was then inactivated by adding 86 µl of 10% SDS , and incubating 30 min at 65°C . Samples were then individually diluted into 7 . 62 ml of ligation mix ( 750 µl of 10% Triton X-100 , 750 µl of 10×ligation buffer , 80 µl of 10 mg/ml of BSA , 80 µl of 100 mM ATP and 3000 Cohesive end Units of T4 DNA ligase ) . Ligation was carried out at 16°C for 2 hours . 3C libraries were then incubated overnight at 65°C with 50 µl Proteinase K ( 10 mg/ml ) , and with an additional 50 µl Proteinase K the following day for 2 hours . The DNA was purified by one phenol and one phenol-chloroform extraction , and precipitated with 0 . 1 volume of 3M NaOAc pH 5 . 2 ( 800 µl ) and 2 . 5 volumes of cold EtOH ( 20 ml ) . After at least 1 h at −80°C , the DNA was centrifuged 25 min at 20 , 000 g at 4°C , and the pellets were washed with cold 70% EtOH . The DNA was resuspended in 400 µl of 1×TE pH 8 . 0 , and transferred to eppendorf tubes for another phenol-chloroform extraction and precipitation with 0 . 1 volume of 3M NaOAc pH 5 . 2 ( 40 µl ) and 2 . 5 volumes of cold EtOH ( 1 . 1 ml ) . DNA was recovered by centrifugation ( 25 min at maximum speed at 4°C ) , and washed eight times with cold 70%EtOH . The pellets were then dissolved in 100 µl of 1×TE pH 8 . 0 , and incubated with RNAse A ( 1 µl at 10 mg/ml ) for 15 min at 37°C . This protocol was used to produce the 5C data for the distal limb , Shh−/− distal limb , and head shown in Figure 4 . The protocol is essentially the same as the one described for samples containing 2 to 10 million cells , with some modifications . Briefly , one million fixed cells were incubated for 15 min on ice in 200 µl of lysis buffer ( 10 mM Tris ( pH 8 . 0 ) , 10 mM NaCl , 0 . 2% NP40 supplemented with fresh protease inhibitor cocktail ) . Cells were then disrupted on ice with a dounce homogenizer ( pestle B; 2×20 strokes ) . Cell suspensions were transferred to eppendorf tubes and centrifuged 5 min at 2000 g . Supernatants were removed and the cell pellets were washed twice with 100 µl of 1×EcoRI buffer ( NEB ) . After the second wash , the cell pellet was resuspended in 50 µl of 1×EcoRI buffer . 1×EcoRI buffer ( 337 µl ) was added to each tube , and the mixture was incubated 10 min at 65°C with 0 . 1% SDS final ( 38 µl ) . Triton X-100 ( 44 µl of 10% Triton X-100 ) was added before overnight digestion with EcoRI ( 400 Units ) . The restriction enzyme was then inactivated by incubating 30 min at 65°C . Ligation was performed in 600 µl ( 450 µl of digestion product , 15 µl of 10% Triton-X-100 , 60 µl of ligase buffer , 6 µl of 10 mg/ml of BSA , 6 µl of 10 mM ATP , and 300 Cohesive end Units of T4 DNA ligase ) . Ligation was carried out at 16°C for 4 hours . 3C libraries were then incubated overnight at 65°C with 15 µl Proteinase K ( 10 mg/ml ) , and with an additional 15 µl Proteinase K the following day for 2 hours . The DNA was purified by one phenol and two phenol-chloroform extractions , and precipitated with 0 . 1 volume of 3M NaOAc pH 5 . 2 ( 64 µl ) and 2 . 5 volumes of cold EtOH ( 1740 µl ) . After at least 1 h at −80°C , the DNA was centrifuged 25 min at maximum speed at 4°C , and the pellets were washed once with cold 70% EtOH . The DNA was resuspended in 50 µl of 1×TE pH 8 . 0 , and incubated with RNAse A ( 1 µl at 10 mg/ml ) for 15 min at 37°C . As 3C products are quantified by PCR amplification of expected ligation junctions with different primer pairs , differences in PCR primer pair efficiencies must be corrected using control 3C libraries . Control libraries were generated from bacterial artificial chromosomes ( BACs ) as previously described [29] and contain equimolar ratios of all possible 3C contacts . Briefly , BAC clones covering the HoxA region ( mm9 , chr6: 51 , 946 , 668–52 , 656 , 241 ) , and one USP22 control region ( mm9 , chr11: 60 , 890 , 403–61 , 093 , 236 ) ) were mixed in equimolar ratio . Mixed BACs were digested with EcoRI and randomly ligated with T4 DNA ligase ( 5700 Cohesive end Units ) overnight at 16°C . BAC libraries were then purified by phenol-chloroform extraction . The libraries were generated with the following BACs: RP23-420L19 , RP24-359H1 , RP24-242G11 , RP-347D14 , RP23-305I5 ( Invitrogen , CHORI ) . These libraries were used only to correct primer pair efficiencies during 3C analysis and not in the 5C experiments . 3C primers were designed using the ‘3CPrimer’ program ( http://dostielab . biochem . mcgill . ca ) , and sequences are listed in Table S2 . Three reactions using the control BAC library and three reactions using each 3C library were generated for each primer pair . The PCR conditions were described elsewhere [29] . 3C PCR products were resolved on agarose gel containing ethidium bromide and quantified using a ChemiDoc XRS system featuring a 12-bit digital camera and the Quantity One computer software ( version 4 . 6 . 3; BioRad ) . Interaction frequencies ( IF ) were measured by dividing the value of each template PCR reactions by the value of each of the three control PCR reactions . The nine values were then average to determine the normalized interaction frequency . Three biological replicates were averaged after normalization for the wt limb and head . Normalization between different libraries was done using the compaction profiles for the USP22 region and an intergenic region within HoxA region as a reference . 5C primers covering the USP22 region ( mm9 , chr11: 60 , 917 , 307–61 , 017 , 307 ) and the HoxA region ( mm9 , chr6: 52 , 099 , 908–53 , 050 , 000 ) were designed using ‘my5C . primer’ [61] and the following parameters: optimal primer length of 30 nt , optimal TM of 65°C , default primer quality parameters ( mer:800 , U-blast:3 , S-blasr:50 ) . The sequences of these primers are listed in Table S3 and S4 . Primers were not designed for large ( >20 kb ) and small ( <100 bp ) restriction fragments . Low complexity and repetitive sequences were excluded from our experimental designs such that not all fragments could be probed in our assays . Primers with several genomic targets were also removed . The universal A-key ( CCATCTCATCCCTGCGTGTCTCCGACTCAG- ( 5C-specific ) ) and the P1-key tails ( ( 5C-specific ) -ATCACCGACTGCCCATAGAGAGG ) were added to the Forward and Reverse 5C primers , respectively . Reverse 5C primers were phosphorylated at their 5′ ends . Two experimental designs were used in our study . In the “cluster R” design ( anchored 5C scheme , Figure 2 , Figure S1 , Figure 4C , Figure S3 ) , Reverse 5C primers covered the HoxA cluster while Forward 5C primers tiled the surrounding upstream region . In this design , we used 142 Forward and 39 Reverse 5C primers ( 133 Forward/30 Reverse for the HoxA region , 9 Forward/9 Reverse USP22 region ) . In the “FR” design ( alternating 5C scheme , Figure 5 , Figure S3 ) , alternating Forward and Reverse 5C primers covering the entire HoxA region were used to generate the 5C libraries . This design used 194 primers ( 86 Forward/90 Reverse for the HoxA region , 9 Forward/9 Reverse USP22 region ) . Primer sequences are listed in Table S3 ( anchored “R” design ) and S4 ( alternating “FR” design ) . 5C libraries were prepared and amplified with the A-key and P1-key primers following a procedure described previously [30] . Briefly , 3C libraries were first titrated by PCR for quality control ( single band , absence of primer dimers , etc . ) , and to verify that contacts were amplified at frequencies similar to what is usually obtained from comparable libraries ( same DNA amount from the same species and karyotype ) [29] , [62]–[63] . We also verified the quality of the 3C libraries by generating a compaction profile in the USP22 region . In general , we used approximately 1 . 5 µg of 3C library per 5C ligation reaction when the libraries were generated from a large number of cells ( 2×106 to 107 cells ) . When 3C libraries were generated from a small cell number ( 106 cells ) , we used approximately 1 µg of DNA . Before adding the 3C libraries to the reaction tubes , 5C primer stocks ( 20 µM ) were diluted individually in water on ice , and mixed to a final concentration of 0 . 002 µM . Mixed diluted primers ( 1 . 7 µl ) were combined with 1 µl of annealing buffer ( 10×NEBuffer 4 , New England Biolabs Inc . ) on ice in reaction tubes . Salmon testis DNA ( 1 . 5 µg ) was added to each 5C reaction , followed by the 3C libraries and water for a final volume of 10 µl . Samples were denatured at 95°C for 5 min , and annealed at 55°C for 16 hours . Ligation with Taq DNA ligase ( 10 U ) was performed at 55°C for one hour . One tenth ( 3 µl ) of each ligation was then PCR-amplified individually with primers against the A-key and P1-key primer tails . We used 28 cycles based on dilution series showing linear PCR amplification within that cycle range . The products from 2 ( for the 3C libraries prepared from a large number of cells ) to 8 ( for the 3C libraries prepared from 106 cells ) PCR reactions were pooled before purifying the DNA on MinElute columns ( Qiagen ) . 5C libraries were quantified on agarose gel and diluted to 0 . 0534 ng/µl ( for Xpress Template Kit v2 . 0 ) or 0 . 0216 ng/µl ( for Ion PGM Template OT2 200 kit ) . One microliter of diluted 5C library was used for sequencing with an Ion PGM Sequencer . Samples were sequenced onto Ion 316 Chips following either the Ion Xpress Template Kit v2 . 0 , and Ion Sequencing Kit v2 . 0 protocols , or the Ion PGM Template OT2 200 Kit , and Ion PGM Sequencing 200 Kit v2 . 0 protocols as recommended by the manufacturer's instructions ( Life Technologies ) . Analysis of the 5C sequencing data was performed as described earlier [30] . The sequencing data was processed through a Torrent 5C data transformation pipeline on Galaxy ( https://main . g2 . bx . psu . edu/ ) . Briefly , the data was mapped against a customized reference file with TMAP . The reference file contained a list of all possible contacts between Forward and Reverse 5C primers covering our regions . The data was then filtered to remove low-quality reads ( MAQ quality score of lower than 30 ) , reads aligning more than two nucleotides away from the reference sequence start site , and reads which do not contain EcoRI restriction sites . This analysis generates an excel sheet containing interaction frequency lists ( IFL ) as well as a text file , which was used to visualize results using ‘my5C-heatmap’ [61] . Limb-enriched 5C interactions were obtained by subtracting limb and head 5C-seq data . Data was normalized by dividing the number of reads of each 5C contact by the total number of reads from the corresponding sequence run . All scales correspond to this ratio multiplied by 103 . The number of total reads and of used reads is provided for each experiment in Table S5 . 5C data are provided in Tables S6 to S20 and can be downloaded from our website: http://dostielab . biochem . mcgill . ca/ The limb p300 and H3K27Ac datasets ( Acc . No . GSE13845 and GSE30641 ) are from E11 . 5 embryos , and were downloaded from the Gene Expression Omnibus ( GEO ) website http://www . ncbi . nlm . nih . gov/geo/ . The my5C-primer and my5C-heatmap bioinformatics tools can be found at http://3dg . umassmed . edu/my5Cheatmap/heatmap . php | Hox genes encode transcription factors with crucial roles during development . These genes are grouped in four different clusters names HoxA , B , C , and D . Mutations in genes of the HoxA and D clusters have been found in several human syndromes , affecting in some cases limb development . Despite their essential role and contrary to the genes of the HoxD cluster , little is known about how the HoxA genes are regulated . Here , we identified a large set of regulatory elements controlling HoxA genes during limb development . By studying spatial chromatin organization at the HoxA region , we found that the regulatory elements are spatially clustered regardless of their activity . Clustering of enhancers define tissue-specific chromatin domains that interact specifically with each other and with active genes in the limb . Our findings give support to the emerging concept that chromatin architecture defines the functional properties of genomes . Additionally , our study suggests a common constraint of the chromatin topology in the evolution of HoxA and HoxD regulation in the emergence of the hand/foot , which is one of the major morphological innovations in vertebrates . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2013 | Clustering of Tissue-Specific Sub-TADs Accompanies the Regulation of HoxA Genes in Developing Limbs |
Progression of a cell through the division cycle is tightly controlled at different steps to ensure the integrity of genome replication and partitioning to daughter cells . From published experimental evidence , we propose a molecular mechanism for control of the cell division cycle in Caulobacter crescentus . The mechanism , which is based on the synthesis and degradation of three “master regulator” proteins ( CtrA , GcrA , and DnaA ) , is converted into a quantitative model , in order to study the temporal dynamics of these and other cell cycle proteins . The model accounts for important details of the physiology , biochemistry , and genetics of cell cycle control in stalked C . crescentus cell . It reproduces protein time courses in wild-type cells , mimics correctly the phenotypes of many mutant strains , and predicts the phenotypes of currently uncharacterized mutants . Since many of the proteins involved in regulating the cell cycle of C . crescentus are conserved among many genera of α-proteobacteria , the proposed mechanism may be applicable to other species of importance in agriculture and medicine .
C . crescentus is a dimorphic bacterium that inhabits freshwater , seawater , and soils , where it plays an important role in global carbon cycling by mineralizing dissolved organic materials [26] . C . crescentus normally undergoes an asymmetric cell division resulting in two different progeny cells ( Figure 1 ) : a motile , flagellated swarmer cell and a sessile stalked cell [22 , 23 , 27] . The nascent stalked cell then enters immediately into a new round of cell division and produces , about 90–120 min later , a new swarmer cell . The nascent swarmer cell swims around for 30–45 min before it differentiates into a stalked cell and initiates the DNA replication–division cycle . In this paper , we restrict our attention to the division cycle of stalked cells . Figure 2 depicts central elements of the cell division regulatory network in C . crescentus . Caulobacter crescentus has 3 , 767 protein-encoding genes [28] , of which 553 are cell cycle regulated [29] . Two master-regulator proteins control more than 25% of cell cycle–regulated genes: the transcription factor CtrA [30] directly regulates 95 genes ( including divK , ccrM , podJ , pleC , ftsZ , and ftsQ ) [31 , 32] , whereas GcrA controls 49 genes [15 , 29 , 32] . There is also weak evidence from microarray data [32] that CtrA can up-regulate dnaA . In addition , DNA synthesis in C . crescentus is under direct control by CtrA [33–35] , which binds to the origin of DNA replication and inhibits initiation of DNA synthesis [30] . CtrA is present at a high level in swarmer cells , whereas in stalked cells , it changes from low to high level during the cell cycle [15 , 36 , 37] . The abundance and activity of the CtrA protein is regulated through gene transcription , protein degradation , and phosphorylation . Expression of ctrA is under control of two promoters , ctrA P1 and ctrA P2 [31 , 36 , 38] . The weaker ctrA P1 promoter is activated in the early stalked cell ( ∼35 min after the initiation of DNA replication [39 , 40] ) by GcrA protein [15] and inhibited by high levels of CtrA itself [36] . The stronger ctrA P2 promoter is activated later , in predivisional cells , by the CtrA protein itself [36] . In addition , the ctrA P1 promoter is only activated from a new strand of hemimethylated DNA [31 , 40] . The ctrA P2 promoter is not active in swarmer cells , even though these cells have high levels of CtrA [36] . Furthermore , expression from ctrA P2 is inhibited in predivisional cells by conditions that inhibit DNA replication [41] . These facts indicate that ctrA P2 has regulators other than CtrA itself [36] . Proteolysis of CtrA ( and CtrA∼P ) is significantly accelerated by the phosphorylated form of DivK protein , DivK∼P , via the ClpXP protease pathway [42] , or with the help of some other ( as yet unknown ) histidine phosphotransferases [43] . Recently , RcdA and CpdR proteins have been reported to be involved in CtrA degradation in combination with ClpXP [44 , 45] . When this proteolysis pathway is activated , the half-life of CtrA in vivo is 5 min or less [38] . CtrA is active when phosphorylated [46 , 47] , a reaction carried out by a histidine kinase , CckA [46 , 48] , and a histidine phosphotransferase , ChpT [49] . In addition , CtrA is also phosphorylated by a tyrosine kinase , DivL [50] . CtrA is rapidly dephosphorylated in vivo . The activity of CckA was shown recently to be down-regulated by a DivK∼P [44 , 45 , 49] , thereby linking the phosphorylation and proteolysis pathways of CtrA . But otherwise , how the kinase and phosphatase reactions are regulated to control the fraction of active CtrA is poorly understood . GcrA is an activator of components of the replisome and of the segregation machinery [15] , and also regulates genes such as ctrA , pleC , and podJ [15 , 19] . GcrA protein concentration varies through the cell division cycle , peaking early in the cycle in stalked cells and reaching its minimum in a swarmer cell , after cell division . The DNA replication-initiating protein , DnaA , is required for gcrA expression [18] . In addition , transcription of gcrA is repressed by the CtrA protein [15] . DNA replication proceeds in three phases: initiation , elongation , and termination . The origin of DNA replication ( Cori ) in C . crescentus has one potential binding site for DnaA , a protein involved in initiating DNA synthesis [51] . The DnaA binding site partially overlaps with five CtrA binding sites in Cori [33–35] . CtrA represses initiation of DNA replication [30] . Thus , DNA replication is only initiated when DnaA level is high and CtrA level is low . In addition , DNA replication cannot be re-initiated until the origin stie has been fully methylated [52 , 53] . These conditions prevail during the swarmer-to-stalked cell transition , and just after division in the stalked cell compartment [34] . Once initiated , DNA synthesis continues bidirectionally along the circular chromosome , with an average speed of ∼20 . 5 kb/min in minimal broth , finishing in the late predivisional cell [54] . Elongation of newly replicating DNA strands requires a complex machinery , many components of which are under GcrA control [15] . Several cell cycle–related genes ( ctrA , gcrA , dnaA , ftsZ , and ccrM ) have GANTC methylation sites in their promoters [19 , 31 , 40 , 52 , 53 , 55 , 56] . Hence , the expression of these genes may be sensitive to the methylation state of the promoter . DNA replication transforms a fully methylated gene ( both strands methylated ) into a pair of hemimethylated genes ( only one strand methylated ) . At some later time , the unmethylated strands become methylated by the action of CcrM to return the genes to the fully methylated state [53] . These methylation transitions may affect the expression of cell cycle–related genes [53] . Methylation of Cori is also necessary for initiating a new round of DNA synthesis [34] . These methylation effects provide feedback from the progression of DNA replication to the cell cycle control system . In C . crescentus and other α-proteobacteria , CcrM is the methyltransferase that accounts for methylation of newly synthesized DNA strands . ccrM transcription is activated by CtrA only from a hemimethylated chromosome for about 20 min , in a late predivisional cell ( its expression peaks at ∼105 min in the 150-min swarmer cell cycle ) [57] . Lon protease is required for CcrM degradation [58] . The half-life of CcrM is less than 10 min in vivo [39] . Thus , CcrM activity is limited to a short portion of the predivisional cell phase , just before cell division . The multicomponent Z-ring organelle , which forms and constricts at the mid-cell plane , plays an important role in compartmentation of the predivisional cell and its subsequent division [27] . Compartmentation lasts about 20 min [59] . After the late predivisional cell is divided into two progeny cells , the Z-ring is disassembled and degraded . The Fts proteins ( FtsZ , FtsQ , FtsA , and FtsW ) have been identified as crucial elements of the Z-ring . ftsZ expression is positively and negatively regulated by CtrA [29 , 60] , and it may also by regulated by DNA methylation since the ftsZ promoter has a methylation site [40 , 53] . The ftsQ gene is expressed only after CtrA-mediated activation in the late predivisional cell [41] . The FtsQ protein localizes predominantly to the mid-cell plane of the predivisional cell , consistently with the appearance of the Z-ring [61 , 62] . The FtsA protein exhibits the time course similar to FtsQ [61] . divK transcription is activated by CtrA in late predivisional cells , which results in a slight elevation of DivK protein , otherwise present throughout the cell cycle at a nearly constant level [42 , 63] . The total amount of DivK∼P , the form that promotes CtrA degradation , does not appear to undergo dramatic changes during the cell cycle . It is 50% ± 20% lower in swarmer cells than in predivisional cells [63] . However , DivK and DivK∼P are dynamically localized during the cell division cycle [63–68] . Membrane-bound proteins DivJ and PleC , which localize at stalked and flagellated cell poles , respectively , regulate this process [64 , 65] by having opposite effects on DivK phosphorylation . DivJ is a kinase that continuously phosphorylates DivK at the stalked cell pole , and PleC promotes the continuous dephosphorylation of DivK∼P at the flagellated cell pole [64 , 67] . Hence , opposing gradients of DivK and DivK∼P are established between the two cell poles . Full constriction of the Z-ring disrupts the diffusion of DivK between the two poles [59 , 64] . As a result , DivK∼P accumulates in the nascent stalked cell compartment and unphosphorylated DivK accumulates in the nascent swarmer cell compartment . High DivK∼P promotes CtrA degradation in the stalked cell compartment [42 , 43] , whereas high CtrA is maintained in the swarmer cell compartment [16] . The nonuniform distribution of DivK and DivK∼P , and their corresponding effects on CtrA degradation , contribute largely to the different developmental programs of swarmer and stalked cells in C . crescentus . In addition , recent investigations indicate that CtrA phosphorylation is also at least partially under the control of DivK∼P ( as mentioned above ) , which shows that DivK∼P not only controls the stability of CtrA , but also its activity [44 , 45] .
To simulate the molecular regulation of a wild-type stalked-cell division cycle , we solve the equations in Table 1 subject to the parameter values and initial conditions in Tables 2 and 3 . Figure 4 illustrates how scaled protein concentrations and other variables of the model change during repetitive cycling of a stalked cell . The duration of a wild-type stalked-cell division cycle in our simulations is 120 min ( ∼90 min for S phase and ∼30 min for G2/M phase ) , as typically observed in experiments [22 , 23 , 59] . The main physiological events of the division cycle can be traced back to characteristic signatures of protein expression , as described in the Introduction . The division cycle starts with initiation of DNA replication ( Figure 4A ) from a fully methylated origin site by elevated DnaA , when CtrA is low and GcrA is sufficiently high ( to induce production of required components of the replication machinery ) ( Figure 4C and 4D ) . Immediately after DNA replication starts , Cori is hemimethylated . As DNA synthesis progresses , certain genetic loci become hemimethylated in order along the chromosome ( Figure 4B ) . Consequently , the regulatory proteins are produced and reach their peak concentrations sequentially . By contrast , dnaA expression seems to be activated by full methylation [55] , so its expression rate declines immediately after DNA replication starts . The effect of methylation on dnaA expression is minor compared to the regulatory signals coming from GcrA and CtrA . When the replication fork passes the ccrM locus , the gene becomes available for transcription , but is not immediately expressed , because CtrA level is low . In a predivisional cell , at approximately 35 min after start of DNA replication , the replication fork passes the ctrA gene ( Figure 4B ) , and its expression is immediately activated by GcrA ( Figure 4C ) and then further up-regulated by CtrA itself . Later on , when CtrA level becomes high , expression of the ccrM gene and , later , hemimethylated fts genes ( at ∼65 min ) , are expressed by the activation from high-level CtrA ( Figure 4D ) . High CtrA down-regulates gcrA expression . When DNA replication is finished , the new DNA strands are methylated by elevated CcrM in about 20 min . DNA methylation shuts down production of CtrA , CcrM , and Fts proteins . Meanwhile , elevated Fts proteins promote Z-ring formation and constriction ( Figure 4D ) , which separates the predivisional cell into two compartments , thereby restricting access of DivK and DivK∼P to only one of the old poles of the cell . As a result , in the stalked cell compartment , most DivK is converted into DivK∼P , accelerating CtrA proteolysis there ( Figure 4C ) . In a nascent stalked cell , low CtrA concentration releases gcrA expression , and GcrA protein level rises . Then , low CtrA , high GcrA , and high DnaA drive the nascent stalked cell into a new round of DNA synthesis from the fully methylated chromosome . These computed properties of the model agree reasonably well with what is known ( or expected ) about cell cycle progression in C . crescentus . In Figure 5 , we compare our simulation with experimental data . The data , collected from literature , were obtained by different research group with various experimental techniques . In most cases , experimental uncertainties of the data were not reported , but it is reasonable to assume that the error bounds are quite generous . Therefore , based on a visual comparison , we conclude that our model is in reasonable agreement with experimental observations . The only serious objection that might be raised is to our simulation of DivK∼P ( Figure 5C , green curve ) , which increases rapidly in the stalked-cell compartment after the Z-ring closes and DivK∼P is cut off from its phosphatase at the swarmer cell pole . Jacobs et al . [62] reported roughly constant levels of DivK∼P in predivisional stalked cells , i . e . , until just before Z-ring constriction , and significant differences of DivK∼P levels between stalked cells and swarmer cells . Our waveform for DivK∼P is consistent with this report and predicts that there should be a distinct peak of DivK phosphorylation in the stalked cell compartment at the end of the division cycle . This peak seems to be an inevitable consequence of the current belief that , upon Z-ring constriction , DivK becomes dephosphorylated in the swarmer cell compartment and remains heavily phosphorylated in the stalked cell compartment . The phenotypes of mutant cells provide crucial hints for deciphering the biochemical circuitry underlying any aspect of cell physiology . A mathematical model must be consistent with known phenotypes of relevant mutants . To make this test , we simulate cell cycle mutants of C . crescentus using exactly the same differential equations , parameter values , and initial conditions as for wild-type cells ( Tables 1 , 2 , and 3 ) , except for those modifications to parameters dictated by the nature of the mutation ( Table 4 ) . Our simulations of 16 classes of mutants are in agreement with experimentally observed phenotypes , as described here .
We propose ( Figure 3 ) a realistic mechanism for regulating the cell division cycle of stalked cells of C . crescentus . The mechanism includes three master-regulatory proteins ( GcrA , DnaA , and CtrA ) , a DNA methylase ( CcrM ) , Z-ring components ( Fts proteins ) , and an end-of-cycle protein ( DivK ) in its inactive and active ( phosphorylated ) forms . Cytokinesis is represented by a phenomenological variable that describes the extent of constriction of the Z-ring . DNA synthesis is described in terms of initiation , elongation , and termination . We assume that initiation of DNA replication requires high DnaA and GcrA , low CtrA , and full methylation of the origin site , and that the rate of DNA elongation is independent of DnaA , GcrA , and CtrA , and is almost linear . Transcription of some genes occurs only from an unmethylated DNA sequence; hence , the expression of such genes depends on their location on the newly synthesized DNA strand . Compartmentation in the predivisional cell is assumed to result in localization of phosphorylated DivK to the stalked compartment of the dividing cell , promoting CtrA degradation there . These assumptions are formulated as a mathematical model ( Table 1 ) consisting of 16 nonlinear , ordinary differential equations for seven proteins , the state of the Z-ring , the progression of DNA synthesis , and the methylation state of five gene sites on the DNA . The rate equations entail 44 parameters ( rate constants , binding constants , and thresholds; Table 2 ) that need to be determined by fitting the model to specific experimental observations . For the present , parameter estimation is done by trial and error , so we can only claim that our model equations and parameter set are sufficient to account for many properties of cell cycle control in C . crescentus . Because we fit the model to many different mutant phenotypes , we have a wealth of data to fix the parameters and to provide meaningful confirmation of the mechanism . Table 2 is in no sense an optimal parameter set , nor can we quantify how robust the system is , although our experience suggests that the model is quite hardy . Our present model is based heavily on an earlier conjecture [17] that the C . crescentus cell cycle is controlled by a bistable switch , created by positive feedback in the molecular circuitry of the ctrA gene . In that conjecture , the switch is flipped from the off-state ( CtrA low ) to the on-state ( CtrA high ) by GcrA accumulation as cells enter S phase , and then switched back to the off-state by DivK activation ( phosphorylation ) as cells divide ( the CtrA–DivK negative feedback loop ) . The original model did not account for the ways in which gene expression is linked to DNA methylation , thereby anchoring the protein interaction network to the progression of DNA replication forks . By incorporating DNA synthesis and methylation into the Brazhnik–Tyson model , the present model provides a more satisfactory account of cell cycle regulation in C . crescentus , and it can be tested by comparison to a broad spectrum of mutant phenotypes . Because the new model successfully reproduces the behavior of wild-type and mutant cells in many quantitative details , we conclude that our present understanding of the control system ( Figure 3 and Table 1 ) , properly interpreted , is accurate and adequate . On the other hand , the proposed mechanism must be considered as an evolving hypothesis that will be continually examined , revised , and improved as new observations tell us more about the control system . Some obvious improvements to the model include refined criteria for DNA initiation , regulated phosphorylation of CtrA , spatial localization of proteins , inclusion of a swarmer cell compartment , and an account of the swarmer-to-stalk cell transition . Finally , most of division-control proteins ( such as CtrA , DivK , CcrM , FtsZ , and FtsQ ) are conserved among α-proteobacteria [72] , suggesting that the computational model proposed here for C . crescentus may prove applicable to other types of α-proteobacteria , including symbiotic nitrogen-fixing genera ( Rhizobia ) and pathogenic genera ( Brucella spp . , Coxiella spp , etc . )
To understand the molecular logic of cell cycle regulation in C . crescentus , we constructed a mathematical model of the temporal dynamics of the regulatory genes and proteins . Following standard rules of chemical kinetics , we converted the wiring diagram in Figure 3 into a set of rate equations describing the temporal dynamics of the model . Justification of our approach is described in detail in [17] . Our model includes: Seven proteins: DnaA , GcrA , CtrA , CcrM , DivK ( inactive ) , and DivK∼P ( phosphorylated , active form ) , and a “representative” Fts protein . Two phenomenological variables , Z ( the state of closure of the septal Z-ring ) and I ( introducing a delay between activation of ccrM transcription and later activation of CcrM protein production ) . The progression of DNA replication ( including initiation , elongation , and termination ) and its methylation ( including probabilities of hemimethylation of ccrM , ctrA , dnaA , and fts genes , and of the replication origin site , Cori ) . Accordingly , our mathematical model consists of 16 nonlinear differential equations presented in Table 1 , including 28 kinetic constants ( k's ) , 11 binding constants ( J's ) , and five thresholds ( θ's ) . Our choice of parameter values is given in Table 2 . A common trend in developing complex models in molecular cell biology is to start from a simple coarse-grained ( “phenomenological” ) model and then refine and expand it step by step ( as data become available ) into an increasingly more comprehensive model . ( A good example is the progression of models of the budding yeast cell cycle [2 , 4 , 73] . ) We have taken this approach in our study of the C . crescentus cell cycle . We have limited the scope of our model so that it can be based largely on experimental observations , is not overwhelmed with assumptions , and is able to make predictions . Obviously , at any stage of modeling there will be facts that have not yet been incorporated and thus are out of the scope of the model . Our modeling assumptions are described here . First , we propose to model , at this stage , only the average behavior of cells and do not address naturally occurring fluctuations in cell cycle progression . Second , the rise of DivK∼P in stalked compartments after constriction of the Z-ring is a necessary , but not sufficient , condition for CtrA degradation . In our coarse-grained model of CtrA proteolysis , we use DivK∼P as a signal for starting rapid degradation of CtrA . In other words , DivK∼P determines when the degradation of CtrA is turned on , but the how ( the machinery that degrades CtrA , involving RcdA , CpdR , and ClpXP ) is assumed to be there when needed and is not modeled at present . Third , CtrA is activated by phosphorylation ( by kinases CckA and DivL ) , and a complete model of the Caulobacter cell cycle should take this into account . Unfortunately , little is known about the phosphorylation and dephosphorylation of CtrA and how these processes are temporally regulated . During the division cycle of wild-type cells , the levels of CtrA and CtrA∼P rise and fall together [22 , 46] , so we need not distinguish between the two forms . Therefore , in the current model , we keep track of CtrA synthesis and degradation only , assuming that CtrA∼P is a fixed fraction of total CtrA . This assumption , though a great oversimplification , is harmless enough for most of the mutants we consider in this paper . But it seems to cause serious problems for exactly those mutants ( ctrAop , ctrAΔ3 , ctrAD51E , and ctrAD51EΔ3 in wild-type background ) that interfere with normal synthesis , degradation , or activation of CtrA [34] . Later versions of the model will have to include CtrA∼P as a variable , when we have a better of idea of the mechanisms controlling CtrA phosphorylation . It is known that DivK∼P promotes the proteolysis of CtrA∼P [42] and negatively regulates CckA activity , thereby reducing phosphorylation of CtrA [49 , 74] . Hence , DivK∼P works to eliminate CtrA∼P activity by two independent pathways . We lump these two effects together as a single DivK∼P promoted reaction for removing active CtrA . Fourth , the dnaA locus is very close to the origin site ( Cori ) [28] . Within its promoter , potential CtrA and DnaA boxes and methylation sites exist for regulating its expression [20 , 34 , 52 , 55] . GcrA is a repressor for dnaA expression [15] , and CtrA seems to be an activator [32] . However , DnaA protein concentration varies very little during the Caulobacter cell cycle [55] . Although we include the regulatory signals in the model , they do not much affect the dynamics of a stalked cell because DnaA level is nearly constant throughout the cell cycle due to DnaA's long half-life . Fifth , initiation of DNA replication is triggered by the combined conditions of low CtrA , high DnaA , and fully methylated DNA origin site . In addition , initiation requires sufficient replication machinery , which is correlated to a high level of GcrA . We combine these regulatory effects into a single term . We assume that once initiation of DNA replication is successful , DNA elongation starts immediately . Elongation of new DNA strands is linear in time until it finishes , based on experimental data indicating that the speed of DNA replication in C . crescentus is almost constant [54] . Sixth , full constriction of the Z-ring requires accumulation and activation of a number of proteins , including FtsZ , FtsQ , FtsA , and FtsW , some of which are stimulated by CtrA . To simplify the model , we use Fts as a combined component to relay the signal from CtrA to Z-ring constriction . The transition from Z-ring open ( = 1 ) to fully constricted ( = 0 ) is modeled as a Goldbeter-Koshland ultrasensitive switch [75] . Seventh , we include the effects of DNA methylation on gene expression in our model because these effects mediate important feedback loops between DNA synthesis and the master regulatory proteins , and because DNA methylation can be a useful target for new drug development . In our model , the genes ccrM , dnaA , ctrA , and fts as well as the origin of DNA replication are regulated by methylation . Methylation plays a minor role in the regulation of GcrA production [19] , so we disregard it in our model . We allow a modest contribution of DNA methylation to regulating the production of DnaA . ccrM gene expression is significantly affected by its methylation state [40 , 57] . The activity of ctrA-P1 is known to depend on hemimethylation [36] , and the activity of ctrA-P2 seems to depend in some other way on DNA replication [37] . For simplicity , we assume that both ctrA promoters are turned on by hemimethylation of the gene . Among fts genes , the ftsZ promoter has a methylation site [40 , 53] , but the ftsQ promoter does not [41] . Scanning the ftsQ gene for the consensus sequence GANTC using the Regulatory Sequence Analysis Tools ( http://rsat . ulb . ac . be/rsat/ ) , we found a GAGTC segment in the coding sequence , suggesting that the ftsQ gene might also be affected by methylation . Since our “Fts” variable is a combination of Fts proteins , we conclude that our fts gene should be regulated by methylation . The effects of methylation on gene promoters and Cori are described by probabilities to be methylated or hemimethylated during the cell cycle . The probabilities ( h . . variables ) are in turn controlled by the progression of DNA replication and by the activity of CcrM [52 , 53] . Eighth , ccrM transcription is tightly regulated by CtrA protein , but accumulation of CcrM protein shows a noticeable delay from the transcriptional activation of its gene [37] , resulting in delayed activation of DNA methylation [57] . This delay is mimicked in our model by an intermediate variable I in the CtrA-to-CcrM pathway . Ninth , we recognize the importance of spatial controls in the Caulobacter cell cycle . However , at this stage , we are trying to model the stalked cell cycle as far as possible without explicitly tracking the spatial localization of regulatory proteins . That would require a more sophisticated mathematical framework and is planned for the next stage of the model . As the result of this simplification , our model makes no distinction between the stalked and swarmer parts of the predivisional cell . Right after compartmentation and before cytokinesis , we keep track of proteins in the stalked cell compartment only . At this stage , the distinction between swarmer and stalked cells is made by the phosphorylation state of DivK ( being completely phosphorylated in the stalked compartment ) . Tenth , we assume cells grow steadily in time , with a mass-doubling time of about 120 min and with the accumulated material shed at each division in the swarmer cell . In the present model , there is no coupling between cell growth and division , as in our models of eukaryotic cell proliferation [10] . Hence , there is no need for us to keep track of cell size , except to notice that if cell division is delayed or blocked , then the stalked cell will grow longer than normal and eventually be described as having a filamentous morphology . Parameter values for our model ( Table 2 ) were determined from available experimental data , wherever possible . Rate constants of degradation were estimated from experimentally observed half-lives of proteins . Rate constants of protein synthesis were adjusted to fit variations of protein concentration observed in experiments . Parameter values of Z-ring dynamics were set to be consistent with observed durations of the open ( ∼100 min ) and constricted ( ∼20 min ) states of the Z-ring [59] . Rate constants of DivK phosphorylation and dephosphorylation were estimated from the difference of DivK∼P concentration before and after Z-ring closing in predivisional cells [63] . Successful initiation of DNA replication depends on satisfying four requirements: low CtrA ( [CtrA] < θCtrA ) , high DnaA ( [DnaA] > θDnaA ) , high GcrA ( [GcrA] > θGcrA ) , and a fully methylated origin site ( hcori < θCori ) . The thresholds were adjusted to position the onset of the S phase correctly in wild-type cells . Replication-fork progression ( elongation ) begins at each successful initiation ( [Ini] = 0 . 05 ) and stops when DNA replication is complete ( [Elong] = 1 ) . The constant rate of elongation is consistent with an 80-min delay for copying the chromosome . Due to the constant rate of DNA replication , those genes that must be hemimethylated in order to be transcribed will be expressed in a temporal sequence determined by their positions on the chromosome from the origin of replication [39 , 76] . To model this effect , the variable hgene is set to 1 ( hemimethylated ) when [Elong] = distance of gene from Cori . Some time thereafter , when CcrM activity is high , the hgene decays exponentially back to 0 ( fully methylated ) . Most Hill function exponents are assumed to be 2 , with a higher value ( nH = 4 ) where sharper switching was required . Initial conditions ( Table 3 ) were taken to represent the beginning of a stalked cell cycle in a wild-type cell . The phenotypes of relevant mutants were collected from the literature . To simulate each mutant , we use exactly the same equations ( Table 1 ) and parameter values ( Table 2 ) except for values of those parameters directly affected by the mutation ( Table 4 ) . Mutations are introduced in our model after 120 min of simulation of the wild-type cell . For gene deletion , the rate of synthesis of the corresponding protein is set to zero . For gene overexpression , an additional constant rate of synthesis of the corresponding protein is introduced into the equations , because proteins are typically overexpressed from an extra copy of the gene under control of an inducible promoter . For heat- or cold-sensitive mutants , the relevant rate constant ( s ) retains its wild-type value at the permissive temperature and is set to zero at the restrictive temperature . For partial deletions , the relevant parameter value is assumed to lie between 0% and 100% of the wild-type value , according to the experimental characterization of the mutation . Equations of the model were solved numerically with Matlab 2006a ( The MathWorks ) . Machine-readable files for reproducing our simulations are made available in Text S1 and on our Web site ( http://mpf . biol . vt . edu/research/caulobacter/pp/ ) . | The cell cycle is the sequence of events by which a growing cell replicates all its components and divides them more or less evenly between two daughter cells . The timing and spatial organization of these events are controlled by gene–protein interaction networks of great complexity . A challenge for computational biology is to build realistic , accurate , predictive mathematical models of these control systems in a variety of organisms , both eukaryotes and prokaryotes . To this end , we present a model of a portion of the molecular network controlling DNA synthesis , cell cycle–related gene expression , DNA methylation , and cell division in stalked cells of the α-proteobacterium Caulobacter crescentus . The model is formulated in terms of nonlinear ordinary differential equations for the major cell cycle regulatory proteins in Caulobacter: CtrA , GcrA , DnaA , CcrM , and DivK . Kinetic rate constants are estimated , and the model is tested against available experimental observations on wild-type and mutant cells . The model is viewed as a starting point for more comprehensive models of the future that will account , in addition , for the spatial asymmetry of Caulobacter reproduction ( swarmer cells as well as stalked cells ) , the correlation of cell growth and division , and cell cycle checkpoints . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"cell",
"biology",
"in",
"vitro",
"microbiology",
"computational",
"biology",
"eubacteria"
]
| 2008 | A Quantitative Study of the Division Cycle of Caulobacter crescentus Stalked Cells |
Considering that mutations in known prostate cancer ( PrCa ) predisposition genes , including those responsible for hereditary breast/ovarian cancer and Lynch syndromes , explain less than 5% of early-onset/familial PrCa , we have sequenced 94 genes associated with cancer predisposition using next generation sequencing ( NGS ) in a series of 121 PrCa patients . We found monoallelic truncating/functionally deleterious mutations in seven genes , including ATM and CHEK2 , which have previously been associated with PrCa predisposition , and five new candidate PrCa associated genes involved in cancer predisposing recessive disorders , namely RAD51C , FANCD2 , FANCI , CEP57 and RECQL4 . Furthermore , using in silico pathogenicity prediction of missense variants among 18 genes associated with breast/ovarian cancer and/or Lynch syndrome , followed by KASP genotyping in 710 healthy controls , we identified “likely pathogenic” missense variants in ATM , BRIP1 , CHEK2 and TP53 . In conclusion , this study has identified putative PrCa predisposing germline mutations in 14 . 9% of early-onset/familial PrCa patients . Further data will be necessary to confirm the genetic heterogeneity of inherited PrCa predisposition hinted in this study .
Prostate cancer ( PrCa ) is the most frequent non-cutaneous cancer diagnosed in men worldwide and the third leading cause of male cancer deaths in Europe [1] . Despite efforts in early detection and screening strategies [2] , PrCa is estimated to be responsible for the death of 27 , 540 men in the United States in 2015 [1] . Contrarily to other cancer types , very little is known about the genetic contribution to the 10–20% of PrCa cases with evidence of familial clustering [3] . In fact , besides age and race , family history is the only other well-established risk factor for PrCa [4] . While familial PrCa is defined by an aggregation of PrCa in families , hereditary prostate cancer ( HPC ) is characterized by a pattern of Mendelian inheritance associated with rare mutations in susceptibility genes [3 , 5] . First-degree relatives of a PrCa patient have a two-fold increased risk of developing the disease compared to the general population . The risk is even higher when the number of affected relatives increases and the age at diagnosis decreases [3 , 5 , 6] . The existence of a genetic component behind PrCa development is strengthened by the four-fold higher concordance rate of PrCa among monozygotic twins compared to dizygotic twins [7 , 8] . Linkage analysis and genome-wide association studies have pinpointed some loci associated with PrCa predisposition , but the majority has not been consistently reproduced [9] . In 2004 , a combined genome-wide linkage analysis of 426 families from four HPC studies identified a locus at 17q21-22 strongly associated with PrCa [10] . Despite previous reports linking mutations in BRCA1 ( at 17q22 ) with PrCa predisposition [11 , 12] , Ewing et al . later identified a rare but recurrent mutation ( G84E ) in the HOXB13 gene ( at 17q21 ) in up to 3% of the patients with both early-onset and family history of the disease , using a next-generation sequencing ( NGS ) approach covering the 202 genes present in the defined region of interest ( ROI ) at the 17q21-22 locus [13] . An increased risk of PrCa for the HOXB13 G84E mutation carriers has been confirmed by several groups [14 , 15] and other HOXB13 variants associated with PrCa have been found in other populations [16 , 17] . Besides HOXB13 , BRCA2 mutation carriers are also at increased risk of developing PrCa [18–20] . Overall , BRCA2 mutations seem to explain about 2% of early-onset PrCa cases [19] , a frequency that can be slightly higher for BRCA2 mutations with a founder effect in specific populations [21 , 22] . Additionally , a higher risk for PrCa in Lynch syndrome families has been proposed [23 , 24] , with some studies reporting a five- to ten-fold increased risk of PrCa development for carriers of MSH2 mutations compared to non-carriers [25 , 26] . However , recent studies of our group found germline mutations in HOXB13 , BRCA2 and MSH2 in only 1 . 5% of early-onset and/or familial PrCa cases [17 , 27] . Mutations in a few additional genes or specific variants , namely in CHEK2 [28–31] , NBN [32 , 33] , ATM [34 , 35] , and BRIP1 [36] , have been reported to increase the risk of PrCa , although some in a population-specific context . Despite these reports , the large majority of prostate carcinomas showing Mendelian inheritance still have no explanation concerning highly penetrant susceptibility variants . In this work , we aimed to evaluate the proportion of cases with early-onset and/or familial/hereditary PrCa that can be attributed to mutations in 94 genes associated with inherited cancer predisposition , using our validated targeted next generation sequencing ( NGS ) pipeline [37] . This approach allowed to identify functionally deleterious/“potentially pathogenic” mutations in nine genes , revealing six genes ( CEP57 , FANCD2 , FANCI , RAD51C , RECQL4 and TP53 ) not previously associated with PrCa predisposition . Overall , a candidate disease-causing mutation in a cancer predisposing gene was identified in 18 patients ( 14 . 9% ) , with ATM and CHEK2 representing 61 . 1% of the cases .
Of the genes previously reported to increase the risk for PrCa development ( after excluding cases with known mutations in HOXB13 , BRCA2 and MSH2 in this series ) , we found a nonsense mutation in ATM and a splicing mutation in CHEK2 ( Table 1 ) . The ATM mutation c . 652C>T , which leads to a premature stop codon at codon 218 , was found in a patient ( HPC177 ) with five brothers diagnosed with PrCa ( Fig 1A ) , including twin brothers diagnosed before the age of 61 years , thus fulfilling the A1 and A2 criteria ( see Material and Methods section for criteria description ) . The family is living abroad , which renders segregation analysis difficult to perform . The CHEK2 mutation c . 593-1G>T , predicted to affect the splice site by three of the four queried in silico predictors ( S1 Table ) and reported as “likely pathogenic” in ClinVar , was found in a patient ( HPC395 ) with a family history of three breast cancer ( BrCa ) cases , two of them diagnosed at early age ( Fig 1B ) , thus fulfilling the B3 criterion . One of the nices with BrCa is carrier of the CHEK2 mutation c . 593-1G>T . To strengthen the causality between these mutations and cancer development , we used KASP genotyping in 710 healthy controls and searched for the variant among 504 samples from non-prostate cancer cases analyzed with the same NGS panel and pipeline in the Department of Genetics of IPO Porto . Among the 504 cancer cases , the same ATM stop mutation was found in a patient diagnosed with bilateral BrCa at early age ( previously described [38] ) and the same CHEK2 splicing mutation was found in an early-onset breast and colon cancer patient . No carriers were found either among our 710 healthy controls or in ExAC , for both mutations . Considering that most of the genes so far associated with an increased risk for PrCa development have previously been described to predispose to breast/ovarian cancer and/or Lynch syndrome , we looked for missense variants in the 18 genes associated with these diseases , namely ATM , BLM , BRCA1 , BRCA2 , BRIP1 , CDH1 , CHEK2 , MLH1 , MSH2 , MSH6 , NBN , PALB2 , PMS2 , PTEN , RAD51C , RAD51D , STK11 , and TP53 . Missense variants predicted to be pathogenic by at least 12 of the 15 in silico pathogenicity predictors ( including at least three conservation tools ) were considered “potentially pathogenic” . Of the 42 missense variants found ( S2 Table [45] ) , ten variants fulfill these criteria ( Table 2 ) and include four variants in ATM , two in CHEK2 , and one in each of the genes BRIP1 , MSH2 , MSH6 and TP53 . The CHEK2 missense mutation c . 349A>G , found in two patients ( HPC188 and HPC289 ) , was the only mutation classified as “pathogenic/likely pathogenic” in ClinVar . Patient HPC188 has family history of PrCa , with three first- and second-degree relatives diagnosed at or before the age of 65 years ( one early-onset ) , thus fulfilling the A1 and A2 criteria ( S1A Fig ) . Patient HPC289 is an early-onset PrCa case , fulfilling the B1 and B3 criteria for having a heavy family history of cancer , with several cases diagnosed at early age ( S1B Fig ) . Both MSH2 and MSH6 variants are reported in public databases and classified as variants of unknown significance ( VUS ) in ClinVar . As mutations in Lynch syndrome predisposing genes are usually associated with loss of protein expression in the tumor , we performed immunohistochemistry for the MSH2 and MSH6 proteins in the prostate tumors of the patients HPC371 and HPC332 , respectively , and no loss of expression was found , rendering the MSH2 and MSH6 mutations as probably not associated with PrCa development in these patients . Apart from the CHEK2 missense mutation c . 349A>G , the remaining missense variants here identified were either not described in the literature or classified as VUS . To increase our understanding on the pathogenic potential of these variants , we screened our 710 healthy controls and the 504 non-prostate cancer samples as described above . For the ATM mutations c . 995A>G and c . 8560C>T and for the TP53 mutation c . 839G>A , no additional carriers were found . The BRIP1 mutation c . 847T>C and the CHEK2 mutation c . 695G>T were found in one of the 504 cancer cases ( each ) and the ATM mutations c . 1595G>A and c . 5750G>A were found in two cases ( each ) of the 710 healthy controls and in six and three cases , respectively , of the 504 cancer cases . Of all these variants , only the ATM mutation c . 8560C>T was found significantly increased in our PrCa patients comparing with our healthy controls ( P = 0 . 024; S3 Table ) , with the CHEK2 mutation c . 349A>G reaching borderline significance ( P = 0 . 057 ) . With the exception of the TP53 mutation c . 839G>A , not found in any of the 504 non-prostate cancer patients , all missense mutations have no significant frequency differences in cancer patients fulfilling criteria for other hereditary cancer syndromes ( P>0 . 05; S3 Table ) . When comparing the frequencies obtained in our PrCa patients with those of the Non-Finnish Europeans ( NFE ) described in ExAC , highly significant associations are obtained for all the ATM and CHEK2 missense mutations ( S3 Table ) . Following the guidelines from the American College of Medical Genetics and Genomics and the Association for Molecular Pathology ( ACMG-AMP ) for variant interpretation and classification [46] using InterVar [47] , all the missense variants here identified in ATM , CHEK2 , BRIP1 and TP53 are classified as VUS . Adding the PS3 criterion ( “well-established in vitro or in vivo functional studies supportive of a damaging effect on the gene or gene product” ) to the classification of the CHEK2 variant c . 349A>G [48 , 49] and the PS4 criterion ( “the prevalence of the variant in affected individuals is significantly increased compared with the prevalence in controls” ) to the classification of the ATM variant c . 8560C>T , supports the “likely pathogenic” nature of these variants in PrCa development . Of the 121 cases enrolled in this study , 45 have criteria to be classified as familial/hereditary PrCa ( A group ) and 86 are cases of early-onset PrCa and/or PrCa associated with clustering of other cancers in the family ( group B ) , with ten cases fulfilling both A and B criteria ( S4 Table ) . Regarding age at diagnosis , 64 cases ( 52 . 9% ) were diagnosed with PrCa at or before the age of 55 years , thus being considered early-onset PrCa cases . Considering the number of cases with prostate carcinomas in the 121 families , 91 cases ( 75 . 2% ) have family history of two or more relatives with PrCa , with 27 cases ( 22 . 3% ) having three and 33 cases ( 27 . 3% ) having at least four . When comparing clinicopathological characteristics of the patients harboring the deleterious/”potentially pathogenic” mutations ( n = 18; excluding the cases with the MSH2 and MSH6 VUS , described above ) with the “negative” group ( n = 103 ) , no statistically significant associations were observed , either considering all cases or considering the subgroups of cases with familial/hereditary PrCa or early-onset PrCa ( S5 Table ) . Consistent with the increasing chance of incidental findings of the NGS approaches , we found a c . 3846_3860del in-frame deletion in MSH6 that falls into this classification . This mutation was found in the patient carrying also the truncating mutation in RAD51C ( HPC186 ) and is classified as pathogenic in two of the Lynch syndrome families diagnosed at IPO Porto . However , we find unlikely its association with the PrCa in this patient , as no loss of MSH6 expression was observed in the tumor ( contrarily to what we observed in the colon carcinomas of our Lynch syndrome families ) . Analyses in the available relatives showed segregation of the variant in the niece with colon cancer ( S2 Fig ) .
In this work we used a NGS approach targeting the full coding-sequence of 94 genes associated with cancer predisposition to identify germline mutations in a selected series of 121 PrCa patients with early-onset disease and/or criteria for familial/hereditary PrCa , alone or associated with other cancers . This strategy is justifiable by the fact that , with the exception of HOXB13 , all the genes so far associated with PrCa hereditary predisposition were previously associated with an increased risk for BrCa , OvCa or other cancers , including those causing the phenotypically heterogeneous diseases hereditary breast/ovarian cancer ( HBOC ) and Lynch syndrome [20 , 24 , 27 , 50] . Using our previously established NGS analysis pipeline [37] , and after excluding the few cases in our series with germline mutations in HOXB13 or in genes associated with HBOC or Lynch syndrome [17 , 27] , we found monoallelic truncating/deleterious mutations in seven genes , of which only ATM and CHEK2 have been previously implicated in PrCa development [28–31 , 34 , 35 , 43] . Deleterious mutations in ATM and CHEK2 thus represent 0 . 8% ( each ) of the cases enrolled in this study , with the nonsense ATM mutation representing 2 . 2% of the cases fulfilling criteria for familial/hereditary PrCa ( A group ) , which resembles the frequency of mutations found in BRCA2 in earlier studies [27 , 35] . Curiously , both mutations occur in families with several BrCa cases and were both found in one case ( each ) of the 504 non-prostate cancer cases analyzed with the same NGS approach in the Department of Genetics for fulfilling criteria for HBOC . We found monoallelic functionally deleterious mutations in three genes of the FA family , namely RAD51C ( FANCO ) , FANCD2 and FANCI genes , the latter two not previously associated with cancer risk . RAD51C , a RAD51 paralog involved in the homologous recombination ( HR ) repair pathway [51] , was first described as a susceptibility gene for BrCa and OvCa , showing complete segregation in six families [44] . Nowadays , RAD51C deleterious mutations are established as a risk factor for OvCa only , with a prevalence of about 0 . 8% in familial OvCa and 0 . 4–1 . 1% in OvCa cases unselected for family history [42] . The family history of the patient harboring the c . 890_899del mutation in RAD51C has no confirmed ovarian cancer diagnosis , but includes a relative with gynecological cancer deceased at young age . FANCD2 and FANCI are involved in the initial steps of the FA pathway , leading to the activation of downstream repair factors , such as FANCD1 ( BRCA2 ) , FANCJ ( BRIP1 ) , FANCN ( PALB2 ) and FANCO ( RAD51C ) , to mediate HR [52] . Considering that mutations in all these four FA members have been associated with risk for BrCa and/or OvCa [41 , 43 , 44] , with mutations in BRIP1 and BRCA2 also associated with PrCa development [19 , 20 , 36] , our report of mutations in RAD51C , FANCD2 and FANCI may increase to five the list of FA members involved in PrCa predisposition . In our series , functionally deleterious mutations in FA genes represent 4 . 4% ( 2/45 ) of the familial/hereditary PrCa cases ( A criteria ) and 1 . 6% ( 1/64 ) of the early-onset PrCa cases . Among the 504 non-prostate cancer cases diagnosed at our institution with the same NGS approach , three carriers of the same mutations were found , one with the FANCD2 mutation and two with the RAD51C mutation , with different cancers occurring in the families . Further studies are required to determine the frequency of germline mutations in these genes in PrCa and other hormone-related cancers , as the pedigrees of both case HPC186 and case HPC447 , with the RAD51C and the FANCD2 functionally deleterious mutations , respectively , include relatives affected with BrCa and/or gynecological cancers . To our knowledge , this is also the first report of heterozygous germline truncating mutations in CEP57 and RECQL4 as possible cancer risk factors . CEP57 encodes a 57 kDa member of the CEP family of centrosomal proteins involved in MVA2 , a rare pediatric syndrome with high risk of development of childhood cancers [53 , 54] . On the other hand , RECQL4 belongs to a family of five RecQ helicases [RECQL1 , WRN ( RECQL2 ) , BLM ( RECQL3 ) , RECQL4 and RECQL5] [55] . Interestingly , monoallelic mutations in RECQL1 and in the Bloom syndrome gene BLM have been described as risk factors for BrCa [56 , 57] , although the BLM association has been contested by others [58 , 59] . According to the Uniprot database , the RECQL4 frameshift mutation c . 2636del we here describe is not expected to affect the known functional domains of the protein , but more downstream ( C-terminal ) mutations in RECQL4 have been shown to cause RTS or GBS [55] and the C-terminal seems to be necessary for RECQL4 nucleolar localization through interaction with PARP-1 [60] , therefore making very likely its deleterious nature . Additionally , the absence of both mutations in public databases , namely ExAC , and in the 504 non-prostate cancer cases analyzed in our institution with the same NGS approach , may reflect their PrCa specificity . In addition to the seven cases with truncating/deleterious mutations , we found “likely/potentially pathogenic” missense mutations in 11 PrCa families . Taking into account the diversity and general high concordance of the in silico tools that were considered for the prediction of variant pathogenicity ( S2 Table ) , along with the low frequency found among the 710 healthy control cases screened ( S3 Table ) and with the fact that other missense mutations in ATM , CHEK2 and TP53 have been linked with cancer development [61 , 62] , it is plausible that the variants in ATM , BRIP1 , CHEK2 and TP53 here identified may explain PrCa susceptibility in the families carrying them . The pathogenic nature of the CHEK2 mutation c . 349A>G found in two cases , was suggested in several studies , showing loss of DNA damage response and impaired activation due to lack of phosphorylation [48 , 49] . Furthermore , this CHEK2 variant has been found in three of 694 BRCA1/BRCA2-negative BrCa families , two from the United Kingdom and one from the Netherlands , being described as a moderate to low penetrance variant [63] . On the other hand , in a large case-control study gathering data from three consortia participating in the Collaborative Oncological Gene-environment Study ( COGS ) , the CHEK2 variant c . 349A>G was associated with increased BrCa risk ( odds-ratio 2 . 26 ) , but not with an increased risk for PrCa or OvCa [43] . Segregation analysis and/or phenotypic evaluation in vitro would be useful to complement the available information concerning the pathogenicity of this and the remaining missense variants here identified . For the variants found significantly associated ( or showing borderline significance ) with PrCa development , namely the ATM variant c . 8560C>T and the CHEK2 variant c . 349A>G , larger cohorts of familial/early-onset PrCa cases would be useful to define cancer risk estimates and the age-standardized PrCa risk attributed to these variants . Looking at the overlap between the patients harboring truncating/functionally deleterious mutations and those harboring “likely/potentially pathogenic” missense variants , the ATM variant c . 8560C>T , found in three patients ( HPC3 , HPC186 and HPC332 ) , is the only variant overlapping with other mutations , namely in patient HPC186 , who carries the RAD51C frameshift mutation , and in patient HPC332 , who carries the MSH6 c . 1729C>T variant . Immunohistochemistry analysis for MSH6 in the tumor of patient HPC332 showed normal MSH6 expression , thus reducing the likelihood that the MSH6 c . 1729C>T variant is a PrCa risk factor and rendering the ATM mutation c . 8560C>T the most likely risk variant in this patient . Excluding the case HPC186 ( with co-occurrence of the RAD51C frameshift mutation ) , ATM represents the most commonly mutated gene in our series , eventually explaining increased risk of PrCa in seven cases ( ~5 . 8% ) , with CHEK2 being the second most frequently mutated gene ( four cases , ~3 . 3% ) . Overall , functionally deleterious/“likely/potentially pathogenic” variants were found in 18 patients ( excluding the two families with missense mutations in MSH2 and MSH6 ) . Of these , eight patients ( 44 . 4% ) fulfill the A criteria and 12 ( 66 . 7% ) fulfill the B criteria ( two cases complying with both ) , representing 17 . 8% and 13 . 9% of the samples enrolled in each group . Seven of the 18 cases ( 38 . 9% ) were diagnosed at early age , representing 10 . 9% of the patients in the early-onset group . Comparing clinicopathological data from patients harboring these variants with the group of patients without an identified “potentially pathogenic” mutation , no statistically significant associations were found . In the context of this study , we identified one truncating variant in a gene that is included in the list of incidental findings recommended for return to patients after clinical sequencing by the guidelines of the American College of Medical Genetics and Genomics [64] . The previously unreported MSH6 in-frame mutation c . 3846_3860del predisposes to Lynch syndrome ( OMIM #120435 ) , as it has been classified as pathogenic in two Lynch syndrome families in our institution , with evidence that included demonstration of loss of expression restricted to MSH6 in the colorectal tumors of carriers , a pattern also observed in the colon cancer of a relative of this patient who is also carrier of this in-frame MSH6 variant . On the other hand , as no loss of MSH6 expression was observed in the prostate tumor , this MSH6 variant is unlikely to explain the PrCa predisposition in this family , which is most likely related to the RAD51C truncation mutation or the ATM missense mutation also found in this patient . Even though targeted sequencing was performed under a research protocol and not as part of clinical sequencing , this incidental finding was reported to the patient during genetic counseling , as recommended by the guidelines of the American College of Medical Genetics and Genomics , and appropriate follow-up is being offered to the family as judged clinically appropriate . In conclusion , we found functionally deleterious/“likely/potentially pathogenic” germline mutations in 18 of the 121 ( 14 . 9% ) familial/hereditary and/or early-onset PrCa cases selected for this study . To our knowledge , this study is the first to report functionally deleterious germline mutations in the three FA genes RAD51C , FANCD2 and FANCI , and in two genes until now only associated with recessive disorders , CEP57 and RECQL4 . Further data will be necessary to confirm the genetic heterogeneity of inherited PrCa predisposition hinted in this study .
This study is in accordance with the ethical standards of the Ethics Committee of the Portuguese Oncology Institute of Porto ( approval number 38 . 010 ) and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards . We selected 121 cases from our previously described series of 462 early-onset and/or familial/hereditary PrCa cases [17] , with two groups being considered: A ) cases with familial/hereditary PrCa , and B ) cases with early-onset PrCa and/or association with other types of cancer . Among the cases in the A group , three criteria were defined: 1 ) cases with at least three first-degree relatives with PrCa independently of the ages at diagnosis , 2 ) cases with two first-degree relatives with PrCa with average age at diagnosis ≤65 years and at least one of the affected cases diagnosed before the age of 61 , and 3 ) cases diagnosed before the age of 61 with at least two first- or second-degree relatives with PrCa and average age at diagnosis of the three younger cases ≤65 . Regarding the cases in the B group , three criteria were considered: 1 ) cases diagnosed before the age of 56 years with at least three first- or second-degree relatives diagnosed with cancer and average age at diagnosis of the three younger diagnoses ≤55 , 2 ) cases diagnosed with second primary cancers besides PrCa and 3 ) cases with relatives diagnosed with either early-onset and/or rare cancer types ( bilateral breast , male breast , brain ) and/or clustering of other cancer types ( e . g . breast , colon , or gastric cancers ) . Cases previously identified as harboring pathogenic mutations in known PrCa predisposing genes ( HOXB13 , BRCA2 and MSH2 ) were excluded from this case priorization [17 , 27] . DNA previously extracted from peripheral blood leucocytes by standard procedures [17] was quantified using Qubit Fluorometer ( Thermo Fisher Scientific , Waltham , MA , USA ) . We used as control samples 710 healthy individuals ( 391 males and 319 females; mean age 55 . 1 years; SD±9 . 4 years ) , including 528 blood donors ( 285 males and 243 females ) from the Portuguese Oncology Institute of Porto with no personal history of cancer at the time of blood collection and 182 healthy relatives ( 106 males and 76 females ) with negative predictive genetic testing ( each from independent families ) . We applied our previously established NGS pipeline [37] using the TruSight Rapid Capture target enrichment workflow and the TruSight Cancer panel , both from Illumina , Inc . ( San Diego , CA , USA ) . For variant analysis , sequences were aligned to the reference genome ( GRCh37/hg19 ) using three different alignment and variant calling software: Isaac Enrichment ( v2 . 1 . 0 ) , BWA Enrichment ( v2 . 1 . 0 ) and NextGENe ( v2 . 4 . 1; Softgenetics , State College , PA , USA ) , as previously described [37] . Briefly , for variant annotation and filtering , . vcf ( variant call format ) files from the three software were imported into GeneticistAssistant ( Softgenetics ) and filtered for variant frequency in our in-house database , excluding variants present in more than 10% of the cases . Additional variant selection included those with coverage >20x , alternative variant frequency between 30% and 70% ( excluding variants in mosaicism ) , and minor allele frequency ( MAF ) ≤0 . 1% [65 , 66] . Synonymous variants and intronic variants at more than 12-bp away from exon-intron boundaries were excluded . For MAF filtering , data was obtained from the 1000 Genomes Project [Based on Project Phase III Data [67]] , Exome Variant Server [from NHLBI Exome Sequencing Project ( http://evs . gs . washington . edu/EVS/ ) , accessed in January , 2017] and Exome Aggregation Consortium [ExAC ( http://exac . broadinstitute . org ) , accessed in January , 2017] databases , whenever available . Variants assigned as not pathogenic , likely not pathogenic , of no clinical significance or of little clinical significance , according to public databases , namely ClinVar ( http://www . ncbi . nlm . nih . gov/clinvar/ , accessed in January , 2017 ) , Breast Cancer Information Core [BIC ( https://research . nhgri . nih . gov/bic/ ) , accessed in January , 2017 ) ] , and InSiGHT ( via the Leiden Open-source Variation Database [LOVD ( http://www . lovd . nl/3 . 0/home ) , accessed in January , 2017] [68] ) , were discarded . All the variants identified were validated by Sanger sequencing . For this purpose , primers ( S6 Table ) were designed using the Primer-BLAST design tool from the National Center for Biotechnology Information ( NCBI ) [69] . For PCR amplification , an initial denaturation step was performed at 95°C for 15min , followed by 35 cycles with denaturation at 95°C for 30s , annealing at appropriate temperature ( 58–62°C ) for 30s and extension at 72°C for 45s . A final extension step at 72°C for 9min was included . For the sequencing reaction , the BigDye Terminator v3 . 1 Cycle Sequencing Kit ( Thermo Fisher Scientific ) was used , according to the manufacturer’s instructions , and samples were run in a 3500 Genetic Analyzer ( Thermo Fisher Scientific ) . For validation of the RECQL4 variant , primers and PCR conditions from Nishijo et al . were used [70] . The TP53 variant was validated following the IARC ( International Agency for Research on Cancer ) protocol for direct sequencing ( http://p53 . iarc . fr/ , update 2010 ) . Primers and PCR conditions for Sanger sequencing validation of MSH2 and MSH6 variants were kindly provided by Professor Michael Griffiths from the West Midlands Regional Genetics Laboratory , Birmingham Women’s NHS Foundation Trust , Birmingham , United Kingdom . To explore the functional consequence of truncating/deleterious variants , MutationTaster [71] and Uniprot [72] were queried . To infer the putative impact on splicing , the splice site predictors Human Splicing Finder 3 . 0 [73] , MaxEntScan [74] , NNSPLICE [75] and NetGene2 [76] were used . To predict the biological impact of missense mutations , we looked at data from the predictor tools embedded in the NGS Interpretative Workbench from GeneticistAssistant , which includes the functional predictors SIFT , PolyPhen2 , LRT , MutationTaster , PROVEAN , FATHMM , CADD , MutationAssessor , MetaLR , MetaSVM and VEST3 , and the conservation analysis tools PhyloP , GERP++ , PhastCons and SiPhy , as previously described [37] . To search for clinicopathological associations between mutation carriers and non-carriers , information on PSA at diagnosis , tumor staging and Gleason Score were gathered from medical records ( S4 Table ) and the Fisher’s exact test was used . To evaluate the frequency in the general Northern Portuguese population of the missense variants identified in our series of PrCa patients we used KASP technology genotyping ( KBioscience , Herts , UK ) in our series of 710 healthy individuals , following manufacturer’s recommendations . KASP assay primers ( S6 Table ) were designed using the Primer-BLAST design tool from NCBI and data were analyzed in the LightCycler 480 Software 1 . 5 . 0 . | Prostate cancer ( PrCa ) is the most frequent cancer diagnosed in men worldwide , estimated to be responsible for the death of 27 , 540 men in the United States in 2015 . Contrarily to other cancer types , the genetic contribution to the 10–20% of PrCa cases occurring in families with aggregation of the disease is largely unknown . Germline mutations in the BRCA2 and the MSH2 breast and colon cancer predisposing genes , respectively , explain only about 1 . 5% of our early-onset/familial PrCa cases . Taking advantage of recent deep sequencing technologies and an analysis pipeline established in our group , we have screened 121 PrCa patients with strong evidence of an hereditary component for mutations in 94 genes involved in several cancer predisposing syndromes . We found truncating/functionally deleterious mutations in seven genes and “likely pathogenic” missense variants in four genes , of which five and one , respectively , have not been previously associated with PrCa predisposition . We believe this study significantly contributes to the understanding of the genetic heterogeneity behind early-onset/familial PrCa . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"methods"
]
| [
"cancer",
"detection",
"and",
"diagnosis",
"deletion",
"mutation",
"medicine",
"and",
"health",
"sciences",
"cancer",
"risk",
"factors",
"genetic",
"diseases",
"oncology",
"mutation",
"hereditary",
"nonpolyposis",
"colorectal",
"cancer",
"nonsense",
"mutation",
"mutation",
"databases",
"frameshift",
"mutation",
"autosomal",
"dominant",
"diseases",
"research",
"and",
"analysis",
"methods",
"biological",
"databases",
"clinical",
"genetics",
"genetic",
"causes",
"of",
"cancer",
"diagnostic",
"medicine",
"heredity",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences"
]
| 2018 | Targeted next generation sequencing identifies functionally deleterious germline mutations in novel genes in early-onset/familial prostate cancer |
We used a bivariate ( multivariate ) linear mixed-effects model to estimate the narrow-sense heritability ( h2 ) and heritability explained by the common SNPs ( hg2 ) for several metabolic syndrome ( MetS ) traits and the genetic correlation between pairs of traits for the Atherosclerosis Risk in Communities ( ARIC ) genome-wide association study ( GWAS ) population . MetS traits included body-mass index ( BMI ) , waist-to-hip ratio ( WHR ) , systolic blood pressure ( SBP ) , fasting glucose ( GLU ) , fasting insulin ( INS ) , fasting trigylcerides ( TG ) , and fasting high-density lipoprotein ( HDL ) . We found the percentage of h2 accounted for by common SNPs to be 58% of h2 for height , 41% for BMI , 46% for WHR , 30% for GLU , 39% for INS , 34% for TG , 25% for HDL , and 80% for SBP . We confirmed prior reports for height and BMI using the ARIC population and independently in the Framingham Heart Study ( FHS ) population . We demonstrated that the multivariate model supported large genetic correlations between BMI and WHR and between TG and HDL . We also showed that the genetic correlations between the MetS traits are directly proportional to the phenotypic correlations .
Obesity associated traits such as central adiposity , dyslipidemia , hypertension , and insulin resistance are major risk factors for type 2 diabetes and cardiovascular complications [1] . The constellation of these traits has been termed metabolic syndrome ( MetS ) . Understanding the genetic factors underlying these traits and how they are correlated is clinically important . Large-scale genotyping investigations such as genome-wide association studies ( GWAS ) are useful tools for identifying genetic factors . However , significant genetic variants discovered in GWAS explain only a small proportion of the expected narrow-sense heritability , h2 , defined as the ratio of additive genetic variance to phenotypic variance [2] . This discrepancy underlies the debate concerning “missing” genetic factors among the common variants [3] , [4] . The main approach of GWAS has been to identify significant single-nucleotide polymorphisms ( SNPs ) by examining each SNP individually for significance . The h2 attributed to that marker is then given by 2f ( 1−f ) a2 , where f is the frequency of the marker and a is the additive effect . To reduce the chance of false positives , a stringent p-value criterion has been adopted ( typically p = 5*10−8 , based on an adjusted p-value of 0 . 05 for one-million tests ) . It has been suggested that this selection criterion is too conservative [5] and that some of the missing heritability may be linked to genetic markers of small effect that fail this stringent cutoff . Alternatively , the narrow sense heritability explained by the common SNPs , hg2 , may be estimated by adapting a linear mixed-effects model [6] , [7] that is used to estimate h2 . This model decomposes the phenotypic variance into genetic and residual variance components . Usually , the model is applied to related individuals where the genetic relationships are estimated by using family pedigree or genetic markers [8] , [9] . Yang et al . [6] , [7] pointed out that hg2 could be estimated using genetic relationships obtained from the common SNPs for unrelated individuals . The main assumed difference between hg2 and h2 is due to the difference in linkage disequilibrium ( LD ) between the common SNP markers and the rest of the genome , with the assumption that closely related individuals would be in greater LD than unrelated individuals . Thus , heritability estimated with the genetic relationships of unrelated individuals is attributed to the common variants while that estimated with genetic relationships of related individuals is attributed to the entire genome . While the method does not identify single variants , it provides the maximum expected variance expected by the set of markers or the relative complement of the set ( e . g . , common versus rare variants ) . Recently , it has been shown that a large proportion of h2 is explained by the common single-nucleotide polymorphisms ( SNPs ) for several traits using this model [6] , [7] . Here , we showed that large proportions of the phenotypic variance for several metabolic syndrome ( MetS ) traits were also captured by the common SNPs . Among these , we validated the height and body-mass index estimates by Yang et al . [6] , [7] in independent GWAS populations . We also quantified the genetic correlation between traits explained by the common SNPs .
We estimated h2 and hg2 for height and body-mass index ( BMI ) in the Framingham Heart Study population ( FHS ) , and height and seven metabolic syndrome traits ( MetS ) traits: BMI , waist-to-hip ratio ( WHR ) , systolic blood pressure ( SBP ) , fasting glucose ( GLU ) , fasting insulin ( INS ) , fasting triglycerides ( TG ) , and fasting high-density lipoprotein ( HDL ) in the Atherosclerosis Risk in Communities population ( ARIC ) ( ARIC MetS estimates shown in Table 1 ) . Our base FHS population consisted of 4 , 240 subjects and our base ARIC population consisted of 8 , 451 subjects ( see Methods and Tables S1 and S2 for a description of the populations ) . The genetic relationship between pairs of subjects was estimated using 436 , 126 genome-wide common SNP markers for ARIC and 320 , 118 SNPs for FHS ( see Methods for details ) . We first estimated h2 for related individuals with relationships between 0 . 35 and 0 . 65 , derived empirically from the SNP markers , for height and BMI in the ARIC and FHS populations ( see Methods for derivation of the relationship matrix ) . This resulted in 3 , 663 subjects ( 6 , 706 , 953 pairs of subjects ) for FHS and 530 subjects ( 140 , 185 pairs of subjects ) for ARIC . We found h2 to be 0 . 77 ( s . e . 0 . 03 ) for height and 0 . 39 ( s . e . 0 . 04 ) for BMI in FHS , and 0 . 88 ( s . e . 0 . 09 ) for height and 0 . 34 ( s . e . 0 . 12 ) for BMI in ARIC . The estimated h2 were consistent with values obtained using phenotypic regression ( data not shown ) and previous results [6] , [7] , [10] , [11] . We then compared these values to estimates for hg2 for unrelated individuals with relationships less than 0 . 025 ( see Methods for derivation of the relationship matrix ) . This resulted in 1 , 489 subjects ( 1 , 107 , 816 pairs of subjects ) for FHS and 5 , 647 subjects ( 31 , 882 , 962 pairs of subjects ) for ARIC . As mentioned above , hg2 provides an estimate of the heritability explained by common variants because of presumed lesser linkage disequilibrium between the common SNPs and the rest of the genome as compared to related individuals . We found hg2 to be 0 . 50 ( s . e . 0 . 18 ) for height and 0 . 10 ( s . e . 0 . 18 ) for BMI in FHS , and 0 . 46 ( s . e . 0 . 05 ) for height and 0 . 14 ( s . e . 0 . 05 ) for BMI in ARIC . These values are consistent with previously estimated values [6] , [7] . Using the average across FHS and ARIC estimates , this implied that the common SNPs accounted for approximately 58% of h2 for height and 33% for BMI . To assess whether including more common SNPs would explain more of the h2 , we examined how hg2 depended on the number of SNPs . As shown in Figure S1 , the mean and standard error of the hg2 estimate for height in the ARIC population appeared to stabilize after approximately 300 , 000 SNPs . We then estimated h2 and hg2 for the MetS traits in the ARIC population using the same subjects as above ( see Table 1 ) . We validated our h2 estimates by using phenotypic regression between related individuals for some of the traits ( data not shown ) . The median h2 was 0 . 33 , the minimum was 0 . 23 ( INS ) , and the maximum was 0 . 48 ( HDL ) . The median hg2 was 0 . 13 , the minimum was 0 . 09 ( INS ) , and maximum was 0 . 24 ( SBP ) . Comparing the medians suggested that hg2 explains ∼39% of the h2 for these MetS traits . We found that the common SNPs explained large proportions of the h2: 41% of h2 for BMI , 46% for WHR , 30% for GLU , 39% for INS , 34% for TG , 25% for HDL , and 80% for SBP . We next estimated the genetic correlations between MetS traits using a bivariate ( multivariate ) model ( see Tables S3 and S4 for covariances ) . Table 2 shows the genetic and residual correlations for related individuals using bivariate models . The genetic correlation is the additive genetic covariance between traits normalized by the geometric mean of the individual trait genetic variances . The residual correlation is similarly estimated using the residual covariance and variances . For related individuals , we found significant genetic correlations for BMI-WHR , WHR-INS , GLU-INS , INS-TG , and TG-HDL and significant residual correlations between BMI-WHR , BMI-INS , BMI-HDL , WHR-INS , INS-HDL , and TG-HDL . Table 3 shows the genetic and residual correlations for the unrelated individuals . We found significant genetic correlations for BMI-WHR and TG-HDL and significant residual correlations for all of the estimates except SBP-HDL . The genetic correlations for unrelated individuals were proportional to the genetic correlations for related individuals ( see Figure S2 ) with a proportionality constant of 0 . 44 ( s . e . = 0 . 15 ; two-tail t-distribution p-value with 20 d . f . = 8 . 2*10−3 ) . The phenotypic correlations between traits were similar for related and unrelated individuals and are shown in Table 4 . These values were also consistent with the reported estimates in the National Heart Lung and Blood Institute-Family Heart Study ( NHLBI-FHS ) , which included Framingham Heart Study and ARIC families [11] . We validated our genetic correlation estimates using bivariate models for each pair of traits by analyzing all 7 MetS traits simultaneously for the unrelated individuals in a single multivariate model . This 7 trait multivariate model was much more expensive computationally so we used a less stringent convergence rule . The results were similar to the bivariate model ( see Table S5 and S6 ) although the genetic correlation increased and their error decreased for a number of the estimates . In addition to the significant genetic correlations in the bivariate models , we also found the genetic correlation for BMI-INS to be significant in the 7 trait model . We then examined the relationship between the genetic and phenotypic correlations ( see Figure S3 ) . For related individuals , we found that the phenotypic correlations rp were proportional to the genetic correlations rg with a proportionality constant of 1 . 2 ( s . e . = 0 . 16; two-tail t-distribution p-value with 20 d . f . = 3 . 1*10−7 ) . For unrelated individuals , we found that the phenotypic correlations were proportional to the genetic correlations with a proportionality constant of 0 . 85 ( s . e . = 0 . 19 ; two-tail t-distribution p-value with 20 d . f . = 2 . 3*10−4 ) . The direct proportionality between rp and rg implies that the ratio rg/rp is approximately constant for the MetS traits .
We used a recently developed approach to analyzing GWAS data and provided new estimates for the total amount of additive genetic information contained in the common SNPs for MetS traits . The approach uses a linear mixed-effects model to estimate the additive genetic variances and correlations between traits . The model relies on knowing the genetic relationships between the individuals analyzed . Previously , this had been obtained from family pedigrees . Visscher et al . [9] and Yang et al . [6] observed that the genetic relationships could be computed from the GWAS SNPs . They also presumed that the heritability estimated for unrelated individuals with low SNP correlation are explained mainly by these common SNPs because the linkage disequilibrium between the common SNPs and the rest of the genome is weak . This would be in contrast to related individuals with high SNP correlation where linkage disequilibrium is strong . Thus , heritability estimated with the genetic relationships of unrelated individuals is attributed to the common SNPs while that estimated with the related individuals is attributed to the entire genome . This then creates a major distinction between h2 and hg2 . We computed both in the same population . However , differences between estimates of h2 and hg2 may also arise due to differences in environmental influences and non-additive genetic effects that may bias the estimates . Provided that these biases are small then the ratio of hg2 to h2 provides an estimate of the proportion of narrow sense heritability captured by the common SNPs . We confirmed previous findings that a large proportion of h2 is explained by the common SNPs . Our hg2 estimates for height and BMI in two independent analyses ( i . e . ARIC and FHS ) were consistent with previously reported values [6] , [7] . Our h2 estimates for BMI , GLU , INS , TG , HDL , and SBP were similar to the findings of the large family National Heart , Lung , and Blood Institute ( NHLBI ) Family Heart Study [11] , which included Framingham Heart Study and ARIC families . We found that hg2 explained a large proportion of h2 across the MetS traits , and hg2 explained approximately 39% of the h2 for these traits . We estimated that the common SNPs explain 58% of h2 for height , 41% for BMI , 46% for WHR , 30% for GLU , 39% for INS , 34% for TG , 25% for HDL , and 80% for SBP . Our hg2 findings are striking compared to traditional GWAS approaches where significant common SNPs have been shown to explain only 4% of h2 for BMI with 32 SNPs , 11% for GLU with 14 SNPs , 20% for TG with 48 SNPs , 25% for HDL with 60 SNPs , 3% for SBP with 10 SNPs , and 12% for height with 180 SNPs [12]–[16] . Height had the largest absolute hg2 , which was consistent with having a large h2 . Surprisingly , SBP had the largest proportion of h2 explained by the common SNPs while only a few percent of this has been uncovered by traditional GWAS . However , the standard error of hg2 for SBP was large and reducing this error will be important for further investigation . Conversely , our analysis suggested that the SNP markers already identified for TG and HDL may contain the maximum heritability expected from the common SNPs . Our analysis of hg2 against the number of SNPs suggested that the mean and standard error of hg2 for height is well estimated by approximately 300 , 000 markers and that including more markers would have little effect for this trait and perhaps others . The standard error of hg2 also increased with SNP number . This may seem paradoxical but can be explained by recalling that the estimate for hg2 is proportional to the regression coefficient of the square of the phenotype differences versus the genetic relationship ( i . e . Haseman-Elston regression ) [8] . The standard error of hg2 is thus inversely proportional to the variance of the genetic relationship . Since the latter is estimated from the common SNPs , this variance is expected to decrease as the number of SNPs increases thereby increasing the standard error [6] . Using the bivariate ( multivariate ) model [17] , [18] we estimated the genetic and residual correlations between the MetS traits . Among these , we found that the genetic correlations in related and unrelated individuals for BMI and WHR were significantly different from zero . This is consistent with both traits as indirect measures of body fat and common health risks [19] . Previously , Rice et al . , 1994 [20] found significant genetic correlations between BMI and SBP among normotensive nonobese families . This suggested a common genetic etiology to their physiological relationship through hyperinsulinemia resulting in increased renal reabsorption of sodium and sympathetic activation [20] . We found a large genetic correlation among related subjects , although it was not significant because of the large error . This was consistent with the large family study by the NHLBI that did not find a significant genetic correlation [8] . Perusse et al , 1997 [21] argued that cross-trait resemblance between BMI and lipids is mostly environmental . In concordance , we did not find significant genetic correlations between either BMI or WHR and TG and HDL for either related or unrelated individuals ( see Table 3 and Table 4 ) while residual ( which includes environmental ) correlations were significant for BMI–HDL . We found that the residual covariance accounted for a minimum of 71% ( derived from the estimates in Table 4 and Table S3 ) of the phenotype covariance between BMI or WHR and the lipid measurements for related individuals . Genetic correlations between TG and HDL were also large , which is consistent with their direct physiological relationship [22] . This is also consistent with the findings from a recent GWAS meta-analysis whose results showed that 50% of the significant markers for TG were also significant for HDL ( derived from Supplementary Tables 6 and 11 in [16] ) , and with a genome-wide LOD correlation analysis [23] . While we found some significant genetic correlations among both related and unrelated subjects , the variance was large for these estimates and greater statistical power is needed for better accuracy . We found that the genetic correlation was directly proportional to the phenotypic correlation , which was an unexpected , empirical finding . Previously , a linear relationship between the correlations was hypothesized by Cheverud for sets of traits with common functions , and shown empirically for a number of traits [8] , [24]–[26] . While this finding is interesting from an evolutionary genetics perspective , it may also serve a useful purpose in the maximum likelihood computation of the linear mixed-effects model by providing initial genetic correlation ( i . e . covariance ) estimates based on the phenotypic correlations . In summary , we provided evidence that the common SNPs explain large proportions of the variance for several MetS traits in agreement with previous findings for some of these traits [6] , [7] . This is consistent with the original premise of GWAS that a large proportion of phenotypic variation for common traits may be due to common variants [27] . However , an amendment to this premise is that it is likely to be many common variants with small effect . This is supported by recent meta-analyses with larger sample sizes that have identified more associated common SNPs . This approach can serve as a first approximation of the total heritability expected from common SNPs given a genome-wide set of markers and requires fewer subjects to achieve significant results . We also found genetic associations that will be useful for single gene and systems biology studies . Future studies with greater power will provide estimates for weaker multivariate genetic associations and provide greater precision for the estimates presented here .
Our main study population was the Atherosclerosis Risk In Communities ( ARIC ) population . The ARIC population consists of a large sample of unrelated individuals and some families across North America . The population was recruited from four centers across the United States: Forsyth County , North Carolina; Jackson , Mississippi; Minneapolis , Minnesota; and Washington County , Maryland . For this study , we restricted our analysis to the European-American group . The population was recruited in 1987 from the general population consisting of subjects aged 45 to 64 years . The ARIC population consisted of 8 , 451 subjects . Quality control and genotype calls for common SNPs were evaluated previously for ARIC using the Affymetrix Human SNP Array 6 . 0 . We selected bilallelic autosomal markers based on the following criteria: missingness <0 . 05 , Hardy-Weinberg equilibrium ( p<10−6 ) and minor allele frequency >0 . 05 . Subjects with missingness >0 . 05 were removed . This resulted in 436 , 126 retained markers . Quality control measurements from dbGAP ( GENEVA ARIC Project Quality Control Report Sept 22 , 2009 ) indicate significant population stratification between self-identified white ( European-ancestory kind group ) and black populations when projected onto HapMap components . Furthermore , principal-components analysis of the European-ancestory group by dbGAP showed that no component explained more than 0 . 1% of the population variance . For this study we only analyzed the European-ancestory group and treated it as a single population . ARIC phenotypes were adjusted for age , sex , and study center . Only single measurements from visit 1 were used for these subjects . We only used subjects with negative diabetes status and with genotype and phenotype information for all traits . This resulted in 8 , 451 subjects . We standardized all the traits . We first log-transformed BMI , glucose , insulin , triglycerides , HDL , and systolic blood pressure . All laboratory measurements are under fasting conditions . Population trait statistics are in Table S1 . We estimated h2 and hg2 for height and BMI in the Framingham Heart Study population ( FHS ) . The FHS population is a large multi-generational dataset that started in 1948 in Framingham , Massachusetts in the United States . It consists of a number of ethnicities predominantly from the United Kingdom , Ireland , Italy , and Western Europe [28] . Markers were screened similarly to ARIC and we also removed any SNPs that did not overlap with the ARIC set , which results in 320 , 118 SNPs . We used principal components analysis of the linkage disequilibrium ( LD ) pruned genetic relationship matrix to identify components with variance >0 . 1% . LD pruning was as in the ARIC 2009 report . This resulted in 73 , 432 retained SNPs . We found three significant components that were then used as covariates in the REML model . For consistency with ARIC , we restricted the age range at time of exam to 45 to 65 years and randomly selected a single measurement in the case of multiple measurements . Phenotypes were adjusted for age , sex , and generation prior to the REML estimation and standardized . We first log-transformed BMI . Population trait statistics are in Table S2 . Our base FHS population consisted of 4 , 240 subjects . We determined h2 using the linear mixed-effects model ( see derivation below ) and related individuals defined as genomic relatedness between 0 . 35 and 0 . 65 . We assume that the common SNPs are in greater linkage disequilbrium among related individuals and , as such , can be used to estimate the total additive-genetic variance across the allele spectrum as suggested by Visscher et al . , 2006 [9] . We constrained the relationship matrix to have at least one related pair per subject . This was done by pruning the entire population relationship matrix by randomly selecting a row and removing the row and its corresponding column if no genomic covariance in the row was between the cutoff values . For all pairs , including unrelated individuals , we used their empirically defined relationship . This resulted in 530 individuals being selected for analysis in ARIC and 3 , 663 individuals in FHS . h2 was estimated with h2 = varg/ ( varg+vare ) , where varg and vare are the genetic and residual variance components estimated by the REML model using related individuals . The error was estimated from the inverse Fisher Information ( see linear mixed-effects model below ) and propagated using a first-order Taylor expansion . We used the linear mixed-effects model and only unrelated individuals to estimate the additive-genetic variance attributable to the common SNPs ( hg2 ) . Unrelated individuals were defined as subjects with maximum genomic correlation of <0 . 025 . The genomic relationship matrix was then produced as above based on this cutoff . The cutoff was taken from Yang et al . 2010 [6] and is less than the expected coefficient of relatedness between 2nd cousins . For these estimates we used the same group of 5 , 647 unrelated individuals for all estimates in ARIC and 1 , 489 individuals in FHS . hg2 was estimated as hg2 = varg/ ( varg+vare ) , where varg and vare are the genetic and residual variance components estimated by the REML model using unrelated individuals . The standard error was estimated as above . The height hg2 versus SNP number analyses were performed over allele frequency range of 0 . 05 to 0 . 5 in order of increasing and decreasing frequency . The genetic correlation ( rg ) is defined as , where ( varg ( ti ) ) is the additive genetic variance of trait i and covariance ( covg ( ti , tj ) ) is the additive genetic covariance between the traits . The variances and covariances are estimated directly in the multivariate linear mixed-effects model . The error was computed from the estimated errors of the variances and covariance using a first-order Taylor expansion . The residual and phenotypic correlations were analogously defined . Phenotype correlations and error were estimated by linear regression of the standardized phenotypes . The mean and errors for proportionality constants between the genetic and phenotypic correlations were determined by randomly sampling over the distributions of the parameter estimates ( i . e . Monte Carlo method ) assuming that the error around the mean parameter estimate was normally distributed and that the parameters were independent . We then fit a linear function with the y-intercept fixed at 0 ( after first confirming that it was not significantly different from zero ) . We assessed significance for correlation coefficients ( r ) using the standardized Fisher transformed estimate of r: arctan ( r ) /arctan ( s . e . ( r ) ) . We estimated the two-tailed p-value from a normal distribution and significance was determined by p<0 . 05 and Bonferroni corrected for 21 hypotheses . Significance for regression coefficient ( ) was estimated using the standardized coefficient . We estimated the two-tailed p-value from a t-distribution and 20 degrees of freedom and significance was determined by p<0 . 05 . Preprocessing of SNPs and phenotypes was done using PLINK [29] ( v1 . 07 , http://pngu . mgh . harvard . edu/purcell/plink/ ) and MATLAB ( 2010b , MathWorks , Natick , MA ) . REML optimization was executed using software written in MATLAB . We considered the following multivariate linear mixed-effects model for m individuals , n loci and t traits [6]–[8] , [17] , [18] , [30]:where yi is a m×1 vector of trait i for m individuals , Xi is an m×s fixed effects matrix for trait i , vi is a s×1 vector of fixed effects parameters for trait i , Z is an m×n matrix of standardized genotypes , ui is an n×1 vector of random effects for trait i satisfying ui∼N ( 0 , G ) and ei is an m×1 vector of residual effects satisfying ei∼N ( 0 , R ) , with matrix blocks Gij = covgijIn and Rij = coveijIm and Il is the l×l identity matrix . This model can be used for single or multiple traits . For two traits , it is called a bivariate model . The model is identical to that used by [6] , [7] , [17] . We considered only bi-allelic SNPs in Hardy-Weinberg equilibrium . Denote the minor allele by q and the major allele by Q . Let the minor allele frequency at locus i have frequency pi . We assign a value of 2 for genotype qq , 1 for genotype qQ and 0 for genotype QQ . The Hardy-Weinberg mean frequency for the genotype at locus i is 2pi and the variance is 2pi ( 1−pi ) . The standardized genotype entries have values of ( 2−2pi ) / ( 2pi ( 1−2pi ) ) 1/2 for qq , ( 1−2pi ) / ( 2pi ( 1−2pi ) ) 1/2 for qQ , and −2pi/ ( 2pi ( 1−2pi ) ) 1/2 for the QQ genotype . The log of the likelihood function is given bywhere the covariance matrix can be expressed as a tensor product with m×m blocks V−1ij and A is the genetic relationship matrix . Following Yang et al . [6] , we used a modified covariance matrix for A , , where the diagonals of A are computed using the formulaWe use the restricted maximum likelihood ( REML ) approach [8] where the gradients of the log likelihood are given bywhere Iij is a tm×tm dimensional matrix with zero entries except for a m×m identity matrix at block location i , j , and , where . We solved the REML equations using an EM algorithm [8] , which was given byfor iteration k+1 in terms of iteration k . We iterated until the rate of change of the log likelihood function was less than about 10−4 . We also checked that the rate of change of the square of the covariance predictions was less than 10−8 . We checked our results against the software developed by Yang et al . ( GCTA ) [31] for the univariate model . For the multivariate model , we transformed to a coordinate system where the covariance matrices were diagonal [8] to speed up the computation . Let zj be the set of phenotypes for individual j . We used the canonical transformation such that and . Q can be computed from the formula where , ( S is the matrix of left eigenvectors of GR−1 ) . The transformed genetic covariances are given by and the residual covariances are It . Each step consisted of taking a single step with the univariate EM algorithm for the transformed additive genetic and residual variance followed by a transformation back to the original coordinates . We iterated until the maximum of the magnitudes of the components of the gradient of the log likelihood function was less than approximately . In our computations , we used both the direct EM algorithm and the canonically transformed algorithm because even though the transformed algorithm was in principle faster , it sometimes had poor convergence properties if the initial guess was not sufficiently close to the maximum likelihood value . We ensured that both give the same results . For computational efficiency , the results shown are computed from the bivariate model for the different trait pairs . We confirmed our results with a multivariate model that included all traits . Our error estimates were given by the inverse of the Fisher information matrix F , which we computed by evaluating the Hessian of the log likelihood at the maximum likelihood predictions . F is a t ( t+1 ) ×t ( t+1 ) dimensional matrix with rows corresponding to the genetic and residual variances and covariances ( where covij was set equal to covji ) and with block elements ( that are not all contiguous ) given byfor and . | The narrow-sense heritability of a trait such as body-mass index is a measure of the variability of the trait between people that is accounted for by their additive genetic differences . Knowledge of these genetic differences provides insight into biological mechanisms and hence treatments for diseases . Genome-wide association studies ( GWAS ) survey a large set of genetic markers common to the population . They have identified several single markers that are associated with traits and diseases . However , these markers do not seem to account for all of the known narrow-sense heritability . Here we used a recently developed model to quantify the genetic information contained in GWAS for single traits and shared between traits . We specifically investigated metabolic syndrome traits that are associated with type 2 diabetes and heart disease , and we found that for the majority of these traits much of the previously unaccounted for heritability is contained within common markers surveyed in GWAS . We also computed the genetic correlation between traits , which is a measure of the genetic components shared by traits . We found that the genetic correlation between these traits could be predicted from their phenotypic correlation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"heredity",
"genetics",
"biology",
"human",
"genetics",
"genetics",
"and",
"genomics"
]
| 2012 | Heritability and Genetic Correlations Explained by Common SNPs for Metabolic Syndrome Traits |
The within-host evolution of influenza is a vital component of its epidemiology . A question of particular interest is the role that selection plays in shaping the viral population over the course of a single infection . We here describe a method to measure selection acting upon the influenza virus within an individual host , based upon time-resolved genome sequence data from an infection . Analysing sequence data from a transmission study conducted in pigs , describing part of the haemagglutinin gene ( HA1 ) of an influenza virus , we find signatures of non-neutrality in six of a total of sixteen infections . We find evidence for both positive and negative selection acting upon specific alleles , while in three cases , the data suggest the presence of time-dependent selection . In one infection we observe what is potentially a specific immune response against the virus; a non-synonymous mutation in an epitope region of the virus is found to be under initially positive , then strongly negative selection . Crucially , given the lack of homologous recombination in influenza , our method accounts for linkage disequilibrium between nucleotides at different positions in the haemagglutinin gene , allowing for the analysis of populations in which multiple mutations are present at any given time . Our approach offers a new insight into the dynamics of influenza infection , providing a detailed characterisation of the forces that underlie viral evolution .
The overall risk to human health posed by the novel H7N9 influenza virus [1] , while potentially severe , is as yet unknown [2] , [3] . Pandemic influenza is a zoonosis [4] , and as such any new pandemic may be expected to arise through a two-step process [5] , [6] , the virus first gaining the ability to cause sporadic , localised infections in humans until , after a second transition , emerging into a global pandemic . Each of these steps are evolutionary in nature , being characterised in turn by the adaptation of a virus to be able to infect a human host , and the development of increased transmissibility between hosts . In the nH7N9 strain , the first of these steps has already taken place , including the acquisition of mutations responsible for human-specific receptor binding [7] . Progression to a global epidemic , therefore , depends upon the evolution of increased transmissibility of the virus , a phenotypic change which can only occur while the virus grows in a host environment . As is true for other viral species [8] , understanding the intra-host evolution of influenza is an important task . A vast array of mathematical modelling approaches have been directed at the questions of influenza infection , transmission , and evolution [9] . Of particular relevance to this study are models which track the dynamics of a single infection . Based upon observed changes in viral titre over time , inferences have been made of many important properties of infection , including the reproductive number for cellular infection , the timescale and numbers of viruses produced during the infection of a cell , and the impact upon the viral population of both innate and adaptive immune responses [10]–[14] . Considering data of intracellular RNA levels , the fine detail of viral replication within a cell has been described [15] . Evolutionary models of competition between viral strains have clarified the relationship between selection for growth and transmission effects , and the dynamics of immune escape [16]–[18] . In the cases above , the viral population was either modelled as a population of identical individuals , or as a set of distinct classes of virus , characterised by differing immune escape or transmission properties . Building upon these approaches , a genetic classification of viruses was used to model H5N1 influenza evolution [19]; the fitness of a virus was defined according to the presence or absence of a set of mutations . Here we divide the viral population in a similar manner , expressing the fitness of a virus as a function of its genetic composition . However , rather than analysing the consequences of a proposed fitness landscape , we here infer how selection was actually at work based upon observed genetic sequence data . In chronic infections such as HIV , time-resolved sequence data from individual hosts is readily available [20] . However , the course of an influenza infection , even in an immunocompromised host [21] , is relatively short . As such , time-resolved genetic data is rare , the main examples having been collected from experimentally-infected animal populations [22] , [23] . In this work , we consider data from one such study , examining the evolution of H1N1 influenza within individuals in a swine population [24] , [25] . The basic principle of our method is to learn the role of selection acting upon a viral population by means of a maximum likelihood method . We adopt a coarse-grained quasispecies model ( cf . [26] ) to describe the evolution of the viral population , in which viruses are classified according to the nucleotides ( here denoted alleles ) present at a limited number of positions ( or loci ) in their genomic sequence . In this model , evolution proceeds deterministically , contingent only upon the initial state of the population , and the role of selection for or against specific alleles . By considering the consequences for the population dynamics of different proposed models of selection , and comparing these to the observed evolution of the system , we estimate how selection was at work . The low rate of recombination within RNA segments of influenza [27] , [28] , combined with a high viral mutation rate , leads to complex evolutionary dynamics , with the fate of mutations being strongly affected by genetic hitchhiking and clonal interference [29]–[31] . As such , discerning the effects of selection requires that interactions between alleles at different loci are taken into account [32] . Here this is achieved by considering the frequencies of haplotypes , sets of sequences with specific alleles at specific loci ( e . g . allele C at locus i and allele T at locus j ) . In our model , the viral population can be described at potentially any genomic resolution , keeping track of the population in terms of haplotyes spanning arbitrary numbers of loci . However , higher-locus models are more computationally demanding . As such , we first apply a filtering process to cut out loci at which alleles do not show statistical evidence of having evolved under selection . For each polymorphic locus , we use a single-locus model of evolution to find alleles that appear to evolve in a non-neutral behaviour , changing in frequency over time . Change in the frequency of an allele may occur as the direct result of selection , or due to linkage disequilibrium with a selected allele , or alleles , at other loci . As such , to distinguish between these cases , wherever apparent non-neutrality is observed at more than one locus , we apply a multi-locus model of haplotype frequency change to the data . This model explicitly accounts for interactions between alleles at different loci , and is used to identify the maximum likelihood explanation for the changes observed in the sequence data . As has been noted elsewhere , the use of viral sequence data to understand population structures requires substantial care ( e . g . [33] , [34] ) . Selective amplification of sequences , or general sequencing bias , can produce a misleading picture of a population as a whole . PCR-induced recombination can lead to false measurements of linkage disequilibrium between alleles at different loci . We discuss the potential impact of each of these factors upon our results .
Viral sequence data collected from a previous transmission experiment [25] were analysed . An overview of the structure of this experiment is shown in Figure 1 . The chain of infection was propagated by a process of housing pairs of uninfected pigs with pairs of infected pigs , the previously-infected pigs being removed after transmission had occurred . Throughout the experiment , samples were collected from pigs using nasal swabs , with viral sequences being amplified via RT-PCR and Sanger sequenced . Viral sequences were collected from the majority of the pigs; for 16 of the 24 pigs involved in the experiment , data was collected at more than one time-point , an essential prerequisite for our method . For the samples collected in these animals the depth of sequencing varied from 6 to 81 sequences ( mean 51 ) from a pig at a given time-point , with data being collected at up to five time-points across the course of an infection . Limited transmission of variants was observed between individual infections . In our analysis , non-neutral behaviour was identified in six populations . In general , signs of selection were relatively rare . While very many individual mutations were observed in the population as a whole , most of the substantial changes in allele frequency occurred at a small number of sites ( e . g . Figure 2 ) . As such , eighteen alleles in the dataset were identified as being potentially non-neutral . Interference effects between alleles were found to be of importance; of these eighteen alleles , a total of nine were identified as being genuinely under selection , changes in frequency at the other nine being explicable in terms of linkage disequilibrium with other selected alleles . In the populations identified to be non-neutral , a variety of forms of selection were found , including evidence for time-dependent selection , and for selection acting simultaneously at more than one locus ( a selection of inferred trajectories are shown in Figure 3; further inferences are presented in Supporting Figure S1 ) . Our multi-locus model discriminated between cases where multiple alleles changed in frequency under independent selection , and cases where selection acting upon one allele led to substantial changes in the frequency of others ( Table 1 ) . In Pig104 , strong evidence [35] was found for negative selection acting against the G → A mutation in locus 114 , with an inferred selection coefficient of −1 . 6 per 12 hours ( h ) . Such a magnitude of selection is relatively large; by comparison , an allele at frequency 50% with a selection coefficient of −1 per 12 h would decrease to 12% frequency after one day and to less than 2% after 2 days . The mutation under selection in this case is synonymous , such that the observation of strongly deleterious selection is perhaps a surprising one . While , using our method , no statistical evidence for selection upon this allele was identified in other pigs , the same polymorphism was found in data collected at the earliest time point for pigs 115 and 116 , but not at subsequent time-points , consistent with a hypothesis of negative selection for this nucleotide across all viral populations . In Pig109 strong evidence was found for positive selection upon at least two of three alleles; in favour of the G → A polymorphism at locus 553 , the A → G polymorphism at locus 696 , or the G → A polymorphism at locus 914 . Fixation of all three of these mutations occurred between two samples , and models with any single one of these mutations as the selected allele performed similarly well , giving estimated selection coefficients between 3 . 0 and 3 . 1 per 12 h for the selected allele . Joint consideration of four-locus haplotype frequencies provided evidence that at least two of these mutations were independently under selection . The most likely model had coefficients of 2 . 8 per 12 h at each of the loci 696 and 914 . However , the difference between two-locus additive models was small , and models in which any two of the three polymorphisms were under selection performed similarly well ( Supporting Table S1 ) . An interesting feature of this result is that the pairs of mutant alleles inferred to be under selection are highly linked , the mutant alleles at loci 696 and 914 appearing only jointly on a sequence , and never in isolation . The inference that selection is acting at two loci , rather than at only one locus , arises from the effect of mutation in the model; this result is explored more fully in Supporting Information . We note that , while the polymorphism at locus 696 is synonymous , those at 553 and 914 are non-synonymous in character , corresponding to the mutations D185N and S305N ( the former being contained within the Ca2 epitope region [36] ) . In Pig115 weak evidence was found for positive selection in favour of the G → A polymorphism at the locus 188 , with an inferred selection coefficient of 1 . 2 per 12 h . This polymorphism is non-synonymous , representing the amino acid substitution G63E . Bootstrapping of this result against inferences from sequence data that had been randomised in time largely supported this inference; from a total of 200 sets of randomised sequence data , a stronger signal in favour of a model of constant selection was identified in only eight cases . Details of the bootstrapping of all results are given in Supporting Text S1 and in Supporting Figure S2 . In Pig405 , strong evidence was found for positive selection acting upon the G →A polymorphism at locus 844 , with a selection coefficient of 0 . 4 per 12 h , along with simultaneous , time-dependent selection acting upon the A → G polymorphism at locus 553 . Selection at this second locus was inferred to be initially positive , with mean strength 0 . 9 per 12 h during the first time-interval , weakly negative during the second time interval , with mean strength −0 . 1 per 12 h , then finally strongly negative , of mean magnitude greater than −2 per 12 h for the final time interval . Each of these polymorphisms are non-synonymous ( corresponding to the mutations V282I and N185D respectively; the mutation at locus 553 is identical to that observed in Pig109 , albeit in the reverse direction ) . Identification of time-dependent selection acting upon the latter , epitope mutation is of particular interest , raising the possibility that this corresponds to an adaptive immune response by the host to the virus . In this population the magnitude of the time-dependent selection inferred for the final time-point was large and negative , but hard to identify with precision . This arises from a time-dependent model of selection being coupled with an observed allele frequency of zero at the final time-point . Excluding the influence of allele frequencies at other loci , the data in such a case can lead to an inference of arbitrarily strong negative selection; the time resolution at which data are collected imposes a limit on the magnitude of selection that can correctly be inferred [37] . In Pig410 we identified weak evidence for time-dependent selection acting upon the synonymous C → T mutation at locus 447; in this case , a bootstrapping calculation produced a stronger signal of selection than that for the real data in only three out of 200 cases ( Supporting Figure S2 ) . Time-dependent selection was also identified in Pig412 , where strong evidence was found for time-dependent selection acting upon the synonymous G → A mutation at locus 696 , with further weak evidence for negative selection acting upon the synonymous A → G mutation at locus 48 . Under the multi-locus model , a selection coefficient of 1 . 8 was identified at locus 696 for the first time interval . The inferred strength of selection at this locus for the second , final time interval was imprecise , but very large and negative; the value of −22 . 8 per 12 h reported in Table 1 again being caused by an observed frequency of zero at the final time-point . Alleles at which selection was inferred were distributed across the HA protein ( Supporting Figure S3 ) . Significant changes in allele frequency were identified in more than one infection at five different loci ( 447 , 553 , 696 , 824 and 844 ) . Of these , selection was inferred to act at the loci 696 and 844 in more than one infection . This repetition of mutations may be explained by the design of the experiment; selection is most likely to be observed when polymorphisms exist at non-negligible frequency in the population , while polymorphisms at higher frequencies are more likely to be transmitted between infections . Under an initial scan for potentially non-neutral alleles , very weak evidence for selection was identified in the data from Pig113 at the three loci 447 , 824 and 844 . However , under the full multi-locus model , a neutral model of evolution was finally preferred . As we discuss further in Supporting Text S1 , our evolutionary model is more conservative in identifying selection in cases where multiple loci are considered simultaneously .
Our evolutionary model assumes that the viral population is genetically well-mixed in the host , and that it evolves in a deterministic manner , both with respect to mutation , and to selection . The first of these assumptions asserts that each sample of viruses collected from the pig is representative of the viral population in the animal at the time . This would not be true if , for example , the viral population was split into diverse subsets , with selection acting in very different ways in each . Study of these effects was not possible given the data studied here . Our assumption of deterministic evolution is based on the underlying viral population being large in number , that is , large enough that and are significantly greater than 1 , where N is the number of viruses in an animal , μ is the mutation rate per locus , and σ is the magnitude of selection [48] . Considering selection , the lowest resolution at which we report selection , of 0 . 1 per 12 h , is , accounting for two rounds of replication in the lifetime of an infected cell [10] , [15] , equivalent to a fitness difference of 0 . 05 per generation . As such , this part of the assumption holds if N is substantially larger than 20 viruses . Considering mutation , the criterion that is stricter than that for selection ( where μ is of order 10−5 [49] , [50] ) , requiring N to be substantially larger than 105 . In influenza , models of replication in a single cell suggest that of the order of 104 virions are produced within each cell [51] , while in the samples from which viruses were sequenced , a viral load of between 30 and 5500 particles per µl [24] was measured; once an infection has progressed to the point where viral sequencing is possible , the population is very likely large enough for this to be fulfilled . In the earliest stages of an infection , stochastic mutational behaviour could potentially lead to an incorrect inference of the initial variant frequencies within the population; however , these values are not used to draw any biological conclusions about the system . Horizontal transmission between co-housed animals was not incorporated into the model; we believe this was unlikely to have greatly influenced the collected data . If the viral populations in the two simultaneously infected pigs were substantially different in composition , transmission of viruses from one animal to the other might alter the composition of the viral population in the second animal . However , the viral populations in this experiment were not sufficiently different in sequence to be able to distinguish superinfection from the growth of de novo mutations . Further , while the viral titre implicated in transmission is unknown , we believe that the incoming titre is likely to be substantially smaller than the pre-existing number of viruses in the second infected animal . A second assumption in our study is that the collected sequence data are relatively accurate . That is , we assert that the sequences obtained from the sample are representative of the sample itself . The basis of our inference upon data means that the accuracy of the data is vital for obtaining useful results . For example , in addition to raw allele counts , our approach makes explicit use of linkages between mutations . Our method allows for the possibility of generic error in the sequencing process , and fully accounts for the statistical noise inherent to a finite data sample . However , there are systematic data biases that may also affect the results obtained . For example , PCR-induced recombination has the potential to alter the observed frequencies of multi-locus haplotypes [52] , [53] . Testing for such an effect , by fitting an exponential model to the observed absolute linkage disequilibrium between pairs of alleles , we found no evidence for such recombination , no decay in this statistic being observed with increasing distance between alleles ( Supporting Figure S4 ) . Sequencing bias also has an effect on whether or not a mutation is recognised as being under selection . Mutations that are preferentially identified by a sequencing method would appear in the sample at higher frequencies , such that changes in their frequencies were amplified , leading to a greater chance that such mutations were found to be under selection . For this dataset , a consistent sequencing method was used to process all of the samples; we therefore assumed sequencing bias to be consistent between samples , such that observed changes in allele frequency were caused either by the finite sampling process , or by a process of mutation and selection . Estimating the extent of sequencing bias in the observed sequences is difficult , the sequences themselves representing the only information about the real viral population . Counting the mutations observed in the data showed a high transition:transversion ratio of 9 . 7 ( Supporting Figure S5 ) . This is broadly consistent with values observed for other RNA viral populations [54] , [55] , albeit that measurements of this ratio in influenza have previously been based upon global , rather than within-host , populations [56] . Biased sampling , whether occurring via the collection of a biological sample that is unrepresentative of the whole population , or as a result of the subsequent PCR amplification , also has the potential to affect our inference . We have here assumed that the data is an unbiased sample of the real population . Our inferences are partially limited by the use of sequences describing only the HA1 region of the influenza virus . While our inferences of deviation from neutrality in a population are not affected by alleles elsewhere in the virus , the attribution of selection to given alleles may be affected by unobserved polymorphisms in the HA2 region of influenza , or if reassortment were limited ( though see [57] ) , with alleles in other viral segments . The potential influence of selection acting upon polymorphisms that have not been observed is of greatest relevance to the cases of apparently time-dependent selection; constant selection acting upon interfering mutations causes time-dependent selection effects [32] . One example is the case of Pig412 where initially positive , then negative selection is inferred . In this infection , many haplotypes which are observed at the intermediate time point are no longer seen in the final time point; this pattern is consistent either with a switch in the direction of selection acting upon the synonymous mutation at locus 696 , as was inferred , or with very strong positive selection acting upon an unobserved mutation on the consensus haplotype causing a selective sweep later in the observation . Such a scenario is much less likely in the case of Pig405 , where the haplotype containing the allele inferred to be under negative selection is outcompeted in the final time interval by four other haplotypes , including that of the initial consensus . We have here described a framework for the inference of selection acting upon a viral population within an individual host , based upon time-resolved sequence data . Within-host selection is of importance for the future evolution of the H7N9 influenza virus , and for understanding the epidemiology of other influenza strains . During an epidemic , both within-host growth , and the transmission of viruses , are important , and potentially competing factors; a mutation which is beneficial for within-host growth may prove deleterious for transmission and vice versa . While we have here considered only the first of these factors , our method could easily be used to infer the role of selection for transmission , given specific conditions . First of all , substantial continuity would be required between the native and the transmitted populations , such that changes in allele frequencies before and after transmission were primarily the result of selection; severe bottlenecking would distort the population structure . Secondly , clarity would be required about the source of each infection; in the experiments considered , where an infection begins with an unknown mixture of viruses from two other individuals , the role of selection in transmission cannot be evaluated . Transmission events in the data analysed here have been discussed elsewhere [58] . In more straightforward cases , where transmission occurs between known individuals , and where continuity between viral populations is more evident ( e . g . [59] ) , use of our method to infer selection acting across transmission events is likely to be achievable . The collection of sequence data describing the within-host evolution of influenza is at present , relatively rare , although we anticipate that improvements in sequencing technology will make such data increasingly accessible . Increased collection of sequence data from patients , and from evolutionary experiments , will greatly add to our understanding of viral infection . Our approach increases the value of such work , characterising in detail the forces that underlie within-host viral evolution .
Quasispecies theory [26] provides a deterministic description of the evolution of mutation-prone , self-replicating organisms; this framework has profoundly influenced studies of RNA viral evolution [60]–[63] . To describe the evolutionary dynamics of the influenza virus within an individual host we apply a coarse-grained quasispecies model , in which the viral population is described as haplotypes spanning a limited set of loci , rather than as complete viral sequences . Specifically , we represent the viral population as a frequency vector , defined at discrete times , and comprised of elements , where is the fraction of sequences in the population with the haplotype ; that is , with the nucleotides at a subset of loci in the viral genome . To model mutation between haplotypes , we assumed a constant rate of mutation , μ , between any two specific nucleotides at a given locus , the probability of mutation from haplotype to haplotype in a single generation being given by ( 1 ) where is the Hamming distance between the two haplotype sequences . Selection was accounted for by ascribing to each haplotype the ( potentially time-dependent ) selection coefficient . The effect of selection on the haplotype frequency between times and was thus defined by the function : ( 2 ) where . Considering the evolution of influenza , we supposed time-points to be spaced at 12-hour intervals , roughly approximating the time required for a round of intracellular growth within a cell [10] . Within such a round of growth , each virus undergoes two rounds of replication , modelled as having equal mutation rates , with the parameter representing an overall rate of mutation per nucleotide per generation of 10−5 [49] , [50] . Selection was assumed to act upon the viral population once it has exited the cell , giving the relation ( 3 ) where is the matrix consisting of elements , modelling a single round of replication . The behaviour of the system is thus specified in a deterministic manner by the selection parameters , and by the initial state of the system , given by the elements of the vector . We note that , while sequence data was collected at known times throughout the course of each infection , the precise moment at which each infection began is unknown . Here , we assumed to be precisely 24 hours before the first observed set of sequence data from the infection . While the uncertainty in this value has consequences for the accuracy of the elements of the inferred vector , no conclusions were finally drawn from these values . An inference of selection was carried out by comparing maximum likelihood values obtained under a hierarchical series of models , each specifying the parameters and . The coarse-grained quasispecies model can be expressed in terms of haplotypes of arbitrary length . We describe the general model below . In order to test for the influence of PCR-induced recombination upon the dataset , we calculated a measure of linkage disequilibrium between loci . For each pair of polymorphic loci in the dataset , we calculated the value , equal to the absolute linkage disequilibrium between these loci , normalised by the maximum potential linkage disequilibrium given the allele frequencies in question ( 8 ) where the labels 0 and 1 represent the consensus and most common minor alleles at each locus , represents the frequency at time of the allele at locus , and represents the frequency at time of the haplotype at loci and . Values of were compared for loci at different positions in the sequence , fitting a model of the form , for all points for which , where is the sequence distance between loci and . Here a greater negative value of would indicate that a higher mean rate of recombination in the viral sequences occurred during the sequencing process . A test of the ability of the method to discriminate between selected and non-selected alleles , and to correctly infer the magnitude of selection acting upon a locus , was performed by running analyses for simulated data . For simulated populations with a single allele under selection , a correlation coefficient of of more than 0 . 95 was found between real and inferred selection coefficients , with an equivalent correlation of 0 . 91 for simulated systems with two alleles under selection . Further details are given in Supporting Text S1 and Supporting Figures S6 and S7 . | The evolution of the influenza virus is of great importance for human health . Through evolution , current influenza viruses develop the ability to infect people who have been vaccinated against earlier strains . New strains of influenza that infect birds and pigs could evolve to infect and spread between people , causing a global pandemic . The influenza virus lives within a human or animal host , so that viral evolution happens within , or in the spread between , individuals . As such , what happens to the virus during the course of an infection is a question of great interest . We here describe a statistical method that uses viral genome sequence data to measure how evolution affects the influenza virus within a single host . Studying data from infections transmitted between pigs , we find evidence for evolutionary adaptation in six of sixteen animals for which data were available . In one case , an immune response mounted by a pig against the virus is apparent . Our method provides a statistical framework for using sequence data to study viral evolution on very short timescales , enabling new research into within-host viral evolution . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"organismal",
"evolution",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"influenza",
"population",
"genetics",
"microbiology",
"population",
"modeling",
"microbial",
"evolution",
"infectious",
"diseases",
"viral",
"immune",
"evasion",
"infectious",
"disease",
"modeling",
"pathogenesis",
"host-pathogen",
"interactions",
"natural",
"selection",
"virology",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"computational",
"biology",
"evolutionary",
"biology",
"evolutionary",
"processes"
]
| 2014 | Identifying Selection in the Within-Host Evolution of Influenza Using Viral Sequence Data |
Current antiretroviral therapy is effective in suppressing but not eliminating HIV-1 infection . Understanding the source of viral persistence is essential for developing strategies to eradicate HIV-1 infection . We therefore investigated the level of plasma HIV-1 RNA in patients with viremia suppressed to less than 50–75 copies/ml on standard protease inhibitor- or non-nucleoside reverse transcriptase inhibitor-containing antiretroviral therapy using a new , real-time PCR-based assay for HIV-1 RNA with a limit of detection of one copy of HIV-1 RNA . Single copy assay results revealed that >80% of patients on initial antiretroviral therapy for 60 wk had persistent viremia of one copy/ml or more with an overall median of 3 . 1 copies/ml . The level of viremia correlated with pretherapy plasma HIV-1 RNA but not with the specific treatment regimen . Longitudinal studies revealed no significant decline in the level of viremia between 60 and 110 wk of suppressive antiretroviral therapy . These data suggest that the persistent viremia on current antiretroviral therapy is derived , at least in part , from long-lived cells that are infected prior to initiation of therapy .
Infection with HIV-1 results in progressive immunodeficiency and death from opportunistic infection or cancer . Current antiretroviral therapy is effective in suppressing plasma viremia to levels below the detection limit of FDA-approved assays ( 50–75 copies HIV-1 RNA/ml ) , restoring immune function , and reducing morbidity and mortality . Antiretroviral therapy does not cure HIV-1 infection , however , and , at a minimum , eradication of HIV-1 infection will require complete suppression of its replication as well as elimination of latent viral reservoirs . Although some reports have shown the persistence of low-level viremia in patients on suppressive therapy [1–6] , the determinants of this viremia , and whether it results from ongoing replication cycles or release from latent reservoirs , are not well defined [2 , 7–9] due in part to the limited sensitivity of prior HIV-1 RNA assays . To investigate these issues , we developed a real-time PCR-based method ( single-copy assay , SCA ) capable of detecting and reliably quantifying HIV-1 RNA with a limit of one copy per ml plasma [10] . Using this assay , we have found that more than 80% of patients on currently recommended antiretroviral therapies have quantifiable viremia for at least 2 y after initiation of therapy . These results have important implications for understanding the mechanism of HIV-1 persistence despite long-term antiretroviral therapy .
We first measured plasma HIV-1 RNA with both an FDA-approved assay ( bDNA; detection limit 75 copies/ml ) and SCA in three patients initiating antiretroviral therapy . As expected [11–13] , therapy produced a rapid decline in plasma HIV-1 RNA ( Figure 1 ) , reaching undetectable levels within 50–260 d . HIV-1 RNA values from the two assays were similar at levels that were detectable by both assays , but the SCA continued to detect HIV-1 RNA throughout the sampling period , well below the limit of detection of the bDNA assay . These initial observations suggested that low-level viremia can persist for years in patients receiving suppressive antiretroviral therapy . To investigate this phenomenon in a larger patient population and to compare the effects of different treatment regimens , we analyzed specimens from study M98–863 , a Phase III randomized clinical trial comparing lopinavir/ritonavir ( LPV/r ) and nelfinavir ( NFV ) , each in combination with stavudine and lamivudine , in previously antiretroviral-naïve HIV-1-infected individuals [14] . We studied a subset of 145 patients ( see Figure S1 ) whose plasma HIV-1 RNA declined to less than 50 copies/ml within 24 wk of initiating therapy and remained at this level at all time points through 60 wk . SCA results were available from 130 patients ( 63 on LPV/r and 67 on NFV ) . As shown in Figure 2 , HIV-1 RNA values at week 60 in the two arms combined ranged from <0 . 6–174 copies/ml with a geometric mean ( median ) of 3 . 2 ( 3 . 1 ) copies/ml and with no significant between-arm differences in the mean ( p = 0 . 56 ) or overall distribution of values ( p = 0 . 82 ) . Plasma HIV-1 RNA was below the limit of quantification by SCA in about 17% of patients ( LPV/r arm: 17%; NFV arm: 18% ) . We also determined the distribution of HIV-1 RNA levels in 28 patients enrolled in National Institutes of Health ( NIH ) studies whose viremia was suppressed to <75 copies/ml on non-nucleoside reverse transriptase inhibitor ( NNRTI ) -containing regimens . As shown in Figure 2 , the distribution of SCA values from these patients was comparable to that of patients in the M98–863 trial ( p = 0 . 22 ) ; specifically , the mean values were not significantly different ( p = 0 . 17 ) and there was no difference between efavirenz- and nevirapine-containing regimens ( p = 0 . 29 ) . We investigated virologic , immunologic , demographic , and clinical parameters for correlates of persistent viremia in samples from both arms of the M98–863 trial . Comparison of the HIV-1 RNA level by SCA at week 60 to pretherapy RNA levels ( Figure 3 ) showed a significant correlation ( p < 0 . 001 ) and a median decrease between pretherapy and week 60 HIV-1 RNA values of 14 , 250-fold ( range 448- to1 , 720 , 000-fold ) . The correlation coefficient ( r2 = 0 . 17 ) suggests pretherapy viral RNA level accounts for 17% of the variability in stable viremia on therapy , and that other factors , such as time since infection , may contribute to the level of on-therapy viremia . No significant differences between treatment arms were detected in median HIV-1 RNA reduction , and additional analyses ( unpublished data ) indicated that both the magnitude of HIV-1 RNA reduction and the level of persistent viremia were not associated with pretherapy CD4+ T-cell count , change in CD4+ T-cell count on therapy , age at enrollment , or gender . To examine the stability of the low level of viremia in the M98–863 study , we next performed a longitudinal analysis of HIV-1 RNA after week 60 . Of the 145 patients with pretherapy samples who had suppression of plasma HIV-1 RNA between weeks 24 and 60 , 117 had samples available and continuous suppression ( all values <50 copies/ml , Figure S1 ) at subsequent time points . Results of SCA testing of one to five samples ( median of three ) from each of these patients during weeks 60–110 of therapy are shown in Figure 4 . No significant decreases in HIV-1 RNA levels were detected over time , regardless of whether patients were analyzed overall or separately by treatment group; i . e . , the HIV-1 RNA slopes were not significantly different from 0 overall or in either treatment group . Among all patients combined , a half-life of less than 67 wk could be excluded with 95% confidence , using a mixed model linear regression analysis . The study presented here represents the first large-scale and long-term analysis of persistent viremia in patients on standard antiretroviral therapy . Residual viremia on therapy ( median viral RNA of 3 . 1 copies/ml plasma ) was more than 10-fold below the limit of detection of FDA-approved HIV-1 RNA assays . Viremia was suppressed to less than 0 . 6 copies/ml in approximately one-fifth of patients , nearly 100-fold lower than the detection limit of FDA-approved assays . As such , SCA permitted a more accurate estimate of the maximal effect of antiretroviral therapy . Comparison of pre- and post-therapy levels revealed a median reduction in HIV-1 RNA of more than 14 , 000-fold . After this maximal reduction in viremia occurred , however , HIV-1 RNA levels remained stable with no significant decline between the first and second years of therapy . In previous studies , investigators using smaller datasets have reported slow declines in plasma HIV-1 RNA corresponding to half-lives of infected cells of 6 mo [1 , 6] or longer [8 , 15] . Our longitudinal analysis , using a much larger dataset , revealed no significant evidence of decline of HIV-1 RNA levels between 60 and 110 wk of therapy . It is possible , however , that a slow decline in HIV-1 RNA with half-life of greater than 67 wk could have been missed with the current dataset . We are currently assembling a new set of samples with longer follow-up to investigate this possibility . Three sources of persistent viremia have been proposed: ( i ) ongoing cycles of viral replication in the presence of antiretroviral drugs because of inadequate drug inhibitory potency; ( ii ) production of HIV-1 from sanctuary sites into which antiretrovirals do not penetrate sufficiently; or ( iii ) reservoirs of long-lived cells infected prior to the initiation of therapy with production of HIV-1 from integrated proviruses , but without complete cycles of replication occurring because of blockade by antiretroviral drugs [8 , 15–17] . The marked stability of viremia detected in this study , in addition to the similar levels of suppression by LPV/r- , NFV- , or NNRTI-containing regimens , strongly suggests that current antiretroviral therapies inhibit HIV-1 replication cycles and virus production to below a background level set by virus released from long-lived cells . More specifically , the residual viremia is likely derived from a reservoir of long-lived cells infected before the initiation of therapy . This reservoir might comprise chronically infected cells that produce virus at a low stable rate and/or latently-infected cells , like resting CD4+ T cells , that can be isolated from infected individuals on suppressive therapy and activated to produce infectious virus [8 , 15 , 18 , 19] . Our data describe only the level of viremia , not other viral characteristics . Others have reported no accumulation of new resistance mutations and little genetic variation over time in patients with suppressed viremia . Recent analyses have suggested that a limited number of genetic variants circulate during suppression [20–22] , also consistent with a model of HIV-1 production from long-lived cells . Analyses of HIV-1 replication during suppressive antiretroviral therapy intended to distinguish between active replication cycles and production from reservoirs have yielded conflicting results [6 , 23 , 24] , and additional treatment intensification trials to distinguish these possibilities are underway ( NIH study 02-I-0232 , http://clinicalstudies . info . nih . gov/cgi/protinstitute . cgi ? NIAID . 0 . html ) . Information from such trials is essential in designing future antiretroviral strategies to fully suppress viremia . This study analyzed only those participants in the M98–863 study whose viremia remained suppressed to <50 copies/ml for 60–110 wk . As demonstrated in prior intent-to-treat analyses of M98–863 , LPV/r had superior efficacy compared to NFV as judged by Kaplan-Meier analysis of plasma HIV-1 RNA rebound to greater than 400 copies/ml through 96 wk [14 , 25–27] . The present results imply that the difference in efficacy is not associated with differences in residual viremia in the two groups . Early therapeutic failures have been hypothesized to result from the emergence of mutations that exist prior to initiation of antiretroviral therapy [16 , 28] . In this regard , the superiority of LPV/r may be related in part to its contribution to a higher genetic barrier to resistance for the combination regimen , as compared to NFV , for which a single mutation confers significant resistance [29 , 30] . Other factors that may have contributed to the differential efficacy observed between LPV/r and NFV are related to differences in pharmacokinetic profiles , as lopinavir plasma concentrations exceed the IC50 for wild-type HIV by a considerably greater margin than observed with NFV . Consequently , individual pharmacokinetic variability and adherence lapses may be more likely to result in loss of suppression with NFV-based regimens compared to LPV/r-based therapy . Similar levels of persistent viremia in patients undergoing NNRTI-containing therapy for prolonged periods ( mean = 111 . 2 wk ) were noted even though the genetic barrier to NNRTI is likely to be low; again , early drug failures due to pre-existing mutations were likely excluded from selection . The presence of persistent viremia has a number of important implications for development and application of antiretroviral therapies . Few patients on antiretroviral therapy have “undetectable” viremia if sufficiently sensitive assays are employed , and greater suppression of viremia can probably not be achieved using current inhibitors of HIV-1 replication because persistent viremia likely arises , at least in part , from chronically infected cells . New therapeutic strategies will be needed to eliminate persistent viremia and its source .
Plasma samples were obtained from stored specimens from patients with plasma HIV-1 RNA <50 copies/ml in the M98–863 trial and from patients on treatment at the NIH Clinical Center as follows . M98–863 was a randomized double-blind , Phase III study comparing LPV/r ( 400/100 mg twice daily ) ( n = 326 ) with NFV ( 750 mg three times daily or 1 , 250 mg twice daily ) ( n = 327 ) , each in combination with stavudine and lamivudine , in previously antiretroviral-naïve HIV-1-infected patients [14] . All patients in M98–863 who achieved plasma HIV-1 RNA values <50 copies/ml by week 24 of therapy and who remained undetectable at that level at all study visits through week 60 of follow-up were identified for analysis ( n = 237 ) . From this group , a subset of 163 participants treated at investigational sites in North America was identified for SCA testing . Of these , six were excluded for lack of archived samples and 12 for inefficient amplification of baseline sample . Pretherapy samples of 145 participants were amplifiable by SCA ( Figure S1 ) , and only these participants were candidates for further analyses . Of these , 130 also had valid assay results for week 60 samples and constituted the week 60 group analysis . All 145 participants with amplifiable pretherapy samples who remained on study and maintained suppressed plasma HIV-1 RNA levels <50 copies/ml through 110 wk of follow-up were considered for longitudinal analysis . As shown in Figure S1 , six participants did not have samples available after week 60; 21 did not maintain complete suppression after week 60 by conventional viral RNA testing; and one had no valid assay results during analysis . Therefore , 117 participants were included in the longitudinal analysis . Prior to participation in the study , all M98–863 patients provided informed consent for viral RNA quantitation . The protocols and procedures for subsequent SCA analysis of these samples by the HIV Drug Resistance Program were reviewed and approved by the NIAID IRB . Patients taking NNRTI-containing regimens with suppressed plasma HIV-1 RNA levels by FDA-approved testing were also recruited from treatment studies performed at the NIH Clinical Center ( n = 28 ) . All patients were taking Department of Health and Human Services ( DHHS ) guideline-approved antiretroviral regimens containing efavirenz ( n = 22 ) or nevirapine ( n = 6 ) and were either sampled frequently by bDNA or SCA following initiation of therapy or had been on therapy for 20 to >250 wk ( mean 111 . 2 wk ) prior to SCA sampling . All patients were well and without history of recent intercurrent illness at the time of phlebotomy , had hemoglobin levels ≥12 g/dl , and provided informed consent for phlebotomy and for research sample storage . The SCA for HIV-1 RNA detection was performed as described previously [10] , starting with 7 ml plasma , except for the M98–863 samples , from which only 3 ml plasma was available . As a result , the lower limit of quantitation for these samples was 0 . 6 copies HIV-1 RNA/ml , compared to 0 . 3 copies/ml when 7 ml plasma was used . In all analyses , values below the assay quantitation limit for study M98–863 ( 0 . 6 copies/mL ) were considered to be equal to the assay limit . Plasma was obtained from whole blood samples within 2–4 h of phlebotomy and immediately frozen at −70 °C . [10 , 31] . For each sample , three separate aliquots of the cDNA product were assayed for HIV-1 RNA and two aliquots for the recombinant avian retrovirus internal standard RNA using real-time PCR of conserved sequences within gag as described [10] . About 10% of M98–863 patients were excluded because SCA and Amplicor assays on pretherapy samples were significantly discordant due to inefficient amplification by SCA ( Figure S2 ) , probably a result of polymorphism in the probe or primer sequences ( A . Wiegand and S . Palmer , unpublished data ) . In all other samples , there was a close correlation between commercial Amplicor RT-PCR and SCA values for pretherapy samples ( r2 = 0 . 61 , Figure S2 ) . Further details of extraction , optimum amplification conditions , and performance characteristics , as well as quality control procedures to prevent artifactual amplification , have been described [10] . Study M98–863 used the PCR-based Amplicor HIV-1 MONITOR assay version 1 . 0 for plasma HIV-1 RNA quantitation , performed according to manufacturer's specifications ( Roche Diagnostics , http://www . roche . com ) . For patients taking NNRTI-containing regimens , plasma HIV-1 RNA was quantified using the bDNA-based VERSANT HIV-1 RNA assay version 3 . 0 according to manufacturer's specifications ( Bayer Diagnostics , http://diagnostics . siemens . com ) in a manufacturer-certified site using the semiautomated Bayer System 340 unit . Versant HIV-1 RNA version 3 . 0 is approved by the FDA for clinical use with a limit of quantification of 75 copies HIV-1 RNA/ml plasma , although our reported experience with this assay indicates a quantification limit of 50 copies/ml [32 , 33] . For log-transformed baseline viral RNA determinations , comparisons between Amplicor HIV-1 RNA assays and SCA were conducted using linear regression and Pearson's correlation , as were comparisons between log-transformed viral RNA determinations at baseline and week 60 . Mean SCA values were compared between groups using a one-way analysis of variance; comparisons of distributions of SCA values were conducted using the Kolmogorov-Smirnov test . A linear mixed effects regression model with a spatial linear correlation structure to account for correlation between repeated measurements within participants was used to assess the relationship between time and log-transformed viral determinations using SCA . In all analyses , values below the assay quantitation limit for study M98–863 ( 0 . 6 copies/mL ) were considered to be equal to the assay limit . Sensitivity analyses using other imputation methods did not alter results meaningfully . | Combination antiretroviral therapy is effective in reducing , but not eliminating , HIV-1 replication . Residual viremia during suppressive antiretroviral therapy may arise from a number of sources , including reservoirs of long-lived virus-producing cells , or ongoing complete cycles of viral replication . Here , we used a new , more sensitive assay of HIV-1 RNA to measure residual viremia in a large cohort of patients with prolonged suppression on antiretroviral therapy . We found a persistent , stable level of viremia in patients on prolonged therapy that correlated with pretherapy levels of HIV-1 . Over 80% of patients had viremia ≥1 copy/ml plasma , and the level of viremia was independent of the drug regimen patients were taking . These data strongly suggest that persistent viremia on antiretroviral therapy is likely derived from reservoirs of long-lived virus-producing cells that are not affected by currently available drugs that target new cycles of viral replication . New antiviral strategies that eradicate this reservoir will be necessary to cure HIV-1 infection . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
]
| [
"viruses",
"infectious",
"diseases",
"virology"
]
| 2007 | ART Suppresses Plasma HIV-1 RNA to a Stable Set Point Predicted by Pretherapy Viremia |
Neuropeptide signaling influences animal behavior by modulating neuronal activity and thus altering circuit dynamics . Insect flight is a key innate behavior that very likely requires robust neuromodulation . Cellular and molecular components that help modulate flight behavior are therefore of interest and require investigation . In a genetic RNAi screen for G-protein coupled receptors that regulate flight bout durations , we earlier identified several receptors , including the receptor for the neuropeptide FMRFa ( FMRFaR ) . To further investigate modulation of insect flight by FMRFa we generated CRISPR-Cas9 mutants in the gene encoding the Drosophila FMRFaR . The mutants exhibit significant flight deficits with a focus in dopaminergic cells . Expression of a receptor specific RNAi in adult central dopaminergic neurons resulted in progressive loss of sustained flight . Further , genetic and cellular assays demonstrated that FMRFaR stimulates intracellular calcium signaling through the IP3R and helps maintain neuronal excitability in a subset of dopaminergic neurons for positive modulation of flight bout durations .
Neuromodulation of animal behavior by neuropeptides is ubiquitous among vertebrates and invertebrates [1 , 2] . Unlike fast acting neurotransmitters , neuropeptides and their receptors influence neuronal activity and circuit dynamics by modulating presynaptic neurotransmitter release . The mechanisms for doing this include changes in ion channel and transporter function as well as regulation of gene expression [1 , 2] . The neural action of neuropeptides can either be local or at long distances by release into circulation and can influence intrinsic behaviors such as feeding , mating , sleep and aggression [3 , 4] . An important and critical behavior in flying insects is flight . Altered flight behavior , in particular the inability to maintain long durations of flight bouts , can impinge on the fly’s ability to optimally search and reach food sources as well as sites suitable for egg laying . Neuromodulation of insect flight has thus far been attributed primarily to biogenic amines [5–7] . A role for neuropeptide-based modulation of flight behavior has remained largely unexplored . In invertebrates , neuropeptides activate G-protein coupled receptors ( GPCRs ) [8] followed by generation of soluble second messengers such as cAMP [9–12] or inositol 1 , 4 , 5-trisphosphate ( IP3 ) [13–15] . IP3 binds to and activates the endoplasmic reticulum ( ER ) localized Ca2+ channel , the IP3 receptor ( IP3R ) , resulting in release of calcium from ER-stores [16] . Invertebrate neuropeptide receptors that stimulate IP3-mediated Ca2+ release include the Pigment Dispersing Factor Receptor ( PdfR ) [14] , FMRFaR [13–15] amongst others [8 , 17] . We identified the PdfR and FMRFaR in a genetic screen for neuronal GPCRs that regulate flight in Drosophila through IP3-mediated Ca2+ release [14] . Further investigation of PdfR demonstrated a role for this receptor in both the developing and adult flight circuit , though the identity of Pdf responsive neurons that affect flight remained ambiguous [14] . FMRFaR has been described earlier in the context of an escape response to intense light in Drosophila larvae [13] and the larval to pupal transition under nutrient-limiting conditions [15] . Adult behaviors implicating the FMRFaR include startle-induced locomotor activity [18] and adaptive sleep following heat stress [19] . In the context of flight behavior , a neuronal requirement for IP3-mediated Ca2+ release was initially described in flight deficient IP3R mutants [20] . Subsequent cellular and molecular studies identified a role for IP3-mediated Ca2+ release in dopaminergic neurons during pupal stages , when the flight circuit matures [7] . Interestingly though , flight deficits were also observed upon temporal knockdown of the FMRFaR exclusively in mature neurons [14] . Thus far , flight deficits arising from adult specific reduction of IP3/Ca2+ signaling have remained unexplored . Here we have investigated if FMRFaR mediated Ca2+ signaling is required for Drosophila flight by generating new CRISPR-Cas9 mediated mutants for the FMRFaR . Flight deficits arising from specific knockdown in adult dopaminergic neurons suggests a role for FMRFaR in modulating adult flight and implicate the IP3R and possibly CaMKII as downstream signaling components . Genetic and cellular assays indicate that the FMRFaR on adult dopaminergic neurons helps maintain optimal membrane excitability which could potentially be required for synaptic release of dopamine .
The neuropeptide receptor , FMRFaR , was identified amongst other G-protein coupled receptors ( GPCRS ) as a positive regulator of Drosophila flight in a pan-neuronal screen , where genetic data supported IP3R mediated Ca2+ signaling and store-operated calcium entry ( SOCE ) as the down-stream effectors of receptor activation [14] . We confirmed these initial observations by RNAi mediated knockdown of the FMRFaR with another pan-neuronal GAL4 ( nSybGAL4 ) in the central nervous system ( CNS ) , followed by measurement of flight bout durations in tethered flies in response to a gentle air-puff . Unlike previous measurements where flight times were recorded for a maximum of 30 seconds [14] , in the current study we monitored flight for 15 minutes ( 900 seconds ) , which allows for resolution of longer flight bout durations . Upon FMRFaR knockdown in nSyb positive neurons , significantly shorter flight bouts were observed as compared to controls ( FMRFaRRNAi/+ , nSyb/+; Fig 1A ) . The efficacy of FMRFaR knockdown by the RNAi strain was confirmed in the adult CNS by pan-neuronal expression of the RNAi with nSybGAL4 . A significant reduction in FMRFaR mRNA levels was observed ( Fig 1B ) . Similar to the pan-neuronal knockdown of FMRFaR , flight deficits were also observed in flies with FMRFaRRNAi targeted to a large number of dopamine-synthesizing cells driven by THGAL4 ( Fig 1A ) . Because non-neuronal expression of THGAL4 has been documented [21] , we tested GAL4s with restricted expression in neurons [22] . For this , we used the THD and THCGAL4 driver lines that mark non-overlapping subsets of central dopaminergic neurons [22] . Knockdown of FMRFaR using either the THD1GAL4 or the THD’GAL4 resulted in flight deficits that were equivalent to the phenotype observed with THGAL4-driven knockdown ( Fig 1A and 1C , S1A Fig; TH>FMRFaRRNAi vs . THD1>FMRFaRRNAi or TH>FMRFaRRNAi vs THD’>FMRFaRRNAi are not significantly different from each other; p>0 . 05; Mann-Whitney U-test ) . However , the THDGAL4 strains are also known to express in some non-dopaminergic neurons [22] . To test the role , if any , of FMRFaR function in non-dopaminergic neurons marked by THDGAL4s , expression of the FMRFaRRNAi was blocked specifically in TH-expressing neurons with the THGAL80 transgene [23] . THGAL80 was made with the same promoter region of the TH gene as that employed for THGAL4 and has been shown to effectively suppress THGAL4 driven GFP expression [23] . In presence of the THGAL80 transgene , THD1 and THD’ driven GFP expression was selectively blocked in dopaminergic neurons , as identified by the presence of anti-TH immunostaining and the absence of GFP in a majority of neurons ( S1B Fig ) . However , a few GFP positive and TH negative neurons remain in both THD1 and THD’GAL4 strains even in the presence of THGAL80 ( S1B Fig; white arrows ) . Presumably , the shorter genomic regions used for making THD1 and THD’ lend themselves to the influence of neighbouring genomic regions resulting in some ectopic expression in non-TH neurons . Importantly , the flight deficits observed in THD>FMRFaRRNAi flies were reversed upon introduction of THGAL80 confirming that such flight deficits arose exclusively from dopaminergic neurons marked by these GAL4s ( Fig 1A; blue bar , S1A Fig; blue bar ) . On the other hand , flies with knockdown of the FMRFaR using the THC1 or THC’GAL4s did not result in significant flight deficits , indicating that dopaminergic neurons unique to THD1 or THD’GAL4 require FMRFaR mediated signaling to regulate flight bout durations ( S1A Fig ) . THDGAL4 strains express in central dopaminergic neurons comprising the Protocerebral Posterior Lateral 1 ( PPL1 ) and Protocerebral Posterior Medial 3 ( PPM3 ) clusters and these two neuronal clusters are not marked by the THC’GAL4 [7 , 22] . Furthermore , the reduced flight deficit observed with FMRFaR knockdown in dopaminergic neurons was not a consequence of overall reduction in motor activity because THD1>FMRFaRRNAi flies showed normal locomotor activity as compared to their genetic controls ( S1C Fig ) . Having identified the dopaminergic subsets where FMRFaR is required for flight , subsequent experiments were performed with THD1GAL4 as the dopaminergic expression driver . RNAi knockdown of FMRFaR with nSybGAL4 reduced FMRFaR levels by about 30% ( Fig 1B ) . To further refine our understanding of the FMRFaR’s role in Drosophila flight , a complete knockout of the FMRFaR ( ΔFMRFaR ) was generated by the CRISPR-Cas9 method [24] . The strategy employed , removed nearly 1 . 5 kb of the 1 . 6 kb coding region of FMRFaR ( Fig 1D ) . Primers spanning the FMRFaR locus ( 5’F+3’R ) amplified only about ~400bp genomic fragment from FMRFaR homozygous and heterozygous knockouts , thereby confirming FMRFaR gene knockout ( Fig 1D ) . Other primer pair combinations ( 5’F+5’R and 3’F+3’R ) helped us distinguish between the homozygotes and heterozygotes . Homozygous knockouts of the FMRFaR showed near complete loss of FMRFaR mRNA as well as reduced flight bout durations as compared to the heterozygotes ( Fig 1E and 1F ) . Next , we tested dopaminergic neuron specific knockout of FMRFaR by expression of UAScas9 in the genetic background of flies with ubiquitous expression of guide RNAs that target the gene for FMRFaR ( THD1>FMRFaRdual;cas9; see Materials and methods ) . Flies with knockout of the FMRFaR in THD1 cells flew for significantly shorter durations as compared to the appropriate genetic controls ( FMRFaRdual/+;cas/+—Fig 1F; 4th bar in black and THD1/+—Fig 1A; last bar in black; p<0 . 01; Mann-Whitney U-test ) ( Fig 1F ) . To account for non-specific effects of cas9 expression , we tested flies expressing only the cas9 transgene in THD1 neurons ( THD1>cas9 ) . The flight deficit observed with FMRFaR knockout in THD1 cells was significantly different from THD1>cas9 flies ( Fig 1F; THD1>cas9 vs THD1>FMRFaRdual;cas9; p<0 . 01; Mann-Whitney U-test ) . The near identical flight deficits observed between FMRFaR null flies and flies with FMRFaR knockout in THD1 cells suggests that modulation of flight bout duration by FMRFaR may derive solely from dopaminergic neurons ( Fig 1F; THD1>FMRFaRdual;cas9 vs ΔFMRFaR Homozygous; p>0 . 05; Mann-Whitney U-test ) . To test this idea , we next investigated whether FMRFaR is enriched in dopaminergic neurons of the brain . Owing to the lack of either an antibody for FMRFaR or GAL4 strains that reliably mark FMRFaR expressing neurons , we chose to employ molecular and functional assays to confirm the presence of FMRFaR on dopaminergic neurons . Towards this end , we sorted adult brain dopaminergic neurons marked by THGAL4 driven eGFP , using fluorescence-activated cell sorting ( FACS ) and measured transcript levels of a few selected genes . Two markers of dopaminergic neurons , ple ( encoding the enzyme Tyrosine Hydroxylase or TH ) and dDAT ( encoding a dopamine transporter ) were highly expressed in GFP +ve cells , confirming homogeneity of the sorted population ( Fig 2A ) . FMRFaR expression , as measured by qPCR , was significantly enriched in TH expressing neurons ( GFP +ve ) , whereas expression of calcium signaling molecules like dSTIM and dOrai was similar in dopaminergic and non-dopaminergic neurons ( Fig 2A ) . The presence of active FMRFaR on central dopaminergic neurons was tested next . THD1 neurons with expression of a genetically encoded calcium sensor , GCaMP6m [25] were tested for receptor activation in adult brain explants and calcium responses were specifically monitored from the PPL1 and PPM3 clusters , previously implicated in the regulation of flight bout durations ( Fig 1A , S1A Fig ) [7] . Stimulation with one of the most abundantly expressed neuropeptide ligands of FMRFaR , DPKQDFMRFa ( henceforth referred to as FMRFa ) [26 , 27] , resulted in a slow calcium rise ( Fig 2B , 2C , 2E and 2F ) . FMRFa peptide stimulated GCaMP6m response was significantly attenuated upon knockdown of the FMRFaR using either the RNAi or FMRFaRdual;cas9 ( Fig 2B , 2C , 2E and 2F ) . Peptide stimulation of THD1 cells in the presence of a sodium channel inhibitor , 2 μM Tetrodotoxin ( TTX ) resulted in calcium responses that were comparable to that obtained in the absence of TTX ( Fig 2D , 2E and 2F ) . This suggested that the rise in FMRFa-stimulated Ca2+ is due to direct activation of the FMRFaR and not a consequence of synaptic inputs from other neurons to the THD1 neurons . Moreover , FMRFa stimulated Ca2+ signals attenuated significantly upon knockdown of itpr in THD1 marked dopaminergic neurons ( Fig 2D , 2E and 2F ) . Taken together , these data suggest the presence of FMRFaR on central brain dopaminergic neurons and that receptor activation likely leads to Ca2+ release from the IP3R . Pan-neuronal knockdown of the FMRFaR in an earlier study suggested a requirement of the receptor both during pupal development and in adult Drosophila neurons [14] . To investigate the developmental stage ( s ) when FMRFaR signaling is required in dopaminergic neurons for maintaining flight , the TARGET ( Temporal And Regional Gene Expression Targeting ) system [28] was used for stage-specific FMRFaR knockdown . This system employs a temperature sensitive GAL80 element ( TubGAL80ts ) which represses GAL4 expression at 18°C . At 29°C the GAL80ts is inactivated , thus allowing GAL4 driven expression of the UAS transgene . Flight durations of flies with FMRFaR knockdown in dopaminergic neurons throughout development ( 29°C ) were reduced to less than 300 seconds as compared to genetic controls that flew for a median time of about 600 seconds ( Fig 3A , S2A Fig ) . Flies with knockdown of the FMRFaR in larval stages displayed normal flight bouts ( 29°C Larval ) , whereas a modest reduction in flight bout durations was observed with pupal knockdown ( 29°C Pupal ) ( Fig 3A ) . Interestingly , knockdown of the FMRFaR in adult stages affected flight duration maximally , with exacerbation of the flight deficit over time , such that flies with knockdown for eight days flew for less than 150 seconds ( 29°C Adults ) ( Fig 3A ) . In contrast , the control genotypes maintained at 29°C for the same adult period could fly for significantly longer and showed a median flight time of over 600 seconds ( S2A Fig ) . Next , we tested the ability of IP3R overexpression to compensate for loss of FMRFaR in adult dopaminergic neurons . Indeed , flight deficits observed with FMRFaR knockdown in adults could be rescued to a significant extent by simultaneous overexpression of an itpr+ transgene ( Fig 3A; blue bar; p<0 . 01; Mann-Whitney U-test ) , but not another UAS transgene , GCaMP6m ( Fig 3A; orange bar; p>0 . 05; Mann-Whitney U-test ) . Rescue of FMRFaR knockdown by overexpression of itpr+ supports earlier findings that IP3R-mediated Ca2+ release functions downstream of FMRFaR [13] and is consistent with observations in Fig 2D , 2E and 2F . Temporal and cell specific knockout of the FMRFaR with FMRFaRdual was attempted , but was not pursued because control flies with expression of cas9 in adult THD1;TARGET cells exhibit flight deficits when maintained at 29°C for 8 days ( S2A Fig , last lane in pink ) . Consistent adult specific phenotypes were observed with the THD1GAL4 strain which has been used for subsequent experiments . The cellular basis for flight deficits arising from loss of FMRFaR in THD1 marked neurons was investigated next . Expression of endogenous TH and membrane-bound GFP driven by THD1GAL4 was visualized in the PPL1 and PPM3 clusters of adult Drosophila brains at 8 days after knockdown of FMRFaR . Both TH and GFP immunoreactivity appeared as in a control genotype , and there was no apparent loss of dopaminergic neurons or their neurite projections ( Fig 3B , 3C and 3D , S2B Fig ) . Taken together these data suggest that the FMRFaR is required in adult THD1 neurons for sustaining flight bout durations . The increasing disability to maintain flight bouts for longer durations upon manipulation of FMRFaR levels in adult THD1 neurons could likely be a consequence of increased RNAi expression over time . Downstream signaling mechanisms that require FMRFaR activation in THD1 dopaminergic neurons were investigated next . There is evidence demonstrating that FMRFa evoked modulation of excitatory junction potentials ( EJPs ) at Drosophila neuro-muscular junctions are dependent on the protein kinase , CaMKII [13 , 29] . Hence , we tested whether overexpression of wild-type CaMKII ( WT-CaMKII ) in dopaminergic cells rescued the deficit in flight bout durations observed upon reduced FMRFaR signaling . Overexpression of WT-CaMKII during pupal stages was insufficient to rescue flight deficits in adults ( Fig 4A ) . However overexpression of WT-CaMKII in adults modestly rescued FMRFaRRNAi induced flight deficits on day 4 , 6 and 8 ( Fig 4A , S2A Fig—THD1;TARGET controls and S3A Fig–WT-CaMKII/+;FMRFaRRNAi/+ controls ) . It should be noted that overexpression of WT-CaMKII in adult dopaminergic neurons also reduced the duration of flight bouts ( S3A Fig; orange bars ) , suggesting that CaMKII levels need to be tightly regulated in adult dopaminergic neurons for maintenance of flight ( see Discussion ) . Lack of robust rescue by CaMKII of flight deficits in FMRFaR knockdown flies may in part arise from an imbalance between levels of FMRFaR and CaMKII in THD1 neurons . Rescue with the constitutively active form of CaMKII ( CaMKIIT287D ) was not attempted because expression of CaMKIIT287D in adult THD1 neurons resulted in strong flight deficits ( S3A Fig; maroon bars ) . Reduced neuronal excitability by overexpression of the mutant transgene , CaMKIIT287D in larval neurons has been observed previously where it elicited behavioral deficits , presumably via modulation of potassium currents [30] ( also see Discussion ) . Overall the partial rescue of flight deficits in the FMRFaR knockdown animals by expression of a WT-CaMKII transgene indicates a genetic interaction between them . To test directly if CaMKII is required in dopaminergic neurons to modulate flight bout durations , we expressed a synthetic peptide inhibitor of CaMKII ( Ala ) [31] in dopaminergic cells of interest . Ala is a peptide analog of the CaMKII autoinhibitory domain that can bind to the catalytic domain of CaMKII , thereby functioning as an exogenous inhibitor . Inhibition of CaMKII in pan-dopaminergic ( TH ) or the subset dopaminergic neurons ( THD1 ) resulted in flies with significantly reduced durations of flight bouts ( Fig 4B ) . Knockdown of CaMKII with a CaMKIIRNAi also elicited a significant reduction of flight bout durations ( S3B Fig ) . Inhibition of CaMKII activity by the Ala peptide in THD1 neurons did not affect their general locomotor ability , indicating that CaMKII function in THD1 neurons is likely to be flight specific ( S3C Fig ) . Additionally , TARGET experiments demonstrated that THD1-driven expression of Ala specifically in either pupal or adult stages significantly compromised the duration of flight bouts , as compared to genetic controls ( Fig 4C , S2A Fig—THD1;TARGET controls , S3D Fig—Ala/+ controls ) . Given that CaMKII is involved in several developmental processes , for example , axon terminal growth [32] , a pupal requirement for CaMKII is not surprising . Maximal reduction in flight bout duration was observed in flies wherein CaMKII was inhibited by Ala peptide expression in THD1 cells for 8 days as adults ( Fig 4C ) . These data support a role for CaMKII activity in the THD1 subset of dopaminergic neurons in the specific context of maintaining flight bout durations . Similar to FMRFaR knockdown , CaMKII inhibition with Ala in adult neurons did not affect levels of TH mRNA ( S3E Fig ) . Thus , unlike previous findings where reduced Ca2+ signaling in pupal dopaminergic neurons resulted in reduced expression of the gene encoding TH ( ple or TH ) [6 , 7] , FMRFaR and CaMKII in adult dopaminergic neurons appear to modulate flight by an alternate cellular mechanism . To test if CaMKII activity affects FMRFaR induced Ca2+ release , THD1 neurons expressing Ala peptide were stimulated with the FMRFa peptide . FMRFa-stimulated cellular calcium responses were not affected by Ala expression in THD1 neurons ( Fig 4D , 4E and 4F ) . These data suggest that the function of CaMKII is either downstream or parallel to FMRFaR induced Ca2+ release . The ability of FMRFa to activate CaMKII in Drosophila central brain neurons was tested next by immunostaining for pCaMKII , a CaMKII modification that occurs upon prolonged calcium elevation in the cell [30 , 33] . For technical reasons , this immunostaining was performed on primary neuronal cultures from larvae , wherein we overexpressed the FMRFaR+ in all nSybGAL4 positive neurons [34] . In addition , the same neurons were marked with mCD8GFP to normalize the pCaMKII signal and to account for variability in strength of expression of the overexpressed FMRFaR+ . A significant increase in pCaMKII/GFP staining was observed upon stimulation of larval nSyb>FMRFaR+;mCD8GFP positive neurons with FMRFa as compared to cells with solvent addition ( S3F–S3H Fig ) . These data support the idea that FMRFa stimulated Ca2+ release activates CaMKII in central brain neurons . Even though a direct test for FMRFa-stimulated CaMKII activation in THD1 neurons has not been possible so far due to technical reasons , the ability of WT-CaMKII to partially rescue the flight deficit in FMRFaR knockdown animals ( Fig 4A ) , taken together with these cellular data suggest that CaMKII may function downstream of FMRFaR signaling . Nevertheless , an FMRFaR-independent and parallel role for CaMKII in THD1 neurons remains possible . The underlying cellular basis for the flight deficits observed upon FMRFaR knockdown and Ala expression in dopaminergic neurons was investigated next . To measure neuronal activity of THD1 neurons , we tested their response to a depolarizing stimulus . Ex vivo brain preparations from THD1 marked dopaminergic neurons were stimulated with 70 mM KCl , a condition known to raise the resting membrane potential and activate voltage-gated calcium channels on the plasma membrane [35] . The stimulation was followed by optical measurements of calcium dynamics . Firstly , using the TARGET system , we expressed GCaMP6m in THD1 marked cells , specifically in adults . Animals reared at 29°C for 8 days ( also the time corresponding to maximal flight loss upon FMRFaR knockdown or Ala expression; Figs 3A and 4C ) , were used in these experiments . Robust Ca2+ elevations were observed in THD1 marked cells from control animals upon addition of KCl ( Fig 5A–5D; black trace ) . Ca2+ responses from neurons with either FMRFaR knockdown or Ala expression , were however significantly reduced as compared to controls ( Fig 5A–5D; red and blue traces ) . Similar experiments on ex vivo brains obtained after FMRFaR knockdown in adults for 2 days , exhibit a minimal decrease in Ca2+ responses upon depolarization with KCl , in agreement with their weak flight deficits ( S4A–S4C Fig , Fig 3A ) . These data suggest that FMRFaR and CaMKII help maintain normal cellular responses to changes in membrane excitability in THD1 neurons . Interestingly , in the FMRFaR knockdown condition , most cells exhibit a dampened Ca2+ response ( Fig 5D , 2nd row—pink arrow ) , but a few cells responded normally ( Fig 5D , 2nd row—yellow arrow ) . Their normal response may be due to insufficient receptor knockdown . Alternately all THD1 marked cells might not express the FMRFaR ( see Discussion ) . Based on the reduced Ca2+ elevation upon depolarization with KCl , observed in THD1 marked dopaminergic neurons of flies with either FMRFaR knockdown or CaMKII inhibition , we hypothesized that such neurons might have reduced membrane excitability . To ascertain if reduced FMRFaR signaling indeed alters the ability of THD1 neurons to undergo membrane depolarization , a genetically encoded fluorescent voltage indicator , Arclight [36] , was expressed in dopaminergic neurons of THD1GAL4 . Stimulation of 8 day old adult brains with KCl showed a robust change in Arclight fluorescence corresponding to neuron depolarization ( Fig 6A–6D; black trace ) . As proof of concept , expression of Kir2 . 1 in THD1 cells for 2 days nearly abolished KCl induced membrane depolarization ( S5A–S5C Fig; blue trace ) . Dopaminergic neurons with FMRFaR knockdown for 8 days displayed a significantly reduced ability to depolarize after KCl addition ( Fig 6A–6D; red trace ) . As was observed previously ( Fig 5D ) , membrane depolarization induced changes in Arclight fluorescence were attenuated in most THD1 neurons ( Fig 6D , 2nd row; pink arrow ) , while few responded like controls ( Fig 6D , 2nd row; yellow arrow ) . Similar experiments with adult brains at 2 days exhibit equivalent changes in membrane potential in both control and with FMRFaRRNAi , in agreement with the weaker flight deficits of 2d old flies with FMRFaR knockdown ( Figs 3A and S5A–S5C ) . Because knockdown of FMRFaR reduced KCl induced Ca2+ entry in dopaminergic neurons ( Fig 5A–5D ) , attenuated membrane depolarization ( Fig 6A–6D ) and correlated with shortened flight bout durations ( Fig 3A ) , we hypothesized that FMRFaR function is likely required to maintain optimal excitability in these neurons . Consequently , increasing neuronal excitability might compensate for reduced FMRFaR function and thereby rescue the flight deficits observed in FMRFaR knockdown flies . To test this idea , we introduced a temperature sensitive cation channel , dTrpA1 [37] in flies with adult specific FMRFaR knockdown . Activation of the dTrpA1 calcium channel can compensate in part for reduced Ca2+ entry through voltage–gated channels [15 , 38] . Indeed , expression of dTrpA1 significantly improved the maintenance of flight bouts in 6 and 8 day old FMRFaR knockdown adults ( Fig 6E , S2A Fig—THD1;TARGET controls and S5D Fig–dTrpA1/+;FMRFaRRNAi/+ controls ) . Likewise , expression of NaChBac ( a bacterial sodium channel ) [39] , moderately alleviated the near loss of flight observed with Ala-dependent CaMKII inhibition in 6 and 8 day old adults ( S5E and S5F Fig ) . Expression of another UAS element ( GCaMP6m ) in the background of FMRFaR knockdown or Ala expression did not rescue the loss of flight ( Fig 3A–orange bar , S5F Fig—green bar , as compared to Fig 4C—orange bar; 8 days as Adults at 29°C; p>0 . 05; Mann-Whitney U-test ) . Expression of just the dTrpA1 or the NaChBac transgenes in adult dopaminergic neurons also affected flight bout durations , supporting the idea that optimal activity of ion channels in THD1 marked adult dopaminergic neurons is required for their normal function ( S5D and S5F Fig ) . Taken together , these data demonstrate that FMRFaR helps maintain optimal excitability of flight bout extending central dopaminergic neurons . We hypothesized that changes in membrane excitability are likely to affect the release of dopamine containing synaptic vesicles from THD1 neurons . A temperature-sensitive mutant transgene for dynamin ( UAS-Shibirets or Shits ) [40] was used to block synaptic vesicle recycling by shifting flies transiently to the non-permissive temperature of 30°C . Indeed , when synaptic vesicle recycling was inhibited in THD1 neurons of adult flies , reduced flight bout durations were observed as compared with controls ( Fig 7A ) . Flies of the same genotype ( THD1>Shits ) maintained at 22°C throughout showed normal flight bout durations ( Fig 7A ) . Thus acute manipulation of synaptic function of adult THD1 neurons affects the maintenance of flight bout duration in the tethered flight paradigm . Finally , we tested requirement for the neurotransmitter , dopamine in THD1 neurons , for modulating the duration of adult flight bouts . Adult-specific expression of an RNAi transgene for the dopamine-synthesizing enzyme Tyrosine Hydroxylase ( THRNAi; [41] ) in THD1 neurons significantly impaired the maintenance of longer flight bouts from as early as 2 days of THRNAi expression ( Fig 7B ) , suggesting that dopamine release from synapses of THD1 neurons is required for maintaining longer flight bouts .
In this study we demonstrate that the FMRFaR and CaMKII signaling in specific central dopaminergic neurons of the adult Drosophila brain helps maintain optimal membrane excitability ( Figs 5A–5D and 6A–7D ) . Previous work by our group demonstrated that loss of IP3R mediated Ca2+ signaling in central dopaminergic neurons during pupal stages similarly affects flight circuit function in adults [6 , 7] . In parallel , neuropeptidergic modulation of flight by the Pdf receptor was investigated , but the exact class of neurons that required this receptor or the downstream effectors remained elusive [14] . For the first time , we describe here a neuropeptidergic signaling mechanism in adult brain neurons that is required for sustained flight ( S6A Fig ) . Our data support a model where FMRFa-modulated activity in central dopaminergic neurons belonging to the PPL1 and PPM3 clusters extends the duration of flight bouts , a behavior that is likely to have significant effects on optimal sourcing of nutrients , finding sites for egg laying and identifying mates in the wild . The role of neuropeptide signaling in modulating animal behavior is well documented [2 , 4] . Indeed , rescue of flight deficits by overexpression of the cation channel dTrpA1 in dopaminergic neurons of FMRFaR knockdown flies directly addresses the importance of peptide driven neuronal excitability , for flight behavior ( Fig 6E ) . Reduced cellular responses to a depolarizing stimulus observed upon FMRFaR knockdown in THD1 neurons may arise from functional modification , possibly by CaMKII , of plasma membrane resident ion channels , or their expression levels ( Figs 5A–5D and 6A–6D ) . However , our data do not exclude other kinases or modifiers as functioning downstream of the FMRFaR . We predict that FMRFaR in dopaminergic neurons is a key neuromodulator of membrane excitability which stimulates release of dopamine containing synaptic vesicles . In fact , stimulation of synaptic transmission downstream of FMRFaR activation has been described previously in Drosophila larval neuromuscular junctions [13] . Though we have not tested synaptic vesicle release directly upon FMRFaR stimulation , our data support an essential role for synaptic transmission and dopamine synthesis in adult dopaminergic neurons for maintenance of flight bout durations ( Fig 7A and 7B ) . Identification of the precise ion channels that are affected by FMRFaR signaling , leading to changes in membrane excitability and very likely synaptic transmission in flight modulating dopaminergic neurons , needs further investigation . Multiple FMRFa peptides exist in Drosophila [26 , 27] . Cells positive for FMRFa peptides have been characterized in the Drosophila central nervous system using antisera specific to some of the FMRFa peptides [42–45] . Immunostaining of adult brains , especially against the peptide , DPKQDFMRFa ( also used in this study ) have revealed extensive neuronal varicosities in the anterior and lateral regions of the protocerebrum [44 , 45] . It is hence conceivable that FMRFa released through these projections activates the FMRFaR on the anatomically proximal PPL1 and PPM3 clusters of dopaminergic neurons . The PPL1 neurons are known to innervate the superior protocerebrum , a region considered to function at the interface of olfactory input and motor output modules of the brain [46] . Thus activation of the FMRFaR on dopaminergic neurons might stimulate dopamine release , which could then reinforce a dopaminergic circuit , required for sustained flight . Furthermore , co-expression of the dopamine synthesizing enzyme Tyrosine Hydroxylase ( TH ) in a few neurons marked by an FMRFaGAL4 [47] , suggests the existence of an autocrine signaling mechanism within the flight modulating dopaminergic neurons . This idea is supported by recent studies of single cell sequencing of Drosophila neurons where FMRFa transcripts were seen in a small number of central brain dopaminergic neurons [48 , 49] . Interestingly , a recent report from Manduca demonstrated that an FMRFa positive neuron lies at the center of a putative sensory-motor circuit for integration of olfactory stimuli with wing movements during flight [50] . The natural context in which FMRFa release is triggered for modulation of flight in Drosophila remains to be identified . Neuromodulatory signals from other receptors in central dopaminergic neurons is also a possibility and would widen the scope of sensory inputs received by these cells for integration with flight behavior . FMRFaR activation by the peptide DPKQDFMRFa stimulates intracellular Ca2+ release through the IP3R ( Fig 2D–2F ) [13] and enhances synaptic transmission at the larval neuromuscular junction and thereby modulates muscle contraction [13 , 29 , 45] . Previous work from our lab has demonstrated that in Drosophila neurons , subsequent to intracellular Ca2+ release , cytosolic calcium is further elevated by store-operated Ca2+ entry ( SOCE ) through dOrai [51] . Thus , we predict that the rise in Ca2+ we see upon FMRFa stimulation of THD1 neurons could possibly be a combination of both IP3R mediated Ca2+ release and SOCE . Rescue of flight deficits in the FMRFaR knockdown flies by an itpr+ transgene ( Fig 3A ) and loss of peptide induced Ca2+ rise in an itpr knockdown background ( Fig 2D–2F ) support the idea that IP3R-mediated Ca2+ release is likely downstream of FMRFaR activation . Activation of overexpressed IP3Rs in the FMRFaR knockdown condition might occur through alternate receptors . It is also possible that overexpression of the IP3R allows for more ER-Ca2+ release from the residual FMRFaRs after RNAi knockdown . Thus , in the context of flight as well , FMRFaR signaling in dopaminergic neurons appears to function upstream of IP3R-mediated Ca2+ release ( Figs 2D–2F and 3A ) . The adult requirement of FMRFaR on dopaminergic neurons supports a primary role for this receptor in mature neurons for modulation of flight , but not as much for maturation of the flight circuit in pupal stages ( Fig 3A ) . In contrast , the dFrizzled2 receptor , which also functions upstream of the IP3R in central dopaminergic neurons , was shown to be required exclusively during the pupal stages , albeit in a different dopaminergic cluster of the brain [6] . These observations support the existence of specific receptors that stimulate IP3-mediated Ca2+ release and function either during flight circuit maturation in pupae or in a neuromodulatory role in adults , to influence flight circuit dynamics . A range of flight deficits in IP3R mutants with differing allelic strength support this idea [20] . As described previously by other groups in larval motor neurons and in larval neuromuscular junctions [13 , 29] , our genetic and cellular data also suggests that CaMKII might be a downstream effector of FMRFaR signaling in neurons ( Fig 4A , S3F–S3H Fig ) . Although we have shown increased pCaMKII immunostaining in larval CNS neurons upon FMRFa stimulation ( S3F–S3H Fig ) , direct activation of either CaMKII or other Ca2+ dependent kinases by FMRFa in adult THD1 neurons , remains to be tested rigorously . CaMKII , thus far has been known to contribute to synaptic plasticity in the context of learning and memory [52 , 53] . Our data , for the first time , demonstrate that CaMKII is required in dopaminergic neurons for maintenance of Drosophila flight over periods of several minutes , possibly by modulating neuronal firing during flight . Multiple mechanisms for CaMKII dependent modulation of membrane excitability have been described in Drosophila such as phosphorylation of the potassium channel , Eag , leading to an increase in the Eag current [54] . In another study , CaMKII dependent phosphorylation of the Ca2+ activated potassium channel binding protein , Slob was shown to favor its binding to 14-3-3 , that eventually altered the voltage sensitivity of slowpoke channels [55] . In both cases , CaMKII acted as a negative regulator of neuronal excitability , and support our data demonstrating flight deficits observed upon overexpression of WT-CaMKII and CaMKIIT287D ( S3A Fig ) . However , the more interesting and compelling explanation for our data demonstrating reduced KCl-induced depolarization upon CaMKII inhibition by Ala comes from in vitro work that has described a role for CaMKII in decelerating the inactivation of voltage sensitive calcium channels , thereby enhancing transmitter release [56] . This was however shown to be independent of its kinase activity . More recently , there is evidence to suggest CaMKII-dependent activation of transcription factors in Drosophila neurons [57 , 58] . Hence , the possibility of transcriptional regulation of voltage-gated membrane channel genes by CaMKII cannot be excluded and needs further investigation .
Drosophila strains used in this study were maintained on cornmeal media , supplemented with yeast . Flies were reared at 25°C , unless otherwise mentioned . WT strain of Drosophila used was Canton-S . The fly lines nSybGAL4 ( BL51635 ) , UASFMRFaRRNAi ( BL25858 ) , UAScas9 ( BL54593 ) , UASdTrpA1 ( BL26263 ) , UASNaChBac ( BL9468 ) , UASKir2 . 1 ( BL6595 ) , UASGCaMP6m ( BL42748 ) , UASArclight ( BL51057 ) , UASmCD8GFP ( BL5130 ) , UASAla ( BL29666 ) , UASWT-CaMKII ( BL29662 ) , UASCaMKIIT287D ( BL29664 ) , UASTHRNAi ( BL25796 ) , UASDicer ( BL24648 ) were obtained from Bloomington Drosophila Stock Centre ( BDSC ) . The UASitprRNAi ( 1063-R2 ) strain was from National Institute of Genetics ( NIG ) and the UASCaMKIIRNAi ( v38930 ) was from Vienna Drosophila Resource Center ( VDRC ) . A strain with two copies of TubGAL80ts on the second chromosome was generated by Albert Chiang , NCBS , Bangalore , India , and has been used for all TARGET experiments . The dopaminergic GAL4 driver , THGAL4 [21] and the THGAL80 strain [23] were kindly provided by Serge Birman ( CNRS , ESPCI Paris Tech , France ) . All the TH subset GAL4s ( THD1 , THD’ , THC1 , THC’ ) were a gift from Mark N Wu ( Johns Hopkins University , Baltimore ) [22] . The UASeGFP transgene was provided by Michael Rosbash ( Brandeis University , Waltham , MA ) . The UASShits strain was obtained from Toshihiro Kitamoto ( University of Iowa , Carver College of Medicine , Iowa City ) . The generation and use of UASitpr+ transgene has been described earlier [15 , 59] . The UASFMRFaR+ strain was generated by microinjection of the pUAST-FMRFaR full-length cDNA into fly embryos and generating stable UAS transgenic lines . Standard fly genetics were followed for making strains and recombinants . Temperature shift experiments were performed as described below . Briefly , larvae , pupae or adults were maintained at 18°C and transferred to 29°C only at the stage when the UAS-transgene needed to be expressed . Firstly , as a control , flies of the genotype THD1;TARGET>UAS-transgene were maintained throughout at 18°C , a condition where the UAS-transgene expression is suppressed ( labeled as ‘18°C’ in figures ) . To observe maximum effect , flies of the same genotype were grown at 29°C throughout development and as adults ( labeled ‘29°C’ ) . For larval specific knockdown , animals of the desired genotypes developed at 29°C from embryos to the wandering larval stage , following which they were maintained at 18°C till the time when they were tested for flight ( labeled as ‘29°C Larval’ ) . Likewise , for pupal specific knockdown , wandering third instar larvae were shifted from 18°C to 29°C . Pupae developed at 29°C and were transferred to 18°C , soon after eclosion ( labeled as ‘29°C Pupal’ ) . In case of adult specific knockdown , larval and pupal development continued at 18°C . Following this , adult flies were maintained at 18°C for two days post eclosion and then transferred to 29°C and used for experiments at 1 , 2 , 4 , 6 or 8 days after transfer ( labeled as ‘29°C Adult’ ) . To generate a FMRFaR null allele , we used the CRISPR-Cas9 methodology [60–64] . Two guide RNAs ( gRNAs ) were designed , at the 5’ and 3’ends of the gene at the following sequences: ( sg1-GGGAGCCATGAGTGGTACAGCGG , sg4-GATCTCTGCATTTCGCGGGCGGG ) so as to delete ~1 . 5 kb from the coding region of the 1 . 6 kb gene . A gRNA dual transgenic fly ( FMRFaRdual ) was made first [24] . FMRFaRdual transgenic males were mated with Act5c-Cas9 virgins [65] . From the F1 progeny obtained , 16 were screened by PCR for the deletion . Because all flies tested were positive for the deletion , 4 F1 flies were individually crossed to balancers and 30 F2 progeny were screened for the deletion . Progeny , that were positive for deletion and negative for presence of dual gRNA ( total 6 F2s ) , were maintained as stable lines . The deletion was confirmed by sequencing of the DNA from the FMRFaR knockout flies . The following primers were used for confirmation of deletion—5'F: GACATAGTCATCAGGTGCTC , 5’R: TGCACCTCCGTGTGGTTAAG , 3’F: GAACAACGGCGATGGAACTC , 3'R: GGTGCTCTAAGTCAACCCCT . For tissue specific deletions of FMRFaR gene , a fly strain containing FMRFaRdual and UAScas9 was made and subsequently mated with GAL4 strains of interest . Single flight tests were performed on adult flies 3–5 days post eclosion , unless specified differently . Flies were anaesthetized on ice for 2–3 minutes and then tethered using nail polish applied to the end of a thin metal wire . The tether was glued onto the region between the neck and thorax of adult flies . A gentle mouth blown air-puff was delivered to test for flight response . Flight time in tethered condition was recorded for a maximum of 15 minutes . In three independent batches of 10 , a minimum of 30 flies were tested for each genotype . All control genotypes tested were obtained by crossing the GAL4 or UAS strain of interest to the wild-type strain , Canton-S . Raw data was plotted as box and whisker plots using Origin ( OriginLab , Northampton , MA ) software . Each bar represents 25th to 75th percentile . The solid diamond and horizontal line within the bars indicate mean and median , respectively . Individual values are shown as open diamonds . Because the distribution of flight data was not normal , we have used the non-parametric Mann-Whitney U-test , which compares medians across the datasets . However , the mean is also represented in all bar graphs and is especially useful for data when flight times are shorter and the distributions are normal . Locomotor activity was measured by modifying a previously published protocol [66] . Briefly , locomotor activity of 4–6 day old singly housed virgin males was tested in a circular chamber of diameter 4 cm . The chamber was placed on a sheet containing a specific pattern . Single flies were aspirated into each chamber and allowed to acclimatize for 5 minutes . Each fly was monitored separately for the number of times it crossed each line in the pattern , over a period of 10 seconds . For every experiment , six single flies were tested sequentially for the span of 10 seconds . This was repeated 5 times , i . e . we measured locomotor activity of every single fly five times in blocks of 10 seconds , so as to randomize activity measurements during the experimental time . Total locomotor activity is represented as the sum total of the number of lines crossed by each fly over the experimental duration of 50 seconds ( plotted as Locomotor Activity Units ) . A minimum of 25 flies were tested from each genotype . Adult brains were dissected in adult haemolymph-like solution ( AHL ) [67] and embedded in ~10 μl of 1% low melt agarose ( Invitrogen ) with anterior side facing upwards . Brains were then bathed in 100 μl AHL until imaging . Images were obtained on an Olympus FV1000 inverted microscope ( Olympus Corp . , Japan ) in time lapse mode using the 20X objective . Both GCaMP6m and Arclight signals were captured using the 488 nm excitation laser line . Fifty frames of basal activity were recorded , followed by stimulation with either 5 μM Peptide ( DPKQDFMRFa , NeoBioLab , Massachusetts , USA ) or 70 mM KCl . A minimum of 5 brain explants were imaged for every experimental condition . Raw fluorescence data for regions of interest were extracted using the Time Series Analyser plugin in Image J 1 . 47v . Change in fluorescence , ΔF/F was calculated using the following formula for each time ( t ) : ΔF/F = ( Ft-F0 ) /F0 , where F0 is the average basal fluorescence of the first 10 frames . Area under the curve and Peak ΔF/F was calculated using Microsoft Excel considering all frames post-stimulation . Values were plotted as box plots . Twenty adult CNS ( THGAL4>eGFP ) were dissected in cold Schneider’s medium ( #21720–024 , Life Technologies ) followed by ~45 minutes incubation in an enzymatic solution containing 0 . 75 μg/μl of Collagenase ( Sigma-Aldrich ) and 0 . 40 μg/μl Dispase ( Sigma-Aldrich ) . Brain lysates were then spun at 3000 rpm for 3 minutes . The supernatant was discarded and the pellet was resuspended in cold Schneider’s media . Single cell suspensions were obtained by gentle tituration and then passed through a 40 μm mesh filter . Fluorescence-activated cell sorting ( FACS ) of samples was carried out as described earlier [7] . A minimum of 4 biological replicates were sorted and ~10 , 000 GFP +ve and GFP -ve cells were collected separately in TRIzol Reagent ( Life Technologies ) for RNA isolation . Total RNA was extracted from sorted cells or dissected adult CNS using TRIzol RNA extraction protocol ( Ambion , ThermoFischer Scientific ) . In case of dissected CNS , a minimum of three biological replicates , each containing 5 brains , was used per genotype . Approximately 500 ng of RNA was then used for DNAse treatment and first strand cDNA synthesis . Quantitative real-time PCR was performed on ABI7500 fast machine ( Applied Biosystems ) using Kapa SYBR FAST qCR Master mix ( KAPA Biosystems , Wilmington , MA ) . rp49 was used as an internal control to quantify relative levels of the target transcripts . qPCR values are represented as normalized fold change of mRNA transcripts of the indicated genes . In case of sorted cells , normalized mRNA levels of genes in GFP -ve cells have been plotted relative to the GFP +ve population . Every qPCR run was followed by a melt curve analysis to confirm primer specificity . The following formula was used to calculate fold change: Fold change = 2-ΔΔCt , where , ΔΔCt = ( Ct ( target gene ) -Ct ( rp49 ) ) Experimental— ( Ct ( target gene ) -Ct ( rp49 ) ) Control Sequences of primers used are as follows: rp49: Forward- CGGATCGATATGCTAAGCTGT Reverse- GCGCTTGTTCGATCCGTA FMRFaR: Forward- GTGCGAAAGTTACCCGTCG Reverse- TAATCGTAGTCCGTGGGCG TH ( ple ) : Forward- GTTGCAGCAGCCCAAAAGAAC Reverse- GAGACCGTAATCATTTGCCTTGC dSTIM: Forward- GAAGCAATGGATGTGGTTCTG Reverse- CCGAGTTCGATGAACTGAGAG dOrai: Forward- GAGATAGCCATCCTGTGCTGG Reverse- CGGATGCCCGAGACTGTC Immunohistochemistry was performed on dissected adult CNS as described earlier [7] . Briefly , brains were dissected in 1x PBS , followed by fixation in 4% paraformaldehyde and blocking in 0 . 2% phosphate buffer , 0 . 2% Triton-X 100 and 5% normal goat serum . Overnight incubation with primary antibodies was followed by 2 hours of incubation with corresponding fluorescent secondary antibodies . The primary antibodies used were: mouse anti TH ( 1:50 , #22941 , ImmunoStar ) and rabbit anti GFP ( 1:10 , 000 , A6455 , Life Technologies ) . The following were the fluorescent secondary antibodies used at 1:400: anti-mouse Alexa Fluor 568 ( #A11004 , Life Technologies ) and anti-rabbit Alexa Fluor 488 ( #A11008 , Life Technologies ) . Images were obtained as confocal stacks of 1 μm thickness using 20X objective on Olympus FV1000 Confocal microscope . Primary neuronal cultures and immunostaining of neurons was carried out as described previously [34] . Briefly , third instar larval CNS were dissected in Schneider’s medium containing 50 μg/ml Streptomycin ( Invitrogen ) , 10 μg/ml Amphotericin B ( Invitrogen ) and 50U/ml Penicillin ( Invitrogen ) . Following this , brains were subjected to 40 minutes of enzyme treatment in 0 . 75 μg/ml Collagenase and 0 . 40 μg/ml Dispase . Neurons were dissociated , spun down and plated in growth medium , DMEM/F12-1065 ( Gibco ) supplemented with the antibiotics mentioned previously and 20mM HEPES . All cell culture reagents were procured from Sigma-Aldrich , unless specified differently . Fourteen-sixteen hours old cultures were washed twice with hemolymph-like saline , HL3 and incubated with either the FMRFa peptide or a solvent control for 20 minutes . Following this , cells were fixed in 3 . 5% paraformaldehyde for 20 minutes at room temperature and then washed in wash buffer ( 1/10th dilution of blocking buffer ) . Cells were then permeabilized for a total of half an hour ( 10 minutes * 3 times ) , blocked in blocking buffer ( 5% BSA , 0 . 5% Triton X in PBS ) for 1 hour , at room temperature and subsequently incubated overnight in the primary antibody diluted in wash buffer ( mouse anti pCaMKII-22B1; 1:100; #sc-32289 , Santa Cruz ) . The next day , cells were washed thrice in wash buffer ( for 10 minutes each ) and incubated with anti-mouse Alexa Fluor 568 ( #A11004 , Life Technologies ) secondary antibody for half an hour at room temperature . Control dishes for the secondary antibody were treated as the experimental dishes and were incubated with the fluorescent secondary , but not the primary antibody . Following another set of three washes ( 15 minutes each ) in wash buffer , cells were covered in 60% glycerol before imaging . An inverted Olympus FV3000 confocal microscope with a 60X oil objective ( 1 . 42 NA ) , pinhole size 150 μm was used to image the cells . GFP and pCaMKII signals were captured with 488 and 561 laser lines respectively . The same settings were used to capture images from all experimental conditions , on any particular day of imaging . The entire cell volumes of cells were captured as optical slices of 0 . 60 μm thickness . Images were analyzed in Image J as described above . After applying background subtraction , fluorescence values of pCaMKII were divided over GFP values to obtain the ratios shown in S3G Fig . Origin ( OriginLab , Northampton , MA ) software was used for plotting raw data and calculation of statistical significance . Raw data were tested for normality . A Student t-test or an ANOVA statistical test was performed on normally distributed data . Non-normal data were tested for statistical significance using the Mann-Whitney U-test . | Neuropeptides play an important role in modulating neuronal properties such as excitability and synaptic strength and thereby influence innate behavioral outputs . In flying insects , neuromodulation of flight has been primarily attributed to monoamines . In this study , we have used the genetically amenable fruit fly , Drosophila melanogaster to identify a neuropeptide receptor that is required in adults to modulate flight behavior . We show from both knockdown and knockout studies that the neuropeptide receptor , FMRFaR , present on a few central dopaminergic neurons , modulates the duration of flight bouts . Overexpression of putative downstream molecules , the IP3R , an intracellular Ca2+-release channel , and CaMKII , a protein kinase , significantly rescue the flight deficits induced by knockdown of the FMRFaR . Our data support the idea that FMRFaR and CaMKII help maintain optimal membrane excitability of adult dopaminergic neurons required to sustain longer durations of flight bouts . We speculate that the ability to maintain longer flight bouts in natural conditions enhances the individual’s capacity to search and reach food sources as well as find sites suitable for egg laying . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"neurochemistry",
"chemical",
"compounds",
"dopaminergics",
"depolarization",
"membrane",
"potential",
"electrophysiology",
"neuroscience",
"insect",
"flight",
"biological",
"locomotion",
"animals",
"flight",
"(biology)",
"organic",
"compounds",
"animal",
"models",
"hormones",
"drosophila",
"melanogaster",
"model",
"organisms",
"experimental",
"organism",
"systems",
"amines",
"molecular",
"biology",
"techniques",
"neurotransmitters",
"catecholamines",
"drosophila",
"dopamine",
"research",
"and",
"analysis",
"methods",
"animal",
"cells",
"neurochemicals",
"hyperexpression",
"techniques",
"chemistry",
"molecular",
"biology",
"insects",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"arthropoda",
"gene",
"expression",
"and",
"vector",
"techniques",
"biochemistry",
"cellular",
"neuroscience",
"eukaryota",
"cell",
"biology",
"organic",
"chemistry",
"physiology",
"neurons",
"biogenic",
"amines",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"organisms"
]
| 2018 | FMRFa receptor stimulated Ca2+ signals alter the activity of flight modulating central dopaminergic neurons in Drosophila melanogaster |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.