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 |
|---|---|---|---|---|---|
Arthropod-borne pathogens account for millions of deaths each year . Understanding the genetic mechanisms controlling vector susceptibility to pathogens has profound implications for developing novel strategies for controlling insect-transmitted infectious diseases . The fact that many viruses carry genes that have anti-apoptotic activity has long led to the hypothesis that induction of apoptosis could be a fundamental innate immune response . However , the cellular mechanisms mediating the induction of apoptosis following viral infection remained enigmatic , which has prevented experimental verification of the functional significance of apoptosis in limiting viral infection in insects . In addition , studies with cultured insect cells have shown that there is sometimes a lack of apoptosis , or the pro-apoptotic response happens relatively late , thus casting doubt on the functional significance of apoptosis as an innate immunity . Using in vivo mosquito models and the native route of infection , we found that there is a rapid induction of reaper-like pro-apoptotic genes within a few hours following exposure to DNA or RNA viruses . Recapitulating a similar response in Drosophila , we found that this rapid induction of apoptosis requires the function of P53 and is mediated by a stress–responsive regulatory region upstream of reaper . More importantly , we showed that the rapid induction of apoptosis is responsible for preventing the expression of viral genes and blocking the infection . Genetic changes influencing this rapid induction of reaper-like pro-apoptotic genes led to significant differences in susceptibility to viral infection .
As a genetically regulated mechanism of cell elimination , apoptosis plays an important role in maintaining tissue homeostasis through the removal of obsolete or potentially dangerous cells . The controlled collapse of intracellular infrastructures and encapsulation of cell bodies associated with apoptotic cell death has long led to the speculation that apoptosis could function as an efficient innate immune mechanism against intracellular pathogens such as viruses [1] , [2] , [3] . The majority of evidence supporting the role of apoptosis as an important immune response has come from the study of viruses . Many viruses encode genes that can interfere with the regulation of apoptosis at various levels [4] . For example , the pivotal upstream regulator P53 is a frequent target of viral inhibition . It can be sequestered by the SV ( Simian virus ) 40 T antigen or degraded by proteins encoded by adenoviruses or human papillomaviruses . In addition , it was found recently that adenovirus E4orf3 can block P53-induced gene expression by promoting de novo heterochromatin formation at P53-targeted promoters [5] . Besides blocking the sensors or upstream regulators , viral proteins can also directly interfere with the apoptotic machinery . For instance , many viruses ( including adenovirus , Epstein-Barr virus , Kaposi's sarcoma-associated γ-herpesvirus , mouse γ-herpesvirus , etc . ) encode functional homologs of the anti-apoptotic regulator Bcl-2 , which can directly inhibit the intrinsic apoptotic pathway . Similarly , key components of the extrinsic pathway are targeted by viruses such as Shope fibroma virus , myxoma virus , smallpox virus , etc . ( reviewed in [6] ) . Last but not least , some viruses , particularly insect baculoviruses , encode caspase inhibitors . Both P35 and IAP ( Inhibitor of Apoptosis ) were initially identified in lepidopteran baculoviruses [7] , [8] . It has been very well demonstrated that these two genes are required for the infectivity of baculoviruses in lepidopteran hosts ( reviewed in [9] ) . While much of the evidence strongly suggests that evading or delaying apoptosis is an important mechanism for viruses to succeed in establishing proliferative infection , it has also been documented that at later stage of infection , viruses induce apoptosis to assist in their dissemination ( reviewed in [10] ) . Despite the evidence from virology studies , the functional role of apoptosis in mediating insect immunity has been under debate . Since insects do not have adaptive immunity , induction of apoptosis could conceivably play an even more prominent role in antiviral defense than in mammalian and other vertebrate hosts . Although induction of apoptosis has been observed following viral infection of mosquitoes [11] , the regulatory mechanisms , i . e . the regulatory pathway and pro-apoptotic genes responsible for the induction of apoptosis following viral infection , remained obscure . This gap of knowledge has prevented mechanistic analysis to evaluate the role of apoptosis as an innate immune mechanism in dipteran insects . In the mean time , a series of studies conducted in cultured insect cells reported that apoptosis was either not observed [12] , [13] , or as is the case for the baculovirus Autographa californica multicapsid nucleopolyhedrovirus ( AcMNPV ) or Flock House virus ( FHV ) in Drosophila cells , only observed relatively late in the infection cycle ( i . e . at or after 24 hrs p . i . ) [14] , [15] . More importantly , blocking apoptosis in these infection systems seems to have little effect on the infection and proliferation of the viruses . These observations raise the question of whether apoptosis is an innate immune response that can prevent/limit the infection , or is simply one of the cellular outcomes associated with late stage viral infection . Genetic studies in Drosophila revealed that the four IAP-antagonist genes , reaper , hid , grim , and sickle ( also referred to as the RHG genes ) together play a pivotal role in mediating developmental cell death [16] ( Figure 1 ) . With the exception of Hid , whose pro-apoptotic activity can be suppressed by the MAP kinase pathway [17] , RHG genes are mainly regulated at the transcriptional level and are selectively expressed in cells destined to die during animal development . Transcriptional activation of the RHG genes is also responsible for mediating the induction of apoptosis following cytotoxic stimuli such as irradiation . Interestingly , the sequences of the RHG genes diverged very rapidly during evolution . Consequently , no RHG ortholog was identified during the initial annotation of the genome of the mosquito Anopheles gambiae . The first RHG gene in mosquitoes , michelob_x ( mx ) , was identified with an advanced bioinformatics approach and verified as a bona fide IAP-antagonist [18] . Although the sequence of mx has diverged greatly from that of reaper , to the extent that it is almost beyond recognition , its transcriptional regulation was surprisingly similar in that it is induced rapidly following UV irradiation [18] . The identification of mx allowed the verification of the potential involvement of reaper-like IAP-antagonists in mediating pro-apoptotic response following viral infection . We found that mx is rapidly induced in Aedes aegypti larvae exposed to the mosquito baculovirus CuniNPV ( Culex nigripalpus nucleopolyhedrovirus ) [19] . This rapid induction of mx was specifically observed in virus-infected larval midgut cells and followed by quick apoptotic cell death and elimination of the infected cells at about 4–6 hr p . i . . Interestingly , the rapid induction of apoptosis was only observed in the A . aegypti larvae that are refractory to CuniNPV infection . There was no rapid induction of apoptosis when larvae of a susceptible species , Culex quinquefasciatus , were exposed to the same infection [19] . While the correlation between apoptosis and resistant phenotype suggested that the induction of apoptosis may play a role in mediating the resistance to viral infection , this hypothesis could not be tested in mosquitoes due to the lack of genetic means to manipulate the pro-apoptotic response in specific cells/tissues . In order to mechanistically test the functional role of rapid induction of pro-apoptotic response as an innate immune response , we established two in vivo virus infection systems in Drosophila melanogaster . We found that , similar to what was observed in A . aegypti larvae exposed to CuniNPV through the native route of infection , injection of either DNA or RNA viruses induced rapid expression of RHG genes at 1–2 hr post infection . The induction of the RHG genes requires the function of dP53 and is mediated by a highly conserved regulatory region in the vicinity of the reaper gene . More importantly , we showed that , in live D . melanogaster larvae and adults , the rapid induction of apoptosis is an important innate immune response that is capable of limiting or blocking viral gene expression and replication .
To probe whether similar rapid induction of RHG genes and apoptosis can be observed following viral infection in the fruit fly , we tried to infect Drosophila larvae and adults with AcMNPV and FHV , respectively . AcMNPV , a lepidopteran baculovirus with a dsDNA genome of about 134 kb , infects the larvae of susceptible lepidopteran hosts [20] . AcMNPV does not complete its replication cycle in Drosophila cells , but it can enter these cells and initiate early gene expression and viral DNA replication [21] . AcMNPV budded virus ( 3×104 PFU per larva ) was introduced into the abdominal hemocoel of 3rd instar Drosophila larvae through micro-injection . Q-PCR analysis indicated that following AcMNPV injection , two RHG genes , hid and reaper , were quickly induced as early as 1 hr post injection ( p . i . ) . By 2 hr p . i . , the level of pro-apoptotic genes had returned to normal ( Figure 2A ) . This rapid induction of RHG genes likely required immediate early gene expression from the baculovirus , since UV-inactivated virus failed to induce hid expression ( Figure S1 ) . In parallel , we also introduced AcMNPV to cultured Drosophila DL-1 cells . Even at MOI ( Multiplicity of infection ) of 20 , AcMNPV failed to trigger significant induction of reaper or hid at early stages of the infection . A moderate level of induction was observed at 6 hr p . i . and significant induction was observed at 24 hr p . i . ( Figure 2B ) . Since the RNA was extracted from homogenized whole larvae , the level of gene induction revealed by Q-PCR cannot fully reflect the magnitude of change of gene expression in specific cells . To monitor the level of hid mRNA in individual cells , FISH was performed with DIG-labeled RNA probes against hid . Several tissues , including fat body , midgut , hindgut , malpighian tubule , ovary/testis , etc . were examined . We found that the induction of hid following AcMNPV infection was mainly observed in scattered fat body cells ( Figure 2C ) . At 1 hr p . i . , 16 . 09%±2 . 42% cells in the larval fat body were positive for hid . A few ( 1 . 77%±0 . 46% ) midgut cells were also found to be hid-positive in the infected larvae ( Figure 2D ) . Flock House virus ( FHV ) is a positive-sense single strand RNA virus of the family Nodaviridae ( reviewed in [22] ) . Originally isolated from grass grubs , it has been shown to replicate in plants , yeast , and a variety of insects [23] , [24] . We injected FHV into the thorax of adult flies at dosages ranging from 2×102 to 2×106 PFU/adult . Q-PCR results indicated that similar to AcMNPV , at 1–2 hr post viral injection , the expression of reaper was significantly induced in FHV-injected Drosophila adults . The level of reaper in FHV infected adults was about 1 . 6 fold higher than in the control-injected sample ( Figure 2E ) . This result demonstrated that in addition to a DNA virus , infection by an RNA virus can also induce rapid induction of RHG genes . This rapid induction of RHG genes was not observed when parallel FHV infection was performed in Drosophila DL-1 cells . Addition of FHV , at 40 PFU/cell , to cultured DL-1 cells did not induce reaper or hid expression until about 36 hr p . i . ( Figure 2F ) . This corresponded well with a previous report which showed that caspase activation and apoptosis occurred after 36 hr p . i . when a similar dosage of FHV was applied to DL-1 cells [14] . We next sought to verify whether the induction of reaper is specific to FHV-infected cells . The quick induction of apoptosis in the fat body of wild type D . melanogaster prevented us from detecting viral gene expression in those cells . To block the apoptotic process induced by viral infection , we injected FHV ( 2×103 PFU/fly ) into adult D . melanogaster that have elevated expression of Diap1 in the fat body ( Lsp-Gal4→UAS-Diap1 ) . DIAP1 is capable of blocking reaper/hid -induced apoptosis . At 1 day post injection , not all cells in the fat body were positive for FHV capsid protein . However , all ( 100% ) of those that were positive for FHV capsid protein also had high levels of reaper mRNA ( Figure 2G ) . In contrast , some cells were positive for reaper mRNA but did not have detectable levels of the FHV capsid protein; these could be cells that were recently infected and in which the capsid protein has not yet been expressed . Taken together , these observations indicated that the rapid induction of RHG genes following viral exposure is a general phenomenon that can be observed in both mosquitoes [19] and Drosophila . In addition , this response is not limited to DNA viruses . Infection by an RNA virus such as FHV can elicit a rapid induction of pro-apoptotic genes as well . The significant difference in the timing of pro-apoptotic response following AcMNPV and FHV infection in D . melanogaster larvae and adults vs . that in cultured cells indicated that the dynamics of the pro-apoptotic response , and possibly the mechanism , differs significantly . To elucidate the role of rapid induction of apoptosis in limiting viral infection in Diptera , we focused our subsequent analyses on in vivo models . The rapid induction of both reaper and hid following viral infection was reminiscent of what was observed following ionizing irradiation of embryos and larvae [25] , [26] , [27] , in which case the function of the transcriptional factor P53 is required for the rapid induction of the RHG genes [27] . To investigate the mechanism responsible for mediating virus -induced pro-apoptotic gene expression , we infected Drosophila larvae of three different genotypes with AcMNPV . The three genotypes were w1118 ( wild type ) , P53[5A-1-4] , and Df ( IRER ) . The P53 loss-of-function allele 5A-1-4 is a deletion generated via homologous recombination [28] . Flies homozygous for this mutant allele have a reduced level of stress-induced apoptosis , but are otherwise viable and have no obvious phenotype . When the expression levels of RHG genes were monitored following injection of AcMNPV infection , we found that the induction of hid and reaper was completely blocked in this P53 null mutant ( Figure 3A ) . Somewhat to our surprise , FHV -induced expression of reaper in the adult was also P53-dependent . Q-PCR results indicated that while in wild type flies , FHV infection can result in about 1 . 6 fold induction of reaper at 1–2 hr post infection , FHV injection failed to induce reaper expression in flies lacking P53 function ( Figure 3B ) . Taken together , these results indicated that P53 plays a pivotal role in mediating the rapid induction of RHG genes following viral infection . The induction of RHG genes following ionizing irradiation requires a regulatory region upstream of reaper/grim/hid , called the IRER ( Irradiation responsive enhancer region ) [25] . In flies deficient for IRER , the induction of RHG genes reaper and hid following irradiation is either completely blocked or significantly suppressed depending on the tissue and developmental stage examined . The proximal breaking point of the deletion is about 3 kb upstream of the reaper promoter . Thus this regulatory mutant specifically blocks stress-induced expression of the RHG genes without deleting any transcribed region . More importantly , other DNA damage –induced responses , such as the induction of DNA repair proteins Ku70/Ku80 , remain intact in this mutant [25] . Our data indicated that similar to what was observed in the P53 mutant strain , AcMNPV and FHV -induced RHG gene expression was blocked in Df ( IRER ) flies , indicating this regulatory region is required for induction of reaper/hid following virus infection ( Figure 3A&B ) . Our previous work showed that Aedes aegypti midgut cells infected by CuniNPV following the native ( oral ) route of infection become TUNEL-positive at 4–6 hr p . i . [19] . In wild type larvae or adults infected by AcMNPV or FHV , the rapid induction of RHG genes in fat body cells was followed by apoptosis at about 2 . 5 hr p . i . ( Figure 3C&D ) . Apoptotic cells were recognized with an antibody developed against activated ( cleaved ) caspase-3 , which labels cells containing the activated form of the initiator caspase Dronc [29] . There were little if any apoptotic cells in fat bodies from larvae or adults injected with control media or suspension buffer , respectively . However , a significant increase of apoptotic cells was observed at 2 . 5 hr after injection in fat bodies of larvae or adults injected with either AcMNPV or FHV , respectively . Corresponding with the absence of induction of the RHG genes following viral infection , the rapid induction of apoptosis in both larval and adult fat body was blocked in the P53 null mutant ( Figure 3C&D ) . The induction of apoptosis was also blocked in homozygous Df ( IRER ) flies . Since IRER is a regulatory region controlling the expression of RHG genes , this evidence indicates that P53-induced expression of the RHG genes is responsible for the rapid induction of apoptosis following viral infection . Significant induction of RHG genes was not observed in P53−/− flies even at later time points after the infection . At 4–7 days p . i . , essentially all cells in the fat body of the P53−/− flies were filled with FHV ( containing capsid protein immune-reactivity ) ( Figure 3E ) . These cells , unlike those in wild type flies , had lost the integrity of cell membranes and became permeable to propidium iodide ( PI ) . Similar loss of membrane integrity , a typical feature of necrotic cells , was also observed in CuniNPV -infected susceptible mosquitoes at 48–72 hr p . i . [19] . To investigate the functional significance of the rapid induction of apoptosis as an innate immune response against viral infection , we first examined the expression of two AcMNPV immediate early genes , ie0 and ie1 [30] at 6 hr post infection in wild type , P53 deficient , and IRER deficient Drosophila strains . Expression of ie0 was not detectable in wild type larvae following virus injection . However , it was reliably detectable in either P53 null or Df ( IRER ) D . melanogaster larvae at 6 hr post injection ( Figure 4A ) . The level of expression of ie1 was very low , but nonetheless detectable , in wild type larvae at 6 hr post infection . Its level of expression was dramatically higher in P53 mutant or Df ( IRER ) larvae that lack the rapid induction of apoptosis . This indicated that the rapid induction of apoptosis mediated by P53 and IRER plays an important role in inhibiting viral gene expression . To test whether the rapid induction of apoptosis can function to block or limit viral proliferation , we monitored the genome levels of FHV at 24 hr post injection of 200 PFU per D . melanogaster adult . In this assay , a group of 5 flies for each genotype was homogenized and relative abundance of the FHV RNA genome was assayed with Q-PCR primers targeting both RNA1 and RNA2 of FHV ( Figure 4B ) . We found that compared to wild type D . melanogaster adults , the relative levels of FHV genome were significantly higher in the P53−/− and Df ( IRER ) mutants . To verify whether the difference in FHV proliferation was indeed due to the lack ( or delay ) of cell death , we next injected D . melanogaster adults of the genotype Lsp-Gal4/UAS-Dronc_RNAi with 200 PFU per adult of FHV . Dronc is an upstream caspase that plays a pivotal role in mediating cell death induced by RHG genes . We found that knocking down dronc specifically in the fat body ( where Lsp-Gal4 is expressed ) allowed the proliferation of FHV viral genomes to a level comparable to that observed in Df ( IRER ) flies at 24 hr p . i . ( Figure 4C ) . These results indicate that rapid induction of apoptosis is responsible for limiting viral proliferation at early stages of the infection . To test whether the lack of rapid induction of apoptosis could lead to establishment and proliferation of FHV , we monitored the amplification of viral genomes following FHV injection in individual wild type or mutant D . melanogaster adults . At 4 days following FHV injection ( 20 PFU per adult ) , the levels of viral genome RNA in most wild type D . melanogaster adults did not increase at all ( Figure 5A ) , indicating that there was no successful proliferation of the virus . In contrast , in both P53 mutant and Df ( IRER ) flies , the levels of FHV RNA increased dramatically , indicating successful proliferation of the virus . When 200 PFU per animal of FHV was injected , the levels of viral RNA were unchanged in most wild type flies at 4 days post injection . In contrast , the levels of viral RNA in P53−/− and Df ( IRER ) flies indicated that significant proliferation had occurred in those mutants that lack rapid induction of apoptosis ( Figure 5B ) . The successful proliferation of FHV in D . melanogaster adults lacking the rapid induction of apoptosis was also verified by visualizing FHV coat protein using an antiserum raised against purified FHV particles ( Figure 5C ) . At 4 days following injection of 200 PFU per animal , no cells were detected positive for FHV in wild type flies . In contrast , almost all cells in the fat body of the P53 mutant ( or Df ( IRER ) ) flies were positive for FHV . Furthermore , as early as 4 days p . i . , cells containing FHV capsid protein were observed in the salivary glands of P53−/− flies ( Figure 5D ) , indicating systemic infection had been established . These results indicated that the rapid induction of apoptosis , observed in wild type Drosophila melanogaster , is capable of blocking the infection when the infecting dose is less than 200 PFU per animal . Conversely , lack of rapid induction of apoptosis , as was observed for the P53−/− and Df ( IRER ) , leads to significantly increased susceptibility . Our previous work found that following exposure to the mosquito baculovirus CuniNPV , the mosquito reaper ortholog mx was rapidly induced ( within 2 hr p . i . ) to mediate apoptosis of midgut cells in A . aegypti larvae [19] . In this study , we asked whether similar rapid induction of pro-apoptotic response can be observed in adult female mosquitoes following exposure to a human pathogen . Blood meals with or without DEN-2 ( dengue virus serotype 2 ) JAM 1409 were fed to adult Aedes aegypti mosquito strains that are either refractory ( MOYO-R ) or a susceptible ( MOYO-S ) to DEN-2 . When the expression level of mx was monitored via Q-PCR , we found that it was significantly induced in the MOYO-R strain following DEN-2 exposure when compared with the control-fed females ( Figure 6 ) . This induction of mx in the refractory strain was rapid , since at 3 hr post blood meal ( p . b . m . ) the level of mx was about 2 . 5 fold higher in the virus-fed compared to the control-fed mosquitoes . The difference of mx levels between virus-fed and control-fed MOYO-R Aedes aegypti strain receded to lower levels at 18 hr p . b . m . . In contrast , there was no difference in the expression of mx between control-fed or DEN-2 fed females of the MOYO-S strain . This indicates that the susceptible MOYO-S strain lacks the rapid induction of mx .
So far , the rapid induction of RHG genes and apoptosis following viral infection have only been observed in vivo . Dengue virus infection of the mosquito cell line C6/36 does not induce apoptosis and there appears to be no significant induction of pro-apoptotic genes [31] , [32] . Studies with lepidopteran animal models have clearly shown that the anti-apoptotic activity of the AcMNPV P35 gene is required for its infectivity ( reviewed in [9] ) . However , when AcMNPV or FHV were applied to the Drosophila DL-1 cells , apoptosis was only observed 24 or 36 hr post infection , respectively , and the effect of blocking apoptosis only had minor effects on the proliferation of the viruses [14] , [15] . Correspondingly , we observed that there was no significant induction of RHG genes before 24 hr p . i . when either virus was applied to DL-1 cells ( Figure 2 ) . It is unclear as to why cell lines , that have been tested so far , lack the rapid pro-apoptotic response observed in both live mosquitoes and fruit flies . One possibility is that only certain types of cells can launch the rapid pro-apoptotic response , and such cell types are not represented in cultured cell lines . Another possibility is that cultured cell lines were unknowingly selected to have reduced sensitivity to stress-induced cell death . The regulatory region required for mediating viral infection induced pro-apoptotic genes , i . e . IRER , serves as a locus control region mediating the induction of RHG genes in response to a variety of stresses , such as x-ray , UV , oncogenic stresses , etc . In addition , the accessibility of IRER is controlled by epigenetic regulation . When IRER is epigenetically blocked , i . e . in heterochromatin-like conformation , the RHG genes are no longer responsive to stresses such as DNA damage [25] . Our analysis of several Drosophila cell lines ( S2 , Kc167 , etc . ) indicated that the IRER region in these cell lines is enriched for heterochromatic modifications and resistant to DNase I treatment [33] ( and unpublished observations ) . It is possible that cells with reduced sensitivity to stress-induced cell death , either through genetic mutation or epigenetic silencing of IRER , are inadvertently selected during in vitro cell culture processes . As a result , the ability to launch the rapid induction of RHG genes following viral infection may have been lost in long term cultured cells . Our data indicate that the rapid induction of apoptosis is capable of blocking infection at its initiation stage when the animals are exposed to relatively small amounts of virus . This response may contribute to the “midgut infection barrier” that has long been observed for arbovirus transmission through insect vectors [34] . Apoptosis of midgut cells following viral exposure has been observed before , when a refractory strain of C . pipiens was orally infected with West Nile virus [11] . Similarly , rapid induction of mx was observed in refractory A . aegypti ( MOYO-R ) females orally infected with DEN-2 ( Figure 6 ) , while there was a conspicuous lack of rapid induction of mx in the susceptible strain ( MOYO-S ) . Our data obtained with FHV –infected P53−/− and Df ( IRER ) animals indicated that the lack of rapid induction of apoptosis following viral infection led to dramatically increased susceptibility to established systemic infection . The rapid induction of apoptosis effectively denies the opportunity for viral gene expression ( Figure 4 & 7 ) . This has been previously demonstrated for CuniNPV infection through a native route of infection [19] , where we showed that viral gene expression was only detected when apoptosis was delayed with caspase inhibitors . In the current study , we showed that the lack of rapid induction of apoptosis in P53−/− and Df ( IRER ) animals allowed viral gene expression and proliferation . In addition , significant viral proliferation was achieved when the level of Dronc in fat body cells was knocked down by tissue-specific RNAi . All of these data indicate that rapid elimination of infected cells is responsible for blocking the infection at the initiation stage , before significant expression of viral genes could take control of the cellular system . Rapid induction of apoptosis as an innate immune response against viral infection is not restricted to insects . For instance , rapid induction of apoptosis was observed at 8 hr following infection of human embryonic stem cells ( hESCs ) with recombinant AAV [35] . hESCs are extremely sensitive to various stresses , and the rapid induction of apoptosis in hESC cells following rAAV infection also requires P53 . Rapid induction of apoptosis was also observed following the infection of primary dendritic cells by the intracellular pathogen Legionella pneumophila , which induces apoptosis within the first hour of infection [36] . Similarly , influenza A viruses mutated for NS1 induce rapid apoptosis in primary macrophages [37] . Similar to what we discovered with Drosophila , P53−/− mice are hypersensitive to influenza A infection [38] . However , the anti-viral effect of P53 in mice may include induction of pro-inflammatory genes in addition to its pro-apoptosis function . Study of innate immunity in C . elegans also revealed that P53 has an ancient role as an immune/stress sensor [39] . In this study , we found that although FHV virus can proliferate in Df ( IRER ) that lacks rapid induction of apoptosis , the titer of FHV was consistently lower than that observed for P53 null mutant animals . The observed difference between P53−/− and Df ( IRER ) flies could be due to the fact that other pro-apoptotic genes are also induced/activated by P53 . Alternatively , it could be due to another anti-viral activity of P53 besides its role in the rapid induction of apoptosis . Both may be true since a whole transcriptome microarray analysis of dengue virus -induced changes in gene expression revealed that many P53 target genes were activated in the refractory strain ( MOYO-R ) but not in the susceptible strain ( MOYO-S ) [40] . The same study also revealed that caspase1 ( AAEL012143 ) and caspase3 ( AAEL005963 ) were significantly up-regulated in the MOYO-R strain , but not in the MOYO-S strain [40] . Caspase1 ( also known as CASPS7 ) has been shown to be involved in apoptosis in an A . aegypti cell line [41] . Together with the finding reported here ( Figure 6 ) , it seems that P53 mediates the induction of multiple pro-apoptotic genes following viral infection in MOYO-R , but not in MOYO-S . Our demonstration that either blocking the induction of the RHG genes in Df ( IRER ) or knocking down the regulatory caspase DRONC increases susceptibility of D . melanogaster to FHV infection ( Figure 4 & 5 ) indicates that the major mechanism of P53-mediated anti-viral activity is through its role in the rapid induction of RHG genes and apoptosis . It is not clear how P53 is activated in virus -infected cells . In Drosophila , three pathways have been well characterized for their role in mediating immune response , i . e . the Toll pathway , the IMD pathway , and the Jak-STAT pathway ( reviewed in [42] ) . The Toll pathway is mainly responsive to fungi and Gram-positive bacteria while the IMD pathway is activated by Gram-negative bacteria [43] . Recent studies indicate that besides anti-fungal and anti-bacterial functions , the Toll pathway is also involved in antiviral response [44] , [45] . The JaK-STAT pathway has been shown to be activated by Drosophila C virus . Loss of function of JaK led to increased viral load and decreased survival rate after viral infection [46] . However , we found that vir-1 , a target gene of JaK-STAT , was not induced by AcMNPV or FHV infection when reaper/hid were significantly induced , which suggested that JaK-STAT pathway is not likely involved in activating P53 ( Figure S2 ) . The fact that both AcMNPV- and FHV- induced rapid transcriptional activation of RHG genes required P53 and IRER suggests that a common mechanism may be responsible . The two viruses are quite different , i . e . dsDNA virus vs . ssRNA virus . AcMNPV cannot fully replicate in Drosophila whereas FHV can replicate in a variety of insects including Drosophila . The fact that these two viruses , and very likely other viruses such as DEN-2 and CuniNPV , induce rapid induction through the same transcription factor and regulatory region strongly suggests that a more general mechanism is involved . Revealing this mechanism should shed great light on our understanding of virus-vector interactions .
Drosophila white 1118 ( w1118 ) strain was used as a standard wild-type strain . p53 deficient line p53[5A-1-4] which has a 3 . 3 k deletion in p53 gene [28] was obtained from the Bloomington Stock Center ( Indiana University , Bloomington , IN , USA ) . The IRER deficient strain B11 was previously described [25] . All strains were maintained on a standard cornmeal medium at room temperature . Baculovirus AcMNPV was produced as previously described [47] . Generally speaking , Spodoptera frugiperda cell line sf9 was cultured with sf900 medium at 28°C . Infectious Autographa californica nucleopolyhedrovirus ( AcMNPV ) was obtained by transfection of sf9 cells with bacmid DNA ( AcWTPG ) containing the AcMNPV genome . Virus was titered in sf9 cells by standard end point dilution assay . For AcMNPV infection , Drosophila 3rd instar larvae were injected with budded AcMNPV at a dosage of 3×104 PFU/larva in the dorsal-posterior area . Injected larvae were kept in sf900 medium with or without virus for indicated times before subjecting to RNA extraction and Q-PCR analysis . FHV was propagated and purified following established protocols [48] . For FHV infection , adult flies at 4–6 days of age were used . FHV was diluted with sf900 medium and the infection was achieved by injection of viral suspension into the thorax of adult flies . The injected flies were then cultured with standard fly food at room temperature . The MOYO-R and MOYO-S strains were used and are refractory ( ∼20% susceptible ) vs . susceptible ( ∼54% susceptible ) to oral infection with DENV , respectively . The origins of these strains , our standard rearing conditions , DENV susceptibility status , cell culture procedures and mosquito infections are described elsewhere [49] . DENV-2 strain JAM 1409 was cultured using Aedes albopictus C6/36 cells wherein a 0 . 1 multiplicity of infection ( MOI ) was used for infecting the mosquito cells . Females were provided an artificial infectious blood meal freshly prepared using defibrinated sheep blood ( Colorado Serum Co . , CS1122 ) mixed with an equal volume of the cell culture suspension . Controls were similar but were prepared with uninfected cell culture suspensions . At 3 h and 18 hr post-infection , total RNA was extracted from 20 females per sample using a Qiagen RNAeasy Kit following manufacturer's instructions . RNA was quantified using a Nanodrop spectrophotometer and RNA quality was assessed using a Bioanalyzer . Three biological replicates were obtained . Larval total RNA was extracted with RNeasy Mini Kit ( QIAGEN , Valencia , CA , USA ) according to the protocol provided by the manufacturer . Adult RNA was extracted with TRIZol Reagent ( Invitrogen , Grand Island , NY , USA ) following manufacturer's manual and purified with RNeasy Mini Spin Column ( QIAGEN ) . RNA samples were treated with DNase I to remove genomic DNA . cDNA was prepared by reverse transcription of total RNA with a High-Capacity cDNA Archive Kit ( Applied Biosystems , Foster City , CA , USA ) . Q-PCR was performed with an ABI 7500 Fast thermocycler ( Applied Biosystems ) following protocols provided by the manufacturer . Triplicates were measured for each gene/sample combination . The oligo sequences of the main target genes are as follows: reaper: 5′-ACGGGGAAAACCAATAGTCC-3′ and 5′-TGGCTCTGTGTCCTTGACTG-3′; hid: 5′-CTAAAACGCTTGGCGAACTT-3′ and 5′-CCCAAAAATCGCATTGATCT-3′; rp49: 5′-GCTAAGCTGTCGCACAAATG-3′ and 5′-GTTCGATCCGTAACCGATGT-3′; AcMNPV ie0: 5′- CGAGACGCGTTGAAGCTAAT-3′ and 5′- CGCAACATTCTTTTGGCTTT-3′; AcMNPV ie1: 5′- GGCAGCTTCAAACTTTTTGG-3′ and 5′- TTCACACCAGCAGAATGCTC-3′; FHV RNA1: 5′- CCAGATCACCCGAACTGAAT-3′ and 5′-AGGCTGTCAAGCGGATAGAA-3′; FHV RNA2: 5′-CGTCACAACAACCCAAACAG-3′ and 5′-GGTCGGTGTTGAAGTCAGGT-3′ . Q-PCR results were normalized to rp49 or GAPDH for Drosophila and mosquito samples , respectively , before further calculation . The amount of FHV genome was estimated using a pre-generated standard curve and regression equation . To get the standard curve of the viral dosage/Ct value , a serial dilution of known dosage of FHV was mixed with wild type adult male ( one fly per dilution ) followed by homogenization . RNA extraction and Q-PCR were performed as described above to get the Ct value of viral RNA1 or RNA2 . Standard curve and regression equation were generated using Microsoft Office Excel ( version 2007 ) . Probes were synthesized using digoxin ( DIG ) -RNA Labeling Mix ( Roche , Madison , WI , USA ) . Drosophila 3rd instar larval cuticles were partially removed to expose inside tissue in 4% paraformaldehyde . After prefixing with 4% paraformaldehyde in PBT_DEPC ( 0 . 3% Triton in PBS made with DEPC pretreated double-distilled water ) for 30 min , the tissue was incubated for 7 min with 50 mg/ml protease K in PBT_DEPC , and reaction was stopped by washing with 4% paraformaldehyde . Samples were incubated with probes diluted in hybridization buffer ( 50% formamide , 25% 2×SSC , 20 mg/ml yeast tRNA , 100 mg/ml ssRNA , 50 mg/ml heparin , and 0 . 1% Tween-20 ) . Hybridization was performed overnight at 60°C . Larvae were incubated with horseradish peroxidase ( HRP ) -conjugated anti-DIG ( Roche ) antibody after hybridization , followed by signal amplification using the Tyramid Signal Amplification Kit ( PerkinElmer , Waltham , MA , USA ) . Rabbit monoclonal antibody to cleaved caspase-3 was purchased from Cell Signaling ( Danvers , MA , USA ) . The antibody was used at a dilution of 1∶200 . AlexaFluor 488 labeled goat-anti-rabbit antibody was purchased from Molecular Probes and was used at a dilution of 1∶1000 . Propidium iodide was purchased from Sigma ( St . Louis , MO , USA ) . To detect cell necrosis , 100 nL of PI ( 1 mg/mL ) was injected into the thoraces of adult flies . Injected flies were cultured with standard fly food at room temperature for 20 min before subjecting to fixation and immunostaining . Fly fat bodies were dissected in PBS containing 4% paraformaldehyde and fixed for 20 min at room temperature . After being washed with PBS containing 0 . 1% Triton X-100 ( PBST ) , the samples were blocked with PBST containing 5% normal goat serum for 30 min . Samples were then incubated overnight with anti-cleaved caspase-3 antibody ( 1∶200 dilution ) at 4°C . Labeling with secondary antibody was done at 25°C for 2 hr . Slides were mounted with Vectorshield Mounting Medium ( Vector Laboratories , Burlingame , CA , USA ) . Pictures were taken with a Leica upright fluorescent microscope ( Leica , Bannockburn , IL , USA ) using OpenLab software ( Improvision , Coventry , UK ) . All quantitative data are shown as mean ± standard deviation unless noted otherwise . Student's t-test was used to evaluate statistical significance . | Arthropod-borne pathogens account for millions of deaths each year . Understanding the genetic mechanisms controlling arthropod susceptibility to pathogens has profound implications for developing novel strategies for controlling insect-transmitted infectious diseases . Although it was postulated that apoptosis ( a genetically controlled form of cellular suicide ) may play a very important role in insect innate immunity against viral infection , direct evidence has been lacking due to the lack of knowledge on the regulatory pathways responsible for the induction of apoptosis following viral infection . In this study , we found that there is a rapid induction of pro-apoptotic genes within 1–3 hours of exposure to virus . This rapid pro-apoptotic response was only observed in live animals but not in cultured cells . Genetic analysis indicated that animals lacking this rapid pro-apoptotic response were hypersensitive to viral infection . Thus our work provides unequivocal evidence indicating that rapid induction of apoptosis plays a very important role in mediating insect resistance to viral infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cell",
"death",
"immunology",
"microbiology",
"animal",
"models",
"mechanisms",
"of",
"resistance",
"and",
"susceptibility",
"model",
"organisms",
"drosophila",
"melanogaster",
"animal",
"models",
"of",
"infection",
"biology",
"cell",
"biology",
"immunity",
"virology",... | 2013 | P53-Mediated Rapid Induction of Apoptosis Conveys Resistance to Viral Infection in Drosophila melanogaster |
Kaposi’s Sarcoma-associated Herpesvirus ( KSHV ) establishes stable latent infection in B-lymphocytes and pleural effusion lymphomas ( PELs ) . During latency , the viral genome persists as an epigenetically constrained episome with restricted gene expression programs . To identify epigenetic regulators of KSHV latency , we screened a focused small molecule library containing known inhibitors of epigenetic factors . We identified JQ1 , a Bromodomain and Extended Terminal ( BET ) protein inhibitor , as a potent activator of KSHV lytic reactivation from B-cells carrying episomal KSHV . We validated that JQ1 and other BET inhibitors efficiently stimulated reactivation of KSHV from latently infected PEL cells . We found that BET proteins BRD2 and BRD4 localize to several regions of the viral genome , including the LANA binding sites within the terminal repeats ( TR ) , as well as at CTCF-cohesin sites in the latent and lytic control regions . JQ1 did not disrupt the interaction of BRD4 or BRD2 with LANA , but did reduce the binding of LANA with KSHV TR . We have previously demonstrated a cohesin-dependent DNA-loop interaction between the latent and lytic control regions that restrict expression of ORF50/RTA and ORF45 immediate early gene transcripts . JQ1 reduced binding of cohesin subunit Rad21 with the CTCF binding sites in the latency and lytic control regions . JQ1 also reduced DNA-loop interaction between latent and lytic control regions . These findings implicate BET proteins BRD2 and BRD4 in the maintenance of KSHV chromatin architecture during latency and reveal BET inhibitors as potent activators of KSHV reactivation from latency .
Kaposi’s Sarcoma-associated Herpesvirus ( KSHV ) is a human gammaherpesvirus responsible for all forms of Kaposi’s Sarcoma ( KS ) and strongly associated with pleural effusion lymphomas ( PELs ) and Castleman’s Disease[1] . KSHV can establish long-term latent infection in B-lymphocytes where it persists as a stable , chromatin-associated circular minichromosome , commonly referred to as an episome [2 , 3] . During latent infection , the viral genome expresses only a few viral genes required for maintaining the latent state and host-cell survival [4 , 5] . The major latency transcripts include the multi-cistronic RNAs encoding LANA ( ORF73 ) , vCyclin ( ORF72 ) , vFLIP ( ORF71 ) , K1 , and 21 miRNAs . The major immediate early genes are also regulated as a cluster of RNAs that can be initiated during the early stage of the reactivation process . These include the immediate early transcriptional activator RTA ( ORF50 ) , KbZip ( ORF51 ) , and a series of transcripts that are made in the opposite orientation that include ORF45-49 . Lytic transcription is repressed during latency , while latency transcription occurs efficiently . How these regions are differentially regulated and how they communicate with each other remains an area of active interest . KSHV latency is maintained by several epigenetic regulatory mechanisms . Lytic cycle regulatory regions , especially the immediate early promoter regions controlling RTA transcription are regulated by bivalent histone modifications that include both euchromatic H3K4me3 and repressive H3K27me3 at the same regulatory locus [6 , 7] . Inhibitors of polycomb-associated H3K27me3 methyltransferase EZH2 are sufficient to induce lytic cycle replication [8–10] . In KSHV positive B-cell pleural effusion lymphomas , KSHV latency can be reactivated by other epigenetic pathways , including histone deacetylase ( HDAC ) inhibitors in combination with phorbol esters [11] . Lytic reactivation may also be induced by other cellular stress pathways , including hypoxia [12] , reactive oxygen species ( ROS ) [13] , cytokine stimulation[14] , and terminal differentiation[15] . During latent infection in PEL cells , the KSHV genome is also regulated by higher-order epigenetic regulatory mechanisms [16] . We have shown that the chromatin organizing factor CTCF colocalizes with cohesins at several locations in the KSHV genome , including the latency control region [17] . Subsequent studies revealed that KSHV latency control region formed a DNA-loop interaction with the lytic control region , mediated in part by the CTCF-cohesin complex [18] . Chromosome conformation capture ( 3C ) revealed that the control regions for the lytic and latent cycle transcripts are in close proximity during latency , and that this is disrupted during the reactivation process . Depletion of cohesin subunits , including RAD21 , SMC1 , or SMC3 led to the reactivation of KSHV [19] . Depletion of CTCF , as well as Rad21 , were also found to be restriction factors for KSHV lytic reactivation , especially when combined with HDAC inhibitor sodium butyrate [20] . CTCF is a sequence-specific DNA binding protein that has multiple functions in gene regulation , including the formation of chromatin boundaries and DNA-loop interactions [21] . Cohesins form a ring-like structure that mediates DNA-DNA interactions important for sister chromatid cohesion , homologous recombinational DNA repair , and promoter-enhancer communication for transcriptional regulation [22] . How other factors may regulate the formation and dissolution of these higher-order DNA structures , and how these impact gene expression patterns is not completely known . The KSHV latent episome is maintained largely by the viral-encoded protein LANA [23 , 24] . LANA is a sequence-specific DNA binding protein that interacts specifically with GC-rich elements in the terminal repeats ( TR ) of KSHV [25] . LANA binding to the TR is necessary for tethering the viral episome to metaphase chromosomes , and can also function as an efficient origin of DNA replication [24 , 26 , 27] . LANA can also regulate transcription of viral and host genes , and interact with other regions of the viral and cellular chromosome , but with lower affinity than at the viral TR [28 , 29] . LANA interacts with many host proteins . Among these are the Bromodomain 2 ( BRD2 ) and BRD4 members of the BET family [30–33] . Bromodomains are known to interact with acetylated lysines , most typically found on histone tails in euchromatic regions [34 , 35] . BRD2 and 4 have been implicated in the regulation of several viruses , including tethering and transcriptional regulation of HPV through the E2 protein , a viral orthologue of KSHV LANA [36–38] . The precise function of BRD2 or BRD4 in mediating LANA function is not clear , but several lines of evidence suggest that these factors facilitate LANA-dependent episome maintenance and transcription regulation . Small molecule inhibitors of BET proteins , especially BRD2 and BRD4 , have been highly effective at disrupting their biochemical function in binding acetylated lysines on histone tails [39 , 40] . A prototype BET inhibitor , JQ1 , has been shown to effectively inhibit transcription mediated by BRD2 and BRD4 dependent super-enhancers , including those regulating the cMyc oncogene in several cancer models [40 , 41] . JQ1 has also been shown to trigger the reactivation of latent HIV [42–46] . The mechanism of HIV reactivation by JQ1 is thought to be through the redistribution of BRD2 and BRD4 to promote RNA polymerase elongation on HIV genomes [43 , 45] . The effects of BET inhibitors on KSHV latency is not as well understood . One study found that BET inhibitors synergize with lenalidomide to selectively kill KSHV positive PEL cells [47] . Here , we screened a library of small molecule epigenetic modulators and found BET inhibitors to be among the most potent for activation of KSHV lytic reactivation , and further investigate the mechanism of their action .
To identify epigenetic regulators of KSHV latent to lytic switch , we developed a simple reporter cell line , BJAB-BAC16 , consisting of BJAB cells stably infected with KSHV Bac16 that carry a constitutively expressed GFP gene . BJAB-BAC16 cells were first tested for their ability to respond to known lytic cycle reactivating reagents consisting of phorbol ester TPA combined with histone deacetylase inhibitor sodium butyrate ( NaB ) ( Fig 1 ) . Addition of TPA with NaB led to a robust stimulation of GFP signal from BJAB-BAC16 cells using FACS ( Fig 1A ) , as well as by high-content imaging using an Operetta ( Fig 1C ) . This indicated that GFP signal could be used as a surrogate for lytic cycle gene activation . Both assays were miniaturized for 384-well high-throughput screening with 25 , 000 cells per well and demonstrated robust statistical properties based on Z-factor >0 . 8 ( Fig 1B ) . We then proceeded to screen a focus library of 24 compounds with known epigenetic targets . Each compound was arrayed for a 10-point dose response curve and an EC50 value was calculated for stimulation of GFP signal ( Fig 1D ) . We found that several compounds stimulated KSHV lytic cycle with EC50 < 3 μM , while only one compound , the BET inhibitor JQ1 , activated GFP signal at EC50 <1 μM in both FACS and Operetta assays . Using the high-content analysis of the Operetta , JQ1 was found to have an EC50 of 0 . 58 μM ( Fig 1E ) . The effect of JQ1 on KSHV lytic cycle gene expression was validated by RT-qPCR for KSHV PAN , ORF50 , and LANA ( Fig 1F ) . We next tested JQ1 for its ability to activate KSHV lytic cycle in various KSHV-positive cell lines . BCBL1 , BC-1 , JSC-1 , or SLK-BAC16 were treated with 4 μM JQ1 daily for 3 days ( 72 hrs ) and then assayed by RT-qPCR for transcription of PAN , ORF50 , and LANA ( Fig 2A–2D ) . We found that JQ1 efficiently activated viral lytic genes for PAN and ORF50 in all PEL cell lines ( Fig 2A–2C ) . JQ1 efficiently stimulated PAN in the non-lymphoid cell line SLK-BAC16 , but had less significant activation on ORF50 ( Fig 2D ) . The effects of JQ1 on LANA transcripts were also cell-dependent , with an ~ 4 fold increase in BCLB1 , a ~2 fold decrease in BC-1 , and no change in JSC-1 or SLK-BAC16 ( Fig 2A–2D , lower panels ) . We next tested whether JQ1 , and another BET inhibitor I-BET151 , could induce KSHV DNA replication ( Fig 2E ) . Viral DNA replication was monitored by qPCR by comparing viral relative to cellular DNA . We found that JQ1 , as well as I-BET151 , efficiently induced DNA replication in BCBL1 , BC-1 , JSC-1 , and SLK-BAC16 cells . JQ1 and I-BET151 induced ~10 fold increase in viral DNA copy number , while positive control sodium butyrate ( NaB ) induced only 3–5 fold in PEL cells ( Fig 2E ) . To further analyze the status of KSHV genomes after treatment with JQ1 , we analyzed BCBL1 cells treated with DMSO or JQ1 by pulse-field gel electrophoresis ( PFGE ) followed by Southern blot analysis . We observed a large amplification of viral linear and sub-linear genomes ( indicative of incomplete genome replication ) , confirming that lytic cycle DNA replication was activated by JQ1 ( Fig 2F ) . JQ1 is known to act rapidly on BRD2 and BRD4 binding to acetylated histones [41] , but its kinetic effects on transcription can be complex . We determined the time course of KSHV transcripts after JQ1 treatment . As expected , cMyc transcription was reduced within 6 hrs after JQ1 treatment ( Fig 3A ) . Interestingly , KSHV transcripts for lytic cycle ( ORF50 and PAN ) were also reduced at 6 hrs , while LANA transcription was modestly elevated . However , by 72 hrs , KSHV lytic transcripts for ORF50 and PAN were 2 and 15 fold relative to DMSO control . Similar results were observed with another pan-BET inhibitor I-BET151 ( Fig 3B ) . On the other hand , BIC-1 , which is more selective for BRD2 , had inhibited cMyc transcription by 24 hrs , but did not activate KSHV lytic gene transcription ( Fig 3C ) . Similar observations were made with another PEL cell line BC3 ( S1 Fig ) . These results indicate that structurally different pan-BET inhibitors can stimulate KSHV lytic transcription , but also raise the possibility that BET inhibitors may contribute indirectly to KSHV reactivation through suppression of myc or other cellular targets . To determine whether BRD2 or BRD4 were the targets of JQ1-mediated reactivation of KSHV , we transduced BCBL1 cells with lentivirus expressing shBRD2 , shBRD4 , or shControl ( shCtrl ) ( Fig 4 ) . shBRD4 reduced BRD4 efficiently ( Fig 4A ) , and produced a ~2 . 5 fold increase in PAN and 2 fold increase in ORF50 ( Fig 4B ) . Similarly , shBRD2 partially reduced BRD2 expression ( Fig 4C ) , and led to a ~6 fold increase in PAN and ~2 . 5 fold increase in ORF50 ( Fig 4D ) . Both shBRD4 and shBRD2 produced a modest ( ~2 fold ) increase in linear and sublinear KSHV genomes in PFGE analyses ( Fig 4E ) . These results indicate that depletion of either BRD4 or BRD2 can partially induce KSHV lytic cycle transcription , and weakly induce lytic DNA replication . BRD2 and BRD4 are known to interact directly with LANA protein , but it is not known how they interact with the viral genome . We used Chromatin-Immunoprecipitation ( ChIP ) assay to measure the relative association of BRD2 and BRD4 with the KSHV genome at the LANA-binding site in the TR , and CTCF binding sites in latency control region and the lytic control region ( Fig 5A ) . We found that BRD2 and BRD4 associated with all three viral genome positions , showing the highest enrichment at the LANA binding sites at the TR . BRD2 and BRD4 were found enriched similarly at the CTCF binding sites within latency control region , while BRD4 was selectively enriched at the lytic control region ( Fig 5B ) . Neither BRD2 nor BRD4 bound to a region within the ORF37 gene ( primer i ) , indicating that the enrichment at control regulatory regions is selective . To determine if JQ1 affected BRD4 or BRD2 interaction with LANA , we assayed the effects of JQ1 on the coimmunoprecipitation ( coIP ) of LANA with BRD4 ( Fig 5C , top panels ) or with BRD2 ( Fig 5C , lower panels ) . We found that addition of JQ1 did not disrupt , and had a modest stimulatory effect , on the interaction of LANA with BRD4 and BRD2 ( Fig 5C ) . We next asked whether JQ1 had any effect on the binding of BRD4 or BRD2 proteins to KSHV genome by ChIP assay ( Fig 5D and 5E ) . We found that JQ1 treatment reduced BRD2 and BRD4 interactions at the TR regions by ~50% ( Fig 5D and 5E ) . BRD2 interactions at the latency control region were also reduced by ~50% by JQ1 ( Fig 5D ) , while BRD4 interactions at the latency and lytic control regions were reduced by only ~20% after JQ1 treatment ( Fig 5E ) . These findings indicate that BRD2 and BRD4 can interact with viral regulatory regions , and that JQ1 can partially disrupt these interactions . To better understand the mechanism through which JQ1 activates KSHV latent to lytic switch in BCBL1 cells , we performed ChIP assays for several key factors known to regulate this process , including CTCF , RAD21 , RNAPII , histone H3K9ac , and LANA ( Fig 6 ) . JQ1 had a small inhibitory effect ( ~20% reduction ) on CTCF binding at the latency control region ( primers e-g ) , but no detectable changes in the lytic control region ( primers a-d ) or the TR ( primer h ) ( Fig 6B ) . On the other hand , JQ1 reduced cohesin subunit Rad21 by 50% at the latency control region ( primers e and f ) , as well as at the lytic control region ( primer d ) ( Fig 6C ) . RNA polymerase II ( RNAPII ) was also reduced to 50% binding after JQ1 treatment at the latency control region ( primer f and g ) , but slightly increased at the lytic control region ( primers a-d ) . Similarly , histone H3 acetylated on K9 ( H3Ac9 ) was reduced by 60% at the latency control region ( primers f and g ) , and to a lesser extent that the TR , but did not change at the lytic control regions ( Fig 6E ) . We also observed that LANA binding was reduced by ~50% at the TR , and by >80% at the latency control region and lytic control region after JQ1 treatment ( Fig 6F ) . These findings indicate that JQ1 treatment alters the interaction with the latent KSHV genome for several key regulatory factors , including Rad21 , RNAPII , and LANA . These findings also suggest that JQ1 leads to the selective loss of histone acetylation at the TR and latency control region , but not at the transcriptionally active lytic control region . DNA-loop formation has been shown to occur between the latent and lytic control regions of KSHV in PEL cells [18] , and disruption of this loop by depletion of Rad21 led to a reactivation from latency [19] . To determine whether JQ1 had any effect on KSHV DNA loop interactions , we performed chromatin conformation capture ( 3C ) assays using an anchor primer at the KSHV latency control region ( Fig 7A and 7B ) . As expected , we observed a selective interaction between the latent control region and the lytic control region ( primer 69163 ) ( Fig 7B ) . Treatment with JQ1 reduced this 3C interaction by 50% , as well as other weaker 3C interactions at positions downstream ( 56293 and 58589 ) and upstream ( 72974 and 77155 ) of the lytic control region . These findings suggest that JQ1 treatment alters the DNA conformation associated with stable episomal latency of KSHV . To determine whether the effect of JQ1 chromatin factor binding was an indirect consequence of viral DNA replication , we tested whether an inhibitor of viral lytic DNA replication prevented the JQ1-induced loss of RAD21 or LANA binding to KSHV genome ( S2 Fig ) . Phosphono-acetic acid ( PAA ) is a potent inhibitor of KSHV lytic replication [48 , 49] . While PAA inhibited KSHV genome replication by ~5 fold ( S2A Fig ) , it did not reverse the effects of JQ1 binding on RAD21 ( S2B Fig ) and further stimulated the loss of LANA binding to TR ( S2C Fig ) . We also show that induction of viral DNA replication by NaB treatment for 72 hrs , led to the loss of BRD2 or BRD4 ChIP with KSHV genome ( S3 Fig ) . However , at 1 hr post-treatment with NaB , BRD4 , but not BRD2 , showed an increase binding to the lytic and latency control regions ( S3C Fig ) . Taken together , these findings suggest that JQ1 disruption of BRD2 and BRD4 leads to a change in RAD21 and LANA binding to KSHV genome that can not be attributed to an indirect consequence of viral DNA replication . To help resolve the question of whether JQ1 has a direct effect on KSHV epigenetic regulation , we assayed its effects at 1 hr post-treatment . Although this time point is too early to detect changes in KSHV transcription and DNA replication , its effect on BRD2 and BRD4 ChIP could be detected at the TR , and latency control region ( Fig 8B and 8C ) . We also found a reduction in CTCF , RAD21 , and LANA ChIP at the latency control region ( Fig 8D , 8E and 8G ) , and a more modest loss of LANA and H3K9ac at TR ( Fig 8E and 8G ) . We also observed a small increase in RNAPII occupancy at the latent and lytic control regions ( Fig 8F ) . This suggests that JQ1 treatment leads to a rapid ( 1 hr ) change chromatin regulatory factors interactions , including the loss of RAD21 and an increase of RNAPII at lytic promoters . This occurs as early as BRD2 and BRD4 disruption can be detected . To investigate whether chromosome conformation was also affected at this early time point , we performed 3C at 1 hr post-treatment with JQ1 ( Fig 8I ) . We observed a small , but statistically significant decrease in 3C linkages at 72974 and 77153 , indicating that the interaction between the latent and lytic control regions show signs of disruption at the earliest time points measured , and preceding viral lytic DNA replication .
Pharmacogenomics is a valuable tool for understanding biological process and pathways affected by small molecules and candidate pharmacological agents . Here , we have screened a focus library of small molecules with known inhibitory activities directed towards cellular epigenetic regulators and assayed these for their ability to stimulate KSHV lytic cycle gene expression in latently infected B-lymphoma cells . We found that bromodomain inhibitors , including JQ1 , were among the more potent activators of KSHV lytic cycle gene expression . JQ1 was found to induce KSHV lytic cycle transcription , as well as DNA replication , in several different PEL cell lines . We investigated the mechanism of action of JQ1 , focusing on the well-characterized JQ1 target proteins BRD2 and BRD4 . BRD2 and BRD4 were found to interact with KSHV episomes at latency control regions , including the LANA binding sites in TR , and CTCF-cohesin sites at the latency and lytic control regions . Depletion of BRD2 or BRD4 partially phenocopied JQ1 activation of KSHV lytic transcription . JQ1 reduced binding of LANA to the TR and latency control region , but did not destabilize the interaction of BRD4 or BRD2 with LANA protein . JQ1 reduced RAD21 binding and disrupted a DNA loop interaction between the latent and lytic control regions . Taken together , these findings suggest that BRD2 and BRD4 contribute to maintaining the KSHV latent state , including a RAD21-dependent chromosome conformation important for KSHV latency control ( Fig 9 ) . In contrast to our findings , others have found that JQ1 and another BET inhibitor I-BET151 show no evidence of KSHV reactivation [47 , 50] . One possible explanation for these different observations is the different conditions used for JQ1 treatment . In our study , we applied 4 μM JQ1 every day for 3 days , while the previous study applied 0 . 5 μM once and assayed 72 hrs later . We confirmed that JQ1 can inhibit cMyc transcription as early as 6 hrs after addition of JQ1 ( Figs 3 and S1 ) . JQ1 has been shown to inhibit B-cell lymphoma proliferation by disrupting the super-enhancer activation of the cMyc gene [41 , 51] . For KSHV infected PEL cells , JQ1 was found to enhance cell killing in the presence of lenalidomide which was found to selectively degrade and inhibit the IKZF1-IRF4 pathway [50] . It is possible that lenalidomide in combination with BET inhibitors prevents KSHV lytic reactivation . It is also possible that JQ1 inhibition of cMyc and other targets may trigger disruption of KSHV latency through additional indirect mechanisms . This would be consistent with the relatively slow kinetics of viral reactivation after JQ1 treatment . However , our finding that JQ1 rapidly reduces BRD2 , BRD4 , LANA and RAD21 interaction with KSHV genomes , and alters KSHV chromatin conformation independently of viral DNA replication , suggest that JQ1 can also act directly on the KSHV epigenome . BET inhibitors , including JQ1 , are known to activate transcription of latent forms of HIV [42] . For this reason , BET inhibitors have been considered for lytic therapy to cure latent HIV [52] . The mechanism for BET inhibitor activation of HIV may involve complicated and indirect mechanisms for BRD4 and BRD2 [42 , 53] . Inhibition of BRD4 has been shown to release its interaction with 7SK repressor complex to activate RNA polymerase elongation factor pTEFb ( cyclin T1 and CDK9 ) to drive RNA polymerase past positioned nucleosome and TAR RNA barriers [54] . In addition , BRD2 can associate with acetylated TAT protein , as well as interact with transcriptional activators and repressors that regulate HIV reactivation [53] . There are interesting parallels between HIV and KSHV latency control . Similar to HIV , KSHV latency transcription is regulated by a strongly positioned nucleosome and RNA polymerase pausing [55 , 56] . The positioned nucleosome and RNA polymerase pausing depends on the cluster of CTCF binding sites in the first intron of the LANA transcript [55] . Moreover , RNA polymerase associated negative elongation factor ( NELF ) has been implicated in the control of KSHV lytic transcripts [57] . We found that JQ1 increased RNA polymerase II occupancy at the LANA transcript at very early times ( 1 hr ) after treatment ( Fig 8E ) , but this decreased at later times ( 72 hrs ) ( Fig 6D ) . JQ1 decreases 3C loop formation partially at early times ( Fig 8I ) , and more significant at later times ( Fig 7B ) . JQ1 disruption of BRD2 and BRD4 interaction with chromatin is known to occur at very early times , but how these early events regulate subsequent transcriptional and conformational events are not known . We suggest that JQ1 direct disruption of BRD2 and BRD4 interaction with KSHV chromatin leads to several subsequent events , including the loss of LANA binding to TR , loss of RAD21-dependent conformational control , and transcriptional derepression of KSHV lytic immediate early genes . LANA is known to bind to BRD2 [58] and BRD4 [31 , 32] , and biophysical analyses revealed a direct interaction with the LANA DNA binding domain [30] . Interaction with BRD2/4 has been implicated in LANA metaphase chromosome tethering , as well as with transcriptional regulation [31 , 32] . Our results provide evidence that inhibition of bromodomain function by JQ1 reduces LANA interaction with viral genomic DNA , suggesting that these interactions are mediated , in part , through BRD2 and BRD4 association with acetylated lysines . While LANA may have acetylated lysines [59] , it is unlikely that BRD4 associated with LANA through acetyl-lysine binding , as this interaction is known to occur through the BET domain independently of the bromodomains [31] . Consistent with this , we show that JQ1 did not disrupt the interaction between LANA and BRD2 or BRD4 ( Fig 5C ) . In contrast , JQ1 interfered with LANA binding at multiple regions of the genome , including sites that lack known consensus DNA recognition sites , such as at the latency and lytic control regions . We also observed that lytic induction by NaB reduced BRD2 binding throughout the KSHV genome ( S3 Fig ) , but increased BRD4 binding to the latent and lytic control regions at early times after treatment ( S3 Fig ) . This suggests that BRD2 and BRD4 have separate functions in KSHV reactivation , although additional experiments will be necessary to sort these out more precisely . BRD2 and BRD4 association with LANA is likely to facilitate LANA interaction with acetylated histones at these sites on the viral genome . This suggests that BRD2 and BRD4 facilitate LANA binding to TR in the context of chromatin , as well as target LANA to some epigenetic modifications associated with latent and lytic control regions on the viral genome . Our findings also suggest that JQ1 disrupts the 3D conformation of the KSHV genome during latency . This disruption was reflected in the loss of 3C DNA interactions between latent and lytic control regions , as well as the reduction in Rad21 binding at the latency control region . Previous studies have indicated that Rad21 is essential for KSHV DNA conformation and loop interactions , consistent with the known function of cohesin in mediating DNA-DNA interactions [19] . LANA is also found to interact weakly with the latency and lytic control regions by ChIP assay ( Fig 6F ) [29 , 59] . Therefore , it is possible that LANA in association with BRD2 and BRD4 mediates additional contacts between the TR and other regions of the viral genome during latency . Since JQ1 has a major effect on LANA-BRD4/BRD2 binding to TR , we propose that LANA binding at TR is important for maintaining the overall conformation of KSHV during latency ( Fig 9 ) . However , we were unable to demonstrate any direct physical interaction between BRD2 or BRD4 with cohesin subunit RAD21 ( S4 Fig ) , suggesting the conformational control of KSHV latency involves additional factors . High-throughput screening identified several other epigenetic modulators that may regulate KSHV lytic reactivation . We found that other BET inhibitors , including IBET-151 ( Figs 3 and S1 ) , PFI-1 and Bromosporine ( general inhibitors of BRD2 and BRD4 ) , showed low micromolar activity for KSHV reactivation . We also observed activity with several other epigenetic modulators , including inhibitors of G9A and GLP histone H3K9 methylation ( UNC0638 , UNC0642 , and A-366 ) . This may suggest that H3K9 methylation is an important regulator of KSHV latency in B-lymphocytes ( e . g . BJAB cells ) and PEL cells . This is consistent with previous studies that found peaks and valleys of H3K9me3 on the KSHV genome in latently infected BCBL1 cells [60] , and may warrant future investigation . Inhibitors of EZH2 H3K27me3 methyltransferase , such as DZNep , has previously been shown to reactivate KSHV [7] . DZNep was not part of our compound library screen , and although we confirmed that H3K27me3 was enriched at lytic promoter regions ( S5 Fig ) , we did not see any effect on H3K27me3 after BRD2 or BRD4 depletion . This suggests that BRD2 and BRD4 may function independently of the H3K27me3 associated Polycomb repression of KSHV . The findings from this study suggest that multiple epigenetic pathways regulate KSHV latent to lytic switch . Various epigenetic modifications and processes are required to maintain the stable latent cycle gene expression program and the associated chromosome conformation . This information may provide some clinical insights into the treatment of KSHV associated disease . As JQ1 can provide a robust lytic initiating signal for latent KSHV , it may serve as an adjuvant for immune-based therapies and in combination with lytic cycle inhibitors , like gancyclovir , for pharmacological treatment of KS .
BJAB ( uninfected B cell lymphoma ) cells ( ATCC ) , SLK ( uninfected ) cells ( NIH AIDS reagent program ) , BJAB-BAC16 cells , KSHV positive PEL cells ( BCBL1 , BC3 ) ( gift of Yan Yuan , UPENN ) , and double positive KSHV and EBV infected PEL cells ( JSC-1 , BC1 ) ( gift of Yan Yuan , UPENN ) were grown in RPMI medium ( Gibco BRL ) containing 10% heat-inactivated fetal bovine serum and the antibiotics penicillin and streptomycin ( 50U/ml ) . 293T cells ( ATCC ) , iSLK ( gift of J . Jung , USC and D . Ganem , Novartis ) and SLK-BAC16 were cultured in Dulbecco’s modified Eagle’s medium with 10% fetal bovine serum and the antibiotics . iSLK cells were cultured in the presence of 1 μg/ml puromycin and 250 μg/ml G418 . KSHV BAC16 and its derivatives were introduced into iSLK cells via Fugene HD transfection as described previously [61] . Two days post transfection , iSLK-BAC16 cell lines were established and maintained in the presence of 1ug/ml puromycin , 250ug/ml G418 , and 1 , 000ug/ml hygromycin B . For transient transfection , actively growing 293T cells were processed with Lipofectamine reagent ( Invitrogen ) , and the cells were harvested 72 hours post transfection . All cells were cultured at 37°C in a 5% CO2 environment . Stable iSLK-BAC16 cells were induced in the presence of both doxycycline ( 1 μg/ml ) and sodium butyrate ( 1 mM ) and the absence of hygromycin , puromycin , and G418 . Four days later , supernatant was collected and cleared of cells and debris by centrifugation ( 1 , 000 g for 15 min at 4°C ) and filtration ( 0 . 45 μm ) . Virus particles were pelleted in 25% sucrose/1x PBS solution by ultracentrifugation ( 100 , 000 g for 1 h at 4°C ) . BJAB and SLK cells were infected with concentrated KSHV-BAC16 viruses derived from induced iSLK-BAC16 cells as described above . BJAB cells were subjected to spin infection for 30 min at 450 x g and 25°C in the presence of 8 ug/ml of polybrene ( Sigma ) . SLK cells were seeded at approximately 2x105 cells/well in 6 well plate 24h prior to infection and then inoculate overnight . After 48 hrs , cells were selected by hygromycin at 200ug/ml for 2 weeks and KSHV episome presence was checked by PFGE . Derived stable lines were designated as BJAB-BAC16 or SLK-BAC16 cells . KSHV LANA was cloned into p3XFLAG-CMV-24 ( Sigma ) as described previously [59] . Human BRD4 and BRD2 expression constructs were a gift from Dr . Jianxin You ( University of Pennsylvania , School of Medicine ) . Lentiviruses were generated and lentivirus infection were performed as described previously [16] . PEL cells were harvested at 6 or 10 days post puromycin selection . shRNA for BRD2 ( TRCN0000006308 and 6309 ) and BRD4 ( TRCN0000021427 and 21428 ) were obtained from Sigma TRC library . A library of 24 compounds with known epigenetic targets were arrayed in 10-point dose response analysis from 20μM to 1nM . Row B17-B22 and O3-O8 received 2 μM sodium butyrate and 50 μg/ml TPA as positive control for lytic reactivation , and the remaining wells of row B and O received DMSO ( 0 . 2% final concentration ) . BJAB-BAC16 cells were generated by infecting BJAB cells with recombinant BAC16 KSHV derived virus and maintained in RPMI with 10% FBS . 50 μl containing 25 , 000 BJAB-BAC16 cells were dispensed using a Biotek Microflo into clear 384-well tissue culture plates ( Greiner Inc . , cat# 781–192 ) . Fifty nanoliters per well of test compound was transferred to assay plates using a Janus MDT equipped with a 384 nanohead ( Perkin-Elmer Inc ) . Cells were incubated with compounds for 48 h at 37°C in 5% CO2 incubator . To determine the lytic population , 10 , 000 GFP-positive live cells were analyzed by fluorescence activated flow cytometry ( FACS ) using a high-throughput sampler attached to a BD FACSCalibur ( BD Biosciences ) . The gate for the lytic population was set by the increment of GFP intensity during lytic reactivation . The percentage of lytic cells was acquired from this gate for data analysis . Similarly , cells were analyzed to by an Operetta high content scanner ( Perkin Elmer , Inc ) with a 20x objective . For Operetta analysis , assay plates were prepared by incubating 5 , 000 BJAB-BAC16 cells with compounds in 50 μl of RPMI at 37°C in 5% CO2 . Plates were centrifuged at 250g for one minute and 3 fields per well were subsequently imaged at 24h and 48h post drug treatment . The acquired GFP fluorescent images were analyzed using Harmony software and GFP intensity per cell was determined . Data of test compounds was normalized to DMSO controls to calculate fold lytic activation ( i . e . , fold lytic activation = test ( compound ) /average ( DMSO ) ) . EC50 values were determined using Spotfire ( TIBCO , Perkin-Elmer ) data analysis . Active compounds of interest for further study were defined as those with a reproducible EC50 of less than 1μM in activating lytic replication . A Z’-factor was calculated to measure the statistical relevance of the screen , where Z’ equals ( 1 − 3 ( σp − σn ) /|μp − μn| ) where σ is variance , μ is mean , with p representing positive controls , and n negative . Chromatin immunoprecipitation ( ChIP ) assays were performed as described previously [19] . Antibodies used in the ChIP assays are listed below . Primers for ChIP assays were used as described previously [19] . PCR data were normalized to input values that were quantified in parallel for each experiment . The following antibodies were used for ChIP assays: anti-IgG ( Santa Cruz Biotechnology ) , anti-CTCF ( Millipore ) , anti-Rad21 ( Abcam ) , anti-acetylated H3K9 ( Millipore ) , RNA polymerase II ( Santa Cruz sc-889x ) , anti-BRD2 ( Bethyl ) , anti-BRD4 ( Bethyl ) antibodies . The mouse monoclonal antibody anti-IgG ( Santa Cruz Biotechnology ) and Rat anti-KSHV LANA antibody ( Advanced Biotechnologies Inc . ) were used for ChIP assays . Rabbit polyclonal anti-BRD4 ( Bethyl ) , mouse monoclonal anti-actin ( Sigma ) and anti-FLAG ( Sigma ) antibodies were used for Western blotting . JQ1 was a gift from the Jay Bradner Lab , I-BET151 from Sigma-Aldrich and BIC1 from Calbiochem and were used at a concentration of 4 uM . Phosphonoacetic acid ( PAA ) was purchased from Sigma and used at a concentration of 400 ug/ml . The amount of intracellular KSHV DNA was determined by quantitative PCR ( qPCR ) analysis of purified total genomic DNA as described previously [19] . Immunoprecipitation ( IP ) was performed as described previously [62] . RT-PCR was performed as described previously [19] . BCBL1 cells infected with lentivirus expressing shControl , shBRD2 or shBRD4 were used for PFGE and PFGE was performed as described previously [19] . Hirt DNA extraction and Southern analysis were performed as described previously [19] . KSHV DNA was quantified by PhorphorImager . p-values were calculated by 2-tailed student t-test using Excel ( Microsoft , Redmond , WA ) . * p<0 . 05 , ** p<0 . 01 , *** p<0 . 001 . | KSHV is an oncogenic human herpesvirus implicated as the causative agent of KS and cofactor in pleural effusion lymphomas ( PELs ) . The latent virus persists in PELs as an epigenetically regulated episome . We found that small molecule inhibitors of BET family have potent activity in triggering the lytic switch during latent infection in PELs . The BET family inhibitor JQ1 disrupted the latent virus from maintaining a closed DNA loop conformation . These findings have implications for treatment of KSHV-associated malignancies with epigenetic modulators of the BET inhibitor family . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"gene",
"regulation",
"pathogens",
"microbiology",
"viruses",
"dna",
"replication",
"dna",
"viruses",
"genome",
"analysis",
"epigenetics",
"molecular",
"biology",
"techniques",
"dna",... | 2017 | BET-Inhibitors Disrupt Rad21-Dependent Conformational Control of KSHV Latency |
The behavior and phenotypic changes of cells are governed by a cellular circuitry that represents a set of biochemical reactions . Based on biological functions , this circuitry is divided into three types of networks , each encoding for a major biological process: signal transduction , transcription regulation , and metabolism . This division has generally enabled taming computational complexity dealing with the entire system , allowed for using modeling techniques that are specific to each of the components , and achieved separation of the different time scales at which reactions in each of the three networks occur . Nonetheless , with this division comes loss of information and power needed to elucidate certain cellular phenomena . Within the cell , these three types of networks work in tandem , and each produces signals and/or substances that are used by the others to process information and operate normally . Therefore , computational techniques for modeling integrated cellular machinery are needed . In this work , we propose an integrated hybrid model ( IHM ) that combines Petri nets and Boolean networks to model integrated cellular networks . Coupled with a stochastic simulation mechanism , the model simulates the dynamics of the integrated network , and can be perturbed to generate testable hypotheses . Our model is qualitative and is mostly built upon knowledge from the literature and requires fine-tuning of very few parameters . We validated our model on two systems: the transcriptional regulation of glucose metabolism in human cells , and cellular osmoregulation in S . cerevisiae . The model produced results that are in very good agreement with experimental data , and produces valid hypotheses . The abstract nature of our model and the ease of its construction makes it a very good candidate for modeling integrated networks from qualitative data . The results it produces can guide the practitioner to zoom into components and interconnections and investigate them using such more detailed mathematical models .
While the genome contains all hereditary information , the decisions that a cell makes are governed by a complex cellular machinery that resides above the genome . Modeling this machinery is both important—as it helps understand proper cellular functioning and the implications of aberrations thereof , and a daunting—given the “known unknowns” ( e . g . , kinetic parameters of given reactions ) and the “unknown unknowns” ( data incompleteness is the rule , rather than the exception , in biological research ) . The cellular machinery can be broken down into three main components—signaling , transcription regulation , and metabolism—each of which consists of a network of molecules and interactions among them . The signaling network is responsible for relaying messages from the external environment of a cell to the nucleus . Inside the nucleus , the transcription regulation network determines , upon receiving signals , which genes are expressed , and to what extent . The metabolic network is the energy and resource management component of the cell , producing energy and products that are required by cellular processes . Various modeling techniques have been used successfully for modeling the dynamics of each of these components individually . The success of modeling each of the three components individually notwithstanding , these components are interconnected within the cell and their dynamics are intertwined , thus creating a complex network whose modeling and understanding are major endeavors in systems biology . Several biological studies and surveys have highlighted this interconnection inside the cell and the significance of analyzing the components simultaneously rather than individually , including , but not limited to , [1]–[6] . Indeed , several approaches were introduced recently for integrated modeling of biological networks: regulatory FBA ( rFBA ) [7] , steady state regulatory FBA ( SR-FBA ) [8] , integrated FBA ( iFBA ) [9] , integrated dynamic FBA ( idFBA ) [10] , probabilistic regulation of metabolism ( PROM ) [11] , the method of [12] , dynamic FBA ( dFBA ) [13] and a recently published whole cell computational model [14] . One common aspect to all the existing models is the use of flux balance analysis ( FBA ) for modeling carbon and energy metabolism . FBA is a widely used method that estimates fluxes of metabolic reactions , thereby making it possible to predict the growth rate of an organism or the rate of production of a metabolite of interest . However , FBA is only suitable for determining fluxes at steady state . With exceptions of some modified forms , FBA does not account for regulatory effects such as activation of enzymes by protein kinases or regulation of gene expression [15] . The methods that use the unmodified version of FBA – all but idFBA and dFBA — only capture the steady state of metabolism , therefore not capturing the full dynamic within the cell . These methods mainly acquire the effects of changes that individual components have on each other . On the other hand , the methods that discretize FBA ( dFBA and idFBA ) , are able to reveal not only a more complete profile of the cell , but also the dynamic behavior of the interconnections between the components . For recent surveys of these methods , please see [16] , [17] . In this paper , we propose a new Integrated Hybrid Model ( IHM ) that aims to capture the dynamic behavior within and between the components of the cell , and which belongs to the class of executable models [18] . This model integrates two types of modeling techniques: Petri nets ( PNs ) , which have been used for modeling metabolic networks and signaling networks [19] , and Boolean networks , which have been used to model regulatory networks as well as protein signaling networks [20] , [21] . One of the first successful Petri net-based models of metabolism was devised by Reddy et al . [22] , [23] . Over the recent years , various types of Petri nets have been introduced and extensively used in modeling different metabolic systems [24]–[27] . Signaling pathways , on the other hand , have posed more of a challenge for Petri nets . Their highly interleaved ( with possible forward- and backward loops ) and parametrized nature makes it a difficult mapping onto a Petri net framework . Despite these limitations , Petri nets have been shown to be applicable in signaling pathways using careful parameterization and execution strategies [28]–[32] . Transcription regulation has been modeled successfully using Boolean networks , starting with the work of [33] . Over the years , with the steady increase in the amount of data on genetic regulation , Boolean networks became a common strategy for modeling this cellular process; e . g . , [34]–[36] . Our integrated hybrid model uses Petri nets to model the metabolic and signaling components , and Boolean networks to model the transcriptional component . Further , the model makes connections between the Petri net and Boolean network component using a special modeling part . Our modeling approach assumes knowledge of the connectivity among the various species in the system , and is then minimally parameterized based on qualitative data . The dynamics of the biological system are then obtained by executing the parametrized model . Of the existing approaches , idFBA is comparable to our approach , as it allows for modeling the dynamics by discretizing time and conducting FBA analyses for short time intervals . However , idFBA is applicable where FBA models have been curated ( e . g . , for single-cell organisms ) , whereas our modeling approach is applicable more broadly in terms of organism selection , and requires only qualitative data . We implemented and tested our modeling methodology on two biological systems: ( 1 ) the transcriptional regulation of glucose in human physiology , with knowledge based on [1] , and ( 2 ) osmoregulation in S . cerevisiae , based on the system in [37] . The two systems differ in temporal and spatial scales . For the transcriptional regulation of glucose , the interactions among different components are reflected in the cooperation among multiple cell types , and the mass transportation is through blood vessels in the human body , thus acting at longer time scales than single cell systems . On the other hand , the modeling of osmoregulation in S . cerevisiae encompasses metabolism , signaling and transcriptional regulation , all within a single cell . The exchange of proteins or metabolites is mediated through diffusion and cellular transportation . We choose the two systems to show the diversity of the biological scenarios to which our integrated hybrid model is applicable . The two systems are very well curated and studied , both experimentally and computationally . This makes them ideal for validating our methodology and for comparing with existing modeling frameworks . Our modeling approach produced results that match experimentally derived data ( in terms of both validation and prediction ) . There is an abundance of qualitative data on biological interaction networks , and developing models and methods that utilize such data is desirable . Our proposed method fits within this category which offers a complementary approach , rather than an alternative one , to the FBA-based category of methods as well as other categories such as kinetics-based methods .
In our context , a Petri net ( PN ) is a 4-tuple that defines a weighted , complete , directed , bipartite graph . The disjoint sets and correspond to two types of nodes , places and transitions , respectively . In modeling signal transduction and metabolism , they correspond to chemical species and biochemical reactions that happen among these species . The element is a mapping defined , where is the set of non-negative real numbers . These mappings could be used to encode , for example , stoichiometries of biochemical reactions . Finally , is the initial marking of the Petri net , which assigns a number of tokens to each place . This correspond to the initial concentration of chemical species . The state of a Petri net is given by a vector of length with being the number of tokens in place . In particular , the initial state , , is given by the initial marking . Additionally , a vector of length provides the transition rates for the system , where denotes the rate of transition to simulate the empirical rate constant used in the law of mass action that governs the corresponding reaction . The Petri net can be executed both deterministically and stochastically [38]–[40] . In this work , we utilize a stochastic protocol based on the Gillespie “first reaction” method [41] . The method characterizes the dynamics of each transition by a propensity function . Let be a transition whose inputs is the set and outputs is the set . In state , the propensity of transition is defined byGiven these propensity values , the method determines the putative time at which the next transition fires based on the probability distribution function given byThe transition with the smallest time is then chosen to fire . Firing transition amounts to updating the number of tokens in every place according to the rule and updating the number of tokens in every place according to the rule . Once a transition is executed , the state of the Petri net changes . The execution time is updated by 1 , which is , in our case , a slight modification from the original algorithms where time is updated by . Consecutive firings of transitions results in a walk through the state space of the Petri net from the start state . The final dynamics of the system is acquired by averaging several full runs of Gillespie starting from the initial state and executing the same number of steps . A detailed description of Petri nets and its application to systems biology can be found in [19] . See Figure 1 for an illustration . A Boolean network is a 3-tuple , where is a vector of Boolean variables ( that is , variables that take values in the set ) and is a vector of Boolean functions with function , for , associated with variable , and is a vector of length that has a Boolean value for each of the variables and denotes the start state . In modeling transcriptional regulation , each Boolean variable indicates whether a gene is being transcribed at a given time and the Boolean functions stipulate how transcriptional factors regulate the transcription of their targets . The state of a Boolean network is a Boolean vector of size , where is the value of variable . The value of of variable is updated by applying function to the current state of the Boolean network . More formally , let be the state of the Boolean network at time . Then , if function is executed at time , the state of the Boolean network one step later is given by , where for every , and . In particular , . Given a Boolean network representing a set of variables , the dynamics of the system can be simulated by repeatedly executing the Boolean functions and updating the “current” state . In the classical synchronous simulation , the states of all variables are updated simultaneously after all of the functions in have executed . In an asynchronous simulation , only one Boolean function is chosen and executed in a given time step . See Figure 1 for an illustration . As described above , gene regulatory networks have been successfully modeled using Boolean networks . Signaling and metabolic networks have been successfully modeled using Petri nets . In our integrated hybrid model , the regulatory components of the biological system are modeled using Boolean networks , whereas the other two components are modeled using Petri nets . To facilitate connections between the two components , our model contains , in addition to the Petri net and Boolean network components , a set of Place-to-Boolean and Boolean-to-Place triplets that create a Boolean value based on binarization of the number of tokens and a number of tokens based on a Boolean value , respectively . We now describe our modeling approach formally .
In order to assess the ability of IHM to capture the dynamics of complex biological systems we implement an IHM model for the system of transcriptional regulation of glucose metabolism , which was surveyed in [1] . Timely uptake of cellular glucose from the blood , a task regulated by the secretion of insulin and glucagon , is crucial to human metabolism . This system involves the interaction of multiple cellular components in cells of different cell types and cells that span a physical distance . We demonstrate that IHM can readily be adapted to model such a biological scenario , and allows us to investigate issues such as the interplay between AKT and FOXO in this system . Given that the modeled system involves more than a single cell type , it is unclear how to apply FBA-based techniques to it . Yeast responds to the environmental osmolarity by adjusting the cellular glycerol concentration [37] . Such response is mediated through signaling pathways that sense the extracellular osmotic pressure as well as transcriptional regulation of about 10% of the yeast genes that manipulate the metabolism of glycerol . The effect of the medium osmolarity is first sensed and transmitted by the well-studied HOG/MAPK pathway [64] , [65] whose upstream involves two redundant branches—Sho1 branch [66] , [67] and Sln1 branch [68] . The HOG signaling pathway is one of the first to sense the osmotic upshift , playing a pivotal role in yeast's adaptation to high osmolarity . Hog1 , the end effector of HOG pathway , activates in the nucleus the central transcriptional factors Hot1 [69] , Msn2/4 [70] and Ptp2/3 [71] . These transcriptional factors turn on the expression of enzymes that promote glycolysis , which leads to the production of glycerol , an inert osmolyte . The surge in the glycerol concentration increases the cytosolic osmolarity , counteracting the osmotic upshift in the environment and protecting the cell from dehydration . While the effect of Gpd1/Gpp2 ( which is a product of Hot1 and MSN2/4 ) gene controls osmoregulation via glycerol production though metabolic pathway , Ptp2/3 is a much stronger mediator of osmotic stress , as it acts on suppressing the activity of Hog1 transcription factor directly .
We proposed a simple , yet effective , integrated hybrid model ( IHM ) that allows for simultaneously modeling signaling , metabolic , and regulatory processes within a single framework , while explicitly capturing the dynamics within each component and the interplay among them . As we applied the integrated model to two biological systems , we demonstrated how much our model can capture by mainly relying on the topology of the system ( given the simple and general rules for setting most of the model's parameters ) . In both systems , we were able to successfully validate our results against both experimental data and other models . In the case of transcriptional regulation of glucose , we compared our model against an ODE-based model that only focuses on glucose-insulin interactions , while in our case we consider a larger system . The results compare well against the experimental data . No comparison was done with other integrated models , since it is not clear how to formulate an FBA-based model for this system . In the case of the osmoregulation system , we compared our model against the idFBA approach [10] . The IHM framework has an intuitive graphical representation that makes the construction of the connectivity map of the model a relatively simple task . Further , as experimental evidence becomes available to provide support for new connections or against existing ones , the connectivity map can be readily updated to accommodate this new evidence without having to recreate the model from scratch . Our model is reconstructible and its parameterization is obtainable from qualitative data , which is abundant in the literature and public databases . It is important to note that while the connectivity map is often easy to obtain from the literature and public databases , parameterizing the IHM poses the biggest challenge in terms of obtaining the executable model . In this paper , we parameterized the IHM for both biological systems manually—a task that took very short time to achieve , given that most of the parameters were set using general rules and only a few of them had to be fine-tuned . The results ( e . g . , the feed/fast cycle in the regulation of glucose metabolism system ) are qualitatively robust to most parameter values that we choose , as tested by executing the model with parameters varied around the chosen value . We identify as a direction for future research the task of devising computational techniques for automated parameterization of our IHM using qualitative experimental data . Some techniques for a similar task were recently introduced [72] and we will build on those . While the aforementioned existing approaches for integrated analysis of biological networks provide promising frameworks , a salient feature of all of them is that they depend on flux-balance analysis ( FBA ) as a main analytical component . This dependence means that an FBA model must be curated for the system under analysis , which is not clear how to obtain for a system such as the regulation of glucose metabolism , which involves more than a single cell type . Further , this dependence necessarily makes the analysis metabolism-centric and shifts the focus from the other two components . Third , as FBA is aimed at understanding the behavior of the system at steady state , the dynamics of the system cannot be studied , except under the idFBA modeling technique , as it takes a step-wise approach to conducting FBA . Our model , on the other hand , is not based on FBA and , consequently , provides a complementary approach to the FBA-based ones . Our model builds on the success of Boolean networks and Petri nets for modeling cellular networks . As advances continue to be made for both modeling techniques , our integrated modeling approach would readily benefit from these advances , as different flavors of of Boolean networks ( e . g . , probabilistic ones ) and Petri nets ( e . g . , colored Petri nets ) can be plugged into our model without having to modify the way the connectivity map is constructed or the system is executed . In other words , our model can be viewed as a reconfigurable model , where different components , along with their execution protocols , can be assembled to generate a model of integrated systems . It is important to note that while we made decisions on the model to fit the two biological systems we studied , other biological systems may require more features in the modeling approach . For example , in the Petri-to-Boolean connections , it might be the case that the state of the Boolean variable is set based on a function of a set of the Petri net places . Our IHM can be easily extended to incorporate such features , with little or no need to modify the execution strategy . That is , the model is easy to extend as long as the syntax of the new features and their effects on the execution strategy are well-defined . Last but foremost , our IHM approach lends itself in a straightforward manner to hypothesis generation . Perturbation experiments can be simulated in silico by setting the numbers of tokens at Petri net places and Boolean variables to a certain value , and the system can be executed to study the effect . For example , a Boolean variable can be set to 0 to simulate its inhibition , or the number of tokens can be set to a large number in place to represent a constitutive enzyme . Further , new components can be added in or existing ones can be removed easily to study the effect of these components on the overall performance of the system . Finally , while we chose to model transcriptional regulation using Boolean networks here , the entire system ( that is , all three types of biological networks ) could be represented using a single Petri net . This allows for a more refined simulation of the transcription factors and their targeted genes , but also requires replacing the Boolean functions by Petri net transitions whose parameters must be learned from the data . | Within the cell of an organism , three networks—signaling , transcriptional , and metabolic—are always at work to determine the response of the cell to signals from its environment , and consequently , its fate . Evidence from experimental studies is painting a picture of complex crosstalk among these networks . Thus , while a wide array of computational techniques exist for analyzing each of these network types , there is clear need for new modeling techniques that allow for simultaneously analyzing integrated networks , which combine elements from all three networks . Here , we provide a step towards achieving this task by combining two population modeling techniques—Petri nets and Boolean networks—to produce an integrated hybrid model . We demonstrate the accuracy and utility of this model on two biological systems: transcriptional regulation of glucose metabolism in human cells , and cellular osmoregulation in yeast . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2013 | Modeling Integrated Cellular Machinery Using Hybrid Petri-Boolean Networks |
Major gaps in our understanding of Plasmodium vivax biology and the acquisition of immunity to this parasite hinder vaccine development . P . vivax merozoites exclusively invade reticulocytes , making parasite proteins that mediate reticulocyte binding and/or invasion potential key vaccine or drug targets . While protein interactions that mediate invasion are still poorly understood , the P . vivax Reticulocyte-Binding Protein family ( PvRBP ) is thought to be involved in P . vivax restricted host-cell selectivity . We assessed the binding specificity of five members of the PvRBP family ( PvRBP1a , PvRBP1b , PvRBP2a , PvRBP2b , PvRBP2-P2 and a non-binding fragment of PvRBP2c ) to normocytes or reticulocytes . PvRBP2b was identified as the only reticulocyte-specific binder ( P<0 . 001 ) , whereas the others preferentially bound to normocytes ( PvRBP1a/b P≤0 . 034 ) , or showed comparable binding to both ( PvRBP2a/2-P2 , P = 0 . 38 ) . Furthermore , we measured levels of total and IgG subclasses 1 , 2 , 3 and 4 to the six PvRBPs in a cohort of young Papua New Guinean children , and assessed their relationship with prospective risk of P . vivax malaria . Children had substantial , highly correlated ( rho = 0 . 49–0 . 82 , P<0 . 001 ) antibody levels to all six PvRBPs , with dominant IgG1 and IgG3 subclasses . Both total IgG ( Incidence Rate Ratio [IRR] 0 . 63–0 . 73 , P = 0 . 008–0 . 041 ) and IgG1 ( IRR 0 . 56–0 . 69 , P = 0 . 001–0 . 035 ) to PvRBP2b and PvRBP1a were strongly associated with reduced risk of vivax-malaria , independently of age and exposure . These results demonstrate a diversity of erythrocyte-binding phenotypes of PvRBPs , indicating binding to both reticulocyte-specific and normocyte-specific ligands . Our findings provide further insights into the naturally acquired immunity to P . vivax and highlight the importance of PvRBP proteins as targets of naturally acquired humoral immunity . In-depth studies of the role of PvRBPs in P . vivax invasion and functional validation of the role of anti-PvRBP antibodies in clinical immunity against P . vivax are now required to confirm the potential of the reticulocyte-binding PvRBP2b and PvRBP1a as vaccine candidate antigens .
The two major malaria parasites , Plasmodium falciparum and Plasmodium vivax , differ in their ability to invade human erythrocytes . While P . falciparum invades both mature ( normocytes ) and young erythrocytes ( reticulocytes ) , P . vivax can only invade the latter [1] . This differential specificity is believed to be mediated by distinct ligand-receptor interactions , though the exact mechanisms remain to be elucidated [1] . For P . falciparum , the merozoite invasion of erythrocytes is a multistep process [2] mediated by the binding of the erythrocyte binding-like ( EBL ) and reticulocyte binding-like ( PfRh ) protein families to receptors on the surface of the host cell [3] . For P . vivax , the only ligand-receptor interaction identified to date is between the Duffy binding protein ( PvDBP ) and the Duffy antigen receptor for chemokines ( DARC ) [4] . The recently identified P . vivax erythrocyte-binding protein ( PvEBP ) also shares a Duffy binding-like domain [5] . However , the presence of DARC in both normocytes and reticulocytes does not explain the restricted host-cell selectivity of P . vivax . The recent observation that P . vivax can invade Duffy-negative cells also indicates the existence of alternative pathways of invasion [6 , 7] . The P . vivax reticulocyte binding brotein family ( PvRBP ) is composed of 11 members [5 , 8 , 9] , and although their precise roles remain largely unknown , their homology to the much better characterized PfRh protein family suggests that they may be important invasion ligands [3] . Members of the PvRBP family have been implicated in erythrocyte binding , and in some cases in reticulocyte recognition [8 , 10] . Variation in expression of PvRBPs genes in different parasite isolates have been described , suggesting that these genes may be redundant in function [11] . The relatively high degree of polymorphism observed in the genes encoding PvRBPs also indicates that they are important for parasite survival and may be under immune selection [10 , 12 , 13] . Collectively , this suggests that the understudied PvRBP family may be of key importance for P . vivax invasion , and like their better studied P . falciparum homologues , potential targets for a vaccine targeting blood stage infections [14] . Antibodies to several P . vivax merozoite proteins have shown associations with reduced risk of vivax-malaria in naturally exposed individuals [15] . Among them , antibodies to PvDBPs have been target of extensive study [15] . While PvDBP is a promising vaccine candidate , several challenges to vaccine development remain , including the presence of highly polymorphic , immuno-dominant epitopes in the DARC-binding region II , and the need to elicit high titers to achieve strain-transcending blocking [14] . It is therefore likely that PvDBP would need to be combined with further antigens targeting alternative invasion ligands . Members of the PvRBP family are recognized by antibodies from vivax-positive patients [11 , 16] , and populations living in endemic areas [17 , 18] . Yet , an in-depth characterization of the immune responses to these proteins , as well as the role of antibodies to PvRBPs in the acquisition of immunity to malaria is lacking . In this study , we have assessed the erythrocyte-binding profiles of five members of the PvRBP family and their specificity to normocytes or reticulocytes , identifying a member of the PvRBP family that exclusively binds to reticulocytes . Furthermore , we measured levels of total and IgG subclasses to these five PvRBPs and a non- erythrocyte binding protein fragment of a sixth PvRBP in a cohort of young Papua New Guinean ( PNG ) children with well-characterized differences in exposure . We identified an association between reduced risk of vivax-malaria and antibodies to two of the PvRBPs , including the reticulocyte-specific binder . Our results provide important insights into the acquisition of immunity to PvRBPs in young children , highlighting this protein family as an interesting target to be further evaluated for their potential as P . vivax vaccine antigens .
Proteins included in this study were PvRBP1a ( amino acids [aa] 160–1170 ) , PvRBP1b ( aa 140–1275 ) , PvRBP2a ( aa 160–1135 ) , PvRBP2b ( aa 161–1454 ) , PvRBP2cNB ( aa 501–1300 ) and PvRBP2-P2 ( aa 161–641 ) Their expression and purification have been described in details elsewhere [10 , 11] . Despite several attempts , we were unsuccessful in expressing recombinant PvRBP2c that includes the conserved erythrocyte-binding domain in E . coli using both native and refolding methods . Thus , the PvRBP2cNB fragment included in this study does not contain the erythrocyte-binding domain , which encompasses residues 128 to 429 . An SDS-PAGE of PvRBP2-P2 recombinant protein is shown in S1 Fig; the purity and stability of the remaining PvRBPs have been verified and presented in a previous publication [11] . Antibody production was performed at the Walter and Eliza Hall Institute Monoclonal Antibody Facility as previously described [10] . 96-well flat-bottomed plates ( Maxisorp , Nunc ) were coated with each of the 6 PvRBPs ( 65 nM/well ) in individual wells and incubated for two hours . For 65 nM of protein in 100 μL , we added 0 . 8 μg , 0 . 9 μg , 0 . 7 μg , 1 μg , 0 . 6 μg and 0 . 4 μg for PvRBP1a , PvRBP1b , PvRBP2a , PvRBP2b , PvRBP2cNB and PvRBP2_P2 respectively . Plates were blocked with 5% skim milk/0 . 1% Tween-20 for one hour . After washing , specific anti-PvRBP polyclonal antibodies ( 1 mg/mL stock ) were added at halving serial dilutions ( from 1:2000 to 1:64000 ) for one hour . Plates were washed three times before the addition of HRP-goat anti-rabbit secondary antibodies ( 1:2000 dilution ) for one hour . Azino-bis-3-ethylbenthiazoline-6-sulfonic acid ( ABTS liquid substrate; Sigma-Aldrich ) was used to detect HRP activity . 1% SDS was used to stop the reaction and absorbance was measured at 405 nm . All experiments were performed at room temperature . All washes were done in PBS/0 . 1% Tween-20 , and dilutions of antibodies in 0 . 5% skim milk/0 . 1% Tween-20 . Samples were tested in duplicates . Reticulocytes were enriched from whole blood and the erythrocyte-binding assays performed as described previously [10] . Binding of PvRBPs was detected using 0 . 025 mg/mL of corresponding anti-PvRBP rabbit IgG . Antibody reactivity to the 6 PvRBPs in naturally-exposed individuals was assessed in samples from a longitudinal cohort of 264 children ( 1–3 years old ) undertaken in Ilaita , East Sepik Province , PNG [19] . Children were enrolled between March-September 2006 , and followed for up to 16 months . Blood samples were collected every eight weeks and at episodes of febrile illness . All P . vivax infections were genotyped , allowing the determination of the incidence of genetically distinct blood-stage infections acquired during follow-up ( i . e . the molecular force of blood-stage infections , molFOB ) [20] . Samples collected at enrolment from 224 children that completed follow-up were included in the present study ( median age 1 . 7 , inter-quartile range [IQR] 1 . 3–2 . 5 ) . To measure antibody levels in the cohort of PNG children , purified proteins were conjugated to Luminex Microplex microspheres ( Luminex Corp . ) as described elsewhere [21 , 22] , using the following concentrations per 2 . 5 x 106 beads: PvRBP1a = 3 μg/mL; PvRBP1b = 11 . 4 μg/mL; PvRBP2a = 6 . 7 μg/mL; PvRBP2b = 0 . 2 μg/mL; PvRBP2cNB = 0 . 8 μg/mL; PvRBP2-P2 = 5 . 4 μg/mL . Bead-array assays were performed as previously described [23] . Plasma samples were diluted 1:50 in PBS , and secondary antibody donkey F ( ab’ ) 2 anti-human IgG Fc R-PE ( 1 mg/ mL , Jackson Immunoresearch ) ; mouse anti-human IgG1 hinge-PE ( 0 . 1 mg/ mL , clone 4E3 , Southern Biotech ) ; IgG2 Fc-PE ( 0 . 1 mg/ mL , clone HP6002 , Southern Biotech ) ; IgG3 hinge-PE ( 0 . 1 mg/ mL clone HP6050 , Southern Biotech ) ; or IgG4 Fc-PE ( 0 . 1 mg/ mL , clone HP6025 , Southern Biotech ) diluted 1:100 in PBS was added to detect total , IgG1 , IgG2 , IgG3 or IgG4 respectively . A dilution series of a pool made of serum collected as part of an earlier study with immune adults living in different villages of high malaria transmission in East Sepik Province , PNG , was included on each plate as positive controls . To correct plate-to-plate variations , the dilutions of the PNG adult pool were fitted as plate-specific standard curves using a 5-parameter logistic regression model [22 , 24] . For each plate , median fluorescence intensity ( MFI ) values were interpolated into relative antibody units based on the parameters estimated from the plate’s standard curve . Associations with age , exposure and correlations between antibody levels of different subclasses and/or different antigens were determined using Spearman’s rank correlation , and differences by infection status using Mann-Whitney U tests . Generalized estimating equation ( GEE ) models with exchangeable correlation structure and semi-robust variance estimator were used to analyze the relationship between antibodies to PvRBPs and prospective risk of P . vivax episodes ( defined as axillary temperature ≥ 37 . 5°C or history of fever in preceding 48 hours with a concurrent parasitaemia >500 P . vivax parasites/μl ) over the 16 months of follow-up [22 , 25] . For this , antibody levels were classified into tertiles ( cut-off values are shown in Table 1 ) , and analyses were done comparing the incidence rate ratio ( IRR ) of clinical malaria in those with medium and high versus low antibody levels . Children were considered at-risk from the first day after the blood sample for active follow-up was taken . For each child , the molFOB was calculated as the number of new blood-stage genetically distinct P . vivax clones acquired/year-at-risk , and square-root transformed for better fit [20] . Adjustments were made for seasonal trends , village of residency , age , and molFOB . In order to study the breadth of anti-PvRBP antibodies , for each antigen antibody levels stratified into tertiles were scored as 0 for low , 1 for medium and 2 for the high tertiles , respectively . Scores were then added up to reflect the breadth of anti-RBP antibodies , yielding a median score of 6 ( IQR 2–9 ) . All analyses were performed using STATA version 12 ( StataCorp ) or R version 3 . 2 . 1 ( http://cran . r-project . org ) . Ethical clearance was obtained from the PNG Medical Research and Advisory Committee ( MRAC 05 . 19 ) , and the Walter and Eliza Hall Institute ( HREC 07/07 ) . Written informed consent was obtained from the parents or guardians all children participating in the cohort study .
Except for the PvRBP2cNB fragment , all PvRBPs proteins expressed encompass the conserved erythrocyte-binding domain [10] . As such , PvRBP2cNB binding serves as a control for background signal in this flow cytometry-based assay . Binding was significantly higher for all binding fragments except for PvRBP2b in normocytes ( not stained with thiazole orange , TO- ) and PvRBP1b in reticulocyes ( stained with thiazole orange , TO + ) ( both P>0 . 05 ) . Among the five binding PvRBPs tested we found three types of binding profiles: i ) binding preferentially to normocytes: PvRBP1a ( Fig 1 , TO- vs . TO+: P = 0 . 034 ) and PvRBP1b ( P = 0 . 017 ) ; ii ) binding to both normocytes and reticulocytes: PvRBP2a ( P = 0 . 38 ) and PvRBP2-P2 ( P = 0 . 38 ) ; and iii ) binding only to reticulocytes: PvRBP2b ( P < 0 . 001 ) . We assumed that the pooled serum from hyper-immune PNG adults represented the equilibrium antibody levels to all proteins achievable under life-long natural exposure . Therefore , by comparison with IgG levels observed in PNG children , we determined how many children have already achieved IgG levels that were >50% , >25% or >10% of the adult levels ( Table 1 ) . Although , semi-immune , young PNG children were reactive to all six PvRBPs tested , there were differences in the immunogenicity of different proteins ( Table 1; S2 Fig ) . Whereas , 47% and 95% of children had reached >50% and >10% of the hyper-immune adult levels for PvRBP2-P2 , respectively , only 7% and 20% reached the same levels for antibodies targeting the non-red cell binding PvRBP2cNB fragment ( P < 0 . 001 ) . The other proteins were intermediately immunogenic , with antibodies to PvRBP2a more rapidly acquired than those to PvRBP2b , PvRBP1a and PvRBP1b ( Table 1; S2 Fig ) . To each PvRBP , total IgG levels correlated moderate to strongly with IgG levels to the other PvRBPs ( rho = 0 . 49–0 . 82 , P < 0 . 001 ) , with the strongest correlation between PvRBP1b and PvRBP2b ( rho = 0 . 82 ) ( S1 Table ) . Rabbit polyclonal antibodies against the different PvRBP constructs showed strong recognition of the specific constructs but only limited cross-reactivity ( Fig 2 ) . The exception was antibodies raised against the PvRBP2cNB fragment , which showed low cross-reactivity with PvRBP1b and PvRBP2a . This indicated that the high correlations in naturally acquired total IgG levels are therefore likely to reflect co-acquisition rather than cross-reactivity ( Fig 2 ) . While antibodies to PvRBP1a , PvRBP1b , PvRBP2b and PvRBP2cNB increased moderately with age ( rho = 0 . 15–0 . 29 , P < 0 . 001–0 . 025 ) , no such association was found for the two most immunogenic proteins , PvRBP2a and PvRBP2-P2 ( Fig 3A; S2 Table ) . Total IgG levels to all proteins except PvRBP1b were significantly higher in the 124 children ( 55 . 4% ) that had a current , PCR-detectable P . vivax infection ( P < 0 . 001–0 . 006 ) ( S2 Table ) . To better understand the effect of age on the acquisition of antibodies to PvRBPs , we stratified children by the presence of infection at sample collection . After stratification , increase in total IgG to PvRBP1b , PvRBP2b , PvRBP2cNB and PvRBP2-P2 with age were stronger in children with current infection ( rho = 0 . 19–0 . 33 , P < 0 . 001–0 . 039 ) suggesting that antibodies to PvRBPs are strongly reflective of recent exposure ( S2 Table ) . Given the young age and large heterogeneity in exposure among children [20] , age is not a good proxy for life-time exposure . As a better measure of life-time exposure to malaria , we therefore calculated the molFOB ( described in methods ) [20] . This estimated P . vivax life-time exposure was a substantially better predictor of antibody levels than age , and total IgG levels to all 6 PvRBPs significantly increased with increasing life-time exposure ( rho = 0 . 18–0 . 36 , P < 0 . 001–0 . 006 ) ( Fig 3B; S2 Table ) . For PvRBP1a and PvRBP2a , the effect of life-time exposure to P . vivax was observed only in children free of infection at study start ( P = 0 . 008–0 . 022 ) , again indicating that the effect of recent infections on antibodies to PvRBPs is strong ( Fig 3C; S2 Table ) . The breadth of anti-PvRBP antibodies ( described in methods ) was higher in children with concurrent P . vivax infection ( median in children free of infection = 4 . 5 , IQR = 2–8 versus median in infected children = 7 , IQR = 4–10; P < 0 . 001 ) , and increased with increasing estimated life-time exposure ( rho = 0 . 29 , P < 0 . 001 ) . In hyper-immune PNG adults , four different patterns of IgG subclass reactivity to PvRBPs were observed i ) predominant IgG1 with sub-dominant IgG3: PvRBP2-P2; ii ) predominant IgG3 with sub-dominant IgG1: PvRBP1a; iii ) predominant IgG1 with sub-dominants IgG2+IgG3: PvRBP2a , PvRPB2b and PvRBP2cNB; and iv ) IgG1+IgG2+IgG3 with no obvious dominance: PvRBP1b . There were no detectable levels of IgG4 to any of the proteins ( Table 1; S3 Fig ) . In comparison to adults , young children had already acquired substantial IgG1 levels to all 6 PvRBPs . Apart from the less immunogenic PvRBP2cNB fragment , >20% and >62% of the children had reached >50% and >10% of the IgG1 levels seen in hyper-immune adults to the different PvRBPs . PvRBP1a , PvRBP2-P2 and PvRBP2a also had detectable levels of IgG3 , but only PvRBP2-P2 had a similarly high prevalence of IgG3 as for IgG1 ( 20 . 5% and 63 . 4% , respectively ) ( Table 1; S3 Fig ) , indicating that IgG1 antibodies were acquired faster than IgG3 . The predominance of IgG1 subclass is further highlighted by the generally stronger correlations of total IgG with IgG1 ( rho = 0 . 91–0 . 94 , P < 0 . 001 ) than IgG3 ( rho = 0 . 55–0 . 71 , P < 0 . 001 ) ( S1 Table ) . Despite the narrow age group in the cohort , there was a weak indication of polarization towards IgG3 with increasing age to PvRBP1a ( rho = -0 . 19 , P = 0 . 005 ) and PvRBP2-P2 ( rho = -0 . 16 , P = 0 . 015 ) , evidenced as a decrease in the IgG1/IgG3 ratio . Children had no detectable IgG2 or IgG4 to any of the proteins ( S3 Fig ) . IgG1 to all 6 PvRBPs ( P ≤ 0 . 003 ) were higher in infected children , although only moderately to PvRBP1b and PvRBP2cNB ( P = 0 . 06 ) . Similarly , those with a current infection had higher IgG3 to PvRBP1a , PvRBP2a and PvRBP2-P2 ( P ≤ 0 . 002 ) ( S2 Table ) . For the PvRBPs with dominant IgG1 subclass , the effect of age and exposure mostly mimic that observed for total IgG ( S2 Table ) . For PvRBP2-P2 and PvRBP1a IgG3 , but not IgG1 , still increased with age ( rho = 0 . 19–0 . 21 , P = 0 . 002–0 . 005 ) ( S2 Table ) . Over the 16 months follow-up of the PNG cohort , each child had an incidence rate of 1 . 25 ( 95%CI 1 . 08–1 . 45 ) P . vivax episodes/year at risk . Following adjustment for confounders , total IgG to all 6 PvRBPs tested were associated with protection against P . vivax malaria ( IRR 0 . 52–0 . 69 , P < 0 . 001–0 . 016 ) ( Fig 4; S3 Table ) . To further understand the contribution of antibodies to specific PvRBPs in the protective effect observed , we fitted a multivariate model to account for the fact that antibodies to the 6 PvRBPs were co-acquired and therefore highly correlated . In multivariate analysis , only total IgG to PvRBP1a ( IRRM 0 . 71 , 95%CI 0 . 53–0 . 94 , P = 0 . 019; IRRH 0 . 63 , 95%CI 0 . 44–0 . 88 , P = 0 . 008 ) , and PvRBP2b ( IRRM 0 . 73 , 95%CI 0 . 54–0 . 99 , P = 0 . 041; IRRH 0 . 63 , 95%CI 0 . 44–0 . 90 , P = 0 . 011 ) remained associated with reduced risk of vivax-malaria , suggesting that antibodies to these two PvRBPs are important correlates of naturally-acquired protective immunity . After adjusting for confounders , both IgG1 and IgG3 to PvRBP1a ( IgG1 IRR 0 . 48 , P < 0 . 001; IgG3 IRR 0 . 51–0 . 67 , P < 0 . 001–0 . 011 ) and PvRBP2a ( IgG1 IRR 0 . 66 , P = 0 . 010; IgG3 IRR 0 . 51 , P < 0 . 001–0 . 011 ) , and only IgG3 to PvRBP2-P2 ( IRR 0 . 60 , P = 0 . 002 ) were associated with reduced risk of vivax-malaria . IgG1 to PvRBP1b ( IRR 0 . 52 , P < 0 . 001 ) and PvRBP2b ( IRR 057–0 . 59 , P < 0 . 001 ) was also associated with protection . No association was found for the non-binding PvRBP2cNB fragment , although it was observed for total IgG levels in univariate analysis ( Fig 4; S3 Table ) . In a model combining both IgG1 and IgG3 levels to a given antigen , both IgG1 and IgG3 remained significantly associated with protection for PvRBP1a ( IgG1 IRRH 0 . 55 , 95%CI 0 . 38–0 . 80 , P = 0 . 001; IgG3 IRRM 0 . 69 , 95%CI 0 . 51–0 . 92 , P = 0 . 011 ) , and only IgG3 for PvRBP2a ( IRRH 0 . 51 0 . 36–0 . 72 , P < 0 . 001 ) and PvRBP2-P2 ( IRRH 0 . 60 , 95%CI 0 . 43–0 . 84 , P = 0 . 002 ) . When combining IgG1 and IgG3 to all PvRBPs in a multivariate model , however , only IgG1 to PvRBP1a ( IRRH 0 . 56 , 95%CI 0 . 39–0 . 80 , P = 0 . 001 ) and PvRBP2b ( IRRM 0 . 64 , 0 . 46–0 . 90 , P = 0 . 011; IRRH 0 . 69 , 95%CI 0 . 48–0 . 97 , P = 0 . 035 ) remained associated with reduced risk of vivax-malaria . There was a very strong association between increasing total IgG to the repertoire of PvRBPs , and increase in protection against vivax-malaria . For each increase in one unit of the breadth score ( see methods for detailed description ) , a reduction of approximately 8% in the incidence rate of P . vivax episodes was observed ( IRR 0 . 92 , 95%CI 0 . 89–0 . 96 , P < 0 . 001 ) . Considering the repertoire of IgG1 antibodies , the reduction in risk was of approximately 7% ( IRR 0 . 93 , 95%CI 0 . 90–0 . 97 , P < 0 . 001 ) . The effect of the IgG1 repertoire is no longer significant after IgG1 to PvRBP1a ( IRRH 0 . 53 , 95%CI 0 . 33–0 . 85 , P = 0 . 008 ) and PvRBP2b ( IRRM 0 . 63 , 95%CI 0 . 44–0 . 89 , P = 0 . 010 , IRRH 0 . 65 , 95%CI 0 . 42–1 . 00 , P = 0 . 053 ) is accounted for , again highlighting importance of these proteins in naturally-acquired protective immunity .
Advances in understanding P . vivax biology and the acquisition of immunity to this parasite , as well as the development of vaccines against P . vivax lag much behind what has been achieved for P . falciparum [14 , 26 , 27] . This is largely due the lack of a stable in vitro culture system for P . vivax that makes functional studies very challenging . In this study , we investigated whether six recombinantly expressed members of the PvRBP family are involved in P . vivax host-cell specificity by testing their erythrocyte-binding preferences . We identified that only PvRBP2b binds solely to reticulocytes . In previous reports , PvRBP1a and PvRBP2c have also been described as reticulocyte-specific binders [8] . Our recombinant PvRBP1a however binds preferentially to normocytes . One explanation for this observation is that C-terminal regions outside of the recombinant construct that is present on native protein governs reticulocyte specificity . In addition , PvRBP1a forms a complex with PvRBP2c [8] in parasites and this complex may be responsible for reticulocyte-binding . Unfortunately , as we were unsuccessful in expressing recombinant PvRBP2c with its binding domain , we were unable to confirm its erythrocyte-binding profile . The molecular mechanisms by which PvRBP2b mediates specific reticulocyte binding , and its reticulocyte-specific receptor are yet to be elucidated . Since our ability to do functional assays with P . vivax is constrained due to the lack of in vitro culture , we sought to investigate whether antibodies to the 6 PvRBPs are targeted by natural-immunity in a population of young children from PNG . The 6 PvRBPs were recognized differently . The PvRPB2cNB fragment had the lowest immunogenicity of all and antibodies to this fragment did not have a strong association with risk of vivax-malaria . This may be a consequence of the absence of the erythrocyte-binding region . The strongest protective effect was observed for total IgG to PvRBP1a and PvRBP2b . As there is poor cross-reactivity between antibodies targeting the binding regions of different PvRBPs , it is highly likely that these antibodies may have additive or even synergistic effects . IgG1 and IgG3 were the predominant IgG subclass to PvRBPs in PNG children . Interestingly , the dominant IgG subclass to PvRBP1a was different between children ( IgG1 ) and adults ( IgG3 ) . For PvRBP1a and PvRBP2-P2 , there was also some early evidence of switching to IgG3 with increasing age . Exposure to malaria parasites , among other antigenic and host characteristics , seems to play a major role in determining the predominant IgG response and for P . falciparum merozoite antigens , both age and transmission intensity have been previously associated with switching in predominance of the IgG subclass towards IgG3 [28 , 29] . Both IgG1 and IgG3 subclasses are cytophilic , T-cell dependent , and bind strongly to Fcγ , mediating phagocyte activation and complement fixation [30 , 31] , and predominance of IgG1 and/or IgG3 , in variable ratios , is common for several P . vivax [32–34] and P . falciparum merozoite proteins [35–37] . For the young PNG children included in this study , a reduction in risk of vivax-malaria was observed with IgG3 to PvRBP1a , PvRBP2a and PvRBP2-P2 but , ultimately , it was IgG1 to PvRBP1a and PvRBP2b that showed the strongest associations with protection . Adults also had detectable levels of non-cytophilic IgG2 to most PvRBPs tested however , the significance of this finding remains to be investigated . In P . falciparum IgG2 antibodies to EBA175 were shown to be short-lived [38] , but nevertheless correlated with lower parasitemia . High levels of IgG2 to RESA and MSP2 have also been associated with a lower risk of P . falciparum infection [39] , indicating that although uncommon , IgG2 antibodies might be important for immunity against malaria . Antibodies to PvRBP2b have also been previously associated with lower parasitaemia in clinical cases [11] . Both proteins seem to be under selective pressure and , in comparison to PvRBP1a , PvRPB2b is less polymorphic and with highly conserved regions , which may be beneficial for vaccine development [12 , 13] . Both PvRBP1a and PvRBP2b are less genetically diverse than PvRBP2c [12 , 13] . The existence of antigenic diversity in the different PvRBP genes however , has never been investigated . A study with Brazilian samples identified two regions of PvRBP1a ( aa 431–748 and 733–1407 ) as the most immunogenic with predominant IgG1 response , but the relationship between antibodies to the different regions and protection was not explored [18] . Further studies of different regions of both PvRBP1a and PvRBP2b molecules would be important to identify the main epitopes targeted by protective antibodies . The major limitation of this study is the lack of inclusion of a recombinant protein containing the binding domain of PvRBP2c , precluding both the investigation of the red-cell binding characteristics of PvRBPc and determining relative importance of antibodies targeting red-cell binding fragments of all main PvRBP proteins . Nevertheless , this is the first study where antibody levels to PvRBPs were investigated in samples from a well-designed longitudinal cohort study , which made it possible to adjust for other factors that confound the relationship between antibody acquisition and risk of disease , most importantly the heterogeneity in individual exposure to P . vivax blood stage infections [20] . The findings of this study provide further insight into P . vivax host-specificity and naturally-acquired immunity to PvRBPs in children . While the molecular functions of PvRBPs in P . vivax invasion are not well understood , the role of the PfRh family , which are homologs of PvRBPs in P . falciparum , have been well characterized in parasite invasion [3] . Several members of the PfRh family have been implicated in recognition of red blood cells , signaling events or creating a pore in the red blood cell membrane during invasion [40–44] . In particular PfRh5 has been a focus of intense research as a leading blood stage vaccine candidate due to its essential function in P . falciparum invasion [45 , 46] . Apart from gene structure and sequence homology , PvRBP2a and PfRh5 also adopt a similar structural fold within their erythrocyte-binding domains [10 , 47 , 48] . Using this structural scaffold with varied surface properties PvRBPs and PfRhs are able to mediate alternate receptor engagement . Monoclonal antibodies against PfRh5 results in the strong inhibition of parasite growth across multiple strains and Aotus nancymaae monkeys immunized with anti-PfRh5 vaccine are protected against severe infection [42 , 49 , 50] . It is likely that the erythrocyte-binding domain of PvRBPs will be able to elicit antibodies that have the ability to block P . vivax invasion . Our results underline the key role of PvRBPs in parasite–host interactions and highlight their potential as P . vivax vaccine candidate antigens . Further immuno-epidemiological studies in broader age groups from areas of different transmission intensities , as well as functional studies in vitro or in animal models , and a better understanding the molecular function of all PvRBPs , including the PvRBP1a/PvRBP2c complex in P . vivax invasion are now required to validate and prioritize one or several PvRBPs for development as vaccine candidates . | In parallel with the tremendous reduction in malaria burden , Plasmodium vivax ( Pv ) is now the predominant malaria species in the Asia-Pacific and Americas . Pv can only invade young erythrocytes ( reticulocytes ) and this restriction is thought to involve the Reticulocyte-Binding Protein family ( PvRBP ) . Given their predicted role , PvRBPs are potentially interesting vaccine targets . However , the acquisition of immunity to Pv in general ( PvRBPs in particular ) is poorly understood , hindering vaccine development . Here , we show that out of five PvRBPs , only one ( PvRBP2b ) binds exclusively to reticulocytes . Furthermore , we measured antibody levels to all six PvRBPs in a cohort of young Papua New Guinean children , assessing the relationship between antibodies to PvRBPs and risk of malaria disease . Both total and specific antibody subclass levels ( IgG1 and IgG3 ) to the reticulocyte-specific binder PvRBP2b , and the non-specific binder PvRBP1a were strongly associated with lower risk of clinical disease . Our findings indicate a diversity of roles of PvRBPs in erythrocyte invasion and highlight their importance as targets of the naturally acquired immunity to Pv . Functional studies of the role of PvRBPs in reticulocyte invasion will be required to fully understand the potential of PvRBP1a and PvRBP2b as vaccine candidates . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"blood",
"cells",
"children",
"parasite",
"groups",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"enzyme-linked",
"immunoassays",
"plasmodium",
"immunology",
"tropical",
"diseases",
"geographical",
"locations",
"parasitic",
"diseases",
"reticulocytes",
... | 2016 | Plasmodium vivax Reticulocyte Binding Proteins Are Key Targets of Naturally Acquired Immunity in Young Papua New Guinean Children |
Discovery of rare or low frequency variants in exome or genome data that are associated with complex traits often will require use of very large sample sizes to achieve adequate statistical power . For a fixed sample size , sequencing of individuals sampled from the tails of a phenotype distribution ( i . e . , extreme phenotypes design ) maximizes power and this approach was recently validated empirically with the discovery of variants in DCTN4 that influence the natural history of P . aeruginosa airway infection in persons with cystic fibrosis ( CF; MIM219700 ) . The increasing availability of large exome/genome sequence datasets that serve as proxies for population-based controls affords the opportunity to test an alternative , potentially more powerful and generalizable strategy , in which the frequency of rare variants in a single extreme phenotypic group is compared to a control group ( i . e . , extreme phenotype vs . control population design ) . As proof-of-principle , we applied this approach to search for variants associated with risk for age-of-onset of chronic P . aeruginosa airway infection among individuals with CF and identified variants in CAV2 and TMC6 that were significantly associated with group status . These results were validated using a large , prospective , longitudinal CF cohort and confirmed a significant association of a variant in CAV2 with increased age-of-onset of P . aeruginosa airway infection ( hazard ratio = 0 . 48 , 95% CI=[0 . 32 , 0 . 88] ) and variants in TMC6 with diminished age-of-onset of P . aeruginosa airway infection ( HR = 5 . 4 , 95% CI=[2 . 2 , 13 . 5] ) A strong interaction between CAV2 and TMC6 variants was observed ( HR=12 . 1 , 95% CI=[3 . 8 , 39] ) for children with the deleterious TMC6 variant and without the CAV2 protective variant . Neither gene showed a significant association using an extreme phenotypes design , and conditions for which the power of an extreme phenotype vs . control population design was greater than that for the extreme phenotypes design were explored .
Cystic fibrosis ( CF ) ( MIM219700 ) is a life shortening , monogenic condition caused by mutations in cystic fibrosis transmembrane conductance regulator ( CFTR ) [1] . CF is associated with dysfunction in multiple exocrine organs ( e . g . pancreas , intestine , liver , lung ) , but the primary cause of morbidity and mortality is progressive obstructive lung disease associated with persistent endobronchial infection and neutrophilic inflammation [2 , 3] . Specifically , individuals with cystic fibrosis are at high risk for Pseudomonas aeruginosa ( P . aeruginosa ) airway infection , with a typical overall pattern of progression from first acquisition , to more frequent infections , to chronic infection and then to chronic mucoid P . aeruginosa . Eventually , the airways of ~80% of adult patients are chronically infected with P . aeruginosa . P . aeruginosa airway infection in individuals with CF is associated with increased morbidity and reduced survival [4–6] . In a large registry-based study , any history of P . aeruginosa airway infection at or before age 2 years was associated with reduced lung function , increased frequency of pulmonary exacerbations and reduced survival [4–6] . Onset of mucoid P . aeruginosa is associated with a sharp decline in lung function [7 , 8] and even higher mortality [9] . Collectively , these observations support a causal relationship between frequent P . aeruginosa infection and morbidity and mortality in CF . Current CF treatment guidelines focus on early detection and attempted eradication of P . aeruginosa airway infection prior to establishment of chronic infection [10] . Aggressive acute and maintenance antibiotic treatment of P . aeruginosa infection in individuals with CF in the U . S . has coincided with a rise in the median predicted survival from 28 to 38 years [11] . The age-of-onset of chronic P . aeruginosa infection of the lungs in individuals with CF varies widely , and the heritability for age-of-onset of chronic/persistent P . aeruginosa infection of the airway , independent of CFTR genotype , has been estimated to be 0 . 85 ( on a scale of 0 to 1 ) [12] . Identification of host genetic factors could help delineate sub-populations of individuals for more aggressive monitoring and treatment; identify new targets to facilitate development of therapeutics; and lead to a better understanding of the pathophysiology of P . aeruginosa infection in CF . Additionally , finding modifiers of P . aeruginosa airway infection in CF could provide a model for studies to find disease modifiers of other Mendelian conditions and perhaps for complex traits , as well . To this end , an intense search to find genetic modifiers of airway P . aeruginosa infection in CF has been underway over the past decade . Variants in MBL2 and SCL9A3 have been associated with first acquisition of airway P . aeruginosa in children with CF [13] , [14] , and nominally significant associations between age-of-onset of chronic P . aeruginosa infection have been reported recently for other lectin pathway genes: FNC1 , FNC2 and MASP3 [15] . More recently , we used exome sequencing and an extreme phenotypes design to discover two variants in DCTN4 associated with early onset of chronic P . aeruginosa airway infection [16] . However , even collectively , these variants explain only a small fraction of the genetic variance of risk for P . aeruginosa airway infection , indicating that other genetic modifiers of P . aeruginosa airway infection remain to be found . To find variants/pathways that influence risk of early chronic P . aeruginosa infection in individuals with cystic fibrosis , we used a study design in which a single phenotypic extreme is compared to a control population . In contrast , in an extreme phenotypes study design , frequencies of variants in the two opposite s of the same phenotype in a study population are directly compared . In the extreme phenotypes design , power depends both on the group sizes and on the difference in variant frequencies between extremes ( i . e . , the empirical effect size ) . However under certain genetic architectures , most or all of the effect size can be due to enrichment ( or depletion ) of casual variants in only one extreme relative to the entire study population . In such cases , comparison of the single enriched ( or depleted ) extreme to the entire study population will increase statistical power . We sampled CF individuals from each of two extremes of phenotype , generated exome data for each sample , and then compared each extreme separately to a large control exome data set in order to discover genetic variants associated with the phenotypic expression in each CF sample . One extreme comprised 85 individuals with cystic fibrosis and extreme early onset chronic P . aeruginosa infection; the other comprised 65 individuals with extreme late onset P . aeruginosa ( S1 Fig ) . Study individuals were drawn from the Early Pseudomonas Infection Control ( EPIC ) Observational Study and the Genetic Modifier Study ( GMS; a member of the North American CF Gene Modifier Consortium [17] , [18] ) . The large control data set included 3 , 239 individuals of European ancestry who were ascertained to study non-lung disease phenotypes as part of the NHLBI Exome Sequencing Project ( ESP ) . In both the extreme early onset P . aeruginosa vs . control population and the extreme late onset P . aeruginosa vs . control population comparisons , we discovered a gene associated with time to chronic P . aeruginosa infection in individuals with CF . Specifically , we found a variant in CAV2 , encoding caveolin-2 , was associated with increased age-of-onset ( i . e . , protective ) and variants in TMC6 , encoding transmembrane-like channel 6 were associated with decreased age-of-onset ( i . e . , deleterious ) . Both of these results were subsequently replicated in a large cohort of independent samples from the Early Pseudomonas Infection Control ( EPIC ) [19] study cohort . Furthermore , we demonstrate that the statistical power to find both genes was greater under the single extreme vs . control population design than under the extreme phenotypes design . Together , these findings provide proof-of-principle that use of exome sequencing and a single extreme phenotype vs . control population design can identify genetic modifiers of Mendelian conditions and may be generalizable to the search for rare variants underlying complex , common diseases .
A per-gene comparison between the extreme early onset chronic P . aeruginosa group ( n = 85 ) and the control group ( n = 3 , 239 ) using the optimal sequence kernel association test with small sample adjustment ( aSKAT-O ) [20] revealed significant differences between the frequencies of variants in two genes , CFTR ( p<10–16 ) and CAV2 ( p = 1 . 1x10-6 ) ( Fig 1A ) . Examination of the eight ( nonsynonymous ) variants included in the CAV2 by-gene test showed that the aSKAT-O signal was likely due entirely to imbalance in frequency between groups for the single common variant ( rs8940; p . ( Q130E ) , MAF = 0 . 19 in European Americans ( EA ) per the NHLBI Exome Variant Server ) ; the seven other variants had a cumulative MAF < 0 . 2% . This prediction was confirmed by a per-variant analysis in which the result for rs8940 replicated the by-gene result for CAV2 , p = 6 . 7 x 10–7 , for rs8940 , which occurs at a highly conserved site ( Genetic Evolutionary Rate Profiling ( GERP ) score = 5 . 9 ) . We attempted to validate the association between early age-of-onset of chronic P . aeruginosa and CAV2 rs8940 by screening an additional 643 unrelated EPIC participants using the Infinium Human Exome BeadChip and manual genotyping of rs8940 among individuals for whom the exome chip genotyping was not performed ( S1 Table ) . Association between the presence of at least one rs8940 minor allele and age-of-onset of chronic P . aeruginosa infection was tested using the Cox proportion hazards model stratified on age at entry to the EPIC study [16] The model was adjusted for CFTR risk group [21] , number of bacterial cultures tested for each individual , presence of a DCTN4 risk allele [16] , age at enrollment and its interaction with rs8940 . This model revealed a significant inverse association ( p = 0 . 01; HR = 0 . 53 , 95% CI = [0 . 32–0 . 88] ) between age-of-onset of chronic P . aeruginosa infection and presence of at least one minor allele of rs8940 ( Fig 1B and S2 Table ) . This effect was stronger among younger enrollees ( p = 0 . 01 for rs8940 x enrollment age interaction . ) Similar results were obtained after adjustment for ancestry using principal components ( PCs ) on the subset of 608 individuals with exome or genome-wide genotype chip data available ( p = 0 . 0054; HR = 0 . 48 , 95% CI = [0 . 29 , 0 . 81] ) ( S2 Table ) . CAV2 is located ~1 Mb from CFTR and higher than expected frequencies of CAV2 rs8940 alternate alleles were found in individuals with various CFTR mutations that cause CF , including F508del-CFTR , N1303K , G542X and W1282X ( S3 Table ) , resulting in an association between CFTR mutation and CAV2 rs8940 . Because age of acquisition and onset of chronic P . aeruginosa infection in CF might vary by CFTR mutation , and because CFTR mutations were associated with CAV2 rs8940 , there is potential for the association between rs8940 and time to chronic P . aeruginosa infection to be due to confounding . To remove any possible confounding by CFTR mutation , we repeated the Cox model analysis restricted to individuals homozygous for F508del-CFTR ( n = 312 ) ; the results were essentially unchanged ( p = 0 . 027; HR = 0 . 44 , 95% CI = [0 . 21–0 . 91] ) . Even stronger results were obtained when the analysis of F508del-CFTR homozygotes was restricted to the individuals with self-declared European ancestry and adjusted for PCs ( n = 286 ) ( p = 0 . 0087; HR = 0 . 35 , 95%CI = [0 . 16–0 . 77] ) ( S2 Table ) , but the difference in HR among the F508del-CFTR homozygous group and remaining individuals did not reach statistical significance ( p = 0 . 13 ) . These findings support a strong association between presence of the CAV2 rs8940 alternate allele and protection against earlier and more severe P . aeruginosa infection , an association that is independent of the CFTR mutation . We next performed a per-gene comparison between the extreme late onset chronic P . aeruginosa airway infection extreme ( n = 65 ) and the ESP control group above using aSKAT-O and found that the frequencies of variants in CFTR ( p<1x10-10 ) and TMC6 ( p = 9 . 5x10-7 ) differed significantly between the two groups ( Fig 2A ) . Examination of variants in TMC6 in the late onset extreme vs . control group revealed a complex pattern of differences among the 36 non-synonymous variants contributing to the by-gene test ( S4 Table ) . Specifically , the late onset group had fewer rare variants ( cumulative observed MAF = 0 . 007 vs . 0 . 014 ) , a higher observed MAF for two ( rs12449858 , rs2748427 ) of three common variants ( rs34712518 , rs12449858 , rs2748427 with MAFs of 0 . 06 , 0 . 09 , and 0 . 19 per EVS , respectively ) in TMC6 and a lower derived allele frequency at rs3471258 . Based on these observations , we next performed a validation analysis in which we collapsed all of the rare variants into one score ( i . e . , TMC6 rare variant score ) while the three common variants were analyzed individually . Thus we performed a total of four tests in the validation analysis using the Cox model as above for the CAV2 validation . The TMC6 validation analysis was performed using data from 580 unrelated EPIC participants genotyped using the Illumina Exome chip . The presence of one or more derived alleles at rs34712518 was significantly associated with age-of-onset of chronic P . aeruginosa infection in the primary test of association ( p = 0 . 012 , HR = 1 . 8 , 95% CI = [1 . 3–2 . 8] , p< 0 . 05 after correction for the four tests ) . The Kaplan Meier plot comparing age-of-onset of chronic P . aeruginosa between TMC6 rs34712518 variant groups showed a striking divergence in risk with progressing age ( Fig 2B ) , indicating the hazard ratio is dependent on age ( i . e . the hazards are not proportional over ages ) and that an age-allele interaction should be fitted in the Cox model for a more accurate description of the effect . At age 7 years , the HR was 4 . 5 ( p = 0 . 01 , [1 . 4–12 . 6] ) , and at 10 years , the HR was 5 . 2 ( p = 0 . 00046 , [2 . 1–43 . 0] ) ( S5 Table ) . Variant rs12449858 was not significantly associated with age-of-onset of chronic P . aeruginosa ( p = 0 . 12 , HR = 0 . 7 , 95% CI = [0 . 44 , 1 . 1] ) , nor was the collapsed rare variant score ( p = 0 . 10 , HR = 0 . 38 , 95% CI = [0 . 12–1 . 2] ) . The derived allele of TMC6 rs2748427 also was marginally associated with age-of-onset of chronic P . aeruginosa infection ( p = 0 . 027 , HR = 1 . 4 95% CI = [1 . 03–1 . 8] , p = 0 . 08 after multiple-test correction ) . However , there was a strong association between the derived allele counts of rs3412518 and rs2748427 ( Pearson correlation = 0 . 5; p<2x10-16 ) , and individuals with a derived rs2748427 allele but not a derived rs34712518 allele showed no independent risk of chronic P . aeruginosa infection ( HR = 1 . 18 , p = 0 . 29 , [0 . 86–1 . 6] ) . This suggests that the variant with a lower MAF , rs34712518 ( MAF = 0 . 06; p . ( G191D ) ) is associated with higher risk while the more common variant , rs2748427 , is not independently associated with increased risk . The result was similar after adjusting for PCs and limiting the analysis to F508del-CFTR homozygotes ( S5 Table ) . There were no significant interactions with enrollment age nor with chronological age in either case . We tested for interactions between risk variants found in DCTN4 , CAV2 , and TMC6 . No significant interaction was identified between DCTN4 and either CAV2 or TMC6 . In contrast , we did find a significant interaction between the derived rs3412518 allele in TMC6 and the rs8920 CAV2 allele ( p = 0 . 007 ) . Specifically , by age 10 , children with the derived rs3412518 allele in TMC6 without the protective CAV2 allele were at much higher risk of chronic P . aeruginosa infection ( p = 2 . 6x10-5 , HR = 12 . 1 , [3 . 8–38 . 8] ( Fig 2C ) , while children with the derived rs3412518 allele in TMC6 with the protective CAV2 variant had no significant excess risk of chronic P . aeruginosa infection at this same age ( p = 0 . 41 , HR = 2 . 1 , [0 . 38–11 . 3] ) ( Fig 2D and S5 Table ) . When the test for the TMC6-CAV2 interaction was limited to the F508del-CFTR group , the test was again significant ( p = 0 . 03 ) with an estimated HR = 29 . 1 at age 10 for F508del-CFTR homozygous children with both the TMC6 and CAV2 deleterious alleles ( 95% CI = [4 . 1 , 204 . 8] , n = 129 , S3 Fig ) compared to HR = 2 . 6 ( 95% CI = [0 . 3 , 23 . 1] , n = 144 ) for those with the TMC6 deleterious allele and a protective ( derived ) CAV2 allele . Because acquisition of P . aeruginosa is associated with worse lung function over time , strong modifiers of P . aeruginosa infection may also be associated with lung function . Accordingly , we tested for association between CAV2 rs8940 and TMC6 rs34712518 cystic fibrosis-specific percentiles of forced expiratory volume at 1 second ( FEV1 ) . Among 572 children with lung function data beyond age 7 years , an age at which lung function studies are considered reliable , the presence of one or more CAV2 rs8940 derived alleles was associated with a 5 . 5 percentile point increase in lung function ( p = 0 . 001 ) ( Fig 3A ) , while presence of one or more TMC6 rs34712518 alleles was associated with an 8 . 0 percentile point decrease in CF-specific FEV1 ( p = 0 . 01 ) ( Fig 3B ) . Both of these effects were in the direction predicted based on their association with time to chronic P . aeruginosa infection . No interaction effect between variants was detected for the FEV1 outcome . We sought to understand to what extent discovery of these associations statistical power afforded by using the single extreme vs . control population design instead of an extreme phenotypes design . We performed a detailed power study for TMC6 , using simulations based on the observed association between TMC6 and age-of-onset of chronic P . aeruginosa airway infection in the validation data set . The results not only show explicit power gains but the methods also serve as a model for power calculations for other studies . Specifically , the distributions of event times for CF individuals with and without the rs3412518 alternate allele were well described by different Weibull distributions ( Fig 4A ) . The Weibull distribution is widely used in time-to-event analysis because of its flexibility and interpretability [22] . We calculated the expected frequency of the TMC6 rs3412518 alternate allele as a function of age among children who had not yet reached chronic P . aeruginosa airway infection by that age ( Fig 4B ) . The results show a striking depletion of the rs3412518 alternate allele among CF individuals who were older and free of chronic P . aeruginosa airway infection . This is expected , intuitively , for a modestly rare allele that confers high risk: individuals carrying the alternate allele will succumb earlier to chronic P . aeruginosa airway infection and will be under-represented among individuals free of P . aeruginosa at older ages . The probability of carrying a TMC6 rs3412518 alternate allele can be calculated for children who became chronically infected with P . aeruginosa at a particular age to obtain the TMC6 allele distributions among our extreme samples for the observed ages in each extreme group ( using the same methods that were used to generate Fig 4B ) . New samples , conditional on the observed ages , were drawn from the Weibull-based distributions , and the power of aSKAT-O to detect a difference between groups was calculated for our sample sizes . The result was an approximately 40% increase in power across a range of test sizes for the single extreme phenotype vs . control design relative to the extreme phenotypes design for TMC6 rs3412518 ( Fig 5A ) . We also calculated the power based on a sample size of 150 ( i . e . , all available exomes originating from one extreme ) in the late onset chronic P . aeruginosa airway infection extreme vs . population control and observed exceptional power for the single extreme vs . control design ( 99% ) when all exomes were devoted to the single extreme ( Fig 5A , red line ) . This is consistent with the increasing differences in the population MAF and the allele frequency in the late age-of-onset extreme ( Fig 4B , red dotted line versus blue line , respectively . ) In a separate power analysis , we drew samples from the observed distributions of variants over TMC6 ( S4 Table ) to better represent a by-gene analysis . Power gains were even greater under this scenario for the single extreme vs . control design relative to the extreme phenotypes design ( S5B Fig ) . Doubling the sample size of the extreme in a single extreme vs . control design provides greater power than the extreme phenotypes design even when the allele frequency of the causal variant in both extremes differs from the population frequency .
We used a single extreme phenotype vs . control population study design to discover variants in TMC6 and CAV2 associated with age-of-onset of chronic P . aeruginosa airway infection in individuals with CF . Examination of the variants within CAV2 and TMC6 allowed us to tailor our validation analysis to provide additional information on the role of specific variants that could not be uncovered in the by-gene discovery analysis . We found that CAV2 rs8940 was associated with protection against chronic P . aeruginosa airway infection and TMC6 rs34712518 was associated with earlier age-of-onset of chronic P . aeruginosa airway infection . Among individuals with both variants in the validation set , the protective effect of CAV2 rs8940 nullified the deleterious effect of TMC6 rs34712518 ( i . e . , a significant interaction was present rather than an additive effect ) . Consistent with these results , we also found a significant difference in lung function between children with and without CAV2 rs8940 and TMC6 rs34712518 . Notably , the FEV effect size for TMC6 rs34712518 is of the same magnitude as that for risk group 1 vs . risk group 2 CFTR genotype groups in this cohort ( no functional CFTR versus some residual function , respectively , in risk groups 1 and 2 [21] ) . Together , these findings suggest that both CAV2 and TMC6 are strong modifiers of age-of-onset of chronic P . aeruginosa airway infection in persons with CF . Neither of these discoveries were possible using the extreme phenotypes ( extreme vs . extreme ) analysis with our samples , with p-values for both genes being greater than 0 . 001 under that design . The observation that CAV2 is a modifier of age-of-onset of chronic P . aeruginosa airway infection in individuals with CF is strongly supported by functional studies of caveolin-2 and its dimeric complement , caveolin-1 . Caveolin-2 is one of a family of three structural proteins of caveolae , flask-shaped invaginations on the surface of lung epithelium and myeloid cells ( Zaas 2009–1 ) [23] . Caveolins contribute to host defenses by regulating lipid-raft-mediated endocytosis and play a substantial role in the inflammatory response to infection and protein trafficking in the lung epithelium of individuals with CF [24 , 25] . In models of invasive infection , P . aeruginosa co-opts the host endocytotic machinery of lung epithelial cells where it replicates . In a variety of human cell lines , caveolin-2 forms a heterodimer with caveolin-1 , which after infection with P . aeruginosa , co-localizes with CFTR and P . aeruginosa [26] . Knock-out of CAV2 in murine epithelial cells prevents P . aeruginosa invasion [27] . Notably , CAV2 is among the genes identified by GeneGO in conjunction with the Cystic Fibrosis Foundation in the MetaMiner-CF database and network tool as potential modifier genes in CF based on its role in endosome formation in lung epithelial cells [28] . PolyPhen2 classifies the CAV2 rs8940 variant as “probably damaging , ” with a score of 0 . 997 , while the Combined Annotation-Dependent Depletion ( CADD ) score puts this variant among the 2% of SNVs most likely to be pathogenic [29] ( CADD-PHRED score = 17 . 1 ) ( http://cadd . gs . washington . edu/download , V1 . 1 ) . The role of lung epithelial cell invasion in chronic P . aeruginosa airway infection in persons with CF remains unknown [30] . However , invasion of host epithelial cells at some point in the development of chronic P . aeruginosa airway infection is suggested by the presence of anti-Pseudomonal antibodies early in life among persons with CF infected with P . aeruginosa [31] and a spike in anti-P . aeruginosa antibody titers in persons with CF prior to diagnosis of chronic P . aeruginosa airway infection [7 , 32] . This suggests that variants in CAV2 might reduce early P . aeruginosa invasion of lung epithelia , thereby reducing the risk of chronic P . aeruginosa airway infection . Alternatively , co-expression of caveolin-2 with F508del-CFTR in murine cell lines can rescue mutant CFTR expression on the cell surface [33] . Isoforms of caveolin-2 might rescue mis-folded CFTR proteins ( e . g . , F508del-CFTR ) resulting in increased cell surface expression of mutant CFTR and improved lung defenses against P . aeruginosa . While the mechanism ( s ) by which CAV2 rs8940 influences risk of chronic P . aeruginosa airway infection in persons with CF remains to be determined , our results are consistent with the known functional roles of caveolins in response to P . aeruginosa infection in CF models and highlight caveolin-2 as a potential therapeutic target to block P . aeruginosa invasion . TMC6 encodes a highly conserved , integral transmembrane protein that interacts with zinc transporter 1 to influence intracellular zinc concentrations . Mutations in TMC6 underlie epidermodysplasia verruciformis , a condition in which persons are unusually sensitive to development of skin lesions caused by invasion human papilloma virus ( HPV ) [34] . One postulated mechanism is that variants in TMC6 dysregulate HPV replication via an effect on AP-1 transcription [35] . AP-1 plays a cooperative role in the P . aeruginosa-dependent induction of IL-8 , a key mediator in human bronchial epithelial cells and the lung pathology of persons with CF [36] . In addition , an AP-1 transcription factor binding site is located in the promoter of DEFB1 , which encodes beta-defensin-2 [37] , another important cytokine involved in protection against P . aeruginosa in persons with CF . Accordingly , variants in TMC6 may influence the regulation of AP-1 and its downstream effectors , ultimately modifying the response to P . aeruginosa . Moreover , HPV gains entry to cells via both clathrin-dependent and caveolin-dependent endocytosis [38 , 39] . TMC6 could affect HPV invasion at the point of caveolin-mediated endocytosis , functionally linking TMC6 and CAV2 and providing a basis for the observed interaction between TMC6 and CAV2 on risk of chronic P . aeruginosa airway infection . These results provide a compelling reason to pursue functional studies of the associated variants . However , it should be kept in mind that causal variants may in LD with the discovery variants . We identified a strong interaction between risk variants in CAV2 and TMC6 . This finding highlights two considerations for studies aimed at identifying risk variants for complex traits . First , to increase the power of association studies for novel risk variants , it might be necessary to test for interactions with known genetic risk variants . This is feasible from a multiple-testing standpoint , given the relatively small numbers of known risk variants . Second , the possibility that a substantial portion of the “missing heritability” might be due to interactions among risk variants should emphasize recent attention to development and use of methods for identifying significant interactions in the face of multiple-testing in discovery analyses , a difficult but important problem . We used a single extreme phenotype vs . control population design to gain power over an extreme phenotypes design to identify genes with differences in variant distributions between groups . Our power studies based on modeling of observed distributions for variants in TMC6 revealed that use of this strategy resulted in substantial power gains over the extreme phenotypes design . The power gain for identifying the association between age-of-onset of chronic P . aeruginosa airway infection and TMC6 variants derives from the observation that the late onset chronic P . aeruginosa airway extreme was depleted of TMC6 risk variants , while the frequency of these risk variants in the extreme early onset chronic P . aeruginosa airway infection did not differ substantially from the control population . Indeed , since the TMC6 risk variant has little effect until age 5 to 7 years , selecting children with age-of-onset before age 7 years did not enrich for this risk variant . That risk variants were depleted in the population of “survivors” ( children with later age-of-onset in this setting ) is an important consideration for studies to discover risk variants for childhood-onset conditions such as CF because the direct implication is that cohorts of adults can be underpowered to detect risk variants that affect the phenotype early in life . The relative power of a single extreme phenotype vs . control population compared to an extreme phenotypes design will depend on the genetic architecture underlying the phenotype studied . In either design , selecting extreme individuals in at least one “case” arm of the study will result in enrichment of causal variants in that arm relative to both population controls and to the opposite extreme , a fact that can be demonstrated by Bayes Theorem , but the degree of relative enrichment ( effect size ) might differ between designs . Given N samples in each extreme , the extreme phenotypes design will usually be more powerful than a single extreme phenotype vs . control for situations in which both extremes exhibit differences in variant distributions relative to population controls , regardless of how large the sample size is for the control population . This is due to the diminishing “rate of return” in power as the control population sample size increases while the “case” sample size stays fixed at N . On the other hand , we found no realistic situation in which an extreme phenotypes design with N samples per arm had better power than a 2N single extreme vs . control population design . The implication is that investigators with access to control exomes/genomes might elect to allocate resources entirely to one phenotypic extreme so as to maximize power . This is fortuitous because only one extreme can be defined well or is of greater biomedical interest for many conditions ( e . g . , high blood pressure , risk of early stroke , etc . ) . Consequently , we think the single extreme phenotype vs . control population design is likely to be a desirable approach , particularly for adult-onset common diseases . Generating exome/genome data from large control populations is expensive and as additional exomes/genomes become publically available , investigators will increasingly rely on public repositories of sequence data ( e . g . , dbGaP ) for discovery studies . Such studies should always be validated with appropriate independent samples , but attention should still be given to avoiding false positives at the discovery stage . Our extreme and control exomes were sequenced contemporaneously at the same lab , but factors such as differences in targets , sequencing platforms , variant calling , etc . can introduce bias ( i . e . , batch effects ) that result in confounding . Correcting for these confounders is critical and results based on the use of such population controls that lack explicit description of how these were managed should be interpreted with caution . Extensive considerations on the use sequence data from population controls are provided by Dercach , et al [40] , along with methods for potential bias correction . Additionally , the large imbalance in samples sizes of the groups compared in the extreme phenotype vs . control population design must be considered . With very different sample sizes , tests that are based on the mean and variance of the test statistic for large samples ( “second order asymptotics” , which is by far the most common measure of operating characteristics of a test ) can fail to provide reliable p-values with test sizes that are too large ( p-values that are too small ) . The aSKAT-O test with its “small sample adjustment” provided a correction for this problem . Permutation tests , including Fisher’s exact test , also retain reliability with imbalanced sample sizes but suffer from lower power , challenges with covariate adjustment and difficulty in computational implementation over the entire exome compared to aSKAT-O . Nevertheless , as data from thousands of exomes / genomes that can serve as controls become available to the scientific community , we think the extreme phenotype vs . population controls will become an increasingly popular study design .
This research was approved by the Seattle Children's Hospital IRB ( approval numbers 12974 and 11686 ) . All study participants provided written informed consent or assent for the use of their DNA in studies aimed at identifying genetic risk variants for cystic fibrosis ( CF ) and for broad data sharing . Institutional certification was obtained for each sample in which exome sequencing was performed to allow deposition of phenotype and genotype data in the database for Genotypes and Phenotypes ( dbGaP ) and BAM files in the NCBI short-read archive . Detailed methods are provided in Emond et al [16] . Single nucleotide variants ( SNVs ) were called using the UMAKE pipeline at University of Michigan , which allowed all samples to be analyzed simultaneously , both for variant calling and filtering . BAM files were summarized using BWA , refined by duplicate removal , recalibration , and indel re-alignment . We excluded all reads that were not confidently mapped ( Phred-scaled mapping quality < 20 ) from further analysis . To avoid PCR artifacts , we clipped overlapping ends in paired reads . We then computed genotype likelihoods for exome targeted regions and 50 flanking bases , accounting for per base alignment quality ( BAQ ) using samtools [41] . Variable sites and their allele frequencies were identified using glfMultiples , and a support vector machine ( SVM ) classifier was used flag probable false-positive variant sites . SNVs at HapMap polymorphic sites and Omni 2 . 5 array polymorphic sites in the 1000 Genomes project data were flagged as likely true positives . A total of 1 , 908 , 614 SNVs passed the SVM filter , with an overall transversion to transition ratio ( Ts/Tv ) of 2 . 84 . After the initial SNV calls were generated , we re-examined the VCF files and applied filters considering total read depth , the number of individuals with coverage at the site , the fraction of variant reads in each heterozygote , the ratio of forward and reverse strand reads for reads carrying reference and variant alleles , and the average position of variant alleles along a read . CFTR genotypes from the exome sequencing results were in agreement with the clinical genotype data . The definition of chronic P . aeruginosa infection was chosen to parallel the “Leeds” criterion suggested by Lee et al . within the sampling frame of the EPIC study . Routine bacterial cultures were scheduled to be taken during quarterly visits ( once every 3-months ) . We defined an individual to have reached the chronic endpoint if s/he had at least two positive P . aeruginosa culture-quarters during any one-year period . Individuals in the analysis set had a median of 3 . 5 culture-quarters per year . Individuals with extreme phenotypes were selected from the EPIC study , which was open to CF-affected individuals of any CFTR genotype , and from GMS , which included only individuals with the F508del-CFTR homozygous genotype . The early age-of-onset group comprised all individuals of European ancestry in either EPIC or GMS who had consented/assented for deposition of exome data to dbGaP and who had reached the chronic infection endpoint by age 7 at the time of exome sequencing ( ancestry was determined using principal components analysis ) . Age 7 years demarks the earliest quartile of onset ages among EPIC DNA Collection study individuals ( median age-of-onset of chronic P . aeruginosa airway infection = 14 . 3 years among the 1322 participants from EPIC DNA Collection study ) . Four individuals with age-of-onset later than 7 years but less than the median were included in the early onset extreme in order to make use of available sequencing , resulting in a total of 11 early onset individuals from GMS and 75 from EPIC . The median age-of-onset among the early extreme was 2 . 6 years . EPIC individuals for the late onset extreme ( n = 46 successful exomes ) were selected from among the oldest individuals who were still free of chronic P . aeruginosa airway infection at the time of selection of the exome sequencing sample . The 19 individuals selected for the late onset extreme from the GMS were free of chronic P . aeruginosa airway infection until at least age 18 years , with one person free until age 58 years . One person in the late onset extreme became chronic at age 14 . 4 years after being selected for sequencing; otherwise , no person in the late onset extreme became chronically infected before age 21 . The number of quarters with positive P . aeruginosa culture results among the early onset extreme was 70 times greater than that for the late onset extreme ( 905 vs 13 ) despite the longer observation times for the latter . The percentage of individuals in the early and late extremes who were identified by newborn screening ( NBS ) was 38 . 9% and 6 . 5% , respectively , compared to 21 . 6% for the EPIC Observational cohort overall [42] ) . Because NBS has a protective effect on acquisition of Pa [42] , a larger percent of NBS individuals in the early-onset extreme makes this sample even more extreme in terms of Pa risk than it would be with a smaller percentage of NBS-identified individuals . Pancreatic enzyme use was similar in both extremes ( 93 . 4% early onset individuals ever used pancreatic enzymes , compared to 94 . 6% of the late onset individuals . ) The adjusted SKAT-O method described Lee et al . [20] was used to obtain p-values for each gene . Non-synonymous variants annotated to a given gene according to the RefSeq/NCBI 37/Hg19 model were included in each by-gene test . The small-sample adjustment was critical for the discovery analysis , as the imbalance in sample sizes resulted in marked overdispersion of the qq-plots ( spuriously low p-values; S2 Fig ) for tests without a small sample correction . Each by-gene test was adjusted for PC1 , PC2 and PC3 from a principal components decomposition of the entire set of exome data after applying the QC filters described above; any control group individuals that could be potential ancestral outliers were liberally trimmed using smartPCA from the EIGENSOFT package ( http://www . hsph . harvard . edu/alkes-price/software/ ) . Powers for the extreme phenotypes design ( n = 65 vs . n = 85 ) and the extreme phenotype vs . control large set design were calculated prior to the analysis assuming an elevated frequency of risk variants in one extreme and not the other; powers specific to TMC6 were calculated post hoc by modeling the time-to-event distributions with Weibull distributions and by using the observed MAFs of variants in TMC6 . Censored data methods ( Kaplan-Meier survival curves and the Cox proportional hazards ( PH ) model ) were used to in the validation analysis to estimate age-of-onset of chronic P . aeruginosa airway infection and to test for differences between genetic groups in the associated hazard ratio ( HR ) for onset of chronic P . aeruginosa infection . The Cox PH model test has maximal statistical power when hazards are proportional , but the test size remains valid when the PH assumption is violated , providing an a priori parsimonious test choice in this time-to-event scenario . All models were stratified into five groups according to quintiles of enrollment age , because enrollment criteria for the validation set included being free of chronic P . aeruginosa airway infection , making it invalid to compare children enrolled at later ages ( selected to be free of chronic P . aeruginosa airway infection ) to those enrolled at earlier ages . Heuristically , stratification in the Cox model results in making comparisons within strata without the need for proportional hazards across strata , and then combines the results from different strata for an overall test statistic [43] . While over-stratification can result in a loss of power , failure to stratify could possibly lead to confounding , which is the more important consideration here . All models included enrollment age , the number of observations on study ( in order to adjust for less sensitivity for detection of the end-point in individuals with fewer observations ) and an indicator for CFTR risk group 2 ( to adjust for severity of the CFTR dysfunction ) [21] . The two group classification system of McKone et al is a parsimonious method for a complex set of genotypes that works well in practice: risk group 2 genotypes include those with at least one functional group IV or V allele , are thought to have residual CFTR function , correlate highly with pancreatic sufficient disease and have a generally milder course of disease [21] , and was significantly associated with age-of-onset of chronic P . aeruginosa in all models ( p ≤ 0 . 01 ) . Sex/gender , pancreatic enzyme use , identification by newborn screening and F508del-CFTR homozygous genotype were not significant predictors of age-of-onset , did not affect variant HR estimates and were not included in the final models . Race was not included in the model because there were not enough individuals on non-European ancestry to attain convergence of the estimation algorithm . However , the models were fitted for individuals of European ancestry only to determine whether any significant effects were driven by the few African American individuals in the analysis ( S1 Table ) . Further , principal components were constructed for the subset of validation individuals for whom chip data were available ( N = 608 ) , and PCs 1 , 2 , and 3 were used to adjust for possible confounding by ancestry among this subset of CF individuals . To even further avoid potential ancestral confounding , models were fitted to the subset of F508del-CFTR homozygous individuals , both with and without PC adjustment , to ensure that residual confounding by happenstance imbalance in CFTR-mutation severity among CAV2 or TMC6 groups did not account for the observed associations . As an additional validation analysis , association between discovered variants and lung function was tested . All EPIC participants underwent lung function tests starting at age 6 years of age as part of routine care , generally on a quarterly basis . Forced expiratory volume in 1 second ( FEV1 ) is regarded as the most informative measure of lung function among CF individuals , and CF-specific percentile equations have been developed for lung function studies in CF . We tested for association between mean CF-specific FEV1 and TMC6 and CAV2 variants using generalized estimating equations ( GEE ) to account for correlation of measurements within subjects after adjusting for age and CFTR clinical risk group . Five hundred seventy-two children had variant calls and lung function measurements at age 7 and beyond ( lung function measurements before age 7 not used due to the “learning” effect and potential unreliable measurements before age 7 ) with a median of 23 lung function measurements and a median of 4 . 4 years of observation time ( min = . 5 , max = 12 ) . Principal components ( PCs ) were calculated for the 557 individuals with exome chip data using ancestry informative markers on the chip . These PCs were used to adjust for potential confounding by population stratification in the validation analysis . In addition , we sought to increase the sample size for the validation analysis with PC adjustment and to compare exome-based PC adjusted with AIMs-based PC adjustment by using genotype information from two different sources . Among the validation individuals who were genotyped for CAV2 manually and who did not have chip data ( n = 86 ) , most ( n = 51 ) had 353 ancestry informative markers ( AIMs ) available from the Illumina Golden Gate AIM ( “fingerprint” ) chip used in another study , and n = 354 individuals had data from both chips . Among subjects who had PCs for both the exome chip and the fingerprint chip , the linear correlation between PCs1 from either source was very high ( r = 0 . 91 ) with higher correlation among individuals not clustering with the main group ( r = 0 . 98 ) . This indicates that one set of PCs can be converted to the other via a linear transformation for the purposes of identifying ancestral group . We regressed the exome chip PCs on the fingerprint chip PCs to obtain a linear conversion equation by which to convert PCs from the fingerprint chip to the scale of those for the exome chip . We converted PC1 for the 53 individuals without exome chip data to the exome chip scale for use in adjusting for ancestry in the validation analyses for sample size of 608 . We examined all three subsets with PC data available: PCs from exome chip only , PCs from fingerprint chip only; and PCs from combined sources . | Whole exome and whole genome sequencing provide the opportunity to test for associations between expressed traits and genetic variants that cannot be tested with chip technology , particularly variants that are too rare to be included on chips designed for genome-wide association analysis . We used exome sequencing to identify variants in CAV2 and TMC6 that modify the age-of-onset of chronic Pseudomonas aeruginosa infection among children with cystic fibrosis , and validated our findings in a large cohort of children with cystic fibrosis . For a fixed number of study participants , it is known that the extreme phenotypes design provides greater statistical power than a random sampling design . In the extreme phenotypes design , one compares the frequency of a given set of genetic variants in one extreme of age-of-onset ( early onset ) to that in the other extreme ( late onset ) . Here , we employed an alternative design that compares genetic frequencies in exomes sampled from one extreme to that among exomes from a large set of controls . We show that this design confers substantially greater statistical power for discovery of CAV2 and TMC6 and provide general conditions under which this single extreme versus control design is more powerful than the extreme phenotypes design . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Exome Sequencing of Phenotypic Extremes Identifies CAV2 and TMC6 as Interacting Modifiers of Chronic Pseudomonas aeruginosa Infection in Cystic Fibrosis |
Mitochondrial oxidative phosphorylation ( OXPHOS ) is responsible for generating the majority of cellular ATP . Complex III ( ubiquinol-cytochrome c oxidoreductase ) is the third of five OXPHOS complexes . Complex III assembly relies on the coordinated expression of the mitochondrial and nuclear genomes , with 10 subunits encoded by nuclear DNA and one by mitochondrial DNA ( mtDNA ) . Complex III deficiency is a debilitating and often fatal disorder that can arise from mutations in complex III subunit genes or one of three known complex III assembly factors . The molecular cause for complex III deficiency in about half of cases , however , is unknown and there are likely many complex III assembly factors yet to be identified . Here , we used Massively Parallel Sequencing to identify a homozygous splicing mutation in the gene encoding Ubiquinol-Cytochrome c Reductase Complex Assembly Factor 2 ( UQCC2 ) in a consanguineous Lebanese patient displaying complex III deficiency , severe intrauterine growth retardation , neonatal lactic acidosis and renal tubular dysfunction . We prove causality of the mutation via lentiviral correction studies in patient fibroblasts . Sequence-profile based orthology prediction shows UQCC2 is an ortholog of the Saccharomyces cerevisiae complex III assembly factor , Cbp6p , although its sequence has diverged substantially . Co-purification studies show that UQCC2 interacts with UQCC1 , the predicted ortholog of the Cbp6p binding partner , Cbp3p . Fibroblasts from the patient with UQCC2 mutations have deficiency of UQCC1 , while UQCC1-depleted cells have reduced levels of UQCC2 and complex III . We show that UQCC1 binds the newly synthesized mtDNA-encoded cytochrome b subunit of complex III and that UQCC2 patient fibroblasts have specific defects in the synthesis or stability of cytochrome b . This work reveals a new cause for complex III deficiency that can assist future patient diagnosis , and provides insight into human complex III assembly by establishing that UQCC1 and UQCC2 are complex III assembly factors participating in cytochrome b biogenesis .
Mitochondrial disorders of ubiquinol-cytochrome c oxidoreductase ( complex III , MIM 124000 ) represent a significant proportion of patients with OXPHOS dysfunction [1] , [2] . Their identification is challenging due to ( a ) the sheer number of candidate genes , ( b ) their complicated interplay and ( c ) an incomplete understanding of complex III assembly . To date , mutations in only eight human genes have been identified as responsible for complex III deficiency . The first mutation was identified in the only mtDNA-encoded subunit of complex III , cytochrome b ( MT-CYB , MIM 516020 ) , in an adult patient with progressive exercise intolerance [3] . Patients with MT-CYB mutations have since been reported with a range of phenotypes and symptoms including Mitochondrial Encephalomyopathy , Lactic Acidosis and Stroke-like episodes ( MELAS , MIM 540000 ) , mitochondrial myopathy , cardiomyopathy and multisystem failure [2] , [4] . While mtDNA mutations are the most common identified cause of complex III deficiency , mutations in nuclear genes can also be causative . Mutations have been reported in four nuclear genes encoding complex III subunits; the supernumerary subunits UQCRB ( MIM 191330 ) [5] and UQCRQ ( MIM 612080 ) [6] and more recently the core subunit UQCRC2 ( MIM 191329 ) [7] and the catalytic subunit CYC1 ( MIM 123980 ) [8] . Patients with UQCRB and UQCRC2 mutations presented with hypoglycaemia and lactic acidosis , the patient with UQCRQ mutations presented with severe psychomotor retardation and patients with CYC1 mutations presented with recurrent ketoacidosis and insulin-responsive hyperglycemia [8] . In addition to complex III subunit genes , mutations have also been identified in two complex III assembly factor genes , BCS1L ( MIM 603647 ) [9] , TTC19 ( MIM 613814 ) [10] and recently LYRM7 [11] . BCS1L mutations were first identified in 2001 [9] and since then more than 20 additional mutations of the BCS1L gene have been reported [2] . The clinical presentation of patients with BCS1L mutations varies greatly with some mutations being associated with tubulopathy , encephalopathy and liver failure [9] , others with GRACILE syndrome ( growth retardation , aminoaciduria , cholestasis , iron overload , lactic acidosis and early death , MIM 603358 ) [12] , with isolated encephalopathy [13] or with Björnstad syndrome characterized by sensorineural deafness associated with short brittle hair ( MIM 262000 ) [14] . Mutations in TTC19 have been shown to cause encephalopathy with variable age of onset and rate of progression , which in some patients is associated with severe psychiatric manifestations [10] , [15] . Mutations in LYRM7 are associated with early onset encephalopathy [11] . Despite the discovery of pathogenic mutations in eight different complex III-related genes , the majority of patients with complex III deficiency remain unsolved . We aimed to identify new genes underpinning complex III deficiency and to elucidate their role in the complex III assembly process . We identified a causative homozygous UQCC2 ( MIM 614461 ) splicing mutation in a patient with severe intrauterine growth retardation , neonatal lactic acidosis and renal tubular dysfunction associated with complex III deficiency . We established the role of UQCC2 as a complex III assembly factor that cooperates with UQCC1 ( MIM 611797 ) to mediate cytochrome b protein expression and subsequent complex III assembly .
We studied a consanguineous Lebanese patient presenting with severe intrauterine growth retardation , neonatal lactic acidosis and renal tubular dysfunction . Spectrophotometric enzyme assays revealed a severe complex III deficiency with residual activity of only 9% in skeletal muscle and 5% in skin fibroblasts when normalized to citrate synthase activity ( Figure 1A ) . Complex I activity was reduced to 29% and 37% and complex IV activity was reduced to 51% and 53% in muscle and fibroblasts respectively , whereas complex II activity was normal . Secondary deficiency in complex I and complex IV have been described in patients with primary complex III deficiency previously [16]–[19] . In our own experience , skeletal muscle from 4 previous patients with pathogenic mutations in genes encoding complex III subunits or assembly factors ( CYC1 , UQCRC2 , and two BCS1L ) had residual activities for complexes I , II , III and IV of 54±17 , 117±17 , 16±6 and 55±13 ( expressed as % of control mean relative to citrate synthase; mean ± S . E . M . ) . Given the severity of the complex III defect in regard to the activities of complexes I and IV , we thus regarded the patient as having a primary complex III defect . In order to uncover the molecular basis for the complex III defect , we performed “MitoExome” sequencing , involving the targeted capture and massively parallel sequencing ( MPS ) of the mtDNA and ∼1000 nuclear genes predicted to encode the entire mitochondrial proteome [20] . This approach identified 829 single nucleotide variants or small insertion/deletions in the patient , which were analyzed to select the disease gene ( Figure 1B ) . Three genes , TYMP , MTCH1 and UQCC2 , harbored rare ( allele frequency <0 . 005 in dbSNP version 132 [21] and the 1000 genomes project release 20100804 [22] ) homozygous or compound heterozygous variants that were predicted to potentially impact protein function . Although TYMP is a known OXPHOS “disease gene” , the clinical and biochemical presentation of the patient , plus the consanguinity , suggested that the rare compound heterozygous TYMP mutations ( c . 242G>A , c . 148A>C ) were not causal . Patients with TYMP mutations typically present with combined OXPHOS deficiency associated with mtDNA deletions/depletion rather than a primary complex III deficiency , and have mitochondrial neurogastrointestinal encephalomyopathy ( MNGIE , MIM 603041 ) as opposed to the primary lactic acidosis and renal tubulopathy observed in our patient [23] . Although the homozygous c . 170C>T mutation in the pro-apoptotic MTCH1 was not reported in dbSNP version 132 [21] or the 1000 genomes project release 20100804 [22] , it was detected with a minor allele frequency of 0 . 005526 in the Exome Variant Server ( NHLBI GO Exome Sequencing Project , http://evs . gs . washington . edu/EVS/ March 2013 ) , suggesting it is likely to be too common to cause a rare mitochondrial disorder . Computational analyses suggested that the third candidate , UQCC2 ( previously called MNF1 , M19 or C6orf125 ) was likely causal based on 1 ) its orthology to the S . cerevisiae complex III assembly factor Cbp6p , and 2 ) its co-expression with complex III subunits ( see below ) . Further analyses indicated a wider conservation of complex III assembly factors , as the interaction partner of S . cerevisiae Cbp6p , Cbp3p , also had a mammalian homolog , UQCC1 ( previously called UQCC ) , which was co-expressed with UQCC2 and similarly co-expressed with complex III subunit genes . The homozygous c . 214-3C>G UQCC2 ( NM_032340 ) mutation fell within a 30 . 7 Mb long contiguous stretch of homozygosity ( LCSH ) ( Figure S1 ) , consistent with both alleles being inherited from a common ancestor . The mutation was verified via Sanger sequencing ( Figure 1C ) and was not reported in dbSNP version 132 [21] , the 1000 genomes project release 20100804 [22] or the Exome Variant Server . To investigate whether the c . 214-3C>G mutation might be a common variant found within the Lebanese population , a Sequenom assay was developed to genotype 86 Lebanese controls . The c . 214-3C>G variant was not detected , suggesting it is rare in the ethnically-matched population . To see if this gene might be the cause of complex III deficiency in other patients , we sequenced the coding regions of UQCC2 in 11 patients with confirmed complex III deficiency who lacked a molecular diagnosis . No potentially pathogenic changes were identified . The c . 214-3C>G UQCC2 mutation is found 3 bases upstream of exon 3 . The third base upstream of exons generally has only moderate conservation , usually being a cytosine or a thymine and never a guanine [24] . In keeping with this , the c . 214-3C>G UQCC2 site is moderately conserved , with no vertebrate having a guanine at this position ( Figure S2 ) . To investigate whether the c . 214-3C>G mutation causes a splicing defect , Reverse Transcriptase ( RT ) -PCR was performed using RNA extracted from patient fibroblasts . Sequencing revealed aberrant mRNA splicing in the patient , with two major mRNA splice variants ( Figure 2A ) , both of which were generated by the use of cryptic acceptor sites ( Figure 2B–C and Text S1 ) . Patient fibroblasts had only 2% residual wild-type UQCC2 expression ( Figure 2D ) , as determined by qRT-PCR , suggesting the c . 214-3C>G mutation almost completely abolishes wild-type splicing . In keeping with this , there was no detectable UQCC2 protein observed by western blot ( Figure S3 ) . The protein encoded by the alternative splice species is likely unstable , as there was also no evidence of elongated or truncated UQCC2 protein ( Figure S3 ) . Human UQCC2 was previously postulated to have a role in mtDNA maintenance and was found to associate with mitochondrial nucleoids [25] , however , PicoGreen staining indicated mitochondrial nucleoids were not disturbed in patient fibroblasts ( Figure S4A ) . The patient also had no significant mtDNA depletion , having 78% mtDNA compared to the mean of 4 control fibroblast cell lines when estimated by qPCR analysis ( Figure S4B ) . Iterative orthology prediction using the Ortho-Profile method [26] revealed that the UQCC2 gene is an ortholog of the S . cerevisiae CBP6 gene that is required for complex III assembly [27] ( Figure 3A ) . We used the Ortho-Profile method to investigate whether there was wider conservation of complex III assembly factors and found that Cbp3p , which cooperates with Cbp6p in complex III assembly , also has a predicted human ortholog , UQCC1 ( Figure 3B ) . Both orthologous groups have diverged significantly among eukaryotes with UQCC2/Cbp6p having an overall low amino acid conservation ( Figure 3A ) , while only the C-terminus of UQCC1/Cbp3p is conserved ( human residues 135–279 ) ( Figure 3B ) . The N-terminal regions of UQCC1/Cbp3p proteins ( residues 1–134 in human and 1–144 in yeast ) are highly divergent in both metazoa and fungi , with homologous sequences recognizable only in closely related species ( vertebrates for the N-terminal fragment of UQCC1 and the Saccharomyceta clade for Cbp3p ) . Interestingly , the amino acids from positions 12 to 96 have been shown to be relatively dispensable for Cbp3p function , explaining the lack of sequence conservation [28] . To support the association of UQCC1 and UQCC2 proteins with complex III in mammals , we investigated the co-expression of the genes with complex III subunit genes in 91 mouse tissues and cell types . UQCC1 and UQCC2 genes co-express highly with each other at the mRNA level ( Pearson correlation 0 . 636 ) . Genes for complex III subunits co-express significantly with both UQCC2 ( average 0 . 66 , Figure 3C ) and UQCC1 ( average 0 . 66 , Figure 3D ) . The co-expression of the two genes with complex III subunits is , on average , 3-fold higher than with genes encoding other mitochondrial proteins ( two-sided Mann–Whitney test P-value <2×10−5 ) . Both genes are conserved in amoebozoa indicating their ancient evolutionary origin , preceding the divergence of human and fungi ( Figure 3A and 3B ) . UQCC1 is additionally conserved in distantly related eukaryotes that contain mitochondria , including the stramenopile Phytophthora infestans . A related stramenopile , Blastocystis hominis , has lost respiratory chain complexes III , IV and V and congruently we did not identify orthologs of the two complex III assembly factors in its genome . To verify that the UQCC2 mutation was indeed responsible for the complex III defect , patient fibroblasts were transduced with a lentiviral construct expressing wild-type UQCC2 mRNA to examine whether it could restore complex III assembly . Lentiviral transduction caused UQCC2 expression in PUQCC2 to increase to a level comparable to controls ( Figure 4A ) . To assess complex III restoration , western blotting for the complex III subunit UQCRFS1 was performed as this subunit was clearly degraded in patient fibroblasts ( Figure 4 ) . As a negative control , fibroblasts with a mutation in a complex III subunit ( manuscript in preparation ) but wild-type UQCC2 were transduced in parallel . Lentiviral transduction of UQCC2 caused a significant increase in UQCRFS1 protein expression in the patient with UQCC2 mutations ( p<0 . 05 , Two-Way ANOVA ) , but caused no significant change in the normal control or the complex III deficient patient with mutations in a complex III subunit gene . After transduction , the level of UQCRFS1 protein expression in PUQCC2 was no longer significantly different from control cells . We also measured complex III activity in fibroblasts before and after transduction with UQCC2 . There was a clear increase in complex III activity after transduction with UQCC2 in PUQCC2 from 29% to 73% of control ( Figure S5A ) . These results prove that the complex III deficiency is due to a lack of UQCC2 , and that the mutations in other candidate genes in the patient , TYMP and MTCH1 , are not causative . To further characterize the biochemical consequence of UQCC2 deficiency , protein levels and complex assembly were analyzed by sodium dodecyl sulphate ( SDS ) -polyacrylamide gel electrophoresis ( PAGE ) and Blue Native ( BN ) -PAGE using patient fibroblasts . BN-PAGE and immunoblot analysis of mitochondria lysed with Triton X-100 revealed that PUQCC2 had a severe complex III defect , with a markedly reduced amount of complex III holocomplex but normal levels of complex II ( Figure 5A and Figure S6A ) . Consistent with enzyme analysis , there was a moderate reduction in complex I holocomplex , likely due to instability of complex I due to complex III deficiency [17] . Interestingly , despite mildly reduced complex IV activity in patient fibroblasts ( Figure 1A ) , the level of complex IV holoenzyme appeared increased ( Figure 5A and Figure S6A ) . We also analyzed patient complexes by BN-PAGE of mitochondria lysed with digitonin , which allows visualization of OXPHOS supercomplexes . Prior to transduction , the small amount of complex III detectable in PUQCC2 appears to be in the supercomplex form with little or no complex III dimer ( Figure S5B ) . However , transduction with UQCC2 restores the relative amounts of complex III in PUQCC2 in the dimer and supercomplex forms to ratios similar to control ( Figure S5B ) , further confirming UQCC2 as the causative gene . SDS-PAGE analysis demonstrated that fibroblasts from PUQCC2 had a mild defect in the level of the complex III subunit , UQCRC2 , and a more pronounced deficiency of the UQCRFS1 and UQCRC1 subunits ( Figure 5B and Figure S6B ) . The UQCC2 binding partner , UQCC1 , was barely detectable by western blot , suggesting UQCC2 is required for its stability . We confirmed that the lack of UQCC1 is due to the UQCC2 deficiency , by repeating SDS-PAGE analysis on fibroblasts transduced with UQCC2 ( Figure S5C ) . We also transfected HEK293 cells with an siRNA targeting UQCC2 that resulted in a 40% knockdown of the UQCC2 protein and 60% reduction in UQCC1 protein , supporting the requirement of UQCC2 for UQCC1 stability ( Figure 5C and Figure S7A ) . In contrast to patient fibroblasts , no obvious defect in complex III subunit levels was observed in cells with UQCC2 knockdown . This is likely a consequence of the less severe UQCC2 deficiency achieved in knockdown experiments , with 60% residual protein compared to no detectable protein in patient fibroblasts . Given that UQCC2 deficiency was associated with loss of UQCC1 , we further investigated the relationship between these proteins . We first confirmed that human UQCC1 is a mitochondrial protein by cellular fractionation and SDS-PAGE ( Figure 6A ) . Proteinase K digestion indicated that UQCC1 localizes to the inner mitochondrial membrane ( Figure 6B ) . In S . cerevisiae , the UQCC1 ortholog Cbp3p interacts with the UQCC2 ortholog , Cbp6p , and together they activate translation of mtDNA-encoded cytochrome b , bind and stabilize the newly synthesized protein and deliver it to an early complex III assembly intermediate [29] , [30] . To address a possible association between UQCC1 and UQCC2 , HEK293 cells expressing C-terminal TAP-tagged UQCC1 or UQCC2 under the control of a doxycycline-inducible promoter were generated and subjected to single step affinity purifications . Subsequent SDS-PAGE and western blot analysis of the UQCC2-TAP purification revealed efficient co-isolation of UQCC1 ( Figure 6C ) . The complex III subunits UQCRC1 , UQCRC2 , UQCRFS1 and the mitochondrial ribosomal subunits MRPL12 and MRPS22 did not co-elute with UQCC2-TAP ( Figure 6C ) . Using a UQCC1-TAP tagged construct we confirmed the interaction with UQCC2 ( Figure 6D ) . To investigate whether UQCC1-deficient cells exhibit a similar biochemical phenotype to UQCC2-deficient cells , we transfected HEK293 cells with siRNA targeting UQCC1 or cyclophilin B , which , along with mock-transfected cells , served as a negative control . The knockdown led to disappearance of UQCC1 protein in mitochondrial lysates and a concurrent loss of UQCC2 ( Figure 7A ) . Although UQCC2 stability appears to depend on UQCC1 , we did not observe that over-expression of either UQCC2-TAP or UQCC1-TAP led to an increase of UQCC1 or UQCC2 respectively ( Figure 6C and Figure 6D ) . Knockdown of UQCC1 led to reduced levels of the UQCRFS1 , UQCRC1 and UQCRC2 subunits of complex III ( Figure 7A and Figure S7B ) . With respect to the presence of individual subunits , the impact of UQCC1 knockdown appears to be limited to complex III , as subunits from complexes I , II , IV and V were unaffected ( Figure 7A , Figure 7B and Figure S8 ) . Nevertheless , with respect to the presence of complete OXPHOS complexes , as measured with BN-PAGE analysis , we do observe besides a marked reduction of complex III , also a reduction of complex I ( Figure 7B and Figure S7C ) . The UQCC1 knockdown profile is thus similar to UQCC2 deficiency in PUQCC2 , with reduced levels of complex III subunits , a reduced level of complex III and , to a lesser extent , a reduced level of complex I ( Figure 5A , 5B ) . Two-dimensional BN-PAGE experiments confirmed lower levels of mature complex III and additionally showed the accumulation of a partially assembled subcomplex containing UQCRC1 . This subcomplex was not detected in the cyclophilin B knockdown control cells ( Figure 7C ) . Subsequent OXPHOS enzyme activity measurements of UQCC1-depleted cells showed reduced complex III , a reduced combined complex II/III activity ( SCC ) and a slight but considerable reduction in complex I , compared to the mock control ( Figure 7D ) . Having established that UQCC1 and UQCC2 are involved in complex III assembly , we next investigated whether they are involved specifically in cytochrome b biogenesis . Mitochondrial translation products from patient and control fibroblasts were subjected to a 35S-pulse-chase assay and analyzed by SDS-PAGE . Even at zero hours chase , a striking and specific defect in cytochrome b protein levels was observed; other mtDNA-encoded subunits were present in normal amounts or , in the case of COX2 and COX3 , an increased amount ( Figure 8A ) . qRT-PCR analysis revealed that MT-CYB mRNA levels were unaffected in patient cells ( Figure 8B ) . To determine whether UQCC1 is involved in the stabilization of newly synthesized cytochrome b , the UQCC1-TAP purification was carried out with 35S metabolically labeled mitochondrial translation products . UQCC1 specifically associated with newly synthesized cytochrome b and not with other newly-translated mtDNA-encoded subunits ( Figure 8C ) . Moreover , inhibition of mitochondrial translation with chloramphenicol for 72 h predictably led to the disappearance of mtDNA-encoded COX1 , but also UQCC1 and UQCC2 proteins , suggesting that UQCC1 and UQCC2 may be stabilized by cytochrome b ( Figure 8D ) . In contrast , SDHA , a nuclear-encoded subunit of complex II , was not affected . We conclude that UQCC1 and UQCC2 are critical factors required for the expression of cytochrome b and complex III biogenesis .
Here we report the first case of complex III deficiency due to UQCC2 mutations . MitoExome MPS identified 3 genes with potentially pathogenic recessive-type variants , of which UQCC2 was strongly linked to complex III by evolutionary and computational analyses . We demonstrated a UQCC2 splicing defect resulting in a lack of UQCC2 protein and verified that this gene was causal by restoring complex III protein and activity levels in patient fibroblasts with lentiviral transduction of UQCC2 . This patient was a singleton from a consanguineous family , and would have been slower to solve using only traditional methods such as homozygosity mapping , which identified 1894 candidate genes in regions of LCSH , at least 86 of which have a putative role in the mitochondria [31] . MitoExome sequencing has already shown promise for the molecular diagnosis of patients with mitochondrial disease [20] , [32] , [33] and is likely to aid diagnoses in years to come . Patients with complex III deficiency present with great clinical heterogeneity . Symptoms typically provide little insight into the underlying genetic cause . Our patient shared some of the clinical features of previously reported complex III deficient patients such as tubulopathy and primary lactic acidosis with BCS1L patients [9] but only identification of other pathogenic UQCC2 mutations in unrelated individuals will provide a complete picture of the clinical spectrum of patients with UQCC2 dysfunction . Complex III deficiencies are commonly accompanied by a reduction in complex I and sometimes complex IV activity [16]–[18] . We also observed reduced presence and activity of complex I in muscle and fibroblasts from PUQCC2 , which prompted us to provide additional evidence that both UQCC1 and UQCC2 are in fact assembly factors specific to complex III . One type of evidence comes from the analysis of the presence-absence patterns of the two genes among sequenced genomes . The presence and function of CBP3 and CBP6 and their orthologs in fungi that encode complex III but have lost complex I ( S . cerevisiae , Schizosaccharomyces pombe ) substantiate their role in complex III biogenesis . Conversely , orthologs of these assembly factors are absent in B . hominis , a species that has lost complexes III–V but still encodes complexes I and II . A relative of B . hominis , Phytophthora infestans , which encodes complex III also encodes an ortholog of Cbp3p/UQCC1 ( Figure 3 ) . Coevolution of orthologs of UQCC1 and UQCC2 with complex III and not with complex I indicates that these assembly proteins function primarily in the assembly of complex III . We elucidated the role of UQCC2 , in cooperation with UQCC1 , in human cytochrome b biogenesis and subsequent complex III assembly and function . No direct link between these proteins and human complex III has previously been described . Previous studies of human UQCC2 suggested that the protein localizes to mitochondrial nucleoids [25] and that it modulates respiratory chain activity in skeletal muscle and pancreatic cells [34] . However , we found no disturbance of mitochondrial nucleoids or mtDNA copy number in patient fibroblasts . Previous studies of human UQCC1 have all focused on variation at this locus being associated with human height [35]–[37] . Our data suggest that UQCC1 and UQCC2 are interacting and interdependent proteins , with the stability of UQCC2 depending on UQCC1 , and vice versa . We also show that UQCC1 plays an important role in early complex III biogenesis via interaction with newly synthesized cytochrome b and recruitment of this mtDNA-encoded subunit into a complex III assembly intermediate ( Figure 9 ) . The direct effect of UQCC1 and UQCC2 dysfunction is an immediate lack of cytochrome b leading to disruption of the downstream complex III assembly process . Mutations in UQCC2 ( Figure 5 ) and UQCC1 depletion ( Figure 7 ) both lead to reduced levels of the complex III subunits UQCRFS1 , UQCRC1 and UQCRC2 . A subcomplex containing UQCRC1 but not UQCRFS1 accumulates upon UQCC1 knockdown , consistent with a defect early in CIII assembly before incorporation of the UQCRFS1 subunit ( Figure 9 ) . The role of UQCC1:UQCC2 in the initial stages of complex III assembly is further supported by the UQCC1 binding to newly synthesized cytochrome b ( Figure 8C ) , although we have not shown that UQCC1 and UQCC2 bind to cytochrome b together . Interestingly , in S . cerevisiae , Cbp3p and Cbp6p have been shown to provide a feedback loop modulating cytochrome b expression in response to complex III assembly [30] . Cbp6p and Cbp3p , bind to MT-CYB mRNA to activate its translation and then deliver newly-synthesized cytochrome b to a complex III assembly intermediate . When early complex III assembly is disrupted , cytochrome b cannot be deposited by the Cbp3p:Cbp6p complex and so these factors remain bound to the cytochrome b protein . While bound in an assembly intermediate , the Cbp3p:Cbp6p complex is unable to activate further cytochrome b translation , thus modulating cytochrome b synthesis in response to complex III assembly . Such coordination between mtDNA translation and nuclear gene expression prevents the build-up of mtDNA-encoded proteins in the absence of functional complexes . It will be interesting to investigate whether UQCC1 and UQCC2 provide a similar feedback loop between cytochrome b translation and complex III assembly in mammalian mitochondria . The role of UQCC1 and UQCC2 in cytochrome b expression like their S . cerevisiae orthologs Cbp3p and Cbp6p , is supported by the fact that no cytochrome b synthesis is detected in PUQCC2 via a mitochondrial translation assay ( Figure 8A ) . Furthermore , mitochondrial translation is a prerequisite for the stability and function of both proteins ( Figure 8D ) . The previously reported co-localization of UQCC2 with mitochondrial nucleoids would be consistent with a role in mitochondrial translation , as factors required for mitochondrial protein synthesis , such as ATAD3 and PHB , are often found to associate with mitochondrial nucleoids [38] . Nevertheless , whether UQCC1 and UQCC2 are directly required for translation activation of MT-CYB remains to be established , specifically because the 5′ UTRs of mitochondrial mRNAs to which translational activators in S . cerevisiae bind are absent from human mitochondrial mRNAs . Furthermore , in the fungus S . pombe the function of the Cbp6p ortholog appears to be only post-translational [39] . One can speculate that the lack of sequence conservation of the N-terminus of the Cbp3p orthologs , even among fungi , could explain the lack of conservation of translation activation . Nevertheless , there is currently no information about which region of S . cerevisiae Cbp3p is required for translation activation and the N-terminus of Cbp3p appears to be relatively dispensable for its function , even in S . cerevisiae itself [28] . To date , only one putative human mitochondrial translational activator , TACO1 , which is required for the translational activation of the COX1 subunit of complex IV , has been described [40] . In summary , here we have used MitoExome MPS in combination with computational and experimental analyses to identify the first case of complex III deficiency due to UQCC2 mutation . We demonstrate that UQCC2 and its binding partner UQCC1 are required for early complex III assembly by mediating the synthesis , stability and/or assembly of the mtDNA-encoded complex III subunit , cytochrome b .
Investigations were performed with ethics approval by the Human Research Ethics Committee of the Royal Children's Hospital , Melbourne . The proband was the first child of first cousin Lebanese parents and was patient P12 in [20] . The pregnancy was complicated by intrauterine growth retardation , and he was born at 36 weeks gestation by emergency caesarean section because of placental compromise . Birth weight was 1280 gm , length 41 cm and head circumference 29 cm . He had good Apgar scores ( 9 and 9 at one and five minutes respectively ) , but by 12 hours of age he became lethargic and had loose stools . He was found to have severe metabolic acidosis ( pH 7 . 16 , lactate 9 . 6 mmol/L; normal range 0 . 7–2 . 0 ) , and CSF lactate at around that time was 3 . 8 mmol/L ( normal <2 . 0 ) . His blood electrolytes suggested he had a proximal renal tubular acidosis . His condition improved with rehydration and bicarbonate supplementation , but blood lactate remained high ( 5–14 mmol/L ) even when well . In addition , he was mildly dysmorphic with synophrys , epicanthic folds , upward slanting palpebral fissures , a depressed nasal bridge and flattened nose , and had a unilateral undescended testis . He also had unilateral postaxial polydactyly , the skin tag being removed in the newborn period . His father had a similar facial appearance and had a history of a cleft palate . Seizures were effectively treated with phenobarbitone for 6 months , after which it was ceased . A CT and MRI scan of the brain revealed no abnormality . A vitamin cocktail including riboflavin , thiamine , vitamin C , biotin and coenzyme Q did not appear to have an effect on his blood lactate levels or clinical condition . He developed acute gastroenteritis at 5 months of age , at which time severe metabolic acidosis again developed , which again resolved with rehydration . Afebrile seizures recurred at two years of age , and were treated with sodium valproate and later with lamotrigine . EEG was normal at this time . Developmental milestones were delayed: he sat unaided at six months , crawled at 10 months , could cruise around furniture at 13 months , walked unaided at 15 months , but still had no formal speech by 2 years 3 months . By that age he could walk unaided , but frequently fell . Stamina was normal . Formal neuropsychological review at two years nine months revealed severe delay in fine motor and visuo-spatial performance , self-care skills and social play , with gross motor skills being only mildly impaired . There were no concerns with vision , but he had a mild sensorineural hearing impairment . Despite speech therapy , at six years of age he still had only several words in Arabic but no meaningful speech . He went on to develop a number of autistic features , including impulsivity , limited eye contact , extreme hyperactivity , aggressive behaviour , night time roaming , and continued to have no real expressive language , although he was felt to have reasonable receptive language . He required two minders at school because of concerns of his lack of regard for his physical safety . Ritalin caused him to become even more agitated , whereas clonidine appeared to be of some benefit . He was lost to follow up at nine years of age . Unless otherwise described below , cell culture , DNA isolation , RNA isolation , cDNA synthesis and sequencing of PCR products were performed as described previously [41] . To sequence unique splice variants , RT-PCR products were first cloned into a pTOPO vector using the TOPO TA Cloning Kit ( Invitrogen ) as per manufacturer's protocol . For mitochondrial nucleoid staining , patient fibroblasts and control cell lines were grown on coverslips and stained with 3 µl/ml PicoGreen [42] ( Invitrogen ) for 1 hour and 10 nM MitoTracker Red CMXRos ( Invitrogen ) for 30 min at 37°C , 5% CO2 . Coverslips were washed with PBS , then mounted on slides for live cell imaging using a Zeiss AxioImager . M1 epifluorescence microscope . Genes encoding the entire predicted mitochondrial proteome ( 1381 nuclear genes and the mtDNA ) were captured and sequenced on an Illumina Genome Analyzer II as described previously [20] . Variant prioritization used previously established criteria for likely pathogenicity [20] , [41] . Spectrophotometric enzyme assays assessing mitochondrial OXPHOS activity were performed as described previously for patient samples ( muscle post-nuclear supernatants and fibroblast mitochondria ) [43] and for functional studies in mitochondria from HEK293 cells , see [44] and references therein . An orthology identification pipeline that uses sequences , sequence-based profiles as well as profile-derived Hidden Markov Models [26] was applied to identify human orthologs of fungal Cpb3p and Cbp6p proteins . As a negative control for the orthology prediction method , sequence-based profiles of the orthologs were also used to search for orthologs in the genome nucleotide sequence of B . hominis [45] , a species with mitochondria-like organelles but without complex III . No orthologs were found in the B . hominis genome despite using sensitive PSI-tblastn [46] to circumvent possible gene annotation errors . To calculate the Pearson correlation of mRNA expression in murine tissues and cell types , 182 microarray sample measurements with Affymetrix Mouse Genome 430 2 . 0 Array [47] were used . The data ( GNF Mouse GeneAtlas V3 ) were downloaded from Gene Expression Omnibus , record GSE10246 [47] . The data were transformed as described previously [48] . Quantitative expression analysis of UQCC2 was performed as previously described [49] , using the MNF1 Hs00942667_m1 gene expression assay ( Life Technologies ) that detects the exon 2/3 junction of UQCC2 ( MNF1 ) with the HPRT1 Endogenous Control Gene Expression Assay ( Life Technologies ) for normalization , and the Cytb ( MT-CYB ) Hs02596867_s1 gene expression assay ( Life Technologies ) with the previously described ND1 assay [50] for normalization . Because the Cytb and ND1 assays cannot distinguish cDNA from mtDNA , an additional DNAse-treatment was performed prior to qRT-PCR using the Turbo DNA-free kit ( Ambion ) as per manufacturer's protocol . Quantitative analysis of mtDNA copy number was performed with a probe targeting ND1 to represent mtDNA and a probe targeting CFTR as the nuclear reference , as described previously [51] . Molecular karyotyping of patient DNA was performed with the Illumina HumanCytoSNP-12 array ( version 2 . 1 ) as previously described [52] . Automated LCSH detection was performed with the CNVPartition v3 . 1 . 6 algorithm in KaryoStudio software . SNP genotypes were generated in GenomeStudio software ( Illumina ) with data from a set of 102 intra-run samples . A Sequenom assay using multi-plexed MALDI-TOF mass spectrometry was designed to genotype 86 Lebanese controls for the c . 214-3C>G mutation . The forward , reverse and extension primers were as follows: 5′ACGTTGGATGCTTCACTTCCTTTCTGCCCC3′ , 5′ACGTTGGATGTGTACTCTTCCAACGACAGG3′ , 5′CACTTCCTTTCTGCCCCGGTGAC3′ . Genotypes were called using the MassARRAY System Typer version 4 . 0 ( Sequenom ) . Full length UQCC2 was amplified from cDNA using high-fidelity Phusion Taq ( Finnzymes ) with a forward primer incorporating a 5′ BamHI recognition site ( 5′CGGGATCCACCATGGCGGCCAGCCGGTACCGGCGTT3′ ) and a reverse primer incorporating a 3′ XbaI recognition site ( 5′GCTCTAGATTATCAGGCCTTATGATCCTCCTCAGGAC3′ ) . The resulting RT-PCR product was cloned into the 4-hydroxytamoxifen-inducible lentiviral vector , pF_5x_UAS_MCS_SV40_puroGEV16-W [53] . UQCC2 viral particles were generated and patient fibroblasts were transduced as described previously [41] . Three independent transductions were performed and cells were harvested 12–18 days after selection with 1 mg/ml puromycin . One-dimensional 5–15% BN gradient and two-dimensional ( 2D ) SDS gradient PAGE were done as described previously [54] , [55] . Whole cell or tissue samples were used for SDS-PAGE and isolated mitochondria were used for BN-PAGE . SDS-PAGE with 10% NuPage gels ( Invitrogen ) and immunoblotting was performed as described previously [41] . Proteins were detected with the following antibodies: α-MNF1 ( ATLAS antibodies or as previously described [25] for detection of UQCC2 ) , Total OXPHOS Human WB Antibody Cocktail containing α-ATP5A1 , α-UQCRC2 , α-SDHB , α-COX2 and α-NDUFB8 ( MitoSciences ) , α-NDUFA9 ( as previously described [55] ) , α-UQCC ( Atlas Antibodies ) , α-CBP ( GenScript ) , α-ND1 ( kindly provided by A Lombes [56] ) , α-cyclophilin B ( Affinity Bioreagents ) , α-OXA1L ( Central Animal Facility Nijmegen ) , α-SDHA , α-SDHB , α-COX1 , α-UQCRC1 , α-UQCRC2 , α-UQCRFS1 and α-ATP5α ( all MitoSciences ) , α-TOM20 ( BD transduction laboratories ) , α-CK-B 21E10 ( kindly provided by the Department of Cell Biology Nijmegen [57] ) , α-MRPL12 ( Abcam ) , α-MRPS22 ( Proteintech ) and α-VDAC1 ( Calbiochem ) . Secondary antibodies were goat α-mouse or swine α-rabbit IgG horseradish peroxidase ( HRP , DakoCytomation ) , goat α-mouse or α-rabbit IgG HRP antibodies ( Invitrogen ) . Quantification of western blots was performed by densitometry using ImageJ software or the Chemidoc XRS+ system ( Biorad ) software . zMitochondrial translation assays were performed as described previously [58] , [59] . Briefly , fibroblasts were cultured with cycloheximide to inhibit cytoplasmic translation and mtDNA-encoded proteins were labeled with a 2-hour pulse of 35S-methionine/35S-cysteine ( EXPRE35S35S Protein Labeling Mix; Perkin Elmer Life Sciences ) prior to washing and a chase with cold methionine for 0 to 24 hours . Mitochondria were then isolated and translation products were analyzed by SDS-PAGE and autoradiography . HEK293 cells were cultured and labeled in the same way , except that labeling was done for 1 hour , emetine was used instead of cycloheximide and Tran35S-Label ( MP Biomedicals ) was used for labeling . Two-way repeated-measures analysis of variance ( ANOVA ) was used for comparisons of groups followed by post hoc analysis with a Bonferroni correction to account for multiple comparisons . The UQCC2 open reading frame was PCR amplified without the stop codon from HEK293 cDNA adding Attb recombination sites ( underlined ) using the following primers: forward 5′-AA AAAGCAGGCTTCGCCACC ATGGCGGCCAGCCGGTACCGGCG-3′ and reverse 5′-GAAAGCTGGGTG GGCCTTATGATCCTCCTCAGG-3′ . After the first PCR , the specific product was used in a second PCR using this primer set: forward 5′-GGGGACAAGTTTGTACAAAAAAGCAGGCT-3′ and reverse 5′-GGGGACCACTTTGTACAAGAAAGCTGGGT-3′ to complete the recombination sites and allow the cloning in the pDONR201 vector using BP clonase enzyme mix ( Invitrogen ) . The pDONR201 vector containing the UQCC1 open reading frame without stop codon was obtained from the Harvard Medical School ( clone ID: HsCD00081684 ) [60] . Next , mammalian expression vectors under the control of a tetracycline-inducible promoter adding a tandem affinity purification ( TAP ) tag at the C-terminus were generated by recombining the pDONR201 vectors with the appropriate expression vector with the aid of LR clonase enzyme mix . All vectors were checked with sequence analysis before further use . T-REx Flp-In Human Embryonic Kidney 293 cells ( HEK293; Invitrogen ) were grown and maintained in Dulbecco's Modified Eagles Medium ( DMEM; Biowhitaker ) supplemented with 10% FCS , 1% [v/v] penicillin/streptomycin , zeocin ( 300 µg/ml , Invitrogen ) and 5 µg/ml blasticidin ( Calbiochem ) . To generate stable cell lines expressing TAP fusion proteins , cells were transfected with the corresponding construct using Superfect transfection reagent ( Qiagen ) and selected for stable transfectants by replacing the zeocine in the culture medium with hygromycin ( 200 µg/ml , Calbiochem ) . Gene expression was induced by adding 1 µg/ml doxycycline ( Sigma ) to the culture medium for a minimum of 24 h . Mitochondrial translation was inhibited with 40 µg/ml chloramphenicol ( CAP ) for a minimum of 72 h . siRNAs were designed using the online available software from the Whitehead Institute for Biomedical Research [61] and synthesized by Biolegio ( Nijmegen ) . The following siRNAs were used: UQCC2 antisense 5′-AGUAGUUUGAAUGGAGUCG dTdT-3′; UQCC1 antisense 5′-UAUGAUACGACACAUGUAC dTdT-3′ as well as control cyclophilin B targeting siRNA ( Thermo Scientific ) . For transfection HEK293 cells were plated in antibiotic-free culture medium and transfected the next day with 10 nM siRNAs using Dharmafect 1 transfection reagent ( Dharmacon ) . At day 3 , cells were split 1∶4 and transfected again the next day . Cells were harvested 96 hours after the first transfection and analyzed with SDS and/or BN-PAGE . The cellular fractionation of HEK293 cells was done as previously described [62] . For determining the submitochondrial localization a proteinase K protection assay was performed as previously described [63] . Mitoplasts were pelleted by centrifugation as described before [64] . The supernatants containing the solubilized proteins were used for further analysis . Protein concentrations of the samples were determined with the microBCA protein kit ( Thermo Scientific ) . HEK293 cells were induced with doxycycline for 24 h to express UQCC1-TAP or UQCC2-TAP fusion proteins before being harvested and processed for a single step affinity purification using the Interplay Mammalian TAP kit ( Agilent Technologies ) as per manufacturer's protocol . | Mitochondrial complex III deficiency is a devastating disorder that impairs energy generation , and leads to variable symptoms such as developmental regression , seizures , kidney dysfunction and frequently death . The genetic basis of complex III deficiency is not fully understood , with around half of cases having no known cause . This lack of genetic diagnosis is partly due to an incomplete understanding of the genes required for complex III assembly and function . We have identified two key proteins required for complex III , UQCC1 and UQCC2 , and have elucidated the role of these inter-dependent proteins in the biogenesis of cytochrome b , the only complex III subunit that is encoded by mitochondrial DNA . We have shown that mutations in UQCC2 cause human complex III deficiency in a patient with neonatal lactic acidosis and renal tubulopathy . This work contributes to an improved understanding of complex III biogenesis , and will aid future molecular diagnoses of complex III deficiency . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Mutations in the UQCC1-Interacting Protein, UQCC2, Cause Human Complex III Deficiency Associated with Perturbed Cytochrome b Protein Expression |
Many organisms navigate gradients by alternating straight motions ( runs ) with random reorientations ( tumbles ) , transiently suppressing tumbles whenever attractant signal increases . This induces a functional coupling between movement and sensation , since tumbling probability is controlled by the internal state of the organism which , in turn , depends on previous signal levels . Although a negative feedback tends to maintain this internal state close to adapted levels , positive feedback can arise when motion up the gradient reduces tumbling probability , further boosting drift up the gradient . Importantly , such positive feedback can drive large fluctuations in the internal state , complicating analytical approaches . Previous studies focused on what happens when the negative feedback dominates the dynamics . By contrast , we show here that there is a large portion of physiologically-relevant parameter space where the positive feedback can dominate , even when gradients are relatively shallow . We demonstrate how large transients emerge because of non-normal dynamics ( non-orthogonal eigenvectors near a stable fixed point ) inherent in the positive feedback , and further identify a fundamental nonlinearity that strongly amplifies their effect . Most importantly , this amplification is asymmetric , elongating runs in favorable directions and abbreviating others . The result is a “ratchet-like” gradient climbing behavior with drift speeds that can approach half the maximum run speed of the organism . Our results thus show that the classical drawback of run-and-tumble navigation—wasteful runs in the wrong direction—can be mitigated by exploiting the non-normal dynamics implicit in the run-and-tumble strategy .
Navigation up a gradient succeeds by finding those directions in which signals of interest increase . This can be difficult when the size of the navigator is small compared to the length scale of the gradient because local directional information becomes unreliable . In this case , cells [1 , 2] , worms [3] , larvae [4] , and even robots [5] often adopt a run-and-tumble strategy to navigate . During runs the organism moves approximately straight , collecting differential sensor information in one direction . Tumbles , or reorientations at zero speed , enable the organism to explore other directions . Signal levels are transduced rapidly to the motility apparatus through an internal state variable , so that increases in attractant transiently raise the probability to run longer ( and tumble less ) before a negative integral feedback adapts it back [6] . Classically , averaging over many runs and tumbles results in a net drift up the gradient , although this is usually rather modest because of the occasional runs in the wrong direction . We here focus on the positive feedback inherent to this strategy wherein motion up the gradient lowers the probability to tumble , which further boosts drift up the gradient . Our analysis reveals an unstudied regime in which rapid progress can be achieved . Small fluctuations in the speed of the organism along the gradient grow into large transients in the correct direction but small ones otherwise . We show that this asymmetric amplification arises from the positive feedback , which causes the eigenvectors near the adapted state of the dynamical system to become non-orthogonal , therefore leading to non-normal dynamics . The resulting large transient are further boosted by a nonlinearity that is intrinsic to the positive feedback . Such non-normal dynamics were first discovered in fluid mechanics where they were shown to play an important role in the onset of turbulence in the absence of unstable modes [7 , 8] . Past theoretical studies of run-and-tumble navigation have mostly focused on what happens when adaptation dominates the dynamics ( e . g . [9–13] ) . In this regime , the internal state of the organism exhibits small fluctuations around its mean , and mean field theory ( MFT ) can be applied to make predictions . This approach has been used to describe the motile behavior or populations of E . coli bacteria in exponential ramps [13–15] and oscillating gradients [16] . Beyond the well-understood negative feedback-dominated regime there is a large portion of physiologically relevant parameter space where the positive feedback between movement and sensation dominates the run-and-tumble dynamics . Agent-based simulations have shown that , in this case , large transient fluctuations can emerge in the internal state of an individual organism climbing a gradient , precluding the use of mean field approaches [13] . While systems of partial differential equations ( PDEs ) can be integrated numerically to reproduce these dynamics [17] , a precise understanding of the role of the positive feedback in generating such large fluctuations and the impact of those on the performance of a biased random walk are fundamental questions that remain largely unanswered because of difficulties in obtaining analytical results . Here we develop an analytical model of run-and-tumble gradient ascent that preserves the rich nonlinearity of the problem and incorporates the internal state , 3D-direction of motion , and position of the organism as stochastic variables . We find that large fluctuations in the internal state originate from two key mechanisms: ( i ) the non-normal dynamical structure of the positive feedback that enables small fluctuations to grow , and ( ii ) a quadratic nonlinearity in the speed along the gradient that further amplifies such transients asymmetrically . Utilizing phase space analysis and stochastic simulations , we show how these two effects combine to generate a highly effective “ratchet-like” gradient-climbing mode that strongly mitigates the classic drawback of biased random walks: wasteful runs in wrong directions . In this new regime an organism should be able to achieve drift speeds on the order of the maximum swim speed . Our results are general in that they apply to a large class of biased random walk strategies , where run speed and sampling of new directions may be modulated based on previously encountered signals .
Consider a random walker with an internal state variable F that follows linear relaxation towards the adapted state F0 over the timescale tM , which represents the memory duration of the random walker . We assume that the perceived signal , ϕ ( X , t ) = ϕ ( C ( X , t ) ) , at position X and time t ( here C represents the signal ) , is rapidly transduced to determine the value of an internal state variable F via a receptor with gain N: F ˙ = - F - F 0 t M + N ∂ ∂ t + X ˙ · ∇ ϕ ( X , t ) . ( 1 ) Stochastic switching between runs and tumbles depends on F and follows inhomogeneous Poisson statistics with probability to run r ( F ) = λT/ ( λR + λT ) = 1/ ( 1 + exp ( −HF ) ) , where H is the gain of the motor , and λR ( F ) and λT ( F ) are the transition rates from run to tumble and vice versa [15 , 18] . During runs the speed is constant ∥ X ˙ ∥ = v 0 and the direction of motion is subject to rotational Brownian motion with diffusion coefficient DR . During tumbles the speed is nil and reorientation follows rotational diffusion DT > DR to account for persistence effects [19] . Taken together , these two processes cause the random walker to lose its original direction at the expected rate t D - 1 = ( n - 1 ) ( r D R + ( 1 - r D T ) where n = 2 , 3 for two- and three-dimensional motion respectively . Note that , in this minimal model we ignore possible internal signaling noise [20 , 21] , and all randomness comes from the rotational diffusions DR and DT as well as the stochastic switchings with rates λR ( f ) and λT ( f ) . The effect of signaling noise is considered below using agent-based simulations . Since ϕ ( C ) can be nonlinear , Eq ( 1 ) includes possible effects of saturation of the sensory system . Consider a static one-dimensional gradient and define the length scale of the perceived gradient as L ( X ) = 1 / | | ∇ ϕ ( X ) | | and the direction of motion as s = u ^ · X ^ . Then from Eq ( 1 ) the internal dynamics satisfies the following equations during runs and tumbles , respectively: F ˙ | r u n = - F - F 0 t M + N v 0 L ( X ) s F ˙ | t u m b l e = - F - F 0 t M . ( 2 ) We are interested in the displacement of the random walker along the gradient over timescales longer than individual runs and tumbles . In the limit where the switching timescale tS = 1/ ( λR + λT ) is much shorter than the other timescales we derive from a two-state stochastic model and Eq ( 2 ) ( Methods Eqs ( 13 ) and ( 14 ) ) : r ˙ = r 1 - r - f ( r ) - f 0 t M ︸ negative feedback + r s L ( X ) / ( N H v 0 ) ︸ positive feedback , ( 3 ) where f = HF . The first term is the negative feedback towards the adapted run probability r0 = r ( f0 ) . The second term shows how motion up the gradient ( s > 0 ) causes the probability to run r to feed back on itself—when the organism is oriented up the gradient ( s > 0 ) , F increases only during runs ( Eq ( 2 ) ) , and this increase in turn raises r ( F ) so that the probability that the dynamics of F follows F ˙ | r u n rather than F ˙ | t u m b l e is increased , and so on . A positive feedback is thereby created with characteristic timescale tE = L/ ( NHv0 ) . Steeper gradient ( smaller L ) , stronger receptor gain N or motor gain H , or faster speed v0 , all lead to stronger positive feedback ( shorter tE ) . This important timescale , tE , together with tM ( memory duration ) and tD ( direction decorrelation time ) , effectively determines the dynamics . Expressing time in units of tM , we introduce the following two non-dimensional timescales: τ E ( X ) = t E ( X ) t M = L ( X ) t M N H v 0 τ D ( f ) = t D t M = 1 / t M ( n - 1 ) r ( f ) D R + 1 - r ( f ) D T . ( 4 ) Here τE quantifies the ratio between the negative and positive feedbacks . ( See Table 1 for a summary of the symbols used . ) From above , we expect that the dynamics will depend on how τE and τD compare with one . To explore how run-and-tumble dynamics depend on τE and τD , we used a previously published stochastic agent-based simulator of the bacteria E . coli that reproduces well available experimental data on the wild-type laboratory strain RP437 ( [15 , 22] and S1 Appendix ) . In this case the internal state F represents the free energy of the chemoreceptors . Since E . coli approximately detects log-concentrations ( S1 Appendix Eq ( S11 ) ) , we simulated an exponential gradient so that τE is a constant . In this case the cells reach steady state with a constant drift speed VD . Calculating VD from 104 simulated trajectories for a range of τE and τD0 = τD ( r0 ) values reveals that cells climb the gradient much faster when the positive feedback dominates ( τE < 1 ) ( Fig 1A ) . The trajectories of individual cells resembled that of a ratchet that moves almost only in one direction ( Fig 1B green ) . In contrast , when the negative feedback dominates ( τE > 1 ) the trajectories exhibit both up and down runs of similar although slightly biased lengths ( Fig 1B red ) . VD also depends on τD and peaks when the direction decorrelation time is on the same order as the memory duration ( τD ≃ 1 ) , consistent with previous studies [11 , 23 , 24] . In these simulations the adapted probability to run r0 = 0 . 8 and the ratio DT/DR = 37 were kept constant . Changing these values did not change the main results ( S1A–S1C Fig ) . In a wild type population , individual isogenic cells will have different values of τE and τD0 due to cell-to-cell variabilities in swimming speed and in the abundance of chemotaxis proteins [22 , 25 , 26] . In a recent experimental study , the phenotype and performance of individual wild type cells ( RP437 strain ) was quantified by tracking cells swimming up a static quasi-linear gradient of methyl-aspartate ( varying from 0 to 1 mM over 10 mm ) . This experiment revealed large differences among the performances of individual cells within the isogenic population [22] , which could be reproduced by complementing the model of bacterial chemotaxis just described with a simple model of noisy gene expression ( Fig 2 in [22] ) . To examine in which region of the ( τE , τD0 ) space these cells might have been operating , we used this same model ( complemented with diversity in rotational diffusion coefficients DR and DT due to variations in cell length; see S1 Appendix ) with the same parameter values to run simulations of 16 , 000 cells climbing the experimentally measured ( Fig 1A ) . We find that even in this relatively shallow gradient some cells might have been operating in the positive-feedback-dominated regime , especially near the bottom of the gradient ( black dots ) . As the cells climb the gradient , τE becomes larger ( white dots ) because , as concentration increases , the log-sensing cells in the quasi-linear gradient face a shallower gradient , and thus weaker positive feedback . To better understand the origin of the fast drift speed and its associated “ratchet-like” behavior , we examine the relationship between the drift speed VD and the statistics of the internal state f . Using tM as the unit of time and v0tM as the unit of length we derive a Fokker-Planck equation for the probability P ( x , f , s , τ ) that at time τ = t/tM the cell is at position x = X/ ( v0tM ) with internal state f and orientation s ( Methods Eq ( 14 ) ) : ∂ τ P = - ∂ f - f - f 0 + r ( f ) s τ E ( x ) P + L ^ s P ( n - 1 ) τ D ( f ) - ∂ x r ( f ) s P . ( 5 ) Here L ^ s = ( 1 - s 2 ) 3 - n 2 ∂ s ( ( 1 - s 2 ) n - 1 2 ∂ s ) is the rotational diffusion operator on the ( n − 1 ) -sphere . All symbols used are summarized in Table 1 . For simplicity we consider a log-sensing organism swimming in a static exponential gradient . In this case , τE ( x ) = τE is constant ( more complex gradient profiles and the effect of receptor saturation are considered later in the paper ) . Therefore the positive feedback becomes independent of position and the system can reach a steady state drift speed . Separating the variable x and integrating over x we obtain ( Methods Eq ( 17 ) ) V D = τ E ⟨ f - f 0 ⟩ ¯ , ( 6 ) where 〈⋅〉 represents averaging over f and s and the bar indicates steady state . Eq ( 6 ) indicates that the drift speed is determined by the steady state marginal distribution p ¯ ( f ) . To find an analytical expression for p ¯ ( f ) , we expand the steady state joint distribution P ¯ ( f , s ) in orthonormal eigenfunctions of the angular operator L ^ s—the first two coefficients are the marginal distribution p ¯ ( f ) and the first angular moment p ¯ 1 ( f ) / n = ∫ P ¯ s d s—and discard higher orders to obtain a closed system of equations . The analytical solution for the steady state marginal distribution p ¯ ( f ) reads p ¯ ( f ) = 1 W r ( f ) τ E 1 n r ( f ) τ E 2 - f - f 0 2 exp - ∫ f f 1 - f 0 τ D ( f 1 ) 1 n r ( f 1 ) τ E 2 - f 1 - f 0 2 d f 1 , ( 7 ) where W is a normalization constant . The full derivation is provided in Methods Eqs ( 25 ) – ( 27 ) , together with an interpretation of the distribution as a potential solution p ¯ ( f ) ∝ exp ( - V ( f ) ) where V ( f ) is the “potential” . We also examine how the shape of the potential depends on τE and τD . The solution p ¯ ( f ) is plotted in Fig 1C . When the negative feedback dominates ( τE ≳ 1 ) the distribution is sharply peaked and nearly Gaussian with variance σ 2 = τ D 0 r 0 2 / n τ E 2 ( Methods Eq ( 32 ) ) and its mean barely deviates from the adapted state f0 ( Fig 1C red and blue ) . Substituting p ¯ ( f ) into Eq ( 6 ) and taking the limit τE ≫ 1 yields known MFT results [11–13] ( Methods Eq ( 43 ) ) . When the positive feedback dominates ( τE ≪ 1 ) the distribution p ¯ ( f ) now exhibits large asymmetrical deviations ( Fig 1C green ) between the lower and upper bounds fL and fU , which satisfy the relations fL = f0 − r ( fL ) /τE and fU = f0 + r ( fU ) /τE . For small τE the lower bound decreases as fL → ln τE whereas the upper bound increases as fU → 1/τE ( Methods Eq ( 46 ) ) . MFT becomes inadequate in this regime , as recently suggested by 1D approximations [17] . When the positive feedback dominates , matching the memory of the cell with the direction decorrelation time becomes important: keeping the direction of motion long enough ( τD ≳ 1 ) allows the distribution to develop a peak near fU ( Fig 1C green ) , which according to Eq ( 6 ) results in higher drift speed ( S2 Fig ) . We verified the approximate analytical solution p ¯ ( f ) captures the run-and-tumble dynamics well by plotting it against the distribution of f obtained from the agent-based simulations ( Fig 1C ) . Integrating p ¯ ( f ) according to Eq ( 6 ) predicts well the drift speed for all τE ( Fig 1D ) , including where the positive feedback dominates ( τE < 1 ) . In the fast gradient climbing regime ( τE ≪ 1 ) trajectories resemble that of a ratchet ( Fig 1B ) . To gain mechanistic insight into this striking efficiency we examined the Langevin system equivalent to the Fokker-Planck Eq ( 5 ) . Defining v = rs as the normalized run speed projected along the gradient , we change variables from ( f , s ) to ( r , v ) and obtain ( Methods Eqs ( 47 ) – ( 52 ) ) : d r d τ = r 1 - r - f ( r ) - f 0 + v τ E d v d τ = v 1 - r - f ( r ) - f 0 + v τ E - v τ D ( r ) + r 2 - v 2 τ D ( r ) η ( τ ) , ( 8 ) where v = dx/dτ and η ( τ ) denotes delta-correlated Gaussian white noise . The nullclines of the system ( Fig 2A and 2C ) intersect at the only stable fixed point ( r , v ) = ( r0 , 0 ) of Eq ( 8 ) where the eigenvalues of the relaxation matrix - 1 1 - r 0 r 0 / τ E 0 - 1 / τ D 0 ( 9 ) are both negative ( Methods Eq ( 53 ) ) . Stochastic fluctuations due to rotational diffusions DR and DT ( heat maps in Fig 2A and 2C ) continuously push the system away from the fixed point . The magnitude of these fluctuations is large near the fixed point , causing the system to quickly move away . Fluctuations are smaller near r = 1 and v = 1 , enabling the organism to climb the gradient at high speed for a longer time . Net drift results from spending more time in the region where v > 0 . Stochastic excursions in the ( r , v ) -plane away from the fixed point exhibit distinctive trajectories depending on the value of τE . When the positive feedback dominates ( τE ≪ 1; Fig 2A ) the eigenvectors of the relaxation matrix , ( 1 , 0 ) T and ( ( 1 - r 0 ) r 0 τ E τ D 0 τ D 0 - 1 , 1 ) T , are highly non-orthogonal . This defines a non-normal dynamics that enables linear deviations to grow transiently [7 , 8] to feed the nonlinear positive feedback ( v2 term second line in Eq ( 8 ) ) leading to large deviations . Importantly , this only happens for runs that start in the correct direction . If the run is in the wrong direction the linear deviation does not grow ( Fig 2B; see also S1 Movie ) . Asymmetry arises because the v2 term is always positive . Similar selective amplification properties are observed in neuronal networks , where non-normal dynamics enables the network to respond to certain signals while ignoring others ( including noise ) [27 , 28] . Thus , a random walker running in the correct direction is aided by the positive feedback , which pushes its internal dynamics towards the upper right corner of the phase plane where r = 1 and v = 1 . If , instead , the run is in the wrong direction ( v < 0 ) , the nonlinearity pushes the system back into the high noise region near the fixed point where it will rapidly pick a new direction ( Fig 2B ) . In contrast , when the negative feedback dominates ( τE ≳ 1; Fig 2C ) , the eigenvectors become nearly orthogonal . Linear deviations from the fixed point simply relax to the fixed point regardless of the initial direction of the run . Thus runs up and down the gradient are only marginally different in length , resulting in a small net drift ( Fig 2D ) . This key difference between the positive and negative feedback regimes is reflected in the flow field ( white curves in Fig 2A and 2C ) . For simplicity in our analytical derivations we assumed the environment was a constant exponential gradient with concentrations in the ( log-sensing ) sensitivity range of the organism . Here we explore what happens when the organism encounters concentrations beyond its sensitivity range . For wild type E . coli the change in the free energy of the chemorecetor cluster due to ligand binding is proportional to ln ( ( 1 + C/Ki ) / ( 1 + C/Ka ) ) ( S1 Appendix Eq ( S8 ) ) . Therefore the receptor is log-sensing to methyl-aspartate only for concentrations between Ki ≪ C ≪ Ka , where Ki = 0 . 0182 mM and Ka = 3 mM are the dissociation constants of the inactive and active states of the receptor . When C < Ki the receptor senses linear concentration [29] , whereas when C > Ka the receptors saturate [30–33]: as a cell approaches a high concentration source its sensitivity decreases ( S1 Appendix Eq ( S10 ) ) . This in turn increases the value of τE . Simulations in an exponential gradient show that this effect results in an eventual slow-down as the cell approaches the source ( Fig 3A–3C ) . Realistic gradients are typically limited in spatial extent and often are not exponential , in which case L and therefore τE are different in different regions . L is long near the source in a linear gradient , for example , and decreases linearly with distance from the source . Simulations show that the cell initially climbs the gradient fast but later slows down as the gradient length scale L increases and τE increases ( Fig 3D–3F ) . On the contrary , for a static localized source in three dimensions , L is short near the source but increases linearly with distance from it ( Methods ) . Thus , τE decreases and the cell accelerates as it approaches the source ( Fig 3G–3I ) . Comparing cells in various dynamical regimes ( different values of τE ) across these different gradients suggests that a lower value of τE results in faster gradient ascent . When entering a food gradient , it is natural to try to climb as fast as possible . However this strategy could create a problem: the longer runs implied by the positive feedback mechanism could propel the accelerating E . coli beyond the nutrient source . This is the case in Fig 3E , where cells with the lowest τE ( green ) reach the source first but overshoot slightly; they settle , on average , at a further distance than those with intermediate τE ( blue ) . Thus there is a trade-off between transient gradient climbing and long-term aggregating , as previously observed [13 , 15 , 23] . In nature , as chemotactic bacteria live in swarms , chasing and eating nutrient patches driven by flows and diffusions while new plumes of nutrients are constantly created by other organisms [2 , 34] , the actual environments experienced by bacteria are far more complex . The trade-off we found here hints that in these random small fluctuating gradients [11 , 16 , 35–37] the bacteria should not aim for maximal drift speed but need to deal with this trade-off to avoid overshoot . In general , natural environments will be complex , with a variety of different sources and gradients , implying different parameter domains will be optimal for E . coli at different times . Such phenotypic diversity may well confer an advantage [15 , 37–41] .
Our results illustrate the surprisingly new capabilities that can emerge when living systems exploit the full nonlinearity inherent within an otherwise simple and widely used strategy . For the particular case of bacterial chemotaxis we showed that cells that swim fast , have long memory ( adaptation time ) , or large signal amplification , are likely to exhibit “ratchet-like” climbing behavior in a positive-feedback-dominated regime , even in shallow gradients . As we showed from simulations using a model that fits experimental data , this regime should be accessible to wild-type bacterial populations . Actually identifying these “ratcheting” cells from experimental trajectories would require observing them for a sufficient time ( T ≫ tM , tD , tE ) and in a sufficiently steep gradient over the distance traveled ( ΔX = VDT ∼ 0 . 5v0T ) . Using parameter estimates from [13 , 20] , for tE < tD ≃ tM ∼ 10 s we take T ∼ 200 s , and for v0 = 20 μm/s we get ΔX = 2 mm . To see how this compares with existing experimental setup with a quasi-linear gradient varying from 0 to 1 mM over 10 mm [22] , we note that the black dots in Fig 1A show that some cells located 1 . 5 mm away from 0 concentration can operate in the positive-feedback-dominated regime . Thus , using the same setup as in [22] these requirements would be satisfied near the bottom of the gradient if the source concentration was increased to 3 mM . It is common to make simplifying assumptions to facilitate analysis , but we do not believe that ours are limiting . We showed with simulations that our results hold ( S1 Fig for details ) when we take into account: ( i ) different values of r0 and DT/DR; ( ii ) the limited range of the receptor sensitivity [15 , 18] ( S1 Appendix Eq ( S10 ) ) ; ( iii ) possible nonlinearities ( S1 Appendix Eq ( S4 ) ) and asymmetries of adaptation rates [14 , 42] . A hallmark of E . coli chemotaxis is that , in the absence of a gradient , run-and-tumble behavior adapts back to prestimulus statistics [6 , 43] . These robust properties of integral feedback control [6] remain in place in our study because the transients originate from non-normal dynamics around the stable fixed point . The boost from positive feedback described here is independent from other mechanisms that can enhance drift up a gradient such as imperfect adaptation in the response to some amino acids [44] and stochastic fluctuations in the adaptation mechanism [20 , 21] . The latter has been shown to enhance chemotactic performance in shallow gradients by transiently pushing the system into a regime of slower direction changing provided it is running up the gradient . There are some similarities between the effect of signaling noise and the positive feedback mechanism presented here: both can affect drift speed by causing long-lasting asymmetries in the internal state when running up the gradient . In S3 Fig we show using simulations that signaling noise in the adaptation mechanism does not change our conclusion that the drift speed is maximal in the positive-feedback-dominated regime . Depending on the region of the ( τE , τD ) parameter space , the signaling noise can either enhance the drift speed by less than 10% or reduce it by up to 30% . The fact that non-normal dynamics might be exploited to boost runs in the correct direction parallels recent findings in neuroscience [27] that suggest neuronal networks use similar strategies to selectively amplify neural activity patterns of interest . Thus , non-normal dynamics could be a feature that is selected for in living dynamical systems . Although we used bacterial chemotaxis as an example , our results do not depend on the specific form of the functions r ( f ) and tD ( f ) , provided they are increasing . Therefore our findings should be applicable to a large class of biased random walk strategies exhibited by organisms when local directional information is unreliable . In essence , any stochastic navigation strategy requires a memory , tM , to make temporal comparisons , a reorientation mechanism , tD , to sample new directions , and external information , tE , relayed to decision-making circuitry through motion and signal amplification . Our theoretical contribution showed the ( surprisingly ) diverse behavioral repertoire that is possible by having these work in concert . In retrospect , perhaps this should not be surprising given the diverse environments in which running-and-tumbling organisms can thrive .
We define P R ( X , u ^ , F , t ) and P T ( X , u ^ , F , t ) as the probability distributions at time t to be running or tumbling at position X in direction u ^ with internal variable F . As described , there is Poisson switching between runs and tumbles with rates λR ( F ) and λT ( F ) , runs and tumbles follow rotational diffusion with DR and DT , and motion is constant in runs and 0 in tumbles . Thus we construct a two-state stochastic master equation model [45] ∂ t P R = - ∂ F F ˙ | r u n P R - ∇ · v 0 u ^ P R + ∇ u ^ 2 D R P R - λ R P R + λ T P T ∂ t P T = - ∂ F F ˙ | t u m b l e P T + ∇ u ^ 2 D T P T + λ R P R - λ T P T , ( 10 ) where F ˙ | r u n , t u m b l e are defined in Eq ( 2 ) . Since the gradient varies in one direction only we focus on motion in the gradient direction and integrate the probability over all other directions . Thus ∇ · u ^ = s ∂ X and ∇ u ^ 2 = L ^ s , the polar angle part of the rotational diffusion operator on the ( n − 1 ) -sphere . To derive the analytical form of L ^ s we note in n-dimensional space we can iteratively write down the Laplace-Beltrami operator [46] as ∇ S n - 1 2 = sin θ 2 - n ∂ θ sin θ n - 2 ∂ θ + sin θ - 2 ∇ S n - 2 2 , ( 11 ) where 0 < θ < π is the polar angle . In a one-dimensional gradient we define the gradient direction as the polar axis , thus s = u ^ · X ^ = cos θ . We can write sin θ = 1 - s 2 and ∂ θ = - 1 - s 2 ∂ s . Then the polar angle part is L s = 1 - s 2 3 - n 2 ∂ s ^ 1 - s 2 n - 1 2 ∂ s . ( 12 ) Using the definitions of the normalized internal state f = HF , of the timescale of switching between runs and tumbles tS = 1/ ( λR + λT ) [45] , and of the probability to run r = λT/ ( λR + λT ) , we obtain ∂ t P R = - ∂ f - f - f 0 t M + N H v 0 L s P R + D R L ^ s P R - 1 - r t S P R + r t S P T - s ∂ X v 0 P R ∂ t P T = - ∂ f - f - f 0 t M P T + D T L ^ s P T + 1 - r t S P R - r t S P T . ( 13 ) If we assume the switching terms with t S - 1 in Eq ( 13 ) dominate , the probabilities to be running and tumbling equilibrate on a much faster timescale than the other ones . Therefore we can let P = PR + PT and can approximate the actual probability to run as PR/P ≈ r . Adding the two equations above yields the Fokker-Planck equation: ∂ t P ≈ - ∂ f - f - f 0 t M + r s L / N H v 0 P + r D R + 1 - r D T L ^ s P - r s ∂ X v 0 P . ( 14 ) This is equivalent to a system of Langevin equations . Considering dr/df = r ( 1 − r ) the internal variable dynamics ( the first term on the right ) gives Eq ( 3 ) which defines tE . The angular dynamics ( the second term on the right ) defines tD . Using the time scale definitions in Eq ( 4 ) and non-dimensionalizing time τ = t/tM and position x = X/ ( v0tM ) , we obtain the Fokker-Planck Eq ( 5 ) . From the Fokker-Planck Eq ( 5 ) we consider the steady state so that ∂τ = 0 . For a log-sensing organism moving in an exponential gradient τE does not depend on x . We can therefore integrate over x to get an equation for the marginal steady state distribution P ¯ ( f , s ) —this removes the ∂x term . Integrating over s gives 0 = - ∂ f - f - f 0 ∫ P ¯ w ( s ) d s + r ( f ) τ E ∫ s P ¯ w ( s ) d s , ( 15 ) where the bar indicates steady state . By the boundary conditions that P → 0 at ±∞ , we must have r ( f ) ∫ s P ¯ w ( s ) d s = τ E f - f 0 ∫ P ¯ w ( s ) d s . ( 16 ) From the −∂x ( rsP ) term of the Fokker-Planck Eq ( 5 ) , the spatial flux is r ( f ) s and the drift speed is its average over the distribution . Thus we get the drift speed as Eq ( 6 ) V D = ⟨ r s ⟩ ¯ = ∫ ∫ r ( f ) s P ¯ w ( s ) d s d f = τ E ∫ ∫ f - f 0 P ¯ w ( s ) d s d f = τ E ⟨ f - f 0 ⟩ ¯ . ( 17 ) Here we use separation of variables and expand the solution to the Fokker-Planck Eq ( 5 ) as a sum of eigenfunctions of the operator L ^ s on s . We then ignore high order terms assuming τD0 ≪ 1 and derive an approximate analytical solution . The eigenvalue problem of the angular operator L ^ s , defined in Eq ( 12 ) , is ( 1 - s 2 ) y ″ - ( n - 1 ) s y ′ = λ y . ( 18 ) We identify this as the Gegenbauer differential equation [47] , with eigenfunctions the Gegenbauer polynomials C k ( n / 2 - 1 ) ( s ) and the corresponding eigenvalues λ k ( n / 2 - 1 ) = - k ( k + n - 2 ) . When n = 3 they are Legendre polynomials with eigenvalues λ k ( 1 / 2 ) = - k ( k + 1 ) . The first few Gegenbauer polynomials are C 0 ( n / 2 - 1 ) ( s ) = 1 C 1 ( n / 2 - 1 ) ( s ) = n - 2 s C 2 ( n / 2 - 1 ) ( s ) = n - 2 2 n s 2 - 1 . ( 19 ) They are orthogonal in the sense that ∫−11Ck ( n/2−1 ) ( s ) Cl ( n/2−1 ) ( s ) ( 1−s2 ) n−32ds=Nk ( n/2−1 ) , ( 20 ) where the normalization constants are N k ( n / 2 - 1 ) = π 2 4 - n ( k + n - 3 ) ! k ! ( 2 k + n - 2 ) ( Γ ( n / 2 - 1 ) ) 2 . When n = 3 they are N k ( 1 / 2 ) = 2 2 k + 1 , those of Legendre polynomials . The weight in the integration above is consistent with the geometry on an ( n − 1 ) -sphere Sn−1 , whose the volume element are iteratively defined [46] as d S n - 1 ω = sin θ n - 2 d θ d S n - 2 ω . ( 21 ) After a change of variable s = cos θ and integrating over all remaining dimensions , we see that any integration of s should carry a weight w ( s ) ds = 1 - s 2 n - 3 2 d s . ( 22 ) From orthogonality and completeness , we write any function of s , in particular the probability distribution P , as a series of Gegenbauer polynomials . When n = 3 this is the Fourier-Legendre Series . P ( x , f , s , τ ) = ∑ k = 0 ∞ p k ( x , f , τ ) C k ( n / 2 - 1 ) ( s ) N 0 ( n / 2 - 1 ) N k ( n / 2 - 1 ) = 1 N 0 ( n / 2 - 1 ) p 0 + p 1 n s + p 2 n + 2 n - 1 n s 2 - 1 2 + ⋯ , p k ( x , f , τ ) = ∫ - 1 1 N 0 ( n / 2 - 1 ) N k ( n / 2 - 1 ) C k ( n / 2 - 1 ) ( s ) P ( x , f , s , τ ) 1 - s 2 n - 3 2 d s , ( 23 ) where we normalize the definitions to ensure p 0 = ∫ - 1 1 P ( 1 - s 2 ) n - 3 2 d s is the same as the marginal distribution . When n = 3 , the above is P ( x , f , s , τ ) = ∑ k = 0 ∞ p k ( x , f , τ ) 2 k + 1 2 C k ( 1 / 2 ) ( s ) = 1 2 p 0 + p 1 3 s + p 2 5 2 3 s 2 - 1 2 + ⋯ , p k ( x , f , τ ) = ∫ - 1 1 2 k + 1 C k ( 1 / 2 ) ( s ) P ( x , f , s , τ ) d s . ( 24 ) From now on we denote the marginal distribution p ( f ) = p0 ( f ) . Also , from this definition p 1 = n ∫ - 1 1 s P ( 1 - s 2 ) n - 3 2 d s . Substitute the expansion Eq ( 23 ) into the Fokker-Planck Eq ( 5 ) and use the orthogonality Eq ( 20 ) , we obtain ∂ τ p k = - ∂ f - f - f 0 p k + r ( f ) τ E s ^ k l p l + λ k ( n / 2 - 1 ) ( n - 1 ) τ D ( f ) p k - ∂ x s ^ k l p l , ( 25 ) where s ^ k l = k ( k + n - 3 ) ( 2 k + n - 4 ) ( 2 k + n - 2 ) δ k - 1 , l + ( k + 1 ) ( k + n - 2 ) ( 2 k + n - 2 ) ( 2 k + n ) δ k + 1 , l ( summation over l implied ) is an operator relating neighboring orders . It comes from the positive feedback term . When n = 3 it is s ^ k l = k 4 k 2 - 1 δ k - 1 , l + k + 1 4 ( k + 1 ) 2 - 1 δ k + 1 , l . The first few equations are ∂ τ p = - ∂ f - f - f 0 p + r ( f ) τ E 1 n p 1 - r ( f ) ∂ x 1 n p 1 ∂ τ p 1 = - ∂ f - f - f 0 p 1 + r ( f ) τ E 1 n p + 2 ( n - 1 ) n n + 2 p 2 - 1 τ D ( f ) p 1 - r ( f ) ∂ x 1 n p + 2 ( n - 1 ) n n + 2 p 2 ∂ τ p 2 = - ∂ f - f - f 0 p 2 + r ( f ) τ E 2 ( n - 1 ) n n + 2 p 1 + 3 n n + 2 n + 4 p 3 - 2 n ( n - 1 ) τ D ( f ) p 2 - r ( f ) ∂ x 2 ( n - 1 ) n n + 2 p 1 + 3 n n + 2 n + 4 p 3 . ( 26 ) In the definition of s ^ k l , when k ≫ 1 the non-zero entries approach a constant 1/2 . This means for large k the coefficients pk in Eq ( 25 ) evolve similarly except that higher orders decay with faster rates k ( k + n − 2 ) / ( n − 1 ) τD . Therefore when τD0 ≪ 1 we can neglect the 2nd and higher orders , which closes the infinite series of moment equations and leaves two equations concerning the zeroth and first marginal moments in s , p ( x , f , τ ) and p1 ( x , f , τ ) respectively . At steady state the approximation gives the analytical solution p ¯ ( f ) = 1 W r ( f ) τ E 1 n r ( f ) τ E 2 - f - f 0 2 exp - ∫ f f 1 - f 0 τ D ( f 1 ) 1 n r ( f 1 ) τ E 2 - f 1 - f 0 2 d f 1 , ( 27 ) where W is a normalization constant . Eq ( 27 ) is the same as Eq ( 7 ) in the main text . We can interpret the steady state distribution as a potential solution p ¯ ( f ) ∝ exp ( - V ( f ) ) where V ( f ) is the “potential” . In this case the equivalent “force” in internal state is F ( f ) = - V ′ ( f ) = d ln p ¯ ( f ) d f = d d f ln r ( f ) τ E 1 n r ( f ) τ E 2 - f - f 0 2 - f - f 0 τ D ( f ) 1 n r ( f ) τ E 2 - f - f 0 2 . ( 28 ) Since τD0 ≪ 1 the second term dominates , making the “force” a spring-like system , with spring constant k ( f ) = 1 τ D ( f ) 1 n r ( f ) τ E 2 - f - f 0 2 . ( 29 ) Three observations can be made from this spring constant in intuitively understanding the steady state distribution p ¯ ( f ) . ( i ) k ( f ) →∞ , i . e . the “spring” becomes infinitely “stiff” , when the denominator approaches 0 . Therefore , the bounds of the distribution p ¯ ( f ) are proportional to 1/τE , the ratio between the positive and negative feedbacks ( Eq ( 4 ) ) . Intuitively , a stronger positive feedback ( smaller τE ) drives the internal state f further away from f0 , so the spring constant k ( f ) is smaller and the distribution p ¯ ( f ) is wider . ( ii ) A slower change in direction ( smaller τD ) leads to a larger spring constant k ( f ) ∝ 1/τD ( f ) , and thus the distribution p ¯ ( f ) is more concentrated near the “origin” f0 . Intuitively , a shorter direction correlation time τD inhibits coherent motion in a single direction , which is required by the positive feedback to consistently drive the internal state f away . Thus the distribution p ¯ ( f ) is more concentrated . ( iii ) Asymmetries are created by the functional dependencies of r ( f ) and τD ( f ) , both increasing in f—a “weaker spring” for higher values of f shifts the distribution p ¯ ( f ) there . Intuitively , more positive feedback ∝ r ( f ) and more coherent motion ∝ τD ( f ) in the positive direction asymmetrically drives the internal state towards higher values . These 3 observations can all be found in Fig 1C . We expand the steady state solution Eq ( 27 ) in orders of 1 τ E ⪡ 1 and τD0 ≪ 1 and obtain a near-Gaussian approximation , from which we integrate using Eq ( 6 ) to obtain MFT results . First , we write the steady state distribution Eq ( 27 ) as p ( f ) ¯ = 1 W B ( f ) exp - ∫ f A ( f 1 ) d f 1 . ( 30 ) From the Taylor expansion of the integrand in the exponent A ( f ) = f - f 0 τ D ( f ) 1 n r ( f ) τ E 2 - f - f 0 2 = n τ E 2 r 0 2 τ D 0 f - f 0 + 1 + O ( 1 τ E 2 ) Σ m = 1 ∞ n m + 1 τ E 2 m + 2 r 0 2 m + 2 τ D 0 f - f 0 2 m + 1 - 1 + O ( 1 τ E 2 ) Σ m = 1 ∞ n m τ E 2 m r 0 ′ r 0 2 m + 1 τ D 0 2 m + r 0 τ D 0 ′ r 0 ′ τ D 0 f - f 0 2 m , ( 31 ) where ′ = d/df , we see that if we define σ 2 = r 0 2 τ D 0 n τ E 2 , ( 32 ) the first term in A ( f ) will give - ∫ A ( f ) d f = - ( f - f 0 ) 2 2 σ 2 + . . . . If we can show that the rest of the terms are small when 1 τ E < 1 and τD0 < 1 , we can write p ¯ ( f ) as a small deviation from a Gaussian . Indeed , if we consider the integration range | f - f 0 | ∼ σ ∼ O ( τ D 0 / τ E ) , we can write - ∫ f 0 f A ( f 1 ) d f 1 = - f - f 0 2 2 σ 2 - Σ m = 1 ∞ τ D 0 m 2 m + 2 f - f 0 2 m + 2 σ 2 m + 2 + Σ m = 1 ∞ r 0 ′ r 0 τ D 0 m - 1 2 m + 1 2 m + r 0 τ D 0 ′ r 0 ′ τ D 0 f - f 0 2 m + 1 σ 2 m + O ( 1 τ E 2 ) = - f - f 0 2 2 σ 2 + r 0 ′ r 0 1 3 2 + r 0 τ D 0 ′ r 0 ′ τ D 0 f - f 0 3 σ 2 - τ D 0 4 f - f 0 4 σ 4 + r 0 ′ r 0 τ D 0 5 4 + r 0 τ D 0 ′ r 0 ′ τ D 0 f - f 0 5 σ 4 + O ( 1 τ E 2 ) + O ( τ D 0 2 ) , ( 33 ) Similarly , the prefactor is B ( f ) = n τ E r 0 ( 1 - r 0 ′ r 0 f - f 0 + τ D 0 f - f 0 2 σ 2 - 3 τ D 0 r 0 ′ r 0 f - f 0 3 σ 2 + O ( 1 τ E 2 ) + O ( τ D 0 2 ) ) . ( 34 ) Substitute Eqs ( 33 ) and ( 34 ) back into Eq ( 30 ) and taking care of the orders of all cross terms , we obtain p ¯ ( f ) = 1 Z e - f - f 0 2 2 σ 2 2 π σ 2 ( 1 - r 0 ′ r 0 f - f 0 + τ D 0 f - f 0 2 σ 2 + r 0 ′ 3 r 0 2 - 9 τ D 0 + r 0 τ D 0 ′ r 0 ′ τ D 0 f - f 0 3 σ 2 - τ D 0 4 f - f 0 4 σ 4 + r 0 ′ 60 r 0 103 + 32 r 0 τ D 0 ′ r 0 ′ τ D 0 τ D 0 f - f 0 5 σ 4 - r 0 ′ 12 r 0 2 + r 0 τ D 0 ′ r 0 ′ τ D 0 τ D 0 f - f 0 7 σ 6 + O ( 1 τ E 3 ) + O ( τ D 0 5 2 ) ) . ( 35 ) with normalization constant Z . We notice from Eq ( 30 ) that the range of distribution is bounded by fL and fU , defined by f L - f 0 = - 1 n r ( f L ) τ E , f U - f 0 = 1 n r ( f U ) τ E . ( 36 ) Since σ = r 0 τ D 0 n τ E ⪡ r 0 n τ E , we see that the integration range is much larger than the standard deviation of the Gaussian factor , and thus can be considered from −∞ to ∞ . Therefore we get the normalization constant Z = 1 + τ D 0 4 + O ( 1 τ E 2 ) + O ( τ D 0 2 ) . ( 37 ) Substitute Eqs ( 35 ) and ( 37 ) into Eq ( 6 ) and carry out the integrals V D = r 0 τ D 0 n τ E 1 - 3 4 τ D 0 r 0 ′ + r 0 τ D 0 ′ τ D 0 + O ( 1 τ E ) + O ( τ D 0 3 2 ) 1 + τ D 0 4 + O ( 1 τ E 2 ) + O ( τ D 0 2 ) . ( 38 ) Finally , noticing that by the definition of τD in Eq ( 4 ) τ D ( f ) = τ D 0 r 0 D R + 1 - r 0 D T r ( f ) D R + 1 - r ( f ) D T , ( 39 ) we can get τ D 0 ′ = τ D 0 D T - D R r 0 D R + 1 - r 0 D T r 0 ′ . ( 40 ) Therefore r 0 ′ + r 0 τ D 0 ′ τ D 0 = τ D 0 ′ τ D 0 r 0 D R + 1 - r 0 D T D T - D R + r 0 = τ D 0 ′ τ D 0 D T D T - D R . ( 41 ) Taking DT ≫ DR , we put this back into Eq ( 38 ) and get V D = r 0 τ D 0 ′ n τ E 1 - 3 4 τ D 0 + O ( 1 τ E ) + O ( τ D 0 3 2 ) 1 + τ D 0 4 + O ( 1 τ E 2 ) + O ( τ D 0 2 ) = r 0 τ D 0 ′ n τ E 1 + τ D 0 1 + O ( 1 τ E ) + O ( τ D 0 3 2 ) . ( 42 ) When converted back to real units ( t instead of τ = t/tM ) , the highest-order term is identical , except for notations , to Eq ( 3 ) in Dufour et al . [13] obtained from a different approach . It can also be reduced to Eq ( 12 ) in Si et al . [12] by assuming a high running probability and a long memory . It agrees with Eq ( 6 . 24 ) in Erban & Othmer [48] and Eq ( 16 ) in Franz et al . [49] with appropriate inclusion of rotational diffusion . In Eq ( 38 ) we expanded the distribution as a near-Gaussian around f0 . From Eq ( 6 ) we see the mean internal state f m = 〈 f 〉 ¯ has a slight shift , so it’s more accurate to expand around fm . From Eqs ( 38 ) and ( 6 ) in the main text , we see 〈 f - f 0 〉 ¯ ∼ O ( 1 / τ E 2 ) . Thus considering the shift in fm the resulting VD has the same form compared to Eq ( 42 ) : V D = r m τ D m ′ n τ E 1 + τ D m 1 + O ( 1 τ E ) + O ( τ D m 3 2 ) . ( 43 ) The first term in Eq ( 5 ) says the flux in f-space is non-negative provided − ( f − f0 ) + rs/τE > 0 , or , noting s ≤ 1 f ≤ f 0 + r ( f ) s / τ E ≤ f 0 + r ( f ) / τ E . ( 44 ) Thus the upper bound fU of the distribution p ( f ) is achieved at equality . Similarly , the lower bound fL is achieved when we take equal signs of f ≥ f 0 + r ( f ) s / τ E ≥ f 0 - r ( f ) / τ E , ( 45 ) noting s ≥ 1 . When τE becomes small we note fU , L deviates far away from f0 as 1/τE → ∞ . Using the definition r = 1/ ( 1 + exp ( −f ) ) , we write f L , U = ∓ 1 τ E 1 + exp - f L , U . ( 46 ) The plus sign gives exp ( −fU ) ≪ 1 and fU ≈ 1/τE . The minus sign gives exp ( −fL ) ≫ 1 and fL = −exp ( fL ) /τE . Taking logarithm , the latter gives fL = ln ( |fL|τE ) ≈ ln τE . To derive Langevin equations from the Fokker-Planck equation we need to consider the geometric weight factor w ( s ) in Eq ( 22 ) for anglular integration . In deriving s-dynamics , we start with the angular part of the Fokker-Planck Eq ( 5 ) ∂ τ P = L ^ s P ( n - 1 ) τ D ( f ) + … . ( 47 ) Multiplying an arbitrary function A ( s ) and integrating over all dimensions , we obtain ∫ d x ∫ d f ∫ - 1 1 w ( s ) d s A ( s ) ∂ τ P = ∫ d x ∫ d f ∫ - 1 1 w ( s ) d s A ( s ) 1 - s 2 3 - n 2 ∂ s 1 - s 2 n - 1 2 ∂ s P ( n - 1 ) τ D ( f ) = - ∫ d x ∫ d f ∫ - 1 1 w ( s ) d s s P τ D ( f ) ∂ s A ( s ) + ∫ d x ∫ d f ∫ - 1 1 w ( s ) d s 1 - s 2 P ( n - 1 ) τ D ( f ) ∂ s 2 A ( s ) . ( 48 ) To apply the standard result of equivalence between Fokker-Planck equations and Langevin equations , we need to change the measure in s-space to unity . This prompts the definition Q ( s , t ) = w ( s ) ∬P ( y , f , s , t ) dxdf so that the above becomes ∫ - 1 1 d s A ( s ) ∂ τ Q = - ∫ - 1 1 d s s Q τ D ( f ) ∂ s A ( s ) + ∫ - 1 1 d s 1 - s 2 Q ( n - 1 ) τ D ( f ) ∂ s 2 A ( s ) = ∫ - 1 1 d s A ( s ) ∂ s s Q τ D ( f ) + ∫ - 1 1 d s A ( s ) ∂ s 2 1 - s 2 Q ( n - 1 ) τ D ( f ) , ( 49 ) where we integrated by parts and discarded boundary terms . Since A ( s ) is an arbitrary function , we can write down the Fokker-Planck equation ∂ τ Q = ∂ s s Q τ D ( f ) + ∂ s 2 1 - s 2 Q ( n - 1 ) τ D ( f ) , ( 50 ) which is equivalent [45] to the Langevin equation d s d τ = - s τ D ( r ) + 2 1 - s 2 ( n - 1 ) τ D ( r ) η ( τ ) . ( 51 ) where η ( τ ) denotes the Gaussian white noise with 〈η ( τ1 ) η ( τ2 ) 〉 = δ ( τ1 − τ2 ) . The other two variables follow standard results [45] from the Fokker-Planck Eq ( 5 ) in the main text d f d τ = - f - f 0 + r ( f ) s τ E , d x d τ = r ( f ) s . ( 52 ) Now we change variables according to the definitions r ( f ) = 1/ ( 1 + exp ( −f ) ) and v = rs , and derive from the above dynamics in Eqs ( 51 ) and ( 52 ) to get the Langevin Eq ( 8 ) . Near the fixed point ( r0 , 0 ) , the eigenvectors and eigenvalues of the linearized Langevin Eq ( 8 ) are: 1 0 for eigenvalue - 1 ; 1 - r 0 r 0 τ E τ D 0 τ D 0 - 1 1 for eigenvalue - 1 τ D 0 . ( 53 ) When τE is large , ( 1 - r 0 ) r 0 τ E τ D 0 τ D 0 - 1 ⪡ 1 and the eigenvectors are almost orthogonal . When τE is small , ( 1 - r 0 ) r 0 τ E τ D 0 τ D 0 - 1 ≫ 1 and the eigenvectors are not orthogonal . In Fig 1A heat map the drift speed VD was calculated by fitting the linear part of the mean trajectory . In Fig 1B the first 50 s were removed to avoid the start up transient . In Fig 1C , the steady state p ¯ ( f ) from agent-based simulations was calculated from the histogram of all the internal values of the 104 simulated cells between τ = 10 and τ = 20 , sampled at regular steps of τ = 0 . 01 . Numerical solutions of the Fokker-Planck Eq ( 5 ) were obtained by expanding the distribution in angles , as in Eq ( 25 ) , and keeping the first 10 orders . The steady state p ¯ ( f ) was found by solving an initial value problem using the NDSolve function in Mathematica , with 104 spatial points and integration time up to τ = 10 . Further orders , finer grid , and longer integration times were checked to ensure solution accuracy . In Fig 1D , VD from agent-based and Fokker-Planck were calculated by plugging into Eq ( 6 ) p ¯ ( f ) obtained from those methods in C . MFT was calculated by combining Eq ( 42 ) with Eq ( 6 ) to find both f m = 〈 f 〉 ¯ and VD [12 , 13] . In the inset , the black curves show the approximate distribution in Eq ( 35 ) . In Fig 2B and 2D the Langevin trajectories were generated using Euler’s method to integrate Eq ( 8 ) . In Fig 3C , 3F and 3I the τE calculation considered receptor saturation as well as the varying gradient length scales , with C and L evaluated at mean positions . Note this is not the average τE over the population . | Countless bacteria , larvae and even larger organisms ( and robots ) navigate gradients by alternating periods of straight motion ( runs ) with random reorientation events ( tumbles ) . Control of the tumble probability is based on previously-encountered signals . A drawback of this run-and-tumble strategy is that occasional runs in the wrong direction are wasteful . Here we show that there is an operating regime within the organism’s internal parameter space where run-and-tumble navigation can be extremely efficient . We characterize how the positive feedback between behavior and sensed signal results in a type of non-equilibrium dynamics , with the organism rapidly tumbling after moving in the wrong direction and extending motion in the right ones . For a distant source , then , the organism can find it fast . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"cell",
"motility",
"medicine",
"and",
"health",
"sciences",
"social",
"sciences",
"mathematical",
"models",
"neuroscience",
"biological",
"locomotion",
"simulation",
"and",
"modeling",
"systems",
"science",
"mathematics",
"algebra",
"research",
"and",
"analysis",
"meth... | 2017 | Feedback between motion and sensation provides nonlinear boost in run-and-tumble navigation |
Kaposi's sarcoma-associated herpesvirus ( KSHV ) is tightly linked to at least two lymphoproliferative disorders , primary effusion lymphoma ( PEL ) and multicentric Castleman's disease ( MCD ) . However , the development of KSHV-mediated lymphoproliferative disease is not fully understood . Here , we generated two recombinant KSHV viruses deleted for the first RBP-Jκ binding site ( RTA1st ) and all three RBP-Jκ binding sites ( RTAall ) within the RTA promoter . Our results showed that RTA1st and RTAall recombinant viruses possess increased viral latency and a decreased capability for lytic replication in HEK 293 cells , enhancing colony formation and proliferation of infected cells . Furthermore , recombinant RTA1st and RTAall viruses showed greater infectivity in human peripheral blood mononuclear cells ( PBMCs ) relative to wt KSHV . Interestingly , KSHV BAC36 wt , RTA1st and RTAall recombinant viruses infected both T and B cells and all three viruses efficiently infected T and B cells in a time-dependent manner early after infection . Also , the capability of both RTA1st and RTAall recombinant viruses to infect CD19+ B cells was significantly enhanced . Surprisingly , RTA1st and RTAall recombinant viruses showed greater infectivity for CD3+ T cells up to 7 days . Furthermore , studies in Telomerase-immortalized human umbilical vein endothelial ( TIVE ) cells infected with KSHV corroborated our data that RTA1st and RTAall recombinant viruses have enhanced ability to persist in latently infected cells with increased proliferation . These recombinant viruses now provide a model to explore early stages of primary infection in human PBMCs and development of KSHV-associated lymphoproliferative diseases .
Kaposi sarcoma-associated herpesvirus ( KSHV , also known as human herpesvirus 8 [HHV8] ) infection is pivotal to the development of Kaposi sarcoma ( KS ) . KSHV is also strongly associated with two lymphoproliferative diseases , primary effusion lymphoma ( PEL ) and Multicentric Castleman's disease ( MCD ) [1] , [2] . During its lifespan , KSHV undergoes latent and lytic cycle replication ( reactivation ) . In comparison to lytic cycle replication , fewer genes are expressed in latent infection and a number of these genes are involved in disruption of the cell cycle , and in maintenance of the viral genome . One of those latent genes is Latency-associated nuclear antigen ( LANA ) , encoded by KSHV open reading frame 73 ( ORF73 ) , which is critical for persistence of the viral episome and maintenance of latent infection in KSHV infected cells [3] . During lytic cycle replication , almost all viral genes are expressed in a staged temporal manner . The replication and transcription activator ( RTA ) is encoded by KSHV ORF50 and plays an essential role in the control of the lytic replication cycle . RTA can activate KSHV lytic genes including ORF6 ( single-stranded DNA-binding , SSB ) , ORF21 ( thymidine kinase , TS ) , ORF57 ( mRNA transcript accumulation . MTA ) , ORF59 ( polymerase processivity factor , PF-8 ) , ORF 74 ( vGPCR ) , K2 ( vIL-6 ) , K5 ( MIR-2 ) , K6 ( vMIP-1 ) , K8 ( k-bZIP ) , K9 ( vIRF ) , K12 ( kaposin ) , K14 ( vOX-2 ) and polyadenylated nuclear ( PAN ) through direct binding with high affinity to RTA-responsive elements ( RREs ) or in combination with cellular transcription factors , RBP-Jκ , Ap-1 , C/EBP-α , Oct-1 , and Sp1[4] , [5] , [6] , [7] , [8] , [9] , [10] , [11] , [12] , [13] , [14] , [15] , [16] , [17] , [18] , [19] , [20] , [21] , [22] . Recombinant viruses that lack RTA establish latency quite efficiently but are unable to reactivate [23] . Our earlier studies also suggest that RTA contributes to the establishment of KSHV latency by activating LANA expression during the early stages of infection through the major effector of the Notch signaling pathway , recombination signal binding protein Jκ ( RBP-Jκ ) . This mutual RTA/LANA feedback regulatory mechanism is likely to be a key event in establishment of KSHV latency and is yet to be completely elucidated . RBP-Jκ , also named CBF1 or CSL , is a member of the CSL family ( CBF1 , Suppressor of Hairless , and Lag ) and is the major downstream effector of the Notch signaling pathway [24] , [25] . RBP-Jκ functions on the target gene by recruiting distinct protein complexes to the promoter . RTA mimics Notch signaling and can activate target promoters by binding to the repression domain of RBP-Jκ , thereby activating promoters [17] . In KSHV-infected cells , RBP-Jκ mediates cooperative transactivation of KSHV genes , ORF57 , K-bZIP , ORF6 ( SSB ) , K14 ( vGPCR ) , LANA , K8 , ORF47 and RTA [12] , [15] , [17] , [26] , [27] , [28] , [29] . The KSHV RTA promoter contains four potential RBP-Jκ binding sites [30] and mutation of the first and third RBP-Jκ sites within the KSHV RTA promoter results in approximately 50% repression of RTA expression in vitro [30] . Understanding the pathogenesis of specific concerns including Kaposi's sarcoma has been aided by development of model systems [31] . It has been difficult to establish lymphoblastoid cell lines in culture by KSHV , which has slowed the understanding of the natural mechanism of KSHV-mediated lymphoproliferative disease . However , the mechanism of differentiation and proliferation due to EBV infection in B lymphocytes are well documented [32] . Recently , two groups have shown that KSHV infects a subset of tonsillar B cells driving plasmablast differentiation and proliferation , and KSHV-encoded viral FLICE-inhibitory protein ( vFLIP ) induces B lymphocytes transdifferentiation and tumorigenesis in an animal model [33] , [34] . Furthermore , Myoung and Ganem reported that T and B lymphocytes in primary human tonsils can be infected by KSHV , with B lymphocytes producing a substantial amount of infectious virions [35] , [36] . In contrast to EBV , transformation and immortalization have not been clearly observed in KSHV infected B or T cells . KSHV infection of peripheral blood mononuclear cells ( PBMCs ) occurs prior to the onset of KS [37] , and KSHV DNA is frequently detected in immune-deficient patients [38] , [39] , [40] , [41] . In addition , PBMCs of KSHV infected marmosets support viral infection and replication [42] . Finally , we recently showed that PBMCs exposed to virions from BAC36-293 cells can effectively model early infection [43] . Our previous studies showed that mutation of the RTA promoter in vitro ( first and all three RBP-Jκ binding sites ) led to significant decreases in RTA activity [30] . RTA is an immediate early protein that serves as the master switch for viral lytic replication . The interaction between RTA and RBP-Jκ controls the activation of multiple viral target genes which are absolutely critical for virus reactivation [17] , [26] , [27] . Based on these observations , we generated two subtly mutated recombinant KSHV viruses based on BAC36 ( wt ) , RTA1st ( deletion of the first RBP-Jκ binding site in the RTA promoter ) , and RTAall ( deletion of all three RBP-Jκ binding sites in the RTA promoter ) . Our results show that RTA1st and RTAall recombinant viruses had enhanced ability to maintain viral latency and decreased capability to drive lytic replication in infected cells . This resulted in an increase in cell growth and proliferation . Furthermore , RTA1st and RTAall recombinant viruses were enhanced in their ability to infect human PBMCs . Both recombinant viruses infected T and B cells during primary infection , resulting in increased infectivity during the early stage of infection . Long-term persistence of the viral episome in infected cells further confirmed that RTA1st and RTAall recombinant viruses had a greater ability to maintain latent infection and enhanced proliferation . Our study provides specific insights into the contribution of KSHV to its associated lymphomas .
Human embryonic kidney 293 ( HEK 293 ) cells were maintained in Dulbecco's modified Eagle medium ( DMEM ) supplemented with 5% bovine growth serum . De-indentified Human peripheral blood mononuclear cells ( PBMCs ) were obtained from the University of Pennsylvania CFAR Immunology Core . The Core maintains an IRB approved protocol in which Declaration of Helsinki protocols were followed and each donor gave written , informed consent . Telomerase-immortalized human umbilical vein endothelial ( TIVE ) cells were a kind gift from Dr . Rolf Renne [44] . The wild-type KSHV BACmid , BAC36 wt , was provided by S . J . Gao ( University of Texas , San Antonio , TX ) . Kanamycin ( Kn ) cassette containing plasmid , pL452 was obtained from the National Cancer Institute Biological Resources Branch . Mutagenesis of BAC36 was performed using the Red Recombination method as described previously [45] . The primers used were BAC 36-RTA1st , the forward PCR primer 5′- caaaactgtgtttagtagcaacacaccctggcgagcccagctgtcgaggcCAATTCCGATCATATTCAATAACCCTTAAT -3′ ( target sequence is low-cased ) , and the reverse primer , 5′- tttgagaagcatctttagagagctagaggcttccgtccccaatttcagtaAGAACTAGTGGATCCCCTCGAGGGACCTA -3′ were used to amplify a Kan resistance cassette ( underlined sequence is Kan cassette primer ) flanked by BAC sequences ( genomic position 69057 and 69171 , NC_009333 ) ; BAC 36-RTAall , the forward PCR primer 5′- caaaactgtgtttagtagcaacacaccctggcgagcccagctgtcgaggcCAATTCCGATCATATTCAATAACCCTTAAT -3′ ( target sequence is low-cased ) , and the reverse primer , 5′- atggcgacgtgcactactcgggacccccgcgcaccccggcatatggagtaAGAACTAGTGGATCCCCTCGAGGGACCTA -3′ were used to amplify a Kan resistance cassette ( low-cased sequence is Kan cassette primer ) flanked by BAC sequences ( genomic position 69057 and 70417 , NC_009333 ) . The Kan cassette flanked with 50bp KSHV genomic sequence was electroporated into EL350 containing BAC36 at 1 . 75kV , and the resulting colonies plated on Kanamycin ( 50 µg/ml ) and chloramphenicol ( 25 µg/ml ) double selection plates followed by incubation at 30°C . BAC plasmid DNA was isolated from 10-ml overnight cultures by the alkaline lysis procedure and characterized by restriction enzyme analysis followed by Southern blot analysis . The transformed single colonies were induced with 10% L ( + ) arabinose ( Sigma-Aldrich , Inc . , St . Louis , MO ) for 1h , then plated to Chloramphenicol and Kanamycin selection plates , separately . The resulting single colonies on Chloramphenicol plates were inoculated in 10-ml of LB with chloramphenicol overnight at 30°C . Small-scale DNA isolation was performed to characterize the DNA of specific mutants , followed by Southern blot analysis . All the clones were confirmed by DNA sequencing using the University of Pennsylvania Perelman School of Medicine sequencing core . Large preparations of KSHV BAC plasmids were obtained from 500-ml E . coli cultures with the Qiagen Large construct kit ( Qiagen , Inc . , Valencia , CA ) according to manufacturer instructions . The DNA probe used for Southern blot hybridizations was amplified as a 1 . 4-kb fragment ( corresponding to the RBP-Jκ locus of RTA promoter ) with the KSHV genome as the template and primer set forward: 5′-ATGCAGCGGGGTGAGCCTGCCTCCAGCC-3′ , and reverse , 5′-TTGCAGAATACTGGACAACAGCGCGTCG-3′ . Purified KSHV BAC plasmid DNA was digested with XhoI and resolved on 0 . 65% agarose gels in 0 . 5X Tris-borate-EDTA buffer for 14 to 18 h at 40 V . DNA fragments were visualized by ethidium bromide staining , denatured , and transferred to Zeta-Probe GT genomic tested blotting membranes ( Bio-Rad Inc , Hercules , CA ) . DNA probes were radiolabeled with [α-32P] dCTP with the NEBlot ( New England Biolabs , Inc . , Ipswich , MA ) . Prehybridization was performed at 63°C for 1 h in hybridization buffer ( 7% sodium dodecyl sulfate , 10% polyethylene glycol , 1 . 5X SSPE [1X SSPE is 0 . 18 M NaCl , 10 mM NaPO4 , and 1 mM EDTA , pH 7 . 7] ) . DNA blots were hybridized with radiolabeled probes in the same solution at 63°C for about 7 h . Blots were washed twice for 15 min with 2X SSC ( 1x SSC is 0 . 15 M NaCl plus 0 . 015 M sodium citrate ) -0 . 1% sodium dodecyl sulfate and twice for 30 min with 0 . 1X SSC-0 . 1% sodium dodecyl sulfate at 63°C . Blots were exposed to a Phosphoimager plate ( Molecular Dynamics , Inc . Sunnyvale , CA ) overnight at room temperature followed by scanning with the Typhoon 9200 ( GE Healthcare Inc . , Piscataway , NJ ) . Junction PCRs were performed to identify the expected deletions . The primers used were RTA1st ( genomic position 69010 to 69291 , NC_009333 ) 5′ TCCCAGCCAAGTCCCTCGTG 3′ and 5′ GTCCCACTGCTGCGATCCAG 3′; RTAall , ( genomic position 69010 to 70537 , NC_009333 ) 5′ TCCCAGCCAAGTCCCTCGTG 3′ and 5′ GCCCGGATACGCGCACATGC 3′ . Purified Bac36 DNAs were transfected into 293 cells via CaPO4 method . Hygromycin B ( 150 ng/ml ) was then added for selection 24 h after transfection . Three weeks after selection , homogenous populations of GFP-positive cells harboring KSHV Bac36 DNAs were obtained . Butyric Acid at a final concentration of 3 mM and TPA ( Sigma ) at 20 ng/ml was used for lytic induction . Cell suspensions were centrifuged at 3000 rpm for 20 min and the supernatant was filtered through a 0 . 45 µm cellulose acetate filter . The viral particles were concentrated by ultracentrifugation at 70 , 000xg at 4°C and stored at −80°C . Infection of PBMCs were performed as described previously [43] . In brief , 1x107 were infected by incubation with virus suspension in 1ml of RPMI 1640 ( with 10% FBS ) medium in the presence of Polybrene at a final concentration 5 ng/µl ( Sigma , Marborough , MA ) and incubated for 4 h in 37°C . Cells were centrifuged for 5min at 1500rpm , the supernatant discarded , pelleted cells were washed by fresh RPMI medium for 2 times and resuspended in fresh RPMI 1640 ( 10% FBS ) medium in 6-well plates and culture at 5% CO2 , 37°C humidified incubator . TIVE cells were cultured in 12-well plates to 60% confluence in Medium 199 supplemented with 20% FBS , 200 mM L-glutamine , 5 mg/ml Penicillin/Streptomycin ( P/S ) and 10 mg/ml Endothelial cell growth supplement from bovine neural tissue . Concentrated viruses were added to the supernatant in the presence of Polybrene ( 4 µg/ml ) and spun at 2 , 500 rpm for 1 h at room temperature . The supernatants were removed and washed twice and then incubated with fresh medium . GFP expression was used to monitor infection under fluorescence microscope ( Olympus Inc . , Melville , NY ) . For quantitation of intracellular viral DNA , cells were harvested and washed twice with 1xPBS to remove the residual viruses . Cells were incubated by HMW buffer ( 10mM Tris-HCl pH 8 . 0 , 150mM NaCl , 10mM EDTA , 0 . 5% SDS ) for 2 hrs at 55°C . 0 . 5 mg/ml proteinase K was added and incubated at 37°C overnight with subsequent extraction in phenol/chloroform/isopropanol . Viral DNA was treated with RNase , then precipitated and resuspended in water . Extracellular viral DNA was extracted from culture supernatants as essentially described previously [35] , [43] . In brief , virions were pelleted down at 70 , 000xg for 2 hrs at 4°C and resuspended . Cellular DNAs and free viral DNAs were removed by treatment with DNase I at 37°C for 1∼2 hrs . Virion DNA was treated with HMW buffer for 20 minutes . Lysates were treated with proteinase K overnight at 37°C with subsequent extraction with phenol/chloroform/isopropanol . Intracellular and extracellular viral DNAs were quantitated by real-time DNA PCR for TR ( 5′ GGCTCCCCCAAACAGGCTCA 3′ , and 5′ GGGGGACCCCGGGCAGCGAG 3′ ) . GAPDH and BAC36 DNAs were the standards for intracellular and extracellular viral DNAs , respectively . Total RNA from infected PBMCs were extracted by using TRIzol ( Invitrogen , Inc . , Carlsbad , CA ) and 1 µg DNase-treated total RNA were used to generate cDNA using the High capacity RNA-to-cDNA kit ( Applied Biosystems Inc . , Foster City , CA ) according to manufacturer's instructions . RT-qPCR was performed on a StepOnePlus Real-Time PCR System ( Applied Biosystems Inc , Carlsbad , CA ) or Opticon 2 Real-Time PCR System . The reactions were carried out in a 96-well plate at 95°C for 10 min , followed by 35 cycles at 95°C for 30 s , 51°C for 30 s and then 72°C for 40 s . The differences of cycle threshold values ( CT ) between the samples ( ΔCT ) were calculated after standardization by GAPDH and converted to fold changes using one of the samples as a standard ( 1-fold ) . The primers used were LANA: 5′ CATACGAACTCCAGGTCTGTG 3′ , 5′ GGTGGAAGAGCCCATAATCT 3′; RTA: 5′ CAGACGGTGTCAGTCAAGGC 3′ , 5′ ACATGACGTCAGGAAAGAGC 3′; GAPDH 5′ GGTCTACATGGCAACTGT GA 3′ , 5′ ACGACCACTTTGTCAAGCTC 3′ . All the reactions were run in triplicates . Cells were applied to a slide well and fixed with 4% paraformaldehyde with 0 . 1% Triton X-100 , and blocked with 10% BSA . Cells were then incubated with a primary antibody ( Mouse anti-LANA ) , and specific signals were detected with a secondary antibody conjugated with Alexa Fluor 594 ( Invitrogen , Carlsbad , CA ) . The cells were counterstained with 4′ , 6′-diamidino-2-phenylindole ( DAPI ) . Images were observed and recorded with a Fluoview FV300 microscope ( Olympus Inc . , Melville , NY ) . BAC36 wt , BAC RTA1st and BAC RTAall transfected 293 cells were selected for 3-4 weeks with Hygromycin B . 300 stably transformed 293 cells were seeded in 6 cm Petri dish in DMEM supplemented with 10% FBS , Hygromycin B ( 150 ng/ml ) and Difco Noble Agar ( BD , Franklin Lakes , NJ ) ( 0 . 5% ) . After 2 weeks of growth , the colony were monitored and photographed under fluorescence microscope ( Olympus Inc . , Melville , NY ) . The plates were scanned by Typhoon 9200 for GFP signals and the colony number was quantitated using the Odyssey V3 . 0 software . BAC36 wt , BAC RTA1st and BAC RTAall stably transfected 293 cells were harvested and fixed in 1% paraformaldehyde for 30–60 min . The fixed cells were washed twice by 1XPBS and analyzed using on FACSCalibur based on GFP signals . Infected PBMCs cells were stained essentially as described previously [46] , [47] . Briefly , PBMCs were harvested and washed at 1dpi , 2dpi , 4dpi and 7dpi . T cells and B cells were detected by using the APC conjugated anti-CD3 and PercpCy 5 . 5 conjugated anti-CD19 mAbs ( BD Biosciences , San Jose , CA ) . GFP signals were used to monitor KSHV positive cells . Data were acquired on FACSCalibur equipped with CellQuest Pro software and analyzed using FlowJo software . Data are shown as mean values with standard errors of the means ( SEM ) . The significance of differences in the mean values was evaluated by 2-tailed Student's t test . P<0 . 05 was considered statistically significant .
Our previous studies showed that RTA contributes to establishment of KSHV latency by activating LANA expression during the early stages of infection via RBP-Jκ , the major effector of the Notch signaling pathway [28] . The activity of the RTA promoter was reduced about 40% in the truncations of first RBP-Jκ and all three RBP-Jκ binding site within RTA promoter compared to the wt promoter [30] . To investigate the roles of the RBP-Jκ binding sites in the RTA promoter , we constructed two KSHV recombinant viruses , BAC RTA1st with a deletion of the first RBP-Jκ binding site and RTAall with deletion of all three RBP-Jκ binding sites . BAC36 wt carries the full KSHV genome , a GFP tag , and a eukaryotic resistance gene , hygromycin [48] . Infectious KSHV can be reconstituted by transfection of BAC36 wt DNA into 293 cells [48] . Using the BAC36 wt as a template , we designed PCR primers so that the RBP-Jκ binding sites in the RTA promoter were removed from the genome . Fig . 1A and B shows a schematic for the generation of the recombinant BAC RTA1st and RTAall using PCR primers that integrated the Kanamycin resistance gene ( Neo: the neo cassette is resistant to Neomycin or Kanamycin in prokaryotes ) and two loxP sites from plasmid pL452 into the BAC36 genome , replacing the RBP-Jκ binding sites of the RTA promoter . LoxP is the substrate sequence of Cre recombinase , so the insert fragment between two loxP sites ( including neo cassette ) can be subsequently removed by expressing Cre recombinase after induction by L-arabinose [49] . A PCR product containing Neor flanked by loxP sites and two fragments of 50-bp KSHV sequences from the two ends of the RBP-Jκ site in the RTA promoter was generated using pL452 plasmid as a template . This PCR product was transfected into BAC36 wt-E . coli 350 to remove the RBP-Jκ site in the RTA promoter after homologous recombination and Cre-mediated excision of Neor . The resulting BAC recombinants were screened and analyzed on 0 . 65% agarose and subsequently by southern blot analysis to show that the RBP-Jκ binding site in the RTA promoter was removed from the KSHV genome ( Fig . 1A and C ) . Digestion of the BAC36 wt DNA with XhoI generated one 12584kb fragment at the RBP-Jκ binding site in the RTA promoter . For BAC RTA1st , replacement of the RBP-Jκ binding site with the Kan cassette changes the two fragment sizes to 9486bp and 4973bp . After induction , the fragment between two loxP sites was removed , so the smaller fragment ( 4973b ) shifted in size to 3176kb , indicating removal of Kan cassette . Southern blot showed the presence of a 5kb band before induction and a unique 3kb band in recombinant BAC RTA1st when hybridized with a probe within the RBP-Jκ binding site ( Fig . 1A and C ) . To further confirm whether the altered digestion pattern of the BAC mutants was the result of the expected recombination , we carried out junction PCR by using the primers designed at the recombination site showing that the junction bands in the BAC RTA1st shifted based on the presence of the remaining loxP site and XhoI site ( Fig . 1D ) . Similarly , For BAC RTAall , replacement of the RBP-Jκ binding sites with the Kanamycin cassette changes the two fragment sizes to 8240bp and 4973bp . After induction , the fragment between two loxP sites was removed , so the smaller fragment ( 4972b ) shifted in size to 3176kb - indicating removal of the Kan cassette . Southern blot showed the presence of a 5kb band before induction and a unique 3kb band in the recombinant BAC RTAall when hybridized with a probe within the RBP-Jκ binding site ( Fig . 1B and E ) . Junction PCR showed that the junction bands in the BAC RTAall shifted based on the presence of the remaining loxP site and XhoI site ( Fig . 1F ) . Finally , the PCR products were sequenced to confirm the expected mutation . To reconstitute recombinant viruses , we transfected BAC36 wt , RTA1st and RTAall DNAs into 293 cells . The transfection efficiencies were monitored by the fluorescence microscopy . GFP positive cells were detected after 24–48h post-transfection ( data not shown ) . Positive cells were enriched by hygromycin selection , generating 293 cell lines bearing BAC36 wt , RTA1st and RTAall ( Fig . 2A ) . Subsequently , the cells were fixed and immunostained against LANA to confirm that BAC 36 wt and RTA1st and RTAall stable cell lines harbored the KSHV genome ( Fig . 2B ) . Similar levels of GFP positive signals and LANA staining were observed for all three stably transfected cell lines generated due to hygromycin selection . To ensure that both RTA1st and RTAall stable cells were able to produce recombinant viruses after lytic reactivation , the cells were treated with TPA and butyric acid to induce lytic reactivation . Whole cell lysates were prepared from the uninduced and induced cells at 24 , 48 , 72 hours post induction ( hpi ) , and the expression of LANA and RTA were analyzed by western blot analysis using the corresponding specific antibodies . The results showed that LANA expression exhibits a slight decrease after induction in BAC36 wt-293 cells at 48 and 72 hpi as more cells switch to lytic replication . However , there was no obvious difference in RTA1st and RTAall-293 cells ( Fig . 3A ) . This may be due to the fact that more cells were infected with RBP-Jκ mutant viruses in latent phase with higher levels of LANA . In the wild type , RTA expression was not affected thus most of the viral genome copies underwent lytic reactivation thereby expression of latent protein , LANA diminished over time . This is further evident by the levels of RTA , which increased , in a time dependent manner after induction with TPA and butyric acid , in wt BAC36-293 cells ( Fig . 3A ) . Notably , though RTA expression was increased in a time-dependent manner after induction in all cell lines , the levels of RTA expression in RTA1st and RTAall-293 was much lower than seen in BAC36 wt-293 cells . This suggests that RTA expression was reduced in RTA1st and RTAall-293 cells ( Fig . 3A ) . In addition , the viruses were collected from supernatant of the induced BAC36 wt , RTA1st and RTAall-293 cells and quantitated for virion particles by quantitative PCR analysis . The results showed that RTA1st and RTAall-293 cells produced fewer KSHV genomes , suggesting a decrease in virion production post-induction ( Fig . 3B ) . This confirms our hypothesis that deletion of the RBP-Jκ binding sites in the RTA promoter of KSHV results in an attenuated lytic cycle and thus decreased viral progeny . Furthermore , total DNA was extracted from BAC36 wt , RTA1st and RTAall-293 cells . Intracellular viral DNA levels were determined by quantitative PCR analysis standardized by GAPDH . The result showed that RTA1st and RTAall-293 cells had a greater number of viral copies compared to BAC36 wt-293 cells , further indicating that RTA1st and RTAall deficient viruses exhibited a decrease in lytic capability although the genome copy numbers were greater ( Fig . 3C ) . BAC36 wt , RTA1st and RTAall are recombinant KSHV viruses harboring a GFP marker which allows us to track viral genome stability in 293 cells . Furthermore , RTA1st and RTAall 293 cells possessed higher intracellular viral DNA copies than BAC36 wt-293 cells . We postulated that both of them should have increased GFP signals . Flow cytometry was performed based on GFP signals for BAC36 wt , RTA1st and RTAall -293 cells . Interestingly , RTA1st and RTAall-293 cells showed approximately 78% and 93% GFP fluorescence intensity , respectively . However , only 41% GFP fluorescence intensity was seen for BAC36 wt-293 cells ( Fig 4A , B ) and pellets from collected RTA1st and RTAall-293 cells also had a more intense green color based on visual inspection when compared to BAC36 wt-293 cells ( data not shown ) . KSHV is a known human oncovirus and is associated with cellular transformation [31] , [50] , [51] . Therefore we wanted to determine the proliferation rate for BAC36 wt , RTA1st and RTAall-293 cells . Cells were starved in DMEM with 0 . 1% FBS overnight . Next day , media were replaced with DMEM supplemented with 5% FBS . Cells were cultured for 24 hrs and harvested for analysis by flow cytometer . Interestingly , RTA1st and RTAall-293 cells had a higher percentage of cells ( 55 . 7% and 59 . 4% ) comparing to BAC36 wt-293 ( 51 . 8% ) in G1 phase which was consistant after multiple repeats ( Fig . 4C ) . These results suggested that RTA1st and RTAall recombinant viruses can promote cell growth and proliferation in the infected cells . In addition , cells harboring more viral genome copies should have an enhanced capability for driving cell growth . Therefore we tested this hypothesis with a colony formation assay . The colonies were photographed using fluorescence microscopy and scanned by a Typhoon 9200 system based on GFP signals . Surprisingly , the average size of hygromycin-resistant colonies for RTA1st and RTAall-293 cells was distinctively bigger ( almost double ) relative to BAC36 wt-293 cells ( Fig . 4D ) , indicating that RTA1st and RTAall recombinant viruses possess enhanced capability for cell growth . Furthermore , GFP positive colonies were scanned and quantitated . The results also showed that RTA1st and RTAall-293 cells showed a 1 . 8 and 2 . 4 fold increase in total colonies in comparison to BAC36 wt-293 ( Fig . 4E , F ) , providing further evidence that mutation of the RBP-Jκ sites enhanced KSHV latent infection and so promoted cell growth and proliferation . KSHV infection is strongly linked to two lymphoproliferative diseases , referred to as PEL and MCD [1] , [2] . Infection of PBMCs with KSHV can provide greater insights into initiation and development of KSHV associated lymphomagenesis . KSHV can infect many cell types , including epithelial cells , keratinocytes , endothelial cells and PBMCs which are permissive cells with varying degrees of infectivity [36] , [44] , [52] . Previous studies show that determination of GFP signals can be effectively used to monitor the efficiency by which KSHV BAC36 infects PBMCs [43] . Our investigation shows that mutations of the RTA promoter , specifically the RBP-Jκ cognate sequences , results in reduced RTA expression as well as reduced RTA-mediated auto-activation of its promoter [30] . Here , we have now developed a BAC system to further explore the feedback regulation of RTA by LANA . As described previously , GFP signals were monitored by green fluorescence from the infected PBMCs [43] . We concentrated BAC36 wt , RTA1st and RTAall viruses from TPA and Butyric acid treated-293 cells . The concentrated viruses were used to infect PMBCs and the infected cells were monitored at 1 , 2 , 4 and 7 days post infection ( dpi ) . The expected result was that virion particles from both BAC36 wt and recombinant viruses would be detected as GFP signals were seen by 2dpi ( Fig . 5A ) . Infected PBMCs were collected at 1 , 2 , 4 and 7dpi and total DNA was extracted as indicated in materials and methods . Viral DNA levels were determined by real-time PCR normalized to endogenous GAPDH . As indicated by GFP signals , viral copies of BAC36 wt showed a time-dependent manner increase in viral DNA , and the viral DNA copies of RTA1st and RTAall recombinant viruses had a similar level at 1 and 2 dpi ( Fig . 5B ) . However , genome copies of RTA1st and RTAall recombinant viruses showed a significant increase in infected PBMCs above that of BAC36 wt at 1dpi , 2dpi , 4dpi and 7dpi , suggesting that more latent viral genomes were tethered to the host genome ( Fig . 5B ) . Extracellular viral DNAs were collected from infected supernatants and quantitated by using KSHV TR primer set . In contrast , viral DNA levels of RTA1st and RTAall recombinant viruses were decreased in comparison to BAC36 wt from 1dpi to 7dpi , indicating that the lytic cycle of the recombinant viruses was not as robust . These data further support our pervious results showing that RTA1st and RTAall recombinant viruses exhibited an enhanced latent infection phenotype and a reduced capability for lytic replication in this system ( Fig 5C ) . LANA and RTA are viral encoded molecular switches for latent and lytic phases in KSHV infection . We wanted to determine the mRNA levels of RTA and LANA during primary infection . Immunofluorescence assays showed that LANA signals were detected in BAC36 wt , RTA1st and RTAall infected PBMCs at 1 dpi , 2 dpi , 4 dpi and 7dpi indicating a successful infection ( Fig . 6A ) . DNase treated total RNA from infected PBMCs at 1 dpi , 2 dpi , 4 dpi and 7dpi were used to generate cDNA and signals of RTA and LANA transcripts quantitated by real-time PCR . The results showed that LANA expression from BAC36 wt infected PBMCs increased in a time-dependent manner . Furthermore , RTA1st and RTAall recombinant viruses infected PBMCs showed higher LANA mRNA expression from 1 dpi to 7dpi , indicating that deletion of the RBP-Jκ cognate sequences within the RTA promoter can result in greater stringency for latent infection ( Fig . 6B ) . Additionally , mRNA of RTA was also analyzed and showed a peak at 2dpi in BAC36 wt and recombinant viruses ( Fig . 6C ) , perhaps promoting lytic infection at an early stage [47] . However , compared to BAC36 wt , RT-PCR results from RTA mRNA showed lower levels in RTA1st and RTAall infected PBMCs from 1 dpi to 7dpi with almost no significant change with RTAall recombinant virus ( Fig . 6C ) . This suggests that deletion of the RBP-Jκ cognate sequences in the promoter led to a dramatic loss in the ability of the recombinant viruses to induce lytic cycle activation , and as a result more tightly maintain latent infection ( Fig 6C ) . Furthermore , KSHV ORF6 which encodes the single-stranded DNA ( ssDNA ) binding protein and is tightly associated with DNA replication , showed increased expression in RTA1st and RTAall infected PBMCs compared to BAC36 wt infected PBMCs ( Fig 7A ) . This suggests that KSHV replication is actively engaged although latent infection is more stringent ( Fig 7A ) . ORF49 encoded by KSHV lies adjacent to and is transcribed in the opposite orientation to RTA . It also co-operates with RTA to activate KSHV lytic cycle [53] . We also monitored the activity of ORF49 , which cooperates with RTA during lytic replication through activation of several lytic promoters containing AP-1 sites [53] . The mRNA level was initially higher for recombinant viruses at 1dpi ( Fig 7B ) . However , at 4dpi and 7dpi , the ORF49 transcript levels for the recombinants were much lower than BAC36 wt with RTAall greater than 2-fold less at 4dpi . By 7dpi , both RTA1st and RTAall were further depressed compared to BAC36 wt to about 3-fold ( Fig . 7B ) . This suggests that active lytic replication of KSHV was dramatically reduced by 7dpi with a greater propensity for maintaining latency . We then investigated changes in the K8 and K9 transcript levels . The early lytic protein K8 [54] , showed a general increase in transcript levels over the 7 day period . However , the levels of K8 transcripts for RTA1st and RTAall were generally lower than that of the BAC36 wt virus ( Fig 7C ) . Interestingly , ORF K9 which encodes for vIRF , a homolog to members of the interferon ( IFN ) regulatory factor ( IRF ) and important for regulating intracellular interferon signal transduction [55] , increased dramatically by 2 days , with BAC36 wt continuing to increase in K9 transcript levels . At 7 days all viruses showed a drastic reduction in K9 transcript levels ( Fig . 7D ) . These results suggest a level of regulation of these transcripts which is important for controlling latent infection in KSHV . Human PBMCs contains lymphoid cells consisting of both T and B cells . The results above showed that RTA1st and RTAall recombinant viruses have an enhanced propensity for latent infection during the early stages of in vitro infection . Recently , Myoung and Ganem showed that 20–40% T and 4%–5% B cells from human tonsillar cultures can be infected [36] . However , only B cells support viral replication and produce progeny . Here , we are interested in the ability of KSHV to infect T and B cells from PBMCs during early infection . APC–conjugated anti-CD3 and PercpCy 5 . 5-conjugated anti-CD19 mAbs were used to detect infected T and B cells , respectively . GFP signals were used to detect KSHV-positive cells . Our results showed that GFP positive T cells were detected as early as 1 dpi ( Fig 8A ) . The GFP positive T cells infected by BAC36 wt virus was slightly changed relative to RTA1st and RTAall recombinant viruses . At 2 dpi , the proportion of GFP+ CD3+ T cells was 1 . 65% , 1 . 86% and 1 . 64% , respectively . The percentage of GFP+ CD3+ T cells infected by BAC36 wt , RTA1st and RTAall recombinant viruses were increased to 1 . 68% , 3 . 32% and 3 . 7% at 4 dpi , respectively . At 7 dpi , T cells were infected continuously in a time-dependent manner and the proportion of GFP+ CD3+ T cells were 3 . 98% and 4 . 48% for RTA1st and RTAall relative to 2 . 04% for BAC36 wt ( Fig 8A ) . These results further confirms that T cells were infected and that mutation of RBP-Jκ sites within RTA promoter can result in an increase in T cell infection as determined by GFP signals . We then investigated the response of B cells exposed to KSHV . At 1 dpi the percentage of GFP+ CD19+ B cells were 0 . 48% for BAC36 wt , 0 . 72% for RTA1st and 0 . 79% RTAall recombinant viruses ( Fig 8B ) . The next day , the proportion of GFP+ CD19+ B cells increased to 1 . 77% for BAC36 wt , 1 . 75% for RTA1st and 1 . 74 % RTAall . At 4 dpi , the percentage of GFP+ CD19+ B cells further increased to 2 . 03% for BAC36 wt , 2 . 49% for RTA1st and 2 . 51% RTAall , and at 7dpi , the GFP+ CD19+ B was similar to that at 4dpi suggesting that little or no further increase in B cell infection was seen . Furthermore , RTA1st and RTAall recombinant viruses infected B cells ( 2 . 13% and 2 . 47% ) show a consistently higher rate of infection compared to BAC36 wt infected B cells ( 2 . 05% ) ( Fig 8B ) . This suggests that PBMCs were continually infected over the 7day period . Interestingly , GFP+ CD3+ T cells had a higher rate of infection compared to GFP+ CD19+ B cells from the BAC36 wt infected PBMCs ( Fig 8C ) . Importantly , RTA1st and RTAall recombinant viruses possessed a higher infectivity compared to BAC36 wt virus for both infected T and B cells ( Fig 8D ) . The overall amount of the GFP positive cells showed a definite increase in a time-dependent manner ( Fig 8E ) . This further supports our previous data showing increased fluorescent signal from 2dpi to 7dpi . In general when compared to BAC36 wt , RTA1st and RTAall recombinant viruses infected PBMCs showed more GFP-positive cells , suggesting an increased ability for infection and maintenance of the KSHV genome within the first 7days after infection ( Fig 8E ) . Typically , infected PBMCs stopped clumping and most cells begin to die after 7 dpi . In our experiments , no transformation and/or immortalization was observed in infected PBMCs in vitro during the 7-day period . Thus it was difficult to monitor the effect of long-term infection of RTA1st and RTAall recombinant viruses . Here , we used recently developed telomerase-immortalized endothelial cells ( TIVE ) to monitor the ability of RTA1st and RTAall recombinant viruses to infect TIVE cells [44] . TIVE cells were infected with BAC36 wt , RTA1st and RTAall recombinant viruses as indicated in the materials and methods . GFP signals confirmed latent infection by BAC36 wt , RTA1st and RTAall viruses and photographs were taken at 1 week post-infection ( wpi ) , 2wpi and 4wpi ( Fig . 9A ) . We determined the copy number of KSHV genomes in infected TIVE as a measure of the persistence of the genomes . Total DNAs were extracted from BAC36 wt , RTA1st and RTAall recombinant viruses infected TIVE cells at 1 , 2 , 4wpi . Intracellular viral DNAs were determined by a quantitative PCR analysis standardized by GAPDH . The result showed that copy numbers of BAC36 wt , RTA1st and RTAall recombinant viruses were a slightly lower at 4wpi compared to 1 and 2wpi suggesting a possible loss of KSHV genome due to genome tethering instability . This phenomenon was also seen in BCBL-1-derived cell-free virus infected TIVE cells [44] . However , RTA1st and RTAall recombinant viruses maintained a similar copy number in the infected TIVE cells and infection levels were consistently greater than the BAC36 wt infected TIVE cells at 1 to 4wpi ( Fig . 9B ) . These results further indicated that RTA1st and RTAall recombinant viruses exhibited a decrease in lytic capability . Previous studies showed that KSHV long-term-infected telomerase-immortalized endothelial cells exhibit a significant increase in number of cells in the S phase [44] . Here , we cultured infected TIVE cells for 4 weeks , harvested , fixed and stained them with propidium iodide . Flow cytometry analysis showed that the RTA1st and RTAall infected TIVE cells had an increased S phase population ( 29 . 4% and 32 . 1% ) , relative to BAC36 wt ( 25% ) at 1wpi . Therefore , the recombinant viruses infected cells showed an enhanced capability to proliferate ( Fig 9C ) . Similar patterns were seen at 2wpi and 4wpi where RTA1st and RTAall all exhibited increased S phase populations , further supporting our previous hypothesis that RTA1st and RTAall recombinant viruses can increase cell proliferation in TIVE cells .
KSHV RTA is an immediate early protein ( IE ) that initiates KSHV lytic reactivation from latent infection . It can directly or indirectly stimulate the transcription of a cluster of lytic genes as a transcription factor through binding to specific promoter sequences . RTA is a key regulator for KSHV reactivation because its expression is sufficient to activate the entire lytic cycle . Therefore understanding the regulation of RTA is to provide a better clue related to KSHV infection and tumorigenicity . RTA-deficient viruses are able to establish latency but are unable to reactivate [23] . Here , in an effort to explore KSHV latency and reactivation we generated two recombinant viruses which possess different latency and reactivation profiles compared to BAC36 wt . They serve as important reagents which allow us to examine the early-stages of KSHV infection . This provides a model with which to understand the development of KSHV-associated lymphoproliferative diseases . RTA is an IE protein and its expression is also affected by other viral or cellular factors . For example , RTA up-regulates its own expression through interaction with the CCAAT/enhancer binding protein alpha ( C/EBPα ) at its promoter [20] , [56] . Furthermore , RTA can down-regulate its ability for activating specific viral promoters by cooperating with the viral protein b-Zip , an early protein encoded by ORF K8 [57] , [58] . Our previous studies showed that loss of one of those sites can potentially affect the overall regulation of all four RBP-Jκ sites within the RTA promoter [30] . Importantly , the observation that RTA1st and RTAall recombinant viruses enhance latency and can promote cell growth in 293 cells caused us to further investigate these mutations during early infection . Many human cancers are associated with tumor viruses [59] and many are detected in PBMCs . Human papillomavirus ( HPV ) DNA has been found as an episomal form in PBMCs , but no transcripts are detected [60] , [61] . Recent studies showed that PBMCs of haematuric cattle are additional reservoir of bovine papillomavirus type 2 [62] . Though Hepatitis C virus ( HCV ) is detected in PBMCs of infected individuals , infected PBMCs are not observed in co-culture with cell culture systems producing HCV virions . This implicates additional reservoirs for the virus and allows for PBMCs infection [63] . However , Polyomavirus BK ( BKV ) , one of the tumor viruses associated with nephropathy in renal allografts , elicits a BKV-specific proliferative response in the PBMCs of healthy individuals and bone marrow transplant recipients [64] , [65] . Another human tumor virus , John Cunningham virus ( JCV ) is detectable in the PMBCs of immunoimpaired and healthy individuals [66] . The well studied EBV infects B lymphocytes and induces their differentiation , proliferation [32] . Furthermore , PBMCs infected EBV lead to immortalization and transformation of B cells [47] . However , the mechanism by which KSHV infects and transforms B cells is not fully understood . Thus , the development of lymphoproliferative diseases which include primary effusion lymphoma and multicentric Castleman's disease is yet to be fully understood . Recently , PBMCs from marmosets orally and intravenously infected with rKSHV . 219 , showed the presence of the viral genome and LANA expression [42] . Additionally , T and B cells isolated from primary human tonsillar cells were shown to be infected by KSHV virions , although more T cells were infected . However , these infections were abortive without further lytic infection [35] , [36] . Other T cells types from human PBMCs can support KSHV infection [67] , [68] , [69] . Similar patterns were observed showing that both T and B cells from human PBMCs are effectively infected up to 7 days , suggesting that KSHV has a different mode of infection compared EBV infection . Obviously , many more T cells were infected in a time-dependent manner , perhaps due to the large population of T cells or a receptor on T cells not highly expressed in B cells . At 4 and 7 dpi , T cells were infected up to 4 . 48% , over 2 times the percentage of B cells infected . This phenomenon was also seen in KSHV infected tonsillar cells , though the life span of infected cells was short [36] . This pattern was similar to transformed 293 cells in that RTA1st and RTAall recombinant viruses still enhanced population of T cells in PBMCs , suggesting that infected T cells may be latently infected . Our data also showed that the percentage of CD19 + GFP+ B cells was increased from 1dpi to 7dpi . Unexpectedly , B cell infection boosted the proliferation of infected cells in a time-dependent manner , though T cells are likely to suppress lytic replication of infected B cells . It may be the case that B cells are undergoing lytic cycle replication , releasing progeny which reinfect both T cells and B cells as there was a general increase in lytic cycle gene expression within 2dpi . The population of infected B cells began to decrease at 7dpi , though the number of infected T cells had peaked . As HIV infection may reduce the CD4+ T cell counts , KSHV infected T cells may maintain a fine balance in overall T cell population as well as promote cell proliferation and immortalization for B cells . Another possibility is that certain subpopulations of cells were significantly overactive and may restrict B cell proliferation . Recently , Myoung and Ganem showed that T cells from infected primary human tonsillar lymphoid cells by KSHV did not support proper viral transcription and did not produce infectious virus . However , activated T cells may promote or stabilize latency of KSHV infected B cells [35] , [36] . Interestingly , we did not see proliferation of B cells but the percentage of infected B cells was decreased at 7dpi . Therefore we think other restrictive signals are involved in controlling latency of infected B cells , and so transformation for B cells was suppressed . After 7days post-infection , most cells died and no immortalization was observed . One reasonable explanation is that B cells can provide same paracrine signal activities to T cells but infected T cells may lack the ability to receive these signals from B cells , thus losing their proliferative capability . However , B cells may have a central role in infection and proliferation of PBMCs . After 7dpi , the efficiency of infection of B cells was decreased which led to a dramatic reduction in proliferation of the KSHV infected PBMCs . Did the presence of more T cells which were continuously infected destroy the population balance of PBMCs ? Do B cells drive essential signaling important for T cells proliferation ? These questions merit further investigation . In addition , GFP positive signals showed that more cells were infected in a time-dependent manner . We ruled out the possibility that TPA may have caused this effect in infected T and B cells , as a side-by-side comparison between BAC36 wt , RTA1st and RTAall recombinant viruses strongly supported our conclusion [70] . We clearly showed that RTA1st and RTAall viruses infected total GFP-positive B and T cells showed a prominent increase relative to BAC36 wt in PBMCs up to 7dpi . These results further support our hypothesis that mutation of RBP-Jκ in the RTA promoter can enhance KSHV latent infection of both of B and T cells in PBMCs during primary infection . The expression of total mRNA from infected PBMCs provides additional information and further reinforces the pattern of KSHV infection . RTA1st and RTAall recombinant viruses showed a decrease in K8 expression , though the down-regulation was not necessarily as strong as in B cells infected with RTA-deficient virus [23] . However , this is reasonable because neither RTA1st nor RTAall recombinant viruses abrogated viral lytic capabilities . RTA1st and RTAall recombinant viruses up-regulated KSHV ORF6 ( SSB , single-stranded DNA binding protein ) indicating that viral DNA replication is active in infected PBMCs . LANA expression was also up-regulated , suggesting a more tightly latent infection due to the action of LANA on RBP-Jκ . ORF 49 expression was down-regulated in RTA1st and RTAall recombinant viruses infected PBMCs , suggesting that these recombinant viruses may in part lose their lytic capability . Interestingly , the mRNA levels of ORF K8 which is a direct target of RTA [71] , was significantly decreased in RTA1st and RTAall recombinant viruses infected PBMCs , strongly suggesting that the two recombinant viruses possess a reduced capability for lytic replication during primary infection . Overall , these data further supports our hypothesis that recombinant RTA1st and RTAall viruses are enhanced in their ability to maintain latency after infection of primary cells . Though PBMCs were effectively infected , long-term infection of B and T cells was not supported . RTA1st and RTAall recombinant viruses showed on increase in genome copies in long-term-infected TIVE cells providing evidence that mutation of the RBP-Jκ sites within the RTA promoter enhanced KSHV latent infection and induces the proliferative capability of the infected primary cells . In conclusion , we have now experimentally shown that KSHV recombinant viruses with mutated RBP-Jκ sites within the RTA promoter possess an enhanced ability to maintain latent infection in transformed-293 cells as well as PBMCs during early infection . Our studies further confirm and define our previously published data which showed the effects on truncation of the RTA promoter and has important implications regarding the development of KSHV-associated lymphoproliferative disease . These recombinant viruses now provide a model which can be used to explore the early stages of primary infection in human PBMCs as well as the development of KSHV-associated lymphoproliferative diseases . | Kaposi's sarcoma-associated herpesvirus ( KSHV ) is tightly linked to at least two lymphoproliferative disorders , primary effusion lymphoma ( PEL ) and multicentric Castleman's disease ( MCD ) . The life cycle of KSHV consists of latent and lytic phase . RTA is the master switch for viral lytic replication . In this study , we first show that recombinant viruses deleted for the RBP-Jκ sites within the RTA promoter have a decreased capability for lytic replication , and thus enhanced colony formation and proliferation of infected cells . Interestingly , the recombinant viruses show greater infectivity in human peripheral blood mononuclear cells ( PBMCs ) . The recombinant viruses also infected CD19+ B cells and CD3+ T cells with increased efficiency in a time-dependent manner and now provide a model which can be used to explore the early stages of primary infection in human PBMCs , as well as the development of KSHV-associated lymphoproliferative diseases . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"virology",
"biology",
"microbiology"
] | 2012 | The RBP-Jκ Binding Sites within the RTA Promoter Regulate KSHV Latent Infection and Cell Proliferation |
Chikungunya virus ( CHIKV ) , mainly transmitted in urban areas by the mosquitoes Aedes aegypti and Aedes albopictus , constitutes a major public health problem . In late 2013 , CHIKV emerged on Saint-Martin Island in the Caribbean and spread throughout the region reaching more than 40 countries . Thus far , Ae . aegypti mosquitoes have been implicated as the sole vector in the outbreaks , leading to the hypothesis that CHIKV spread could be limited only to regions where this mosquito species is dominant . We determined the ability of local populations of Ae . aegypti and Ae . albopictus from the Americas and Europe to transmit the CHIKV strain of the Asian genotype isolated from Saint-Martin Island ( CHIKV_SM ) during the recent epidemic , and an East-Central-South African ( ECSA ) genotype CHIKV strain isolated from La Réunion Island ( CHIKV_LR ) as a well-characterized control virus . We also evaluated the effect of temperature on transmission of CHIKV_SM by European Ae . albopictus . We found that ( i ) Aedes aegypti from Saint-Martin Island transmit CHIKV_SM and CHIKV_LR with similar efficiency , ( ii ) Ae . aegypti from the Americas display similar transmission efficiency for CHIKV_SM , ( iii ) American and European populations of the alternative vector species Ae . albopictus were as competent as Ae . aegypti populations with respect to transmission of CHIKV_SM and ( iv ) exposure of European Ae . albopictus to low temperatures ( 20°C ) significantly reduced the transmission potential for CHIKV_SM . CHIKV strains belonging to the ECSA genotype could also have initiated local transmission in the new world . Additionally , the ongoing CHIKV outbreak in the Americas could potentially spread throughout Ae . aegypti- and Ae . albopictus-infested regions of the Americas with possible imported cases of CHIKV to Ae . albopictus-infested regions in Europe . Colder temperatures may decrease the local transmission of CHIKV_SM by European Ae . albopictus , potentially explaining the lack of autochthonous transmission of CHIKV_SM in Europe despite the hundreds of imported CHIKV cases returning from the Caribbean .
Chikungunya virus ( CHIKV ) is a mosquito-borne alphavirus that causes an acute febrile illness characterized by severe arthralgia . International travel and the global expansion of the two main CHIKV urban mosquito vectors , Aedes aegypti and Aedes albopictus , have enhanced the ability of the virus to spread to new regions where environmental conditions are permissive for viral transmission . Phylogenetically , CHIKV strains are classified into three distinct genotypes: Asian , West African , and Central/East/South African ( ECSA ) . Over the past decade , the ECSA genotype significantly expanded its geographical range resulting in epidemics throughout India , Africa , Asia , and temperate Europe [1] . The ECSA genotype has also been repeatedly imported into the Americas , but autochthonous transmission has not been detected despite the ability of local vector species to transmit CHIKV [2 , 3] . Between 1995 and 2009 , 109 imported CHIKV cases were identified in the United States alone , and among those , 13 ( 12% ) developed a viremia high enough to infect mosquitoes [4] . Following the La Réunion epidemic in 2004 , nine imported CHIKV cases were reported during 2006 in the French overseas departments of America [5] . In late 2013 , the first locally-acquired CHIKV infections in the Americas were reported from Saint-Martin Island in the Caribbean [6 , 7] . The virus successively spread to other Ae . aegypti-infested islands . At the time of writing , this epidemic has caused more than 1 , 000 , 000 suspected cases in 43 countries from the Americas [8] , with the potential for further spread to the rest of the continent . Surprisingly , the CHIKV strain responsible for this epidemic belongs to the Asian genotype [6] and not to the ECSA genotype as might have been predicted , based upon the high numbers of imported cases reported in recent years [4] , [5] . Until now , only Ae . aegypti mosquitoes have been implicated in CHIKV transmission in the Americas [9] . Previous studies have suggested that the Asian genotype of CHIKV is constrained in its ability to adapt to Ae . albopictus via negative epistatic interactions of a single residue ( E1-98T ) with the E1-A226V substitution [10] , which could limit viral spread to regions where this mosquito species is dominant [9] . Nevertheless , the potential role of Ae . albopictus as a vector in the Americas for the currently circulating CHIKV strain must also be considered . Indeed , Ae . albopictus is present in at least 19 countries in the Americas [11] , and have previously been shown to experimentally transmit Asian strains of CHIKV [3] . This recent CHIKV epidemic in the Caribbean was also a threat for Europe . In France , from May 2 through July 4 , 2014 , the number of laboratory confirmed imported cases of CHIKV was much higher ( 126 cases ) than in previous years [12] . This increase could potentially enhance the risk of local transmission in Ae . albopictus–infested European regions . Since its first report in Albania in 1979 , Ae . albopictus has progressively spread and today is found in 20 European countries , causing a major public health concern [13] . Previous autochthonous transmissions of CHIKV in Italy and in France highlight the potential to establish transmission cycles involving temperate Ae . albopictus populations [14 , 15] . Furthermore , several studies revealed that European Ae . albopictus transmit ECSA CHIKV strains efficiently , at both 28°C and at lower temperatures [16 , 17] . Without a vaccine or specific treatment available , the only strategy for control of CHIKV outbreaks remains the suppression of vector populations , the use of individual protections ( e . g . repellents ) and the reinforcement of epidemiological surveillance in areas at high epidemic risk . The determination of vector competence in mosquito populations , defined as the ability of the vector to ingest , disseminate and transmit a pathogen , is essential in evaluating the risk of CHIKV transmission and spread into new areas as well as to design appropriate control strategies . Examination of virus population diversity by deep sequencing revealed strong bottlenecks when CHIKV passes mosquito anatomical barriers ( midgut and salivary glands ) leading to select variants with high epidemic potential in mosquito saliva [18] . Vector competence can be highly variable in natural populations and is determined by genotype-genotype interactions , in which successful transmission depends on some specific combination of mosquito and viral genetic characteristics [19] , under specific environmental conditions [20] . Ambient temperatures and even daily fluctuations of temperature play a key role in shaping mosquito vector competence for pathogens [21 , 22] . Moreover , it has been shown that the potential of CHIKV transmission by Ae . albopictus strongly depends on the three-way combination of mosquito population , virus strain and temperature or Genotype x Genotype x Environment ( G x G x E ) interactions [17] . Additionally , knowledge about factors shaping vector capacity corresponding to the ability of a mosquito to act as a vector in the field ( e . g . , mosquito densities , mosquito trophic preferences , mosquito survival rate… ) will be informative for a more accurate appraisal of CHIKV transmission . This study aims to evaluate the potential for the Asian CHIKV strain currently circulating in the Caribbean to initiate outbreaks in other countries of the Americas and to examine the effect of temperature on viral transmission in temperate Europe .
The Institut Pasteur animal facility has received accreditation from the French Ministry of Agriculture to perform experiments on live animals in compliance with the French and European regulations on care and protection of laboratory animals . This study was approved by the Institutional Animal Care and Use Committee ( IACUC ) at the Institut Pasteur . No specific permits were required for the described field studies in locations which are not protected in any way and did not involve endangered or protected species . Eleven mosquito populations from the Caribbean , continental America and metropolitan France were used: eight populations of Ae . aegypti and three of Ae . albopictus ( Table 1 ) . The mosquitoes were field-collected in 2013 using ovitraps ( 10–58 per collection site ) . The field-collected eggs were immersed in water for hatching; larvae were reared at densities of 100–150 individuals per pan and fed with yeast tablets . Emerged adults were identified according to morphological criteria and maintained in cages at 28°±1°C with a 16h:8h light:dark cycle , 80% relative humidity , and supplied with a 10% sucrose solution . The F1 generation of mosquitoes was used for all infection assays except for Ae . aegypti from Saint-Martin ( F3 generation ) and from United States ( F10 generation ) . Two CHIKV isolates of different genotypes were used: one CHIKV isolate from Saint-Martin Island belonging to the Asian lineage and one from La Réunion belonging to the East-Central-South African lineage . The isolate from La Réunion was the strain CHIKV 06 . 21 ( CHIKV_LR ) isolated in 2005 [23] , and provided by the French National Reference Center for Arboviruses at the Institut Pasteur in Paris . The CHIKV strain from Saint-Martin was the CHIKV 20235 [6] , isolated in 2013 from the serum of the first confirmed CHIKV local case in the New World . This strain was kindly provided by the French National Reference Center for Arboviruses in Marseille . The CHIKV_20235 ( CHIKV_SM ) is phylogenetically related to strains recently identified in Asia , i . e . China ( 2012 ) , and the Philippines ( 2013 ) , most of them sharing a specific four amino-acid deletion in the nsp3 gene [6] . The consensus sequences of CHIKV_LR and CHIKV_SM diverge by ~7% at the amino-acid level including two interesting changes at E1-226 ( valine for CHIKV_LR and alanine for CHIKV_SM ) [24] and E1-98 ( alanine for CHIKV_LR and threonine for CHIKV_SM ) [10] . The residue E1-98T exerts a negative epistatic interaction which blocks the ability of Asian strains to adapt to Ae . albopictus via the E1-A226V substitution [10] . None of the CHIKV strains harbor the E2 Ae . albopictus-adaptive mutations ( E2-K252Q , E2-L210Q , E2-K233Q ) described or predicted for CHIKV ECSA strains [25] . Stocks of CHIKV_LR were produced following three passages on Ae . albopictus C6/36 cells and CHIKV_SM was obtained after two passages on Vero cells . Supernatants were harvested and stored at -80°C until used for mosquito experimental infection assays . Viral titers were determined by serial 10-fold dilutions on Vero cells . Five to seven day-old female vectors were fed on an infectious blood-meal containing 1 . 4 mL of washed rabbit erythrocytes and 700 μL of viral suspension supplemented with a phagostimulant ( ATP ) at a final concentration of 5 mM . Four to six boxes of 60 mosquitoes were tested for each population . All 11 mosquito populations were challenged with CHIKV_LR and CHIKV_SM , each viral strain was provided separately in the blood meal . The titer of infectious blood-meals was 106 . 5 pfu/mL in agreement with viremia levels detected in patients [26] . After the infectious blood-meal , fully engorged females were transferred to cardboard containers and maintained with 10% sucrose at 28°±1°C , a 16h:8h light:dark cycle and 80% humidity . After infection , Ae . albopictus from Bar-sur-Loup in southeastern France ( FRA in Table 1 ) were maintained in climatic chambers ( KB 53 , Binder , Tuttlingen , Germany ) with a 16h:8h light:dark cycle under three different temperature regimes: ( i ) a constant temperature of 28°C ± 0 . 1°C , ( ii ) a constant temperature of 20°C±0 . 1°C or ( iii ) at temperatures displaying daily fluctuations between 17°C±0 . 1°C and 23°C± 0 . 1°C ( average: 20°C±0·1°C ) . The constant temperature of 28°C was chosen both because it serves as a typical mean temperature in tropical regions and because this temperature is commonly used in our vector competence assays [3 , 27] . The constant temperature of 20°C and the temperature regime with daily fluctuations around 20°C were chosen as representative of the low-temperature threshold recorded during the Italian epidemic of CHIKV between June and September 2007 [14 , 28] and in southeast France in September 2010 ( http://www . meteociel . fr ) [15] . Batches of ~20 mosquitoes of each combination of mosquito population-virus strain ( and temperature regime for Ae . albopictus from Bar-sur-Loup , FRA in Table 1 ) were analyzed at days 3 , 5 and 7 post-infection ( pi ) for the two CHIKV strains tested . Time-points were chosen based on the kinetics of CHIKV transmission efficiency obtained with mosquitoes from Rio de Janeiro , Brazil [3] . Additionally , mosquitoes from Saint-Martin ( Ae . aegypti , SMAR ) and from Bar-sur-Loup ( Ae . albopictus , FRA ) were also analyzed at day 2 pi to estimate the extrinsic incubation period [27] . To estimate viral transmission , saliva was collected from individual mosquitoes as described by Dubrulle and colleagues [27] . Briefly , wings and legs were removed from each mosquito and the proboscis was inserted into a 20 μL tip containing 5 μL of Fetal Bovine Serum ( FBS ) . After 30 min of salivation , FBS containing saliva was expelled into 45 μL of Leibovitz L15 medium for further titration . Transmission efficiency was determined by the proportion of mosquitoes with virus in the saliva among tested ones ( i . e . , surviving females including females unable to disseminate the virus and those able to disseminate ) . The number of infectious particles per saliva was determined by titration using focus fluorescent assay on C6/36 Ae . albopictus cells . Saliva samples were serially diluted and inoculated onto C6/36 Ae . albopictus cell culture in 96-well plates . After incubation at 28°C for three days , plates were stained using hyper-immune ascetic fluid specific to CHIKV as primary antibody . Alexa Fluor 488 goat anti-mouse IgG was used as the second antibody ( Life technologies ) . The lower detection limit of an assay was 2 FFU/saliva . About 20 saliva samples from each mosquito population infected with CHIKV_SM were collected at day 7 pi , pooled and deep sequenced . Before pooling , saliva were diluted to obtain comparable numbers of CHIKV particles allowing to have an appropriate representation of the overall viral population . Whole genome sequences ( excluding the first 19 nucleotides of the 5’UTR , the 3’UTR and the 25 last nucleotides of the second open reading frame ) were determined for pooled saliva using the Ion PGM Sequencer ( Life Technologies ) as described by Rothberg and colleagues [29] , and sequence analysis was conducted using CLC Genomics Workbench 6 software . For deep sequencing , a set of four primer pairs ( S1 Table ) was used to generate amplicons with 3 μL of nucleic acid extract and the Superscript III One-Step RT-PCR Platinum TaqHifi kit ( Life Technologies ) according to manufacturer’s instructions using the following cycling parameters: 50°C for 30 min , 94°C for 2 min followed by 45 cycles of 94°C for 15 sec , 56°C for 30 sec and 68°C for 4 min . PCR products were verified by gel electrophoresis , and amplicons were purified using Amicon Ultra—0 . 5 mL 30K kit ( Millipore ) according to the manufacturer's instructions . For each sample , an equimolar mix of all amplicons was used to build a library and produce the corresponding sequences for the Ion PGM Sequencer according to the manufacturer's instructions . The reads obtained were trimmed: first using quality score and then by removing the primers used for amplification . Reads were mapped to the genome sequence of CHIKV_SM produced following two passages on Vero cells , which was used as a reference . Mutation frequencies ( proportion of viral genomes with a specific mutation ) at each position were calculated as the number of reads with the mutation compared to the reference divided by the total number of reads at that site . Only substitutions with a mutation frequency ≥ 5% were considered significant for further analysis . The 5% cut-off was considered in our NGS experimental protocol applied to CHIKV plasmid sequence to avoid any background minor variants due to the sequencing method . All statistical tests were conducted using the STATA software ( StataCorp LP , Texas , USA ) . P-values >0 . 05 were considered non-significant . Frequencies were compared using Fisher’s exact test and sample distributions with the Kruskal-Wallis test . If multiple Fisher's tests were applied to the same data set , then the significance level for each test was adjusted by the sequential Bonferroni method to accommodate the multiple tests .
To characterize the ability of Ae . aegypti from Saint-Martin to transmit CHIKV_SM , we evaluated transmission efficiencies and viral loads in saliva at days 2 , 3 , 5 , and 7 pi . As a control , we also infected mosquitoes with CHIKV_LR . Our studies indicate that CHIKV_LR could be detected from day 2 pi ( transmission efficiency: 10%; viral load: 0 . 8±0 . 5 log10 ) and CHIKV_SM from day 3 pi ( transmission efficiency: 35%; viral load: 1 . 5±0 . 9 log10 ) when provided in blood-meals to Ae . aegypti SMAR from Saint-Martin . In addition , we found that at a given day pi , no significant difference in transmission efficiencies were detected between the two viruses , CHIKV_SM and CHIKV_LR ( P-value > 0 . 05 ) ( Fig 1A ) . When examining viral loads in saliva , a similar pattern was obtained except at day 5 pi where the distribution of viral loads in Ae . aegypti SMAR was significantly higher when infected with CHIKV_SM than with CHIKV_LR ( P-value < 0 . 05 ) ( Fig 1B ) . To evaluate the risk of CHIKV spread throughout the Americas , we compared transmission efficiencies among Ae . aegypti mosquitoes from localities in the Americas . All mosquitoes were susceptible to CHIKV_SM ( Fig 2A and 2B ) . At each day pi , we did not find any significant difference in transmission efficiencies between mosquito populations ( P-value > 0·05 ) , except for mosquitoes from French Guiana and Macapá ( P-value < 0 . 05 ) . At day 3 pi , transmission efficiency was significantly reduced ( ~10% ) in these populations ( P-value < 0 . 05 ) . When examining mosquito susceptibilities to CHIKV_LR , no differences were found between mosquito populations except for Ae . aegypti from Macapá which displayed low transmission efficiencies not exceeding 25% ( P-value < 0 . 05 ) ( S1 Fig ) . When examining viral loads , no significant differences were detected regardless of population ( P-value > 0 . 05 ) ( Fig 2A and 2B ) . To characterize the susceptibility of the alternative vector Ae . albopictus to CHIKV_SM , we compared transmission efficiencies and viral loads in saliva between Ae . aegypti and Ae . albopictus collected from two regions ( Brazil and the United States ) . CHIKV_LR was used as a control . Both mosquito species exhibited similar transmission efficiencies for CHIKV_SM at each day pi ( P-value > 0 . 05 ) ( Fig 3A ) . Ae . aegypti was significantly more susceptible to CHIKV_LR than Ae . albopictus at day 3 pi for mosquitoes from Brazil , and at days 5 and 7 pi for mosquitoes from the United States ( Fig 3B ) . When comparing Ae . albopictus transmission efficiencies between viral strains , differences were only observed for Ae . albopictus from Rio de Janeiro at day 3 pi ( 47 . 06% for CHIKV_LR versus 5% for CHIKV_SM ) . Overall , viral loads were similar ( Fig 3C and 3D ) except at day 7 pi for mosquitoes from the United States infected with CHIKV_LR ( Fig 3D ) . To determine if temperate Ae . albopictus were more susceptible to CHIKV_SM than to CHIKV_LR , we evaluated transmission efficiencies and viral loads in saliva at days 3 , 5 , and 7 pi ( Fig 4A ) . We found that transmission efficiency was significantly higher with CHIKV_LR at day 3 pi ( P-value < 0 . 05 ) , while no difference was found between viral strains at days 5 and 7 pi ( P-value > 0·05 ) . Viral loads in saliva were not significantly different ( P-value > 0 . 05 ) at any day pi . To define if transmission of CHIKV_SM was lowered at a colder temperature , we compared transmission efficiencies and viral loads in saliva at each day pi between temperate Ae . albopictus incubated at 28°C and at 20°C after oral infection . The virus was detected from day 3 pi when mosquitoes were incubated at 28°C and only at day 7 pi when incubated at 20°C ( Fig 4B ) . At day 7 pi , transmission efficiencies were significantly higher at 28°C ( 57 . 5%±7 . 9 ) than at 20°C ( 10%±6 . 9 ) ( P-value < 0 . 05 ) ( Fig 4B ) . When examining viral loads in saliva , we obtained the same pattern with a higher viral load detected in saliva of mosquitoes incubated at 28°C ( 1 . 7±0 . 8 log10 ) than at 20°C ( 0 . 7±0 . 3 log10 ) ( Fig 4A and 4B ) . When mosquitoes were incubated with daily fluctuations of temperature with a mean value of 20°C , transmission efficiencies and viral loads in Ae . albopictus saliva were slightly increased ( Fig 4B ) . Using pools of saliva from each mosquito populations , mutation frequencies were estimated from nucleotide polymorphisms at each position of the viral genome using the CHIKV_SM produced on Vero cells as reference . Thirty-four nucleotide substitutions ( of frequency higher than 5% ) were detected throughout the viral genome: 25 in non-structural genes , and nine in structural genes . ( Table 2 ) . Notably , we did detect neither E1-A226V nor E1-T98A substitutions , or other described Ae . albopictus-adaptive mutations in any pool of mosquito saliva . Overall , results were heterogeneous regardless of Ae . aegypti or Ae . albopictus populations , with one exception , Ae . aegypti SMAR from Saint-Martin . While the number of mutations for other mosquito populations ranged from 0 to 6 ( median = 3 ) , we detected 13 mutations for the Ae . aegypti SMAR . In addition , 62% ( 8/13 ) of these mutations were synonymous while the proportion of synonymous mutations for other mosquito populations ranged from 0% to 50% ( median = 17% ) . This specific mutation pattern was also associated with a particular distribution in the viral genome: 54% ( 7/13 ) of them were located in the structural genes ( almost all of them were synonymous ) representing 78% of the mutations detected in this region . Only two other mutations were detected in the structural genes of the virus population from saliva of Ae . aegypti MACA from Brazil . Saliva from Ae . aegypti SAIN and Ae . albopictus USA were not analyzed due to technical problems .
In December 2013 , local transmission of CHIKV was reported on Saint-Martin Island in the Caribbean [6] . Although multiple imported cases of CHIKV had been reported in the New World , epidemic spread of the virus did not occur until approximately ten years after its expansion from costal Kenya in 2004 [1] . Surprisingly , it was the Asian genotype of CHIKV which caused this epidemic , as opposed to the more widespread ECSA genotype . Additionally , the mosquito Ae . aegypti has been implicated as the main vector in this epidemic . Although this combination of mosquito and virus has been documented in sporadic outbreaks [30] , our understanding of virus-vector interactions remains limited . Here , we demonstrate that ( i ) Ae . aegypti from Saint-Martin Island were able to efficiently transmit both Asian and ECSA genotype CHIKV strains isolated from Saint-Martin and La Reunion respectively , ( ii ) Aedes aegypti from the Americas display similar and moderate transmission efficiency for CHIKV_SM , and ( iii ) Ae . albopictus from the Americas were as competent as Ae . aegypti in transmission of CHIKV_SM . Taken together , our findings highlight the potential for further spread of CHIKV within the Americas as well as a potential role of Ae . albopictus in this context . Additionally , our findings that European Ae . albopictus are capable of transmitting both CHIKV_SM and CHIKV_LR raises concerns about the potential for future CHIKV epidemics in Europe . First isolated in 1952 in Tanzania , CHIKV dramatically expanded its geographic distribution over the last decades , the first wave spreading from Africa to India and Southeast Asia , and the second wave from the coastal Kenya to the Indian Ocean region [9] . In 2004 , the ECSA genotype was responsible for the spread of CHIKV beyond its traditional geographic distribution into many tropical regions [31] . In addition , CHIKV strains from the ECSA genotype have previously been identified in Europe: in Italy in 2007 and France in 2010 [14 , 15] . A common theme for this second wave of expansions is the role of Ae . albopictus as the primary arthropod vector . This is due in large part to a single amino-acid mutation in the CHIKV E1 glycoprotein ( E1-A226V ) which increased the vector competence of Ae . albopictus approximately 50-fold compared to the more traditional vector Ae . aegypti [32 , 33] . Second-step Ae . albopictus-adaptive mutations such as E2-K252Q and E2-L210Q , detected in CHIKV isolated from India in 2007 and 2009 respectively , may also have contributed to the spread and rapid diversification of CHIKV lineages [34] , 10 , 24] . Our results contrast with previous findings since we found that Ae . aegypti was more susceptible to CHIKV_LR than Ae . albopictus at day 3 pi for mosquitoes from Brazil , and at days 5 and 7 pi for mosquitoes from the United States ( Fig 3B ) . Geographically distant mosquito populations can correspond to genetically differentiated populations presumably causing the differences observed . While previous studies were mainly based on field-collected mosquitoes [32] or laboratory-adapted strains [33] , our study reports on transmission by detecting virus in saliva of mosquitoes of the F1 generation . While the ECSA genotype has continued to spread throughout Southeast Asia [1] , the Asian genotype has affected limited regions in the Pacific region with the virus progressing in small jumps from East Asia to the Western Pacific [30 , 35] . As early as 2007 , the potential of CHIKV emergence in the Americas was strengthened by human populations mostly naïve to CHIKV combined with high densities of competent Ae . aegypti and Ae . albopictus vectors [1 , 3] . The threat became reality in late 2013 with the detection of the first autochthonous CHIKV cases on the Caribbean island of Saint-Martin . The virus belonged to the Asian genotype and was closely related to strains from East Asia ( Philippines and China ) [6] . At the time of writing , local transmission had been identified in 43 countries in the Americas with more than 1 , 000 , 000 cases reported ( http://www . cdc . gov/chikungunya/geo/index . html ) . The mosquito Ae . aegypti has been identified as the main vector in this epidemic , spreading the virus from Saint-Martin , throughout the Caribbean . Here , we showed that Ae . aegypti SMAR were able to efficiently transmit both ECSA and Asian genotypes of CHIKV at rates similar to Ae . aegypti from New Caledonia [35] . Deep sequencing reveals that this efficient transmission was not associated with the emergence of viral populations harboring the E1-A1226V and/or the E1-T98A in any Ae . aegypti or Ae . albopictus population . Intriguingly , we found a large number of synonymous mutations in the saliva of Ae . aegypti SMAR seven days after oral challenge with CHIKV_SM ( Table 2 ) while up to three variants were identified in Ae . aegypti collected from neighboring islands ( Martinique and Guadeloupe ) . This may suggest that bottlenecks induced by mosquito internal barriers ( i . e midgut , salivary glands ) were less constraining for the CHIKV_SM in Ae . aegypti SMAR mosquitoes than in Ae . aegypti from neighboring islands . It is tempting to propose that CHIKV_SM is naturally well adapted to Ae . aegypti SMAR , and therefore able to rapidly express a particular mutant spectrum consisting of a majority of synonymous mutations . Thus , CHIKV emergence and rapid spread through the Caribbean is due to a CHIKV_SM well adapted to Ae . aegypti from Saint-Martin Island , and to the ability of mosquitoes from the West Indies to transmit Asian strains of CHIKV . Similar to previous epidemics of other vector-borne diseases , CHIKV has expanded outside its traditional range of distribution ( and unusually , emerging in temperate regions ) following the worldwide expansion of Ae . albopictus . This vector is an invasive species currently found in temperate and tropical regions [11 , 13] . Here , we found that regardless of population origin , susceptibility of Ae . albopictus from the Americas was similar for both the ECSA genotype ( CHIKV_LR ) and the Asian genotype ( CHIKV_SM ) at days 5 and 7 pi . Differences were only detected at day 3 pi where , surprisingly , transmission by Ae . albopictus collected from Rio de Janeiro ( Brazil ) was 10 times higher with CHIKV_SM ( ~ 50% ) than with CHIKV_LR ( ~5% ) ( Fig 3A and 3B ) , corroborating the threat of CHIKV_SM for this country . However , temperate Ae . albopictus exhibited the opposite pattern of transmission: with CHIKV_LR , transmission efficiency remained high ( ~ 50% ) from day 3 pi while it reached a similar level only at day 5 pi with CHIKV_SM ( Fig 4A ) . Differences on transmission efficiencies of CHIKV_SM and CHIKV_LR observed between Ae . albopictus from the Americas and Europe highlight the need for further genetic analysis in order to elucidate the phylogenetic relationships for these mosquito populations . Since ambient temperature plays a key role in modulating mosquito vector competence for pathogens [21 , 22] , we also incubated temperate Ae . albopictus at lower temperatures . When infected mosquitoes were incubated at 20°C , corresponding to a mean temperature recorded where local CHIKV transmission was detected in Italy and in southeast France [14 , 28] , the virus was detected very late at day 7 pi in mosquito saliva ( transmission efficiency , ~ 10% ) . When mimicking daily fluctuations of temperature around the mean value of 20°C , transmission efficiency was slightly enhanced though no significant differences were observed ( Fig 4B ) . This result contrasts with our previous findings where Ae . albopictus from southern France transmitted the ECSA genotype better at 20°C compared to 28°C [17] . This was supported in late October 2014 with the detection of 5 CHIKV autochthonous cases in Montpellier , Southern France ( http://www . invs . sante . fr/ ) . Viral isolates from these patients belonged to the ECSA genotype ( Leparc-Goffart , personal communication ) . Despite the hundreds of infected people returning to France from the Caribbean during the summer and later [12] , no autochthonous transmission of the imported Asian CHIKV genotype was detected , supporting that transmission is strongly dependent on the mosquito population genetics , the viral genotype and environmental conditions such as the temperature [17] . According to our results based mainly on vector competence , it is crucial for American and European countries to be prepared for more vector-borne disease epidemics . Vector control measures should be triggered very quickly to prevent transmission of the virus by local mosquitoes . With CHIKV_LR and CHIKV_SM , only three days after infection are needed to initiate an outbreak . Although our results show differences in vector competence , other factors ( mosquito densities , feeding behavior , mosquito survival rate…s ) composing the vector capacity , are needed to assess more accurately the risk of CHIKV transmission . | More than one million chikungunya cases have been reported in the Americas since October 2013 , when the Asian genotype of chikungunya virus ( CHIKV ) was imported by a traveller returning from Asia . CHIKV is mainly transmitted in urban areas by the domestic mosquitoes Aedes aegypti and Aedes albopictus . In this study , we evaluate the potential for the CHIKV circulating in the Caribbean to initiate outbreaks in Aedes-infested regions of continental America and Europe by assessing the ability of local mosquitoes to experimentally transmit the virus . Mosquitoes were exposed to a blood-meal containing the virus which must overcome several barriers to infect various tissues in the vector before being secreted in the mosquito saliva when biting a host . We found that Ae . aegypti and Ae . albopictus transmitted similarly the virus . When exposing Ae . albopictus from Europe at a temperature of 20°C after infection , we detect a significant drop of CHIKV transmission potential . Our results suggest that the CHIKV outbreak in the Americas could potentially spread throughout Ae . aegypti- and Ae . albopictus-infested regions of the Americas however with a limited risk of spillovers in Ae . albopictus-infested regions in Europe . These data will be useful for adapting vector control strategies and epidemiological surveillance . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Chikungunya Virus Transmission Potential by Local Aedes Mosquitoes in the Americas and Europe |
Genome-wide association study ( GWAS ) methods applied to bacterial genomes have shown promising results for genetic marker discovery or detailed assessment of marker effect . Recently , alignment-free methods based on k-mer composition have proven their ability to explore the accessory genome . However , they lead to redundant descriptions and results which are sometimes hard to interpret . Here we introduce DBGWAS , an extended k-mer-based GWAS method producing interpretable genetic variants associated with distinct phenotypes . Relying on compacted De Bruijn graphs ( cDBG ) , our method gathers cDBG nodes , identified by the association model , into subgraphs defined from their neighbourhood in the initial cDBG . DBGWAS is alignment-free and only requires a set of contigs and phenotypes . In particular , it does not require prior annotation or reference genomes . It produces subgraphs representing phenotype-associated genetic variants such as local polymorphisms and mobile genetic elements ( MGE ) . It offers a graphical framework which helps interpret GWAS results . Importantly it is also computationally efficient—experiments took one hour and a half on average . We validated our method using antibiotic resistance phenotypes for three bacterial species . DBGWAS recovered known resistance determinants such as mutations in core genes in Mycobacterium tuberculosis , and genes acquired by horizontal transfer in Staphylococcus aureus and Pseudomonas aeruginosa—along with their MGE context . It also enabled us to formulate new hypotheses involving genetic variants not yet described in the antibiotic resistance literature . An open-source tool implementing DBGWAS is available at https://gitlab . com/leoisl/dbgwas .
The aim of Genome-Wide Association Studies ( GWAS ) is to identify associations between genetic variants and a phenotype observed in a population . They have recently emerged as an important tool in the study of bacteria , given the availability of large panels of bacterial genomes combined with phenotypic data [1–7] . GWAS rely on a representation of the genomic variation as numerical factors . The most common approaches are based on single nucleotide polymorphisms ( SNPs ) , defined by aligning all genomes of the studied panel against a reference genome [1 , 3 , 4] or against a pangenome built from all the genes identified by annotating the genomes [8] , and on gene presence/absence , using a pre-defined collection of genes [5 , 7] . The use of a reference genome becomes unsuitable when working on bacterial species with a large accessory genome—the part of the genome which is not present in all strains . On the other hand , methods focusing on genes are unable to cover variants in noncoding regions , including those related to transcriptional and translational regulation [9 , 10] . Moreover , some poorly studied species still lack a representative annotation [11] . To circumvent these issues and make bacterial genomes amenable to GWAS , recent studies have relied on k-mers: all nucleotide substrings of length k found in the genomes [2 , 5 , 6] . The presence of k-mers in genomes can account for diverse genetic events such as the acquisition of SNPs , ( long ) insertions/deletions and recombinations . Unlike SNP- or gene-based approaches , k-mer analyses do not require a reference genome or any assumption on the nature of the causal variants and can even be performed without assembling the genome sequences [12] . While k-mers can reflect any genomic variation in a panel , they do not themselves represent biological entities . Translating the result of a k-mer-based GWAS into meaningful genetic variants typically requires mapping a large and redundant set of short sequences [2 , 5 , 6 , 13] . Recent studies have suggested reassembling the significantly associated k-mers to reduce redundancy and retrieve longer marker sequences [6 , 13] . Nonetheless , k-mer representation often loses in interpretability what it gains in flexibility , and the best way to encode the genomic variation in bacterial GWAS is not yet clearly defined [14 , 15] . Our approach , coined DBGWAS , for De Bruijn Graph GWAS , bridges the gap between , on the one hand , SNP- and gene-based representations lacking the right level of flexibility to cover complete genomic variation , and , on the other hand , k-mer-based representations which are flexible but not readily interpretable . We rely on De Bruijn graphs [16] ( DBGs ) , which are widely used for de novo genome assembly [17 , 18] and variant calling [12 , 19] . These graphs connect overlapping k-mers ( here DNA fragments ) , yielding a compact summary of all variations across a set of genomes . Fig 1 illustrates the construction of such a graph for a simple example , where the only variation among the aligned genomes is a point mutation . DBGs also accommodate more complex disparities including rearrangements and insertions/deletions ( S1 Fig ) . DBGWAS relies on the ability of compacted DBGs ( cDBGs ) to eliminate local redundancy , reflect genomic variations , and characterise the genomic environment of a k-mer at the population level . More precisely , we build a single cDBG from all the genomes included in the association study ( in practice , up to thousands ) . The graph nodes—called unitigs—represent , by construction , sequences of variable length and are at the right level of resolution for the set of genomes considered , taking into account adaptively the genomic variation . The unitigs are individually tested for association with the phenotype , while controlling for population structure . The unitigs found to be phenotype-associated are then localised in the cDBG . Subgraphs induced by their genomic environment are extracted . They often provide a direct interpretation in terms of genetic events which results from the integration of three types of information: 1 ) the topology of the subgraph , reflecting the nature of the genetic variant , 2 ) the metadata represented by node size and colour , allowing us to identify which unitigs in the subgraph are associated to a particular phenotype status , and 3 ) an optional sequence annotation helping to detect unitig mapping to—or near—a known gene . We benchmarked our novel method using several antibiotic resistance phenotypes within three bacterial species of various degrees of genome plasticity: Mycobacterium tuberculosis , Staphylococcus aureus and Pseudomonas aeruginosa . The subgraphs built from significant unitigs described SNPs or insertions/deletions in both core and accessory regions , and were consistent with results obtained with a resistome-based association study . In addition , novel genotype-to-phenotype associations were also suggested .
For P . aeruginosa levofloxacin resistance , the subgraph obtained with the lowest minq highlighted a polymorphic region in a core gene ( Fig 3A ) . Indeed , it showed a linear structure containing a complex bubble , with a fork separating susceptible ( blue ) and resistant ( red ) strains . The annotation revealed that all unitigs in this subgraph mapped to the quinolone resistance-determining region ( QRDR ) of the gyrA gene . gyrA codes for a subunit of the DNA gyrase targeted by quinolone antibiotics such as levofloxacin and its alteration is therefore a prevalent and efficient mechanism of resistance [20 , 21] . In all our experiments related to quinolone resistance , DBGWAS identified QRDR mutations in either gyrA or parC , which codes for another well-known quinolone target: P . aeruginosa levofloxacin ( first subgraph , gyrA: minq = 7 . 21 × 10−29 and second , parC: 5 . 68 × 10−06 ) , S . aureus ciprofloxacin ( first , parC: minq = 8 . 67 × 10−104 and second , gyrA: 2 . 21 × 10−76 ) , and ofloxacin resistance in M . tuberculosis , whose genome does not contain the parC gene [22] ( first , gyrA: minq = 9 . 66 × 10−144 ) . For P . aeruginosa amikacin resistance , the top subgraph ( minq = 5 . 86 × 10−9 ) highlighted a SNP in an accessory gene ( Fig 3B ) . As in Fig 3A , it contained a fork separating a blue and a red node . However , other remaining nodes were not grey: they represented an accessory sequence because they were not present in all the strains . Most of these nodes were pale-red , showing that the accessory sequence was more frequent in resistant samples . The annotation revealed that this subgraph corresponded to aac ( 6’ ) , a gene coding for an aminoglycoside 6-acetyltransferase , an enzyme capable of inactivating aminoglycosides , such as amikacin , by acetylation [23] . Most unitigs in this gene had a low association with resistance , except for the ones describing this particular SNP . Mapping the sequence of these unitigs on the UniProt database [24] revealed an amino-acid change at L83S , right in the enzyme binding site . This SNP was previously shown to be responsible for substrate specificity alteration in a strain of Pseudomonas fluorescens [25] . It appears to increase the amikacin acetylation ability of aac ( 6’ ) , making its association to amikacin resistance more significant than the gene presence itself . Finally , for M . tuberculosis ethionamide resistance , the top subgraph ( minq = 7 . 86 × 10−11 , Fig 3C ) represented a polymorphic region in a core gene promoter . The subgraph was mostly grey and linear with a localised blue and red fork . The most reliable annotation for this subgraph was fabG1 ( also known as mabA ) , a core gene previously shown to be involved in ethionamide and isoniazid resistance [26 , 27] . None of the significantly associated unitigs mapped to the fabG1 gene , but their close neighbours did ( highlighted in Fig 3C by black circles ) , suggesting that the detected variant was located in the promoter region of the gene . This was confirmed by mapping the significant unitig sequences using the Tuberculosis Mutation database of the mubii resource [28] . For S . aureus resistance to methicillin , the top subgraph ( minq = 7 . 68 × 10−188 ) , shown in Fig 3D , revealed a gene cassette insertion . It contained a long path of red nodes , and a branching region including another red node path . The first path mapped to the mecA gene , extensively described in this context and known to be carried by the Staphylococcal Cassette Chromosome mec ( SCCmec ) [21 , 29 , 30] . The other part of the subgraph represented a >5 , 000 bp fragment of the cassette . It was less linear because it summarised several types of the cassette differing by their structure and gene content [29] . The next subgraphs represented other regions of the same cassette . Interestingly , retaining a greater number of unitigs to build the subgraphs leads to merging these individual subgraphs , representing related genomic regions , into a single one . This can be done by increasing the Significant Features Filter ( SFF ) parameter value , which defines the unitigs used to build the subgraphs . By default , the unitigs corresponding to the 100 lowest q-values are retained ( SFF = 100 ) . Increasing the SFF value to 150 ( 150th q-value = 1 . 60 × 10−27 ) allowed us to reconstruct the entire SCCmec cassette , as shown in S3 Fig . For S . aureus erythromycin resistance , a unique subgraph was generated ( minq = 2 . 69 × 10−100 ) . As shown in Fig 3E , the subgraph described the circular structure of a 2 , 500 bp-long plasmid known to carry the causal ermC gene together with a replication and maintenance protein in strong linkage disequilibrium with ermC [30 , 31] . For P . aeruginosa amikacin resistance , the third subgraph ( minq = 2 . 21 × 10−6 ) represented a 10 , 000 bp plasmid acquisition . Using the NCBI nucleotide database [32] , most of the unitigs in this subgraph mapped to the predicted prophage regions of an integrative and conjugative plasmid , whose structure corresponds to a plasmid , pHS87b , recently described in the amikacin resistant P . aeruginosa HS87 strain [33] . S4 and S5 Figs provide more examples of MGEs recovered by DBGWAS , and the Interpretation of significant unitigs ( step 3 ) subsection of the Methods section discusses SFF default value and tuning . Although resistance determinants are not perfectly or exhaustively known for all species , some resistance mechanisms are well described . This is the case of gyrA and parC alteration in fluoroquinolone resistance in P . aeruginosa [20] , and of the alteration of two streptomycin targets: the ribosomal protein S12 ( coded by rpsL ) and the 16S rRNA ( coded by rrs ) in M . tuberculosis [34] . Here we verify the ability of bacterial GWAS methods to recover these known mechanisms . We compared DBGWAS results to those obtained by applying the same association model to a collection of known resistance genes and SNPs [7 , 35] ( see the Resistome-based association studies subsection of the Methods section ) , and to two other recent k-mer-based methods: pyseer [6 , 36] , and HAWK [13] . For P . aeruginosa levofloxacin resistance ( Table 2 ) , both DBGWAS and pyseer identified the two expected known causal determinants reported by the prior resistome-based study: gyrA and parC , while HAWK only reported gyrA . pyseer reported 224 k-mers , all mapping to gyrA and parC , while the other methods reported less than 10 features ( subgraphs or reassembled k-mers ) , among which were several unknown , potentially new candidate markers . For M . tuberculosis streptomycin resistance ( Table 3 ) , the four methods reported the two expected known causal determinants rpsL and rrs . However , while the resistome-based study and DBGWAS methods ranked the causal rpsL determinant first , pyseer and HAWK reported their lowest p/q-values for the false positive katG determinant . katG and other false positives caused by co-resistance were among the top-ranked features for all methods and this is a well described phenomenon in M . tuberculosis species [34 , 37] . Additional results for all antibiotics can be found in S6 and S7 Tables for resistome-based association studies , and in S3 and S5 Tables for DBGWAS . In addition to resistance markers , all three k-mer-based approaches reported several unknown variants , not described in the context of resistance . Among them , in the context of streptomycin resistance , a noncoding region between a transposase and a PPE-family protein was reported by the three methods but , as expected , not by the resistome-based approach , as only resistance genes were included in this analysis . More generally , knowledge-based approaches such as SNP- , gene- or resistome-based GWAS can be limited in the context of new marker discovery , since any causal variant absent from the chosen reference would remain untested . Besides being time-consuming , preparing such a list of genetic variants can be problematic for bacterial species without extensive annotation or reference availability . Here we describe associations identified by DBGWAS and which were never described in the antibiotic resistance literature . In our P . aeruginosa panel , the second subgraph obtained for amikacin resistance ( minq = 1 . 37 × 10−6 ) gathered unitigs mapping to the 3’ region of a DEAD/DEAH box helicase , known to be involved in stress tolerance in P . aeruginosa [38] . The unitig with the lowest q-value was present in 13 of 47 resistant strains and in only 1 of 233 susceptible strains and represented a C-C haplotype summarising two mutated positions: 2097 and 2103 . This annotation was not an artefact of the population structure , properly taken into account by the linear mixed model . Indeed the 13 resistant strains corresponded to distinct clones belonging to two phylogroups , one of them containing the susceptible strain . In P . aeruginosa levofloxacin resistance , the third subgraph ( minq = 1 . 87 × 10−2 ) represented a L650M amino-acid change in a hybrid sensor histidine kinase/response regulator . Such two-components regulatory systems play important roles in the adaptation of organisms to their environment , for instance in the regulation of biofilm formation in P . aeruginosa [39] , and as such may play a role in antibiotic resistance . In S . aureus , polymorphisms within genes not known to be related to resistance were identified for several antibiotics: purN ( minq = 2 . 02 × 10−22 ) for fusidic acid , odhB ( minq = 1 . 49 × 10−33 ) for gentamicin , ybaK and mqo1 ( minq = 9 . 30 × 10−18 , resp . 6 . 82 × 10−10 ) for trimethoprim . None of these genes have been associated with antibiotic resistance before , to the best of our knowledge . In M . tuberculosis , polymorphisms in two genes encoding proteins involved in cell wall and cell processes , espG1 and espA , were found associated with streptomycin ( seventh subgraph , minq = 9 . 43 × 10−4 ) and XDR phenotype ( third subgraph , minq = 9 . 58 × 10−36 ) , respectively . Again , these genes have never been reported in association with antibiotic resistance before . Although experimental validation would be required to tell whether these hypotheses are false positive ( e . g . , in linkage with causal variants ) or actual resistance mechanisms not yet documented , DBGWAS is a valuable tool to screen for novel candidate markers . Moreover it provides a first level of variant description ( SNPs in gene or promoter , MGE , etc ) which can directly drive the biological validation . Other k-mer-based approaches are as agnostic as DBGWAS and were also able to provide novel hypotheses , but interpreting their output can prove more challenging than a SNP/gene-based GWAS . In the M . tuberculosis streptomycin resistance experiment for example , they reported several thousands of features , while DBGWAS reported only 24 annotated subgraphs without missing any expected determinant ( see Table 3 ) . The thousands of k-mers generated by HAWK and pyseer are of course also amenable to interpretation: to build our Table 3 , we mapped these k-mers to references and extracted annotated variants which showed at least one hit . However , doing so required additional efforts and a working knowledge of the most appropriate annotated references . In addition , k-mers which do not map to the chosen reference cannot be interpreted . By contrast , DBGWAS always returns a subgraph containing these k-mers . Even when no annotation exists , the topology and colours of the subgraphs may hint towards the nature of the causal variant . In addition to providing context for significant k-mers and guiding their interpretation as SNPs or MGEs , DBGWAS clustering of close variants into a subgraph can describe hypervariable regions as single entities , and highlight highly associated haplotypes . As an example , the top subgraph for rifampicin resistance ( minq = 4 . 84 × 10−70 ) contained 36 significant unitigs , distinguishing between susceptible ( blue ) and resistant ( red ) strains . Instead of a single point mutation , this subgraph represented a polymorphic region known as the rifampicin resistance-determining region ( RRDR ) of the rpoB gene . The unitig with the lowest q-value covered several mutant positions , defining a particular haplotype strongly associated with rifampicin susceptibility . Where DBGWAS reported in this case only one subgraph , pyseer , for instance , reported 470 k-mers with the rpoB annotation , and the resistome-based association study reported in this case 4 distinct SNPs in rpoB ( S6 Table ) . In another user-submitted example , DBGWAS identified mosaic alleles of three pbp genes involved in beta-lactam resistance of Streptococcus pneumoniae . Like in the RRDR example , it returned five subgraphs corresponding to the three genes—three subgraphs were annotated pbp2x and represented three distinct polymorphic regions of the gene . Each subgraph summarised the polymorphism of the gene , as opposed to one separate feature for each SNP . Admittedly , some subgraphs output by DBGWAS are not readily interpretable: they are neither coloured bubbles highlighting SNPs , nor long single-coloured paths denoting MGE insertions . This was the case of several subgraphs produced for P . aeruginosa amikacin resistance , and presented in S6 Fig . Genetic variants inserted in variable regions , for example , lead to subgraphs with a high average degree , or to very large subgraphs . The fourth subgraph for instance ( minq = 2 . 21 × 10−6 ) contains a path of three red ( positively-associated ) nodes lying in a noncoding region between variable accessory genes . Consequently , their neighbour unitigs branch to various other unitigs , making the structure complex and hard to interpret . Complex subgraphs also arise when several associated variants have overlapping neighbourhoods ( as defined in the Graph neighbourhoods subsection in the Methods section , and tuned with the nh parameter ) in at least one strain . This is the case for the subgraph with the smallest minq which aggregates aac ( 6′ ) acetyltransferase and the CML efflux pump . The interpretation of such subgraphs is not straightforward . We often found it helpful to tune the nh and SFF parameters to break large subgraphs into a set of smaller ones , as discussed in the discussed in the Methods section . For the aac ( 6′ ) subgraph , where nearby variants are aggregated into a large subgraph , reducing the SFF value to 15 provided a much smaller and easier-to-interpret subgraph focusing on the aac ( 6′ ) mutation ( Fig 3B ) . Otherwise , we recommend to focus on the topology of the most significant unitigs and their close neighbours . To assess the scalability of DBGWAS to large datasets , we retrieved 5 , 000 genomes from M . tuberculosis , 9 , 000 genomes from S . aureus and 2 , 500 genomes from P . aeruginosa , as described in the Large panels subsection of the Methods section . We present in S9 Fig the runtime and memory usage performances for these panels . All 180 runs took less than 5 days and 250 GB of RAM on 8 cores . Both the computational time and memory usage increase log-linearly with the panel size . Moreover , at equal panel size , DBGWAS performance also depends on the genome complexity , requiring less computational resource for more clonal genomes such as M . tuberculosis . We also compared the computational performance of DBGWAS with pyseer and HAWK . The benchmark was performed on 13 datasets , including one large dataset of 2 , 500 genomes for each of the 3 species ( see the Datasets subsection in the Methods section for details ) . Detailed results are presented in S2 Table . DBGWAS was the fastest tool in 11 out of 13 experiments , always taking less than 2 hours . HAWK ran in less than 10 hours in 12 out of 13 experiments , and was a little faster than DBGWAS on two of the large-scale datasets . pyseer took from 13 to 53 hours on 9 experiments , and failed on the 4 others: one exceeded the disk space limit of 1TB , three exceeded the runtime limit of five days . It was brought to our attention during the reviewing process that piping the output of fsm-lite through gzip would decrease the disk space usage . HAWK was more parsimonious in memory usage than DBGWAS on the large scale panels . This can be explained by the fact that the 0 . 8 . 3-beta version of HAWK which we are using does not take into account the population structure , and as such does not have to compute an n × n covariance matrix , providing it a large gain in memory usage—and , to a lesser extent , runtime—for large panels . On the other hand , disregarding the population structure could also lead to spurious discoveries . HAWK v0 . 9 . 8-beta offers an adjustment but failed to recover the known true positives , which is why we chose to present the results of the 0 . 8 . 3-beta version . DBGWAS and HAWK typically used one order of magnitude less memory than pyseer . The most memory-consuming step for pyseer was the k-mer counting step relying on fsm-lite .
In this article we introduce an efficient method for bacterial GWAS . Our method is agnostic: it considers all regions of the genomes and is able to identify potentially new causal variants as different as SNPs in noncoding regions and MGE insertions/deletions . It performs as well as the current SNP- and gene-based gold standard approaches for retrieving known determinants , from genome pre-assemblies and without relying on annotations or reference genomes . DBGWAS exploits the genetic environment of the significant k-mers through their neighbourhood in the cDBG , providing a valuable interpretation framework . Because it uses only contig sequences as input , it allows GWAS on bacterial species for which the genomes are still poorly annotated or lack a suitable reference genome . DBGWAS makes bacterial GWAS possible in two hours using a single-core computer ( see S1 Table ) , outperforming other state-of-the-art k-mer-based approaches . Underlying our method , graph-based genome sequence representations such as DBGs , extend the notion of the reference genome to cases where a single sequence stops being an appropriate approximation [40 , 41] . As demonstrated in this paper , they pave the way to GWAS on highly plastic bacterial genomes and could also be useful for microbiomes [42] or human tumours [13] . DBGWAS currently relies on the Benjamini-Hochberg procedure to control the FDR and offers no advance exploiting the dependence among presence/absence patterns . An important improvement would be to control the false discovery rate at the subgraph level instead of the unitig level . DBGWAS could be extended to different statistical tasks by adapting its underlying association model , to allow for continuous phenotypes or identify epistatic effects , for instance . The interpretability of the extracted subgraphs could also be improved by training a machine learning model to predict which types of event they represent [43] . This automated labelling could guide users in their interpretation and allow them to search for specific events , such as SNPs in core genes or rearrangements . Several recent studies describe in silico models for defining a genomic antibiogram and hopes are high that such technologies will complement the classic phenotypic methods [44] . Several studies have already demonstrated that in some cases , genomic antibiograms can be at least as good as phenotypic ones [30 , 45–47] . Contrary to our approach , these studies require extensive resistance marker databases . DBGWAS will surely contribute to the extension of such databases or to the development of agnostic genomic antibiograms . In conclusion , we demonstrate for three medically important bacterial species that resistance markers can be detected rapidly with relative ease , using simple computer equipment . Our integrated software and visualisation tools offer an intuitive variant representation , hence will provide future users with an enhanced insight into genotype to phenotype correlations , in all domains of microbiology , beyond that of antibiotic resistance . This will include complex traits such as biofilm formation , epidemicity and virulence .
DBGs are directed graphs that efficiently represent all the information contained in a set of sequences . Nodes represent all the unique k-mers ( genome sequence substrings of length k ) extracted from the input sequences . Edges represent ( k − 1 ) -exact-overlaps between k-mers: an edge connects a node n1 to a node n2 if and only if the ( k − 1 ) -length-suffix of n1 equals the ( k − 1 ) -length-prefix of n2 ( Fig 1A ) . These graphs can be compacted into cDBGs by merging linear paths ( sequences of nodes not linked to more than two other nodes ) into a single node referred to as a unitig [48–50] ( Fig 1C ) . Compaction yields a graph with locally optimal resolution: regions of the genome which are conserved across individuals are represented by long unitigs , while regions which are highly variable are fractioned into shorter unitigs ( S1 Fig ) . Human GWAS literature extensively discusses how testing procedures can result in spurious associations if the effect of the population structure is not taken into account [53–55] . Population structures can be strong in bacteria because of their clonality [5 , 6 , 56 , 57] . An additional performance analysis comparing several models for population structure , on both simulated and real data , showed that correcting for population structure using LMMs is often preferable to using a fixed effect correction or not correcting at all ( S1 Appendix ) . We thus rely on the bugwas method [5] , which uses the linear mixed model ( LMM ) implemented in the GEMMA library [58] , to test for association with phenotypes while correcting for the population structure . This method also offers the possibility to test for lineage effects , by calculating p-values for association between the columns of the matrix representing the population structure , and the phenotype [5] . DBGWAS optionally provides bugwas lineage effect plots when the user specifies a phylogenetic tree using the -newick option . An example of the generated figures is available at http://pbil . univ-lyon1 . fr/datasets/DBGWAS_support/full_dataset_visualization/ . Formally , the LMM represents the distribution of the binarized phenotype Yi , given the j-th minor allele pattern Xij and the population structure represented by a set of factors W ∈ R n ≤ p , by: Y i = X i j β + W i T α + ε i j , j = 1 , … , p . ( 1 ) β is the fixed effect of the tested candidate on the phenotype , α ∼ N ( 0 , σ a 2 ) , σ a 2 > 0 is the random effect of the population structure , and ε i j ∼ iid N ( 0 , σ 2 ) are the residuals with variance σ2 > 0 . W is estimated from the Z matrix , which includes duplicate columns representing both core and accessory genome . More precisely , denoting Z = USV⊤ the singular value decomposition of Z , we use W = US . We test H0: β = 0 versus H1: β ≠ 0 in Eq 1 for each pattern using a likelihood ratio procedure producing p-values and maximum likelihood estimates β ^ . To tackle the situation of multiple testing caused by the high number of tested patterns , we compute q-values , which are the Benjamini-Hochberg transformed p-values controlling for false discovery rate ( FDR ) [59] . The LMM is used to identify de-duplicated minor allele presence patterns significantly associated with the phenotype at a chosen FDR level . While the testing step is done at the pattern level , the interpretation of the selected features is done at the unitig level . As a result of the de-duplication procedure , a given pattern may correspond to several distinct unitigs . To faithfully interpret the results , all the unitigs corresponding to the significant patterns are retrieved and are assigned the q-value of their pattern . We now show how the initial cDBG can be used in the interpretation step . We used in our experiments genome sequences from three bacterial species with various degrees of genome plasticity , from more clonal to more plastic: M . tuberculosis , S . aureus , and P . aeruginosa . We also built large datasets with random phenotypes for these 3 species , and used them only for time performance and memory usage assessment . All panels are summarised in Table 4 . We benchmarked DBGWAS against a targeted approach to ensure its ability to retrieve all expected resistance determinants . We thus performed association studies under the same model , using as input a collection of known causal resistance SNPs and genes , defining the resistome . In this validation study , we used bugwas with the same phenotypes and population structure matrix W , so the resistome-based analyses and DBGWAS only differ by their input variant matrix ( unitigs versus SNPs or genes presence/absence ) . For P . aeruginosa resistome , we use a variant matrix previously described [7] , which includes presence/absence of known resistance gene variants , as well as the SNPs called against these reference gene variants . For M . tuberculosis resistome , we built the variant matrix using the same approach as for P . aeruginosa [7]: we called the SNPs from a list of 32 known resistance genes and promoters [34 , 67 , 73] . The time and memory usage required for the complete analysis ( from the mapping of the resistance genes and positions on the genome assemblies to the association study ) are provided in Tables 2 and 3 . We sort the annotated features by q-values . S6 and S7 Tables summarise all top variants using their q-value ranks , while Tables 2 and 3 report the annotations of all variants with a q-value < 0 . 05 for P . aeruginosa levofloxacin and M . tuberculosis streptomycin resistance , respectively . | Genome-wide association studies ( GWAS ) help explore the genetic bases of phenotype variation in a population . Our objective is to make GWAS amenable to bacterial genomes . These genomes can be too different to be aligned against a reference , even within a single species , making the description of their genetic variation challenging . We test the association between the phenotype and the presence in the genomes of DNA subsequences of length k – the so-called k-mers . These k-mers provide a versatile descriptor , allowing to capture genetic variants ranging from local polymorphisms to insertions of large mobile genetic elements . Unfortunately , they are also redundant and difficult to interpret . We rely on the compacted De Bruijn graph ( cDBG ) , which represents the overlaps between k-mers . A single cDBG is built across all genomes , automatically removing the redundancy among consecutive k-mers , and allowing for a visualisation of the genomic context of the significant ones . We provide a computationally efficient and user-friendly implementation , enabling non-bioinformaticians to carry out GWAS on thousands of isolates in a few hours . This approach was effective in catching the dynamics of mobile genetic elements in Staphylococcus aureus and Pseudomonas aeruginosa genomes , and retrieved known local polymorphisms in Mycobacterium tuberculosis genomes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"bacteriology",
"antimicrobials",
"genome-wide",
"association",
"studies",
"infographics",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"drugs",
"microbiology",
"pseudomonas",
"aeruginosa",
"antibiotic",
"resistance",
"... | 2018 | A fast and agnostic method for bacterial genome-wide association studies: Bridging the gap between k-mers and genetic events |
Lymphatic filariasis ( LF ) and soil-transmitted-helminths ( STH ) are co-endemic in 58 countries which are mostly in Africa and Asia . Worldwide , 486 million school-age children are considered at risk of both diseases . In 2000 , the World Health Organization ( WHO ) established the global programme to eliminate LF by 2020 . Since then , the LF elimination programme has distributed ivermectin or diethylcarbamazine citrate ( DEC ) in combination with albendazole , thereby also treating STH . Consequently , many school-age children have been treated for STH through the LF programme . As treatment targets towards the 2020 LF elimination goal are achieved , many countries are implementing the transmission assessment survey ( TAS ) and , if the LF prevalence is estimated to be less than 1% , scaling down mass drug administration ( MDA ) . We analysed the 2014 data on preventive chemotherapy ( PC ) reported from LF STH co-endemic countries and projected the year and location of TAS expected to be conducted between 2016 and 2020 to assess the impact of this scaling down on STH PC . Eighty percent of all co-endemic countries that have already stopped LF MDA nationally were able to establish STH PC through schools . It is estimated that 14% of the total number of children presently covered by the LF programme is at risk of not continuing to receive PC for STH . In order to achieve and maintain the WHO 2020 goal for STH control , there is an urgent need to establish and reinforce school-based deworming programmes in countries scaling-down national LF elimination programmes .
Lymphatic filariasis ( LF ) and soil-transmitted helminth ( STH ) infections are co-endemic in 58 countries worldwide: 34 in Africa region , 12 in Western Pacific region , six in South-East Asia region , four in America region and two in Eastern Mediterranean region ( Fig 1 ) . Mass drug administration ( MDA ) is the recommended preventive chemotherapy ( PC ) strategy of delivering a combination of albendazole and ivermectin or diethylcarbamazine citrate ( DEC ) for LF elimination . Since albendazole also treats STH infections [1] , MDA for LF has often replaced specific interventions targeting STH in co-endemic areas . In 2014 , it was estimated that 46% of those school-age children ( SAC ) who received PC for STH received anthelminthics through MDA targeting LF while 54% were dewormed through school-based interventions ( Fig 2 ) . LF is targeted for global elimination as a public health problem by 2020 [2] . At least 5 years of MDA with effective coverage is expected to have reduced LF infection to a level at which transmission can no longer be sustained and thus MDA can stop . WHO recommends the transmission assessment survey ( TAS ) as the decision-making tool to determine when to stop MDA [3] . Where the number of LF infected children is less than or equal to the critical cut-off value the assessment is considered ‘passed’ and MDA may cease . After stopping MDA , countries are expected to conduct TAS twice over a 4–5 year period to confirm whether LF elimination has been achieved and sustained [4] . On the other hand , for STH , the WHO target for 2020 is to treat at least 75% of SAC living in STH endemic countries [5] . Using a school-based platform to distribute anthelminthic treatment is one of the most cost-effective approaches that WHO recommends in STH control strategies . The aim of this study is to assess the effect of the progressive scaling down of MDA for LF on STH control and the status of the measures in place by different countries to mitigate this effect .
The study is a secondary analysis of aggregated data officially reported annually by Ministries of Health of endemic countries to WHO . The data are publicly available and anonymous .
In 2014 , of the 34 co-endemic countries for LF and STH in Africa region , LF MDA was implemented at least partially in 20 countries . Six countries ( Benin , Cameroon , Cote d’Ivoire , Madagascar , Mozambique and Senegal ) reached at least 75% of STH treatment coverage . More than half of SAC in need of STH PC were treated through LF MDA or PC with schools as drug distribution channels ( Table 1 ) . Two countries , Togo and Malawi , stopped LF MDA nationwide after they implemented and passed the first TAS ( i . e . TAS1 ) , STH PC for school-age children was conducted in both countries ( Table 2 ) but Malawi was not able to reach the 100% geographic coverage . Other countries like Benin , Burkina Faso , Ghana , Mali , Nigeria , and Tanzania stopped LF MDA in 87 IUs ( Table in S1 Table ) and the data analysis in those IUs demonstrated that STH PC for school-age children was conducted in all countries except in Burkina Faso and Mali . Sierra Leone , Guinea and Liberia did not implement LF MDA in 2014 due to the Ebola virus epidemic . In the period 2016–2020 , TAS1 is expected to be conducted in 1866 IUs in the 12 countries that have started LF MDA scaling down . The total number of SAC living in those areas is 73 million of which 18 million are at risk of not continuing to receive PC for STH . In this region , there are four LF STH co-endemic countries: Brazil , Dominican Republic , Guyana and Haiti . In 2014 , Brazil stopped LF MDA in 27 endemic IUs and remained with 2 IUs that are still treating for LF until 2017 , while the Dominican Republic stopped LF MDA in five IUs . Moreover , these two countries were able to implement STH PC for school-age children nationally while in Haiti , SAC were dewormed through both LF MDA and school intervention . In Guyana , school-age children were dewormed for STH only through LF MDA ( Table in S1 Table ) . Two countries , Dominican Republic and Haiti surpassed WHO target for STH control by treating 100% and 91% of SAC respectively . In the period 2016–2020 , TAS1 is expected to be conducted in 134 IUs in four countries . The total number of SAC living in those areas is 3 million of which 900 , 000 are presently not covered by a school programme targeting STH . Among the nine LF-endemic countries , six are co-endemic with STH infections: Bangladesh , India , Indonesia , Myanmar , Nepal and Timor-Leste . All of these except Timor-Leste have implemented TAS and stopped MDA in some IUs . The most significant scale-down has occurred in India and Bangladesh with 71 and 18 IUs having already stopped LF MDA , respectively . Looking at the 2014 STH PC data in those IUs , school-aged children were dewormed for STH in Bangladesh and in one state in India . Nepal and Myanmar conducted LF MDA in endemic areas and dewormed school-age children for STH nationally . Both Bangladesh and Myanmar achieved the 75% minimum treatment coverage in this age group . Indonesia reported deworming of SAC for STH only in LF-endemic areas while Timor-Leste did not conduct any MDA in 2014 ( Table in S1 Table ) . By the end of 2016 , the region will have more than 100 million SAC at risk of STH in IUs where LF MDA will stop , most of whom are from India ( Fig 3 ) . However , India launched a national deworming programme in 2015 which aims to scale up in areas where LF MDA will scale down [9] . In the period 2016–2020 , TAS1 is expected to be conducted in 467 implementation units in 5 countries . The total number of SAC living in those areas is 132 million of which 9 million are presently not covered by a school programme targeting STH . Sudan and Yemen are the two countries in the region that require both LF and STH PC . Yemen is in post LF MDA surveillance while Sudan is finalising the mapping to refine the population requiring LF MDA . In 2014 , Yemen dewormed SAC and achieved WHO minimum target for STH control while Sudan implemented STH PC for school-age children in 12 IUs ( Table 2 and Table in S1 Table ) . In this region , among the 22 LF-endemic countries , 12 are co-endemic with STH , 6 of which have already stopped LF MDA nationally and are under post-MDA surveillance . Among these , only one country , Tonga , did not conduct STH PC for school-age children in 2014 . Philippines , Cambodia and Vietnam have the highest number of school-age children requiring treatment and in 2014 they all conducted STH PC for school-age children ( Table 2 ) . The last two countries attained the 75% and 100% STH treatment coverage and geographic coverage targets respectively and have already stopped the LF MDA . Moreover , Philippines has scaled down LF MDA significantly: 23 IUs have already stopped LF MDA and school-age children were dewormed for STH in 18 IUs ( Table in S1 Table ) ) . Micronesia and Papua New Guinea dewormed school-age children only through the LF programme . In the period 2016–2020 , TAS1 is expected to be conducted in 17 IUs in four countries . The total number of SAC living in those areas is 6 million of which 2 million are not covered by a school programme targeting STH . Globally , nine STH LF co-endemic countries have already stopped LF MDA nationally . Seven of them ( 80% ) have been able to transition successfully to STH deworming for school-age children through schools and three were able to achieve and maintain the WHO minimum STH PC target . However , Malawi and Tonga were not able to reach the total number of school-age children that were routinely dewormed through LF ( Table 2 ) . Among the 41 countries that have stopped LF MDA , at least partially , five ( 12% ) have been able to successfully transition to school deworming for STH in all IUs that stopped LF MDA , ten ( 24% ) treated at least 75% of SAC requiring PC . Of 15 countries that are expecting to stop LF MDA in some IUs by 2020 , more than 85% have school deworming programmes which have taken on deworming of SAC in all IUs that stopped LF MDA . However , three countries ( Burkina Faso , Mali and India ) that have school deworming programmes were unable to deworm school-age children in all areas where LF MDA has already stopped in 2014 . From 2016 and onward , the number of SAC at risk of STH in areas stopping LF MDA will increase significantly reaching 160 million . Since the greatest proportion will be in countries that have already put in place a national STH deworming programme expecting to cover all SAC in need of treatment , the expected number of children not covered by school deworming programmes is estimated to be around 30 million globally ( Fig 3 ) . To facilitate the identification of countries that need to consider this imminent need for scaling-up school-based deworming , we divided them into three categories: category one includes countries that have completely or partially stopped LF MDA and which were already able to successfully continue STH PC for school-age children through schools; category two includes countries that have completely or partially stopped LF MDA and which conducted STH PC for school-age children , but did not reach those that were routinely dewormed through LF and category three includes countries that do not have a national school deworming programme ( Table 3 ) .
LF MDA has been used for many years as a platform for integrated PC to control diseases including STH and as a community-based drug distribution programme; it has reached billions of people [10] which includes SAC . The analysis of 2014 data shows that in all LF-endemic WHO regions , LF MDA has started scaling down , with nine STH co-endemic countries having already stopped MDA at the national level . School-based deworming is a safe , simple and cost-effective control strategy for STH infections recommended by WHO , that can reach both enrolled and non-enrolled school-age children [11] . With LF MDA scaling down whether these two infections co-exist , countries should effectively transition from LF MDA to school-based deworming for STH control after assessing the epidemiology of STH infection to determine whether STH PC should continue in the absence of LF MDA . The results of 2014 data analysis demonstrate that only a minority of the total number of children presently covered by LF programmes are not currently covered by a school deworming programme . Most of these SAC children are in African countries . This is the geographical region where most attention should be focused . The results also demonstrate that some countries have been more successful than others in achieving this transition from LF MDA dependent deworming of SAC to school-based deworming programmes . Three factors contributing to this success included the establishment of a school-based deworming programme before the end of LF MDA , integration of STH data collection with TAS and intersectoral mobilisation of resources to sustain STH deworming activities . Countries that have transitioned effectively to school- based PC are mainly the ones that had simultaneously implemented school and community based PC interventions in collaboration with other sectors , notably with the education sector ( e . g . countries in category 1 , Table 3 ) . Case studies from World Bank [12] describe Cambodia and Vietnam as examples of countries that have achieved and sustained WHO target of treating at least 75% of SAC , attributing their success to multisectoral collaboration . Consequently , WHO/NTD is working with multiple partners ( e . g . World Bank , Global Partnership for Education , NGOs ) to forge collaboration between health and education sectors at all levels . Such collaborations facilitate the mobilization of resources required to support school-based deworming programmes . In the meantime , some countries are establishing national school-based deworming programmes on their own initiative . For example India launched a national school-based deworming programme in 2015 that is expected to cover all SAC in the country , a total of 136 million , which is 24% of SAC population in need of STH PC worldwide . Impact studies on the epidemiology of STH during LF TAS allow national programme managers to assess the prevalence and intensity of infection in the target population and thereby adjust the frequency of deworming interventions in the school-age population . WHO has published a protocol to help countries integrate STH data collection in TAS [13] . Two studies conducted in Sri Lanka and Burkina Faso on assessment of STH morbidity and prevalence during LF TAS enabled these countries to review their STH control strategies leading up to 2020 [14 , 15] . STH data collected during TAS showed that in Burkina Faso SAC do not need deworming in area where LF MDA stopped; the STH prevalence and intensity had dropped significantly and the country is planning to review its strategy to consolidate these gains . Additionally , WHO urges partners that were supporting LF MDA to continue to support countries in their transition to conduct school-based deworming for SAC and STH epidemiological surveys during LF TAS as countries work to establish sustainable STH control strategies post LF MDA . In conclusion , the scaling down of LF MDA will affect STH PC . Many countries have school based deworming programmes that will take over and sustain the gains acquired during the LF elimination programme . WHO urges countries to undertake STH epidemiological surveys along with LF TAS which will allow them to determine their appropriate strategy for the control of STH in SAC with an emphasis on integration into existing government structures and multisectoral collaboration . Additionally , future analysis should examine the public health effects of LF MDA deworming on women of childbearing age and similarly review the implications of the ending of LF MDA for this important age-group also at risk of STH infection and its associated complications . | Lymphatic filariasis ( LF ) and soil-transmitted helminths ( STH ) ( i . e . intestinal worms ) are two tropical diseases that are found together in 58 countries in the world . School-age children are most affected by intestinal worms , albendazole , one of the two drugs used for LF , also treats STH . For this reason , large-scale delivery of LF drugs in the community has been used as a means to also treat school-age children for intestinal worm infections in many countries . In line with the WHO goal to eliminate LF by 2020 , countries that have achieved that objective have started stopping community-based LF treatment . Therefore , we analysed treatment data from 2014 to quantify the effect of this reduction on treatment of school-age children for intestinal worms . The results show that 80% of countries that have already stopped LF treatment were able to administer deworming drugs for STH to school-age children within school-based treatment programmes . There is an urgent need to continue to establish and strengthen deworming through school health programmes in endemic countries in order to meet the WHO established goal to treat at least 75% of at risk school-age children for STH by 2020 . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"education",
"cancer",
"treatment",
"sociology",
"tropical",
"diseases",
"geographical",
"locations",
"india",
"social",
"sciences",
"parasitic",
"diseases",
"oncology",
"pharmaceutics",
"drug",
"administration",
"neglected",
"trop... | 2016 | The Impact of Lymphatic Filariasis Mass Drug Administration Scaling Down on Soil-Transmitted Helminth Control in School-Age Children. Present Situation and Expected Impact from 2016 to 2020 |
Repressors are frequently deployed to limit the transcriptional response to signalling pathways . For example , several co-repressors interact directly with the DNA-binding protein CSL and are proposed to keep target genes silenced in the absence of Notch activity . However , the scope of their contributions remains unclear . To investigate co-repressor activity in the context of this well defined signalling pathway , we have analysed the genome-wide binding profile of the best-characterized CSL co-repressor in Drosophila , Hairless , and of a second CSL interacting repressor , SMRTER . As predicted there was significant overlap between Hairless and its CSL DNA-binding partner , both in Kc cells and in wing discs , where they were predominantly found in chromatin with active enhancer marks . However , while the Hairless complex was widely present at some Notch regulated enhancers in the wing disc , no binding was detected at others , indicating that it is not essential for silencing per se . Further analysis of target enhancers confirmed differential requirements for Hairless . SMRTER binding significantly overlapped with Hairless , rather than complementing it , and many enhancers were apparently co-bound by both factors . Our analysis indicates that the actions of Hairless and SMRTER gate enhancers to Notch activity and to Ecdysone signalling respectively , to ensure that the appropriate levels and timing of target gene expression are achieved .
Growth , patterning and differentiation during development are coordinated by the activity of conserved signalling pathways whose action is largely realised through the transcriptional programmes they regulate . This necessitates mechanisms that ensure appropriate gene expression programmes are initiated , depending on cellular context , and that allow the fine tuning of gene expression in response to signalling levels . One way this is achieved is through the deployment of repressors [1 , 2] [3] . Originally thought to be the primary factor that renders enhancers and promoters silent when signalling is absent , the role of repressors has become more enigmatic since several have been found to reside at sites of active chromatin [4–6] . The Notch pathway is one example where the outcome of signalling is tuned by repressors . When activated , the Notch receptor becomes cleaved , releasing the intracellular domain , NICD , which collaborates with a DNA binding protein , CSL , to regulate gene expression [7–9] . Several different co-repressors have been found to interact directly with CSL and are proposed to keep target genes silenced in the absence of Notch activity [10] . In Drosophila , the best characterized co-repressor , Hairless , is a large unstructured protein that binds directly to Suppressor of Hairless ( Su ( H ) ) , the Drosophila CSL [11–15] . As implied by their names , Su ( H ) was first identified on the basis of loss of function alleles that suppressed the phenotypes caused by a reduction in Hairless levels ( H/+ ) , highlighting the intimate relationship between these two proteins [16 , 17] . Hairless itself functions as an adaptor , binding to Groucho and CtBP that in turn have the capability to interact with histone deacetylases ( HDACs ) [18–20] . It is thought that the HDACs prevent gene activity by modifying the local chromatin environment rendering it refractory to transcription . While the model that CSL is complexed with repressors in the absence of Notch activity is a widely accepted one , there remain many uncertainties . First , although Hairless is a well established partner of Su ( H ) in Drosophila , it is not well conserved outside the Diptera [21] and its role appears to be fulfilled by multiple different proteins in mammals , including KyoT2 [22 , 23] , MINT/SHARP [24–26] and SMRT [27] . While these proteins clearly bind to CSL and form a complex with HDACs and demethylases that modify chromatin [25 , 28] , their functional roles have only been demonstrated at a few specific loci . Thus it remains questionable how widely such repression mechanisms operate . Second , with the diversity of CSL partners it is unclear what contributions the different types of complex might make , for example , whether they are recruited to different targets or have different sensitivities to NICD . Indeed , even in Drosophila the SMRT orthologue SMRTER , has also been reported to bind to CSL [29] and might contribute to repression at a subset of targets [30 , 31] . To gain insight into the regulatory contribution from CSL interacting co-repressors , we analysed the genome-wide binding profiles of Hairless in Drosophila Kc cells and in wing imaginal discs . As predicted , Hairless binding showed substantial overlap with that of Su ( H ) , with a clear correlation in binding intensities . However , analysing the regulation of specific targets also revealed: ( i ) that the role of Hairless in silencing Notch targets is limited and is not a pre-requisite for a gene’s ability to respond to Notch activity; ( ii ) silencing of Hairless insensitive genes cannot be accounted for by binding of SMRTER , another co-repressor . The genome-wide profiles together revealed that many enhancers are bound by all 3 factors , Su ( H ) , Hairless and SMRTER , and detailed investigations into an enhancer from thread/Diap1 suggest this is indicative of co-regulation by Notch and Ecdysone signalling .
To investigate the extent of Hairless recruitment to chromatin in Kc cells , two strategies were taken . First , GFP-tagged Hairless and GFP-Su ( H ) were generated in the context of genomic fragments spanning the gene loci , so that the proteins were expressed at close to physiological levels when introduced into Kc cells ( e . g . S1 Fig ) . The chromatin association of GFP-Hairless in the Kc cells was then measured by chromatin immunoprecipitation ( ChIP ) . With no DNA binding domain , Hairless is recruited to DNA indirectly and its binding was best detected using a two-step crosslinking method [6 , 32] . Second , a Hairless-Dam fusion was generated , so that any sites of Hairless recruitment would become methylated [33] . Both methods yielded profiles with similar distributions across the genome ( S1 Fig ) . We therefore generated a high confidence profile of Hairless occupancy by intersecting data from the two , although we note that the DamID data will be biased by the distribution of the target GATC sites . Over 50% of the regions identified by DamID were also detected by ChIP , a greater proportion than for similar data-sets from Groucho and GAGA-factor ( S1 Fig [33] ) . This parsimonious approach identified 1406 Hairless bound regions ( S1 Table ) . For comparison , a profile of GFP-Su ( H ) generated in a similar manner to GFP-Hairless , identified 376 bound regions ( 1% FDR , S1 Table ) . Su ( H ) is known to form a complex with Hairless [11] and , consistent with this model , the binding profiles of the proteins were found to overlap at many loci ( Fig 1A and 1B ) . For example , at the well characterized Notch regulated E ( spl ) -complex genes , there was excellent correspondence of GFP-Su ( H ) and GFP-Hairless binding , while Hairless-Dam produced a broader profile centred on the same regions ( Fig 1B ) . Overlapping the genome-wide binding profiles revealed that 50% of Su ( H ) bound regions corresponded with high confidence Hairless binding ( Fig 1A ) . These co-occupied positions were also highly enriched for the Su ( H ) DNA binding motif ( p = 3 . 36e-09 ) . Furthermore , when the broader regions surrounding each Su ( H ) peak were considered , 83 . 5% had Hairless binding in the vicinity and in the majority ( 70% ) of loci with Hairless and Su ( H ) binding more than one region of significant Hairless binding was detected . This is suggestive of extra contacts made through DNA looping or through the formation of large complexes . Although many Su ( H ) bound regions exhibit Hairless binding in close proximity , the converse is not the case: 87% of regions enriched for Hairless do not overlap with Su ( H ) bound sites . There are two possible interpretations: one is that Su ( H ) binding at those positions was not captured ( i . e . false negatives ) , the other is that additional factors besides Su ( H ) can recruit Hairless to DNA . Motif analysis suggests that the former explanation is likely to make , at best , a minor contribution , since enrichment for Su ( H ) motifs in the Hairless-only regions was only marginally significant ( rank 407 , p = 0 . 022 ) . This suggests that other factors may contribute to Hairless recruitment . Many of the Hairless-only regions ( >37% ) also exhibit binding to the Hairless partner Groucho [6] ( S2 Fig ) , consistent with these being bona fide co-repressor bound sites ( e . g . CHES-1-Like ) . Finally , in cells depleted for Hairless , there were significant changes in RNA levels from genes that were bound by both Su ( H ) and Hairless as well as from those bound by Hairless only ( S2 Fig ) . Notably however , the former were all increased , indicative of de-repression , whereas the Hairless-only genes exhibited both increased and decreased expression in Hairless depleted cells ( S2 Fig;S3 Table ) . Two genes bound by Su ( H ) only were also deprepressed and are potentially examples of co-bound genes where we failed to detect Hairless binding ( false negatives; S2 Fig; S3 Table ) . While there may be different modes of recruitment , 63% of regions occupied by Hairless exhibited chromatin modifications associated with active enhancers ( signature 3 , Enh; Fig 1C; S2 Fig ) , characterized primarily by H3K4me1 and H3K27ac marks amongst others [34] . A smaller fraction ( 21% ) mapped to regions with characteristics of active transcription start-sites ( Signature 4 , aTSS ) , which are also enriched in active chromatin marks . The remainder were located in intronic or primed chromatin . This distribution is broadly similar to that of Su ( H ) ( Fig 1C; [34] ) , and the regions that were co-bound had similar profiles to those bound by Hairless only ( S2 Fig ) . A slightly larger fraction of Su ( H ) -only regions were associated with active transcription start-sites ( Signature 4/ aTSS; S2 Fig ) : a property also described for the mammalian homologue [35] . Given that Hairless containing complexes are proposed to repress transcription , the fact that it predominantly occupies active enhancer regions seems at first surprising . However , similar recruitment across active loci has also been observed for other co-repressors where , in some cases , they modulate the transcriptional response by curtailing the extent of active histone modifications [4–6] . To assess whether Hairless could do likewise , we examined the consequences of Hairless depletion on acetylated Histone H3 ( H3K56ac ) , a modification that is increased following Notch activation [34] . We found no global changes in the levels of H3K56ac in Hairless depleted cells ( S3 Fig ) . Strikingly however , there was a significant ( 1% FDR ) increase at the E ( spl ) locus , similar to that observed when Notch signalling is activated , indicating that the presence of Hairless can suppress histone acetylation ( Fig 1D ) . A small number ( <10 ) of other loci also exhibited significantly increased H3K56ac ( e . g . S3 Fig ) while the remainder showed no change . Thus Hairless can suppress histone acetylation , but only at a subset of bound loci . Taken together , these data support a model in which Hairless is recruited to enhancers by interactions with Su ( H ) and that it may , under some circumstances , modulate chromatin modifications at those positions and contribute to repression . However , the broad profile of Hairless binding indicates that it is also likely to be brought to chromatin independently of Su ( H ) , and its effects may be complex as some bound regions exhibited increased gene expression in cells depleted of Hairless , whereas others showed decreases . To determine whether Hairless exhibits similar occupancy profiles in tissues , transgenic flies containing a genomic construct carrying GFP-tagged Hairless were generated . The line was capable of rescuing the viability of H[P8]/H[36] mutant flies , indicating that the GFP-Hairless protein was functional and expressed at physiological levels . The chromatin association of GFP-H in wing discs was determined by ChIP and compared to existing data for Su ( H ) occupancy [37] . Fewer regions ( 493; S4 Table ) were detectably bound by Hairless in wing discs than in Kc cells , possibly because the profile represents the mean of binding in several different cell-types so that only regions bound in the majority of cells can be reliably identified . Nevertheless , a significant proportion ( 30% ) of the 493 bound regions overlapped with Su ( H ) binding ( Fig 2A ) , and were enriched for the Su ( H ) motif ( p = 2 . 01e-6 ) . Besides the well-characterized E ( spl ) locus , such regions included deadpan ( dpn; Fig 2 ) , Serrate and Notch . For example , both Su ( H ) and Hairless were bound at an intronic enhancer of the deadpan ( dpn ) gene that has been shown to confer expression in wing discs ( Fig 2B; [38] . Consistent with the binding , we found that dpn expression was de-repressed throughout the wing-disc in clones of cells lacking Hairless ( Fig 2C and 2D ) . As with Kc cells , a significant proportion ( 70% ) of the Hairless associated regions in wing discs did not overlap with the sites bound by Su ( H ) . These Hairless-only regions were not enriched for the Su ( H ) motif ( rank = 145 p = 0 . 045 ) , suggesting that other factors are involved in Hairless recruitment . Indeed , a Hairless-GFP mutated in the Su ( H ) binding domain appears to still be recruited to chromatin , since bands indicative of binding to salivary gland polytene chromosomes are detected in live imaging experiments ( S4 Fig ) . While one other partner for Hairless has been identified in embryos ( Runt [39] ) , this is not expressed in wing discs . We examined the Hairless-only regions for binding motif enrichments and identified matches to Aef1 ( p = 1 . 21e-24 ) and Grainyhead ( p = 4 . 04e-10 ) motifs , which are both widely expressed in the wing disc and are proposed to act as repressors in at least some contexts , making them plausible Hairless partners [40–42] . Two of the best-characterized Notch regulated genes in the wing disc are cut and wingless [43–46] . Both are normally expressed at high levels at the dorsoventral ( D/V ) boundary . However , they can be induced by ectopic Notch activity at other locations in the wing disc [45] . Strikingly , neither of these genes exhibited detectable binding of Hairless or Su ( H ) at their characterized wing-disc enhancers ( Fig 3A and 3B ) , suggesting that these factors are not required to keep cut or wg silent at most positions in the disc . In agreement with this , and unlike the case of dpn , neither cut nor wg were generally de-repressed in Hairless mutant cells ( Fig 3C and 3D ) . De-repression of these genes was only detected when clones were located close to the D/V boundary , the site where these genes are normally responsive ( Fig 3C and 3D ) . At those positions , the loss of Hairless led to slight de-repression in some boundary-flanking cells ( Fig 3C and 3D ) . Thus Hairless is likely to be recruited to these genes only in a limited subset of cells , where it is important to limit their expression , but is not required to repress their enhancers more broadly in the wing disc . Both wg and cut could be induced much more widely when ectopic Notch was provided ( Fig 3E and 3F; [45] , even though Hairless and Su ( H ) do not appear to occupy those loci in most cells . This suggests that stable binding of Su ( H ) /Hairless is not a pre-requisite to render the genes responsive . Indeed , in discs over-expressing NICD , both cut and wg enhancers showed significant enrichment for Su ( H ) binding ( Fig 3A and 3B ) , indicating Su ( H ) can be recruited to those positions . Furthermore , reporters containing these enhancers exhibited the ability to respond widely to ectopic NICD ( e . g . Fig 3F ) . Notably , the cut and wg enhancers are located in regions enriched for the H3K4me1 chromatin mark ( Fig 3A and 3B ) , indicating that the chromatin is “primed” . Other Notch-regulated genes with similar characteristics included scalloped and vestigial , which are also enriched in H3K4me1 despite the lack of broadly detectable Su ( H ) occupancy . SMRTER is a co-repressor that , similar to Hairless , has been suggested to contribute to repression of Su ( H ) /Notch regulated targets [30 , 31] . Since several loci did not exhibit Hairless binding , we reasoned that at least some of these might instead be dependent on SMRTER regulation . We therefore determined the genome-wide profile of SMRTER occupancy in wing discs , using an in-frame YFP fusion generated by a protein-trap transposon insertion [47] . We detected 1165 enriched regions ( 1% FDR; S5 Table ) , of which 16% overlapped with Su ( H ) bound regions ( Fig 4A and 4B ) . This was similar to the number of Su ( H ) bound regions co-occupied by Hairless and included several genes potentially regulated by Notch , such as thread/Diap1 , E2f and anterior open ( aop ) . It is thus plausible that SMRTER could contribute to repression at Su ( H ) bound enhancers . However , no significant binding was detected at dpn , indicating that SMRTER is unlikely to be an obligate partner of Su ( H ) /Hairless , nor was binding detected at cut or wg enhancers . Proteins that occupy the DNA together in a complex might be expected to exhibit similar enrichment profiles . Pairwise comparisons were therefore made of the top 100 bound regions , ranked by fold enrichment , for Su ( H ) , SMRTER and Hairless . Only the profiles for Su ( H ) and Hairless exhibited a significant correlation ( Fig 4D ) , no correlation was seen for Su ( H ) and SMRTER profiles ( Fig 4C ) . However , the majority of Su ( H ) and SMRTER co-bound regions were also occupied by Hairless ( Fig 4B ) . This “triple’ state was more frequent than would be expected by chance ( p = 0 . 0004 ) , suggesting that it is of significance and that these regions may be bound by more than one type of co-repressor complex . One region that was enriched for SMRTER , Su ( H ) and Hairless binding was within an intron of thread/Diap1 ( Fig 5A ) . This region exhibited enhancer activity , conferring a similar expression pattern to the endogenous gene when placed upstream of a GFP reporter ( th1 . 2-GFP; Fig 5B; [37] ) . We therefore examined the effects on the expression of this reporter when depleting Hairless and SMRTER , by targeting RNAi to the posterior part of the wing-disc ( using engrailed-Gal4 Gal80ts ) . Knock down of either of the repressors resulted in de-repression of th1 . 2-GFP ( Fig 5B and 5C ) . In the case of SMRTER depletion , the resulting reporter pattern was still highly modulated whereas with Hairless depletion the expression was more uniform ( Fig 5B ) . When both SMRTER and Hairless were knocked-down together we observed an additive effect on th1 . 2-GFP , although the overall expression levels were not significantly elevated in the double combination ( Fig 5B and 5C ) . Similar effects were seen with enhancers from the reaper and cut genes , which were de-repressed when Hairless or SMRTER were depleted ( S5 Fig ) . To investigate whether the functions of Hairless and/or SMRTER are mediated through Su ( H ) , the Su ( H ) binding motif in the th1 . 2 enhancer was mutated ( th1 . 2[mut] ) and its expression analysed in similar knock-down experiments . Under wild type conditions , low levels of residual expression were observed with the mutated enhancer , which were little changed by the knock-down of Hairless ( Fig 5C and 5D ) . In contrast , SMRTER knock-down resulted in a significant de-repression ( Fig 5C and 5D ) . This implies that SMRTER contributes to the repression of the th1 . 2 enhancer , but that it does so independent of Su ( H ) . Since SMRTER can exert repression at an enhancer that lacks a Su ( H ) binding motif , it is likely that it is recruited to chromatin by another DNA binding protein . In this respect , SMRTER has been proposed to act as a co-repressor for the Ecdysone receptor ( EcR; [29 , 30] ) and we found that SMRTER binding peaks were significantly enriched for one of the EcR motifs ( EcR::USP , p = 5 . 2e-7 ) . We therefore compared the profile of SMRTER binding to published EcR binding data from a similar stage [48] . 38% of SMRTER peaks overlapped with EcR bound regions and there was a significant correlation between the SMRTER and EcR peaks when the top 100 peaks ( ranked by enrichment ) were compared ( cor = 0 . 19; p = 1 . 5e-05 ) . Target genes that were highly enriched for SMRTER and EcR binding included Hr39 and Blimp-1 ( Fig 6A and 6B ) . We examined whether EcR was involved in the regulation of the th1 . 2-GFP reporter by analysing the effects of EcR depletion , which should mimic those observed with SMRTER depletion [29 , 49] , and of expressing the EcR co-activator , Taiman [50] . As predicted , EcR depletion resulted in a mild increase in th1 . 2GFP expression , similar to SMRTER knock-down ( Fig 6C ) , while Taiman expression resulted in high levels of uniform th1 . 2GFP expression , as did expression of NICD ( Fig 6D and 6E ) . The up-regulation induced by Taiman was independent of the presence of the Su ( H ) motif , as a similar response was detected with the mutated th1 . 2[mut] enhancer , while the response to NICD was abolished ( Fig 6E and 6F ) . These data therefore support a model where SMRTER is recruited to the th1 . 2 enhancer through EcR and suggest that the co-occurrence of SMRTER , Hairless and Su ( H ) is likely indicative of enhancers that are co-regulated by Ecdysone and Notch .
In prevailing models for Notch mediated regulation of target genes , target loci are bound by a CSL-co-repressor complex , from which the co-repressors are displaced by NICD to activate transcription [8 , 10] . Our data support the notion that the co-repressor Hairless is bound with Su ( H ) at several Notch regulated enhancers , since their genome-wide binding profiles in both Kc cells and wing discs exhibited considerable overlaps . Furthermore , depletion of Hairless in Kc cells resulted in transcriptional de-repression and an increase in histone acetylation at some loci . Likewise , the absence of Hairless led to de-repression of target-genes such as deadpan and thread/Diap1 throughout the wing disc . Our results , however , challenge the view that pre-binding of the Su ( H ) /Hairless repressor complex is required for genes to be able to respond to NICD . Several well-known Notch regulated genes , such as cut , wg and vg did not generally display Hairless ( or Su ( H ) ) binding in the wing disc despite the fact that they are strongly induced throughout the disc when ectopic NICD is provided . This implies that their silencing is not generally dependant on the Hairless/Su ( H ) repressor complex . In agreement with this idea , no de-repression of cut or wg occurred in most cells lacking Hairless and only a very modest de-repression was observed at the dorsoventral boundary where these genes are normally expressed . It is therefore likely that Hairless is recruited in only a small population of cells where it may help to dampen or refine the response of target enhancers . This is borne out by the limited number of genes that are de-repressed by knock-down of Hairless in Kc cells . Since relatively few Notch regulated genes were bound by Hairless , this raised the possibility that an alternative co-repressor might be recruited at other loci . For example , one of the CSL partners in mammalian cells is SMRT [27] , whose homologue SMRTER was found to co-IP with Su ( H ) in flies [31] , making it a plausible candidate [30] . Analysis of genome-wide SMRTER binding indicated that a significant fraction of Su ( H ) bound regions were also occupied by SMRTER , although the binding intensities were not well correlated . However , many of the co-bound sites were also regions enriched for Hairless , including thread/diap1 , which was robustly enriched for all 3 proteins . Further analysis of thread regulation demonstrated that both Hairless and SMRTER are important for suppressing its expression , albeit in slightly different ways . When the Su ( H ) motifs in the thread enhancer were mutated , only the Hairless regulation was lost while that of SMRTER remained . These data , along with the strong correlation between SMRTER and Ecdysone Receptor binding , imply that the co-repressors are recruited to the same enhancer via different DNA-binding partners: Hairless via Su ( H ) and SMRTER via EcR . Crosstalk between Notch and EcR also occurs at the cut enhancer , although this involves Broad rather than EcR [51] . Thus we conclude that much of the co-localization of SMRTER and Su ( H ) is likely indicative of genes that are regulated by inputs from the EcR and Notch pathways , although we cannot fully rule out the possibility that SMRTER could act via Su ( H ) in some circumstances . A striking feature of the Hairless bound regions is that they correlate with regions of active chromatin , despite Hairless involvement in repression . Similar features have been noted for Groucho [6] , a key partner of Hairless , and for another class of repressors/co-repressors [5] . Likewise , Dorsal repressed enhancers exhibited some active chromatin marks ( H3K4me1 ) although these were found to be hypoacetylated compared to their active counterparts [52 , 53] . It is proposed these repressive factors contribute to polymerase pausing [6 , 52] rather than more long-term silencing [53] , so that gene transcription would be kept in check until/unless conditions change and RNA polymerase is released . Another suggestion is that modulation by repressors enables a fine regulatory control over transcriptional output from a given target gene so that it can be graded over range of values . Indeed , the mammalian CSL co-repressor SHARP has been suggested to control a permissive chromatin state at Notch target genes by concomitantly promoting H3K27 deacetylation and H3K4 methylation ( the former would dampen activity while the latter would enhance it; [54] ) . The modest changes in acetylated histones that occurred at some highly regulated loci in Hairless depleted cells would fit with these models , as would the varied changes in gene expression that occur in the Kc cells . Thus we propose that Hairless , like SHARP and other similar factors , is likely to interact with a range of partners through which it will modulate , rather than silence , the response of a target enhancer to the levels of Notch activity , most likely via local effects on chromatin .
To generate a genomic Hairless construct , AttB plasmids containing the genomic region 3R: 20621141–20628985 ( using primers 5’-GCATTCGTCTCAATAACTAACGTCG 5’-CGCAATAAAAAGACACCTGCAACC ) were constructed , with the coding sequences of eGFP ( enhanced Green Fluorescent Protein ) and Dam methylase [33] inserted in-frame before the stop codon located at exon 4 , to generate a protein fusion at the C-terminus . This plasmid was used in transfection experiments ( see below ) and subsequently injected into strains containing an AttP40 site to generate transgenic Hairless-GFP flies . The functionality was assessed by its ability to rescue H[P8]/H[1] flies; the viability of the mutant flies was rescued ( although they were not fully fertile ) indicating that the plasmid confers functional Hairless activity . For the genomic GFP-Su ( H ) , an AttB plasmid containing the genomic region 2L 15038840–15045039 ( using primers 5’ -CAAGTTAGATATGGCAATGCACCG 5’-ACTGCATATCTGTACTGATGACG ) was constructed , with the coding sequences of eGFP inserted in frame at the start of exon 1 , to generate a protein fusion at the N-terminus . This plasmid was used in transfection experiments ( see below ) . Kc cells were cultured at 25 OC in Shields and Sang M3 insect medium ( Sigma , S3652 ) , supplemented with 5% FBS ( Sigma , F9665 ) , 1g/L yeast extract ( Oxoid , LP0021 ) , 2 . 5g/L bacto-peptone ( BD Biosciences , 211677 ) and 1x Antibiotic-Antimycotic ( Gibco , 15240–062 ) . For Hairless RNAi , Kc cells were transfected with 20μg dsRNA in Fugene 6 ( Promega , E2691 ) in 10 cm plates according to the standard protocol and then incubated for 72h . RNA isolation was performed using Qiagen Rneasy Midi kit ( cat N 75142 ) . To 50μg total RNA in 28μl DEPC water 1μl of 500 ng/μl oligo ( dT ) 23 anchored primer ( Sigma ) was added and incubated at 65°C for 10 minutes , then placed on ice . 8μl of 5x first strand buffer ( Invitrogen ) , 2μl of low-C dNTP mix ( 5mM dATP , dGTP , dTTP , 2mM dCTP ) , 2μl of 1mM Cy3 or Cy5 dCTP ( GE Healthcare ) , 2μl of 0 . 1M DTT ( Invitrogen ) , 0 . 5μl of RNAsin ( Promega ) and 2μl of Superscript III reverse transcriptase ( Invitrogen ) were added and incubated at 46°C for 2 hours . The reaction was stopped with 20μl mix of 0 . 5M EDTA and 1M NaOH and incubated at 65°C for 15 minutes . Sample was neutralised with 25μl of 1M Tris-HCl ( pH 7 . 5 ) . Combined Cy3- and Cy5-labelled probe ( sample and control ) were purified with AutoSeq G-50 column ( GE Healthcare ) and the volume reduced to between 2-5μl in a speed vac with medium heat . Then 2μl of 10mg/ml sonicated salmon sperm DNA ( Invitrogen ) and 140μl of Ocimum hybridisation buffer were added to the labelled mixture and boiled at 100°C for 2min . 140μl of the labelled sample was loaded to 70-mer long oligo microarrays ( FL003 , Flychip ) , and hybridised for 16 hours at 51°C . Post-hybridisation washes were performed as per PowerMatrix slides protocol ( FMB ) . Arrays were scanned at 5μm resolution with a GenePix 4000B ( Axon ) dual laser scanner . Images were processed and spot quantified by the Dapple software ( http://www . cs . wustl . edu/~jbuhler/research/dapple/ ) . Raw data were then loaded into limma ( Bioconductor; [55] ) and normalised with the vsn package ( Bioconductor; [56] ) . The resulting ratios are log2 ratio of sample/control . Four biological replicates were analysed , data are available from GEO as part of super series GSE97603 . Genes with altered expression were assigned to a ChIP peak ( s ) if the peak was within 10kb ( upstream or downstream ) . For Chromatin immunopreciptation Kc cells were transfected with Hairless-GFP or GFP-Su ( H ) plasmids in combination with a plasmid containing puromicyn resistance gene ( pMT-Puro ) using Fugene 6 and grown under puromycin selection ( 2μg/ml ) . Chromatin preparation and ChIP were essentially as described previously [34] [57] , except the cells were fixed in 1% formaldehyde , 1mM EGS for 15min . A library was generated from the immunoprecipitated DNA using a complete whole genome amplification ( WGA ) Kit ( GenomePlex #WGA2-50RXN ) . The amplification reaction was supplemented by adding 0 . 75ul of 10mM dUTP . The amplification was performed for 14 cycles . After purification with Qiagen PCR cleanup columns , 7 . 5μg of purified DNA was hybridized to GeneChip Drosophila Tiling 2 . 0R Array according to Affymetrix specifications . Three independent replicates were performed for each construct . For DamID experiments , KC cells were transfected with Hairless–Dam using Fugene 6 and incubated for 72 hours before harvesting . Genomic DNA was extracted using Qiagen DNeasy Blood and Tissue Kit ( #69504 or #69506 ) , following manufacturer's instructions . 2 . 5μg of genomic DNA were digested in 10μl with DpnI overnight at 37°C . DpnI was inactivated at 80°C for 20min . Genomic DNA fragments were ligated to Phosphorylated AdR adapters ( AdRt 5' CTAATACGACTCACTATAGGGCAGCGTGGTCGCGGCCGAGGA , aligned to AdRb 5' TCCTCGGCCG ) in 20μl reaction for 2 hours at 22°C using FERMENTAS ( cat# EL0011 ) , supplemented with PEG-4000 . The reaction was inactivated at 65°C for 10 min and subsequently digested with DpnII , the resulting DNA fragments were cleaned using Qiagen PCR cleaning kit ( Cat#28104 ) . DNA was amplified for 15 cycles by PCR using Expand DNA polymerase ( Cat#11732641001 ) supplemented with 0 . 1mM dUTP and Adr-PCR primer ( 5' GGTCGCGGCCGAGGATC ) . After amplification , adapters were removed by DpnII digestion and cleaned using Qiagen PCR cleaning columns . Samples were then fragmented and hybridized to GeneChip Drosophila Tiling 2 . 0R Array according to Affymetrix specifications . Three replicates of the Hairless-Dam and a Dam-only control were performed . Dissected heads from 30 third instar larvae ( Hairless-GFP or SMRTER-YFP ) were fixed with 1ml of 4% Formaldehyde +1 . 5mM of EGS ( 25mins at room temperature ) and the reaction quenched with 200μl of 2M-glycine solution . After rinsing 3x ( 1 x PBS + Roche complete protease inhibitor cocktail ) the wing discs were removed and transferred to 50μl of nuclear lysis buffer ( 50 mM Tris-HCl pH 8; 10 mM EDTA; 1% SDS; 1x Roche complete protease inhibitor ) for 10 min on ice before homogenising the wing discs . 250μl of IPDB ( ImmunoPrecipitation Dilution Buffer ) was then added to the mixture and the sample was sonicated for 7 min , with a 30 second “ON/OFF” cycle , in a pre-cooled Bioruptor ( Diagenode UCD-200 ) . After sonication , the sample was centrifuged at 13200rpm at 4°C for 10 min , and the chromatin containing supernatant then diluted with 200μl of IPDB . ChIP was performed as previously described [37 , 57] , using polyclonal rabbit anti-GFP ( Abcam ab290 ) . Three ChIP samples were then pooled for library preparation using the GenomePlex Complete WGA kit . 14 rounds of amplification were performed and the DNA purified using Qiagen PCR cleanup columns . 250μg DNA was then subjected to a further 10 rounds of amplification in the presence of 0 . 75μl of 10mM dUTP . After purification with Qiagen PCR cleanup columns approximately 7 . 5μg of DNA was hybridised to Affymetrix GeneChip Drosophila Tiling 2 . 0R Arrays according to Affymetrix specifications . Three ( Hairless ) or two ( SMRTER ) independent replicate experiments were performed . Window smoothing and peak calling were performed using the Bioconductor package Ringo [58] with a winHalfSize of 700bp and min . probes = 10 . Probe levels were then assigned p-values based on the normalNull method , corrected for multiple testing using the Hochberg-Benjamini algorithm and then condensed into regions using distCutOff of 200bp . All data are available from GEO , super-series accession number GSE97603 . Dataset overlaps were analysed using GenomicRanges ( Bioconductor R; [59] ) and for overlap with chromatin signatures , the profile for Kc cells was used [34] ( https://github . com/rstojnic/notch-chromatin ) . Motif enrichment analysis was performed using the Bioconductor package PWMEnrich [60] , which assesses the enrichment of each motif from a library of 650 experimentally derived DNA motifs for Drosophila transcription factors . Accessible chromatin in BG3 and Kc167 cells was used as a background for calculating P-values of enrichment . The correlation between ChIP signals was calculated on enhancers that are a union of all ChIP peaks from the considered datasets . The average signal over the whole enhancer was then plotted for pairs of datasets and the Pearson correlation and corresponding P-value calculated . Analysis of K56 acetylation changes in Kc cells following Hairless RNAi was carried out using ChIP-chip with the conditions and methods described previously [34] . For controls , cells were treated with GFP RNAi . To identify regions of differential K56 acetylation an algorithm was used to compare the variance between replicates to the variance between samples within sliding 2kb regions [34] . This method identifies regions where the differences cannot be explained by noise , using two replicates for each sample ( e . g . con1 , con2 to HRNAi1 , HRNAi2 where 1 , 2 indicate replicates ) . The noise distribution was used to convert differences between con and HRNAi within individual 2kb regions into a P-value , which was then corrected for multiple testing using the Hochberg-Benjamini algorithm . The resulting P-value represents the False Discovery Rate ( FDR ) . An FDR threshold of 1% was used to define regions of significant difference . The significance calculation of combinatorial overlaps strongly depends on the number of unoccupied enhancers [61] , and the binding model used . We used a null model in which Su ( H ) -H binding is dependent on each other ( as estimated from available binding data ) , but SMRTER binding is independent of both . We assumed there is between 0 and 5000 additional sites in the genome that are unoccupied but that could in principle be bound . We inferred the number of unoccupied enhancers by maximising the likelihood of the fit . Finally , we calculated the P-value of the difference between the observed and expected individual patterns using Fisher’s exact test . We found that 70% more enhancers are bound by all three factors ( Su ( H ) , H , SMRTER ) than expected by our best-fitting model ( P-value 0 . 0004 ) . All alleles and stocks are described in FlyBase ( www . flybase . org ) unless otherwise indicated . The following mutant and reporter lines were used: H[P8] , H[P1] SMARTER-YFP[CPTI001385]; and th1 . 2-GFP [37]; cut1 . 3-GFP and wg2 . 2-GFP . In RNAi and overexpression experiments , the Gal4 driver stock engrailed:Gal4 Tub:Gal80ts was combined with UAS lines and larvae were shifted to 30°C 48 hours after egg laying . The following RNAi and overexpression lines were used: UAS-w RNAi ( TRiPGL00094 ) , UAS-Hairless RNAi ( TRiPJF02624 ) , UAS-SMRTER RNAi ( KK1026110 ) , UAS-Notch-intra[79 . 2]; UAS-Taiman [50]; UAS-EcR-RNAi; UAS-GFP-RNAi ( BL-9330 ) ; UAS-lacZ . Note that UAS-wRNAi , UAS-GFP RNAi or UAS-lacZ were included where appropriate to ensure the number of UAS sequences was kept consistent in each experiment . To generate MARCM clones [62] lacking Hairless , H[P8] FRT82B/TM6B were crossed to hs-FLP tubGal4 UAS-GFP; FRT82B tubGal80 . Progeny were heat shocked at 37°C for 1 hour , 72 hours after egg-laying then kept at 30°C until dissection . Fixation and immunostaining conditions were as previously described . Antibodies used were: rat anti-Ci ( DSHB 2A1; 1/25 ) , mouse anti-Cut ( DSHB 2B10; 1/20 ) , mouse anti-Wg ( DSHB 4D4; 1/20 ) , guinea-pig anti-Dpn ( gift from Christos Delidakis; 1/50 ) , rabbit anti-GFP ( Invitrogen; 1/2000 ) . Alexa conjugated secondary antibodies were from ThermoFisher . Fluorescence quantifications were performed using ImageJ64 . Identical laser and confocal settings were used throughout a single experiment , and in all cases measurements were from at least two independent experiments . An experiment-specific-ROI ( Region Of Interest ) was pre-defined using ImageJ64 , and , for a given experiment , consistent ROI settings were used across genotypes and they were placed at similar positions . Mean grey value measurements for the ROI within the manipulated posterior region were normalised to measurements for the ROI in the wild type anterior compartment for each wing disc . Individual data points are the relative values for each wing disc analysed , and the mean value was calculated for comparison between genotypes . The significance of differences in the expression ratios was assessed using a T-test; p<0 . 05 is indicated by single asterisk , p<0 . 01 is indicated by double asterisk and p<0 . 001 is indicated by triple asterisk . | The communication between cells that occurs during development , as well as in disease contexts , involves a small number of signalling pathways of which the Notch pathway is one . One outstanding question is how these pathways can bring about different gene responses in different contexts . As gene expression is co-ordinated by a mixture of activators and repressors , we set out to investigate whether the distribution of repressors across the genome is important in shaping whether genes are able to respond to Notch activity . Our results from analyzing the binding profile of two repressors , Hairless and SMRTER , show that , in many cases , they are not essential for preventing a gene from responding . Instead they are deployed at a limited number of genetic loci where they gate the response , helping to set a threshold for gene activation . Perturbations to their function lead to enhanced gene expression in limited territories rather than to new programmes of gene expression . Their main role therefore is to restrict the time or levels of signal that a gene needs to receive before it will respond . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"cell",
"binding",
"infographics",
"cell",
"physiology",
"rna",
"interference",
"invertebrate",
"genomics",
"notch",
"signaling",
"epigenetics",
"chromatin",
"computer",
"and",
"information",
"sciences",
"chromosome",
"biology",
"genetic",
"interference",
"gene",
"express... | 2017 | Role of co-repressor genomic landscapes in shaping the Notch response |
Ribonucleotide reductases ( RRs ) are evolutionarily-conserved enzymes that catalyze the rate-limiting step during dNTP synthesis in mammals . RR consists of both large ( R1 ) and small ( R2 ) subunits , which are both required for catalysis by the R12R22 heterotetrameric complex . Poxviruses also encode RR proteins , but while the Orthopoxviruses infecting humans [e . g . vaccinia ( VACV ) , variola , cowpox , and monkeypox viruses] encode both R1 and R2 subunits , the vast majority of Chordopoxviruses encode only R2 subunits . Using plaque morphology , growth curve , and mouse model studies , we investigated the requirement of VACV R1 ( I4 ) and R2 ( F4 ) subunits for replication and pathogenesis using a panel of mutant viruses in which one or more viral RR genes had been inactivated . Surprisingly , VACV F4 , but not I4 , was required for efficient replication in culture and virulence in mice . The growth defects of VACV strains lacking F4 could be complemented by genes encoding other Chordopoxvirus R2 subunits , suggesting conservation of function between poxvirus R2 proteins . Expression of F4 proteins encoding a point mutation predicted to inactivate RR activity but still allow for interaction with R1 subunits , caused a dominant negative phenotype in growth experiments in the presence or absence of I4 . Co-immunoprecipitation studies showed that F4 ( as well as other Chordopoxvirus R2 subunits ) form hybrid complexes with cellular R1 subunits . Mutant F4 proteins that are unable to interact with host R1 subunits failed to rescue the replication defect of strains lacking F4 , suggesting that F4-host R1 complex formation is critical for VACV replication . Our results suggest that poxvirus R2 subunits form functional complexes with host R1 subunits to provide sufficient dNTPs for viral replication . Our results also suggest that R2-deficient poxviruses may be selective oncolytic agents and our bioinformatic analyses provide insights into how poxvirus nucleotide metabolism proteins may have influenced the base composition of these pathogens .
Critical for the replication of all organisms and DNA viruses is the conversion of ribonucleotides to deoxynucleotides to serve as building blocks for genome synthesis and repair . Ribonucleotide reductase ( RR ) is a key enzyme involved in this process , catalyzing the reduction of rNDPs to dNDPs [1] , [2] . RRs can be grouped into one of three classes , based on their requirement for oxygen and the mechanism by which a catalytically-important thiyl radical is generated [1] . Mammals typically encode class I RR proteins while class II and III proteins are found only in microorganisms [1] , [3] . Class I RR enzymes are assembled from both large ( R1; 80–100 kDa ) and small ( R2; 37–44 kDa ) protein subunits , which associate to form enzymatically-active R12R22 tetrameric complexes [1] . These complexes require oxygen to generate a tyrosyl radical found within R2 subunits [1] , [4] , which is ultimately transferred to R1 subunits to generate a thiyl radical used in rNDP reduction . Transfer of the tyrosyl radical from R2 to R1 subunits is thought to occur through a “radical transfer pathway” that uses a series of at least eleven highly-conserved amino acid residues to promote long-range electron transfer [4] , [5] , [6] , [7] , [8] . Mutant proteins containing amino acid substitutions at either the tyrosine involved in radical formation [9] or any of the proposed transfer pathway residues [4] , [6] , [8] , [10] , [11] form inactive RR complexes , indicating that both radical formation and transfer are required for catalysis . Mammalian cells encode a single R1 gene that is only transcribed during S-phase [12] . However , due to the long half-life ( ∼15 h ) of R1 proteins , R1 levels remain essentially constant throughout the cell cycle [13] . The primary small subunit , R2 , is also only expressed during S-phase [12] , [14] however , this protein has a short half-life ( ∼3 h ) and is rate-limiting for R1-R2 complex formation [13] . The short half-life of R2 is due to its polyubiquitination by the anaphase-promoting complex ( APC ) -Cdh1 ubiquitin ligase , which leads to its degradation during mitosis [15] . This degradation is dependent upon APC-Cdh1 recognition of a “KEN” box sequence in the N-terminus of R2 ( Figure 1 ) . Mammals also encode a second small subunit , p53R2 , so named because its elevated expression in response to DNA damage is dependent upon the tumor suppressor p53 [16] . Although p53R2 is 80–90% identical to cellular R2 and can form active complexes with R1 [17] , it lacks ∼33 N-terminal amino acid residues found in R2 , including those containing the KEN box ( Figure 1 ) [15] . The absence of the KEN box sequence likely explains why p53R2 levels are relatively constant throughout the cell cycle in the absence of DNA damage [18] . It has been hypothesized that p53R2 plays some role in supporting mitochondrial DNA synthesis and/or DNA repair outside of S-phase [17] , [18] , [19] , [20] . Therefore , despite their similarity , R2 and p53R2 appear to be differentially regulated and probably serve different purposes during the cell cycle . Many Chordopox- , herpes- , asfra- and iridoviruses , also encode their own class I RR proteins [21] , [22] , [23] , [24] , [25] . These enzymes are generally thought to support viral replication since ribonucleotide reduction is normally the rate-limiting step in de novo dNTP biogenesis [26] . Although most Chordopoxviruses encode RR proteins , many only encode one of the two RR subunits with a clear bias towards the conservation of R2 proteins ( Table S1 ) . Only the Suipox- and Orthopoxviruses contain both R1 and R2 genes . The latter group contains viruses of medical importance including variola virus , the causative agent of smallpox , as well as monkeypox and cowpox viruses , which are responsible for zoonoses in humans [27] , [28] . Most understanding of poxvirus RR proteins comes from studies with another Orthopoxvirus , vaccinia virus ( VACV ) . Slabaugh and Mathews [29] were the first to show that RR activity increased in VACV-infected cells and subsequent studies identified the I4L [25] , [30] and F4L [24] genes as those encoding the 87 kDa I4 ( R1 ) and 37 kDa F4 ( R2 ) proteins , respectively . Biochemical studies showed that VACV and cellular RR enzymes share many features , including a similar tertiary architecture , similar pH dependence , allosteric modulation of activity by nucleotides , and comparable specific activities on most rNDP substrates [31] , [32] , [33] . However , unlike cellular RR , the viral enzyme is less sensitive to allosteric modulation and shows little activity on UDP substrates , indicating that viral and cellular RR enzymes also differ in important ways [33] . The similarities between VACV and mammalian RR are not unexpected given that VACV ( and other poxvirus ) RR subunits typically share >70% sequence identity with their mouse and human homologs ( Figure 1 ) . Previous studies have shown that inactivating VACV I4L does not affect plaque size and that I4L-deficient mutant strains replicate their DNA and produce viral particles to levels comparable to wild-type VACV in cell culture [34] , [35] . Due to these observations , the I4L locus has been suggested to be an excellent site for insertion of foreign genes into VACV [36] . Furthermore , I4L mutants are only mildly attenuated in their virulence in mouse models , exhibiting an ∼10-fold increase in lethal dose 50 ( LD50 ) values when compared to wild-type virus [34] . Paradoxically , another group reported that targeted inactivation of F4L attenuated VACV in mice by ∼1000-fold compared to wild-type virus [37] , but the reason for this attenuation was unknown . Although these were separate studies using different strains of VACV , they suggested that the F4 subunit is more important for virus replication and pathogenesis than I4 , despite the fact that both subunits are needed for RR activity [32] . We initiated our studies of VACV RR because VACV recombination appears to be catalyzed by poxviral DNA polymerases in vivo [38] , and we wanted to determine if perturbing dNTP pools would affect this process . However , it soon became apparent that some of the mutant strains we generated exhibited previously uncharacterized replication defects . This prompted us to revisit how VACV RR affects viral replication and pathogenesis . To do this , we generated a panel of mutant strains containing mutations in the VACV RR genes . We also generated strains lacking a functional J2R ( thymidine kinase or TK ) gene , thus unable to access the parallel viral salvage pathway of dTTP biogenesis [39] . Our studies show that both the VACV R1 and R2 proteins can form a diversity of virus-virus and virus-host protein-protein interactions in vivo , but that the VACV R2 subunit is far more critical for VACV replication and pathogenicity than is the R1 subunit . Our model suggests that poxvirus R2 subunits form active complexes with host R1 proteins in order to ensure a sufficient dNTP supply to support viral replication . This model is substantiated by previous biochemical studies that found a chimeric RR enzyme consisting of VACV F4 and mouse R1 ( MR1 ) to be more active than strictly viral or mouse RR complexes [33] . Our studies also provide insights into why poxviruses have often conserved their R2 but not their R1 genes ( Table S1 ) . To our knowledge this is the first report of a chimeric , virus-host RR forming in vivo . Our study provides further evidence that poxviruses recruit cellular enzymes , in addition to those previously identified such as topoisomerase II [40] and DNA ligase I [41] , to support viral replication . Our bioinformatic analysis of other large DNA viruses suggests that recruitment of host RR subunits may represent a more widespread viral strategy to parasitize host nucleotide biosynthetic machinery .
A series of mutant strains were generated in which one ( ΔI4L; ΔF4L ) or both ( ΔI4L/ΔF4L ) of the VACV RR genes were deleted from the viral genome ( Figure 2A ) . We also constructed VACV encoding an insertional inactivation of the J2R ( TK ) gene in combination with ΔI4L and/or ΔF4L mutations generating ΔI4L/ΔF4L/ΔJ2R and ΔF4L/ΔJ2R strains . These strains provide insights into the relative biological importance of the de novo ( RR-dependent ) and salvage ( TK-dependent ) pathways in VACV replication . We also constructed VACV ΔF4L strains encoding a His6-tagged F4L gene , or a His6-tagged F4L gene encoding a Y300F amino acid substitution , inserted into the J2R locus . These viruses are referred to as VACV strains ΔF4L/ΔJ2RHisF4L and ΔF4L/ΔJ2RHisY300FF4L , respectively . The marker rescue strategies used to generate these mutant strains are depicted in Figure 2A . PCR-based analysis confirmed the deletion or inactivation of the targeted loci in constructed VACV strains ( data not shown ) . Western blotting confirmed the presence or absence of viral RR subunit expression in each of the isolates ( Figure 2B ) . The ΔF4L/ΔJ2RHisF4L strain appeared to express elevated levels of F4 compared to wild-type virus , whereas the ΔF4L/ΔJ2RHisY300FF4L strain had slightly reduced F4 expression ( Figure 2B ) . The former case is likely a result of the F4L gene being under the control of an early/late promoter present on the pSC66 transfer vector whereas the endogenous F4L promoter is activated only at early times during infection [42] . The lower F4 expression of the ΔF4L/ΔJ2RHisY300FF4L strain is likely due to its poor replication in culture ( see below ) . These and other VACV strains are summarized in Table 1 . See Text S1 for details of virus construction . Plaque size and morphologies of the generated strains were analyzed on BSC-40 cells as an initial step to characterize their growth properties . The wild-type and ΔI4L strains exhibited similar plaque morphologies . These plaques typically had large central clearings and were accompanied by smaller secondary plaques that are formed when extracellular enveloped virus are released from infected cells and initiate new infections near primary plaque sites ( Figure 3A ) . Upon measurement of primary plaque areas , no significant differences were found between wild-type and the ΔI4L strain ( Figure 3B ) . In contrast , the ΔF4L , ΔF4L/ΔJ2R , and ΔI4L/ΔF4L/ΔJ2R strains all produced significantly smaller plaques ( P<0 . 05 ) that were only 55–60% of the plaque size exhibited by wild-type virus ( Figure 3B ) . In addition , the primary plaques produced by ΔF4L strains were typically devoid of nearby secondary plaques ( Figure 3A ) . Incorporation of a His6-tagged form of F4L into the TK locus appeared to complement ΔF4L strain replication as the ΔF4L/ΔJ2RHisF4L strain displayed plaques characteristic of wild-type virus in terms of size and the presence of secondary plaques ( Figure 3A and B ) . Strikingly , ΔF4L strains rescued with a His6-tagged F4L gene encoding the Y300F substitution produced plaques that were not only significantly smaller than wild-type virus [ ( P<0 . 05 ) ; Figure 3B] , but were only 35–40% the size of plaques produced by any of the strains with F4L deleted and these differences were statistically significant ( P<0 . 05 ) . These results suggested that deletion of F4L has a more detrimental effect on plaque size than deletion of I4L . They further suggested that re-introduction of a His6-tagged F4L gene into the TK locus can rescue the small plaque phenotype of ΔF4L strains . However , this rescue effect is lost and the ΔF4L strain replication defect is exacerbated when the re-introduced F4 protein encodes the Y300F amino acid substitution . Y300 represents a highly-conserved tyrosine residue found in essentially all mammalian small RR subunits ( Figure 1 ) . The homologous residue in mouse R2 ( MR2; Y370 ) is required for the transfer of radicals from MR2 to MR1 subunits , which is necessary for catalysis [4] . A Y370F substitution abolishes catalysis but does not impede physical interaction of MR1 and MR2 subunits [4] . The same substitution of the homologous tyrosine residue in human p53R2 ( Hp53R2 ) also abolishes RR activity of Human R1 ( HR1 ) -Hp53R2 complexes [43] . Therefore , the Y300F substitution in F4 is predicted to inhibit catalysis while still allow for R1-R2 subunit interaction . These predicted properties of the Y300F F4 protein may explain the dominant negative-like phenotype exhibited by the ΔF4L/ΔJ2RHisY300FF4L strain . We also examined the growth kinetics of these mutant strains in HeLa cells . As previously reported [34] , deleting I4L had little effect on viral replication , with the ΔI4L strain replicating to titers that were 2-fold lower than those produced by wild-type virus 48 h post-infection ( Figure 3C ) . In contrast , large ( >5-fold ) differences between wild-type and ΔF4L strains were readily apparent by 18 h post-infection . This trend continued to the end of the experiment , with the wild-type strain producing ∼15–50-fold more virus than ΔF4L strains 48 h post-infection . This growth defect could be complemented with a His6-tagged F4L gene but not if this gene encoded the Y300F substitution ( Figure 3D ) . In fact , the ΔF4L/ΔJ2RHisY300FF4L strain was unable to undergo productive replication in HeLa cells ( Figure 3D ) . These results suggested that deletion of the F4L gene impairs VACV replication to a higher degree than deletion of I4L , and that concomitant deletion of F4L and J2R does not impede replication further ( Figure 3C and 3D ) . Furthermore , the fact that one can rescue the ΔF4L growth defect with a His6-tagged form of F4L inserted at the TK locus , implies that the defect seen in a ΔF4L strain is not due to other possible idiosyncratic effects caused by deleting the F4L locus ( Figure 3D ) . Finally , these studies further illustrate the dominant negative effect on virus growth imposed by a catalytically-inactive , Y300F-substituted F4 protein . One explanation for the properties of virus encoding a Y300F-substituted F4 protein is that the mutant protein could be competing with cellular R2 proteins for binding to cellular and/or viral R1 subunits . However , the studies shown in Figure 3C and D suggested that deleting I4L does not result in significant replication defects and so the dominant negative phenotype was likely mediated by interaction with cellular R1 proteins . To rule out a role for I4 interaction in this dominant negative phenotype , the His6-tagged wild-type or Y300F-encoding F4L gene was inserted into the J2R locus of ΔI4L/ΔF4L strains . The ΔI4L/ΔF4L/ΔJ2RHisF4L strain produced plaques indistinguishable in size from those formed by wild-type virus ( P>0 . 05; Figure 3B ) . However , deleting I4L had no further effects on the plating properties of the ΔI4L/ΔF4L/ΔJ2RHisY300FF4L strain . This strain still produced plaques that were significantly smaller than those produced by wild-type ( P<0 . 05 ) or ΔF4L strains ( P<0 . 05 ) and were not significantly different from ΔF4L/ΔJ2RHisY300FF4L virus plaques ( P>0 . 05; Figure 3B ) . These observations implied that the plaque properties of ΔF4L strains are not influenced by the presence or absence of I4L . We also tested the ability of other , His6-tagged Chordopoxvirus or host R2 proteins to rescue the small plaque phenotype of the ΔF4L strain . The R2 genes encoded by ECTV , MYXV and SFV R2 genes were all able to rescue the small plaque phenotype , but interestingly the Hp53R2 gene failed to rescue this phenotype ( Figure 3B ) . These results implied that Chordopoxvirus R2 proteins have conserved a specific function and/or activity level that is not recapitulated by Hp53R2 . We hypothesized that the reduced replication of the ΔF4L strains was due to impaired genome replication . This is because RR plays a key role in dNTP biogenesis and our initial studies found that ΔF4L ( Figure S1A ) , but not ΔI4L strains ( Figure S1B ) , exhibited reduced late gene expression , which is a common consequence of defects in DNA replication . To test this hypothesis , BSC-40 cells were infected with wild-type or ΔF4L viruses and genome replication was measured in parallel with viral yields . The results of these experiments are shown in Figure 4 . As in HeLa cells , the ΔF4L strain exhibited impaired replication kinetics in BSC-40 cells , generating only 15% of the total titer observed with the wild-type strain at 24 h post-infection ( Figure 4A ) . This growth defect was associated with impaired DNA synthesis , with the ΔF4L strain exhibiting an ∼3 h delay in genome synthesis as well as an ∼5-fold reduction in DNA production at 24 h post-infection when compared to wild-type infections ( Figure 4B ) . We also tested what effect the drug hydroxyurea ( HU ) would have on these strains , since previous studies have correlated HU resistance with changes in F4 expression [44] . Addition of 0 . 5 mM HU to ΔF4L strain-infected cultures completely blocked virus DNA synthesis . In contrast , wild-type virus still produced detectable amounts of genomic DNA , albeit with delayed kinetics , at levels comparable to what is seen in cells infected with the ΔF4L strain in the absence of HU ( Figure 4B ) . These results suggested that the reduced yields observed with ΔF4L strains are at least partially due to impaired genome synthesis . Furthermore , because sensitivity to RR inhibitors is directly correlated to RR activity levels [45] , the hypersensitivity of the ΔF4L strain to HU suggests that these effects on DNA replication are caused by a reduction in RR activity . We hypothesized that the impaired genome replication of the ΔF4L strain was due to reduced dNTP pool sizes as a result of decreased RR activity . However , it is difficult to interpret the meaning of biochemical measurements of pool sizes because of uncertainties surrounding how dNTPs are distributed in infected cells . Instead , we tested whether VACV RR mutants exhibit an altered sensitivity to the antiviral drug cidofovir ( CDV ) . CDV is converted by cellular kinases to the diphosphoryl derivative ( CDVpp ) [46] which is competitive with respect to dCTP [47] and inhibits VACV E9 DNA polymerase activity [48] , [49] . Thus , CDV sensitivity can be used as an indirect probe for changes in dCTP pool sizes . Table 2 summarizes how RR mutations affect CDV sensitivity as assessed by plaque reduction assays and calculated 50% effective concentration ( EC50 ) values . Wild-type and ΔF4L/ΔJ2RHisF4L strains exhibited similar mean EC50 values of 42 . 0 and 41 . 2 µM , respectively . The ΔI4L strain was significantly more sensitive than the aforementioned strains ( P<0 . 05 ) having a mean EC50 value of 25 . 1 µM . However , loss of F4L ( or F4L and J2R ) resulted in greater hypersensitivities to CDV ( P<0 . 05 ) with EC50 values ∼5–7-fold lower than wild-type values . The ΔF4L/ΔJ2RHisY300FF4L virus was even more sensitive to CDV ( EC50 = 3 . 5 µM ) than either wild-type ( P<0 . 05 ) or ΔF4L ( P<0 . 05 ) strains . As noted previously [50] , [51] , inactivation of J2R did not further alter VACV sensitivity to CDV ( Table 2 ) . The trends in CDV sensitivity closely mirrored those found in measurements of HU sensitivity using a plaque reduction assay ( Table 2 ) . The order of resistance to HU ( from measurements of EC50 ) was wild-type ≥ΔF4L/ΔJ2RHisF4L>ΔI4L>ΔF4L>ΔF4L/ΔJ2RHisY300FF4L and seemed unaffected by the presence or absence of the J2R gene ( Table 2 ) . In order to determine if the hypersensitivities of ΔF4L and ΔF4L/ΔJ2RHisY300FF4L strains to CDV and HU were specific and not simply due to the reduced replicative abilities of these viruses , we performed a plaque reduction assay using phosphonoacetic acid ( PAA ) . PAA is a pyrophosphate analog and DNA polymerase inhibitor that is noncompetitive with dNTPs [52] . Therefore , the efficacy of PAA in inhibiting virus replication would not be expected to be dependent upon RR activity or dNTP pool sizes . Consistent with this , RR mutant VACV strains were not hypersensitive to PAA when compared to wild-type virus ( Table 2 ) . These mutant strains were also not hypersensitive to isatin-β-thiosemicarbazone ( IBT ) , which causes aberrant late viral mRNA biogenesis [53] ( data not shown ) . Collectively , these data all point to a deficiency in dNTP pools as being the cause of the ΔF4L strain growth deficiency ( Figure 3 ) and suggest that F4 , and not I4 , is the critical determinant of growth efficiency and drug sensitivity . Our data suggested that F4 may form functional complexes with host R1 proteins to support viral replication . This hypothesis was strengthened by previous studies that found purified mouse and VACV RR subunits to form functional chimeric RR complexes in vitro [33] . To determine if virus-host RR interactions could occur in vivo , co-immunoprecipitation experiments were performed with VACV-infected HeLa cell lysates using antibodies against HR1 , Human R2 ( HR2 ) or Hp53R2 RR subunits . F4 co-immunoprecipitated with each of the host RR subunits ( Figure 5A ) and the efficiency of “pull-down” was the same in extracts prepared from cells infected with wild-type and ΔI4L strains ( Figure 5B ) , suggesting that the presence of I4 does not significantly impede F4 interaction with host RR subunits . Interaction of F4 with cellular R2 subunits , while unexpected , may not be that surprising given that R2 subunits interact with one another in addition to interacting with homodimers of R1 [1] . We thought these interactions may be in part due to enhanced cellular RR subunit expression after infection . However , we were unable to observe induction of cellular RR expression by 24 h post-infection ( Figure S2 ) . To further confirm the immunoprecipitation results , VACV strains expressing either Flag-tagged HR1 ( ΔJ2RFlagHR1 ) or Flag-tagged I4 ( ΔI4L/ΔJ2RFlagI4L ) were constructed and used in new immunoprecipitation experiments . Immunoprecipitation with anti-Flag antibodies confirmed the interaction of HR1 and I4 with F4 as well as with HR2 and Hp53R2 ( Figure 6A ) . We typically observed weaker R2 bands in immunoprecipitations of Flag-tagged HR1 compared to Flag-tagged I4 despite similar amounts of these two proteins being immunoprecipitated ( Figure 6A ) . This result was likely due to competition between the Flag-tagged HR1 protein and endogenous HR1 , whereas Flag-tagged I4 is expressed in the ΔI4L background and thus does not have to compete for binding to R2 proteins with endogenous I4 . We also prepared extracts from cells infected with ΔF4L/ΔJ2RHisY300FF4L or ΔF4L/ΔJ2RHisF4L viruses and observed that these His6-tagged proteins could also be co-immunoprecipitated with HR1 protein ( Figure 6B ) . Reciprocal co-immunoprecipitation experiments confirmed an interaction between F4 and HR1 proteins ( Figure S3 ) . Other Chordopoxvirus R2 proteins rescued the replication defect of VACV ΔF4L strains ( Figure 3B ) . Therefore , we determined whether these proteins could also interact with HR1 . ECTV , MYXV , and SFV R2 proteins all co-immunoprecipitated with HR1 ( Figure 6B ) . Although there appeared to be differences in the efficiency of HR1 association , western blotting of lysates showed that this reflected differences in R2 expression levels ( Figure 6B ) . These results confirm that RR subunits from poxviruses that infect a diversity of mammalian hosts have conserved the capacity to interact with HR1 . In uninfected cells , mammalian RR subunits show an exclusively cytoplasmic distribution [54] , [55] , [56] . Confocal microscopy studies with antibodies directed against endogenous ( Figure S4A ) or epitope-tagged ( Figure S4B ) RR subunits suggested that VACV infection did not alter host RR localization and VACV RR subunits were also found to exhibit a similar cytoplasmic distribution . The previous studies showed that F4 interacts with HR1 but did not prove whether such an interaction was essential for viral replication . Numerous structural and peptide-inhibition studies of class I RR proteins have identified a C-terminal peptide ( boxed in Figure 1 ) in R2 subunits as critical for interaction with R1 proteins [11] , [57] , [58] , [59] , [60] , [61] . Since this C-terminal peptide is well conserved in F4 ( Figure 1 ) , we speculated that HR1-F4 interactions were also dependent on this peptide . To test this hypothesis , we generated the VACV strain ΔF4L/ΔJ2RHisF4LΔR1BD , encoding a truncation mutant of F4 that lacks the C-terminal seven residues representing the putative R1-binding domain ( R1BD ) . We also generated an R1BD mutant that also encodes the Y300F substitution , ( ΔF4L/ΔJ2RHisY300FF4LΔR1BD ) . As shown in Figure 7A , His6-tagged F4 co-immunoprecipitated with HR1 in HeLa cell extracts . However , there was a clear reduction ( by ∼90% ) in co-immunoprecipitation of His6-tagged F4 proteins lacking the R1BD , despite comparable levels of these two forms of F4 in lysates and immunoprecipitates . Thus , F4 appears to have conserved the R1-binding peptide encoded by class I RRs . We used plaque area measurements to determine if deleting the R1BD would alter VACV plating properties ( Figure 7B ) . The control viruses exhibited the same relative plaque sizes noted previously ( i . e . wild-type = ΔF4L/ΔJ2RHisF4L>ΔF4L>ΔF4L/ΔJ2RHisY300FF4L ) and the differences were all significant ( P<0 . 05 ) . However , the ΔF4L/ΔJ2RHisF4LΔR1BD and ΔF4L/ΔJ2RHisY300FF4LΔR1BD strains produced plaques no different in size from those produced by ΔF4L strains ( P>0 . 05 ) . This suggested that the F4 R1BD was not only required for RR activity , but that the HR1-F4 interaction was also responsible for the dominant negative effects observed with strains encoding the Y300F-substituted F4 protein with an intact R1BD . We also confirmed in these studies that inactivation of J2R alone had no significant effect on plaque size ( Figure 7B ) . Our results suggested that deleting the F4L gene renders VACV highly dependent upon the host cell for provision of a complementing RR activity . This leads to the prediction that the efficiency of growth of a ΔF4L virus will depend upon the level of cellular RR activity . To test this hypothesis , we used two pancreatic cancer cell lines that have been previously reported to exhibit high ( PANC-1 ) and low ( CAPAN-2 ) levels of RR subunit expression and activity [45] , [62] . We prepared cell-free extracts from wild-type virus-infected ( or mock-infected ) PANC-1 and CAPAN-2 cells , and used western blots to measure the levels of RR proteins . This study confirmed that HR1 , HR2 , and Hp53R2 are expressed at lower levels in CAPAN-2 cells , relative to PANC-1 cells , and that this phenotype is unaffected by VACV infection ( Figure 8A ) . We then seeded approximately equal numbers of PANC-1 and CAPAN-2 cells into culture dishes and infected them with wild-type and mutant strains . The total titers for each of these infections at 48 or 72 h post-infection are plotted in Figure 8B . Division of the mean titers obtained in PANC-1 cells by those obtained in CAPAN-2 cultures for each virus gave an estimate of the fold difference in replication efficiencies for each strain in these cells ( Figure 8C ) . These data showed that PANC-1 cells support a relatively normal level of replication of most mutant VACV strains . For example , the wild-type virus grew only ∼3–6-fold better on PANC-1 cells than did ΔF4L , and ΔI4L/ΔF4L , and ΔF4L/ΔJ2R strains ( Figure 8B ) . One exception to this rule is that the wild-type virus produced titers ∼16-fold higher than the ΔI4L/ΔF4L/ΔJ2R strain on PANC-1 cells ( Figure 8B ) . This suggested that in certain cell types and in the absence of J2R and F4L , I4L may play some role in supporting VAC replication . A more notable feature of this experiment is that all virus tested grew better on PANC-1 cells compared to CAPAN-2 cells . The wild-type , ΔI4L , and ΔF4L/ΔJ2RHisF4L strains produced yields 6–8-fold higher on PANC-1 cells than CAPAN-2 cells 48 h post-infection and this difference was greatly exacerbated by deletion or mutation of F4L ( Figure 8C ) . For example , the ΔF4L strain grew 18–30-fold better on PANC-1 cells and the ΔF4L/ΔJ2RHisY300FF4L strain yielded a 113-fold increase in titer on PANC-1 cells compared to CAPAN-2 cells . In fact , titering of input inocula indicated that the ΔF4L/ΔJ2RHisY300FF4L strain did not productively replicate in CAPAN-2 cells ( data not shown ) . This suggested that the reduced RR activity of CAPAN-2 cells imposes a barrier to replication of this mutant . Collectively , these results suggested that the replication defects exhibited by ΔF4L and ΔF4L/ΔJ2RHisY300FF4L strains can be complemented in human cancer cell lines over-expressing cellular RR subunits . However , direct evidence for the linkage between cellular RR levels and mutant rescue requires further studies . We used an animal model to determine if the apparent differential requirement for VACV RR subunits for replication in culture would be recapitulated in vivo . We infected groups of five NMRI mice with equal doses of wild-type , ΔI4L , ΔF4L , or ΔI4L/ΔF4L strains and tracked changes in animal body weight over 24 days . The wild-type and ΔI4L strains exhibited a similar degree of virulence , causing the death of 5/5 and 4/5 animals , respectively , within seven days of infection . In contrast , both ΔF4L and ΔI4L/ΔF4L strains were highly attenuated , with all animals displaying little to no signs of disease and surviving the infections ( Figure 9A ) . There were small , transient drops in body weight for animals infected with the ΔF4L strain around days 5 and 7 , otherwise these animals , and those infected with the ΔI4L/ΔF4L strain , showed no obvious signs of morbidity when compared to the mock-infected control group ( Figure 9A ) . To obtain a more quantitative measurement of the pathogenic nature of these infections , we isolated lung tissues from mice infected with the aforementioned strains on day 5 post-infection . Wild-type and ΔI4L strains clearly had a replication advantage over ΔF4L and ΔI4L/ΔF4L strains with lung titers approximately 4 logs higher than the latter two strains ( Figure 9B ) . These results indicate that VACV RR subunits are differentially required for virulence in mice .
Acquisition of a suitable supply of dNTPs to support replication is a challenging feat for mammalian DNA viruses because most host cells exist predominantly in a terminally-differentiated and quiescent state [63] . The S-phase-specific nature of host R2 expression leaves quiescent cells with only p53R2-R1 complexes to maintain a low ( ∼2–3% the level of cycling cells [64] ) level of RR activity to meet the demands of DNA repair and mitochondrial genome synthesis [16] , [19] , [65] . Since ribonucleotide reduction is the rate-limiting step in mammalian dNTP biogenesis [26] , low RR activity may pose a barrier to productive infection . Therefore , DNA viruses must replicate only in cycling cells , induce host RR activity upon infection , and/or encode their own RR enzymes [63] . Many large DNA viruses , including herpes- , irido- , asfra- and poxviruses have evolved the later strategy . It is clear that herpesvirus-encoded RR proteins are important because inactivation of viral RR genes leads to replication defects in vitro and in animals [66] , [67] , [68] , [69] . Furthermore , inhibiting complex formation by herpes simplex virus ( HSV ) R1 and R2 proteins with a C-terminal R2 peptide mimic has been shown to prevent HSV replication in culture [70] , [71] , [72] . Interestingly , β-herpesviruses , only encode an R1 gene , although it is still required for virulence [63] . However , it is unlikely to play a catalytic role in dNTP biogenesis as it encodes mutations at key catalytic residues that would render this subunit inactive in RR complexes [73] , [74] , [75] . Recent evidence suggests that β-herpesviruses may induce host RR protein expression , possibly explaining why viral RR function was not conserved [75] , [76] . What biological purpose is served by β-herpesvirus R1 proteins is unclear , although it has been suggested that these proteins might play some role in inhibiting apoptosis [77] . The increasing availability of virus genome sequences has revealed that differential conservation of viral RR genes is actually a widespread phenomenon among eukaryotic DNA viruses . However , in contrast to the case of β-herpesviruses , most DNA viruses that encode a single RR subunit encode R2 proteins while R1 is frequently absent . For example , iridoviruses in the Megalocytivirus genus only encode an R2 subunit while all other iridoviruses encode both RR subunits [78] . Furthermore , certain members of the Phycodnaviridae and Ascoviridae viral families also only encode R2 subunits [79] . This R2 bias is also seen in bacteriophage belonging to the Siphoviridae and Myoviridae families , suggesting that even prokaryotic viruses have biased conservation of RR genes [79] . Perhaps the most biased conservation of RR genes is found in poxviruses , with a clear favoring of R2 over R1 ( Table S1 ) . Even Orthopoxviruses , which typically encode both R1 and R2 genes , contain a member ( horsepox virus ) that encodes a fragmented R1 gene [80] . Many other Chordopoxviruses show no evidence of ever having encoded an R1 activity , including the Leporipoxviruses MYXV and SFV which we sequenced ten years ago [81] , [82] . At that time , the close similarity of MYXV and SFV R2 subunits to mammalian R2 proteins , and absence of a viral R1 homolog , led us to suggest that Leporipoxvirus R2 subunits were likely forming chimeric complexes with host R1 proteins [82] . Subsequent biochemical studies of VACV F4 and I4 by Chimploy and Mathews [33] found that mixing purified F4 and I4 with MR1 and MR2 proteins , respectively , resulted in functional , chimeric RR enzymes . However , the two kinds of chimeric RRs did not display identical properties . Whereas the native complexes ( i . e . I42F42 and MR12MR22 ) were about equally active , the I42MR22 enzyme exhibited ∼5-fold less activity and the MR12F42 enzyme showed up to 2-fold more activity than either native complex [33] . These observations may explain why Child et al . reported their ΔI4L strain to exhibit no observable replication defect in culture and only a small ( ∼10-fold ) increase in LD50 for mice compared to wild-type VACV [34] . However , in an attempt to develop new vaccine strains , Lee et al . generated a F4L insertional inactivation mutant and reported significant increases of ∼1000-fold in the LD50 of this mutant in a similar mouse model used by Child et al . with their ΔI4L strain [34] . Collectively these independent bioinformatic , biochemical , and molecular genetic studies all suggested that poxvirus R2 subunits might be more important for viral replication than R1 subunits . However , the contributions of poxvirus R1 and R2 subunits to viral replication and pathogenesis had never been directly compared nor was it clear why R2 subunits may be more important to the poxvirus life cycle . We examined this issue in detail by generating a panel of VACV RR mutant strains and analyzing their plaque , growth , and pathogenic properties . Our studies clearly show that these properties are far more affected in ΔF4L strains compared to ΔI4L strains ( Figures 3 and 9 ) . Combining F4L and I4L deficiencies caused no further inhibition of virus growth , suggesting that the phenotype is dominated by the integrity of the F4L locus . We also showed that inactivating VACV J2R in the ΔF4L background did not further impede replication ( Figure 3 ) , suggesting that the salvage pathway for dNTP production is either not required for replication in culture , or is sufficiently complemented by host TK enzymes . This result made it possible to use the J2R locus as a site for introducing ectopic copies of different recombinant R2 proteins . The ΔF4L strain's phenotype can be completely complemented by a gene encoding His6-tagged F4 and by genes encoding other Orthopoxvirus or Leporipoxvirus R2 proteins ( Figure 3B ) . Why Hp53R2 failed to rescue this phenotype is unclear but several possibilities exist . For one , R1-p53R2 complexes exhibit only 40–60% the activity of R1-R2 complexes [17] and this reduced activity might not meet some activity threshold required for efficient viral replication . Secondly , significant fractions of Hp53R2 proteins are bound in inactive complexes by p53 and p21 proteins , and are only released after appropriate signaling pathways have been activated [83] , [84] . Therefore , even if one over-expresses Hp53R2 , it may not produce a sufficient level of “free” Hp53R2 that could complex with R1 proteins . Finally , Hp53R2 has recently been shown to inhibit MEK2 , a kinase involved in the activation of the Ras-Raf-MAPK signaling pathway [85] . This inhibition could be detrimental as activation of the MAPK pathway is required for VACV replication [86] . These possibilities are currently being addressed . Attempts to generate a VACV strain over-expressing HR2 have thus far been unsuccessful and so it is unclear whether HR2 can complement a ΔF4L strain . The hypothesis that F4 protein can compete with ( or replace ) cellular small subunits to form chimeric RR complexes in vivo is strongly supported by the dominant-negative phenotype exhibited by Y300F-substituted F4 protein ( Figure 3 ) . Viruses encoding these mutant proteins replicate very poorly and produce extremely small plaques . Furthermore , this phenotype is not altered by the presence or absence of I4 ( Figure 3B ) . The genetic data are fully concordant with our immunoprecipitation experiments , which showed that F4 interacted with cellular RR subunits ( Figure 5A ) and that this interaction was unaffected by I4 ( Figure 5B ) . We also found that other Chordopoxvirus R2 proteins co-immunoprecipitated with HR1 ( Figure 6B ) , which is consistent with the ability of these proteins to rescue the replication defect of the ΔF4L strain . The ability of various poxvirus R2 subunits to interact with HR1 might be explained by the high degree of sequence conservation amongst mammalian RR subunits and the fact that Chordopoxvirus R2 subunits are typically >70% identical to mammalian subunits . The observation that poxvirus RR genes are generally more similar to cellular RR genes than other virus RR genes has led to the suggestion that poxviruses have acquired RR genes through horizontal transfer events with their host [87] , [88] . Interestingly , other viral and bacterial pathogens have also likely acquired RR enzymes through host gene capture [89] , [90] , [91] . It would be of interest to determine if Chordopoxvirus R2 proteins exhibit a quantitative binding preference for R1 proteins isolated from their natural hosts ( e . g . MR1 with ECTV R2 and rabbit R1 with MYXV and SFV R2 ) , as that would be an expected consequence of evolutionary adaption to a particular host . Potential differences in binding affinities between poxvirus and host subunits may provide further insight into factors that contribute to poxvirus host range which remain poorly defined . During infection , ∼8-fold more F4 than I4 subunits are synthesized [92] . In tissue culture , levels of mammalian R1 are constant during the cell cycle due to its long half-life [13] , while R2 subunits are quickly degraded late in mitosis leading to a much shorter half-life [15] . Given the relatively reduced activity of R1-Hp53R2 complexes [17] , it is possible that production of F4 in excess allows these subunits to form needed complexes with both viral and host R1 subunits . Interestingly , poxvirus R2 subunits , like Hp53R2 , lack much of the N-terminal sequences found in HR2 including phosphorylation and ubiquitination sites that may regulate HR2 function and degradation ( Figure 1 ) [15] . This may explain why F4 protein levels are stable for at least 12 h after infection [92] . It seems likely that adaptive changes during evolution has led to conservation of poxvirus R2 enzymatic function yet has resulted in a loss of regulatory sequences that may restrict viral subunit levels in the host . We hypothesized that the ability of Chordopoxvirus R2 proteins to interact with HR1 was due to the high degree of conservation of the C-terminal seven residues between poxvirus and mammalian R2 subunits ( Figure 1 ) . This C-terminal motif has been well-characterized in R1-R2 interaction studies of various class I RR enzymes [11] , [57] , [58] , [59] , [60] , [61] and an oligopeptide mimic ( 7FTLDADF1 ) of mammalian R2 C-termini has been shown to inhibit RR activity [57] . Positions 1 , 5 , and 7 in this mimic are the most critical determinants of RR inhibition [57] , and the residues at these positions are conserved in the C-terminus of F4 ( FSLDVDF ) suggesting that VACV and mammalian RR share a common R1-R2 subunit interaction mechanism . The large differences between C-terminal sequences of HSV ( YAGAVVNDL ) and mammalian R2 subunits likely explains why no evidence could be found for interaction of HSV RR proteins with host subunits [23] and why peptide mimics of the HSV R2 C-terminus are highly selective antivirals [71] . Previous studies have used the F4 heptapeptide to generate an affinity column for I4 purification [93] . Therefore , we thought it was likely that F4 interacted with R1 proteins in a similar manner as found with cellular R2 subunits . Indeed , interaction of F4 proteins lacking the putative R1BD with HR1 was clearly impaired ( Figure 7A ) and strains expressing these truncated proteins were unable to rescue the small plaque phenotype of the ΔF4L mutant ( Figure 7B ) . However , deleting the R1BD from Y300F F4 did suppress the dominant negative phenotype ( Figure 7B ) , which further implied that F4 functionally interacts with HR1 through the C-terminus of F4 . Collectively our data show that VACV F4 proteins ( and likely other poxvirus R2 proteins ) are required for efficient viral replication in culture as well as for pathogenesis ( Figure 9 ) . While our studies of CDV and HU sensitivities ( Table 2 ) suggest a defect in RR activity and subsequent dNTP pool biogenesis as the underlying cause for the defect of ΔF4L strains , it is possible that these are only indirect consequences of inactivation of F4L and other functions of F4 are required for replication . However , the dominant negative phenotype of the Y300F-encoding strains in the presence or absence of I4 ( Figure 3B ) , the requirement of the R1BD to produce this phenotype and interact with HR1 , and the similar localization observed with viral and cellular RR subunits in infected cells ( Figure S4 ) all support the simple conclusion that F4 must form complexes with host R1 proteins to facilitate dNTP biogenesis . The critical importance of this interaction for VACV replication suggests that ΔF4L strains may act as selective oncolytic agents . A wide variety of human cancers exhibit elevated RR expression patterns and prolonged treatment of patients with RR inhibitors can lead to drug resistance as result of HR2 gene amplification [94] , [95] , [96] . For example , a recent study of patients with non-small lung cancer found that elevated host RR expression levels in patients' tumors were directly correlated with reduced response to chemotherapy and poorer prognoses [94] . Interestingly , hrR3 , a HSV mutant strain with an inactivated viral R1 gene , replicates more efficiently in cancers with elevated host RR expression [97] and shows promise as an oncolytic agent in mouse models [98] . The enhanced replication of ΔF4L and ΔF4L/ΔJ2RHisY300FF4L strains in PANC-1 cells relative to CAPAN-2 pancreatic cancer cell lines ( Figure 8C ) correlates well with the higher levels of RR subunits in PANC-1 cells ( Figure 8A ) and documented differences in RR activities between these two cell lines [45] . While further evidence will be needed to prove that host RR proteins complement the ΔF4L strain replication defect in PANC-1 cells , our results build a strong circumstantial case for the dependence of these F4L mutant strains on host RR activity . Since VACV and other , non-Orthopoxviruses ( e . g . Leporipoxviruses [99] , [100] , and Yatapoxviruses [101] ) that encode R2 subunits have shown potential for use in cancer virotherapy , we suggest that deletion of R2 subunit genes from these viruses may create more selective oncolytic agents . During our studies we noted that poxviruses that encode R2 and TK genes tend to have higher A+T base content in their genomes than poxviruses lacking these genes ( Table S1 ) . This strong correlation suggests that hybrid virus-host RR complexes and/or poxvirus TK proteins have contributed to the establishment of unique dNTP pools that have influenced viral genome composition during the co-evolution of poxviruses with their hosts . Recently an APC mimic has been identified in poxviruses that lack R2 and TK genes including Molluscipoxviruses , Parapoxviruses , and crocodilepox virus [102] . The APC mimic in orf virus , termed “PACR” ( poxvirus APC/cyclosome regulator ) inhibits APC activity and causes mammalian cells to accumulate in G2/M phases of the cell cycle [102] . Since APC targets mammalian TK [103] and R2 [15] proteins for degradation during mitosis , poxvirus APC mimics may serve to prevent degradation of these host nucleotide metabolism proteins during viral replication . This is supported by the finding that PACR inhibits host TK degradation during the cell cycle [102] . This reliance on host nucleotide metabolism proteins may explain why poxviruses encoding APC mimics have low A+T content in their genomes ( Table S1 ) . Strict reliance on host nucleotide metabolism machinery may also explain why GC-rich Mollusci- and Parapoxviruses have a rather limited host range when compared to AT-rich Orthopoxviruses which encode their own RR and TK [104] ( Table S1 ) . Therefore , poxviruses appear to have acquired different mechanisms to obtain dNTPs for replication , which may ultimately influence their genomic composition and host tropism .
Cell and virus culture methods have been described elsewhere [105] . Wild-type VACV and its mutant derivatives were derived from strain Western Reserve ( WR ) originally acquired from the American Type Culture Collection . Non-transformed African Green Monkey kidney cells ( BSC-40 ) were normally cultured in modified Eagle's medium ( MEM ) supplemented with 5% fetal bovine serum ( FBS ) . HeLa human cervical adenocarcinoma and human embryonic lung ( HEL ) cells were cultured in Dulbeccos MEM ( DMEM ) supplemented with 10% FBS . PANC-1 and CAPAN-2 cells are human pancreatic epithelioid carcinoma and adenocarcinoma lines , respectively and were also cultured in DMEM supplemented with 10% FBS . All of the above cell lines were originally obtained from the American Type Culture Collection . A U20S human osteosarcoma cell line that expresses Cre recombinase was a kind gift from Dr . J . Bell ( University of Ottawa ) . These cells were maintained in DMEM supplemented with 10% FBS . Cells were cultured in Opti-MEM media ( Invitrogen; Carlsbad , CA ) for experiments requiring transfections . ( S ) -1-[3-hydroxy-2- ( phosphonomethoxy ) propyl]cytosine or cidofovir ( CDV or HPMPC ) was from Dr . K . Hostetler ( University of California , San Diego ) . Hydroxyurea ( HU ) was obtained from Alfa Aesar ( Ward Hill , MA ) . X-gal and X-glu substrates were obtained from Sigma Chemical Co . ( St . Louis , MO ) and Clontech ( Palo Alto , CA ) , respectively . Phosphonoacetic acid ( PAA ) was from Sigma Chemical Co . Isatin-β-thiosemicarbazone ( IBT ) was from Pfaltz and Bauer ( Waterbury , CT ) . Mycophenolic acid ( MPA ) and xanthine were obtained from Sigma Chemical Co . Hypoxanthine was obtained from ICN Biomedicals , Inc . ( Aurora , OH ) . Compounds were diluted to their final concentration in MEM ( CDV; HU; PAA; IBT ) or in a 1∶1 mixture of MEM and 1 . 7% noble agar ( X-gal; X-glu ) immediately prior to use . Taq and PfuUltra DNA polymerases were obtained from Fermentas ( Burlington , ON ) and Stratagene ( La Jolla , CA ) , respectively . Normal mouse and goat serum and goat polyclonal antibodies against human R1 ( HR1 ) , human R2 ( HR2 ) , and human p53R2 ( Hp53R2 ) were from Santa Cruz Biotechnology , Inc . ( Santa Cruz , CA ) . Mouse monoclonal antibodies against HR1 and HR2 were from Millipore ( Billerica , MA ) and Santa Cruz Biotechnology , Inc . , respectively . Mouse monoclonal antibodies against Flag and His6 ( His ) epitopes were from Sigma and Roche ( Mississauga , ON ) , respectively . Rabbit anti-Flag epitope polyclonal antibodies were obtained from Sigma . A mouse monoclonal antibody was raised against bacterially-expressed , recombinant ECTV R2 antigen by ProSci ( Poway , CA ) . The resulting antibody also recognizes VACV F4 and was used for western blotting . In some cases , a rabbit anti-F4 polyclonal antibody was also used for western blotting . The plasmid used to express recombinant ECTV R2 antigen and the rabbit anti-F4 antibody were kindly provided by Dr . M . Barry ( University of Alberta ) . A rabbit anti-VACV I4 polyclonal antibody was obtained from Dr . C . Mathews ( Oregon State University ) . Although this antibody recognizes VACV I4 , it also cross-reacts with cellular R1 on western blots [92] . The mouse monoclonal antibody against VACV I3 has been described [40] and the mouse monoclonal antibody against cellular actin was from Sigma . Protein extracts for western blots and immunoprecipitations were prepared from cell cultures by lysing cells on ice in a buffer containing 150 mM NaCl , 20 mM Tris ( pH 8 . 0 ) , 1 mM EDTA , and 0 . 5% NP-40 along with freshly-added phenylmethylsulfonyl fluoride ( 100 µg/mL ) and protease inhibitor tablets ( Roche ) . For western blots , 20–40 µg of total protein were subjected to SDS-PAGE and subsequently blotted with appropriate antibodies after transfer to nitrocellulose membranes . Membranes were scanned using an Odyssey scanner ( Li-COR Biosciences ) . Protein extracts for immunoprecipitations were recovered as described above 6–8 h post-infection from 107 HeLa cells infected with indicated strains at a multiplicity of infection ( MOI ) of 10 . Extracts were then pre-cleared by incubation with normal mouse or goat serum along with protein G sepharose beads ( GE Healthcare Life Sciences; Piscataway , NJ ) for 30 min at 4°C with constant inversion . The samples were subsequently centrifuged ( 2 , 500 rpm , 1 min , 4°C ) and supernatants were transferred to fresh tubes . These extracts were then incubated with the primary antibodies overnight at 4°C with constant inversion . Fresh protein G beads were then added to the extracts and incubated for 2 h at 4°C after which the beads were collected ( 2 , 500 rpm , 1 min , 4°C ) and washed four times with lysis buffer . The resulting bead-protein complexes were resuspended in SDS-PAGE loading buffer , boiled for 15 min and subjected to SDS-PAGE . Western blotting was then performed as described above . Whole cell extracts ( lysates ) were also blotted with indicated antibodies and represented ∼5% of the input material used for immunoprecipitations . Plaque dimensions were measured on 60-mm-diameter dishes of confluent BSC-40 cells infected with ∼100 plaque-forming units ( PFU ) of the indicated strain . After 48 h of infection , triplicate plates were stained with crystal violet and scanned using an HP ScanJet 6300C scanner . The resulting image files were analyzed using ImageJ v1 . 04g software ( National Institutes of Health , USA ) . Unpaired t-tests or one-way ANOVA tests were performed on mean plaque areas between wild-type and each of the various RR mutant strains using GraphPad Prism ( San Diego , CA ) software ( version 4 . 0 ) . In some cases two different RR mutant strains were also compared for differences in mean plaque areas . A P value of <0 . 05 was considered to be statistically significant . Growth analyses were conducted in BSC-40 , HeLa , PANC-1 and CAPAN-2 cell cultures using the indicated MOI and strains . Cells were harvested by scraping monolayers into the culture media at the indicated time points followed by three rounds of freeze-thawing . Virus stocks were titered on BSC-40 cells . For viral genome replication analyses , at the indicated times post-infection , BSC-40 cells were harvested by scraping , collected by centrifugation ( 800 rpm , 10 min , 4°C ) washed once with PBS , and resuspended in 500 µL of 10× saline-sodium citrate ( SSC ) loading buffer containing 1 M ammonium acetate [106] . The cells were then disrupted by three cycles of freeze-thaw and 50-µL aliquots of the lysates were applied to a Zeta probe membrane using a slot-blot apparatus ( Bio-Rad , Richmond , CA ) . Samples were denatured with 1 . 5 M NaCl and 0 . 5 M NaOH and washed twice with 10× SSC loading buffer . The membrane was then hybridized with a 32P-labeled E9L gene probe . After the membrane was washed with SSC buffer and air dried , it was exposed to a phosphorimager screen , imaged using a Typhoon 8600 phosphorimager and the data were processed using ImageQuant software , ( version 5 . 1 ) [40] . In some cases 0 . 5 mM HU was added to the media 1 h post-infection . Plaque-reduction assays were performed as previously described [105] . Briefly , 35-mm-diameter dishes of confluent BSC-40 cells were inoculated with ∼100 PFU of the indicated virus strains , and 1 h after infection either drug-free medium or medium containing the indicated doses of CDV or HU was added to the cultures and the plates were incubated at 37°C for 48 h . Plates were then stained with crystal violet to visualize and count plaques . Mean 50% effective concentration ( EC50 ) values and their 95% confidence intervals ( CIs ) were calculated using nonlinear regression analyses with GraphPad Prism software after three independent experiments had been performed . In cases where the 95% CIs of two different EC50 values did not overlap , these two EC50 values were considered to be statistically different ( P<0 . 05 ) . HeLa cells were grown on coverslips in 24-well plates and infected with the indicated virus strains at a MOI of 5 for 10 h . The cells were fixed for 30 min on ice with 4% paraformaldehyde in PBS . The fixed cells were blocked and permeabilized for 1 h at RT in PBS containing 0 . 1% Tween ( PBS-T ) as well as 10% BSA . The coverslips were then incubated with the primary antibodies diluted in PBS-T ( 1% BSA ) for 2 h at RT , washed three times and then incubated with secondary antibodies conjugated to Alexa 488 or 594 ( Invitrogen ) for 1 h at RT . The cells were then counterstained with 10 ng/mL 4′ , 6′-diamidino-2-phenylindole ( DAPI ) in PBS-T for 15 min . The specimens were examined using a Zeiss 710 Laser-Scanning confocal microscope equipped with DAPI , Alexa 488 , and Alexa 594 filters . Images were captured and processed using ZEN 2009 software and Adobe Photoshop ( version 10 . 0 . 1 ) . BSC-40 cells were grown to confluence and then infected for 1 h with the appropriate VACV strain ( see below ) at a MOI of 2 in 0 . 5 mL of PBS . The cells were then transfected with 2 µg of linearized plasmid DNA using Lipofectamine 2000 ( Invitrogen ) . See Text S1 and for details regarding the primers and transfer vectors used to generate recombinant VACV strains . The cells were returned to the incubator for another 5 h , the transfection solution was replaced with 5 mL of fresh growth medium , and the cells were cultured for 24–48 h at 37°C . Virus progeny were released by freeze-thawing , and the virus titer was determined on BSC-40 cells . To identify recombinant virus , plaques were stained with X-gal or X-glu ( both at 0 . 4 mg/mL ) in solid growth media , or cultured in media containing 25 µg/mL MPA supplemented with xanthine ( 250 µg/mL ) and hypoxanthine ( 15 µg/mL ) for selection of yfp-gpt-encoding strains ( see Text S1; [107] ) . The PCR was used to confirm insertions/deletions in the resulting recombinant viruses . The primers: 5′-GATGAATGTCCTGGATTGGA-3′ & 5′-ATTCCAAAGATCCGACGGTA-3′ were used to PCR amplify ∼700 bp of I4L sequence that should not be present in ΔI4L strains . The primers: 5′-ATGGAACCCATCCTTGCACC-3′ & 5′-ATCTTCTTGAGACATAACTC-3′ were used to amplify ∼930 bp of F4L sequence that should not be present in ΔF4L strains . Disruption of J2R sequence was detected with primers: 5′-TCCTCTCTAGCTACCACCGCAATAG-3′ & 5′-GTGCGGCTACTATAACTTTTTTCC-3′ that bind to regions of J2R flanking the insertion site of pSC66 vector [108] sequences ( see below ) . Primers TGGATTCGTACAAATTGGATTCTAT & AATTGCTATTTCAGAGATGAGGTTC were used to amplify an ∼800 bp fragment from VACV DNA polymerase ( E9L ) sequence to serve as a positive control for amplification . In some cases western blotting was used to confirm the presence or absence of gene expression in the described VACV strains . Details of how each VACV strain was constructed are provided in Text S1 and the marker rescue strategies used in these studies are depicted in Figure 2A . PfuUltra DNA polymerase ( Stratagene ) was used to PCR-amplify DNA for cloning whereas Taq DNA polymerase ( Fermentas ) was used for PCR diagnostic purposes . Plasmid constructs were verified by sequencing and all virus strains were plaque-purified a minimum of three times in BSC-40 cells or Cre recombinase-expressing U20S cells . All VACV strains were characterized by PCR ( data not shown ) and sometimes by western blotting ( Figure 2B ) , although for brevity only characterization of the main viral strains discussed throughout this study is shown . Female NMRI mice , 3 to 4 weeks of age , were obtained from Charles River Laboratories ( Brussels , Belgium ) . Mice were utilized at 5 mice per infection or control group for morbidity studies . Mice were anesthetized using ketamine-xylazine and inoculated intransally ( or mock-inoculated ) with 4×104 PFU of virus diluted in 30 µL of saline . Animal body weights were recorded over the next 24 days or until the animals had to be euthanized because of more than 30% loss in body weight . To determine viral titers in lungs , two ( wild-type infections ) or five animals ( ΔI4L , ΔF4L , and ΔI4L/ΔF4L infections ) were euthanized on day 5 . Lung samples were removed aseptically , weighed , homogenized in MEM , and frozen at −70°C until assayed by titrations on HEL cells . For mouse pathogenicity experiments , the ΔI4L/ΔF4L strain was generated with the pDGloxPKODEL vector . This allowed for removal of the yfp-gpt marker cassette from the I4L locus after passage in U20S cells expressing Cre recombinase . No differences were found in replication in culture between strains expressing the yfp-gpt cassette and those with this cassette deleted by Cre recombination ( Figure S5 ) . See Text S1 for further details . All animal work was approved by the K . U . Leuven Animal Care and Use Committee . All animal guidelines and policies were in accordance with the Belgian Royal Decree of 14 November 1993 concerning the protection of laboratory animals and the European Directive 86-609-EEC for the protection of vertebrate animals used for experimental and other scientific purposes . | Efficient genome replication is central to the virulence of all DNA viruses , including poxviruses . To ensure replication efficiency , many of the more virulent poxviruses encode their own nucleotide metabolism machinery , including ribonucleotide reductase ( RR ) enzymes , which act to provide ample DNA precursors for replication . RR enzymes require both large ( R1 ) and small ( R2 ) subunit proteins for activity . Curiously , some poxviruses only encode R2 subunits . Other poxviruses , such as the smallpox vaccine strain , vaccinia virus ( VACV ) , encode both R1 and R2 subunits . We report here that the R2 , but not the R1 , subunit of VACV RR is required for efficient replication and virulence . We also provide evidence that several poxvirus R2 proteins form novel complexes with host R1 subunits and this interaction is required for efficient VACV replication in primate cells . Our study explains why some poxviruses only encode R2 subunits and identifies a role for these proteins in poxvirus pathogenesis . Furthermore , we provide evidence that mutant poxviruses unable to generate R2 proteins may become entirely dependent upon host RR activity . This may restrict their replication to cells that over-express RR proteins such as cancer cells , making them potential therapeutics for human malignancies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"biology/dna",
"replication",
"virology/virulence",
"factors",
"and",
"mechanisms",
"virology/viral",
"replication",
"and",
"gene",
"regulation",
"virology/virus",
"evolution",
"and",
"symbiosis",
"virology/animal",
"models",
"of",
"infection",
"microbiology",
... | 2010 | Vaccinia Virus–Encoded Ribonucleotide Reductase Subunits Are Differentially Required for Replication and Pathogenesis |
This qualitative study aimed to provide an in-depth understanding of the meaning of dengue fever ( DF ) amongst people living in a dengue endemic region , dengue prevention and treatment-seeking behaviours . The Health Belief Model was used as a framework to explore and understand dengue prevention behaviours . A total of 14 focus group discussions were conducted with 84 Malaysian citizens of different socio-demographic backgrounds between 16th December , 2011 and 12th May , 2012 . The study revealed that awareness about DF and prevention measures were high . The pathophysiology of dengue especially dengue haemorrhagic fever ( DHF ) and dengue shock syndrome ( DSS ) were rarely known; as a result , it was seen as deadly by some but was also perceived as easily curable by others without a basis of understanding . Young adults and elderly participants had a low perception of susceptibility to DF . In general , the low perceived susceptibility emerged as two themes , namely a perceived natural ability to withstand infection and a low risk of being in contact with the dengue virus vector , Aedes spp . mosquitoes . The barriers to sustained self-prevention against dengue prevention that emerged in focus groups were: i ) lack of self-efficacy , ii ) lack of perceived benefit , iii ) low perceived susceptibility , and iv ) unsure perceived susceptibility . Low perceived benefit of continued dengue prevention practices was a result of lack of concerted action against dengue in their neighborhood . Traditional medical practices and home remedies were widely perceived and experienced as efficacious in treating DF . Behavioural change towards attaining sustainability in dengue preventive practices may be enhanced by fostering comprehensive knowledge of dengue and a change in health beliefs . Wide use of unconventional therapy for DF warrants the need to enlighten the public to limit their reliance on unproven alternative treatments .
Since the beginning of the 21st century , dengue fever ( DF ) have been the most important vector-borne arboviral disease in humans , occurring mainly in tropical and sub-tropical countries where over 2 . 5 billion people are at risk of infection [1] . With an estimated 50–100 million dengue infections worldwide , the disease is currently endemic in more than 125 countries in Africa , the Americas , the Eastern Mediterranean , South-east Asia , and the Western Pacific [1] . The incidence of DF in Malaysia has increased dramatically in recent decades and has remained a serious public health problem . Since the first reported outbreak in 1902 , several major outbreaks were reported including in 1974 , 1978 , 1982 , and 1990 [2] . The dengue incidence rate increased from 44 . 3 cases/100 , 000 population in 1999 to 181 cases/100 , 000 population in 2007 , which exceeded the national target for the incidence rate of DF and dengue haemorrhagic fever DHF of less than 50 cases/100 , 000 population . An increase in dengue deaths in the adult population has been observed since 2002 . The case fatality rates , however , for both DF and DHF remained below 0 . 3% [3] . The highest age-specific morbidity rates were in the 15 to 35 years age groups [4] . Preventing or reducing dengue virus transmission depends largely on controlling the mosquito vectors . As such , in most dengue endemic countries , anti-dengue campaign involved media bombardment with messages to eradicate the Aedes spp . mosquitoes to the point that Aedes spp . is synonymous with dengue . Human behaviours and activities , as well as demographic , social and possibly climate changes also contribute greatly to the increased incidence and geographical spread of the disease . This has led to the introduction of Communication for Behavioral Impact ( COMBI ) . Although community-wide effort is the key to eradicate dengue , commitment and participation at individual level such as emptying flower pots and practice of regular removing of water collecting containers and rubbish from their homes , play equally critical role . The Health Belief Model ( HBM ) is one of the most widely used social cognition models to predict health behaviours . The HBM posits that individual's health behaviour is determined by four main elements: i ) consideration of the likelihood ( susceptibility ) ; ii ) consideration of the seriousness ( severity ) of illness; iii ) perceived benefits of taking health action; iv ) perceived barriers to taking health action . These four perceptions are elements that determine the readiness to take action , and are activated by: i ) cues to action and ii ) self efficacy [5]–[6] . There have been a handful of studies that have specifically applied the HBM in attempts to understand perceptions of risk and sustained dengue prevention [7]–[9] . HBM has also been used as a framework for understanding how to effectively structure health communication messages in order to change individual behaviour to prevent dengue [10] . To date , exploration of the constructs of HBM in the context of dengue prevention in the majority of research literature is limited to the realm of quantitative research . Only a few qualitative studies have examined health perceptions of DF [7] , [11]–[12] . These studies underline the importance of public health beliefs in determining preventive measures against DF . However , these studies have been limited by lack of in-depth explorations of the main four dimensions of the HBM and its specific link to dengue prevention practices . Additionally , within the study of health behaviours , knowledge about dengue and its association with preventive behaviours have not been adequately elucidated . Findings have shown that increased understanding of dengue virus transmission was positively attributed to better dengue prevention practices [13]–[14] . Nevertheless , there were evidences which imply that knowledge about DF does not always result in the adoption of recommended preventive behaviours [15]–[16] and thus , further in-depth understanding among those living in dengue endemic country is warranted . Due to its relative scarcity in the research literature , we undertook this qualitative study as a first step in gaining in-depth understanding of the meaning of dengue , dengue prevention and treatment-seeking behaviours in a dengue endemic country . The HBM was used as a framework to explore and understand dengue preventive health behaviours . Qualitative exploration of attitudes and beliefs using constructs from the HBM has the potential to develop an in-depth understanding of the complex interplay of psychological , social , cultural and individuals factors that give rise to differences in dengue related health behaviours .
Participation in this study was voluntary and all participants provided written informed consent . All information was collected anonymously and the outcomes were used for research purposes only . The study was approved by the Medical Ethics Committee of the University of Malaya Medical Centre , Kuala Lumpur , Malaysia ( MEC Ref No . 896 . 15 ) . Participants were members of the public in Malaysia . A sample of multi-ethnic Malaysian citizens with diverse educational and socio-economic backgrounds were recruited in the Klang Valley area of Malaysia , the locale with the highest annual incidence of dengue . The target participants were selected based on the convenience sampling method at the research area . The first step in the recruitment of focus group participants began with word of mouth referral and through personal contacts by research assistants . Once recruited , participants were screened for eligibility using the following inclusion criteria: i ) aged 18 years or over , ii ) willing and able to provide written informed consent , iii ) living in the Klang Valley and iv ) Malaysian citizen . Subsequently , the focus group participants were asked to refer us to other participants they knew who also met the inclusion criteria , such as their friends or acquaintances , via the snowball sampling method . A semi-structured focus group moderator's guide corresponding to the research questions was developed . This semi-structured guide allowed the moderator to pose questions that flowed from one issue to the next . The guide consisted of questions about i ) the meaning of DF and general knowledge about dengue prevention and treatment , 2 ) attitudes about dengue with probes on perceived severity and susceptibility of dengue , 3 ) prevention practices and barriers to prevention , and 4 ) treatment-seeking behaviours . Discussion probes based on HBM constructs were developed to facilitate discussion on barriers to prevention . Focus group discussions ( FGDs ) were conducted in community settings and at places that were convenient for the participants , such as their home or workplace . Groups were separated into the three main ethnic groups of Malaysia , Malay , Chinese and Indian and were conducted in the local languages of the participants . Besides ethnic diversity , participants representing a broad array of socio-economic backgrounds within each ethnic group were recruited to allow exploration of differences in groups from different socio-economic backgrounds . Written informed consent was obtained from all participants prior to the FGDs . All discussions lasted approximately 45 minutes , and were audio-taped and transcribed verbatim . FGDs conducted in languages other than English were forward translated into English . Notes taken by the moderator and note taker supplemented the audio-taped transcripts to glean details from the discussion . After group discussion , a brief questionnaire was administered to participants to gather information regarding their demographic backgrounds . The sampling process , data collection and analyses were continuous and iterative . All group discussions were immediately analysed and compared with the analysis of the previous discussions , which , in turn , further shaped the subsequent sampling , data collection and analysis . The FGDs continued until data saturation was reached or no new information was uncovered . After transcription and cleansing , the transcripts were converted to rich text format and imported into NVivo software ( QSR International Pty Ltd , Doncaster , Victoria , Australia ) for coding and categorising [17] . A directed content analysis approach was used to analyse the data and to identify key themes relating to the HBM constructs , knowledge and treatment-seeking behaviours [18] . The data were first segregated using predefined initial codes corresponding to the determinants of HBM . Subsequently open coding was employed to identify themes emerged under the concept , and more specific axial codes were thereafter developed from the open codes [19] . The codes were analysed using an interpretive descriptive method , where interpretative description goes beyond a mere description and aims to provide an in-depth conceptual understanding of a phenomenon [20] . Coding was performed by a single coder and the consistency of coding was assessed by intra-coder reliability . The researcher coded segments of the data at two different periods and the intra-coder reliability was calculated as the number of agreements divided by the total number of agreements and disagreements . The calculated intra-rater agreement was in the 90th percentile range . Finally , the data were interpreted and presented using the participants' own words as illustrations .
A total of 14 FGDs comprising 5 Malay , 5 Chinese and 4 Indian groups were conducted between 16th December , 2011 and 12th May , 2012 . Each FGD was composed of between 5 and 8 participants of the same ethnic group ( total 84 participants ) . The mean ( ± SD ) age of the sample was 39 . 8 ( ±15 . 8 ) years , age range 21 to 70 years old . The demographic distribution of the study sample is shown in Table 1 . Most of the study participants had at least a high school education . The FGDs comprised housewives , students , unemployed and employed persons of various occupational categories in managerial , professional and technical-unskilled workers . Among all participants in the FGDs , 7 persons reported that they had dengue experience . The average time for one FGD was approximately one hour . All the groups contributed valuable and important information on all the issues being discussed during the discussions . Table 2 summarizes the major themes derived from the FGDs . Some participants described DF as “dangerous” but were not clear about how dengue can lead to death . The vast majority of participants knew that DF fever is caused by mosquitoes . Participants above the age of 60 years with low levels of education had limited knowledge that DF is specifically caused by a mosquito infected with the dengue virus . Many were unable to differentiate between DF and DHF , including some of the highly educated participants . Those who knew about DHF described DHF as “difficult to be cured” . The most salient theme emerging across the focus groups was the notion that the dengue virus will not spread to a person with a strong immune system when bitten by an infected mosquito . Alternatively , they viewed people with a weak immune system will not be able to resist dengue infection after being bitten by a dengue mosquito . The focus group participants were easily able to identify the breeding source of mosquitoes . Almost all knew that dengue mosquitoes breed in stagnant water and that they should frequently check and remove stagnant water to prevent mosquito breeding . The majority noted that the signs and symptoms of dengue was a raised body temperature or prolonged fever . Some knew and cited that DF can be confused with other febrile respiratory illnesses as they share some of the same symptoms . Participants with a higher education were more likely to be able to correctly identify other symptoms such as rashes , joint pains and muscle pain . Dengue-experienced participants had relatively greater knowledge of the signs and symptoms than those who never had dengue . Overall , there were no differences between groups of different ethnic composition in knowledge about dengue and dengue prevention . When the study participants were asked to share their knowledge regarding treatment for dengue , there appeared to be a tendency for participants of various ethnic and demographic levels across all discussion groups to mention the efficacy of 100Plus ( a brand of carbonated isotonic sports drink ) in treating DF . Although participants in general did not know the mechanisms of action of 100Plus in treating DF , many appeared to have heard of its efficacy through word of mouth . Other commonly mentioned treatments for DF were extracts or juices of vegetables or fruits . . Across the focus groups , a considerable number of participants reported that they had heard of extract from papaya leaves as effective for curing DF . Many also cited that they had heard of watermelon juice as another cure for DF . Several other participants mentioned that bitter gourd juice has also been used to treat DF . Frog soup was commonly cited by Chinese participants , whereas the Indian and Malay participants were more likely to have heard that crab soup helps to relief symptoms of DF . A considerable number of Chinese participants strongly believed that porcupine bezoar stone is effective in treating DF . Some of the focus group participants implied that the natural remedies that they cited have been commonly used as treatment for dengue instead of as a relief of DF symptoms . Some of the responses implied that patients have been cured of dengue by these natural remedies . As one woman said , It was also revealed that among the participants who knew or had heard of the efficacy of relieving DF symptoms using natural remedies , or traditional or unconventional treatments , they tended to have preconceived ideas that dengue is not as dangerous as other diseases that warrant professional medical treatments . The perception of severity of dengue described by the participants fell into two main themes: serious or highly deadly , and not a threat . Participants who described dengue as serious and highly deadly comprised approximately one-third of the total number of FGD participants , most of whom knew of neighbours , friends or counterparts who had died from dengue . They viewed dengue as a fast killer illness and the person that they knew or had heard about who had contracted dengue died within a few days after being admitted to the hospital . The remaining FGD participants who viewed dengue as not a threat were relatively younger , mainly between the ages of 18 and 35 years old . They viewed that dengue is not dangerous if a person seeks treatment early and perceived that death only occurs among those who do not seek proper medical care . The focus group participants reported a mixed view when queried on their perceived susceptibility of contracting dengue . They viewed dengue as very common in Malaysia and dengue infection as widespread in many areas . A minority of the focus group participants perceived little or no chance that they would contract dengue . They believed that dengue occurs because of bad luck , chance , fate or uncontrollable factors . They reasoned that dengue is unlike other contagious or infectious diseases where a person is at a particularly high risk of becoming infected upon contact with an infectious agent on the body surface . Only the Aedes mosquito and not all mosquito species can transmit dengue , and even if a person is bitten by an Aedes mosquito , only dengue virus infected mosquitos will cause dengue . Some participants stated that their risk of getting dengue was low as they were confident that the mosquitoes in their area were not Aedes mosquitoes . Furthermore , some viewed that if a person has a strong body defense , a dengue virus infected mosquito will not cause DF in the person . There were also several participants , a majority of whom were young male , who were confident that they have strong body defenses against dengue and will not contract dengue . They viewed that many people who contracted dengue were because their body defenses were not as strong . An elderly participant noted that mosquitoes prefer young children over elderly people . And that the hardened skin layer of an elderly serve as barrier against mosquito bites . When asked about measures to avoid mosquito bites , personal protective measures mentioned by the participants were the use of insecticide sprays , electric rackets , mosquito coils , electric vaporising mats , a mosquito net around bed or installed in windows , protective clothing ( wearing a long-sleeved shirt and long trousers when they go to mosquito-infested areas ) , and avoiding being outdoors at dusk and early evening . Concerns about undesirable hazards relating to mosquito coil smoke or electric mat vapour inhalation were expressed across the focus groups and posed barriers to their consistent use . Many participants told of their preference for non-chemical control alternatives or natural methods to repel mosquitoes , as implied by the following quotations: Various means of destroying mosquito breeding were also cited , such as frequent checking and removing stagnant water from containers in their homes and covering water containers to reduce mosquito breeding . When the focus group participants were probed using a series of questions regarding their barriers to dengue preventive practices , four themes emerged during coding . The barriers to sustained dengue prevention that emerged across all the focus groups of different ethnic composition , in the order of the most commonly appear themes , were; i ) lack of self-efficacy , ii ) lack of perceived benefit , iii ) low perceived susceptibility , and iv ) unsure perceived susceptibility . A perceived lack of self-efficacy emerged in almost all focus groups and across all ethnic and demographic characteristics . The participants admitted to the challenge of constantly keeping the environment clean and free of mosquito breeding sites . Findings from the focus groups indicated that most participants , regardless of ethnicity , failed to constantly change stagnant water in pots and vases and check for mosquito breeding sites . Participants reasoned that they either forgotten or lazy to practice . Many revealed that if dengue cases are reported , the community gears up and cleans the surrounding environment . Taken together this implies lack of self efficacy and low perceived severity of outbreak influence prevention practices . The lack of perceived benefit manifested itself as a perceived lack of control of their chance of getting dengue . The focus group participants viewed that prevention practices may not minimise their chances of getting dengue . They reasoned that the root cause for the spread of dengue is beyond their control . A participant whose home is a mosquito-affected area due to its location next to a forest said , Some stated that despite good self-practices to prevent dengue infection around their homes , such behaviour does not yield benefit as there is lack of a concerted effort from neighbours or other community members . As two participants stated , Low perceived susceptibility manifested itself as a perceived lack of changes of getting dengue even if they are in a high risk area or exposed to mosquito bites . A participant responded , when asked about the perceived susceptibility of getting dengue: Uncertainty about how susceptible a person is to contracting dengue has also emerged as a theme , such that some participants felt that they were not in total control in avoiding dengue . One such participant explained that despite a good dengue preventive practice at home , it is still possible to contract dengue elsewhere in other areas , and therefore practicing preventive measures may not minimise the chances of contracting dengue . Dengue cases reported in the community appeared as cue to action in seeking immediate medical attention if they fall sick , as illustrated in the following quote Most participants stated spontaneously that people who are ill with symptoms of suspected dengue infection should seek modern medical treatment . All the participants who had dengue experience reported that they sought modern medical treatments . While the majority emphasized the importance of modern medical treatment , several participants noted that , to some extent , it was necessary to simultaneously try various alternative treatments as they knew there are no specific treatments for dengue . Many related their experiences of how natural remedies do indeed have a healing effect against dengue . As one participant stated , When participants who had never experienced dengue were asked for their perceived treatment-seeking practices for dengue , many of the participants' narratives suggested that there was an interest and belief in effective means for seeking modern medical care . Although the majority favoured modern over traditional practices or folk medicines in treating dengue , a considerable number of participants noted that they were likely to try various alternative treatments if they were suspected to have dengue . Across the focus groups , most of the participants who had never had dengue unanimously opined that folk medicines or natural remedies should be used in addition to conventional Western practices . The participants' reasons for using folk medicines and natural remedies for dengue can be classified into three main categories , ordered from most to least frequently occurring themes: i ) perceived helpfulness , ii ) trust of natural treatment and iii ) pragmatic to use . The narratives from the focus group participants suggested that high perceived helpfulness or efficacy of alternative treatment was related to word of mouth from local communities that had experience DF . Despite the lack of scientific evidence to support the efficacious claims of these natural remedies , also referred to as folk medicines , many participants reported that they have heard from others that the treatments effectively heal DF and therefore , they have a strong belief in the effectiveness of these treatments . These themes particularly emerged from focus group participants of lower educational level . The second theme , trust of natural treatment , derived from the notion that natural treatments have no potential harmful side effects . Many of these are home remedies that are easily available and prepared at home . One participant stated , Thirdly , the pragmatic use of natural and traditional therapies , a theme that emerged in the focus groups with higher educated participants , largely derived from the notion of knowing that there is no specific cure for the dengue . One participant stated that it is sensible to turn to complementary medicines and natural remedies as only intravenous drips were given if a person is admitted to hospital . He said ,
From the present study , several conclusions can be inferred . Firstly , with regard to knowledge about dengue , the study found that despite good knowledge about DF and prevention and control of dengue , preventive practices to control the vector were influenced by the main constructs of health belief in the HBM . The evidence from this study suggests that some basic information on the pathophysiology of DF and DHF/DSS may be beneficial . The study found that low perceived susceptibility forms part of the threat for young adults and the elderly . Reasons for the low perceived susceptibility to dengue infection emerged as two themes , namely perceived natural ability to withstand infection and low risk of coming into contact with Aedes mosquitoes . Self-efficacy has been found to be a critical barrier to self-regulatory prevention practices which could be enhanced with a consistent message about dengue . As perceived susceptibility and severity influences self-efficacy , education should be specific about high vulnerability to dengue and severity of DHF/DSS and the importance of avoiding secondary dengue infection . The current findings add substantially to the understanding of health beliefs in prevention and control of dengue in dengue endemic region . Therefore , the HBM factors could be used as the theoretical basis of a health educational programme on dengue fever prevention and control in Malaysia . Unconventional treatment practices , including using home remedies and traditional medicine distinctive to cultural beliefs are prominent , thus warrant consideration when designing education messages of dengue for the region . In short , results obtained may be used to contextualize barriers to , or facilitators of , the adoption of dengue prevention behaviours in Malaysia . | In-depth understanding of health beliefs and behaviors may provide insights into sustainable community-based dengue prevention and control . This study uses qualitative method to explore dengue prevention and treatment-seeking behaviours . Focus group discussions were conducted with Malaysian public of various demographic backgrounds in Klang Valley , Malaysia . General knowledge about dengue fever was good but many lack of knowledge of dengue haemorrhagic fever ( DHF ) and dengue shock syndrome ( DSS ) . Many had low perceived susceptibility of getting dengue . Barriers to prevent dengue were categorised as low self efficacy to execute preventive measures , perceived lack benefit of individual preventive measures , and unsure susceptibility of getting dengue . Low perceived benefit of continued dengue prevention practices was a result of lack of concerted action against dengue in their neighborhood . Traditional medicinal practices , though unproven , were common and viewed as efficacious . The findings add valuable insights into how health beliefs could affect dengue prevention and control in endemic region . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"social",
"and",
"behavioral",
"sciences",
"sociology",
"of",
"knowledge",
"neglected",
"tropical",
"diseases",
"sociology",
"social",
"research"
] | 2013 | Health Beliefs and Practices Related to Dengue Fever: A Focus Group Study |
Even in the post-genomic era , the identification of candidate genes within loci associated with human genetic diseases is a very demanding task , because the critical region may typically contain hundreds of positional candidates . Since genes implicated in similar phenotypes tend to share very similar expression profiles , high throughput gene expression data may represent a very important resource to identify the best candidates for sequencing . However , so far , gene coexpression has not been used very successfully to prioritize positional candidates . We show that it is possible to reliably identify disease-relevant relationships among genes from massive microarray datasets by concentrating only on genes sharing similar expression profiles in both human and mouse . Moreover , we show systematically that the integration of human-mouse conserved coexpression with a phenotype similarity map allows the efficient identification of disease genes in large genomic regions . Finally , using this approach on 850 OMIM loci characterized by an unknown molecular basis , we propose high-probability candidates for 81 genetic diseases . Our results demonstrate that conserved coexpression , even at the human-mouse phylogenetic distance , represents a very strong criterion to predict disease-relevant relationships among human genes .
In the last two decades , positional cloning has been remarkably successful in the identification of genes involved in human disorders . More recently , our ability to map genetic disease loci has strikingly increased due to the availability of the entire genome sequence . Nevertheless , once a disease locus has been mapped , the identification of the mutation responsible for the phenotype still represents a very demanding task , because the mapped region may typically contain hundreds of candidates [1] . Accordingly , many phenotypes mapped on the genome by linkage analysis are not yet associated to any validated disease gene ( 850 OMIM entries for phenotypes with unknown molecular basis had at least one associated disease locus on July 2nd , 2007 ) . Therefore , the definition of strategies that can pinpoint the most likely targets to be sequenced in patients is of critical importance [1] . Many different strategies have been proposed to prioritize genes located in critical map intervals . Some of the methods so far developed rely on the observation that disease genes tend to share common global properties , which can be deduced directly by absolute and comparative sequence analysis [2] . However , most of the available prioritization strategies are based on the widely accepted idea that genes and proteins of living organisms deploy their functions as part of sophisticated functional modules , based on a complex series of physical , metabolic and regulatory interactions [3] , [4] . Although this principle has been extensively used even in the pre-genome era to identify the critical players of many different biological phenomena , the present availability of genome-scale information on gene function , protein-protein interactions and gene expression in different experimental models allows unprecedented opportunities for approaching the prioritization problem with greater efficiency . In theory , the use of functional gene annotations would represent the most straightforward approach for candidate prioritization . However , although this strategy may be very useful in selected cases [5] , [6] , at the present stage it has clear limitations , either because it overlooks non-annotated genes [6] , [7] or because it is not evident how the annotated functions of the candidates relate to the disease phenotype . Therefore , computational methods less biased toward already consolidated knowledge , may have strong advantages [1] . In particular , protein-protein interaction maps and gene coexpression data from microarray experiments represent extremely rich sources of potentially relevant information . Recently , the direct integration of a very heterogeneous human interactome with a text mining-based map of phenotype similarity has allowed the prediction of high confidence candidates within large disease-associated loci [8] . Although this approach is highly efficient , it is clearly not exhaustive because very close functional relationships between genes and proteins are possible in the absence of direct molecular binding . In addition , the protein-protein interaction space is currently under-sampled and many genuine biological interactions have not yet been identified in experiments . Conversely , high-throughput experiments are known to result in a large fraction of false positives . The consistently low overlap of protein-protein interactions between large-scale experiments , even when the same proteins are considered , is testament to these problems [9] . Finally , many of the known protein-protein interactions have been ascertained through low-throughput experiments and are thus strongly biased towards better-studied proteins [10] . Since genes involved in the same functions tend to show very similar expression profiles , coexpression analysis could be a very powerful approach for inferring functional relationships , which may correlate with similar disease phenotypes . Accordingly , global analyses have shown that genes highly coexpressed across microarray experiments display very similar functional annotation [11] . However , with notable exceptions [12] , so far the coexpression criterion has not been employed very successfully for the prediction of genetic disease candidates and has been used to this purpose only in combination with other independent evidence [5] , [13] . The noisy nature of high-throughput gene expression datasets may represent one of the possible explanations for this shortcoming . Moreover , even when the coexpression of two genes is reproducibly observed under a high number of experimental conditions , this does not necessarily imply that the genes are functionally related . For instance , extensive meta analysis of microarray data across different species has revealed that neighboring genes are more likely to be coexpressed than genes encoded in distant genomic regions , even if they are not functionally related in any obvious manner [14] , [15] . Phylogenetic conservation has been previously proposed as a very strong criterion to identify functionally relevant coexpression links among genes [16] , [17] . Indeed , significant coexpression of two or more orthologous genes is very likely due to selective advantage , strongly suggesting a functional relation . Therefore , conserved coexpression could be a much stronger criterion than single species coexpression to relate genes involved in similar disease phenotypes . In this report we show that conserved coexpression and phenome analysis can be effectively integrated to produce accurate predictions of human disease genes . Using this approach we were able to select a small number of strong candidates for 81 human diseases , corresponding to a wide spectrum of different phenotypes .
We have generated two human-mouse conserved coexpression networks ( CCN ) , based on cDNA and oligonucleotide microarray platforms , respectively . In the first case , the starting data were log-transformed values of human and mouse ratiometric experiments , downloaded from the Stanford Microarray Database ( SMD ) [18] ( 4129 experiments for 102296 EST probes for human and 467 experiments for 80595 EST probes for mouse ) . The resulting network will be referred to as ‘Stanford’ in the following ( Text S1 ) . In the second case ( ‘Affy’ , Text S2 ) , the network was based on previously described series of normal tissue Affymetrix microarray experiments from human [19] ( 353 experiments corresponding to 65 different tissues for 46241 probe-sets associated to a known gene ) and mouse [20] ( 122 experiments corresponding to 61 tissues for 19692 probe-sets ) . In the human case , the Affymetrix experiments corresponding to the same tissue were averaged to compensate for the different number of replicates available for the various tissues . In both cases , we used the same procedure to generate a final CCN . In particular , we first generated single species gene coexpression networks ( SCN ) and then integrated them on the basis of human-mouse orthology , as detailed below . SCNs were generated by first calculating the Pearson correlation coefficients of every row in the expression matrix ( cDNA probe or Affymetrix probe-set ) with all other rows . A directed edge was established from row r1 to row r2 if r2 fell within the top 1% rows in terms of correlation with r1 . The threshold was first chosen on the basis of a previous study , showing that such 1% interval is most significantly enriched in terms of functionally relevant coexpression [16] . Moreover , we confirmed that using a more stringent 0 . 5% threshold results in strongly reduced sensitivity ( data not shown ) . These directed networks where then converted into undirected SCNs by mapping the rows to the corresponding Entrez Gene identifiers [21]: an edge is established between two Entrez gene IDs G1 and G2 if there is at least one edge from a row assigned to G1 to a row assigned to G2 and vice versa . The correspondence between probe-sets and Entrez gene IDs for the SMD data was established using Unigene ( build 190 and 152 for human and mouse , respectively ) . For Affymetrix data we used the annotation files provided by Affymetrix for each platform . Finally , CCNs were built from SCNs by mapping every Entrez Gene identifier to the corresponding Homologene cluster ( build 55 ) and retaining only the cases in which a one-to-one correspondence could be established between the human and mouse Entrez gene IDs appearing in the SCNs . Genetic disease phenotypes described in OMIM where correlated on the basis of MimMiner [22] . MimMiner assigns a similarity score to all pairs of OMIM phenotype records , based on the text mining analysis of their phenotype descriptions [22] . Two phenotypes were defined to be similar if their score was at least 0 . 4 , because biologically meaningful relationships were mostly detected in phenotype pairs with a similarity score equal or greater than this value [22] . About 1% of all possible pairs of phenotypes included in MimMiner pass this threshold . The analysis of the CCNs was based on the construction of coexpression clusters , defined as a given gene ( the center of the cluster ) plus its nearest neighbors in the conserved coexpression network , thus obtaining one cluster for each gene . The prevalence of genes joining functionally related genes in the CCNs was tested by analyzing the prevalence of Gene Ontology terms within coexpression clusters , compared to the same prevalence in randomized coexpression clusters . We counted the number of coexpression clusters for which at least one Gene Ontology term was significantly overrepresented ( P-value less than 10−4 with exact Fisher test ) , and compared this number with the same number averaged over 100 randomized CCNs . This was done separately for the Affy and Stanford networks , and the results are shown in Figure 2A . The overlap between CCNs and protein interaction networks was evaluated by downloading the list of known interactions between human proteins from HPRD [23] , [24] . To take into account the different experimental methods on which the HPRD interactions are based and their varying degree of reliability , we analyzed separately in-vivo , in vitro and yeast double hybrid interactions . In each case , separately for the Affy and Stanford networks , we compared the overlap between the CCNs and the protein interaction network to the same overlap averaged over 100 randomized CCNs . The results are shown in Figure 2B . Finally to verify whether the CCNs were enriched in edges joining genes causing similar phenotypes , we constructed a network of human genes in which an edge was placed between every pair of genes known to be involved in the same disease or in diseases with MimMiner similarity score at least equal to 0 . 4 . The mapping between OMIM phenotypes and genes known to cause them was obtained from Ensembl , version 45 [25] . We then evaluated the overlap between this network and the CCNs , again compared to the same overlap averaged over 100 randomized CCNs ( Figure 2C ) . The lists of genes contained in OMIM loci of unknown molecular basis were obtained from Ensembl , version 45 [25] . To identify likely candidates for a given disease-associated genomic locus , we first extracted from the networks disease-relevant conserved coexpression clusters . A cluster was considered relevant to a given disease d if it contained at least two genes experimentally known to cause phenotypes similar to d . These clusters will be called ‘disease clusters’ in the following . The genes of the disease clusters , which are also contained in the map interval of the locus associated to d , were retained for scoring as described below . The size of the coexpression clusters and of the loci are very heterogeneous: coexpression clusters contain between 2 and 186 genes while loci associated to OMIM diseases with unknown molecular basis vary between 3 and 2153 genes . Therefore the genes selected in the previous step were assigned a probabilistic score based on the null hypothesis in which the coexpression clusters are random sets of genes . The score is essentially the probability that the two events leading to the identification of a candidate gene for a disease occur by chance . The two events are ( 1 ) the presence in a coexpression cluster of at least 2 genes known to be involved in phenotypes similar to the disease in question and ( 2 ) the presence in the same coexpression cluster of a gene located within the locus relevant to the disease . First , we computed a P-value p1 for associating a cluster to the given disease by chance . This P-value is given by the cumulative hypergeometric distribution considering: the number RDC of genes linked to similar phenotypes that have been found in the cluster ( at least 2 to associate the cluster with the disease ) ; the number Rall of genes in the network that are linked to similar phenotypes; the number GDC of genes in the cluster; and the total number of genes Gall in the network:Second , we computed the P-value p2 of the overlap between the disease cluster and the genetic locus associated to the disease . This P-value is given by the cumulative hypergeometric distribution considering: the number LDC of genes in the locus that are also present in the given disease cluster; the number Lall of genes in the locus that are present in the network; and GDC and Gall are defined as above:In the null hypothesis , associating a cluster with a disease and finding a gene in the cluster that belongs to the appropriate orphan locus are independent events . Thus , the total score for a predicted candidate is given by the product p1 • p2 . When a candidate is found in more than one disease cluster , we consider only the lowest ( best ) score . The cutoff on such scores was determined by estimating the false discovery rate ( FDR ) for each possible cutoff using 100 randomized CCNs per dataset: The false discovery rate is defined as the ratio between the average number of predictions made using randomized CCNs and the number of predictions made using the real CCN , with the same cutoff . The cutoffs thus obtained for a 10% FDR were 4 . 49·10−6 for the Affy and 2 . 67·10−6 for the Stanford network , respectively . To estimate the precision of our procedure we used a leave-one-out strategy: For every gene experimentally associated to an OMIM phenotype we constructed artificial loci centered on the disease gene , and removed all associations between this particular phenotype and all the genes known to cause it , so as to simulate a phenotype with unknown molecular basis . The association between phenotypes similar to the one under examination and the corresponding genes was instead retained . In order to take into account the variability of locus sizes we constructed artificial loci of various sizes , by taking the disease gene plus the N closest genes on each side of the chromosome ( according to their start position on the chromosome ) . The artificial loci thus contained up to 2N+1 genes , but could contain fewer genes when the disease gene was close to one of the chromosome ends . In the following discussion , such artificial loci will be denoted as N20 , N50 , … N500 for locus sizes up to 41 , 101 … 1001 genes , respectively . This range of locus sizes was chosen based on the observed size distribution of orphan loci: OMIM loci for diseases with unknown molecular basis contain an average number of about 273 genes ( median: 180 genes ) . We considered the disease gene as correctly identified if it was selected as a candidate by our method with the same score threshold that we used for the orphan loci . The precision is defined as the ratio between the number of cases with correctly identified disease genes and the number of cases with at least one selected candidate , that is , the fraction of cases with selected candidates in which the disease gene was among the candidate list .
Conserved coexpression has been previously reported to be an efficient criterion to identify functionally related genes [16] , [17] . Therefore , to discover new relationships between human genes with a high potential relevance for disease phenotypes , we produced the two human-mouse gene coexpression networks described above , covering different platforms and experimental conditions . In particular , the Stanford network ( supporting file S1 ) was generated from data based on cDNA platforms , corresponding mostly to experiments performed on tumor cell lines . In contrast , the Affy network ( Text S2 ) was derived from normal tissue data , generated on Affymetrix platforms in two independent studies [19] , [20] . The Stanford network has 8512 nodes ( genes ) and 56397 edges , with an average connectivity of 13 . 2 edges per node . The Affy network is composed of 12766 nodes and 155403 edges , with an average connectivity of 24 . 3 edges per node . Both networks contain a large connected component of 2305 and 4122 genes , respectively , with some other small connected components containing only a few nodes . As expected from previous studies on gene coexpression networks [26] , the two networks are topologically similar to other biological networks , characterized by the existence of a few highly connected nodes ( hubs ) , but they show a connectivity distribution more similar to an exponential law than to a scale-free one ( data not shown ) . More importantly , if compared with 100 random permutations , both networks show a strong prevalence of edges between genes that are annotated to the same Gene Ontology ( GO ) keyword ( Figure 2A ) . This confirms that human-mouse conserved coexpression is a valuable criterion to identify functionally related genes . Accordingly , both networks show a highly significant overlap with protein-protein interactions reported in the Human Protein Reference Database ( HPRD ) [23] , [24] ( Figure 2B ) . Since many genes , such as those involved in basic cellular functions , should be coexpressed regardless of the particular experimental situation , we would expect the two networks to have many common links . Indeed , they share 2305 edges , between the 7332 common nodes , which represents a striking overlap ( the randomized Affy networks had on average 88 . 4 edges in common with the Stanford network , with a standard deviation of 8 . 7 ) . On the other hand , the large number of specific links that characterize the two networks indicates that they provide highly complementary information . Finally , to evaluate the capability of conserved coexpression to link genes involved in similar disease phenotypes , we measured in both networks the prevalence of links between genes associated to phenotypes with similar descriptions [22] . Interestingly , both networks showed a strong enrichment , if compared with the average number obtained from the randomized networks ( Figure 2C ) . We concluded that the two networks represent complementary resources that could efficiently predict disease-relevant relationships among human genes . The high prevalence of links between genes involved in similar disease phenotypes , observed in both networks , suggests that they could provide valuable information to identify likely candidates in mapped disease loci . Therefore , we devised an algorithm that integrates our CCNs with phenotype and mapping information to predict candidate disease genes in large genomic regions ( Figure 1 ) . As detailed in Materials and Methods , the procedure is based on the extraction from the network of the disease clusters , which we consider to be associated to a given disease since they contain at least two genes involved in similar phenotypes . The genes that are present in both the OMIM phenotype loci and the corresponding disease clusters are considered as candidates and assigned a score based on the size of the locus and of the disease cluster . Randomized runs allowed us to select a score threshold corresponding to a 10% FDR . To evaluate how our procedure could perform on the loci characterized by unknown molecular basis , we applied a leave-one-out strategy to all the Ensembl genes associated to at least one OMIM disease ID , by constructing artificial loci of variable size around each gene . We then measured the fraction of artificial loci for which we obtained at least one candidate ( Figure 3A ) , the average number of candidates found for these loci ( Figure 3B ) and the precision ( Figure 3C ) , defined as explained in Materials and Methods . Of the 1762 disease genes contained in OMIM , we could analyze 1426 , whose associated phenotypes are present in the MimMiner similarity matrix . The precision obtained obviously decreases when the size of the artificial loci is increased . However , it is interesting to notice that , while for the smallest artificial loci the precision was excellent ( about 68% for both networks ) , even with the largest artificial loci it was still remarkable ( 37 . 5% for the Affy and 29 . 1% for the Stanford networks , respectively ) . For N90 ( artificial loci with a maximum of 181 genes , very close to the median size of 180 genes for OMIM phenotype loci with unknown molecular basis ) , for example , the leave-one-out validation yielded at least one candidate for 47 . 8% ( Affy ) and 12 . 7% ( Stanford ) of the disease loci ( Figure 3A ) . In these cases , an average of 3 . 67 ( Affy ) and 2 . 17 ( Stanford ) candidates were returned ( Figure 3B ) , that contained the true disease gene in 49 . 3% ( Affy ) and 43 . 6% ( Stanford ) of the cases ( Figure 3C ) . That is , for both networks , when candidates were returned , the very short candidate lists contained the disease-causing gene with a probability of over 40% . We next determined how our method performs compared to other existing approaches . This comparison was based on the enrichment in correctly identified disease genes ( by leave-one-out ) with respect to randomized networks . In our case the fold enrichment is defined as the ratio between number of disease genes correctly recalled in the leave-one out procedure and the same number averaged on 100 randomized CCNs . For this evaluation we used the N90 artificial loci ( see Materials and Methods ) , which are closest in size to the actual orphan loci ( median size 180 ) . Fold enrichment values were computed for several published methods by Lage et al [8] ( see their Supplementary material ) . It must be noticed that the exact definition of the fold enrichment depends on the method under consideration , thus comparison must be taken with caution . However , Table 1 shows that our methods compares favorably with many previously published ones: when considering that , as discussed above , the use of gene expression data allows a less biased analysis compared to most other methods , we conclude that our approach provides a significant and original contribution to the identification of disease genes . The above results indicate that conserved coexpression can be efficiently combined with phenotype correlation data to provide high confidence candidates within genetic disease loci . Therefore , we applied our procedure to 850 OMIM phenotype entries with at least one mapped disease locus but unknown molecular basis . In Table 2 we provide the list of all the 321 candidates ( gene-locus pairs ) obtained with 10% FDR . We obtained predictions for 81 loci , 67 of which where only from the Affy network , 5 only from the Stanford and 9 from both . Interestingly , in 4 of the latter cases , the list of candidates from the two networks contained at least one common gene ( Table 2 ) . Notably , for three OMIM phenotypes ( 163000 , familial multiple nevi flammei; 268700 , saccharopinuria; 300195; AMMECR1 ) our predictions include the actual disease genes that , although not yet correctly annotated in OMIM , have been found to be mutated in patients ( see Table 2 ) . For 22 loci , at least one of the candidates obtained from either network was already known to be involved in phenotypes similar to those described for the locus . These genes represent the most obvious candidates and our results should be considered as further , independent evidence for their possible involvement in the disease . However , it must be noted that some of them were previously excluded , either by the direct identification of crossovers or by the negative results of mutation screenings . Nevertheless , since mutations have most likely been searched only within the annotated exons , we think that the decision to definitively rule out the involvement of such candidates should be taken cautiously . Moreover , even silent exonic mutations , although often considered innocuous polymorphisms , can have severe effects on proteins by disrupting splicing patterns [27] , [28] . In most cases only few candidates are given for a locus , thus providing extremely focused working hypotheses for the identification of the actual disease genes , which in many cases are made even stronger by the available sequence or functional information . For instance , one of the two candidates provided for the OMIM phenotype entry 607221 ( partial epilepsy with pericentral spikes , located on 4p15 ) corresponds to KCNIP4 ( Figure 4 ) . This protein has been show to specifically modulate the activity of Kv4 A-type potassium channels [29] , which are well known regulators of membrane excitability [30] and have been recently involved in epilepsy [31] . Another interesting example is given by the 605285 phenotype entry ( hereditary motor and sensory neuropathy , Russe type , mapped to 10q23 . 2 , a locus comprising only 26 candidate genes ) . The only prediction for this locus is gamma-synuclein ( SNCG ) , which is a very strong candidate both for the low P-value and for the known role of synucleins in neurodegenerative disorders [32] . Even when the number of candidates for a particular locus is substantially higher , our results may provide a strong restriction of the experimental search field , which can be further narrowed by additional evidences . For instance , the phenotype with OMIM ID 130080 ( Ehlers-Danlos syndrome , type VIII ) , is mapped to 12p13 , containing 277 genes . In this case , the Affy and Stanford networks provide 8 and 4 candidates , respectively . Interestingly , the candidate with the lowest associated P-value is the Alpha-2-macroglobulin precursor ( A2M ) , whose absence was previously reported in a patient with Ehlers-Danlos syndrome [33] . A second interesting protein for this locus is CD9 that is the only candidate provided by both networks and that is known to regulate collagen matrix organization by interacting with Beta1 integrin [34] . In general , given the highly stringent criteria that we adopted and considering that the starting data underlying the two networks are completely independent , we propose that the 4 common candidates ( Table 2 ) should be considered as those having the highest priority for experimental validation . Since a recent study has identified high confidence candidate genes by integrating protein-protein interaction with phenotypic information [8] , we evaluated the number of common predictions , and found that a candidate is proposed by both approaches for 7 loci . In 5 cases the candidate proposed by Lage et al . did not overlap with our predictions . The only , remarkable exception was Filamin C ( FLNC ) , which was found as a candidate on chromosome 7q for both the 608423 ( limb-girdle muscular dystrophy type 1F ) and the 603511 ( limb-girdle muscular dystrophy type 1D ) OMIM phenotype entries . Interestingly , the FLNC gene has been previously implicated in myopathy , but mutations have not been reported in families mapping to the above loci . Thus , our results can be considered as further supporting evidence , pointing to the actual involvement of this gene in limb-girdle muscular dystrophy .
In the present report we have shown that the integration of massive gene expression data with phenotype similarity maps can allow to efficiently identify high-probability candidates for many orphan human genetic disease loci , even when these comprise hundreds of genes . The comparison of the two different networks clearly showed that the Affy network is characterized by a much better signal/noise ratio than the Stanford network . Although this may be partially due to the different source material ( normal tissues in the first case , mostly tumor cell lines in the second case ) we think that this could also depend on the higher technical standards reached by oligonucleotide-based platforms . Nevertheless , it is also important to notice that the analysis of the Stanford network allowed the prediction of many candidates that were not obtained from the Affy network , indicating that different datasets can result in complementary predictions . Interestingly , in both cases the gene coexpression criterion proved to be much more effective than it could be expected from previous work , where it has been mostly used in combination with other high-throughput information sources . We think that these results strongly underscore two critical points . The first is the importance of using a restrictive filter to select biologically relevant coexpression links , such as our phylogenetic filter selecting links which are under selective pressure and therefore more likely to imply functional relationships . The second is the usefulness of systematic phenotype analysis methods , which may capture disease similarities that could easily escape human operator-based approaches . Although very significant under the statistical point of view , the overlapping between conserved-coexpression links and physical protein-protein interaction data appeared to be rather limited in absolute terms , strengthening the idea that these criteria may cover partially overlapping subsets of the functional interaction space . Our results strongly suggest that , in most of the cases , requiring the concordance of coexpression data and protein-protein interaction data may worsen , instead of improving , the performances of both methods . Therefore , we envisage the independent use of both types of evidence to predict functional relationships and candidate disease genes . However , in the limited cases for which these approaches provide convergent results , they can be used as strong additive evidence . In conclusion , we propose that our method and our list of candidates will provide a useful support for the identification of new disease-relevant genes . | One of the most limiting aspects of biological research in the post-genomic era is the capability to integrate massive datasets on gene structure and function for producing useful biological knowledge . In this report we have applied an integrative approach to address the problem of identifying likely candidate genes within loci associated with human genetic diseases . Despite the recent progress in sequencing technologies , approaching this problem from an experimental perspective still represents a very demanding task , because the critical region may typically contain hundreds of positional candidates . We found that by concentrating only on genes sharing similar expression profiles in both human and mouse , massive microarray datasets can be used to reliably identify disease-relevant relationships among genes . Moreover , we found that integrating the coexpression criterion with systematic phenome analysis allows efficient identification of disease genes in large genomic regions . Using this approach on 850 OMIM loci characterized by unknown molecular basis , we propose high-probability candidates for 81 genetic diseases . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"computational",
"biology/metagenomics",
"computational",
"biology/comparative",
"sequence",
"analysis",
"computational",
"biology/molecular",
"genetics",
"computational",
"biology"
] | 2008 | Prediction of Human Disease Genes by Human-Mouse Conserved Coexpression Analysis |
De novo mutation is highly implicated in autism spectrum disorder ( ASD ) . However , the contribution of post-zygotic mutation to ASD is poorly characterized . We performed both exome sequencing of paired samples and analysis of de novo variants from whole-exome sequencing of 2 , 388 families . While we find little evidence for tissue-specific mosaic mutation , multi-tissue post-zygotic mutation ( i . e . mosaicism ) is frequent , with detectable mosaic variation comprising 5 . 4% of all de novo mutations . We identify three mosaic missense and likely-gene disrupting mutations in genes previously implicated in ASD ( KMT2C , NCKAP1 , and MYH10 ) in probands but none in siblings . We find a strong ascertainment bias for mosaic mutations in probands relative to their unaffected siblings ( p = 0 . 003 ) . We build a model of de novo variation incorporating mosaic variants and errors in classification of mosaic status and from this model we estimate that 33% of mosaic mutations in probands contribute to 5 . 1% of simplex ASD diagnoses ( 95% credible interval 1 . 3% to 8 . 9% ) . Our results indicate a contributory role for multi-tissue mosaic mutation in some individuals with an ASD diagnosis .
DNA is constantly exposed to natural and artificial mutagenic processes and therefore continually develops lesions and undergoes subsequent error-prone repair . In multi-cellular organisms , these mutations may arise at any time during development resulting in diverse organismal and cellular phenotypes , including disease . The severity of these phenotypes is dependent upon not only the particular genetic change but also the affected cell type and time in development at which the mutation occurs . Obligatory somatic disorders , in which prenatally lethal germline mutations occur post-zygotically , are one extreme [1] . In contrast to obligatory somatic mutation , de novo mutation is thought to primarily occur in the parental germline , typically resulting in genetic variation that is heterozygous in every cell of an organism . Such mutation is de novo in the sense that it is below the limit of detection in a parental sample ( usually DNA derived from blood ) . An early report using comparative genomic hybridization indicated that large de novo copy-number variants are enriched in ASD probands [2] . From these results it was hypothesized , and subsequent microarray and whole-exome sequencing experiments have shown , that a substantial fraction of genetic liability arises de novo in every generation [3–15] . The exact developmental time at which de novo mutations occur however , is under active investigation . Some de novo variants discovered though whole-exome sequencing have properties consistent with mosaicism [8 , 11 , 16 , 17] . Recent experiments using high-depth targeted sequencing have indicated that eight of 27 likely causal variants in individuals with cortical malformations are present as mosaics , occasionally at very low alternate-allele read fractions ( AARF ) [18] . Mosaic mutations have been found to occur in single individuals of monozygotic twin pairs [19 , 20] . Furthermore , 6 . 5% of identified de novo mutations in individuals with severe intellectual disability occur as mosaics [21] . Here we show that de novo variation in a large whole-exome sequencing dataset is frequently mosaic and that such mosaic variation is likely to contribute to disease diagnoses in some affected individuals .
Obligatory somatic mutations typically occur in a localized fashion in tissues that share a common developmental origin . To examine the contribution of tissue-specific mutations in ASD , we generated whole-exome sequence data from paired postmortem frontal cortex ( n = 16 ) and heart ( n = 14 ) or kidney ( n = 2 ) samples from individuals diagnosed with ASD ( n = 12 ) and controls ( n = 4 ) . Sequence data were generated using the Illumina HiSeq platform with an average sequence depth of 95x across capture targets ( S1 and S2 Tables ) . Variants were detected using the somatic variant callers Strelka and Mutect [22 , 23] . These programs detect mutations unique to single tissues from paired samples . Analyzing frontal cortex/heart or frontal cortex/kidney pairs resulted in the identification of 373 mosaic variants in the 32 samples . However , validation experiments indicated that all potential mutation loci were homozygous for the reference allele ( i . e . all mutations chosen for validation were false positives; S1 Fig and S3–S6 Tables ) . Our findings agree with previous results that tissue-specific mosaic mutation in brain rarely occurs at the level of detection afforded by standard whole-exome sequencing experiments [24] . In next-generation sequencing data , reads supporting the alternate allele at variant sites are known to be under-represented due to biases against non-reference alleles [25] . Further , many variant callers explicitly assume a diploid model . Therefore , the extent to which existing germline variant callers accurately genotype mosaic mutations in unpaired samples is uncertain . To evaluate our ability to discover mosaic mutation occurring in single samples , we obtained the Illumina Platinum Genomes sequence including NA12878 , an individual for whom a high-confidence callset is available ( ERP001960 ) [26] . We then characterize the sensitivity of the GATK HaplotypeCaller through in silico mixture experiments ( S2 Fig ) . For this experiment , we utilized 200x sequence data from the Illumina Platinum Genomes ( ERP002490; See Materials and Methods ) . Sequence reads from NA12878 were mixed with sequence reads from her son NA12882 over regions known to harbor variants from the high-confidence GIAB callset . Mixtures were then subsampled to depths of 30x and 50x with random fractions of reads from N12878 and NA12882 . Variants were then called from the mixture and the sensitivity of the variant caller was assessed for variants known to be present in NA12878 but not NA12882 . Sensitivity was also assessed with different values of–ploidy argument which alters the expected AARFs of heterozygous variants . These results demonstrated that higher–ploidy settings improved sensitivity for low frequency mosaic variants at the cost of higher memory usage and longer runtimes ( S3 Fig ) . Given that mosaic variants may be identified with germline variant callers , we sought to determine the mosaic status of variants in the Simons Simplex Collection ( SSC ) , a large collection of simplex autism pedigrees [27] . Extensive phenotypic data and whole-exome sequence data have been generated for all members of the collection and two non-overlapping callsets have been generated from the SSC exomes [13 , 15] . To increase sensitivity for detection of mosaic variants we performed a complete re-calling of all samples in the SSC with a -ploidy 5 setting ( S4 Fig ) . Variant filtration was performed using the GATK’s variant quality score recalibration ( VQSR ) pipeline and de novo variants were identified using find_denovo with default parameters . This resulted in the identification of 6 , 408 de novo variants , of which 3 , 355 and 228 were present in the Iossifov or Krumm callsets , respectively and 2 , 825 were unique to our callset ( S7 Table ) . Average coverage of high quality sequence reads at positions with identified de novo variants was 94 . 6 , indicating that mosaic variants are likely to be accurately detected , when they occur . Of variants identified by Iossifov et al . or Krumm et al . but excluded from our callset the majority were filtered by VQSR ( S8 Table ) . The complete set of variants in the Iossifov or Krumm callsets , annotated with their inclusion or reason for exclusion from the current callset are listed in S9 and S10 Tables . We excluded fifteen families who had a child with more than 10 de novo mutations since these likely occur due to technical artifacts . After exclusion of these families , our callset contained 4 , 909 de novo variants . Variant effects were annotated using SnpEff and mosaic variants were identified from a binomial test with false-discovery protection using the Benjamini-Hochberg procedure [28] . This resulted in the identification of 1 , 036 mosaic variants at an FDR of 5% . Hereafter , we will refer to the variants identified as de novo but not mosaic as germline de novo variants . To ensure that the variants in our callset were present in the samples , we performed Sanger sequencing of 97 ( 47 mosaic and 50 germline de novo ) variants ( Table 1 , pre-filter ) . For all of our validation methods , we present both “detection precision” and “classification precision” , where applicable . We define “detection precision” as our precision for the presence of the variant in the sample , while we define “classification precision” as our precision for the presence of the variant in the sample and correct classification of the variant as either mosaic or germline de novo . Of the 97 reactions , sequencing was informative for 76 . The variant of interest was identified in 100% of samples when the variant was annotated as germline de novo . However , in samples harboring a mosaic variant , precision for the presence of the variant was modest ( 54% ) . Mosaic variants that failed validation were often called with few reads supporting the alternate allele and were frequently called uniquely in our callset . To improve downstream analyses , we made the conservative choice of requiring that identified mosaic variants be present jointly in our callset and in the Iossifov or Krumm callsets . This filter greatly improves the precision of our callset ( 100% for variant presence; Table 1 , post-filter ) with little change in sensitivity . However , the precision of the classification of the mosaic status of variants remained modest ( 68% ) . While Sanger sequencing provides an accurate assessment of the presence of de novo variants , examination of the chromatograms can provide only approximate estimation of mosaic status . To more accurately assess the mosaic status of the identified variants , we performed sequence read phasing of all identified de novo variants . Phasing of the potential mosaic variants relative to nearby inherited heterozygous variants using sequence reads may rigorously confirm the presence of mosaicism . This occurs when three parental haplotypes are inferred: a single haplotype from one parent ( e . g . having the minor allele of the neighboring SNP ) , and two haplotypes from the other parent ( e . g . having the major allele of the neighboring SNP ) which resolve into a haplotype with the mosaic allele and a distinct haplotype lacking the mosaic allele ( Fig 1 ) . We wrote a program called phase-mosaic to perform phasing validation ( see Materials and Methods ) . Of the variants passing filters , phasing was informative for 51 mosaic variants of which 29 were validated as mosaic ( 57%; Table 2 , Pre-filter ) . Mosaic variants identified by next-generation sequencing that failed phasing confirmation tended to be variants called at high depth with a high AARF . We suspect that these variants appear to be mosaic due to preferential capture of the reference allele during exome enrichment . To correct for this effect , we modified our criteria for the identification of mosaic variants to require that mosaic variants have AARF less than 34% . With these adjusted parameters , the precision of our classification of mosaic status improved to 87% ( Table 2 , Post-filter ) . Validation of mosaic variants was also performed using pyrosequencing ( S11 Table ) . Likely-gene disrupting ( LGD ) and missense variants in probands across a range of allele frequencies were chosen for pyrosequencing validation ( see Materials and Methods ) . Consistent with the post-filter results of Sanger sequencing and physical phasing , pyrosequencing validation demonstrated high precision for variant detection and variant classification ( Table 3 ) . Of all variants validated by an orthogonal sequencing technology , 16 were validated with multiple validation methods . The results were consistent , except for apparent inaccuracy in the classification of mosaic status by Sanger sequencing . After the application of filters , we identified a total of 4 , 095 de novo variants in our high-confidence callset , 221 of which were classified as mosaic . Based on our validation experiments , we estimate that our precision for the presence of the called variants is near 100% with the precision of the classification of mosaic variants measured at 87% or 82% by phasing or pyrosequencing , respectively . Of the variants in our final callset , 3 , 351 appear jointly in the current callset and the callset produced by Iossifov et al . while 228 appear jointly in the current callset and the callset produced by Krumm et al . ( Fig 2 ) . In the high confidence callset no mosaic mutations were identified that were shared between a sibling pair . To better understand the properties of variants in our callset , we examined the mutational spectra of the identified mosaic variants relative to germline de novo variants ( S12 Table ) . We find that mosaic variants have significantly more deletions than germline de novo variants ( Fisher’s exact test , p = 5 . 2e-4 ) . However , the rate of occurrence of other types of mosaic mutations is approximately equal to the rate of occurrence of the corresponding de novo mutation . The relative enrichment of mosaic mutations for deletions may indicate an increased rate of false-positive mutation as the identification of indels from next-generation sequence data is known to be difficult . However , our precision when validating mosaic mutation was quite high ( see above ) . We attempted validation of four deletions in the high-confidence callset using Sanger sequencing ( S7 Table ) . All four deletions were found in the sample and three of four were confirmed as mosaic . Besides false-positives , the enrichment of mosaic mutations may indicate a non-reference allele bias , where germline de novo deletions are occurring in the samples but are incorrectly classified as mosaic due to mapping errors . Phasing assessed the mosaic status of eight mosaic deletions , six of which were confirmed as mosaic resulting in a classification precision of 75% , slightly less than the overall classification precision of 87% from sequence read phasing . Therefore , inaccurate classification of the mosaic status of de novo deletions may contribute to the observed enrichment . An additional hypothesis is that the mechanism underlying mosaic mutation wholly or partly differs from that of germline de novo mutation and the relative enrichment of deletions may be attributed to these differences in underlying mechanisms . Previous studies have indicated that de novo mutations occur at higher rates in probands relative to controls leading to the implication of de novo variants as contributing to disease diagnoses [13 , 17] . We utilized our high-confidence mosaic variant callset to compare the rates of mosaic and germline de novo mutation in probands relative to unaffected siblings . Following the protocol of Iossifov et al . we defined regions of joint 40x coverage in children of quad families and extrapolated rates of mutation within these joint 40x regions to the entire capture region ( Fig 3; S13 Table ) . Consistent with previous results , we find that germline de novo LGD mutations are significantly enriched in probands relative to controls ( p = 0 . 001 ) . In addition we find that all classes of mosaic mutations are significantly enriched in probands ( p = 0 . 003 ) . Interestingly , we observe contribution to disease from all classes of mosaic variation , whereas the contribution of germline de novo variation to disease is primarily from LGD mutations . To account for errors in the classification of variants as either mosaic or germline de novo , we extended our model of contributory variation to include incorrectly classified variants . In this model , mosaic variants incorrectly classified as germline de novo account for a substantial portion of the genetic contribution of variants classified as germline de novo ( Fig 4 ) . In total , mosaic variation contributes to 5 . 1% of ASD cases ( 95% credible interval [CI] , 1 . 3% to 8 . 9% ) while all classes of germline de novo variation contribute to 5 . 6% of ASD cases ( 95% CI , 1 . 8% to 9 . 4% ) . The percent of contributory variants to total variants are measured as 6 . 0% ( 95% CI , 2 . 0% to 10% ) and 33% ( 95% CI , 9 . 6% to 54% ) for germline de novo and mosaic variants , respectively . While differences in the rates of mutation in affected individuals may implicate mutations in disease , it may also be the case that mutations in probands occur in more functionally conserved genomic regions . Using all of the mutations in our high-confidence callset , we test the hypothesis that mutations in probands occur at more conserved genomic regions . For this analysis , we use three measures of conservation: base-level conservation as measured by PhyloP , and gene-level conservation as measured by HomoloGene or ExAC ( S14 Table ) [29–31] . The gene-level conservation measures from ExAC and Homologene are complimentary as HomoloGene provides a measure of evolutionary conservation while ExAC provides a measure of conservation in extant human populations . We find that germline de novo LGD mutations occur at more highly conserved positions in probands relative to controls as measured by PhyloP score ( p = 0 . 048 , effect = 0 . 54 ) . For mosaic missense mutations , we observe a stronger effect ( 0 . 91 ) , although the test does not reach statistical significance due to the small sample size ( p = 0 . 179 ) . We find that germline de novo missense variants occur significantly more often in genes thought to be intolerant of loss-of-function mutation as annotated by ExAC ( p = 0 . 013 ) . While our analysis does not show that germline de novo missense mutations occur at significantly higher rates in probands relative to siblings ( p = 0 . 30 ) , germline de novo missense variants likely target genes less tolerant of functional mutation more frequently in affected individuals relative to their siblings . The initial publication of the SSC exome sequencing data demonstrated enrichment of mutations in specific classes of gene targets [13] . To find insight into the mutational mechanisms and functional consequences of mosaic mutation , we replicated this analysis ( with modification ) using our high-confidence callset . This analysis confirmed the significant enrichment of germline de novo LGD mutations from probands in FMRP targets , chromatin modifiers , and genes with known LGD mutations in intellectual disability or schizophrenia ( S15 Table ) . In addition , we observed enrichment of mosaic missense and LGD mutations in probands and siblings in genes involved in embryonic development ( 18 observed versus 12 . 6 expected for probands; 9 observed versus 6 . 7 expected for siblings ) , however this enrichment did not reach statistical significance ( p = 0 . 12 for probands; p = 0 . 30 for siblings ) . We also tested for overlap between genes targeted by mosaic missense and LGD mutations and a set of 107 genes that had been strongly implicated in ASD using the null-length model ( see Methods ) [32] . We found three of the 98 genes with mosaic missense or LGD mutations in probands have been previously implicated in ASD ( KMT2C , NCKAP1 , and MYH10 ) . The presence of the NCKAP1 was confirmed by Sanger sequencing , but did not confirm its mosaic status . However , the number of mosaic mutations in ASD genes does not reach statistical significance for enrichment in the set of 107 ASD genes ( 3 observed; 1 . 15 expected; p = 0 . 109 ) . Zero of 52 genes targeted by mosaic missense or mosaic LGD mutations in siblings were previously implicated in ASD .
There are three major conclusions from this study . First , we show that mosaic mutations occur frequently in individuals diagnosed with ASD and their unaffected siblings . We identify a total of 4 , 095 de novo mutations , of which 221 ( 5 . 4% ) are classified as mosaic . This is similar to previously reported estimates for the fraction of mosaic variants in individuals with intellectual disability [21] . In light of previous work demonstrating the presence of mosaic mutation in diverse body tissues we believe that the mosaic mutations we identified are not unique to blood but are dispersed throughout the body [18] . Although the early steps of our pipeline were performed explicitly to increase our sensitivity for mosaic mutation , many of our filtering steps were conservative and we likely underestimate the true fraction of mosaic mutations in the Simons Simplex Collection . Our filtering approach combined with recent improvements in variant detection algorithms likely accounts for most of the differences between our variant callset and the callsets published by Iossifov et al . and Krumm et al . [13 , 15] Second , we find that mosaic mutations are significantly enriched in probands relative to their siblings . Using our model of contributory variation we estimate that 33% of mosaic mutations contribute to 5 . 1% of ASD diagnoses . As mosaic mutations arise post-zygotically in only a fraction of the cells of an individual , we expect that these results have implications for the interpretation of twin studies , especially observed cases of phenotypic discordance between monozygotic twins . Third , we find that tissue-specific mosaic mutations do not occur in the paired samples at our limit of detection . Given the lack of publications on validated tissue-specific mutations in tissues without visual abnormality and the absence of brain-specific mutation in a centenarian [24] , we do not believe that this finding is unexpected . While a recent study reported tissue-specific mosaic mutation [33] , the results presented here include validation of detected mutations showing that , in our study , these were false positive findings . It is possible that tissue-specific mutations do contribute to ASD in at least some cases . However , discovery of such variation and its implication in disease may require larger numbers of samples or more sensitive approaches ( such as single-cell sequencing ) . Together , these results indicate that mosaic mutations are an identifiable subset of de novo mutation . As heritable factors that may arise in a single twin of a monozygotic pair [19 , 20] , contributory mosaic mutation implies some expected level of discordance between monozygotic twins due to heritable factors arising post-zygotically . Furthermore , high-confidence identification of contributory mosaic mutation in affected probands implies a lower risk of familial recurrence in some families .
Paired samples were obtained from the University of Maryland Brain and Tissue Bank as detailed in S1 Table . Individuals were diagnosed with ASD ( n = 12 ) or were controls; criteria for diagnosing ASD included the Autism Diagnostic Interview-Revised ( ADI-R ) , Childhood Autism Rating Scale ( CARS ) , and Autism Diagnostic Observation Schedule ( ADOS ) as detailed S1 Table . DNA was extracted from tissue dissections according to protocols in the QIAGEN Genomic DNA Handbook . Exonic regions were selectively captured using Agilent SureSelectXT Human All Exon V5 . Sequencing was performed at the Center for Inherited Disease Research at Johns Hopkins generating 100 bp sequence reads on an Illumina HiSeq . CIDRSeqSuite version 3 . 0 . 1 was used for processing of the raw data files . BCL files were converted to qseq format using Illumina’s BCL converter . qseq files were then demultiplexed and converted to FASTQ files using a custom demultiplexer . Paired-end alignment was performed using BWA aln to the 1000 genomes hg19/GRCh37 reference genome [34] . SAM files were sorted , converted to BAM , and duplicates were marked with Picard . GATK was used for local realignment and base quality score recalibration [35 , 36] . Quality metrics for these data are provided in S2 Table . Tissue-specific variants were called from paired samples using MuTect 2 . 7–1 for SNV discovery and Strelka 1 . 0 . 13 for indel discovery [22 , 23] . Input to these programs requires specifying a “tumor” and a “normal” sample . For each paired sample , variants were called twice so that mosaic variants were identified in both the brain and heart/kidney tissue . Validation of these variant calls was performed as indicated in S5 Table ( Targeted Sequencing 1 ) , with variants with the most severe functional effect selected for validation . To more carefully examine the properties of the mosaic variants called by MuTect , variants were recalled jointly in all samples using the GATK’s HaplotypeCaller in the “GENOTYPE_GIVEN_ALLELES” mode [35 , 36] . These variant calls were converted to a text based file format ( S3 Table ) and allelic noise was annotated as an additional quality metric . Allelic noise was measured as the fraction of reads supporting the alternate allele relative to the total number of reads in all samples genotyped as homozygous for the reference allele by the HaplotypeCaller . Samples with called somatic variants were excluded from the calculations of allelic noise . If multiple alternate alleles were present , only the highest alleleic noise was recorded . Using allelic noise and the quality metrics annotated by the GATK’s HaplotypeCaller , the overall quality of the variants was assessed manually . Validation of the highest quality variants was attempted and the results of the validation are shown in S5 Table ( Targeted Sequencing 2 ) and S6 Table . For visualization , the properties of variants including "BaseQRankSum" , "FS" , "MQ" , "MQRankSum" , "ReadPosRankSum" and "SOR" were examined in the Illumina Platinum Genomes ( see below ) and variants called by MuTect . These properties were collectively scaled , principal components analysis was performed and the original variant features were transformed into the principal components . The variants were then plotted along these principal components as shown in S1 Fig . 200x sequence data from NA12878 and NA12882 were downloaded from EBI ( ERP001775 ) . The specific runs chosen were ERR174324 , ERR174325 , ERR174326 , ERR174327 , ERR174328 , ERR174329 , ERR174330 , ERR174331 , ERR174332 , ERR174333 , ERR174334 , ERR174335 , ERR174336 , ERR174337 , and ERR174338 for NA12878 and ERR174347 , ERR174348 , ERR174349 , ERR174350 , ERR174351 , ERR174368 , ERR174369 , ERR174370 , ERR174371 , ERR174372 , ERR174373 , ERR174374 , ERR174375 , ERR174376 , and ERR174377 for NA12882 . Sequence data were aligned to the 1000 Genomes phase 2 human reference genome using BWA MEM version 0 . 7 . 9a and aligned sequence reads were sorted using SAMtools [34 , 37] . Aligned sequence data were then combined into single files and in silico mixing with subsampling was performed using submixbam ( https://github . com/DonFreed/submixbam ) version 290fda over GIAB high-confidence regions ( http://ftp-trace . ncbi . nih . gov/giab/ftp/data/NA12878/variant_calls/GIAB_integration/union13callableMQonlymerged_addcert_nouncert_excludesimplerep_excludesegdups_excludedecoy_excludeRepSeqSTRs_noCNVs_v2 . 19_2mindatasets_5minYesNoRatio_AddRTGPlatGenConf_filtNISTclustergt9_RemNISTfilt_RemPartComp_RemRep_RemPartComp_v0 . 2 . bed . gz; accessed Oct 14th 2015; currently available from ftp://ftp-trace . ncbi . nih . gov/giab/ftp/data/NA12878/analysis/GIAB_integration/ ) . A variant calling and assessment pipeline was written using Snakemake [38] . This pipeline called variants from mixtures using the GATK’s HaplotypeCaller version 3 . 4–46 . The sensitivity of the HaplotypeCaller for variants known to be present in NA12878 and absent from NA12882 was then evaluated using hap . py ( https://github . com/Illumina/hap . py ) . Scripts are available in S1 Code . Analysis of the data in the Simons Simplex Collection made use of cloud computing via Amazon Web Services ( AWS ) for variant calling ( GATK HaplotypeCaller ) , merging gVCFs ( GATK MergeGVCFs ) and genotyping ( GATK GenotypeGVCFs ) . Starcluster ( http://star . mit . edu/cluster/ ) was used for deployment and configuration of clusters of virtual machines on AWS Elastic Cloud Compute ( EC2 ) and a customized Amazon Machine Image ( AMI ) was created containing GATK 3 . 5–0 , samtools 1 . 2 , Python 3 . 5 . 1 , and nda_aws_token_generator version 20b72 ( https://github . com/NDAR/nda_aws_token_generator ) [35–37] . c3 . xlarge , r3 . xlarge and r3 . 2xlarge instances were used for variant calling , merging gVCFs and genotyping , respectively . During variant calling and merging , the available disk space on each node was used as a complex resource to aid in job allocation . With c3 . xlarge instances , ephemeral storage partitions were combined into single logical volumes using RAID 0 . During genotyping , node ephemeral disk partitions were combined into a single network attached storage volume using GlusterFS ( https://www . gluster . org/ ) . Aligned whole-exome sequence data from 8 , 950 individuals in the SSC was accessed through the National Database for Autism Research ( NDAR ) on Amazon Web Services Simple Storage Service ( AWS S3 ) ( https://ndar . nih . gov/study . html ? id=334 ) . We excluded 16 individuals from families 11366 , 11368 , 11377 and 11380 due to data processing issues . Variants were called using the GATK ( v . 3 . 5–0 ) HaplotypeCaller in gVCF mode with standard variant annotations and additional arguments -ploidy 5 , -A GCContent and–A AlleleBalance over NimbleGen EZ-SeqCap v2 . 0 targets with 50 bp of padding [35 , 36] . gVCFs of 20 families were combined using GATK MergeGVCFs resulting in 120 merged gVCF files . All gVCF files were genotyped across capture regions in parallel using the GATK GenotypeGVCFs command with the arguments -stand_call_conf 25 . 0 , -stand_emit_conf 20 . 0 along with the arguments used with the HaplotypeCaller as described above . The genotyping step had high memory requirements over some target regions , causing some jobs to fail even with 116 GB of memory allocated to the java virtual machine . Failed capture regions were repeated with the additional argument—max_alternate_alleles 5 . However , we excluded 53 capture targets due to persistent memory errors ( S16 Table ) . These capture targets were highly enriched for overlap with known simple repeats ( UCSC Simple Repeats Track in BED format; tested using BEDtools fisher; Fisher’s Exact Test , p < 0 . 00001 ) . Variant calls over each capture target were then concatenated and duplicate calls due to overlapping padded targets were removed . In addition to the variant annotations produced by the GATK , raw variants were annotated with the number of sequencing reads supporting the reference allele relative to total number of sequence reads . This information was added to the VCF’s INFO field as the annotation “AbHetUser” . Variants were filtered using the GATK variant quality score recalibration pipeline . The recommended parameters for whole-exome sequencing were used minus the–an QD parameter and with the additional parameter–an AbHetUser . These parameters were chosen for their superior sensitivity and specificity for validated de novo variants in the SSC ( S17 Table ) . SNPs were filtered with a sensitivity tranche of 99 . 3% while indels were filtered with a sensitivity tranche of 98% . De novo variants were identified using the tool find_denovo , a tool we wrote in the C programing language , with default parameters ( https://github . com/DonFreed/find_denovo ) . find_denovo identifies alleles which are present in children but absent from their parents . It then applies a number of filters including a minimum number of reads for all trio members ( 20 ) , a minimum number of reads supporting the alternate allele in the child ( 3 ) , a minimum phred-scaled confidence for the presence of the de novo allele in the child ( 20 ) and the absence of the de novo allele in the parents ( 20 ) , and a maximum number of individuals genotyped for the allele in the cohort ( 2 ) . De novo variant effects were then annotated using SnpEff [28] . Families 11060 , 11431 , 11628 , 11714 , 11905 , 12173 , 12230 , 12401 , 12456 , 12809 , 12879 , 13143 , 13949 , 14025 , and 14355 were excluded as more than 10 de novo mutations were observed in at least one child in the family . Mosaic variants were identified from de novo variants using the binomial test to examine the alternative hypothesis that the de novo allele is supported by significantly fewer sequence reads than expected from the read depth . We use p = 0 . 5 as the expected fraction of sequence reads supporting the de novo allele . p-values were adjusted using the Benjamini-Hochberg procedure with a FDR of 0 . 05 and variants with q < 0 . 05 were called mosaic . In the final callset we add the requirement that mosaic variants must have an AARF of less than 34% . In addition , mosaic variants that were identified uniquely in our callset and not in the callsets produced by Iossifov et al . or Krumm et al . were filtered . Variants identified as de novo in the Simons Simplex Collection were phased to nearby inherited variants to validate mosaic status and to determine the parental haplotype of the variant allele using phase-mosaic , a tool we wrote in Java and Python ( https://bitbucket . org/donald_freed/phase-mosaic , version f47bcd ) . For each identified de novo variant , sequence data 500bp upstream and downstream of the variant was downloaded to the local machine from AWS Simple Storage Service ( S3 ) for each member of the pedigree . Variants were then recalled using the GATK version 3 . 5–0 compiled with the VariantReadIds annotation . Phasing was then performed on the resulting VCF files . Regions of 40x coverage were defined for each individual in quad families using BEDtools genomecov–bga with the resulting BedGraph file converted to a BED file using a custom script ( S1 Code ) [39] . BEDtools was then used to intersect the 40x BED file for each member of a trio and the target capture file to produce a joint 40x BED file for the trio . Variants in the callset were annotated based on their presence or absence in the joint 40x region using custom scripts ( S1 Code ) . The length of the genome present in the joint 40x region was recorded for each child in a quad family . Finally , the rate of de novo mutation for each individual and each class of mutation was calculated from the size of the joint 40x region and the number of mutations in joint regions identified in the child . These rates were then extrapolated to the entire capture region . The mean and standard deviation of the rates observed in probands and siblings are reported in S13 Table . Iossifov et al . previously reported a model of de novo variation in which siblings have a baseline rate of de novo mutation while probands have the same baseline rate and additional mutation due to their affected status [13] . We expand this model to distinguish between germline de novo and mosaic variation while incorporating errors in classification of mosaic status . In siblings the observed rate of mosaic or germline de novo variation was modeled as the sum of correctly and incorrectly classified baseline variation . In probands the models included correctly and incorrectly classified contributory variation in addition to the baseline variation . Classification error rates were for siblings modeled as either incorrectly classified baseline variation over correctly and incorrectly classified baseline variation . For probands , classification error rates were modeled as incorrectly classified baseline and contributory variation over correctly and incorrectly classified baseline and contributory variation . These models were solved to obtain the rate and fraction of contributory variation using the observed rates of mutation and classification errors as measured by phasing validation . Classification error rates were calculated separately for probands and siblings and for germline de novo and mosaic classification . Uncertainty in classification error rates was modeled using the beta-binomial distribution with phasing validation results as model parameters . A 95% credible interval was obtained through 10 , 000 permutations with classification error rates obtained by random draws from their respective distributions . Variants were annotated with PhyloP conservation score , taxonomic conservation as reported by NCBI’s HomoloGene database , and the probability of null mutations being deleterious ( “pNull” ) as reported by ExAC [29–31] . BigWig files containing genome-wide PhyloP scores were downloaded from UCSC ( ftp://hgdownload . cse . ucsc . edu/goldenPath/hg19/phyloP100way/hg19 . 100way . phyloP100way . bw; accessed Nov . 10th , 2015 ) and were used to annotate variant conservation . Gene-level conservation was annotated by querying the NCBI’s HomoloGene database using Biopython and Entrez to find the earliest taxonomic unit reported to sharing the gene containing the mutation [40] . These taxonomic units were then converted to numeric scores where 0 corresponds to conserved in Homo while 31 corresponds to conserved to the root of the HomoloGene taxonomic tree . ExAC gene summary data were downloaded from ( ftp://ftp . broadinstitute . org/pub/ExAC_release/release0 . 3/functional_gene_constraint/README_fordist_cleaned_nonpsych_z_data_pLI_2016_01_13 . txt; accessed Feb . 11th 2016 ) and variants were annotated with the reported probability of their respective gene being intolerant of loss-of-function mutation . Using these data , mutations present in probands were compared to mutations present in siblings with the Wilcoxon rank sum test . The results are reported in S14 Table . Methods for analysis of gene target overlaps and recurrence were adopted , with modification , from Iossifov et al . [13] . RefSeq genes were downloaded from the UCSC Table Browser ( https://genome . ucsc . edu/cgi-bin/hgTables; accessed Jul . 16th 2015 ) . The “chr” prefix was removed from the chromosome names and the raw table was sorted by chromosome and position . Coordinates of coding sequence starts and stops were extracted from the RefSeq table in BED format using custom scripts ( S1 Code ) and overlapping coding sequences were merged using BEDtools [39] . This file was intersected with the BED file of the target capture region and the length of each gene in the target region was calculated . These data were then combined with data of gene membership in gene sets from S7 Table of Iossifov et al . and the high-quality callset to produce a table describing the number of observed mutations in each gene and each gene’s set membership [13] . Given their observed contribution to ASD diagnosis ( S13 Table ) , only mosaic missense and germline de novo LGD mutations were analyzed and these mutations were analyzed in both probands and siblings . These analyses were performed using a null length model where the probability of a mutation occurring within a gene is proportional to its length targeted for exome capture relative to the total size of the capture target . For every mutation-type , individual combination , we calculate the following: ( 1 ) The expected number of recurrent mutations and a p-value for the observed number of recurrent mutations from 10 , 000 simulations using sampling with replacement . ( 2 ) For each gene set from Iossifov et al . we calculate the expected number of genes harboring mutation present in the gene set , given the length of capture targets of genes within the set relative to the total length of all gene capture targets . Using a two-sided binomial test , we test for observed enrichment or depletion from the expectation based on the null length model . For testing the enrichment of mosaic missense and LGD mutations in genes implicated in ASD , we used the approach described above with the target gene set of 107 candidate genes identified by De Rubeis et al . [32] . Amplification and sequencing primers were designed for all loci using PyroMark software and the NCBI’s Primer-BLAST [41] . Additionally , primers were checked for overlap with common SNPs using the UCSC Genome Browser [42] . Samples were amplified according to protocols in the Qiagen Pyromark PCR kit with a single biotinylated primer . Pyrosequencing was performed and data were analyzed by the Johns Hopkins Genetic Resources Core Facility . Sequence libraries were generated from purified DNA according the Nextera XT DNA Library Preparation Guide . Sequence data were then generated on an Illumina MiSeq using a MiSeq Reagent Kit v2 . Sequence reads were aligned to the human reference genome ( UCSC hg19 ) using BWA and reads supporting the reference or alternate alleles were counted [34] . Mosaic variants and germline de novo variants for validation were chosen at random from variants present in samples on hand . In total 97 variants were chosen for validation , 50 germline de novo variants and 47 mosaic variants . Primers for polymerase chain reaction amplification were designed using Primer-BLAST [41] . Amplification was performed using DNA isolated from whole blood and Sanger sequencing was performed at the Johns Hopkins University School of Medicine Synthesis and Sequencing Facility . For paired samples from the University of Maryland Brain and Tissue Bank , BAM files and corresponding phenotypic data are available from the Database of Genotypes and Phenotypes ( dbGaP ) at the National Institutes of Health ( phs000337 ) . The 50x Illumina Platinum Genomes dataset is publicly available from EBI with accession ERP001960 and ERP002490 . All runs in ERP001960 were used while select runs were used from ERP002490 ( see Materials and Methods ) . Sequence data from the Simons Simplex Collection were obtained via controlled access through the National Database for Autism Research ( http://ndar . nih . gov/study . html ? id=334 ) . | Recent sequencing experiments have shown that genetic mutations present in children but not their parents contribute to autism diagnoses in a large fraction of affected families . Here we address the question of whether mutations occurring uniquely in the children arise in the parents’ sperm or egg , or as mosaics in the child after conception . Using a dataset of 2 , 388 families , we find that while these mutations are primarily inherited from parental germ cells , 5 . 4% of these mutations appear to arise after conception . Mosaic mutations occur more frequently in probands relative to their unaffected siblings and from this enrichment we estimate that mosaic mutations contribute to 5 . 1% of autism diagnoses . In addition , we show that brain-specific mutations are not frequently detectable in individuals with an autism diagnosis . Our results indicate that some fraction of identical twins is expected to be discordant for an autism diagnosis due to genetic factors ( post-zygotic mutation ) . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"sequencing",
"techniques",
"medicine",
"and",
"health",
"sciences",
"autism",
"social",
"sciences",
"developmental",
"psychology",
"alleles",
"sequence",
"assembly",
"tools",
"genetic",
"mapping",
"neuroscience",
"mutation",
"genome",
"analysis",
"molecular",
"biology",
... | 2016 | The Contribution of Mosaic Variants to Autism Spectrum Disorder |
Host cell factors can either positively or negatively regulate the assembly and egress of HIV-1 particles from infected cells . Recent reports have identified a previously uncharacterized transmembrane protein , tetherin/CD317/BST-2 , as a crucial host restriction factor that acts during a late budding step in HIV-1 replication by inhibiting viral particle release . Although tetherin has been shown to promote the retention of nascent viral particles on the host cell surface , the precise molecular mechanisms that occur during and after these tethering events remain largely unknown . We here report that a RING-type E3 ubiquitin ligase , BCA2 ( Breast cancer-associated gene 2; also called Rabring7 , ZNF364 or RNF115 ) , is a novel tetherin-interacting host protein that facilitates the restriction of HIV-1 particle production in tetherin-positive cells . The expression of human BCA2 in “tetherin-positive” HeLa , but not in “tetherin-negative” HOS cells , resulted in a strong restriction of HIV-1 particle production . Upon the expression of tetherin in HOS cells , BCA2 was capable of inhibiting viral particle production as in HeLa cells . The targeted depletion of endogenous BCA2 by RNA interference ( RNAi ) in HeLa cells reduced the intracellular accumulation of viral particles , which were nevertheless retained on the plasma membrane . BCA2 was also found to facilitate the internalization of HIV-1 virions into CD63+ intracellular vesicles leading to their lysosomal degradation . These results indicate that BCA2 accelerates the internalization and degradation of viral particles following their tethering to the cell surface and is a co-factor or enhancer for the tetherin-dependent restriction of HIV-1 release from infected cells .
The human immunodeficiency virus ( HIV ) exploits the host cell machinery to maximize viral particle production [1] . In contrast , there are multiple systems in host cells that render them resistant to viral infection through the actions of innate host cell restriction factors [2] , [3] . This intracellular innate system can in turn be antagonized by certain viral proteins , creating a conflict between host cells and pathogens . There is accumulating evidence to now suggest that the balance between host and viral factors influences the susceptibility of the host cells to HIV infection and ultimately AIDS progression [4] . A human transmembrane protein , tetherin ( also known as BST-2 , CD317 or HM1 . 24 ) has been identified as an interferon-induced antiviral host factor in HIV-1-infected cells . During the late phase of the viral replication pathway , tetherin retains nascent HIV-1 virions at the plasma membrane and prevents viral spread [5]–[7] . Tetherin has been shown not only to block the release of lentiviruses such as HIV-1 or SIV , but also other viruses such as MLV , HTLV-1 , Lassa virus and the Marburg virus [8]–[10] . These results indicate that tetherin has broad antiviral properties through the inhibition of viral particle release , and therefore that the activation of this protein might be an effective strategy as an anti-viral therapy . Viral Protein U ( Vpu ) is a 16 kD phosphoprotein that is encoded almost exclusively by SIVCPZ and its descendants , including HIV-1 [11]–[13] . Vpu is a factor that facilitates viral particle release by antagonizing tetherin-mediated viral restriction [6] , [7] , [14] , [15] , in addition to its effects upon CD4 degradation [16]–[18] . The expression of Vpu has been shown to downregulate the tetherin levels on the plasma membrane resulting in effective virion release [7] , [19] . Indeed , Vpu-defective HIV-1 virions are efficiently retained on the plasma membrane and fewer viral particles are released compared with wild-type virions in tetherin-positive cells , including T cells and macrophages . [14] , [20] . On the other hand , in tetherin-negative cells , viral particle release is much less affected by either the presence or absence of Vpu [6] , [7] . These results suggest that Vpu antagonizes the function of tetherin , which otherwise restricts the release of HIV-1 from infected host cells . Following cell surface tethering , HIV-1 virions are subjected to internalization into CD63-positive endosomal compartments , thereby limiting the extent of virus spread [6] , [15] , [21]–[25] . Although tetherin can hold nascent viral particles on the cell surface of the host cells , the precise molecular events following the virion tethering and identity of the related host factors that regulate these processes remain largely unknown . In our current study , we identify a RING-type E3 ubiquitin ligase , BCA2 ( breast cancer associated gene 2; identical to Rabring7 , ZNF364 or RNF115 ) as a novel tetherin-interacting protein that enhances tetherin-dependent viral restriction . BCA2 was found to facilitate the internalization of HIV-1 particles captured by tetherin on the plasma membrane by associating with the cytoplasmic tail of tetherin and directing the degradation of viral particles in lysosomes . Significantly , the targeted depletion of BCA2 was found to reduce the intracellular accumulation of viral particles and to increase the persistence of nascent virions on the plasma membrane . Our current results thus reveal that BCA2 is a potential antiviral host factor through its collaboration with tetherin and is therefore a potential new therapeutic target for AIDS and its related disorders .
The precise mechanism in which HIV-1 particles undergo internalization and/or degradation in cells following tetherin-mediated capture on the plasma membrane has not been well characterized . However , accumulating evidence now suggests that plasma membrane-tethered virions transit the small G protein Rab-dependent endocytotic pathway [6] , [15] . To delineate the molecular determinants that regulate this process , we attempted to identify the Rab family member or its effector proteins that functionally interact with tetherin . As our initial screening test , we performed immunoprecipitation and GST-pull down analyses to examine the interaction of approximately 60 Rab family proteins with either tetherin or HIV-1 Gag protein . These in vitro interaction assays revealed that a Rab7-interacting protein , BCA2 , could interact with tetherin ( data not shown ) . To further confirm this interaction , we performed GST-pull down analysis with recombinant GST-BCA2 . 293T cells were transfected with either N-terminal Myc-epitope-tagged wild-type tetherin or its deletion mutant devoid of the cytoplasmic tail domain ( tetherinΔ1–20 ) . Cell lysates were then subjected to GST-pull down analysis with either GST alone or GST-BCA2 . Consequently , GST-BCA2 was found to interact with full-length tetherin in cell lysates , but to interact less efficiently with tetherinΔ1–20 ( Fig . 1A ) . This result was further confirmed by immunoprecipitation analysis using 293T cells transfected with either Myc-tetherin or Myc-tetherinΔ1–20 together with an N-terminal HA-tagged BCA2 construct ( Fig . 1B ) . BCA2 contains an N-terminal Rab7 binding domain and a C-terminal RING domain [26] . To investigate which of these is involved in the interaction with tetherin , we constructed BCA2 derivatives lacking these domains for use in immunoprecipitation analysis . Our results demonstrated that Myc-tetherin is efficiently coimmunoprecipitated with the full length BCA2 , the N-terminal truncation mutant , BCA2ΔN ( 148–305 aa ) or the RING domain deleted mutant , BCA2ΔRING ( 1–227 aa ) ( Fig . 1C ) . However , the C-terminal truncation mutant , BCA2ΔC ( 1–147 aa ) , showed no detectable interaction with Myc-tetherin ( Fig . 1C ) . These results suggest that tetherin can interact with the middle portion of BCA2 ( 147–227 aa ) located between the Rab-interacting domain and RING finger domain . We also confirmed an interaction between endogenous BCA2 and tetherin in HeLa cells ( Fig . 1D ) , where these proteins were verified to be endogenously expressed ( Fig . S1 ) . To further verify the association between BCA2 and tetherin in cells , we examined the intracellular localization of these two proteins using confocal microscopy . Immunofluorescent analysis revealed that N-terminal GFP-tagged tetherin and HA-BCA2 show a similar distribution in cells and form multiple cytoplasmic dots when they are expressed alone ( Fig . 1E ) . When GFP-tetherin and HA-BCA2 are co-transfected however , these proteins show a significant co-localization predominantly in the cytoplasm , but also in part at the plasma membrane ( Fig . 1E ) . These results together indicate that BCA2 is a tetherin-interacting protein that associates with the cytoplasmic tail of tetherin in cells . We next examined the effects of BCA2 upon HIV-1 particle production in both endogenously tetherin-positive HeLa cells and in tetherin-negative HOS cells . Endogenous tetherin expression on the cell surface was confirmed by flow cytometric analysis ( Fig . S1A ) . Cells were transfected with different amounts of HA-BCA2 together with either the HIV-1 proviral plasmid ( pNL4–3 ) [27] or a Vpu-deleted version of this construct ( pNL4–3ΔVpu ) [14] . After 48 hours , cell supernatants were assayed for the Gag p24 antigen . Strikingly , the expression of BCA2 in tetherin-positive HeLa cells led to a strong restriction of HIV-1 particle production . Importantly , the restriction of Vpu-deleted HIV-1 was more prominent than that of the WT virus in HeLa cells ( Fig . 2A ) . However , there was no significant suppressive effect of BCA2 on viral particle production in tetherin-negative HOS cells ( Fig . 2A ) . This indicated that BCA2 reduces HIV-1 particle production in the presence of tetherin . Consistent with this observation , HOS cells exogenously expressing relatively low amounts of tetherin , but not the tetherinΔ1–20 mutant , showed BCA2-mediated restriction of HIV-1 particle production ( Fig . 2B ) . Since Vpu has been shown to antagonize the antiviral activity of tetherin [6] , [7] , we next investigated whether Vpu could also counteract the antiviral effects of BCA2 . As expected , Vpu-defective HIV-1 particle production was almost completely recovered by the expression of Vpu in HeLa cells ( Fig . 2C ) . However , the co-expression of BCA2 significantly suppressed the recovery of virus particle production by Vpu ( Fig . 2C ) . Conversely , the expression of a truncated Vpu mutant ( Vpu1–50 ) , the function of which is partly impaired [6] , only partially counteracted HIV-1 restriction by BCA2 ( Fig . 2C ) . These results indicate that a functional Vpu antagonizes the restrictive activity of BCA2 . Together with our finding that BCA2 can restrict HIV-1 particle production only in tetherin-expressing cells , these data indicate that the function of tetherin , which is antagonized by Vpu , is likely required for the BCA2-mediated restriction of HIV-1 particle production . Previous studies have demonstrated that BCA2 has E3 ubiquitin ligase activity which is essential for the downregulation of EGFR expression [28] . We therefore examined whether this activity is necessary for the anti-viral effects of BCA2 . We created a RING finger-defective mutant BCA2 ( C228A/C231A ) [29] and investigated its effect upon virus particle production . Although a tetherin-interacting motif defective BCA2 mutant ( BCA2ΔC ) failed to restrict viral particle production , both WT and C228A/C231A BCA2 were capable of doing so ( Fig . 2D ) . Moreover , the effect of C228A/C231A BCA2 mutant was modest increase than that of WT BCA2 ( Fig . 2D ) , probably due to the inhibition of both auto-ubiquitination and following degradation of this mutant as reported previously [28] . These results indicate that the ubiquitin ligase activity of BCA2 is dispensable for its function in the restriction of virus particle formation . To next investigate the effects of BCA2 upon virus particle restriction in T cells , we created Jurkat cells stably expressing untagged BCA2 ( Fig . 2E ) . FACS analysis with a tetherin antibody revealed that these cells indeed express tetherin on their cell surface ( Fig . S1A ) . The cells were then infected with either HIV-1NL4–3 or HIV-1NL4–3ΔVpu at a low multiplicity of infection ( m . o . i . = 0 . 05 ) . In agreement with a previous report [11] , we found that Vpu-deleted virus replicated slightly less efficiently than WT-virus ( Fig . 2E ) . Our results showed that BCA2 expression reduces HIV-1 particle production in both WT- and ΔVpu-virus infected cells , although this effect was more prominent in cells infected with ΔVpu-virus ( about 4-fold ) than with WT-virus ( about 2-fold ) ( Fig . 2E ) . Interestingly , immunoblotting analysis revealed that the expression levels of endogenous BCA2 in HeLa and Jurkat cells were relatively lower than in HOS cells ( Fig . S1B ) , implying that exogenous BCA2 expression would tend to impact virus particle restriction in these cells in the presence of functional tetherin . To delineate the molecular mechanism by which BCA2 suppresses virus production , we performed immunoblotting analysis with a p24 antibody . Interestingly , the expression of BCA2 in HeLa cells significantly reduced the Gag protein levels , particularly cell-associated p24 , but had no effect upon the expression of Vpu ( Fig . 2F ) . Our results also indicate that BCA2 expression has modest effects on viral release efficiency as compared with its drastic effects on the cell-associated p24 protein levels ( Fig . 2F ) . Of note , the BCA2-induced depletion of cell-associated p24 in the absence of Vpu was more prominent than in the presence of Vpu ( Fig . 2F ) . These data together suggest that BCA2 may enhance the degradation of nascent HIV-1 virions captured by tetherin on the plasma membrane . To rule out the possibility that BCA2 affects the expression of HIV-1 proteins , we next performed pulse-chase analysis with pNL4-3ΔVpu-transfected HeLa cells . Our results demonstrated that BCA2 expression induces the rapid degradation of the HIV-1 Gag protein ( Fig . 2G ) . Consistent with our immunoblotting data ( Fig . 2F ) , the degradation of p24 was shown to be more prominent than that of Pr55 ( Fig . 2G ) . Furthermore , our RT-PCR analysis revealed that BCA2 expression does not significantly affect the mRNA levels of HIV-1 Gag ( Fig . S2 ) . These results together indicate that BCA2 facilitates the intracellular degradation of virus particles rather than the suppression of HIV-1 protein expression . As described above , the expression of BCA2 significantly reduces the level of cell-associated p24 protein , raising the possibility that it facilitates the intracellular degradation of unreleased virions . To test this possibility , we performed transmission electron microscopy ( TEM ) analysis of HeLa cells transduced with proviral plasmid together with either HA-BCA2 or a control vector . In control cells , nascent assembled virions were observed on the plasma membrane and relatively little accumulation of virions was observed in intracellular compartments ( Fig . 3A ) . In BCA2-expressing cells , however , substantial numbers of mature virions could be observed in the intracellular vesicles , and a significant reduction of mature viral particles on the plasma membrane was found ( Fig . 3B ) . This suggests that BCA2 facilitates the internalization of mature viral particles into intracellular vesicles for degradation . Consistent with our TEM results , immunofluorescent and confocal microscopic analysis further revealed that BCA2 expression promotes the accumulation of p24 in CD63+ intracellular compartments when compared with the vector control ( Figs . 4A , B ) . Various proteins that are sorted into CD63+ intracellular compartments are destined for lysosomal degradation [30] , [31] . To address whether virion degradation is mediated by this pathway following internalization , we co-transfected HeLa cells with the HIV-1 proviral plasmid together with either empty vector or HA-BCA2 , and then treated the cells with lysosome inhibitors ( leupeptin and NH4Cl ) . Strikingly , treatment with lysosome inhibitors significantly blocked the decrease in intracellular Gag in BCA2-expressing cells ( Fig . 4C ) . Importantly also , parallel ELISA analysis of the supernatants from these transduced cells revealed that lysosome inhibitors had no effect upon viral release ( Fig . 4D ) . These results suggest that BCA2 promotes the lysosomal degradation of HIV-1 virions following their retention on the plasma membrane and subsequent internalization into CD63+ endosomes . To further delineate the role of endogenous BCA2 in HIV-1 particle release , we next performed experiments in which HeLa cells were transduced with either control or two different BCA2-specific siRNAs ( BCA2-I , II ) and then transfected with pNL4-3 or pNL4-3ΔVpu . Immunoblotting analysis with a BCA2 antibody demonstrated that both of the siRNAs targeting BCA2 could significantly reduce its endogenous expression ( Fig . 5A ) . Measurement of the p24 antigen levels in the cell supernatant further revealed that viral particle production was only slightly increased in both pNL4–3 and pNL4–3ΔVpu transfected cells , although the effect was more significant in pNL4–3ΔVpu transfected cells ( approximately 2-fold ) ( Fig . 5A ) . Immunofluorescent analysis by confocal microscopy additionally revealed that although the localization of Gag proteins was observed predominantly in CD63+ intracellular structures in control-siRNA treated cells , this profile was dramatically shifted to the plasma membrane in BCA2-siRNA treated cells ( Figs . 5B , C ) . This indicated that the silencing of BCA2 blocks the relocation of virions into the intracellular compartments and increases the persistence of virions captured by tetherin on the cell surface . To further investigate this possibility , siRNA-transduced cells were subsequently treated with the protease subtilisin , which liberates cell surface-captured virions by abolishing virion-tetherin interactions [15] . In the case of WT-virus , subtilisin stripping had only slight effects upon virion release ( Fig . 5D ) , in agreement with a previous report [15] . However , in the case of Vpu-defective virus , viral release from BCA2-deplation cells was significantly recovered by subtilisin stripping , reaching the level of WT-virus infected cells ( Fig . 5D ) . These data suggest that BCA2 depletion inhibits the intracellular accumulation of Gag proteins and , consequently , increases the fraction of virions retained at the cell surface by tetherin . Overall , the results of our current study indicate that BCA2 facilitates the internalization of HIV-1 virions that have not been released , thereby enhancing their degradation . This internalization and degradation of cell surface-retained virions may represent rate limiting steps in the tetherin-mediated restriction of viral release that are accelerated by BCA2 .
In our current study , we identify BCA2 as a functional tetherin-interacting protein . Although BCA2 is widely expressed in various cell lines [26] , [32] , its antiviral effects have been observed only in cells expressing tetherin , suggesting that BCA2 cooperates with tetherin to achieve efficient restriction of viral particle production . BCA2 was found in our current analyses to play a crucial role in the internalization and degradation of nascent HIV-1 virions , following their tethering to the host cell plasma membrane . These internalization and degradation steps may be rate limiting during restriction by tetherin because the targeted depletion of BCA2 can shift the distribution of Gag to the plasma membrane and can partly overcome the release inhibition of Vpu-minus virions in HeLa cells . Importantly in this regard , BCA2 directs HIV-1 particles to CD63+ endosomes or lysosomes for degradation . The molecular mechanisms by which this is achieved have not yet been fully characterized . However , previous studies have demonstrated that BCA2 directly binds a small G protein , Rab7 , and thereby plays crucial roles in vesicle trafficking to the late endosomes or lysosomes , in addition to lysosome biogenesis [26] , [33] . Indeed , the aberrant expression of BCA2 not only affects epidermal growth factor receptor ( EFGR ) degradation , but also induces the perinuclear aggregation of lysosomes and increased acidity within lysosomes [26] , [28] . Given our current data , these results indicate that BCA2 coordinates the trafficking of intracellular vesicles containing internalized viral particles to the lysosomes in conjunction with Rab7 , resulting in the effective degradation of these virions . Consistently , in tetherin-positive HeLa cells , Gag protein has been shown to co-localize with the GTP-bound active form of Rab7 in the absence of Vpu ( our unpublished observation ) . Furthermore , a dominant negative mutant of Rab5 can inhibit the internalization of nascent HIV-1 particles [15] . These findings raise the possibility that plasma membrane-tethered virions may go through a Rab5- and/or Rab7-dependent endocytotic pathway from the cell surface to the endosomes or lysosomes and eventual degradation . Notably , our immunoprecipitation data indicate that BCA2 interacts with tetherin at a region distinct from the Rab7 binding site . Consistently , an N-terminal truncation mutant of BCA2 can still interact with tetherin . These results suggest that BCA2 may simultaneously interact with Rab7 and tetherin at distinct regions and might therefore act as a physical scaffolding protein between these two proteins . During endocytosis , tetherin-BCA2 complexes might therefore recruit Rab7 to vesicles containing virions . Although the function of BCA2 during HIV-1 restriction is likely to be dependent on tetherin , the antiviral effects of BCA2 were found to be still active against Vpu-positive viruses . However , the overexpression of Vpu can abrogate the antiviral effects of BCA2 , indicating a potentially stoichiometric relationship between BCA2 and Vpu during BCA2-mediated viral restriction . Importantly , our current results suggest that BCA2 is not involved in regulating the expression of Vpu . However , it is possible that BCA2 antagonizes the function of Vpu in counteracting tetherin , although further analysis is needed to address this question . Our current results additionally demonstrate that the effects of BCA2 depletion on particle production are about two-fold , which is a relatively modest impact compared with the 5-to-10-fold effects of Vpu in tetherin-positive cells . This indicates that the inhibition of BCA2 cannot fully restore Vpu-defective HIV-1 particle production to the level of the WT-virus . Apparently , capture of virions on the plasma membrane by tetherin provides restriction even when BCA2-depletion suppresses the internalization and degradation of nascent virions . These effected were further revealed by our subtilisin stripping assay; BCA2-depletion plus subtilisin treatment recovered ΔVpu-virus particle production to the level of the WT-virus . These results indicate that BCA2 very likely functions downstream of virus tethering on the plasma membrane ( i . e . post-tethering stages ) . In summary , the results of our current study demonstrate that BCA2 is a potential anti-HIV-1 host factor that partners with tetherin to facilitate the internalization and degradation of nascent viral particles . Our present findings thus shed new light on the molecular machinery underlying the tetherin-dependent HIV-1 restriction pathway . BCA2 and other molecules of this pathway may thus be potential new therapeutic targets for AIDS and its related disorders .
HeLa , HOS and 293T cells were cultured in DMEM supplemented with 10% fetal bovine serum ( FBS ) . Jurkat cells were maintained in RPMI-1640 containing 10% FBS . Plasmid transfections into adherent or suspended cells were performed using Lipofectamine 2000 ( Invitrogen , Carlsbad , CA ) or Amaxa nucleofector ( Program S-18; Amaxa biosystems , Cologne , Germany ) , respectively , according to the manufacturer's instructions . Human BCA2 and tetherin/CD317 coding sequences were amplified from HeLa total RNA by RT-PCR using the following pairs of oligonucleotides containing restriction enzyme BamHI sites ( underlined ) or a stop codon ( boldface ) : 5′-GGATCCGGATGGCGGAGGCTTCGGCGGC-3′ ( BCA2 forward primer , sense ) and 5′-TCAGAAAGTCCATCGGTCATG-3′ ( BCA2 reverse primer , antisense ) ; 5′-GGATCCGGATGGCATCTACTTCGTATGA-3′ ( tetherin forward primer , sense ) and 5′-TCACTGCAGCAGAGCGCTGAGGC-3′ ( tetherin reverse primer , antisense ) . The purified PCR products were inserted into the pCR4Blunt-TOPO vector ( Invitrogen ) , and cDNA inserts were then subcloned into pCMV-HA , pCMV-Myc , pEGFP-C1 , pIRESpuro ( Clontech , Palo Alto , CA ) or pGEX-KG ( Amersham Bioscience , Sunnyvale , CA ) vectors . A human codon-optimized HIV-1 Vpu expression vector ( pcDNA-Vphu ) [34] and Vpu-deleted HIV-1 molecular clone ( pNL4–3/Udel , herein called pNL4–3ΔVpu ) [14] were kindly provided by Dr . K . Strebel ( National Institutes of Health , Bethesda , MD ) . The ΔRING ( 1–227 aa ) , ΔN ( 148–305 aa ) , ΔC ( 1–147 aa ) , and C228A/C231A derivatives of BCA2 , the tetherin mutant Δ1–20 ( 21–180 aa ) and the truncated Vpu mutant ( 1–50 aa ) were constructed using standard molecular cloning procedures . The WT-virus or ΔVpu-virus stocks were produced by transient transfection of 293T cells with the pNL4–3 or pNL4–3ΔVpu proviral plasmids , respectively . Culture supernatants containing virus were collected 48 hours after transfection , filtered through a 0 . 45 µm Millex-HV filter ( Millipore , Billerica , MA ) and immediately stored at −80°C until use . An anti-BCA2 polyclonal antibody was produced by UNITECH ( Chiba , Japan ) . An anti-p24 monoclonal antibody has been described previously [35] . The rabbit anti-Vpu and mouse anti-HM1 . 24 ( tetherin ) antibodies were kindly donated by Dr . K . Strebel ( National Institutes of Health , Bethesda , MD ) [36] and Chugai Pharmaceutical Co . ( Kanagawa , Japan ) [37] , respectively . Other antibodies used in this study were as follows: mouse anti-HA ( Roche , Basel , Switzerland ) , mouse anti-Myc ( Roche ) , mouse anti-α-tubulin ( Sigma , St . Louis , MO ) , rabbit anti-CD63 ( Santa Cruz Biotechnology , Santa Cruz , CA ) and Alexafluor-conjugated anti-IgG ( Invitrogen ) . For GST pull-down assays , GST-tagged BCA2 was expressed in Escherichia coli BL21 ( DE3 ) cells and purified using standard protocols . Myc-tetherin-expressing 293T cell lysates were incubated with glutathione-beads that had been coupled with GST-BCA2 proteins . The beads were then washed , and bound proteins were visualized by Coomassie Brilliant Blue R-250 staining and analyzed by immunoblotting . For immunoprecipitation analysis , 293T cells expressing Myc-tetherin and HA-BCA2 were lysed and incubated with an anti-HA affinity gel ( Sigma ) . Alternatively , to detect endogenous tetherin-BCA2 complexes , HeLa cell lysates were co-incubated with protein A/G-mixed Sepharose ( GE Healthcare , UK ) and either anti-tetherin antibody or control mouse IgG . Bound proteins were analyzed by SDS-PAGE and immunoblotting . Cells in 12-well plates were co-transfected with pNL4–3 or pNL4–3ΔVpu ( 300 ng ) and either pCMV-HA-BCA2 or empty vector ( 0–300 ng ) , in the presence or absence of vectors encoding Vpu ( 30 or 75 ng ) or Myc-tetherin ( 100 ng ) . Two days after transfection , virus-containing supernatants were harvested and filtrated to remove debris , and p24 antigens were measured by Lumipulse ( Fujirebio , Tokyo , Japan ) . For immunoblotting assays , the virus-containing supernatants ( 400 µl ) was layered onto 600 µl of 20% sucrose in PBS and centrifuged at 20 , 000 g for 2 hours at 4°C . The cell lysates were prepared using RIPA buffer by incubation at 4°C for 10 minutes and centrifugation at 16 , 000 g for 30 minutes . In experiments using lysosomal inhibitors , each drug was added 18 hours before harvesting . Immunoblotting band intensities were quantitated with ImageJ software . For multi-cycle replication assays , Jurkat cells ( 1×106 ) were transfected with either empty vector or pIRESpuro-BCA2 ( 3 µg ) . After the selection of transfectants with puromycin for 24 hours , cell aliquots were then infected with either HIV-1NL4–3 or HIV-1NL4–3ΔVpu at an m . o . i of 0 . 05 . Viral supernatants were collected periodically , and p24 levels were measured as described above . One day prior to transfection , HeLa cells were seeded onto glass-bottom dishes coated with poly-L-lysine ( Matsunami , Osaka , Japan ) . At 48 hours after transfection , the cells were fixed with 4% paraformaldehyde and permeabilized with 1% Triton X-100 . Cells were than stained with primary antibodies and Alexa-conjugated secondary antibodies . Confocal microscopic imaging was performed using a Zeiss LSM510 instrument equipped with a 63× oil-immersion objective . For electron microscopy , transfected HeLa cells were fixed with 2 . 5% glutaraldehyde and subjected to transmission electron microscopy , as described previously [38] . Cells in 6-well plates were co-transfected with pNL4–3ΔVpu ( 1 µg ) and either pCMV-HA-BCA2 or empty vector ( 3 µg ) . Two days after transfection , the cells were washed and starved in Met-/Cys-depletion medium ( Invitrogen ) for 30 min and pulse-labeled for 15 min with 0 . 25 mCi/ml of [35S]Met-Cys medium , and chased in unlabeled medium for 4 . 5 hours . Cells were harvested periodically , and cell lysates were immunoprecipitated with anti-p24 antibody , and then analyzed by SDS-PAGE and autoradiography . BCA2-targeted siRNAs were obtained from Invitrogen as Stealth Select RNAi constructs ( Oligo ID #HSS120532 and #HSS120534 ) . A Stealth RNAi Luciferase reporter control ( Invitrogen ) was used as the negative control siRNA . Cells in 12-well plates were transfected with these siRNAs at a final concentration of 50 µM using Lipofectamine RNAiMAX ( Invitrogen ) . The following day , the cells were re-transfected with 300 ng of either pNL4–3 or pNL4–3ΔVpu , and two days later were harvested and analyzed by immunoblotting or confocal microscopy . For protease subtilisin stripping assays , viral supernatants ( 1 . 2 ml ) from siRNA/DNA-transfected HeLa cells were harvested as a “first supernatant” . After harvesting , the cells were washed once with pre-warmed PBS and then incubated with 300 µl of either PBS or Tris/HCl ( pH 8 . 0 ) buffer containing 1 mg/ml subtilisin ( Sigma ) for 15 min at 37°C . To stop the reaction , 900 µl of DMEM containing 5 mM PMSF were added to the cells , and supernatants ( total 1 . 2 ml ) were again harvested as a “second supernatant” . Both the first and second supernatants were then mixed and the p24 levels were measured as described above . The GenBank accession numbers for human BCA2 ( Rabring7/ZNF364/RNF115 ) and human tetherin ( CD317/BST-2/HM1 . 24 ) are BC054049 and D28137 , respectively . | Human cells possess multiple systems that render them resistant to viral infection . Recently , a transmembrane protein , tetherin , has been identified as an antiviral host factor in HIV-1-infected cells . Tetherin retains newly assembled virions at the plasma membrane and prevents viral release from the infected cells . However , the precise molecular mechanisms following the virion tethering remain largely unknown . In our current study , we have identified a RING-type E3 ubiquitin ligase , BCA2 , which co-localizes and interacts with tetherin in human cells . BCA2 was found to facilitate the internalization of HIV-1 particles captured by tetherin on the plasma membrane and to enhance the targeting of viral particles to the lysosomes . Conversely , the targeted depletion of endogenous BCA2 reduces the intracellular accumulation of viral particles . Additionally , the expression of a small viral protein Vpu , an antagonist of tetherin , counteracts the antiviral effects of BCA2 . These results suggest that BCA2 is a potential antiviral factor that collaborates with tetherin to facilitate the degradation of nascent HIV-1 particles during “post-tethering” processes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"virology/host",
"antiviral",
"responses"
] | 2009 | BCA2/Rabring7 Promotes Tetherin-Dependent HIV-1 Restriction |
The identification of alternatively spliced transcript variants specific to particular biological processes in tumours should increase our understanding of cancer . Hypoxia is an important factor in cancer biology , and associated splice variants may present new markers to help with planning treatment . A method was developed to analyse alternative splicing in exon array data , using probeset multiplicity to identify genes with changes in expression across their loci , and a combination of the splicing index and a new metric based on the variation of reliability weighted fold changes to detect changes in the splicing patterns . The approach was validated on a cancer/normal sample dataset in which alternative splicing events had been confirmed using RT-PCR . We then analysed ten head and neck squamous cell carcinomas using exon arrays and identified differentially expressed splice variants in five samples with high versus five with low levels of hypoxia-associated genes . The analysis identified a splice variant of LAMA3 ( Laminin α 3 ) , LAMA3-A , known to be involved in tumour cell invasion and progression . The full-length transcript of the gene ( LAMA3-B ) did not appear to be hypoxia-associated . The results were confirmed using qualitative RT-PCR . In a series of 59 prospectively collected head and neck tumours , expression of LAMA3-A had prognostic significance whereas LAMA3-B did not . This work illustrates the potential for alternatively spliced transcripts to act as biomarkers of disease prognosis with improved specificity for particular tissues or conditions over assays which do not discriminate between splice variants .
Alternative splicing is the process by which cells can selectively include different sections of pre-mRNA during RNA processing . If these transcripts are translated , this results in a set of closely related , but different , proteins expressed from a single locus [1] , [2] . Alternative splicing is prevalent ( the majority of human genes are alternatively spliced , with an average of about 5 . 4 transcripts per gene [3] ) , and tightly regulated . It is a key player in many molecular pathways , and is known to be involved in many of the ‘hallmark’ processes of cancer [4] , including resistance to apoptosis , invasion , angiogenesis and differentiation [5] . Until recently , a lack of appropriate tools has made it impossible to perform routine global surveys of alternative splicing , making it relatively understudied . Recently , a set of exon microarrays has been developed by Affymetrix . These feature probesets targeting at intervals throughout each transcript , rather than simply at the 3′ end interrogated by most other arrays . This enables the assembly , in silico , of expression levels across genes providing a more complete representation of transcription for each gene , and allowing the identification of loci where there are changes in the splicing pattern across experimental samples . Another useful consequence of the increased resolution of the arrays is that since most transcripts are targeted by multiple probesets , their signals can be combined in order to increase statistical power [6] . This leads to the identification of differentially expressed genes with smaller effects sizes than can be found using other , less comprehensive platforms . However , this increased performance is not without additional challenges , since detailed analysis of the arrays requires annotation describing the known relationships between genes , transcripts and exons , and the ability to combine this with appropriate statistics [7]–[9] . Here we use Affymetrix Exon 1 . 0ST arrays to study hypoxia in human cancers . Hypoxia can lead to an altered , invasive tumour phenotype through wide ranging changes in gene transcription within a cell . Perhaps the best known mediator of this is hypoxia inducible factor ( ) , a transcription factor that regulates the expression of many tumorigenic genes involved in a wide range of cellular processes , including angiogenesis , cell proliferation , apoptosis and cell migration [10] . Several well known regulated ( and cancer associated ) genes ( e . g . VEGFA [11] , CA9 [12] ) are already known to be alternatively spliced , although the relationship between hypoxia , differential transcript expression , alternative splicing , and tumour phenotype has yet to be fully determined . Given the ubiquity of alternative splicing , it is likely that there are many more such events to be discovered . To date , most published metrics used to analyze alternative splicing compute an overall gene level summary that is used as a baseline against which the behaviour of its constituent exons can be compared . The most popular metric is the splicing index [2] , [13] , [14] , which aims to identify probesets that have different inclusion rates ( relative to the gene level ) between two sample groups . MIDAS is an extension of the splicing index that uses ANOVA instead of a t-test to evaluate significance , allowing comparisons between multiple sample groups . In FIRMA [15] , the popular linear model Robust Multichip Analysis ( RMA ) ( used for normalization and summarization of probeset intensities ) is fitted to each gene in order to estimate overall expression level in each sample , while the median of the residuals in an exon is used to generate a summary statistic of each exon's alternative splicing . The method was developed to evaluate situations where there are neither replicates nor pre-defined groups . MADS [16] calculates splicing indices and p-values of individual probes separately , prior to summarization at probeset level . The Pattern-based Correlation ( PAC ) [17] , [18] algorithm is based on the correlation across samples between exon expression levels and the overall gene expression level . PAC is limited by the number of samples , since it works best when there are enough differentially spliced samples to significantly weaken the correlation between gene and exon . Genes , exons or probesets scoring highly in an algorithm/metric ( and usually accompanied by a low probability score ) are identified as promising candidates of alternative splicing events and are suitable for further tests . As discussed elsewhere [15] , alternative splicing is an analogue process , with no threshold above which alternative splicing can be said to occur . Results are therefore usually reported as a ranked list . We used exon microarrays to study alternative splicing events in hypoxia-associated genes in a set of ten Head and Neck Squamous Cell Carcinomas ( HNSCC ) . These samples are a subset of 59 HNSCC collected and analysed previously [19] . The 10 samples comprised the 5 most and 5 least hypoxic samples , determined by their Hypoxia Score ( HS ) , a gene signature derived metric of tumour hypoxia [19] . To identify hypoxia-associated genes , we developed a novel approach that increases the power of detection of differentially expressed genes by exploiting the fact that most transcripts are targeted by multiple , independent probesets . Using this approach , we identified 146 genes with significant hypoxia-related changes in exon expression across their loci , the set includes a higher number of known hypoxia-induced genes compared to the equivalent analysis on HG-U133A Plus2 arrays . To identify alternative splicing events we used a combination of the splicing index and a new metric , proposed here , based on the variation of reliability weighted fold changes ( VFC ) . The weights are based on Detection Above the Background ( DABG ) scores [20] , [21] , which relate to the probability that the observed probeset signal is higher than the background noise distribution . We show here that the inclusion of probeset reliability information improves the detection of alternative splicing events when applied not only to our hypoxia data but also when applied to an independent dataset . The proposed strategy identified SLCO1B3 ( Organic anion-transporting polypeptide 8 ( OATP8 ) , ENSG00000111700 ) , WDR66 ( WD repeat-containing protein 66 , ENSG00000158023 ) , COL4A6 ( Collagen chain precursor , ENSG00000197565 ) and LAMA3 ( Laminin subunit precursor , ENSG0000053747 ) as potentially involved in alternative splicing events related to hypoxia . The strongest evidence was for LAMA3 which was successfully validated by RT-PCR . We also found that expression of the LAMA3-A splice variant in head and neck cancers was strongly associated with poorer survival following primary surgical treatment , showing that our methodology can be used to identify novel splicing events with prognostic significance .
In order to ensure the validity of our approach , it was applied to the colon cancer sample dataset ( 10 paired normal-cancer samples ) from Affymetrix ( http://www . affymetrix . com ) . In [2] the dataset was analysed to identify alternative splicing events and RT-PCR validation of 49 genes ( chosen based on splicing index p-values of filtered genes/probesets , manual inspection and literature information ) was performed . Out of these , differential AS events in colon cancer relative to normal colon tissue were confirmed as either present or absent in 27 genes: eleven genes showed clear differential AS and 16 showed no evidence of AS . Of the remaining 22 genes , 5 showed positive results but with some ambiguity and 17 exhibited AS but were not distinctive between normal and cancerous tissues . The pipeline described in Materials and Methods ( Figure 1 ) was used to identify genes for which there were significant changes in expression in one or more exons across their length . We refer to these as differentially expressed ( DE ) genes , and do not at this stage consider whether expression changes are uniform across their length . We identified 1091 DE genes , 892 up-regulated in colon cancer relative to normal colon tissue and 198 down-regulated . We set the FDR cut-off of the paired t-test to 10% to include as many genes validated by RT-PCR as possible . Fifteen of the 27 genes for which AS events were confirmed by PT-PCR as either present or absent can be found in the set of DE genes obtained . A plot of the ranked VFC values versus the ranked splicing indices ( metrics described in Materials and Methods ) of the DE genes shows that most of the genes successfully confirmed by RT-PCR are placed in the top ranks by VFC apart from FN1 and SLC3A2 which are in the bottom 50% ( Figure 2A ) . Interestingly , SLC3A2 had also a low rank in the analysis performed in [15] . Most of the genes that did not show alternative splicing events are given lower ranks by the VFC but not by the splicing index . COL11A1 was surprisingly high in both ranks even thought it was found to have no alternative splicing event by RT-PCR validation . The inclusion of probeset reliability information in the VFC enables a better differentiation between the true and false positives ( Figure 2B ) . While Hypoxia causes a general down- rather than up-regulation in gene expression [22] , most up-regulated genes are dependent [22] . This study focused on up-regulated changes , as they represent a more specific target group . The pipeline described in Materials and Methods was used to identify genes with hypoxia-induced changes in expression across their loci ( i . e . hypoxia-associated genes ) in the Head and Neck dataset . Essentially , the pipeline aims to find genes for which at least one exon shows a big change or for which many exons show a smaller but consistent change . The filtering stage identifies exon targeting probesets predicted to hybridize to a single locus within the genome , which are significantly differentially expressed between high and low HS samples ( DE probesets ) . The analysis then takes place in two parallel tracks , one that identifies genes targeted by at least one DE probeset with a significantly large change ( Track 1 ) and the other that seeks genes targeted by a high proportion of DE probesets ( Track 2 ) . A combination of the splicing index and the VFC identified alternative splicing events in hypoxia-associated genes . Figure 3 plots the ranked splicing indices versus the ranked VFC values , before and after the inclusion of the DABG information . Six genes ( PYGL , BNC1 , HMGA2 , SLCO1B3 , SNAI2 and CA12 ) show a high splicing index rank but low VFC rank . On further inspection , all of these have one or two probesets found to be absent in all samples , with low fold change ( e . g . Figure 4A ) . This could indicate an alternative splicing event that is not correlated with hypoxia . We found that in five of the six genes these absent probesets were consistently absent in all replicates of the tissue panel sample dataset from Affymetrix ( available at http://www . affymetrix . com/ ) , suggesting that the absent probesets are very likely to be poorly performing probesets . An alternative splicing event was therefore considered unlikely , and these genes to be false positives . In the case of SLCO1B3 , the three replicates of liver are present and highly expressed across all probesets , showing that the probeset absent in HNSCC is not faulty ( Figure 4B ) . SLCO1B3 has two known transcripts ( Figure S3 ) and the information provided by the arrays indicates that samples with high HS express only the shorter transcript of SLCO1B3 , while in liver the longer transcript is expressed . However , nothing can be inferred from the low HS samples because most probesets are absent in these samples . The gene in the left-top of Figure 3B ( DVL1 ) is ranked high by VFC and low by splicing index . Manual comparison to the Affymetrix sample dataset , revealed that most of the FC variation is due to probesets with a low range of response . Genes with high ranks on both metrics ( SI and VFC ) are likely to be alternative spliced . Five genes remained at the top of the ranking of both metrics ( top 0 . 5 quantile of the combined ranking ( SI+ VFC ) ) after inclusion of the DABG information ( LAMA3 , WDR66 , NRG1 , NDUFA4L2 and COL4A6 ) . LAMA3 is targeted by a large number of probesets , generally present across samples . Detailed examination shows that one of its three known transcripts ( ENST00000269217 LAMA3-A ) is differentially expressed in response to hypoxia , while the other two transcripts ( ENST00000313654 LAMA3-B and ENST00000399516 LAMA3-C ) show no difference ( Figure 5 ) . Expression levels of LAMA3 in the Affymetrix sample dataset show that the low FCs in probesets targeting LAMA3-B and LAMA3-C are not due to faulty probesets . Thus , there is strong evidence that LAMA3 is a good candidate for further validation of alternative splicing . Similar analysis provided additional support for WDR66 ( Figure 4C ) . NRG1 featured 3 probesets , with high absolute expression and low fold-change , that were subsequently found to show similar patterns across all samples in the Affymetrix sample dataset; suggesting that these probesets may be saturating and that NRG1 may be a false positive ( Figure 4D ) . NDUFA4L2 presents a similar case to NRG1 . COL4A6 is targeted by a large number of probesets . These show high variation in FC with regions of similar FC values . Most probesets are absent in the five low HS samples , making it unfeasible to infer alternative splicing with respect to HS values , based on the information available . However , this information suggests that the possibility of a shorter isoform of COL4A6 being expressed in high HS values should not be discarded . Using our pipeline we identified SLCO1B3 , WDR66 , COL4A6 and LAMA3 as potentially involved in alternative splicing events associated with hypoxia in HNSCC . Laminin 3 ( LAMA3 ) forms the subunit of laminin-332 , an extracellular glycoprotein , known to be important in cell migration and tumour invasion [25] . Laminin-332 , detected by immunohistochemistry to the gamma 2 subunit , has been found at the invasive edge of squamous cell carcinomas and has been associated with a poor prognosis in a wide range of epithelial carcinomas including oral , cervical and oesophageal cancers [26]–[28] . LAMA3 is known to be alternatively spliced [29] with the shorter transcript , Laminin 3A , encoding the protein subunit for the well characterised laminin-332 .
In [19] 59 Head and Neck Squamous Cell Carcinoma ( HNSCCs ) samples , obtained prior to any treatment at the time of primary surgery , were processed onto Affymetrix HG-U133 Plus2 arrays and a set of 99 genes up-regulated in hypoxia was obtained by analysis of genes whose in vivo expression clustered with the expression of 10 well-known hypoxia-regulated genes ( e . g . CA9 , GLUT1 , and VEGF ) . A Hypoxia Score ( HS ) was defined as the median value of expression for these 99 genes ( ‘HS genes’ ) . High HS values indicated higher hypoxia relative to lower values and were an adverse prognostic factor in an independent microarray dataset . HS was a continuous variable well spread across the samples ( Figure S6 ) . Here , we first eliminated samples from the study with a high percentage of absent calls [32] by removing the top 10-th quantile of the samples ordered by number of absent calls , and then selected the 5 least and 5 most hypoxic samples as defined by the HS values . Confirmation of hypoxia status was carried out by investigating CAIX protein expression [10] in histological sections . There was a statistically significant increased CAIX expression in the samples with high HS values ( p = 0 . 024 , Figure S7 ) . These 10 samples were then processed onto Affymetrix Human Exon 1 . 0ST arrays using manufacturers' standard protocols , as described in [7] . Following hybridization , we investigated the similarity in expression profiles among the 10 exon arrays . Multidimensional scaling and hierarchical clustering of the samples based on a reduced set of probesets ( exonic probesets flagged present; DABG in at least half of all the samples; N = 172 , 204 ) confirmed that the samples are partitioned by high and low HS values , as expected ( Figure S8 ) . Exon array data have been deposited in NCBI's Gene Expression Omnibus [33] and are accessible through GEO Series accession number GSE18300 . Figure 1 shows the analysis pipeline used to identify DE genes in exon array data . Data were first summarised using RMA [34] ( there is no significant difference between RMA and PLIER in terms of alternative splicing identification [35] ) and then filtered to include only exon targeting probesets , predicted to hybridize to a single locus within the genome . A DABG score filtering ( in all samples of at least one replicate set , see Text S1 ) and a t-test are then applied to each probeset . Here , we use the t-statistic for simplicity of implementation; however , any other suitable test could be used . An FDR[36] of 5% was used as a cut-off for statistical significance . The starting point for further analysis is then the set of differentially expressed exonic , non-multiply targeting probesets that passed the DABG score filtering . As such , it is similar to the set of DE probesets that would emerge from a standard analysis of 3′IVT arrays . Alternative splicing occurs as a result of the differential inclusion or exclusion of one or more exons from a gene , and can also involve the retention of intron sequence or the use of alternative 5′ and 3′ splice sites [37] . In this work we concentrated on events related to differential exon usage , therefore , introns and intergenic regions were not considered . We used the combination of two alternative splicing metrics to identify genes alternatively spliced with respect to high and low HS values: the splicing index and the VFC ( Variation of reliability weighted Fold Changes ) . These are described in detail below . The splicing index ( SI ) of probeset relative to gene is defined as ( 1 ) where and are the means of the inclusion rates of probeset in gene across all replicates for sample groups 1 and 2 , respectively . The inclusion rate of probeset , in gene , in group , in replicate is given by ( 2 ) where is the expression level of probeset and is the gene level of gene . The gene level can be calculated by taking the mean or median across all exonic probesets . Overall , the splicing index is highly dependent on the gene level calculation , and is reported to work best when the gene has a large number of constitutive exons and a small number of alternative exons [38] . To calculate the VFC , the range of FCs for all exonic probesets across a gene is calculated . The range is used because it is sensitive to extremes . Alternatives , such as the coefficient of variation or the standard deviation minimise the effects of these outliers , reducing the algorithm's ability to identify single probeset changes . An obvious problem when using FCs is that each FC has different degrees of reliability specified by the DABG p-values . To incorporate this information , we centre the FC values around the median FC and weight them by the number of samples flagged present , mapped through a sigmoid transformation . We define the weight of probeset as: ( 3 ) where is the number of present samples of probeset and is the total number of samples . Finally , we normalise the absolute range of weighted FCs by the weighted-mean of FCs across all probesets . This normalisation is necessary to eliminate the bias resulting from the relationship between mean and range of FCs per gene . A positive correlation is observed because many genes have at least one probeset whose value is not detectable above background ( high DABG p-value ) in most of the samples and has a reduced difference of the mean value between the two sample groups ( producing low FCs ) inducing a large range of FCs across the gene by lowering the minimum FC value . The VFC of gene is thus: ( 4 ) where is the number of non-multiply targeting exonic probesets targeting gene , is the FC of probeset , and and are the median and weighted-mean of FCs in gene .
There are several publications showing a good correspondence between fold change values in the Exon and the 3′IVT arrays ( e . g . [2] , [6] , [41] ) . These comparisons are usually done on a reduced set of genes with overlapping probeset locations . However , these analyses have not compared the relative ability of the platforms to detect differential expression in a supervised analysis . In part this is because the main focus in exon array analysis is the study of alternative splicing . Our work highlights how the analysis of differential expression is enhanced by using the probeset multiplicity offered by exon arrays . We took a novel approach to the handling of DABG p-values in the identification of alternative splicing events . Typically , when filtering is performed at all , probesets absent in more than a predefined number of samples are filtered out . We retain all exonic probesets per gene when calculating the alternative splicing metric , but weight their contribution by the number of present samples . This approach allows a continuous scoring of the reliability of the probesets based on the DABG p-values across the samples , avoiding an abrupt ‘in-or-out’ filtering . We also found that on a number of occasions , a single probeset was responsible for a gene being flagged as alternatively spliced , but that on further investigation , that probeset showed little change across a set of independent experiments , leading us to conclude that the findings were likely to be spurious . We first tested our methodology on a sample dataset for which predicted alternative splicing events where explored by RT-PCR and we were able to confirm that the inclusion of probeset reliability information in the VFC metric enables a better differentiation between the true and false positives . We then used the method to analyse our Head and Neck dataset and four hypoxia-associated alternative spliced candidates were identified ( SLCO1B3 , WDR66 , COL4A6 and LAMA3 ) . We further analysed and validated LAMA3 , which showed the strongest evidence . The finding was successfully confirmed by RT-PCR and an informed re-analysis of the original microarray data allowed probes matched to the LAMA3 transcripts to be identified and a hypoxia-associated , splice variant dependent prognostic relationship with outcome to be determined . Antibodies specific to the different splice variants of LAMA3 were not available , precluding analysis of the different LAMA3 transcripts at the protein level , but identification of the prognostic significance of expression of the LAMA3-A versus LAMA3-B splice variant illustrates the potential for alternatively spliced transcripts to act as biomarkers of disease . The additional information provided by splicing data has the potential to lead to improved specificity for particular tissues or conditions , over assays that do not discriminate between splice variants . This also emphasizes the importance of identifying specific splice variants when interpreting gene expression data . Cell line experiments at 1% hypoxia failed to demonstrate convincing hypoxic induction of LAMA3-A , with only low levels of hypoxia induced expression seen , despite confirmation of a transcriptional hypoxic response through SLC2A1 ( GLUT1 ) expression . Our initial stratification of samples for exon array analysis was based upon the expression of a gene signature of hypoxia associated genes; direct measurement of hypoxia in these tumours in vivo was not performed . Instead , additional confirmation of hypoxia status was carried out by investigating CAIX protein expression in histological sections . CAIX expression was indeed elevated in samples with high HS values supporting the use of the hypoxia associated gene expression score as a surrogate marker for tumour hypoxia , and supporting the hypothesis that differential LAMA3-A expression is related to tumour hypoxia . It may be that greater or more prolonged hypoxia , lower pH or lower glucose levels are required for LAMA3 induction in cell lines or that this simply represents differences between cell line experiments and the situation in tumour . LAMA3 has independently been shown to be regulated in human keratinocyte wound response experiments , using Cobalt Chloride to induce in this case instead of direct hypoxia exposure , and to have a hypoxia response element associated with the promoter for LAMA3-A [42] . This represents the likely mechanism underlying any hypoxia associated differential expression of this transcript . An earlier study however had shown decreased laminin-332 expression in human keratinocytes in response to 0 . 2% or 2% hypoxia exposure [43] . Laminin-332 is known to interact with several components of the extracellular matrix; particularly its interaction with Collagen VII has been shown to be vital for tumour development in skin cancers [44] . Our data would suggest that LAMA3 induction in HNSCC tumours is influenced by hypoxia but the lack of expression seen in our HNSCC cell lines implies that expression may also be dependant upon other factors found in tissues but not in cell culture . Hypoxia is inherently associated with treatment resistance and a more aggressive tumour phenotype [10] . It is possible that LAMA3-A expression is dependent upon factors related to this relationship rather than being independently hypoxia inducible . Whilst the exact pathways involved in the expression of this transcript are unclear this study emphasizes the importance of identifying individual transcript expression in future biomarker research . | Alternative splicing is the process by which cells express a set of different , but related , transcripts from a single gene . When translated , each transcript results in a different protein , resulting in additional cellular complexity . Affymetrix Exon microarrays , which feature multiple probesets targeting different locations throughout each gene , allow the changes in transcription that result from alternative splicing to be investigated in a single genome-wide assay . In addition , the increased number of probesets targeting each gene offers the potential to combine signals in order to increase statistical power , allowing smaller changes to be detected reliably . We developed a novel algorithm to exploit both these aspects of exon arrays and applied it to tumour hypoxia in clinical samples . Our method identified 4 potential transcript variants upregulated in hypoxic cancers , including a splice variant of the Laminin alpha 3 gene , which we were then able to validate by other methods . On further investigation , we found that expression of this particular isoform in head and neck cancers was a strong adverse prognostic factor for survival following primary surgical treatment . This shows that exon arrays can be used to identify clinically relevant splicing events with potential utility as prognostic biomarkers . | [
"Abstract",
"Introduction",
"Results",
"Materials",
"and",
"Methods",
"Discussion"
] | [
"genetics",
"and",
"genomics/functional",
"genomics",
"genetics",
"and",
"genomics/gene",
"expression",
"computational",
"biology/alternative",
"splicing",
"molecular",
"biology/bioinformatics",
"computational",
"biology/genomics",
"oncology/head",
"and",
"neck",
"cancers",
"ma... | 2009 | Exon Array Analysis of Head and Neck Cancers Identifies a Hypoxia Related Splice Variant of LAMA3 Associated with a Poor Prognosis |
For decades the soil nematode Caenorhabditis elegans has been an important model system for biology , but little is known about its natural ecology . Recently , C . elegans has become the focus of studies of innate immunity and several pathogens have been shown to cause lethal intestinal infections in C . elegans . However none of these pathogens has been shown to invade nematode intestinal cells , and no pathogen has been isolated from wild-caught C . elegans . Here we describe an intracellular pathogen isolated from wild-caught C . elegans that we show is a new species of microsporidia . Microsporidia comprise a large class of eukaryotic intracellular parasites that are medically and agriculturally important , but poorly understood . We show that microsporidian infection of the C . elegans intestine proceeds through distinct stages and is transmitted horizontally . Disruption of a conserved cytoskeletal structure in the intestine called the terminal web correlates with the release of microsporidian spores from infected cells , and appears to be part of a novel mechanism by which intracellular pathogens exit from infected cells . Unlike in bacterial intestinal infections , the p38 MAPK and insulin/insulin-like growth factor ( IGF ) signaling pathways do not appear to play substantial roles in resistance to microsporidian infection in C . elegans . We found microsporidia in multiple wild-caught isolates of Caenorhabditis nematodes from diverse geographic locations . These results indicate that microsporidia are common parasites of C . elegans in the wild . In addition , the interaction between C . elegans and its natural microsporidian parasites provides a system in which to dissect intracellular intestinal infection in vivo and insight into the diversity of pathogenic mechanisms used by intracellular microbes .
The intestine is a major route for pathogens to invade the body . Pathogens have evolved sophisticated mechanisms to exploit the host cell machinery of intestinal cells in order to survive and replicate in this environment [1–4] . For example , the bacterium Listeria monocytogenes uses the mammalian endocytic pathway to invade intestinal epithelial cells and then induces actin polymerization , which propels the invading bacteria both within and between host cells [5] . However , many questions remain about how pathogens interact with the intestine in vivo . For convenience , studies of intracellular pathogens of the intestine are often performed in tissue culture cells , which lack the characteristic features of intact intestinal cells . In vivo , intestinal epithelial cells are polarized and contain a “brush border” on their apical side . The brush border is decorated with finger-like microvilli , which are anchored by a conserved cytoskeletal structure called the terminal web . Little is known about what role these differentiated features play during the infection process or how they are manipulated by pathogens . The nematode C . elegans has become an attractive model for exploring host/pathogen interactions in the intestine . In its natural environment , C . elegans feeds on a variety of microbes . C . elegans does not appear to have professional immune cells , and therefore relies mainly on epithelial immunity to fight off microbial infections in the intestine . Several different microbial pathogens have been shown to infect and kill C . elegans in the laboratory and defense against these pathogens involves conserved innate immune signaling pathways [6 , 7] . Many pathogens infect the C . elegans intestine , which is composed of only 20 large epithelial cells that are easily visible because C . elegans is transparent [8] . These epithelial cells contain apical microvilli anchored into a terminal web of actin and intermediate filaments . Such morphological features are characteristic of intestinal cells in mammals , thus making C . elegans an excellent model for understanding interactions between enteric pathogens and intestinal cells . No pathogens have so far been described that reside intracellularly in C . elegans intestinal cells , even though several intracellular pathogens of mammals have been shown to establish lethal intestinal infections [9 , 10] . The failure of human intracellular intestinal pathogens to invade C . elegans intestinal cells is most likely a consequence of the fact that these pathogens are not natural pathogens of C . elegans . Little is known about the natural pathogens of C . elegans and so far , no pathogen has been isolated from wild-caught C . elegans individuals . To learn more about the ecological pressures on C . elegans and to develop an accessible in vivo model for intracellular infection of the intestine , we characterized a natural intracellular pathogen of C . elegans . We show here that this pathogen is a novel species of microsporidia . Microsporidia are obligate intracellular eukaryotic parasites most closely related to the fungi [11 , 12] . They infect a wide range of hosts , including vertebrate and invertebrate animals , as well as some protists . Microsporidia commonly infect insects , and have agricultural significance both as a parasite of honeybees , but also as a biocontrol agent for insect pests . There has been a recent surge of medical interest in microsporidia . At least 15 different microsporidian species have been shown to infect humans , and microsporidia have recently been added to the National Institutes of Health list of priority pathogens , as well as the Environmental Protection Agency list of waterborne microbial contaminants of concern [13] . Microsporidia frequently infect the intestine , which can cause self-limiting diarrhea in immunocompetent patients , but can lead to severe , persistent diarrhea in immunocompromised patients [14] . Microsporidia survive outside of their hosts as spores . Microsporidian spores are distinguished by a striking infection apparatus called the polar tube , which everts to pierce the membrane of the host cell [15] . The polar tube then acts as a syringe to directly inject nuclei and sporoplasm into the host . Once inside the host cell , microsporidia replicate in a cell-wall deficient form called a meront , which eventually differentiates to re-generate the spore form . Despite the medical and agricultural relevance of these parasites , little is known about the pathogenic mechanisms used by these ubiquitous microbes , including how microsporidian spores exit intestinal cells to go on to infect new hosts . In our study of a natural intracellular infection of the C . elegans intestine , we have discovered a new microsporidian species that we have named Nematocida parisii , which defines a new genus and a new species . N . parisii is transmitted horizontally , i . e . , from animal to animal . Infection with N . parisii proceeds through a distinct series of stages within the intestinal cell , starting with a meront stage , in which irregularly shaped microbes called meronts cause “grooves” in the intestine . Later in the infection when these meronts develop into spores , striking gaps appear in the terminal web underlying the intestinal microvilli . These changes occur when animals become infectious to others , so this may be part of an exit strategy for infectious microsporidian spores . Defense against microsporidian infection does not appear to involve the p38 mitogen-activated protein kinase ( MAPK ) or insulin signaling pathways , which are involved in defense against a variety of bacterial and fungal pathogens in C . elegans . In addition to the original microsporidian isolate we characterized , we have also found microsporidia infections in several other natural isolates of Caenorhabditis nematodes from diverse geographical locations . The C . elegans/N . parisii model provides a new system for exploring the specific mechanisms of microsporidia pathogenicity , as well as the more general strategies used by intracellular microbes to survive and replicate in the intestine of animals .
A wild-caught C . elegans strain isolated from a compost pit in Franconville , France ( near Paris ) was found to harbor small , rod-shaped microbes in its intestinal cells ( compare Figure 1C–1F to uninfected animals in Figure 1A ) [16] . These microbes are transmitted horizontally at very high efficiency from animal to animal: if donor “infected” animals are incubated on the same plate as uninfected “recipient” animals , 100% of the recipient animals become infected with the rod-shaped microbes ( n > 30 donors tested , for each donor , 20 recipients were examined ) . Because the infection is readily transferred to the standard N2 laboratory strain of C . elegans , further characterization has been performed in this strain background . The infection does not appear to be transmitted vertically , because noninfected progeny could be isolated from their hermaphrodite infected parents in two different ways . First , eggs could be isolated from infected parents by bleaching gravid adults , and these eggs developed into uninfected adults ( n = 263 animals examined ) . Second , manually separating progeny from their infected parents just before or immediately after hatching allowed them to develop into uninfected adults ( n = 27 animals examined ) . Because we have been unable to culture an infecting microbe ( s ) outside of the host ( see next section ) , we generate infectious extracts that can be used to reproducibly transmit the infection . Infected animals are mechanically disrupted , filtered , and then frozen as a glycerol stock ( see Materials and Methods for details ) . Using these extracts , all C . elegans postembryonic developmental stages have been shown to be susceptible to infection , except for dauer larvae , which do not feed . The first sign of infection appears within a day or two: distinct regions devoid of gut granules appear in the intestinal cells . These initial symptoms can arise at any point within the intestine . Early in infection , these regions are small circles , and later extend to longer grooves , as shown in Figure 1B . After the appearance of these grooves , rod-shaped microbes become visible ( 2 . 18 ± 0 . 15 μm long , 0 . 8 ± 0 . 08 μm wide , n = 47 , Figure 1C–1E ) . Subsequently , slightly larger rod-shaped microbes are sometimes also observed ( 3 . 17 ± 0 . 22 μm long , 1 . 31 ± 0 . 15 μm wide , n = 41 , Figure 1E ) . When the intestine becomes heavily infected with microbes , the rod-shaped microbes are found in discrete vesicles , as shown in Figure 1F . Eventually all infected worms die prematurely compared to noninfected controls ( see Figure 5A ) . While the exact cause of death is difficult to determine , one likely cause of death is that intestinal cells become completely filled with rod-shaped microbes and thus are no longer able to absorb and transmit nutrients . When animals are initially infected with rod-shaped microbes , they appear grossly normal by examination with a dissecting microscope . At 48 h postinoculation , animals with rod-shaped microbes in their intestines had normal feeding rates based on a pharyngeal pumping assay: uninfected animals pumped at 250 ± 70 pumps/minute ( n = 21 ) , while infected animals pumped at 247 ± 60 pumps/minute ( n = 18 ) . Later in the infection animals became more sluggish when their entire intestine became filled with rod-shaped microbes ( e . g . , Figure 1F ) . Infection occurred almost exclusively in the intestine , except in rare cases where heavily infected worms also had microbes in the anal region . To determine the identity of the intracellular microbes , we performed PCR on lysate from infected animals using universal rDNA primers . Infected animals were rinsed , lysed , and then used as template for PCR . The resulting PCR product was cloned , sequenced , and analyzed by BLAST . Using primers that hybridize to prokaryotic rDNA , we found sequence corresponding to two species of Gram-negative bacteria: Pseudomonas putida and a previously unknown Ochrobactrum strain , both of which could be cultured on defined media ( see Materials and Methods ) . However , following repeated treatment with antibiotics ( kanamycin , tetracycline , and gentamicin ) , a strain of animals was obtained that still contained a highly infectious intracellular pathogen , but no longer contained any detectable bacteria , as determined by culturing or by PCR ( see Figure S1 and Text S1 for more information on Ochrobactrum ) . When PCR was performed on lysate from infected animals using primers that hybridize to both prokaryotic and eukaryotic rDNA , sequence was obtained that corresponds to the phylum Microsporidia , which comprises eukaryotic intracellular pathogens . This rDNA sequence is most closely related to rDNA sequence from Ovavesicula popilliae , which is a beetle-infecting microsporidian species [17] , although the two sequences are fairly divergent ( see phylogenetic comparison in Figure 2 ) . Because of this divergence and the fact that the putative microsporidian species infecting C . elegans is also morphologically distinct ( O . popilliae spores are larger and oval-shaped ) we gave the C . elegans microbe a new genus and species name: Nematocida parisii for nematode killer from Paris ( see Taxonomic Summary for the generic and species descriptions ) . Because conditions have not been developed to culture microsporidia independently of host cells , we used RNA fluorescence in situ hybridization ( FISH ) , which is a culture-independent method , to determine that N . parisii corresponds to the C . elegans intracellular microbes . As shown in Figure 3A , oligonucleotide FISH probes for N . parisii rRNA showed strong , specific staining in the intestinal grooves of infected animals , corresponding to the stage of infection shown in Figure 1B . This staining indicated that the intracellular microbe corresponds to a microsporidian species . Moreover , 4′ , 6-diamidino-2-phenylindole ( DAPI ) counterstaining to label DNA often revealed several spots within one FISH-staining cluster in the grooves , indicating these were multinucleate meronts ( Figure 3D ) . As controls , a universal bacterial probe for rRNA did not label infected worms ( Figure 3B ) , nor did the N . parisii-specific probes label intestinal cells of uninfected worms ( Figure 3C ) . After microsporidian meronts replicate , they form a thick cell wall and differentiate into the spore form , which is often oval or rod-shaped . FISH staining of infected animals ( corresponding to the stage of infection shown in Figure 1E ) resulted in labeling of the rod-shaped structures , both small and large ( Figure 3E and 3F ) , which likely correspond to N . parisii spores . We refer to microbes in the groove stage as “meronts” and microbes in the rod-shaped stage as “spores” by analogy with other microsporidian species . Transmission electron microscopy ( TEM ) analysis was used to more closely examine N . parisii at different stages of infection . During the groove stage of infection multinucleate meronts are visible as irregularly shaped structures ( compare Figure 4B and 4C to uninfected animal in Figure 4A ) . In several cases , meronts appear to be attached to intracellular vesicles: of 124 meronts examined at 30 to 32 h postinoculation , 48% have curvature around at least one-quarter of the circumference of an intracellular vesicle ( six sections examined from two different experiments ) . While it is difficult to assign identity to these vesicles without immunostaining , one possibility is that they contain nutrients that the microsporidia are able to absorb ( note the lack of distinct membrane between meront and host vesicle in Figure 4C ) . Later in the infection , multinuclear structures with more regularly shaped membranes are observed , which are also likely meronts ( Figure 4D ) . In both the early meront form ( Figure 4C ) and the late meront form ( Figure 4D ) , meronts could contain one or multiple nuclei , which did not appear to be paired , suggesting this species is monokaryotic . Later in the infection , a variety of microbe shapes are observed . These include microbes that have shapes intermediate between meronts and spores , which are likely to be sporoblasts ( precursors to spores ) ( Figure 4E and 4F ) . Also visible are microbes that appear to be more differentiated spores ( Figure 4G–4L ) . As microbes differentiate into spores , a structure likely to be the polar tube infection apparatus becomes visible ( Figures 4G–4I ) . Also visible are lamellar membranes that likely are the polaroplast membranes , in which the polar tube is positioned . Polar tubes in many microsporidian species are so long that they wrap around several times near the posterior end of the spore and are visible as coils in cross-section . In smaller sized N . parisii spores , one cross-section of the polar tube coil was sometimes observed near the posterior end , often at the interface of a punctate region ( perhaps containing ribosomes ) and a more uniform region ( e . g . , Figure 4I ) . Smaller-sized spores as shown in Figure 4I , 4J , and 4L averaged 1 . 89 ± 0 . 24 μm long and 0 . 54 ± 0 . 07 μm wide , as measured by TEM ( n = 65 ) . Larger-sized spores as shown in Figure 4K had as many as five polar tube coils visible , often near the edge of the spore . Posterior vacuoles were sometimes observed in spores as well ( Figure 4I and 4J ) . In general , these morphological characteristics are considered diagnostic for identification of microsporidian infections [11] . Also observed were membranes forming around clusters of mature spores ( Figure 4L ) . To precisely measure the kinetics of infection we inoculated a staged population of wild-type animals with infectious extract and then examined for different stages of infection over several days by differential interference contrast ( DIC ) microscopy . A typical time course is illustrated in Figure 5A . Meronts appear within the first day , and spores are visible a day later , followed by vesicles filled with spores . Animals begin dying around the fourth day of infection . In many assays , a biphasic curve is observed for spore infection . The first part of the curve ( before about 60 h ) may reflect animals infected by the initial infectious dose , with the second phase ( after about 60 h ) reflecting animals being infected by their contagious neighbors . Next we determined when the infection was transmitted from donor animals . “Donor” infected animals were incubated on the same plate as “recipient” uninfected animals for 12 h , then donor animals were removed from the plate and examined by DIC microscopy for the presence of meronts or spores . All donor animals were alive and active when removed from the plate and their intestinal cells appeared intact by light microscopy . Two days later recipients were examined to determine whether they had received the infection ( Figure 5B ) . We found that animals infected only with meronts were not infectious to others . In contrast , a majority of donor animals with spores transmitted the infection , including some animals that only had a small number of spores visible . The fact that animals with only a small number of spores could transmit the infection suggests that infectious spores can exit intestinal cells without causing much damage , since donor animals had grossly normal intestinal cells as assessed by light microscopy . We also found that animals containing only the small-sized spores could transmit the infection to others and this transmission resulted in recipients that were infected with both small and large-sized spores ( n = 3 donors ) . To investigate how spores escape from intestinal cells , we examined structural features of intestinal cells infected with spores . Electron microscopic analysis of infected intestines indicated that N . parisii infection causes specific damage to the terminal web , a cytoskeletal structure found in many polarized epithelial cells . The terminal web is thought to be composed of actin and intermediate filaments that provide structural support for the actin-rich microvilli [18] . A normal terminal web in C . elegans intestinal cells can be seen as an electron-dense structure underlying the finger-like microvilli , as shown in Figure 6A . In electron micrographs from N . parisii infected animals , parts of the terminal web appear to be missing ( Figure 6B ) , although the microvilli appear largely intact , and cells contain normal-looking apical junctions ( unpublished data ) , which are cytoskeletal structures used to connect neighboring intestinal cells . To further examine this apparent structural damage to the terminal web , immunofluorescence was performed on infected worms to label components of the terminal web with specific molecular markers . The monoclonal antibody MH33 specifically labels the intermediate filament protein IFB-2 , which is a major component of the terminal web in C . elegans intestinal cells [18] . IFB-2 immunostaining in animals infected with spores revealed gaps in the normally continuous sheet corresponding to the terminal web ( compare Figure 6C and 6D ) . To determine whether other cytoskeletal structures are damaged in infected cells , we examined the cytoskeleton membrane linker protein , ERM-1 , which is localized to the apical side of intestinal cells [19] , likely in the microvilli . Visualizing an ERM-1::green fluorescent protein ( GFP ) translational fusion in animals that were also immunostained against IFB-2 ( Figure 6E and 6F ) showed that ERM-1::GFP had occasional gaps in staining , but these were much less consistent and less dramatic than the gaps in IFB-2 immunostaining . These results suggest that N . parisii infection causes more damage to intermediate filaments and the terminal web than to microvilli , consistent with electron micrographs . Because these intermediate filament gaps occur in animals infected with spores , which is the infectious stage , these gaps may be an exit strategy for N . parisii spores . To rule out the possibility that the observed gaps in the terminal web in N . parisii infected animals are caused by nonspecific damage to the intestine , we examined IFB-2 immunostaining in animals infected with other intestinal pathogens . The Gram-negative bacterial pathogen P . aeruginosa and the Gram-positive bacterial pathogen Staphylococcus aureus kill C . elegans rapidly via intestinal infection ( time to 50% animals dead [TD50] approximately 50 h ) [20 , 21] . IFB-2 immunostaining was continuous in P . aeruginosa infected animals ( Figure 6G ) and S . aureus infected animals ( Figure 6H ) even in those that were near death . Therefore , gaps in the terminal web appear to be a specific consequence of N . parisii infection , and not the result of nonspecific damage by intestinal pathogens in general . The C . elegans p38 MAPK mutant pmk-1 is significantly more sensitive to killing than wild-type worms by most bacterial and fungal pathogens tested to date [21–24] . However , pmk-1 mutants were not found to be infected more rapidly by N . parisii as assessed by microscopy , either when infected as first stage ( L1 ) larvae ( Figure 7A ) or as fourth stage ( L4 ) larvae ( Figure 7C ) . pmk-1 mutants appeared to be modestly more susceptible to killing by N . parisii early in infection ( Figure 7B and 7D ) . This slight sensitivity occurs at a time when pmk-1 mutants die slightly more quickly than wild-type animals in the absence of infection ( Figure 7B ) , suggesting the slight sensitivity is not specific to N . parisii . This subtle phenotype is distinct from the robust immunocompromised phenotype seen with pmk-1 mutants when infected with a variety of other pathogens [21–24] . The daf-2/daf-16 insulin/insulin-like growth factor ( IGF ) signaling pathway regulates resistance to a variety of pathogens in C . elegans [25] and appears to act in parallel to the p38 MAPK pathway [26] . Mutations in the daf-2 insulin receptor cause animals to be resistant to many pathogens , and this enhanced resistance requires the downstream daf-16 FOXO transcription factor . Because daf-2 is required for normal larval development , we could not test the role of daf-2 in early larvae , and instead examined the role of daf-2 in older animals ( L4 larvae ) . We found that neither daf-2 nor daf-2;daf-16 mutants appeared to have substantially altered resistance to N . parisii infection as assessed by microscopy ( Figure 7E ) . daf-2 mutants did survive somewhat longer when infected with N . parisii , and this resistance required daf-16 , since daf-2;daf-16 mutants did not survive longer than wild-type animals ( Figure 7F ) . However , daf-2 mutants live more than twice as long as wild-type animals in the absence of infection , so it is difficult to say whether this effect is due to resistance against N . parisii , or is due to a general increase in lifespan . The data in Figure 7 indicate that the PMK-1 p38 MAPK and the DAF-2/DAF-16 pathways , which are important for defense against pathogens like P . aeruginosa and S . aureus , do not play a significant role in defense against microsporidian infection . Consistent with this conclusion , we found that many genes robustly induced by P . aeruginosa and S . aureus were not robustly induced by N . parisii ( Figure S2 ) ( J . I . Irazoqui , E . R . Troemel , F . M . Ausubel , unpublished data and [26 , 27] ) . In addition , nlp-31 , which is a gene induced by the pathogenic fungus Drechmeria coniospora [28] , was not induced by N . parisii ( unpublished data ) . Similarly , a D . coniospora-induced nlp-29::GFP reporter was not induced by N . parisii infection ( unpublished data ) . These results suggest that the C . elegans response to N . parisii is distinct from responses to previously studied fungal and bacterial pathogens . We found several other wild-caught nematodes from a variety of geographical locations that harbored intestinal rod-shaped microbes similar in appearance to N . parisii . Infected nematodes were isolated from multiple regions of France , from Portugal , and from India ( see Materials and Methods ) . These infections were observed directly in individuals coming from the wild , and infected animals were found at a variety of developmental stages ( Figure 8 ) . JU1247 is a wild-caught C . elegans strain isolated from a park near Paris that is about 30 km from the original microsporidia-infected isolate ( found in Franconville ) . JU1395 is a strain of C . elegans isolated from a compost heap in Montsoreau , France , which is 300 km south of Paris . Both of these C . elegans strains harbored intracellular infections in their intestinal cells with similar characteristics to N . parisii , i . e . , they exhibited grooves and had rod-shaped microbes of two distinct sizes . rDNA fragments were isolated from infected JU1247 and JU1395 animals that exactly matched the sequence of N . parisii ( unpublished data ) . In addition , we isolated JU1348 , a strain of C . briggsae from a nature preserve in Kerala , India , which also has an intestinal microsporidian infection with characteristics similar to N . parisii ( grooves and rod-shaped microbes of two distinct sizes ) . A fragment of rDNA was isolated from this strain that is approximately 95% identical to N . parisii rDNA . This rDNA sequence is most closely related to N . parisii in phylogenetic analysis ( Figure 2 ) , and we refer to it as Nematocida sp . 1 . In order to confirm that the three Caenorhabditis strains mentioned above are infected with microsporidia , we performed RNA FISH using an N . parisii probe ( which hybridizes to a region that is identical in both N . parisii and Nematocida sp . 1 . rRNA ) and saw positive signal for all three strains tested ( Figure 8G–8I ) . Microsporidia have been placed into five major clades , based on the phylogenetic analysis of Vossbrinck and Debrunner-Vossbrinck [29] . We have performed phylogenetic analyses using three independent methods and broadly recover these five clades from our analyses ( Figure 2 and Materials and Methods ) . In all three analyses , the two Nematocida sequences ( N . parisii from France and Nematocida sp . 1 from India ) group together and share a most recent common ancestor . The Nematocida lineage was always placed sister to O . popilliae with high support in all three analyses . These three sequences were in turn placed sister to a group comprising Paranosema whitei , P . grylli , Antonospora locustae , and A . scoticae in all three analyses . The group comprising all of these sequences corresponds to one of two major lineages within Clade II of Vossbrinck and Debrunner-Vossbrinck [29] .
Here we describe a natural intracellular parasite of the nematode C . elegans , which we show is a new species of microsporidia and have named N . parisii . To our knowledge , N . parisii is the first pathogen isolated directly from a wild-caught C . elegans strain , and is the first pathogen shown to invade and reside within C . elegans intestinal cells . Our discovery of multiple natural isolates of C . elegans infected with microsporidia presents a new perspective on the challenges that C . elegans faces in its natural habitat . The lifecycle of N . parisii infection of C . elegans is modeled in Figure 9 . Once nematodes have ingested N . parisii spores , these spores likely use a polar tube infection apparatus to directly inject the intestinal host cells with microsporidia nuclei and sporoplasm . While this event has not been directly observed with N . parisii , it is quite likely to occur , because a presumptive polar tube was observed in TEM cross-sections of N . parisii spores , and since all microsporidian species are thought to use a polar tube to inject host cells with spore material . Material directly injected into host intestinal cells form multinucleate meronts , which appear in the C . elegans intestine as irregularly shaped microbes by electron microscopy and stain with a microsporidian-specific RNA FISH probe peppered with DAPI-staining nuclei . Replication of meronts displaces or consumes gut granules inside the intestinal cells , causing “grooves , ” which are visible by light microscopy . These meronts eventually differentiate into spores , which show the characteristic features of microsporidia , including a polar tube . When spores have formed , gaps become visible in the terminal web . At this time , animals become infectious to others , likely through shedding of virulent spores . We found that C . elegans p38 MAPK pmk-1 mutants do not have substantially altered resistance to N . parisii infection , whereas pmk-1 mutants are highly susceptible to a wide of variety of human bacterial and fungal pathogens that have been tested [21 , 22 , 24] . One possible explanation for this observation is that because of a long-term evolutionary arms race with nematodes , N . parisii has evolved mechanisms to suppress PMK-1-mediated immune responses , analogous to the suppression of basal immune responses in plants mediated by Type III secretion system effectors of highly evolved plant pathogens such as P . syringae [30–32] . Alternatively , the lack of p38 MAPK involvement in resistance against microsporidia may be due to the intracellular lifestyle of N . parisii , which could allow this pathogen to evade detection of immune receptors or evade the activity of antimicrobials controlled by the p38 MAPK pathway that are secreted into the intestinal lumen [26] . The extracellular stage of N . parisii is the spore form , which has a tough coat that could be resistant to antimicrobials active in the lumen . We also examined the daf-2/daf-16 insulin/IGF signaling pathway for its role in resistance to N . parisii infection . While the daf-2/daf-16 pathway did not appear to affect the rate of infection by N . parisii , it did affect the survival of infected animals . daf-2 insulin receptor mutants were modestly resistant to infection , and this required the downstream transcription factor daf-16 . It may be that in daf-2 mutants , hyperactivation of daf-16 induces expression of genes that are able to restrict the damage caused by N . parisii infection and thus allow animals to survive longer . DAF-16 controls expression of hundreds of genes , including antimicrobials and cellular stress response genes [33] . However , because daf-2 mutants are generally long-lived , it is difficult to say whether the effect of daf-2 is truly due to pathogen resistance , or to a more general effect on viability . Despite the widespread use of C . elegans as a model organism for many biological processes including innate immunity , little is known about the natural ecology of this animal . C . elegans can be found in the soil , often on rotting fruit , and is thought to undergo a “boom and bust” lifestyle [34] . When food becomes available population levels rapidly increase , and when those food sources are exhausted , population levels crash and animals arrest in varying larval stages , which are able to withstand harsh conditions for long periods of time . In this context , N . parisii is particularly well suited to persist . During times of high population density , N . parisii can rapidly be transmitted from animal to animal . Once a N . parisii infection has been established , the infection can persist in arrested larvae ( Figure 8A and unpublished data ) to survive through times of less food and lower C . elegans population levels . N . parisii appears to be a widespread parasite of Caenorhabditis nematodes , as we found multiple natural isolates in France from widely separated localities that are infected with N . parisii as determined by rDNA sequence analysis . It will be interesting to examine the specificity of interaction between host/parasite pairs from different geographical regions , as well as the defense pathways used against these pathogens . The phylogenetic analyses we performed on N . parisii and the related Nematocida species from C . briggsae indicate that they are unequivocally members of the microsporidia . The Nematocida lineage is closely related to O . popilliae , and tentatively to a group comprising Paranosema and Antonospora , which are members of Clade II [29] . The sister relationship between O . popilliae and the Paranosema+Antonospora group was first suggested by Vossbrinck and Andreadis [17] and our analysis suggests that Ovavesicula+Nematocida is sister to Paranosema+Antonospora . Interestingly , Vossbrinck and Andreadis [17] point out that O . popilliae , Paranosema , and Antonospora are all pathogens of terrestrial insects and that this group likely represents an independent origin of insect parasitism within the microsporidia ( the other insect-infecting clade includes Nosema and Vairimorpha species within Clade IV of Vossbrinck and Debrunner-Vossbrinck [28] ) . Our analyses suggest that species in this group also attack terrestrial nematodes in the genus Caenorhabditis . It is noteworthy that Nematocida and Ovavesicula are both found in soil-dwelling invertebrates . One intriguing feature of N . parisii microsporidian infection is that nematodes can sustain an impressive microbial load—virtually all of the intestinal cells can be filled—while still moving and appearing grossly normal . A recent study of malaria infection in mice suggested that animals tend to evolve either resistance ( the ability to limit parasite burden ) or tolerance ( the ability to limit the damage caused by a given parasite burden ) [35] , a concept that has also been studied in plants [36 , 37] and Drosophila [38 , 39] . It seems likely that C . elegans has evolved a strategy of tolerance in response to microsporidian infection . For example , in lieu of destroying intestinal cells to exit the host , N . parisii infection appears to cause a subtle restructuring of the intestinal cells . Pathogens often evolve to cause only specific damage in order to minimize their impact on the host [40 , 41] . Evolutionary pressure for pathogen and host to coexist is likely to be especially strong in the case of obligate intracellular pathogens , which are completely dependent on their host to survive and replicate . The restructuring of the terminal web caused by N . parisii infection in C . elegans may also occur in mammalian microsporidian infections , but has gone unrecognized because pathogens were assumed to exit only through the regular shedding of intestinal cells . Perhaps other intracellular pathogens of the intestine use a similar strategy to manipulate the intestinal cytoskeleton and then exit host cells . Several pathogens have been shown to manipulate the host cytoskeleton: for example Listeria has been shown to polymerize host actin in order to propel itself through the cell and infect neighboring cells [5] . However , little is known about pathogen interactions with intermediate filaments , which may be the target of N . parisii infection . There is evidence that other microsporidian infections may alter the structure of host cytoskeletal networks , perhaps via microsporidia-expressed intermediate filaments [42] . In addition to providing insight into the ecological pressures on the nematode C . elegans , the discovery of the microsporidian N . parisii may provide practical applications for agricultural pests . Parasitic nematodes are responsible for significant damage to multiple crop plants around the world and there is a need for more environmentally friendly management strategies [43 , 44] . Perhaps N . parisii could be used as a biocontrol agent to limit the spread of such parasitic nematodes . Microsporidia have already been used successfully as biocontrol agents for insects: in particular , the microsporidian A . locustae , which is in the same clade as N . parisii ( Figure 2 ) , is sprayed onto agricultural fields to control grasshoppers ( Orthoptera ) [45 , 46] . The discovery of N . parisii infection of C . elegans also provides a relatively inexpensive whole animal system in which to develop treatments for microsporidian infections in humans . Microsporidia are increasingly appreciated to be a serious medical problem . For example , there are no treatments available for infection by the species Enterocytozoon bieneusi , which is responsible for most of the microsporidian infections in humans [47] . Because microsporidia are obligate intracellular pathogens , screening for antimicrosporidia drugs requires the host to be present . The C . elegans/microsporidia model we have developed is an important advance in this regard . C . elegans are tiny ( 1-mm-long ) hosts that can be used in high throughput screens . High throughput screens for antimicrobial compounds using C . elegans as a host have recently been developed in our laboratory and could be adapted to screen for molecules that prevent or cure microsporidian infection [48] .
Nematodes were isolated as described [49] . Briefly , compost or rotting fruit samples were placed onto a C . elegans culture dish around an Escherichia coli OP50 lawn and individuals were isolated as they came out of the sample . Some individuals were immediately observed by Nomarski optics ( Figure 6A–6D ) . The original N . parisii-infected strain , CPA24 , was isolated in 2004 from a compost pile in Franconville , which is about 15 km north/northwest of Paris , France . The JU1247 , JU1248 , and JU1256 strains were established from infected individuals ( a L2d larva , an egg-laying defective adult and a L2d larva , respectively ) isolated from a rotting apple sampled on October 14 , 2007 in a natural regional park in Santeuil , which is 50 km north/northwest of Paris . The population of Caenorhabditis nematodes in this apple was proliferating ( non-dauer stages ) and contained many C . elegans individuals and at least one C . briggsae individual that were infected by microsporidia: out of a total of 17 animals isolated , at least eight were infected based on examination by Nomarski ( three more died without being examined ) . Similar infections were also observed in two proliferating C . elegans populations from a Ficus isophlebia and an unidentified fruit in the Botanical Garden in Lisbon , Portugal ( July 2005 ) and in C . elegans from mushroom compost outside a farm in Montsoreau , Maine-et-Loire , France ( 300 km south of Paris; March 2008 ) . Strain JU1395 was established from an infected Montsoreau isolate . Out of a total of 14 locations sampled from 2006–2008 , infected C . elegans were found in four locations ( Franconville , France; Lisbon , Portugal; Santeuil , France; and Montsoreau , France ) . Infected C . elegans were not found in the other ten locations ( Hermanville , France; Le Blanc , France; Le Perreux , France; Primel-Sainte-Barbe , France; Merlet , France; Kakegawa , Japan; Concepcion , Chile; Sevilla , Spain; Carmona , Spain; Barcelona , Spain ) . JU1348 is a microsporidian-infected C . briggsae strain established from a sample obtained in the Periyar Natural Preserve in Kerala , India . This area was one of eight areas sampled in Kerala , India where a Caenorhabditis species was found; infected animals were not found in the other seven areas . ( Uninfected C . briggsae and other uninfected Caenorhabditis isolates were found in Estuary Island , Poovar; Botanical Garden , Trivandrum; near Meenmutti Waterfalls; Ponmudi Natural Preserve; Plantation near Kanjirapalli; Angela Spice Garden near Periyar . Uninfected C . brenneri were found in Allepey ) . rDNA was either isolated from pure cultures of bacteria as described below , or directly from infected worms . To isolate bacterial 16S rDNA from infected worms , one to two infected worms were placed in a PCR tube with single egg/worm lysis buffer ( SEWLB ) ( 10 mM Tris-HCl [pH 8 . 3] , 50 mM KCl , 2 . 5 mM MgCl2 , 0 . 045% NP40 , 0 . 045% Tween 20 , 20 ng/μl proteinase K ) , and washed six to eight times with SEWLB over the course of an hour . Then , worms were lysed with a proteinase K treatment at 65 °C for 30 min to 1 h , followed by 15 min at 95 °C to inactivate the proteinase K . This extract was then subject to PCR using universal 16S primers . Bacterial 16S rDNA was isolated with the following primer sets: 8F ( AGAGTTTGATCCTGGCTCAG ) and 1492R ( GGTTACCTTGTTACGACTT ) [50]; 16SUF ( CCGAATTCGTCGACAACAGAGTTTGATCCTGGCTCAG ) and 16SUR ( CCCGGGATCCAAGCTTACGGCTACCTTGTTACGACTT ) [51]; 9FA ( GAGTTTGATCITIGCTCAG ) and 1513R ( TACIGITACCTTGTTACGACTT ) ( Jacob Russell , personal correspondence ) ; N331F ( TCCTACGGGAGGCAGCAGT ) and N797R ( GGACTACCAGGGTATCTAATCCTGTT ) [52]; 337F ( CTCCTACGGGAGGCAGCAG ) and 1100R ( AGGGTTGCGCTCGTTG ) [53] . With these primers , rDNA sequence was isolated that corresponded to an Ochrobactrum sp . and P . putida , which were isolated as described below . For isolation of microsporidian rDNA , infected worms either were treated as above or were disrupted with silicon beads , the extract was filtered through Whatman filter paper number 1 and then subjected to PCR . Microsporidian sequence was isolated with 530F ( GTGCCAGCMGCCGCGG ) [53] and 1391R ( GACGGGCGGTGWGTRCA ) [54] . Further sequence was obtained with Micro308F ( CCGGAGARGGAGCCTGAGA ) , which was designed based on alignment of other microsporidia species , and Micro648R ( CGGTTCCGCACGGGCATC ) , which is specific to N . parisii , to obtain a 1 , 424-bp contig . A buffer-only PCR was always performed in parallel as a negative control , and samples were only used for cloning and sequencing if the negative control gave no signal by gel electrophoresis . PCR products were cloned into the TOPO TA vector ( Invitrogen ) and inserts were sequenced using primers that flank the insert of the TA vector . Sequences were analyzed by BLAST ( and by the Ribosomal Database Project to find their closest match if they were bacterial ) . Infected worms were washed with M9 buffer several times , and then disrupted by vortexing with silicon beads . Extract was plated on several types of rich media , including tryptic soy agar , nutrient agar , and brain heart infusion agar . For universal 16S PCR 1 μl of overnight culture was mixed with 4 μl of SEWLB buffer , incubated at 65 °C for 30 to 60 min , then 95 °C for 15 min to inactivate the protease . This lysate was then used in PCR reactions . Pure cultures of an Ochrobactrum sp . and P . putida were obtained with this method . Feeding these bacteria to uninfected worms , either alone , in combination , or in a variety of dilutions did not confer the intracellular infection . The P . putida infection was lost through regular propagation of infected worms on E . coli strain OP50 , however , Ochrobactrum sp . could only be removed from infected worms by repeatedly incubating worms in buffer containing the antibiotic gentamicin ( 15 μg/ml ) for 1–2 h at a time over several generations ( see Text S1 for further information on Ochrobactrum ) . C . elegans intestines were dissected as described [55] . Briefly , adults were transferred to a drop of M9 with levamisole on a microscope slide and the heads and tails were cut with a 25-gauge needle to extract the intestines . Samples were transferred to a microfuge tube and fixed with 4% paraformaldehyde for 1–2 h . FISH was then performed essentially as described for bacteria [56] . Samples were washed with PBS + 0 . 1% Tween 20 and then transferred to hybridization buffer ( 900 mM NaCl , 20 mM Tris [pH 7 . 5] , 0 . 01% SDS ) containing 5 ng/μl probe . Probes were designed against regions of the ribosomal sequence specific to N . parisii and were synthesized with a Quasar 570 ( Cy3 ) 5′ modification and HPLC purified by Biosearch Technologies , Inc . Two N . parisii-specific probes were tested and gave similar results: MicroA ( CTCTGTCCATCCTCGGCAA ) and MicroB ( CTCTCGGCACTCCTTCCTG ) . These probes also cross-react with the Nematocida sp . 1 sequence isolated from JU1348 , the infected C . briggsae strain from India . The universal bacterial probe EUB338 ( GCTGCCTCCCGTAGGAGT ) was synthesized the same way . Hybridization was performed at 46 °C overnight . Intestines were washed at 48 °C for 1 h in wash buffer ( 900 mM NaCl , 20 mM Tris [pH 7 . 5] , 0 . 01% SDS , 5 mM EDTA ) and then mounted for microscopy with Vectashield containing DAPI ( Vector Laboratories ) . For Figure 3A–3C , staining was performed in parallel and exposure times were the same for all . FISH staining was also performed as above , except without intestine dissection and with acetone fixation for 15 min at room temperature [57] instead of paraformaldehyde fixation . These conditions allowed for better staining of spores and were used for the FISH staining shown in Figure 3E and Figure 8G–8I . In order to place Nematocida in phylogenetic context small subunit ribosomal RNA sequences from 57 placeholder taxa ( downloaded from GenBank ) were chosen from each of the major clades of microsporidia based on the phylogenetic hypotheses of Vossbrinck and Debrunner-Vossbrinck [29] . This included two fungi ( Basidiobolus ranarum and Conidobolus coronatus ) sensu Vossbrinck and Debrunner-Vossbrinck [28] as outgroups , which is consistent with the hypothesis that the Microsporidia's closest living relatives are fungi . We included the N . parisii sequence and the Nematocida sequence ( Nematocida sp1 . ) isolated from Indian C . briggsae strain JU1348 . The small subunit ribosomal RNA sequences were aligned using the default parameters in the Clustal W2 web interface ( http://www . ebi . ac . uk/Tools/clustalw2/index . html ) [58] , and the alignment was visually inspected ( the alignment is available upon request ) . We trimmed all unalignable insertions and deletions from the entire alignment . The resulting alignment comprised 940 characters and 59 taxa . We then used the three mostly widely used phylogenetic methods ( Bayesian inference , maximum likelihood , and parsimony ) to place the Nematocida accessions phylogenetically . To evaluate nodal support for evolutionary relationships among the 59 taxa , we performed a parsimony bootstrap heuristic search in Paup ( phylogenetic analysis using parsimony [and other methods] , version 4 . 0b10; Swofford , D . L . 4 . 0 Beta . Sinauer Associates , Inc . ) [59] , with 1 , 000 replicates using the 59 taxa alignment . Of the 940 total characters , 255 were constant , 69 variable characters were parsimony uninformative , and 616 were parsimony informative . Gaps were treated as missing data . Starting trees were obtained via stepwise addition , with simple sequence addition ( reference taxon was Encephalitozoon lacerate ) , the branch-swapping algorithm was tree-bisection-reconnection and the steepest descent option was not in effect . A bootstrap consensus tree was computed from this heuristic search . We determined the most likely model of nucleotide substitution across these sequences using the Modeltest program , version 3 . 71 [60] . The TrN+I+G model was chosen as the most likely ( −lnL = 18 , 797 ) . We then used the PhyML [61] algorithm through the PhyML 3 . 0 webserver ( http://www . atgc-montpellier . fr/phyml/ ) . This program is based on the algorithm of Guindon and Gascuel [62] and constructs an initial tree using distance methods and then performs a phylogenetic search using the maximum likelihood optimality criterion . It is especially useful with large datasets . We chose the GTR model of nucleotide substitution ( the TrN+I+G model is derived from the GTR model ) in PhyML and performed 100 bootstrap replicates on the dataset to determine support values for relationships in the phylogeny ( on the single most-likely tree , −lnL = 18 , 727 ) . We used the program MrModeltest 2 . 2 [63] to estimate the most likely model of sequence evolution ( MrBayes utilizes a subset of the models used by Paup ) prior to estimating the phylogeny using Bayesian inference . The GTR+I+G model of nucleotide substitution was chosen as the most likely ( −lnL = 32 , 299 ) . We then used the program MrBayes 3 . 1 . 2 [64] to estimate the phylogeny using Bayesian inference , which also yields posterior probabilities of nodes in the phylogeny using a Markov chain Monte Carlo ( MCMC ) approach . The analysis was performed using two independent runs , with four chains each ( temp = 0 . 2 ) , and each chain ( sampled every 100 ) was run for ten million generations . After completion of the run , we examined convergence rates of posterior split probabilities in the MCMC using the program AWTY [65] . Convergence was obtained after ∼6 million generations . A consensus tree for each run was obtained that gave average branch lengths and posterior probability values; the two consensus trees were topologically identical . In accordance with section 8 . 6 of the ICZN's International Code of Zoological Nomenclature , we have deposited copies of this article at the following five publicly accessible libraries: National Museum of Natural History , Smithsonian Institute , Washington ( D . C . ) , United States of America; Museum of Comparative Zoology , Harvard University , Cambridge , Massachusetts , United States of America; Museum National d'Histoire naturelle , Paris , France; California Academy of Sciences , San Francisco , California , United States of America; San Diego Natural History Museum , San Diego , California , United States of America . The new genus and species names established herein have been registered in ZooBank [66] , the official online registration system for the ICZN . The ZooBank publication LSID ( Life Science Identifier ) for the new species described herein can be viewed through any standard web browser by appending the LSID to the prefix http://zoobank . org/ . C . elegans were fixed in 2 . 5% glutaraldehyde , 1 . 0% paraformaldehyde in 0 . 05 M sodium cacodylate buffer , ( pH 7 . 4 ) plus 3 . 0% sucrose . The cuticles were nicked with a razor blade in a drop of fixative under a dissecting microscope to allow the fixative to penetrate . After 1 h fixation at room temperature , the worms were fixed overnight at 4 °C . After several rinses in 0 . 1 M cacodylate buffer , the samples were postfixed in 1 . 0% osmium tetroxide in 0 . 1 M cacodylate buffer for 1 h at room temperature . They were rinsed in buffer and then in double distilled water and stained , en bloc in 2 . 0% aqueous uranyl acetate for 1 h at room temperature ( for lighter staining of mature spores , this first uranyl acetate staining step was sometimes omitted ) . After rinsing in distilled water , the last rinse was carefully drawn off and the worms were embedded in 2 . 0% agarose in PBS for ease of handling . The agarose blocks were dehydrated through a graded series of ethanol to 100% , then into 100% propylene oxide , and finally into a 1:1 mixture of propylene oxide:EPON overnight on a rocker . The following day , the agarose blocks were further infiltrated in 100% EPON for several hours and then were embedded in fresh EPON overnight at 60 °C . Thin sections were cut on a Reichert Ultracut E ultramicrotome and collected on formvar-coated gold grids . They were poststained with uranyl acetate and lead citrate and viewed in a JEOL 1011 TEM at 80 kV equipped with an AMT digital imaging system ( Advanced Microscopy Techniques ) . To measure spore size by light microscopy , infected nematodes containing spores were photographed using Nomarski optics on a Zeiss AxioImager microscope and then spores were measured using the measurement function on Zeiss Axiovision software . Values given are the average ±SD . To measure spore size by TEM , sections of infected nematodes were photographed and then measured using AMT version 5 software . Infected or noninfected animals were placed individually on NGM plates seeded with OP50 and allowed to acclimate for 10–30 min . Videos of pharyngeal pumping were then recorded of each animal for 15–60 s using a Sony Handycam attached to a Zeiss M2 microscope . Videos were analyzed using iMovie and pumps were counted using the slow motion feature . Rates described are the average ±SD and are from two separate experiments . C . elegans intestines were dissected as described [55] . Briefly , adults were transferred to a drop of M9 with levamisole on a microscope slide and the heads and tails were cut with a 25-gauge needle to extract the intestines . Samples were transferred to a microfuge tube with M9 + 0 . 1% Tween 20 to prevent sticking , and fixed with 4% paraformaldehyde for 1–2 h . Samples were washed three times with PBS 0 . 1% Tw20 and then incubated in block for 1–2 h . Block was PBS , 0 . 5% TX100 , 1 mM EDTA , 0 . 1% BSA , 0 . 05% sodium azide , adjusted to pH 7 . 2 with hydrochloric acid . Next , samples were incubated in the primary antibody MH33 ( Hybridoma Bank ) , 1:100 in block , overnight at 4 °C . Samples were then washed four times in PBS , then incubated in the secondary antibody Cy3-labeled goat anti-mouse IgG ( Jackson Immunoresearch ) , 1:1 , 000 in block at room temperature for 2 h . Samples were washed three times in PBS , then mounted in Vectashield with DAPI for viewing by fluorescence microscopy . Infectious N . parisii extract was made as follows . Infected worms were washed with M9 several times over the course of 1 h in a 2-ml microfuge tube . Silicon carbide beads ( BioSpec Products , Inc . ) were added to the tube , and the tube was vortexed for 1 min , four to five times . The worm extract was then filtered through Whatman filter paper number 1 to remove eggs and any remaining intact worms , glycerol was added to a final concentration of 15% , and aliquots were frozen at −80 °C . Extracts were tested for contamination by plating onto TSA bacterial media plates . To begin an infection assay , aliquots were thawed on ice , and then 50 μl of extract ( usually diluted 1:5 or 1:10 ) was added to a lawn of OP50 seeded on a 6-cm NGM plate . Synchronized L1s or L4/young adults were then added to these plates . For survival assays animals were transferred to new plates approximately 2 d after becoming adults and then every day afterward while they were producing progeny , in order to prevent progeny from obscuring the assay . Time to 50% of animals exhibiting symptoms was analyzed with Prism software using nonlinear regression analysis with the error described as standard error of the mean . For meronts , vesicles , and survival curves in Figure 5A , a Boltzmann sigmoidal provided the best fit . A Boltzmann sigmoidal curve did not provide a good curve fit for spore appearance ( Boltzmann sigmoidal fit calculated an aberrantly long time to 50% of animals with spores ) , so a sigmoidal curve was used , although this does not provide a perfect curve fit . In order to determine when there was a statistically significant difference between infected and uninfected animals as described in the legend of Figure 5 , an unpaired two-tailed t-test in Excel software was used to compare these two populations . Strains were compared using the same batch of infectious N . parisii in parallel , with three to four plates tested per strain , per experiment . Each strain was analyzed in at least three independent experiments . For each timepoint 120 to 150 worms of each strain ( 40 to 50 worms from each plate ) were mounted on agarose pads and scored by DIC microscopy for the presence of grooves , rod-shaped microbes , or vesicles of microbes . Animals were tested for survival by prodding with a platinum wire . N2 , pmk-1 ( km25 ) , daf-2 ( e1368ts ) , daf-2 ( e1368 ) ;daf-16 ( mgDf47 ) strains were raised at 20 °C and then shifted to 25 °C for assays . In experiments including daf-2 mutants , strains were transferred to 25 °C ( the restrictive temperature for daf-2 ) several hours prior to inoculation . Statistical analysis of symptoms in different strains was performed with an unpaired two-tailed t-test in Excel software . Statistical analysis of survival of different strains was performed with log-rank analysis in Prism software . For initial characterization of horizontal transmission , single adult donor animals of a marked genotype ( e . g . , Dpy ) were placed on plates with several adult recipient animals , and then recipient animals were examined for the presence of infection several days later . For transmission analysis graphed in Figure 5B , a single infected donor adult animal was co-incubated on the same plate as 200–300 L1 recipient animals for 12 h , then removed and examined by DIC microscopy for the presence of meronts or spores . 2–3 d later recipients ( n > 30 animals ) were examined by DIC microscopy for signs of infection . Animals were grown until the L3/L4 stage , and then transferred to OP50-seeded NGM plates with or without N . parisii . Animals were harvested 34 h later , because this was when the first signs of infection ( grooves ) were apparent in the majority of animals under these conditions . For P . aeruginosa and S . aureus infections , animals were treated as described ( J . I . Irazoqui , E . R . Troemel , F . M . Ausubel , unpublished data ) [26 , 27] . Total RNA was then extracted using TRI Reagent , and reverse transcribed using the Retroscript kit ( Ambion ) . This cDNA was then subjected to quantitative reverse transcriptase ( qRT ) -PCR analysis using SYBR green detection on an iCycler machine ( BioRad ) . Primers for qRT-PCR were designed using Primer3 ( MIT ) , checked for specificity against the C . elegans genome , and tested for efficiency with a dilution series of template . All values are normalized against nhr-23 , which is a control gene that does not vary under these conditions . Fold difference was calculated using the Pfaffl method . Statistical significance was assessed by a one-sample t-test . Primer sequences are available upon request . Small subunit ribosomal RNA sequences have been uploaded to Genbank for N . parisii ( FJ005051 ) , Nematocida sp . 1 ( FJ005052 ) , and Ochrobactrum sp . ( FJ005053 ) . | The small roundworm Caenorhabditis elegans is an important model system for many areas of biology , but little is known about its natural ecology . We have identified an intracellular parasite from C . elegans in its natural habitat isolated near Paris and have named it Nematocida parisii , or nematode-killer from Paris . N . parisii defines a new genus and species of microsporidia . Microsporidia are ubiquitous eukaryotic pathogens that are thought to be highly reduced fungi and are emerging pathogens of humans . The microsporidian N . parisii invades and resides in C . elegans intestinal cells where it goes through a multistep life cycle and eventually escapes out of intestinal cells , leaving holes in the terminal web , an important cellular structure . We have found N . parisii and a related Nematocida species in several wild-caught roundworms , indicating that microsporidian infections may be relatively common for C . elegans in the wild . The C . elegans/N . parisii interaction provides a valuable system in which to study microsporidian infections in a whole animal , and a convenient and inexpensive system in which to screen for anti-microsporidian drugs . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases",
"cell",
"biology",
"ecology",
"immunology"
] | 2008 | Microsporidia Are Natural Intracellular Parasites of the Nematode Caenorhabditis elegans |
The modulation of pentameric ligand-gated ion channels ( pLGICs ) by divalent cations is believed to play an important role in their regulation in a physiological context . Ions such as calcium or zinc influence the activity of pLGIC neurotransmitter receptors by binding to their extracellular domain and either potentiate or inhibit channel activation . Here we have investigated by electrophysiology and X-ray crystallography the effect of divalent ions on ELIC , a close prokaryotic pLGIC homologue of known structure . We found that divalent cations inhibit the activation of ELIC by the agonist cysteamine , reducing both its potency and , at higher concentrations , its maximum response . Crystal structures of the channel in complex with barium reveal the presence of several distinct binding sites . By mutagenesis we confirmed that the site responsible for divalent inhibition is located at the outer rim of the extracellular domain , at the interface between adjacent subunits but at some distance from the agonist binding region . Here , divalent cations interact with the protein via carboxylate side-chains , and the site is similar in structure to calcium binding sites described in other proteins . There is evidence that other pLGICs may be regulated by divalent ions binding to a similar region , even though the interacting residues are not conserved within the family . Our study provides structural and functional insight into the allosteric regulation of ELIC and is of potential relevance for the entire family .
The pentameric ligand-gated ion channels ( pLGICs ) are ionotropic neurotransmitter receptors , which are activated by the binding of ligands to specific sites of the protein . The family includes both cation-selective channels , such as nicotinic Acetylcholine- ( nAChRs ) and Serotonin receptors ( 5HT3Rs ) , and anion-selective channels , such as GABA- ( GABARs ) and Glycine receptors ( GlyRs ) [1] . Despite these differences in ion selectivity , the overall molecular architecture and the mechanism by which ligands open the ion conduction path are conserved [2]–[8] . pLGIC subunits form either homo- or hetero-pentamers that consist of at least two functional units , an extracellular ligand-binding region and a transmembrane pore [9] , [10] . Agonists open the channel by binding to a conserved site in the extracellular domain , at the interface between two subunits [11] , [12] . A homomeric receptor contains five equivalent agonist binding sites , several of which need to be occupied for maximum channel activation and this makes the process highly cooperative [5] , [13]–[16] . Agonist binding is accompanied by conformational rearrangements that are transmitted over a distance of tens of angstroms from the extracellular domain , via the domain interface to the pore [17] . These receptors have thus become important model systems for the study of allosteric mechanisms [18] . Many pLGICs are important drug targets and all aspects of their function can be influenced by pharmacological agents . These are a diverse set of molecules that include agonists and competitive antagonists ( which act on the agonist binding site itself ) , pore blockers that inhibit ion conduction , and allosteric modulators that interact with regions distinct from the agonist-binding site . Modulators such as benzodiazepines [19] , general anesthetics [20] , alcohol [21] , and the antiparasite ivermectin [22] can either enhance or inhibit pLGIC activation . pLGIC function is affected also by divalent cations ( such as calcium and zinc ) in two distinct ways . Cation-selective pLGICs are somewhat permeable to divalents , but the strong interaction between these ions and the pore decreases or blocks conduction in a voltage-dependent manner [23] , [24] . In addition to that , divalent cations can also modulate channel gating . For instance , calcium potentiates the agonist responses of nAChRs [25]–[27] and inhibits those of 5HT3Rs [28] , [29] , and zinc can either potentiate or inhibit channel activation , depending on the type of pLGIC and the ion concentration [30]–[35] . Here we show that both the modulatory and the channel block effects of divalent cations are present also in ELIC , a prokaryotic pLGIC channel whose structure was determined in a nonconducting conformation [36] . Agonists of ELIC include primary amines such as cysteamine , propylamine , and the vertebrate neurotransmitter GABA . In ELIC , these agonists occupy the canonical ligand-binding site of the family and open a cation-selective pore with permeation properties similar to those of eukaryotic channels [37] . Here we describe how divalent cations permeate and block the ELIC pore , and how they also inhibit ELIC gating , by binding in the extracellular domain , to a site remote from the ligand-binding region .
We have investigated the effects of divalent cations on ELIC by electrophysiology and X-ray crystallography . Divalent cations can influence ELIC function in several different ways depending on concentration ( Figure 1 ) . The traces in Figure 1A show that low mM concentrations of the alkaline earth metal ion Ca2+ decrease the single channel conductance of ELIC when added to the extracellular medium at negative holding potentials . ELIC single channel currents are progressively reduced by increasing Ca2+ concentrations and decrease by approximately 25% of their control amplitude at 5 mM Ca2+ ( Figure 1C ) and by a maximum of about 50% at high Ca2+ concentration [37] . This effect is due to tight interactions of divalent ions with the channel pore and has been thoroughly characterized for different pLGIC family members [23] , [24] including the homologous channel GLIC [38] , whose structure was determined by X-ray crystallography in a conducting conformation [39] , [40] . Low extracellular calcium ( greater than 100 µM ) produces also a voltage-independent decrease in agonist potency . This effect is detectable at Ca2+ concentrations too low to decrease channel conductance and is manifested as a parallel rightward shift in the agonist dose–response curve ( Figure 1D , Table S1 ) . A similar effect on agonist binding in the presence of calcium is observed in isothermal titration calorimetry experiments ( Figure 1E ) . Up to 1 mM calcium , the shift in the agonist dose–response curves is truly parallel , as the maximum agonist current does not decrease more than the single channel conductance does ( Figure 1B and 1C ) . This pattern appears to reproduce the effects of competitive antagonists , which bind to the ligand-binding site and reduce its occupancy by the agonist in a surmountable way ( e . g . , their effect can be overcome by increasing agonist concentration ) . This resemblance is obvious if the effects of Ca2+ are compared with those of the competitive antagonist acetylcholine , which is known to bind to the agonist-binding site of ELIC ( Figure 1F ) [41] . The Schild plot for acetylcholine [42] , [43] is linear with a slope of unity and a binding affinity of 1 . 6 mM ( Figure 1G , Table 1 ) . The Schild plot for Ca2+ is also linear , with a potency of 260 µM , but a shallower slope of 0 . 8 ( Figure 1G , Table 1 ) . The similarity between the effect of calcium and that of a competitive antagonist disappears as Ca2+ concentrations are increased above 1 mM . The current traces in Figure 1B show that the reduction in agonist potency is now associated with a decrease in the maximum agonist response . This decrease is too big to be explained by the effect of Ca2+ on conductance: at 5 mM Ca2+ the single channel conductance is reduced by 25% and the maximum agonist response by 55% ( Figure 1C ) . At progressively higher concentrations of the divalent cation , the maximum current response continues to decline and this decrease can be described by a fit to a Langmuir equation with an IC50 of 6 mM ( Figure 1H ) . Despite the strong reduction in the maximum currents , the shift in EC50 remains linear over a wide concentration range ( Figure S1 ) . The pronounced drop in maximum current strongly suggests that at higher concentrations calcium impairs the opening of the channel and reduces agonist efficacy . Next , we tried to establish whether calcium impairs the maximum rate of ELIC gating ( e . g . , when the channel is fully bound to the agonist ) by measuring the on-relaxation of currents elicited by rapid propylamine applications to outside-out patches from HEK293 cells . Figure 1I shows that increasing Ca2+ from 50 to 200 µM does slow the onset of the current elicited by a saturating agonist concentration ( 20 mM propylamine , red trace ) but that this effect is overcome by increasing agonist concentration to 50 mM ( green trace ) . Only minor changes in the time course of deactivation were detected ( Table 2 ) . Thus the maximum rate with which the agonist-bound channel opens is unchanged , which is unexpected given the observed change in agonist efficacy . This could be because we could test only low calcium ( in high calcium the concentrations of agonist required to saturate channel gating are too high to be experimentally feasible ) . Alternatively , calcium impairs gating by affecting a step in the channel activation that controls the size of the maximum agonist response , but not the speed of overall gating ( see Discussion ) . Finally , we found that divalent cations other than Ca2+ also affect ELIC responses . In particular , other alkaline earth metal ions , such as Mg2+ , Sr2+ , and Ba2+ , are slightly weaker than Ca2+ in inhibiting ELIC ( Figure 2A–C and E , Table 1 , Table S1 ) , whereas the transition metal ion Zn2+ is considerably more potent ( i . e . , Schild plot x-intercept 8 µM , Figure 2D and E , Table 1 , Table S1 ) . In order to understand the structural basis of the effects of divalent ions we aimed at identifying the region of interaction by X-ray crystallography . Since the crystal form that was used for the structure determination of ELIC contains high concentrations of sulfate , which forms insoluble salts in the presence of most alkaline earth metal ions , we had to identify novel crystallization conditions compatible with divalent ions . In a broad screen we observed crystals growing in Ba2+-acetate . Ba2+ can be readily located in the electron density by its strong anomalous scattering properties , and since it has comparable effects on channel function as Ca2+ ( Figure 2A , Figure S2A ) , it is reasonable to assume that it will occupy the same sites in the protein . Crystals of the ELIC/Ba2+ complex belong to two different , yet related crystal forms , one similar to the original barium-free form of ELIC that was used for structure determination ( space group P21 ) and another growing in a higher symmetry space group ( P43 ) ( Table 3 ) . Datasets for both crystal forms were collected to 3 . 8 Å ( P21 ) and 3 . 3 Å ( P43 ) resolution and provide equivalent views of the channel and its interaction with divalent cations . The structures show a conformation of the channel that is overall very similar to the structure of ELIC already described . Strong peaks in the anomalous difference density allow us to detect the presence of Ba2+ ions bound to three distinct sites of the protein ( Figure 3 ) . Firstly , a single Ba2+ ion per channel is located on the 5-fold axis of symmetry at the extracellular end of the pore and is coordinated by the side-chains of Asn251 ( position 20′ of the second transmembrane domain in the numbering system developed for the nAChR , Figure 3A–C ) . Throughout the article we will refer to this site as Spore . There are two additional sets of binding sites for Ba2+ in the structure shown in Figure 3B . Both are found at the interface between subunits in the extracellular domain in five symmetry-related locations . One set of sites faces the channel vestibule and will be referred to as Sin . The barium ion in Sin is coordinated by Ser84 of the principal subunit and Asp86 of the complementary subunit ( Figure 3D ) . Barium ions are bound also to a set of five equivalent sites on the outer rim of the extracellular domain ( Figure 3B and E ) . These sites , which we will call Sout , are about 15 Å below the ligand-binding pocket , towards the membrane plane and are formed by the side-chains of acidic amino acids contributed by both subunits . These residues include Asp113 at the end of β6 on the principal side and Glu150 and Asp158 on the loop connecting β8 and β9 on the complementary side ( Figure 3A and E ) . The refined 2Fo-Fc electron density map of this region indicates a direct interaction of the respective carboxylate groups with the bound ions resembling Ca2+-binding sites observed in other proteins ( Figure 3E , Figure S2B and C ) . Remarkably , in none of the collected datasets did we find any evidence for Ba2+ in the ligand-binding pocket itself . The structure of ELIC in complex with Ba2+ has revealed the location of three distinct sites for the interaction with divalent cations . If binding to any of these sites is relevant for the inhibition of the channel , we would expect that mutating the interacting residues should affect the functional modulation by divalent ions . Thus we mutated the residues that contact Ba2+ in the structure and measured again the effects of Ca2+ by two-electrode voltage-clamp electrophysiology ( Figures 4 and 5 , Table 1 ) . Given that the effects of low Ca2+ concentrations resemble those of competitive antagonists , we tested also whether the agonist binding site can play a role ( even though we have no structural evidence that divalents bind there ) . Our functional data show that the agonist binding site is unlikely to be involved , because Ca2+ inhibition is not changed by a mutation here ( R91A ) that increases agonist potency by 3–4-fold ( [37] , Figure 4A and F ) . We then proceeded to investigate whether the inhibitory effects of Ca2+ are produced via binding to the Spore site by truncating the side-chain of the Asn residue in contact with the divalent ion . Our X-ray crystallography data show that the structure of this N251A mutant is on the whole similar to WT but lacks the anomalous difference density in Spore . The structure of this mutant still shows strong density of ions bound to Sout ( and weaker density for Sin ) , thus suggesting that effects of the mutation are local ( Figure 4B ) . Electrophysiological recording shows that agonists activate WT and mutant N251A channels with similar potency and that the inhibition by Ca2+ of these responses is only modestly decreased in N251A ( Figure 4C and Schild plots in 4F ) . This suggests that Spore is not the major site responsible for the Ca2+ inhibition . Figure 4F shows also that mutating the binding residues in another set of divalent ion sites , Sin ( which face the extracellular vestibule ) , has little effect on Ca2+ inhibition . Mutation S84A ( on the principal side ) changes neither the potency of the agonist nor the inhibition by Ca2+ ( Figure 4D ) . Similarly , in the mutant D86A there is only a modest decrease in agonist potency , and the inhibitory effect of Ca2+ is virtually unchanged ( Figure 4E and F ) . Thus we have shown that neither Spore nor Sin mediate the functional effects of calcium on channel activation . In contrast to that , we found that Ca2+ modulation is greatly decreased when we change any of the residues that coordinate divalent cations in Sout . This is seen both when the residues with acidic side chains ( Asp 113 , Glu150 , and Asp158 ) are individually replaced with their uncharged isosteric counterparts ( Asn or Gln ) and when the acidic side-chains are truncated to Ala ( Figure 5 , Figure S3 ) . All of these mutations cause a variable but strong decrease in the potency of Ca2+ , which suggests that they weaken the interaction with the ion and thus its inhibitory effects ( Figure 5E and F ) . The strongest effect among single mutants is observed for residues Asp113 and Asp158 ( Figure 5A , B , and E ) . Combining these two mutations in the double mutant D113A/D158A virtually abolishes the effects of both calcium and barium on the agonist dose–response curves ( Figure 5D , Figure S3E and S3F ) . Remarkably , and in contrast to our observations in WT , in this double mutant the decrease in Imax at high Ca2+ concentration appears entirely due to the reduction in single channel conductance ( Figure 5G ) . The binding of Ca2+ to Sout is thus responsible for both functional effects on the shift of the EC50 and the decrease of Imax . Figure 5 also shows that mutations in Sout shift the EC50 towards higher agonist concentrations , an effect that is not surprising given that this region is thought to be important in transducing agonist binding into channel activation ( Figure 5A–D , Figure S3 , Table S1 ) . The X-ray structure of the double mutant D113A/D158A in complex with Ba2+ is on the whole unperturbed . The double mutation has removed the density of ions bound to Sout , while leaving the strong anomalous difference density in Spore unchanged . This confirms that in this mutant divalents fail to modulate channel activation because they cannot bind to the Sout site ( Figure 5H ) . Given that the same mutations abolish also the modulation by Zn2+ ( Figure 5I ) , it is very likely that Zn2+ inhibits ELIC by binding to the same site . This finding is somewhat unexpected as Zn2+ usually interacts with histidine or cysteine residues . However , since the ligand binding domain of ELIC does not contain any cysteines and since mutations of the two histidines , which are both located on β10 , did not affect the inhibition by Zn2+ ( Figure S4 ) , it is likely that the interaction of this transition metal ion with ELIC occurs at this site and therefore deviates from common binding modes . The results of our mutational analysis strongly suggest that the observed inhibition of ELIC by divalent cations is mediated by the specific interaction with a site that is located at the outer rim of the extracellular domain , at the interface between neighboring subunits . Since this site is distant from the agonist-binding region , we wanted to explore whether there is any direct competition between the effect of divalent ions and that of competitive antagonists binding to the ligand-binding site . Such competitive antagonists include quaternary ammonium compounds such as tetramethylammonium , a weak antagonist ( Figure S5 ) , or acetylcholine , which inhibits the channel with higher affinity . The X-ray structure of ELIC in complex with the heavy atom analogue tetramethylarsonium ( Figure 6A ) and the recently determined structure of ELIC in complex with acetylcholine [41] show that both antagonists bind to the ligand-binding pocket and prevent the binding of the agonist to the same site . The overlap of agonist- and antagonist-binding sites is also reflected in the 10-fold increase in the Schild affinity of acetylcholine in the mutant R91A . This is similar to the increase in agonist potency in the same mutant ( Figure 6B and D ) . In contrast to the mutation in the binding site , the Sout double mutant D113A/D158A abolishes the modulatory effect of Ca2+ but does not alter the affinity of acetylcholine ( WT 1 . 6 mM , D113A/D158A 2 . 1 mM ) , confirming that calcium and acetylcholine act via distinct sites ( Figure 6C and F , Table 1 ) . Finally , in order to probe whether the presence of one antagonist would alter the effect of the other , we have studied the inhibition of ELIC by acetylcholine in the presence of different concentrations of Ca2+ and vice versa . In no case did we find any significant change in the potency of either antagonist , which suggests that the inhibitory effects are additive and the two compounds thus act independently ( Figure 6E–H ) .
X-ray structures of ELIC crystals grown in the presence of barium have allowed us to identify five structurally equivalent binding sites ( Sout ) located at subunit interfaces on the extracellular domain about 15 Å from the agonist-binding region . These are likely to be responsible for the observed inhibition , as mutations at this site have a strong effect on the potency of both Ca2+ and Zn2+ . The sites resemble regulatory calcium-binding pockets found in other ion channel proteins , where the divalent ions interact with the side chains of acidic residues that are often organized in clusters on the protein sequence ( Figure 7A ) [44]–[46] . The interaction found in ELIC is , however , not typical for zinc-binding sites , as these usually contain either histidines or cysteines for ion coordination [47]–[49] , residues that are unlikely to play this role in ELIC ( Figure 7A , Figure S4 ) . While the residues that interact with divalent cations in ELIC are not conserved across pLGICs , there is evidence that equivalent modulatory effects in other pLGICs involve the same ( Sout ) region . In the α7-nAChR , the residue Glu 172 , which has been identified as a key residue in the interaction with calcium [50]–[52] , resides on the same loop as Glu 150 and Asp 158 ( loop 8 ) in ELIC . Similarly , histidine and glutamate residues contributing to the interaction with Zn2+ in GABAARs were mapped to the same location , at the interface between two subunits [53] , thus indicating that the Zn2+-dependent inhibition of GABAARs may follow a similar mechanism . Residues in the same loop of 5HT3Rs have also been proposed to participate to calcium regulation of this receptor [54] . Interestingly , a study on the 5HT3R has identified an aspartate residue in the pore domain as an important determinant for calcium-dependent inhibition . The equivalent Asn residue in ELIC coordinates the barium ion in the site Spore [55] . We investigated this site by mutagenesis but did not find any indication for a similar role in the calcium regulation of ELIC . The phenotypic difference may be due to a stronger interaction with a divalent ion in the 5HT3R where the respective residue is an aspartate and thus carries a negative charge ( cf . , an uncharged asparagine in ELIC ) . The effect of calcium and other divalent cations on gating of ELIC results in a complex functional phenotype . At low extracellular calcium concentrations , we see a reduction in agonist potency that resembles competitive inhibition ( with a linear Schild plot with a slope near unity ) . Despite this resemblance , the agonist binding site is not involved in this process and the presence of the antagonist acetylcholine ( which binds in the canonical agonist site ) has no effect on the action of calcium . Finally , higher calcium concentrations reduce the maximum agonist response ( to a greater extent than can be accounted for by a conductance decrease ) . At first sight , these effects appear to be too complex to be explained by a single microscopic action of divalents ( i . e . , the binding of Ca2+ to the site Sout ) . However , they can all be accounted for , if calcium impairs a single step of ELIC activation , for example channel opening , provided gating is efficient in wild-type ELIC ( i . e . , the agonist efficacy E is high to start with , Figure 7B ) . This is a plausible hypothesis , given the high open probability of the single channel activity in Figure 1A . In first approximation , the relation between maximum open probability Pmax and efficacy E is:and our observations of an ELIC Pmax greater than 95% are compatible with values of E that are greater than 20 ( as reported for other pLGIC such as nicotinic and glycine receptors ) . If the value of E is high to start with , the reduction in efficacy produced by divalents must be substantial before a decrease in maximum response becomes apparent . That is why it is seen only at high calcium concentrations . More modest decreases in efficacy , at low calcium concentrations , will cause only a decrease in agonist potency . This is because agonist EC50 is directly affected by the value of E . In the simplest del Castillo-Katz model , EC50 is given by:where Kd is the microscopic dissociation constant of the agonist ( Figure 7B ) [56] . It can also be shown ( Text S1 ) that the effects of calcium and those of a competitive antagonist are expected to be independent , if we model equilibrium channel activation with a simple scheme , where calcium binding impairs gating ( by affecting E ) and the antagonist binds to the resting form of the channel . This model not only predicts Schild-like behavior for the effect of calcium but suggests also that the Schild intercept is a reasonable estimate for the microscopic affinity of divalents ( Text S1 ) . These conclusions are unchanged if we model channel activation by a more detailed and realistic activation scheme , incorporating an intermediate state between agonist binding and channel opening . The existence of one or more gating intermediate states for channels in the nicotinic superfamily is supported by several lines of evidence . For instance , φ analysis in muscle nicotinic AChRs [57] indicates that blocks of residues move asynchronously in the gating conformational change . In addition to that , mechanisms with reaction intermediates ( referred to as flip , primed , or catch-and-hold [13] , [58]–[61] ) are needed to explain several aspects of the function of the GlyR and the muscle nicotinic AChR , such as agonist efficacy ( Figure 7B ) . In our experiments , the presence of an additional intermediate step that limits the maximum rate of current onset in agonist-bound ELIC channels is required to explain the results of our agonist concentration jumps . This is because we observed that low calcium increased the agonist concentration needed to achieve the maximum rate of current onset , but did not change the limiting rate of channel gating . If activation went through a single conformational step as the channel gates ( as in a simple del Castillo-Katz mechanism ) , this single step would control both the rate of current onset for the agonist-bound channel and the maximum response , and any changes in this would be experimentally detectable ( see Text S1 ) . In our study we have shown how the binding of calcium to a single site remote from the ligand binding pocket modulates the activation of the pLGIC ELIC . Given that divalent ions impair ELIC gating , they are expected to bind more tightly to the resting state of the channel and stabilize it . The location of the divalent binding site at the interface between adjacent subunits is an intriguing mechanism to stabilize distinct states in an allosteric protein , given that these regions are involved in conformational changes ( Figure 7C ) . Thus , occupancy by divalent ions of sites at a similar location in the different pLGICs will result in potentiation or inhibition , depending on whether the equilibrium is shifted towards conducting or nonconducting conformations . Allosteric modulation is important for the pharmacology of pLGICs , as many of pLGIC drugs in therapeutic use act by this mechanism , although by binding to sites distinct from those of divalent ions . Modulation by divalent ions of pLGICs occurs at concentrations that are physiologically relevant in vertebrates and may regulate the activity of channels in their natural environment [31] , [62] . It is not known whether such regulation is important for ELIC activity in its natural host Erwinia chrysanthemi , but it is remarkable that the observed mechanism has been conserved during evolution .
ELIC WT and point mutants were expressed and purified as described [36] , [37] . E . coli BL21DE3 containing a vector encoding for a fusion protein consisting of the pelB signal sequence , a His10 tag , maltose binding protein , a HRV 3C protease site , and ELIC were grown in M9 minimal medium at 37°C to an OD of 1 . 0 and subsequently cooled to 20°C . Expression was induced by addition of 0 . 3 mM IPTG and carried out overnight . All the following steps were performed at 4°C . The protein was extracted from isolated membranes in a buffer containing 1% n-Undecyl-β-D-Maltoside ( UDM , Anatrace , Inc . ) and purified by Ni-NTA chromatography ( Qiagen ) . The purified MBP-ELIC-fusion protein was digested with HRV 3C protease to cleave the His10-MBP protein . His10-MBP and 3C protease were subsequently removed from solution by binding to Ni-NTA resin . ELIC was concentrated and subjected to gel-filtration on a Superdex 200 column ( GE Healthcare ) . The protein peak corresponding to the ELIC pentamer was pooled and concentrated to 10 mg/ml and used for crystallization . The purified protein was crystallized in sitting drops at 4°C . Protein containing additional 0 . 5 mg/ml E . coli polar lipids ( Avanti Polar Lipids , Inc . ) was mixed in a 1∶1 ratio with reservoir solution ( 50 mM ADA pH 6 . 5 , 50 mM BaAc2 , and 10% ( w/v ) PEG4000 ) . The crystals were cryoprotected by transfer into solutions containing 30% ethyleneglycol . All datasets were collected on frozen crystals on the X06SA beamline at the Swiss Light Source ( SLS ) of the Paul Scherrer Institut ( PSI ) on a PILATUS detector ( Dectris ) . The data were indexed , integrated , and scaled with XDS [63] and further processed with CCP4 programs [64] . The structure of WT and mutants in space groups P43 and P21 were determined by molecular replacement in PHASER [65] using the ELIC pentamer ( 2VLO ) as a search model . G164 , which was not included in the original model ( 2VLO ) , was introduced according to the structure of the ELIC acetylcholine complex ( 3RQW ) . The absence of this amino acid had only a local effect and did not influence the location of neighboring residues . The model was rebuilt in Coot [66] and refined maintaining strong NCS constraints in PHENIX [67] . R and Rfree were monitored throughout . Rfree was calculated by selecting 5% of the reflection data in thin slices that were selected for the initial dataset of ELIC and that were omitted in refinement . Binding of the agonist propylamine to ELIC in the presence and absence of calcium was measured by isothermal titration calorimetry ( ITC ) with a MicroCal ITC200 system ( GE Healthcare ) . The syringe was loaded with agonist solution containing 30–37 mM propylamine dissolved in measurement buffer ( containing 25 mM Tris-HCl pH8 . 5 , 150 mM NaCl , and in certain experiments 0 . 6 mM CaCl2 ) . The sample cell was loaded with 300 µl of purified ELIC in measurement buffer containing 0 . 9 mM UDM at a concentration between 80 and 110 µM . Agonist was applied by sequential injections of 2 µl aliquots followed by a 180 s equilibration period after each injection . The data were recorded at 4°C and analyzed by a fit to a single-site binding isotherm . Constructs containing the gene of either the WT or mutant channels preceded by the signal sequence of the chicken α7nAchR were cloned into the pTLN vector for expression in X . laevis oocytes [68] . After linearization of the plasmid DNA by MluI , capped complementary RNA was transcribed with the mMessage mMachine kit ( Ambion ) and purified with the RNeasy kit ( Qiagen ) . For expression , 1–50 ng of RNA was injected into defolliculated oocytes . Two-electrode voltage clamp measurements were performed 1 d after injection at 20°C ( OC-725B , Warner Instrument Corp . ) . Currents were recorded in bath solutions containing 10 mM HEPES ( pH 7 ) , 130 mM NaCl , and the indicated concentrations of cysteamine and divalent cations . In case of solutions containing Zn2+ , cysteamine was replaced by propylamine . The membrane potential in all dose–response measurements was set to −40 mV . As ELIC is permeable to divalent cations , we tested if endogenous calcium-activated chloride channels affected our measurements . To chelate intracellular calcium ions , the oocytes were incubated for 15 to 30 min in bath solutions lacking divalent ions but containing 10 µM BAPTA-AM . Dose–response curves in the presence of calcium obtained from BAPTA-AM-treated oocytes did not differ from the measurements of the untreated oocytes even at elevated Ca2+ concentration ( Figure S6 ) . The lack of a significant effect is likely due to the strong outward-rectification of calcium-activated chloride channels , which do not pass significant currents at negative voltages . X . laevis oocytes were transferred to a hyperosmotic solution to manually remove the vitelline layer . Membrane patches were recorded in the excised outside-out configuration 3–5 d after injection of mRNA with an Axopatch 200B amplifier ( Axon Instruments ) at 20°C . Data were sampled at 100 µs , filtered with 1 , 000 Hz , and analyzed using Clampfit ( Axon Instruments , Inc . ) . Bath solutions contained 10 mM HEPES ( pH 7 . 0 ) , 150 mM NaCl , and indicated concentrations of ligands and divalent cations . Electrodes had a resistance of 3–5 MΩ . Pipette solutions contained 150 mM NaCl , 10 mM EGTA , 5 mM MgCl2 , and 10 mM HEPES at pH 7 . 0 . Bath electrodes were placed in 1 M KCl solution connected to the bath solution by Agar bridges . The agonists were applied to the patch using a stepper motor ( SF77B Perfusion fast step , Warner ) . Human embryonic kidney 293 cells ( American Type Culture Collection-CRL-1573;LGC Promochem ) were maintained at 37°C in a 95% air/5% CO2 incubator in DMEM supplemented with 0 . 11 g/l sodium pyruvate , 10% ( v/v ) heat-inactivated fetal bovine serum , 100 U/ml penicillin G , 100 µg/ml streptomycin sulfate , and 2 mM L-glutamine ( Invitrogen ) . Cells ( passaged every 2 d , up to 30 times ) were plated and transfected by calcium phosphate-DNA coprecipitation [69] , with a total amount of DNA of 3 µg/dish ( 82% ELIC and 18% eGFP DNA , both subcloned in pcDNA3 ) . Cells were bathed in an extracellular solution containing ( mM ) : 150 KCl , 0 . 05 or 0 . 2 CaCl2 , and 10 HEPES , pH adjusted to 7 . 4 with KOH ( osmolarity 310 mOsm ) . Patch pipettes were pulled from thick-walled borosilicate glass ( GC150F; Harvard Apparatus ) and fire polished to a resistance of 8–12 MΩ . Intracellular solution contained ( mM ) : 150 KCl , 0 . 5 CaCl2 , 5 EGTA , and 10 HEPES , pH adjusted to 7 . 4 with KOH . Agonist-evoked currents were recorded at 20°C with an Axopatch 200B amplifier ( Molecular Devices ) from outside-out patches held at −100 mV . Patches were stepped to this holding voltage 0 . 2 s before the agonist was applied and otherwise held at −40 mV . No correction for junction potential was applied ( calculated value 0 . 2 mV ) . Currents were filtered at 5 kHz , digitized at 50 kHz with Digidata 1322A , and saved directly on computer with Clampex software ( all MDS Analytical Technologies ) . All concentration jumps were performed using a piezo stepper ( Burleigh instruments ) with an application tool made from theta tube glass ( Hilgenberg; final tip diameter , 150 µm ) . Voltage commands for the piezo stepper were 200 ms square pulses conditioned by low-pass eight-pole Bessel filtering ( −3 dB frequency 5 kHz ) to smooth oscillations . Actual exchange time was estimated by recording the open-tip response to the application of diluted extracellular solution ( 70% water ) after rupture of the patch . Only patches in which the 20%–80% exchange time was faster than 250 µs were included in the analysis . Agonist solutions were freshly prepared every day from 1 M stock solutions . Propylamine was applied at a concentration known to elicit maximum response ( 20 mM and 50 mM , for 50 and 200 µM Ca2+ , respectively ) . Traces shown are averages of 5 or 10 individual agonist currents , separated by at least 10 s . Responses were averaged , and the time course of activation and deactivation ( between 95% and 5% of the peak current level ) was fitted with one exponential component ( program Clampfit 9 . 0 ) . The coordinates of the P43 crystal form of ELIC in complex with Ba2+ have been deposited with the Protein Data Bank under code 2yn6 . | Pentameric ligand-gated ion channels ( pLGICs ) are ionotropic neurotransmitter receptors that mediate electrical signaling at chemical synapses . The pLGIC family includes receptors for acetylcholine , serotonin , GABA and glycine , which share a similar structural organization and activation mechanism: the channels are closed in the absence of ligands and open when neurotransmitters bind to a conserved site in the extracellular domain . In many family members , activation by the neurotransmitter can be affected by modulators ( including several drugs in therapeutic use ) , which bind to different sites on the channel . Channel function can be modulated also by divalent cations , which either potentiate or inhibit pLGICs at physiological concentrations . Here , we analyze this mechanism in the pLGIC ELIC , a prokaryotic family member of known structure . We show that divalent cations such as calcium or zinc inhibit ELIC by occupying an extracellular site remote from the ligand-binding region thereby interfering with gating . Although the site of interaction is not conserved between different family members , we present evidence that regulation of other pLGICs involves the same region . Our study has thus provided insights into a regulatory process that appears to be general for the pLGIC family in both eukaryotes and prokaryotes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry",
"biology",
"biophysics"
] | 2012 | Inhibition of the Prokaryotic Pentameric Ligand-Gated Ion Channel ELIC by Divalent Cations |
Schistosomiasis , one of the world’s greatest neglected tropical diseases , is responsible for over 280 , 000 human deaths per annum . Praziquantel , developed in the 1970s , has high efficacy , excellent tolerability , few and transient side effects , simple administration procedures and competitive cost and it is currently the only recommended drug for treatment of human schistosomiasis . The use of a single drug to treat a population of over 200 million infected people appears particularly alarming when considering the threat of drug resistance . Quantitative , objective and validated methods for the screening of compound collections are needed for the discovery of novel anti-schistosomal drugs . The present work describes the development and validation of a luminescence-based , medium-throughput assay for the detection of schistosomula viability through quantitation of ATP , a good indicator of metabolically active cells in culture . This validated method is demonstrated to be fast , highly reliable , sensitive and automation-friendly . The optimized assay was used for the screening of a small compound library on S . mansoni schistosomula , showing that the proposed method is suitable for a medium-throughput semi-automated screening . Interestingly , the pilot screening identified hits previously reported to have some anti-parasitic activity , further supporting the validity of this assay for anthelminthic drug discovery . The developed and validated schistosomula viability luminescence-based assay was shown to be successful and suitable for the identification of novel compounds potentially exploitable in future schistosomiasis therapies .
Parasitic flatworm trematodes or flukes of the genus Schistosoma cause schistosomiasis , one of the world’s greatest neglected tropical diseases . The three main species infecting humans , S . mansoni , S . haematobium and S . japonicum , can penetrate intact skin upon contact with water contaminated with parasite larvae . The World Health Organization has listed schistosomiasis as an illness for which new therapies are urgently needed [1] . There are over 200 million people living in the endemic areas of 77 countries worldwide , representing a major health and economic burden in tropical and developing nations [2 , 3] . To date , no vaccine is available against schistosomiasis , so that treatment and most of control initiatives rely on the long-term application of a single drug , praziquantel ( PZQ ) . PZQ has high efficacy , excellent tolerability , few and transient side effects , ease of distribution and competitive cost . However , the use of PZQ is limited by its stage-specific activity [4–6] , since it is active on adult parasites ( 6–7 weeks and over ) and it has minimal activity against juvenile worms ( 1–5 weeks old ) . The latter drawback can partially explain the low cure rates in high transmission areas where patients are likely to harbor juvenile and adult parasites concurrently [7] . Furthermore , the use of a single drug to treat a population of over 200 million infected people and over 700 million people at risk world-wide , appears particularly worrisome when considering the threat of drug resistance . Alarmingly , it is possible to induce resistance of S . mansoni and S . japonicum to PZQ in mice under laboratory conditions . In addition , resistance or reduced susceptibility to PZQ in field isolates of S . mansoni has been sporadically reported [8–12] . For all the above reasons , the search for new schistosomicidal agents represents today a compelling priority [13] . Modern drug discovery pipelines employ target-based screens , using in vitro assays of individual molecules and/or phenotypic screens of entire organisms . Similarly , efforts have been initiated towards the development of bioassays for high throughput screening ( HTS ) of compound libraries [14 , 15] and for automated high content phenotypic screens ( HCS ) for schistosomiasis [16–19] . In this work , we report the development and validation of a medium-throughput , luminescence-based assay for the detection of schistosomula viability . This method is automation compatible and enables the screening of compound collections on schistosomula , thus hopefully contributing to the development of novel therapeutic strategies against schistosomiasis .
Auranofin , gambogic acid ( GA ) , disulfiram , menadione , oltipraz , parthenolide , plumbagin from Plumbago indica , PZQ , thonzonium bromide , sanguinarine chloride hydrate , dimethyl sulphoxide ( DMSO ) , percoll and fetal bovine serum ( FBS ) were from Sigma-Aldrich . The Ro 15–5458 compound was a kind gift from Dr H . Stohler ( Hoffman-La Roche , Basel , Switzerland ) and oxamniquine was provided by Pfizer , London . Drugs were dissolved in DMSO to obtain stock solutions at 10 mM and were then diluted into culture medium . CellTiter-Glo ( CTG ) reagent , used in the schistosomula viability luminescence-based assay , and CellTox green dye , used in the schistosomula staining , were from Promega . BioWhittaker Dulbecco-Modified Eagle’s Medium ( DMEM ) lacking phenol red and containing 4500 mg/l glucose , 1 mM Hepes pH 6 . 98–7 . 30 , 2 mM L-glutamine , 1x antibiotic-antimycotic reagent ( Life Technologies ) and 10% heat inactivated FBS , was used as tissue culture medium for schistosomula . Adult worms were cultured in BioWhittaker DMEM containing 4500 mg/l glucose , 2 mM L-glutamine , 100 U/ml penicillin , 100 μg/ml streptomycin , 0 . 5 μg/ml amphothericin B and 10% heat inactivated FBS . Ethics statement . All animals were subjected to experimental protocols as reviewed and approved by the Public Veterinary Health Department of the Italian Ministry of Health ( Rome , Italy ) ( Authorization N . 25/2014-PR ) , according to the ethical and safety rules and guidelines for the use of animals in biomedical research provided by the relevant Italian laws and European Union’s directives . Maintenance of the S . mansoni life cycle . A Puerto Rican strain of S . mansoni was maintained by passage through the intermediate snail host Biomphalaria glabrata and ICR ( CD-1 ) outbred female mice ( Harlan Laboratories ) as definitive host . Cercariae were shed by infected snails placed under direct light for 1–2 hours . The cercarial suspension was collected , placed on ice and used for the preparation of schistosomula . Adult parasites were harvested by reverse perfusion of the hepatic portal system of infected mice previously euthanized with intra-peritoneal injections of Tiletamine/Zolazepam ( 800 mg/kg ) and Xylazine ( 100 mg/kg ) . Animal infection with S . mansoni . Female ICR ( CD-1 ) outbred 4–7 weeks old mice ( Harlan Laboratories ) were housed under controlled conditions ( 22°C; 65% relative humidity; 12/12 hours light/dark cycle; standard food and water ad libitum ) . Mice were infected transcutaneously with approximately 80 ( mixed sex ) or 200 ( single sex ) S . mansoni cercariae , for life cycle maintenance and adult parasites production , respectively . Preparation of schistosomula for compound screening . Cercariae were shed from infected snails and subsequently converted to schistosomula by mechanical transformation using an optimized version of the protocol of Brink et al . [20] , previously described by Protasio et al . [21] . Briefly , the cercaria6l suspension ( approximately 50 , 000 cercariae ) was placed in a 40 ml glass tube on ice for 0 minutes in order to reduce parasite motility . Tail detachment was obtained by shaking cercariae vigorously for approximately 30 seconds on a vortex mixer before passing them 10–12 times through a 22G syringe needle . Next , schistosomula were purified from cercarial tails by centrifugation on a 70% Percoll gradient ( starting density 1 . 13 g/ml ) . Finally , schistosomula were washed twice with DMEM complete medium lacking FBS and microscope examination was used to assess the quantity and quality of purified organisms ( less than 1% tails ) . Schistosomula were cultured in DMEM complete tissue culture medium at 37°C and 5% CO2 for 24 hours prior to drug treatment . Schistosomula were plated into flat-bottom 384-well black tissue culture treated plates ( PN: 781086 , Greiner Bio-ONE , AU ) for compound assays . Screening of compounds and bioassay setting . A compound collection of 1 , 280 molecules comprising drugs approved by FDA , EMA and other agencies ( Prestwick Chemicals , France ) was tested according to the following procedure . Compounds dissolved in DMSO , DMSO alone ( low control ) and GA ( high control ) were transferred to 384-well , black , tissue culture treated plates using the acoustic droplet ejection technology ( ATS-100 , EDC Biosystems , USA ) to reach a concentration of 10 μM in the final assay volume . A suspension of schistosomula in complete DMEM medium was transferred to assay plates with a multidrop dispenser ( Thermo Fisher , USA ) in order to have a defined number of schistosomula per well in a final volume of 30 μl . After 24 hours incubation at 37°C and 5% CO2 , a volume of 30 μl of CTG reagent ( Promega , USA ) was added resulting in cell lysis and generation of a luminescence signal proportional to the amount of ATP present in the well . Sample luminescence levels ( proportional to ATP levels ) were detected 30 minutes after CTG addition and quantified as RLU ( Relative Luminescence Unit ) by a charge-coupled device ( CCD ) -based detector ( ViewLux , PerkinElmer USA ) . Staining of schistosomula with the CellTox green dye and confocal laser scanning microscopy . Schistosomula were incubated with an equal volume of CTG reagent containing a membrane-impermeant DNA-binding dye , CellTox green ( Promega ) ( 2x final concentration ) , prepared as suggested in the manufacturer’s protocol . Schistosomula were stained for 30 minutes at room temperature and observed with a laser scanning confocal microscope , TCS SP5 ( Leica Microsystems , Mannheim ) using a 40x ( NA = 1 . 25 ) oil-immersion lens with optical pinhole at 1AU . For bright field light and fluorescence images Argon laser at 488 nm was used as excitation source . Confocal Z-stacks were collected at 0 . 5 μm intervals to a total optical depth of 22 μm . Confocal images were processed with Volocity software ( Improvision , Perkin Elmer ) for image rendering and representation of x/y view . Images for direct comparison were collected under same parameters and representative images were chosen . Schistosomula treated with DMSO and incubated with the CellTox green dye without CTG reagent were observed with an Olympus AX70 fluorescence microscope and images were recorded with the XM10 CCD-camera ( Olympus ) and analysed with the Olympus cellSens standard Image software . Images were processed by Adobe Photoshop software . Confirmation of hit compounds . Additional amounts of hit molecules were purchased from Sigma-Aldrich and quality controlled by liquid chromatography-mass spectrometry ( LC-MS ) . Each compound was serially diluted in DMSO and transferred to assay plates in order to produce a concentration range between 40 nM and 50 μM in the final assay volume . The schistosomula viability by luminescence readout was assessed as described above . Schistosomula viability by fluorescence microscopy . The assay was carried out according to Peak et al . [15] . Briefly , schistosomula were treated in microtiter plates for 24 hours with DMSO or GA and then washed three times using DMEM to remove test compounds and culture media supplements . Finally , they were stained with propidium iodide ( PI ) and fluorescein diacetate ( FDA ) at the final concentration of 2 . 0 μg/ml and 0 . 5 μg/ml , respectively . The microtiter plates , containing fluorescently labeled parasites , were subsequently analyzed by the Acumen explorer ( TTP Labtech , UK ) plate based cytometer for the simultaneous detection of PI ( 544 nm excitation/620 nm emission ) and FDA ( 485 nm excitation/520 nm emission ) . Image analysis was carried out with the Acumen explorer software . In vitro studies with S . mansoni adult worms . Male worms were recovered from mice infected , only with male cercariae , by perfusion of mesenteric veins from 8 weeks after infection and cultured in DMEM complete tissue culture medium at 37°C in a 5% CO2 atmosphere . For all treatments , parasites were placed overnight in the presence of the drug ( 10 μM ) and the following day they were washed and then cultured in 3 ml of DMEM complete medium for up to 5 days . Worm status was checked on days 1 , 2 , 3 and 5 using a stereomicroscope and viability was recorded considering phenotypic changes such as loss of mobility , tegumental damages and dark appearance . Images from each treatment were captured using a stereomicroscope Leica MZ12 and a digital camera Leica D500 controlled by Leica Firecam software ( version 1 . 7 . 1 ) . For adult worms , we converted the type and number of phenotypic responses recorded manually into a ‘severity score’ ranging from 0 ( severely compromised ) to 3 ( no effect ) . The following phenotype scoring criteria were used: 3 = worms attached , good movements , clear; 2 = some movements , dark , some tegumental damages; 1 = Sick , little movements , dark , tegumental damages; 0 = Dead . For each sample the following formula was used: ∑ ( worm scores ) number of worms The data are expressed as % severity score ( viability ) relative to DMSO . All tests were repeated at least three times . Data handling and statistical analysis . ATP signal percentage normalization ( % live parasites ) was calculated using the following equation: % live parasites = 100 ( sample average − medium average DMSO average − medium average ) The PI and FDA signals percentage normalization ( % death parasites and live parasites respectively ) were calculated using the following equations: % death parasites = 100 ( 1 − sample average − GA 50 μ M average DMSO average − GA 50 μ M average ) % live parasites = 100 ( sample average − GA 50 μ M average DMSO average − GA 50 μ M average ) Data handling and statistical analysis were carried out using the GraphPad Prism software ( GraphPad , USA ) .
In an attempt to establish a correlation between the number of schistosomula and the ATP signal , serial dilutions of parasites were cultured in 384-well plates for 24 hours . We found the ATP quantitation in these samples to be in strong correlation with the parasite numbers ( Fig . 1 ) . This correlation was linear in the range between 5 and 200 schistosomula per well . The “hook effect” that was observed in cultures with more than 200 schistosomula per well can be possibly explained by reduced efficiency in the lysis of parasites , a prerequisite for ATP detection , and/or the parasite themselves being less vital due to the limited space and nutrients within the 384-well volume . Considering that schistosomula production is a rather labor-intensive process , and according to the limits of the linear range of ATP quantitation , the amount of 100 parasites per well was regarded as the most suitable for the viability assay . In fact , even though 50 parasites per well might be considered a suitable number , the chosen density was preferred in order to obtain a robust readout for single ( no replicas ) library screening . Although the ATP luminescent signal was shown to linearly correlate with the parasite number , one could argue that the number of parasites may change during the assay incubation . However , the parasites are not replicating within the cultures and are not disintegrating upon death . In order to determine the correlation between ATP luminescent signal and viability of a whole organism , such as the schistosomulum , GA ( positive control ) and other selected schistosomicidal compounds i . e . auranofin , oltipraz , oxamniquine , plumbagin , Ro 15–5458 and PZQ , known to be effective on the larval stage of parasites and/or on adult worms were assayed [14 , 15 , 22 , 23] . GA is a natural product that is known to induce apoptosis and cell cycle arrest at the G2/M phase in mammalian cells [24] . It was previously shown that 10 μM GA is also able to kill in vitro-cultured schistosomula after 24 hours incubation [15] . Serial dilutions of GA ranging from 40 nM to 50 μM were delivered to in vitro-cultured parasites and incubated for 24 hours . In order to better characterize the impact of parasite numbers on the ATP quantitation with a known toxic compound , a variable number of schistosomula ranging from 25 up to 200 parasites/well was used ( Fig . 2 ) . In accordance with previous studies , treatment with GA led to a dose response curve having a LD50 comprised between 2 . 30–3 . 52 μM . In addition , samples treated with high GA concentrations recapitulated the no-schistosomula controls average signal , proving that the assay is indeed able to detect the parasite death . Notably , while parasite numbers , as expected , did not influence the potency of the GA the RLU values correlate with the number of parasites ( Fig . 2 ) . Importantly , to investigate the penetration of CTG reagent and its effect on schistosomula in the ATP assay , parasites treated with DMSO were incubated with the membrane-impermeant fluorogenic DNA-binding dye , CellTox green , previously added to the CTG reagent . Confocal laser fluorescent microscopy images showed robust penetration of the CTG reagent . Importantly , the bright field light images clearly demonstrated that the CTG reagent is not destroying the schistosomula and that the overall integrity of parasites is preserved ( Fig . 3 ) suggesting that the ATP quantitation reflects the overall metabolic state of the parasites . Schistosomula incubated with the CellTox green dye without the CTG reagent , as expected , did not show any staining ( S1 Fig . ) . Next , to further investigate the value of the ATP quantitation as a mean to determine schistosomula viability serial dilutions of other well known schistosomicidal compounds , ranging from 40 nM to 50 μM , were delivered to in vitro-cultured parasites ( 100 schistosomula/well ) and assayed using 24 and 72 hours readouts . As shown in Fig . 4 , auranofin , oltipraz , plumbagin and to a minor extent Ro 15–5458 impair schistosomula viability , while oxamniquine and PZQ have no effect on parasite survival . Among the active compounds , in particular with auranofin and oltipraz , a slight increase in potency was observed at 72 hours . However , at compound library screening concentration ( 10 μM ) all the active compounds would have been scored as positive at 24 hours . This incubation time is particularly suited for an HTS due to limited medium evaporation and reduced compound degradation . The results indicate that by using the ATP quantitation the activity of all compounds with the exception of PZQ and oxamniquine were detected . PZQ and oxamniquine in vitro do not induce death of larval stages at day 1 and 3 [14] . We have so far demonstrated that the ATP quantitation methodology can be applied in a viability screening of schistosomula . Thus , this simple and fast detection technology could represent a valid alternative to fluorescence-based microscopy bioassays . Recently , fluorescein diacetate ( FDA ) and propidium iodide ( PI ) have been successfully used to detect and quantify the fluorescent signal of living and dead schistosomula , respectively [15] . In order to verify the equivalence of the two approaches , a head to head comparison of both methodologies was carried out . To this aim , serial dilutions of GA were titrated against schistosomula in vitro . The fluorescence-based viability assay was carried out by staining schistosomula with both PI and FDA as previously reported [15] . As shown in Fig . 5 , the GA potency , determined by the two methods , is comparable and calculated to be 2 . 32 μM and 3 . 5 μM for the ATP and PI/FDA assays , respectively . With regard to sample reproducibility , the ATP quantitation was found to be superior , showing smaller error bars , especially at high GA concentrations . Also , the ATP test has better fitting properties , such as a narrower 95% confidence interval ( Fig . 5 , dotted lines ) and a correlation coefficient r2 of 0 . 971 versus 0 . 8572 of the PI and 0 . 9235 of FDA . Following the preliminary assessments described so far , a set of 1 , 280 drugs approved for human use were tested in this assay . The use of this library offers two major advantages: it provides an increased chance to find hit compounds since it is a collection of cell-active molecules and possibly allows the repurposing of existing drugs . All compounds were screened at the concentration of 10 μM with 100 schistomula/well and the effect of each compound was calculated by normalizing the raw data between 0% toxicity ( DMSO-treated controls ) and 100% toxicity ( 10 μM GA-treated controls ) . DMSO- and GA-treated parasites were also used to determine the “Z’-factor” , a dimensionless , simple statistical characteristic ranging from-∞ to 1 [25] . Values comprised between 1 > Z’ ≥ 0 . 5 indicate an excellent assay [25] . In our screening the Z’ value resulted greater than 0 . 5 for all the microplates tested , thus confirming the high quality of the readout . Considering the relatively small number of molecules tested and the biased collection composition , a statistic approach for the identification of the hit compounds was not envisaged . Moreover , since the drug concentration ( 10 μM ) was at the upper limit of the range commonly accepted for a screening , it was established that hit molecules should have their predicted LD50 at a concentration lower than the tested one; thus the threshold was set to 70% toxicity at 10 μM . Five molecules ( disulfiram , menadione , parthenolide , sanguinarine chloride hydrate and thonzonium bromide ) proved to be active against schistosomula after 24 hours incubation . In order to confirm the initial findings , hit compound potencies were determined in a dose response manner and their LD50 are reported in Table 1 . Of all , only one compound , disulfiram , resulted inactive . The screening was replicated to assess its robustness and false positive/negative rates . Within the second set of results , disulfiram and parthenolide were not identified as active compounds . While disulfiram was already classified as false positive in the dose-response curve , parthenolide was an actual false negative within the second run . With regard to false negatives in the first run , no additional compounds above hit thresholds were identified in the second run , thus demonstrating the reliability of the pilot screen . Since adult schistosomes are the main target of schistosomiasis treatments , the last step in the screening was an in vitro testing against mature parasites . To this end , S . mansoni adult male worms ( 8–10 weeks old ) were recovered from infected mice and treated with the selected compounds at the concentrations of 10 and 20 μM . Included in the screening were also GA and PZQ as positive controls . Following 24 hours of incubation in presence of the drugs , parasites were washed , placed in fresh medium and observed for 5 days . During this time , significant reduction in viability was detected in parasites treated with 10 μM sanguinarine chloride hydrate , menadione and thonzonium bromide , compared to DMSO treated worms ( negative control ) as shown in Fig . 6 . In particular , we found that treatment with thonzonium bromide resulted in 100% lethal phenotype whereas a strong decrease in viability ( approximately 70% ) was recorded in worms exposed to sanguinarine chloride hydrate and menadione . Already 24 hours after exposure to sanguinarine chloride hydrate and menadione , worms appeared no longer attached to the petri dish and showed tegumental damages and movement defects . Such “sick” phenotypes lasted until day 5 of culture . Finally , we found that disulfiram and parthenolide did not impair S . mansoni worm viability even when tested at 20 μM .
In the present study we describe the development and validation of a novel medium- throughput assay to detect viability of S . mansoni schistosomula . Schistosomiasis is a chronic parasitic disease with a mortality estimated at 280 , 000 deaths every year in Sub-Saharan Africa [13 , 26] . As chemotherapy relies on a single drug ( PZQ ) , many initiatives have been promoted aiming to search for novel anti-schistosomal drugs that can represent a valid alternative to the current treatment or could be used in case of emerging resistance . Unfortunately , such drug discovery process is often slow and lacking a uniform and quantifiable evaluation method [27] . Here , we have established the optimal conditions for the application of a luminescence-based assay for the medium-throughput screening of a compound library using S . mansoni schistosomula . This assay is based on the quantitation of the parasite ATP by means of luminescence detection . The use of this technology is widely accepted in the study of the cytotoxic potential of compounds on proliferating cells , since ATP is the primary energy source in cells , a fact that well correlates with their proliferation and metabolic activity . In addition , the detection of ATP is made extremely simple by commercial kits based on the use of an exogenous luciferase whose light signal is proportional to ATP concentration in the sample . ATP-based viability assays have also been used , with high-throughput formats , in unicellular parasitic protozoa such as Trypanosoma brucei [28] , Entamoeba histolytica [29] , Plasmodium berghei ANKA [30] and in Leishmania donovani for the study of a limited number of compounds [31] . However , to our knowledge , this methodology was never , applied before to medium-high throughput compounds screening in multicellular organisms , such as schistosomes . Taking advantage of schistosomula handy characteristics such as their small size and availability in large numbers , we initially focused on setting the best conditions of this assay in the larval stage of the parasite . Only a limited number of assays suitable for objective high-throughput methods are presently in use and their advantage and limitations have been recently highlighted by others [19] . Briefly , the assays are based on assessment of: i ) metabolic activity ( MTT , Alamar Blue and Acid phosphatase ) ; ii ) viability through generated heat flow by isothermal microcalorimetry [32] or fluorescence-based assays with single ( i . e . resazurin ) [14 , 33] or multiple dyes selectively taken up by damaged or healthy organisms [15]; iii ) motility by electrical impedance through a real-time cell monitoring device , xCELLigence system; iv ) high-content systems , image-based methods that can record morphological and motility changes [19] . Nonetheless these methods present some problems especially if used in a screening campaign: i ) fluorescence based metabolic assays are affected by spurious signals due to compounds auto-fluorescence; methods ii ) and iii ) are hard to automate , low throughput and require special detection devices; finally iv ) methods are protocol intensive and subject to automated image analysis biases [34] . In this work , the ATP based viability assay was compared head to head to the fluorescence-based microscopy assays where the quantitation relies on the differential uptake of PI or FDA by dead and live parasites respectively [15] . The latter technology is very labor-intensive , as several washes are required before staining; in addition fluorescence image analysis , results in low-throughout and highly variable results . Moreover , the fluorescence readout is often affected by interferences produced by test compounds , especially when screening random libraries . Finally , image analysis is not easily automated , limiting its use to relatively small compound collections . Comparing these two technologies , the ATP-based detection demonstrated its ability not only to discern between different amounts of parasites , but also to probe their metabolic status while they are still intact . Furthermore , although we have demonstrated that both techniques are accurate and result in a comparable GA LD50 , we found the ATP-based assay more reliable in terms of reproducibility and rapidity . We next applied this new luminescence-based assay to a pilot screening exercise in which five potential killing agents ( sanguinarine chloride hydrate , disulfiram , parthenolide , thonzonium bromide and menadione ) were defined as hits from a compound collection of 1 , 280 approved drugs for human use . Remarkably , four of these compounds have been previously investigated for their anti-parasitic activity . In particular , sanguinarine chloride hydrate , which is a natural benzophenanthridine alkaloid derived from the root of Sanguinaria canadensis [35] , well known for its anti-inflammatory and anti-cancer properties [36 , 37] , was also found to exert a potent anti-schistosomal activity on S . mansoni cercariae and adult worms ( 100% mortality in 48 hours at 10 μM ) [23] . A second compound , parthenolide , the main sesquiterpene lactone ( STL ) isolated from Tanacetum parthenium and Tanacetum vulgare plants , proved to be active against parasites such as Trypanosoma cruzi and Leishmania amazonensis [38] . Interestingly , the crude extract and the essential oil of the aerial parts of T . vulgare resulted also effective against S . mansoni adult worms [39] while STL showed molluscicidal properties against the snail vector B . glabrata [40] . With regard to disulfiram , it was observed that chronic administration of the drug in the diet produced a 60% reduction in the mortality of mice carrying a heavy schistosome burden . This reduction in mortality was associated with an 80% decrease in granuloma formation [41] . A similar effect was also observed in Trichuris muris ( phylum Nematoda ) for which disulfiram treatment of infected mice led to the production of malformed eggs incapable of infecting naive mice [42] . Moreover , disulfiram has also shown toxicity toward the malaria parasite [43] and efficacy against Giardia lamblia [44] , Trichomonas vaginalis and Trichomonas foetus [45] infections . Finally , menadione had highly toxic effects on trophozoites and cysts of Giardia intestinalis [46] . Taken together , these studies suggest that our findings are in accordance with the anti-parasitic activity reported with different organisms , thus supporting the efficiency of our methodology for the discovery of novel anti-schistosomal compounds . We finally tested all the hit compounds on ex vivo adult worms . Of the five compounds tested only thonzonium bromide exerted a lethal effect ( 100% mortality ) after 24 hours , whereas menadione and sanguinarine chloride hydrate caused reduced viability ( 70% mortality ) 5 days after treatment . These results are not surprising as they are in accordance with previous studies showing that the activity of some drugs , e . g . PZQ , is dependent on the age of infection , sex of the worms and on the paired or unpaired status of parasites [4–6] . In conclusion , we demonstrated that our methodology enables the objective measurement of schistosomula viability , it has high sensitivity and permits simple and fast screenings , thus representing a valid alternative to fluorescence-based microscopy assays . | Schistosomiasis , one of the world’s greatest human neglected tropical diseases , is caused by a parasitic flatworm trematode of the genus Schistosoma . Among human parasitic diseases , schistosomiasis ranks second behind malaria in terms of socio-economic and public health importance in tropical and subtropical areas . More than 200 million people are currently infected in 77 countries , 85% of whom live in sub-Saharian Africa . To date no vaccine is available against schistosomiasis . As chemotherapy relies on a single drug , praziquantel , many initiatives have been promoted aiming to search for novel anti-schistosomal drugs that can represent a valid alternative to the current treatment or could be used in case of emerging resistance . Quantitative , objective and validated methods for compound collections screening are needed for the discovery of novel anti-schistosomal drugs . Here , we report the development and validation of a medium-throughput , luminescence-based assay for assessing viability at the schistosomulum stage of the human parasite S . mansoni . Our methodology enables a simple , reproducible , highly sensitive and objective quantitation of parasite viability . It is also automation compatible and enables the screening of compound collections thus hopefully contributing to the discovery of novel therapeutic strategies against schistosomiasis . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Development and Validation of a Luminescence-based, Medium-Throughput Assay for Drug Screening in Schistosoma mansoni |
The mammalian skin epidermis is a stratified epithelium composed of multiple layers of epithelial cells that exist in appropriate sizes and proportions , and with distinct boundaries separating each other . How the epidermis develops from a single layer of committed precursor cells to form a complex multilayered structure of multiple cell types remains elusive . Here , we construct stochastic , three-dimensional , and multiscale models consisting of a lineage of multiple cell types to study the control of epidermal development . Symmetric and asymmetric cell divisions , stochastic cell fate transitions within the lineage , extracellular morphogens , cell-to-cell adhesion forces , and cell signaling are included in model . A GPU algorithm was developed and implemented to accelerate the simulations . These simulations show that a balance between cell proliferation and differentiation during lineage progression is crucial for the development and maintenance of the epidermal tissue . We also find that selective intercellular adhesion is critical to sharpening the boundary between layers and to the formation of a highly ordered structure . The long-range action of a morphogen provides additional feedback regulations , enhancing the robustness of overall layer formation . Our model is built upon previous experimental findings revealing the role of Ovol transcription factors in regulating epidermal development . Direct comparisons of experimental and simulation perturbations show remarkable consistency . Taken together , our results highlight the major determinants of a well-stratified epidermis: balanced proliferation and differentiation , and a combination of both short- ( symmetric/asymmetric division and selective cell adhesion ) and long-range ( morphogen ) regulations . These underlying principles have broad implications for other developmental or regenerative processes leading to the formation of multilayered tissue structures , as well as for pathological processes such as epidermal wound healing .
Skin epidermis is a highly organized tissue that forms an essential barrier between an organism and its surrounding environment to protect the organism from dehydration , mechanical trauma , and microbial assaults . The mammalian epidermis is divided into four distinct compartments ( from the innermost to the outermost ) : stratum basale ( basal ) , stratum spinosum ( spinous ) , stratum granulosum ( granular ) , and stratum corneum ( cornified ) [1] . The formation of the epidermis is a complex yet robust process , relying on the coordinated regulation of a number of cellular events including but not limited to stem cell self-renewal , proliferation , cadherin-mediated cell-to-cell adhesion , integrin-mediated cell-to-basement membrane adhesion , differentiation , and migration [2–6] . Formation of the different layers of epidermis ( i . e . , the stratification process ) occurs during embryonic development , ensuring the production of a functional barrier at birth . In mice , stratification occurs in several stages over a period of less than 10 days ( Fig 1 ) [7] . First , cells of the single-layered surface ectoderm commit to an epidermal fate . The embryonic basal layer then gives rise to the periderm that covers the developing epidermis until the cornified cell layer is formed [7 , 8] . The intermediate cell layer develops between the basal layer and the periderm . Development of the intermediate layer is associated with asymmetric divisions of embryonic basal keratinocytes , which occur perpendicularly to the basement membrane giving rise to one basal cell maintaining its attachment to the basement membrane and one suprabasal cell [3] . The intermediate cells are capable of transient proliferation , and the loss of this proliferative capacity is associated with the maturation of intermediate cells into spinous cells . Spinous cells subsequently undergo further differentiation into granular and cornified cells . Transcriptional regulation is central to epidermal morphogenesis . Our previous studies have revealed important roles of Ovol transcription factors in epidermal proliferation and differentiation . Loss of Ovol1 in mice delays the cell cycle exit of epidermal progenitor cells during late embryogenesis [9] , whereas loss of both Ovol1 and Ovol2 results in severe defects in epidermal development characterized by sluggish exit from a progenitor cell state , defective terminal differentiation , and failed barrier acquisition [10] . Conversely , transgenic overexpression of Ovol2 leads to precocious terminal differentiation at the expense of a progenitor cell compartment [10] . These findings provide a useful entry point to explore the strategic control of progenitor cell proliferation and differentiation during epidermal morphogenesis . Mathematical modeling provides a powerful tool to reveal regulatory mechanisms that cannot be discovered by experiments alone . There are a number of modeling paradigms that can be utilized to model layer stratification . Previously , we developed continuum models to study the mouse olfactory epithelium , including a non-spatial model to explore how multi-lineage stages and feedback regulation govern tissue development and regeneration [11] , as well as a one-dimensional ( 1D ) spatial dynamic model of multi-stage cell lineages to study tissue stratification [12] . In such continuum models , cells are treated as a continuous density distribution , which provides a reliable approximation when large numbers of cells are present and individual properties of the cells are appropriately averaged . In systems such as the epidermis , where stratification entails proliferation and differentiation of a few layers of cells and their interactions with each other as well as the basement membrane , individual cell-based models are needed . A recent review has examined five popular individual-based approaches to model the self-organization of multicellular tissues [13] . For the epidermis , an agent-based modeling framework and a lattice-free cell-center method have been used to produce comprehensive 2D or 3D simulations for the spatiotemporal dynamics of epidermal homoeostasis [14–16] . An anisotropic subcellular element method was used to investigate the roles of complex cell morphology and biophysical anisotropic cell-cell interaction during basal-suprabasal layer formation [17] . These investigations focused on studying the roles of the basal layer stem cells in epidermal formation and homeostasis . Many questions surrounding the development from a single layer of stem cells to a stratified epithelia remain . For example , how do the different types of cells form layers with well-defined boundaries between them ? When cells proliferate and differentiate , how do the different types of cells know where to go and when to start or stop to produce the correct size of different layers ? How do Ovol , key transcription factors that regulate epidermal proliferation and differentiation , control the spatial organization of epidermal layers ? Here we present a 3D multiscale model of epidermis comprised of multiple cell types ( basal , spinous , and granular ) and incorporating regulation of cell proliferation and differentiation via a gene regulatory network . Major components in the 3D multiscale model include the discrete and stochastic Subcellular Element Model ( SEM ) , which are used to describe individual cells , cell divisions , and physical cell-cell interactions . These models also incorporate dynamic systems such as gene regulatory network within each cells and a continuum model for extracellular morphogens . Specifically , symmetric and asymmetric cell divisions , selective cell-cell adhesion , diffusive signaling molecules , feedback regulations of extracellular morphogens , and cell lineage transitions are incorporated to allow natural emergence of multiscale interactions for layer formation . GPU algorithm is implemented to enable efficient parameter exploration . Our study begins with analyzing known epidermal phenotypes of the Ovol mutant mice using both non-spatial and spatial models to interrogate the regulation of the epidermal cell lineage . We find critical regulations exerted by Ovol , namely inhibition of stem cell proliferation and stimulation of spinous cell differentiation , to be the most effective at recapitulating the experimentally observed epidermal phenotypes . Using this as a foundation , we expand our 3D multiscale model to examine the involvement of several spatial elements in epidermal morphogenesis . Our results show that short-range spatial mechanisms of symmetric/asymmetric division and selective cell adhesion in tandem with long-range regulation of the extracellular morphogen leads to robust epidermal stratification .
We first consider a non-spatial cell lineage model [11 , 18 , 19] consisting of three different cell types with four stages: basal stem cells , proliferative intermediate spinous cells , mature non-proliferative spinous cells , and granular cells . Previous work has suggested that Ovol1 and Ovol2 repress the expression of one another [9 , 10 , 20] . Therefore we include an Ovo1-Ovol2 cross-repression loop in our model . In order to make a direct comparison with the Ovol perturbation data , we first explored various possible cellular effects of this loop , such as regulating the probability of stem cell self-renewal , proliferation rates , and maturation times of committed cells to obtain parameters that are consistent with the experimentally observed phenotypes of Ovol-deficient and -overexpressing mice ( summarized in Table 1 ) . To investigate interactions among three major biological scales: transcriptional regulations , individual cells , and the epidermis consisting of many cells and diffusive regulatory molecules secreted from the cells , we next develop a new 3D multiscale model that accounts for intracellular regulatory networks with noises , stochastic cell division , extracellular morphogens , individual cells , and cell population . In previously published experiments , genetic deletion of Ovol1/Ovol2 or overexpression of Ovol2 resulted in an epidermal tissue with altered size and abnormal proportions of different epidermal cell types [9 , 10 , 23] ( Table 1 ) . To investigate how embryonic epidermis achieves appropriate stratification , a four-stage cell lineage model ( Eq 1 ) similar to [18 , 19] was used to explore the transcriptional mechanisms governing growth and differentiation incorporating the Ovol transcription factors , and the model is constrained by the experimental data . Using the cell lineage model , we first evaluated the sensitivity of the different epidermal layers with respect to the cellular parameters in Eq 1 ( details in S1 Text ) . Based on the skin phenotypes of gain- and loss-of function Ovol mutants in Table 1 , we hypothesized that Ovol1 and Ovol2 may down-regulate p0 ( i . e . , promote basal cell to spinous cell transition ) , p1 ( i . e . , promote growth arrest of spinous cells ) ; v0 ( i . e . , inhibit the proliferation rate of basal cells ) , v1 ( i . e . , inhibit the proliferation rate of spinous cells ) , and/or up-regulate d2 ( i . e . , promote terminal differentiation of progenitor cells into granular cells ) ( red dashed lines in Fig 2A , and see S1 Text for more details ) . By exploring the various possible effects of the Ovol regulation loop on the dynamics of cell lineages ( red dashed lines in Fig 2A ) , we found that the simplest form of lineage regulations which can recapitulate the in vivo epidermal phenotypes of all four Ovol mutants , is Model 1 ( dashed lines 1+4 ) where Ovol proteins inhibit p0 and stimulate d2 , or Model 2 ( dashed lines 1+5 ) where Ovol proteins inhibit v0 and stimulate d2 ( see Table 2 and S1 Text ) . More complicated models with additional regulators ( dashed lines 2 or 3 in Fig 2A ) to Model 1 or Model 2 are also able to recapitulate the epidermal phenotypes ( see S1 Text ) . Overall , the modeling studies suggests that Ovol1 and Ovol2 exert pleiotropic effects with two key components: decreased epidermal stem/progenitor cell self-renewal/proliferation and increased terminal differentiation . Consistent with Models 1 and 2 , our previous experimental work has suggested a role for Ovol1 and Ovol2 in suppressing an epidermal progenitor cell fate while facilitating their terminal differentiation [10] . In particular , the number of mitotic cells in the basal layer increases from wild-type and Ovol1-/- to Ovol DKO epidermis [10] . To seek further supporting evidence for a specific effect of Ovol2 on epidermal basal cells , we stained skin from Ovol2-overexpressing mice [10] with basal cell marker K14 . A significant reduction in the size of the K14-positive compartment was indeed observed upon Ovol2 overexpression ( Fig 2B ) , supporting the notion that Ovol2 may suppress p0 or v0 . Moreover , gene set enrichment analysis ( GSEA ) of microarray data obtained from skin of Ovol2-overexpressing and control embryos [10] revealed that gene signatures associated with proliferation and cell cycle are de-enriched upon Ovol2 overexpression , supporting the possibility that Ovol2 suppresses v0 ( Fig 2C ) . Together , these experimental findings provide validation for our model , and lay the foundation for us to next explore the model to assess the contribution of additional cellular and molecular processes to robust epidermal stratification . Next we incorporated the four-stage cell lineage model into a 3D multiscale model that only includes cell proliferation and differentiation ( and thus denoted Base Model ) . Table 3 shows that this Base Model , which involves Ovol’s inhibition of v0 or p0 and stimulation of d2 , is also able to recapitulate the epidermal phenotypes , particularly the alterations in the numbers of various cell types , that result from genetic manipulations of Ovol . As expected , however , without additional spatial regulation , epidermal formation in the Base Model exhibits highly heterogeneous distribution of cells in the layers resulting in strong stochastic variation ( Fig 3 ) . Previous work has demonstrated that polar distribution of cell-cell and cell-substrate adhesions coupled with a developmental switch to asymmetric division leads to robust , predictable stratification between the basal and suprabasal cells [17] . Following this idea , the spatial regulation of asymmetric division and polarized cell adhesion is implemented into the 3D Base Model ( Fig 4A ) , and this is referred to as Asymmetric Division Model . Basal cells undergo asymmetric division to produce a basal stem cell and a spinous cell , with the latter naturally assuming a suprabasal position [3] . If there is only symmetric division , the perpendicular division plane will produce two daughter cells side by side in basal layer , which will grow and compete for the limited space , yielding large contact pressure and possibly dramatic variation of the neighboring layer formation . A comparison of time course images ( Fig 4B ) shows that asymmetric division increases the natural stratification between basal stem and spinous cells , yielding a clear boundary between the two cell types . To quantitatively assess epidermal stratification , we introduced two measurements ( see Materials and Methods ) : Sharpness Index , which resembles 1 minus the standard deviation within each slice along the z-axis , and Isolation Ratio , which quantifies the proportion of cells that are far away from their own target layer for each cell type . Time course evolution of Sharpness Index and Isolation Ratio for the Base Model and the Asymmetric Division Model is presented in Fig 5 . In the Base Model , the Sharpness Index over time is less than or equal to 0 . 5 , and the stacked bars for cell type proportion are evenly distributed along z-axis , indicating a stochastically well-mixed layer due to lack of spatial regulation in layer formation . In the Asymmetric Division Model , the Sharpness Index is overall improved to be greater than 0 . 7 for most layers , and the red stack bars representing the basal stem cell proportion are restricted to the basal layers . This clearly shows that with incorporation of asymmetric division , the stratification level significantly increases for the basal and proliferative spinous layers of the epidermis . With basal cells being restricted to the basal layer and not intermingling with spinous and granular cells of the tissue , the Sharpness Index of spinous and granular layers consequently increases . The observation of a decreased Isolation Ratio is consistent with the notion that basal stem cells stay within the basal layer , and spinous and granular cells migrate closer to their own target locations . A drawback to the Asymmetric Division Model is the introduction of stochastic perturbations during layer formation with spinous and granular cells passively transported upwards from the lower layers by proliferation pressure . This is also captured with a decreased Sharpness Index due to the intermingling of spinous and granular cells , as well as a relatively higher Isolation Ratio of the two cell types . To further analyze the spatial pattern of the same type of cells at multiple distances , we calculated another measure of spatial structure: Ripley’s K function [26] in each slice along the z direction . We compared a typical simulation from the Base Model ( Fig C in S2 Text ) and the Asymmetric Division Model ( Fig D in S2 Text ) , and found that the Ripley’s K function calculations show consistent results for the spatial distribution of the different cell types . In our model , cell interactions are represented through the potential function between subcellular elements . Subcellular elements both within and between cells will be mutually repulsive if their separation is below the equilibrium size of an element . For separations larger than this size , the elements will be mutually attractive , but with the strength of attraction falling off rapidly with separation . Evolution of the system is then prescribed by a large coupled system of Langevin equations for all elements . When the same adhesion strength is assigned to all cell types , the intracellular force will always be isotropic . In order for the reorganization among spinous and granular cells to occur , differential adhesive strength between different cell types is needed . Two types of intercellular adhesion strength are defined: Fa for the same type cells and Fb for the different type cells ( Fig 6A ) . We assume that adhesion strength is different among cell types , and the adhesion force within the same cell type is greater than that between different cell types , i . e . , Fa > Fb . Initially , we set Fa = 4Fb . To restrict intercellular adhesion within short range , the adhesion strength falls to zero once the cell distance is beyond a two-cell diameter . This model is indicated as Selective Adhesion Model . The time course snapshots show that the proliferative spinous cells , the mature spinous cells and the terminally differentiated granular cells do not intermingle as long as selective cell adhesion is maintained ( Fig 6B ) . The quantitative analysis ( Fig 6C and 6D ) showing improvements in Sharpness Index for the spinous and granular layers of the tissue together with reduced Isolation Ratio confirms that selective cell adhesion between epithelial cells , introduced by the differential activation of cadherin receptors and ligands , is necessary to regulate layer formation . The calculation of Ripley’s K function of a typical simulation with the stratified tissue ( Fig E in S2 Text ) shows that the different types of cells are separated into different layers and that the cells distribute regularly within their own layer . To examine the performance of selective cell adhesion in detail , snapshots from several typical 3D simulations are presented . Most simulations end up with a clearly stratified layer structure ( Fig 6B ) . Some simulations fail to form a proper stratified epithelium , specifically in the differentiated cell types ( Fig B in S2 Text ) . In some scenarios , the thickness of differentiated cell layers shows large variation ( Fig B in S2 Text , scenarios 1 , 2 ) . With stochastic effects , this variation decreases in certain cases with time evolution leading to even and flat layers , but in some cases , variation increases leading to unconnected pieces of spinous or granular cells ( scenarios 2 , 3 , 4 in Fig B in S2 Text ) . A corresponding Ripley’s K function calculation ( Fig F in S2 Text ) shows the similar results that there exist a cluster of spinous cells above the granular layer . Such failures reflect the limitation of selective cell adhesion in two ways . First , selective adhesion tends to make the same type of cells form clusters with spherical shape , and these spherical cell clusters break the flat layer pattern ( scenarios 1 , 2 , 4 in Fig B in S2 Text ) . Second , selective cell adhesion is a short-range mechanism , in simulation we assume that attraction between cell elements drops to zero when they are farther than two-cell diameter length , as a result the isolated cells , which are far from their target layer and endure very small even no attraction force that would lead them to join their target layer , will stay where they are for a long time until they differentiate ( scenarios 3 , 4 in Fig B in S2 Text ) . For example in scenario 3 ( Fig B in S2 Text ) , granular cell , trapped within the spinous layer , will either be removed from the system at the end of its cell cycle , or slowly migrate upwards into granular layer by proliferation pressure or intercellular interaction , and scenario 3 will get resolved within half cell cycle time ( 12 hours ) . In scenario 4 ( Fig B in S2 Text ) , when spinous cells are trapped above the granular layer , they are far from the spinous layer and experience no attraction from other spinous cells , therefore they will stay at the top for a very long time ( about two cell cycle time , ≈48 hours ) . Although most simulations end up with normal stratified epidermis as in Fig 6 , once the scenarios in Fig B in S2 Text occur , selective cell adhesion is no longer able to or takes a very long time to resolve this issue . To investigate the effect of adhesion strength , Fa and Fb are varied between 0 . 5 and 5 while other parameters are kept the same . When Fa < Fb , the effect of selective cell adhesion is reversed , and the distribution of spinous and granular cells shows a salt-pepper pattern ( Fig 7 , Pattern i ) . When Fa = Fb , cell adhesion is isotropic and the resulting layer formation resembles the results obtained with the Asymmetric Division Model ( Fig 7 , Pattern ii ) . When Fa > Fb and they are set at relatively small values , the effect of selective adhesion is not strong enough to reorganize the mixed layers ( Fig 7 , Pattern iii ) . When Fa > Fb and both are set to be relatively large values , the strong intercellular adhesion largely reduces the tissue volume , the layers are compact , and the freedom of isolated cells is highly constrained ( Fig 7 , Pattern iv ) . When Fa > Fb and Fa is set to be relatively large , the same type of cells move extensively and the clusters are prone to detach from each other , generating a large inner cavity within the aggregate ( Fig 7 , Pattern v ) . Only when Fa > Fb and the ratio of Fa/Fb is bounded within a zone ( Fig 7 , Pattern vi ) , the simulations yield well-stratified epidermis . Combined , simulations of varied adhesion strength suggests that the selective cell adhesion works most efficiently when ratio Fa/Fb is between 2 and 6 ( Fig 7 , Pattern vi ) , which allows cells not only enough energy to progressively generate and maintain stratified layers but also flexibility to get rid of isolation scenarios . To better investigate the short-range feature of selective cell adhesion , we began simulations by varying the total cell number within the epidermis . The simulation snapshots thus obtained are summarized in Fig 8A . When the total cell number is slightly decreased ( X0 . 75 ) , although consequently the cell number in each cell type decreases and the thickness of the corresponding layers shrinks , layer formation is still maintained . When the total cell number is further decreased to a greater extent ( X0 . 5 ) , the reduction in cell number results in the formation of cavities among the different layers . On the other hand , when the total cell number is increased ( X1 . 5 ) , the layer thickness for each cell type increases accordingly . Once the thickness of granular layer is greater than 4 cell diameter , the isolated spinous cells above the granular layer are not able to , or take a very long time to , return to the spinous layer . Instead of artificially altering the total cell numbers , we adjusted the proliferation rate p1 ( Fig 8B ) . When p1 is increased from 0 . 2 to 0 . 3 , the total number of spinous cells increases accordingly leading to a thickened spinous layer , which greatly decreases layer formation . To investigate the effect of various layer thickness on selective cell adhesion , we changed the total cell number ( Fig 8C ) . With decreased cell number , the stratified layer formation was still maintained . In general , due to the short-range feature of selective cell adhesion , this mechanism works under the restriction of appropriate cell number and layer thickness , generating an appropriately stratified epidermis when the cell layer thickness is around or below a certain range ( 4 cell diameter thick ) . As previous models , which contain only short-range spatial mechanisms such as symmetric/asymmetric division and selective cell adhesion , are not always capable to generate and maintain robust layer formation ( Figs 7 and 8 , Fig B in S2 Text ) , we turned to incorporate a global regulation mechanism through the use of a morphogen gradient . Extracellular signals are believed to play a major role in a cell's decision to either proliferate or differentiate . Previous experiments [27] have suggested calcium as a possible extracellular morphogen to regulate epidermal differentiation . In the developing epidermis , Ovol1 is expressed in suprabasal layers , spinous layers in particular [23] , whereas Ovol2 is expressed predominantly in the basal layer [20] . We conducted simulations exploring the relationship between the epidermal calcium gradient and epidermal development , assuming that ( 1 ) both mature spinous cells and terminally differentiated granular cells are able to secret calcium , and the calcium production rate by granular cells are two times higher than that by spinous cells; ( 2 ) calcium in turn stimulates Ovol1 expression inside spinous cells ( hence upregulate d2 ) and is permissive to Ovol2 expression inside basal stem cells ( hence inhibit v0 or p0 ) ( Fig 9A ) . This is denoted the Signal Model . Simulation results demonstrate that with calcium upregulating d2 , the stratification increases between spinous and granular layers . When spinous cells are misplaced above granular layer ( Fig B in S2 Text , scenario 4 ) , or when spinous cells form cluster and intermingle with granular layer ( Fig B in S2 Text , scenarios 2 ) , they will sense high calcium concentration secreted by themselves and neighboring granular cells; then with morphogen upregulating d2 , spinous cells speed up differentiation into granular cells , leading to decreased Isolation Ratio of spinous cell and improved layer stratification . Simulation results also demonstrate that in normal stratified tissue , morphogen regulation does not decrease stratification . When spinous cells are adjacent to the lower edge of granular layer , i . e . , when they are at the upper edge within the mature spinous layer , they also sense high calcium concentration . However , calcium concentration in well-stratified tissue has near-uniform distribution . Increased differentiation , as a result of calcium upregulating d2 , usually happens among neighboring spinous cells that are adjacent to the granular layer , and the granular cells from differentiation become part of the lower edge of granular layer . This process shifts the boundary between mature spinous cells and granular cells by one cell , and still maintains the boundary under the short-range mechanism of selective cell adhesion . Time course for the Signal Model ( Fig 9B ) , as well as increased Sharpness Index for spinous and granular layers and decreased Isolation Ratio ( Fig 9C and 9D ) , suggest the morphogen regulation mechanism improves the boundary stratification between spinous and granular cells . Analysis based on Ripley’s K function also indicates that the Signal Model yields a well-stratified tissue ( Fig G in S2 Text ) . These results indicate that inclusion of long-range spatial mechanism-morphogen regulation enhances the overall epidermis stratification . Signaling regulation also improves epidermis layer size control . In four-stage non-spatial lineage models , the sensitivity analysis and simulation prediction reveal that in order to recapitulate the experimentally observed changes in the basal layer and to prevent the loss of the basal stem cell population , p0 needs to be greater than ½ ( see S1 Text ) , yielding a trend of ever-growing cell numbers . The 3D spatial model simulation results are consistent with the non-spatial lineage model: in Fig 10A–10C , the pattern that cell number of every cell type keep increasing with time is observed in the first three models ( Base Model , Asymmetric Division Model , and Selective Adhesion Model ) . In the Signal Model ( Fig 10D ) , p0 is inhibited through Ovol regulation due to increased extracellular morphogen calcium secreted by the growing spinous and granular cell population , and this system is able to approach a steady state where the populations of different cell types are well maintained . The simulation results reveal that precise regulation of proliferation and differentiation through Ovol is required to ensure the proper numbers of differentiated cells at the appropriate time , maintaining overall tissue architecture and homeostasis . To better understand the interactive components of the multiscale model , additional simulations were performed to investigate functions of each submodel ( Section E in S2 Text ) . The results ( Figs I , J of Section E in S2 Text ) show how each submodel works and how interactions among the submodels together determine the final outcome ( i . e . the overall behavior of the multiscale method ) . First , based on the Signal Model , the parameter values of p0 , v0 , p1 , v1 , d2 or d3 are varied to study tissues with different cell numbers and cell proportions . If we decrease p0 , v0 or p1 , or increase v1 , d2 or d3 , the cell number will decrease , and the spatial mechanisms still forms stratified tissue . If the total cell number increase is within a certain range , the asymmetric division , selective adhesion and signaling mechanism are still able to stratify the whole tissue . If the total cell number increases too fast ( large v0 , small v1 , d2 or d3 ) or becomes too large ( large p0 , p1 ) , the spatial mechanisms will not be sufficient to form the proper pattern ( Fig I in S2 Text ) . This observation indicates that the spatial mechanisms are relatively robust for pattern formation and that the regulation of proliferation and differentiation by Ovol is critical to forming a stratified tissue . Next , the asymmetric division component was reduced to 0 in the Signal Model , with the other parameters kept in an appropriate range . With the regulation of selective adhesion and extracellular morphogen , the cells are still able to form a stratified tissue . However , symmetric division will yield cells distributed less regularly than for asymmetric division . As a result , even with other spatial regulation mechanisms , the tissue always presented with uneven boundaries between different cell layers ( Fig J in S2 Text ) . The outcome is similar as scenarios 1 , 2 in Fig B in S2 Text , but with a much higher frequency . This suggests that without asymmetric division , other spatial mechanisms fail to produce even and flat layers , consistent with the previous analysis of the functions of selective adhesion and extracellular morphogens . The results demonstrate that asymmetric division is critical to forming even and flat layers within the epidermis tissue . Overall , we find that the results presented here are relatively robust , and that appropriate coupling of the behavior of each submodel–which we used to explore the biological observations–is critical to the overall result .
The simulations of the non-spatial cell lineage model on Ovol have shown that transcription factor regulation of proliferation and differentiation during lineage progression is crucial for the development and maintenance of the epidermal tissue . Using 3D multiscale model to incorporate the multistage cell lineage model and the spatial regulations , we discovered that basal cell asymmetric division , selective intercellular adhesion-derived aggregation , and the influence of morphogen regulation are responsible for both epidermis layer size control and layer pattern formation . Most importantly , simulation results mirror experimental data [9 , 10 , 23] . Selective adhesion is important for epithelial tissues and many other simple tissues derived from precursor cells . This mechanism enables the cells to be selectively connected with the extracellular matrix , or/and other cells . Selective adhesion is also important in dynamic tissue developmental involving cell migration . Intercellular adhesion in the mammalian epidermis is thought to be mediated in part by cadherins ( e . g . , E-cadherin ) —adhesive components of the adherens junctions . Our model assumes that the accumulating cells do not simply remain passively stuck together; instead they adjust adhesion progressively based on cell types . Our results demonstrate that tissue architecture is generated and actively maintained by selective cell adhesions . The multiscale model contains several submodels that are connected to determine the overall behavior of the system . Specifically , 1 ) the multistage cell lineage submodel was designed to be constrained by experimental data ( Tables 1–3 ) , and we found that the population size and proportion of each cell type is controlled by this submodel through a gene regulatory network; 2 ) We systematically explored the effect of symmetric vs . asymmetric division on the overall multiscale model . The asymmetric cell division was found to be critical to robust basal-suprabasal boundary formation , consistent with our previous work [17 , 28] on the effects of symmetric vs . asymmetric divisions in different biological systems; 3 ) The submodel for the selective adhesion , as well as its behavior for various values of other parameters , was explored , showing both its importance in layer stratification and its limitation as a short-range spatial mechanism; and 4 ) The submodel on morphogen dynamics has two roles: a ) providing feedback on controlling the size and the proportion of different cell layers; and b ) providing a long-range spatial gradient for signaling from one cell type to another . The accuracy of the overall multiscale model depends substantially on the functionality and capability of each submodel and how they are connected . Different spatial or temporal scales associated with each model present significant challenges in computation and may need more detailed analytical investigations on their accuracy and integration . To fully explore the interplay among submodels , one may need to investigate wider ranges of ( combinations of ) parameter values , going beyond the current approach in which parameters were chosen submodel by submodel , Although optimizing each submodel is computationally effective , it is possible that some emerging multiscale features of the system may not be captured using this approach . It is important to note that a number of aspects of epidermal biology have not been accounted for in our models . For example , we do not account for the various shape and volume of epidermal cells , the resulting function of these cells ( e . g . providing a barrier against pathogen infections ) , and the possibly deformable nature of the basement membrane to which the basal cells adhere . Moreover , the modeling at this point is primarily limited to understanding embryonic epidermal development and does not account for the influence of damage-induced factors on adult epidermis . Future work is needed to study adult repair and regeneration , damage-related effect , cell shape-related functions , and other interesting issues to identify the similarities and differences in basic regulatory principles that govern epidermal development and homeostasis . Modeling multicellular organisms requires effective tools in describing cells in space , cell-cell interactions , cell-environment interactions , cell division , gene regulatory networks within cells , and communication signals among cells . The computational approach presented in this work provides a good starting point for modeling complex multicellular systems consisting of gene regulatory networks , multiple cell types , cellular lineage hierarchy , cellular mechanics , and their interplays with environments . While each spatial scale and major component of this model , in principle , can be replaced by a method different from this work , the current multiscale coupling allows easy GPU implementation as well as incorporating more complex modules such as dynamic cell fates or more complex gene regulatory networks . One major challenge will be to include multiple multicellular systems that have different spatial and temporal complexities . For example , within the skin , one critical component not modeled in this work is the hair follicle , which introduces another spatial scale and many more cell types . Furthermore , as crosstalk between epidermis and dermis , which consists of many cell types and other functionally important cellular constitutes , is critical to epidermis regeneration , modeling dermis and its interplay with epidermis will naturally require a more complex modeling framework . Our multiscale model of epidermis in this work has laid the foundation for future pursuit in these directions .
The subcellular element method divides an individual cell into a set of discrete elements or subcellular elements . Biomechanical forces are then defined as interactions consisting of intracellular dynamics among elements of the same cell and intercellular dynamics between elements of different cells . We assume that the equation of motion of the position vector Yai of element ai for cell i is ( similar to [21 , 28] ) dYaidt=−∇ai∑i∈IVc ( i ) , c ( j ) , t ( i ) , t ( j ) ( |Yai−Yβj| ) −∇ai∑i∈IVexternal ( Yai ) ( 4 ) where I is the set of all elements in the system , V is a pairwise force interaction between elements ai and βj , Vexternal is any external force that affects that element , c ( i ) and t ( i ) represent the cell type and the element type , respectively . The pairwise force V encompasses both intra- and inter-cellular forces . In the absence of external forces , the intra-cellular forces will scatter the inner elements to the minimum energy configuration with a roughly spherical shape of preferred size . That size is determined by the rest length r0 for Vintra , defining a volume of sorts for the cell . All elements within a cell interact according to the spring potential Vintra=μ ( rij−r0 ) 22 , ( 5 ) where rij is the distance between element i and element j of the same cell and r0 is a rest length . The inter-cellular force interactions are described by Lennard-Jones type potentials Vinter=Fa , bε ( ( σ|rij| ) 12− ( σ|rij| ) 6 ) , ( 6 ) where rij is the distance between element i and element j . The parameter ε determines the strength of interaction . Fa , b represents the intercellular adhesion strength: Fa for the same type cells and Fb for the different type cells . σ is the equilibrium separation where the inter-element potential is zero and two elements are at relative balance position . If the distance between two elements is smaller than σ , they experience a repulsion force to prevent overlap of the cell bodies . When the distance between the elements is greater than σ , but less than a cutoff value , an attraction exists between the elements . Beyond this cut-off value , we assign zero interaction between cells . These medium range interactions are designed to represent the surface interactions of cadherin-mediated cell-cell adhesion . The adherent force between cell elements and the basement membrane is defined by Vexternal ( Yαi ) =εexternal|ri| . ( 7 ) Here , εexternal is the strength of external force , and ri is the distance between element i and the basement membrane . This force has a cut-off distance of half the rest diameter of a cell , to ensure that only the elements “attached” to the basement membrane experience the attraction . To couple cell dynamics and signaling pathway , a regular , rectangular grid for chemical diffusion is superimposed on subcellular element model domain such that each cell element can find its index coordinates of the chemical field . Each simulation time step consists of a substep of subcellular element model followed by a substep of evolution of states of PDEs . During the substep of the cell-based subcellular element model , cells move to a new location , undergo growth and division , make lineage decisions , and produce signals , which modifies the local signal field . An interpolation operator is used to project concentration of signals generated from the subcellular element model domain to the PDE grid blocks: Each element of a cell secrets a constant amount of signal into the PDE grid it locates based on the element coordinates , then the locally produced signal is incorporated into chemical diffusion equation as source term to update the signal field . When cell makes division decision , it will sense the local signal level: each element will record the signal concentration from the PDE grid it is in , and compute a linear combination of the information as the local signal level for cell division decision . During the substep of the PDEs , steady state of signal field is obtained for lineage classification . When updating the chemical diffusion field , diffusivity is based on local element number density . For the diffusion coefficient of the target grid , we first get the element number density of the grid itself and its first degree neighbor grids in 3D , then the diffusivity in each grid is the reciprocal of count density times baseline diffusivity , then the diffusion coefficient of the target grid is just average over the diffusivity of itself and all first degree neighbor grid blocks . In such a way , we defined a simple rule to model the particle density influence on diffusivity . To solve Eq 3 , we apply a second-order central difference for the spatial derivatives , and a forward Euler scheme to the temporal discretization . Step size in space is chosen to be 1 μm . For each step of chemical diffusion evolution , the chemical field is updated for 1000 times , giving dt = 0 . 0036s for chemical equation updates . A flowchart ( Fig 11 ) is listed here for event coordination . The code is written in OpenCL allowing it to take full advantage of current advances in parallel processing . The most computationally intensive component of this method is computation of cell dynamics , which is an N-body simulation involving all-pairs approach to compute the all pair-wise force interactions among cell elements . It scales as O ( n2 ) where n is the total number of sub-cellular elements in the system . This step is a compute-intensive part but also highly parallel and suitable for GPU application . We followed [17 , 21 , 29] to provide a parallel implementation of SEM , and include memory layout of data structures and functional decomposition for efficient implementation . Here , the highly-parallel parts like cell dynamics , cell lineage decision and PDE evolution are handled on GPU , while the less frequent activities like cell growth and division are handled outside of the GPU in the CPU code , which helps to minimize the communication between CPU memory and GPU memory as much as possible . To quantify the level of stratification of epidermal layers , we introduce the definition of Sharpness Index as a function of layer height . The simulation results are firstly divided into several slices of about one cell size thickness along z-axis . Then each slice is processed into a 10 * 10 pixel image , where each pixel is 0 , 1 , or 2 to represent that this pixel is occupied by basal stem cells , spinous cells or granular cells , respectively . Image sharpness measure is calculated as one minus the mean square of the horizontal and vertical derivatives , evaluated as finite differences [30]: SI ( z , g , I ) =1−12# ( I ) ∑ ( x , y ) ∈I ( g ( x+1 , y ) −g ( x−1 , y ) ) 2+ ( g ( x , y+1 ) −g ( x , y−1 ) ) 2 , ( 8 ) where I is the whole image domain except for the image boundaries , g ( x , y ) represents the pixel value at grid position ( x , y ) for slice at height z . This measure of image sharpness is not used for absolute sense , but only to measure the relative sharpness of similar images . According to this definition , sharpness measure for layer boundary will always be less than 1 , and SI = 1 corresponds to an extreme polarization of the tissue . To quantify the efficiency of same type cell self-aggregation by selective adhesion and external signaling regulation , we introduce the definition of Isolation Ratio . At some time point , the coordinates of same type cells are extracted from simulation data , and cells are grouped into same clusters if they are in touch with each other . Then one cluster is marked as the target cluster if it contains the largest number cells and locates at the appropriate position . And the Isolation Ratio is the proportion of the number of cells out of the target cluster over the total cell number . According to this definition , Isolation Ratio = 0 corresponds to a perfect self-aggregation situation . Back skins were freshly frozen in optimal cutting temperature ( OCT ) compound ( Tissue Tek ) , sectioned ( 8 μm ) , fixed with 4% paraformaldehyde , and stained using a rabbit anti-K14 antibody at 1:1000 dilution ( gift from Julie Segre , National Institute of Health , Bethesda , MD ) . Total RNA was extracted from skin using TRIzol reagent ( Invitrogen ) according to the manufacturer's instructions . One μg of total RNA was reverse transcribed into cDNA using the High Capacity cDNA Reverse Transcription Kit ( Applied Biosystems ) according to the manufacturer's instructions . Hybridization of arrays ( GeneChip Mouse Exon 1 . 0 ST Array; Affymetrix ) was performed in duplicate using independent biological samples . Affymetrix GeneChip Analysis Suite software ( MAS5 . 0 ) was used to generate raw data , and genes with normalized expression levels over detection threshold were called and analyzed . K5-tTA;TRE-Ovol2 ( Ovol2 overexpression , Ovol2 BT ) , Ovol1-/-;Ovol2f/-;K14-Cre ( Ovol1/Ovol2 double knockout , Ovol DKO ) , Ovol2f/-;K14Cre ( Ovol2 knockout , Ovol2 SSKO ) , and Ovol1-/- mice have been described previously by [10] . All experiments have been approved , and conform to the regulatory guidelines of the University of California-Irvine International Animal Care and Use Committee . | Epidermal morphogenesis , which occurs during the second half of embryogenesis , is the developmental process that generates a skin permeability barrier essential for terrestrial survival . Defects with this barrier are associated with common skin disorders such as atopic dermatitis . Study of mechanisms that control epidermal development and differentiation is therefore highly relevant to human health . Motivated by recent experimental observations on the role of Ovol transcription factors in regulating epidermal development , we developed a multiscale model to investigate the underlying mechanisms responsible for epidermal layer formation and homeostasis . We report that regulation of proliferation and differentiation by Ovol plays an important role in epidermal development . In addition , our computational analysis shows that asymmetric cell division , selective cell adhesion , and morphogen regulation work in a synergetic manner to produce the well-stratified epidermal layers . Taken together , our results demonstrate that robust epidermal morphogenesis involves a balance between proliferation and differentiation , and an interplay between short- and long-range spatial control mechanisms . This principle may also be applicable to other complex systems of tissue development or regeneration . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"skin",
"medicine",
"and",
"health",
"sciences",
"integumentary",
"system",
"granular",
"cells",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"cell",
"differentiation",
"developmental",
"biology",
"stem",
"cells",
"molecular",
"development",
"epide... | 2018 | Multiscale modeling of layer formation in epidermis |
Candida albicans Ssa1 and Ssa2 are members of the HSP70 family of heat shock proteins that are expressed on the cell surface and function as receptors for antimicrobial peptides such as histatins . We investigated the role of Ssa1 and Ssa2 in mediating pathogenic host cell interactions and virulence . A C . albicans ssa1Δ/Δ mutant had attenuated virulence in murine models of disseminated and oropharyngeal candidiasis , whereas an ssa2Δ/Δ mutant did not . In vitro studies revealed that the ssa1Δ/Δ mutant caused markedly less damage to endothelial cells and oral epithelial cell lines . Also , the ssa1Δ/Δ mutant had defective binding to endothelial cell N-cadherin and epithelial cell E-cadherin , receptors that mediate host cell endocytosis of C . albicans . As a result , this mutant had impaired capacity to induce its own endocytosis by endothelial cells and oral epithelial cells . Latex beads coated with recombinant Ssa1 were avidly endocytosed by both endothelial cells and oral epithelial cells , demonstrating that Ssa1 is sufficient to induce host cell endocytosis . These results indicate that Ssa1 is a novel invasin that binds to host cell cadherins , induces host cell endocytosis , and is critical for C . albicans to cause maximal damage to host cells and induce disseminated and oropharyngeal disease .
The fungus , Candida albicans is a significant human pathogen . In hospitalized patients , this organism disseminates hematogenously and infects virtually all organs . Even with currently available therapy , bloodstream infections with C . albicans are associated with a 37% mortality [1] . C . albicans is also part of the normal oral flora and it usually grows as a harmless commensal . However , when local or systemic host defense mechanisms are impaired , this organism can proliferate and cause debilitating oropharyngeal candidiasis . To persist within the human host and cause disease , C . albicans must be able to adhere to and invade host cells or tissues while resisting the stress caused by host-derived reactive oxygen intermediates and antimicrobial peptides [2]–[5] . In other organisms , heat shock proteins play an important role in each of these activities . For example , some heat shock proteins are expressed on the cell surface of microorganisms , where they function as adhesins [6]-[9] . Also , in some bacteria and parasites , members of the Hsp70 and Hsp100 family of heat shock proteins are required for resistance to host-induced stress [10]–[12] . Ssa1 and Ssa2 are the only two members of the Hsp70 family in C . albicans , and both proteins are expressed on the cell surface of yeast and hyphae [13] , [14] . Previously , we found that histatin 5 , one of the main antimicrobial proteins found in saliva , binds with high affinity to Ssa2 and with lower affinity to Ssa1 . After histatin 5 is bound to Ssa proteins , it is transported into the cytoplasm , where it kills the fungal cell [15] , [16] . C . albicans Ssa1 and Ssa2 are also required for maximal fungicidal activity of human β defensins 2 and 3 [17] . As reported here , we investigated the roles of Ssa1 and Ssa2 in C . albicans virulence in murine models of hematogenously disseminated and oropharyngeal candidiasis . We found that Ssa1 , but not Ssa2 is essential for normal virulence in both models . Through in vitro experiments , we discovered that surface-expressed Ssa1 likely contributes to virulence by acting as an invasin and directly mediating C . albicans invasion of both endothelial and oral epithelial cells in vitro .
To evaluate the contribution of Ssa1 and Ssa2 to C . albicans pathogenicity , the virulence of ssa1Δ/Δ and ssa2Δ/Δ null mutants was tested in a murine model of disseminated candidiasis . Over the 21 day observation period , none of the mice infected intravenously with the ssa1Δ/Δ mutant died ( Figure 1A ) . In contrast , the median survival of mice infected with the wild-type and ssa1Δ/Δ::SSA1Δ/Δ complemented strain was 7 to 8 days . Ssa2 did not appear to influence virulence in this model because the survival of mice infected with the ssa2Δ/Δ null mutant was similar to that of mice infected with the wild-type strain ( p = 0 . 13 ) . Next , we compared the organ fungal burden at various time points in mice infected with the different strains . After 8 h of infection , the kidney and brain fungal burdens of mice inoculated with the ssa1Δ/Δ mutant were significantly lower than mice infected with either the wild-type or ssa1Δ/Δ::SSA1Δ/Δ complemented strains ( Figures 1B–D ) . After 1 and 2 days of infection , the kidney and brain fungal burdens of mice challenged with the ssa1Δ/Δ mutant either declined or were stable . In contrast , the fungal burden of these organs steadily increased in mice infected with either the wild-type or ssa1Δ/Δ::SSA1Δ/Δ complemented strains . The fungal burdens of the livers of mice infected with the ssa1Δ/Δ mutant were also significantly lower than those of the control mice after 1 and 2 days of infection . These results indicate that SSA1 is required for normal levels of C . albicans infection in the kidney and brain as early as 8 h after inoculation . SSA1 also appears to be necessary for the organism to persist in the tissues at later time points . To investigate why mice infected with the ssa1Δ/Δ mutant had no mortality over the course of the infection , we determined their organ fungal burden after 4 , 7 and 14 days post-inoculation . We were not able to analyze the organ fungal burden of mice infected with either the wild-type strain or the ssa1Δ/Δ::SSA1Δ/Δ complemented strain at these later time points because a significant percentage of these mice had died . We found that mice infected with the ssa1Δ/Δ mutant progressively cleared the infection so that their organs became sterile by 14 days ( Figure 1B–D and data not shown ) . This clearance provides an explanation for the prolonged survival of mice infected with the ssa1Δ/Δ mutant . The low fungal burden in the kidneys of mice infected with the ssa1Δ/Δ mutant was verified by histopathology . After 1 day of infection , the kidneys of these mice contained very small foci of organisms . These foci usually consisted of a single fungal element surrounded by a few inflammatory cells ( Figure 1E ) . Interestingly , many of the ssa1Δ/Δ cells were located in the glomeruli . The kidneys from mice infected with either the wild-type or ssa1Δ/Δ::SSA1Δ/Δ complemented strain contained the expected microabscesses , which contained numerous neutrophils surrounding multiple C . albicans filaments . It was not possible to determine whether the filaments of the various C . albicans strains were true hyphae or pseudohyphae in these histopathologic specimens . However , the filaments of the ssa1Δ/Δ mutant appeared to be similar in length to those of the wild-type and ssa1Δ/Δ::SSA1Δ/Δ complemented strain . Also , we have previously found that the ssa1Δ/Δ mutant forms hyphae similar to the wild-type strain in the presence of serum in vitro [16] . Therefore , the attenuated virulence of the ssa1Δ/Δ mutant was unlikely due to a defect in filamentation . During the construction of the ssa1Δ/Δ mutant , URA3 was used as the selectable marker . The chromosomal locus at which URA3 is integrated can sometimes influence the virulence of C . albicans mutants in the mouse model of disseminated candidiasis [18]–[20] . In the ssa1Δ/Δ mutant , URA3 was integrated at the SSA2 locus [16] . However , in the ssa1Δ/Δ::SSA1 complemented strain , URA3 was integrated at the RPS10 locus , which is known to result in normal activity of the URA3 gene product , orotidine 5′-monophosphate decarboxylase [21] . To verify that the observed attenuated virulence of the ssa1Δ/Δ mutant was due to the absence of SSA1 and not the result of the chromosomal locus of URA3 , we tested the virulence of a second ssa1Δ/Δ strain in which URA3 was integrated at the RPS10 locus . As expected , all mice infected intravenously with this mutant survived ( Supplemental Figure S1A ) , thus confirming that SSA1 is required for the normal virulence of C . albicans . Next , we investigated the contributions of SSA1 and SSA2 to virulence in a mouse model of oropharyngeal candidiasis . There was a trend towards reduced oral fungal burden in mice infected with the ssa1Δ/Δ mutant after 1 day of infection , but this difference was only significant when compared to the ssa1Δ/Δ complemented strain ( p = 0 . 04 ) , but not the wild-type strain ( p = 0 . 13 ) ( Figure 2A ) . However , after 2 and 5 days of infection , mice infected with the ssa1Δ/Δ mutant had markedly lower oral fungal burdens compared to mice infected with either the wild-type or ssa1Δ/Δ::SSA1 complemented strain . In contrast , the oral fungal burden of mice infected with the ssa2Δ/Δ mutant was similar to that of mice infected with the wild-type strain ( p = 0 . 57 ) ( Figure 2B ) . Therefore , Ssa1 is required for maximal virulence during oropharyngeal candidiasis , but Ssa2 is dispensable for virulence during this infection . The results of histopathologic examination of the tongues of the mice infected with the various strains verified the reduced virulence of the ssa1Δ/Δ mutant . The tongues of mice infected with the ssa1Δ/Δ mutant had relatively small lesions that contained fewer fungal cells and neutrophils than the lesions of mice infected with either the wild-type or the ssa1Δ/Δ::SSA1 complemented strain ( Figure 2C ) . Importantly , the filaments of the ssa1Δ/Δ mutant were similar in length to those of the wild-type strain . Collectively , these results indicate that Ssa1 is necessary for maximal C . albicans virulence during both hematogenously disseminated and oropharyngeal candidiasis . During hematogenously disseminated candidiasis , blood-borne C . albicans cells must adhere to and penetrate the endothelial cell lining of the blood vessels to invade the deep tissues [3] . C . albicans also adheres to and invades oral epithelial cells during oropharyngeal candidiasis [22]–[24] . After the organism invades endothelial or oral epithelial cells in vitro , it damages these cells . Moreover , C . albicans mutants with impaired capacity to cause host cell damage in vitro frequently have attenuated virulence in mice [25]–[27] . Therefore , we used a 51Cr release assay to determine the capacity of the ssa1Δ/Δ mutant to damage monolayers of human umbilical vein endothelial cells and the FaDu oral epithelial cell line . In this assay , the host cells are incubated with 51Cr , which is taken up by the cells and binds to cytoplasmic proteins . When the host cells are damaged , there is loss of membrane integrity and the labeled proteins leak out of the cells into the medium . The amount of 51Cr that is released is proportional to the extent of cellular damage [26]–[31] . We found that the ssa1Δ/Δ mutant caused 50% less damage to endothelial cells and 89% less damage to epithelial cells compared to the wild-type strain ( Figure 3 ) . Complementing the ssa1Δ/Δ mutant with a wild-type copy of SSA1 restored its capacity to damage these cells . As predicted by the virulence data , the ssa2Δ/Δ mutant caused the same extent of damage to the endothelial and epithelial cells as did the wild-type strain . All strains of C . albicans formed true hyphae that were of similar length on both endothelial and epithelial cells ( data not shown ) . Therefore , the host cell damage defects of the ssa1Δ/Δ mutant were not due to impaired hyphal formation [27] , [32] , [33] . These results indicate that Ssa1 is necessary for C . albicans to cause maximal damage to endothelial and oral epithelial cells in vitro . This defective capacity to damage host cells likely contributed to the attenuated virulence of the ssa1Δ/Δ mutant . Although endothelial cells grow as a monolayer in vivo , epithelial cells grow in multiple , stratified layers in the oropharynx . Therefore , we tested the interactions of the ssa1Δ/Δ and ssa2Δ/Δ mutants with oral epithelial cells in a three-dimensional culture model , which more closely mimics the normal human oral mucosa and submucosa [34] . Consistent with the intraoral animal findings , the ssa1Δ/Δ mutant grew only in small foci and caused little visible epithelial cell damage , even after 2 days of infection ( Figure 4A ) . There was only slight cellular edema in the epithelium , which otherwise maintained its barrier function and prevented the organisms from crossing the basal layer of cells into the collagen gel . In contrast , the ssa1Δ/Δ::SSA1 complemented strain formed a thick biofilm on the epithelial cells , which resulted in degradation of most of the epithelial layers and subsequent submucosal invasion . As predicted by our previous results , the ssa2Δ/Δ strain induced extensive epithelial destruction and invaded into the submucosal compartment similarly to the ssa1Δ/Δ::SSA1 complemented strain . To quantify the extent of epithelial cell damage caused by the various strains of C . albicans in the 3-dimensional model , we measured the leakage of lactate dehydrogenase ( LDH ) into the medium . The ssa1Δ/Δ mutant induced less LDH release than did the ssa1Δ/Δ::SSA1 complemented or ssa2Δ/Δ strains ( Figure 4B ) . These results provide further support for the importance of Ssa1 in the capacity of C . albicans to damage oral epithelial cells . To cause maximal damage endothelial or oral epithelial cells , C . albicans must adhere to and then invade these host cells [27] , [30] , [31] . One mechanism by which C . albicans invades endothelial cells and oral epithelial cells is by inducing its own endocytosis [27] , [31] , [33] , [35] . Therefore , we used our standard differential fluorescent assay to determine if the ssa1Δ/Δ mutant had a defect in its capacity to adhere to and/or be endocytosed by endothelial and FaDu epithelial cells in vitro . Compared to the wild-type strain , the ssa1Δ/Δ mutant was endocytosed poorly by both endothelial cells and oral epithelial cells ( Figure 5A ) . The ssa1Δ/Δ::SSA1 complemented strain and the ssa2Δ/Δ mutant interacted with both the endothelial and epithelial cells similarly to the wild-type strain . In addition , endocytosis defects of the ssa1Δ/Δ mutant persisted when URA3 was integrated at the RPS10 locus ( Supplemental Figure S1B ) . Therefore , Ssa1 , but not Ssa2 , is required for C . albicans to invade endothelial cells and oral epithelial cells in vitro . C . albicans induces its own endocytosis by endothelial cells in vitro by binding to N-cadherin on the endothelial cell surface [36] . The corresponding receptor for C . albicans on oral epithelial cells is E-cadherin [37] . To determine the mechanism of host cell invasion defects of the ssa1Δ/Δ mutant , we examined its capacity to bind to N-cadherin and E-cadherin in cell membrane extracts from endothelial cells and epithelial cells , respectively . Hyphae of the ssa1Δ/Δ mutant bound poorly to N-cadherin and E-cadherin relative to both the wild-type and ssa1Δ/Δ::SSA1 complemented strain , ( Figure 5B ) . This reduced binding to cadherins likely contributed to the impaired host cell invasion and subsequent reduced host cell damage of the ssa1Δ/Δ mutant . Ssa1 is expressed either on or near the cell surface of both hyphae and yeast phase C . albicans [13] . Therefore , we investigated the possibility that Ssa1 by itself could mediate pathogenic host cell interactions . Latex beads were coated with recombinant Ssa1 ( rSsa1 ) , rSsa2 , or bovine serum albumin ( BSA ) , after which their endocytosis by host cells was measured . Both endothelial cells and epithelial cells endocytosed significantly more beads coated with either rSsa1 or rSsa2 compared to the control BSA-coated beads ( Figure 6 ) . These results suggest that both Ssa1 and Ssa2 can directly induce endocytosis by endothelial and epithelial cells . To verify that Ssa1 was expressed on the surface of C . albicans hyphae that were interacting with host cells , we infected FaDu epithelial cells with a strain of C . albicans that expressed an Ssa1-GFP fusion protein . After a 90 min incubation , the cells were fixed , stained with an Alexa 594-conjugated anti-C . albicans antibody to label the cell surface , and then imaged by confocal microscopy . We found that Ssa1-GFP fluorescence was strongest at the periphery of the hyphae ( Figure 7 ) . Importantly , this fluorescence co-localized with that of the fluorescently-labeled anti-C . albicans antibody , indicating that Ssa1 was located in the outermost part of the cell wall . Interestingly , when a similar experiment was performed using organisms grown as yeast cells in suspension , Ssa1 was located slightly internal to the cell surface ( Figure S2 ) . These results suggest that the location of Ssa1 is dependent on the morphology and culture conditions of the organism . Importantly , when C . albicans hyphae are in contact with host cells , Ssa1 is likely expressed on the cell surface where it can interact with host cell receptors . One C . albicans ligand that binds to N-cadherin and E-cadherin is Als3 , and als3Δ/Δ mutants have significantly reduced adherence to and invasion of both endothelial cells and oral epithelial cells in vitro [37]–[39] . Because heat shock proteins can potentially influence the expression of other proteins on the surface of C . albicans , we investigated the possibility that the host cell interaction defects of the ssa1Δ/Δ mutant were due to decreased surface expression of Als3 . Flow cytometric analysis of hyphae stained with a polyclonal anti-Als3 antibody demonstrated that the ssa1Δ/Δ mutant expressed a similar amount of Als3 on its surface as did the wild-type and ssa1Δ/Δ::SSA1 complemented strains ( Figure 8A ) . Furthermore , using indirect immunofluorescence with the anti-Als3 antibody , we determined that Als3 was distributed along the entire length of ssa1Δ/Δ mutant hyphae , similar to the wild-type and ssa1Δ/Δ::SSA1 complemented strains ( Figure 8B ) . Thus , the host cell interaction defects of the ssa1Δ/Δ mutant are unlikely to be due to abnormal amount or location of Als3 on the cell surface . To further investigate possible interactions between Ssa1 and Als3 , we constructed an ssa1Δ/Δ als3Δ/Δ double mutant and compared its host cell interactions with those of ssa1Δ/Δ and als3Δ/Δ single mutants . When a 90 min incubation period was used , the host cell interaction defects of the both the ssa1Δ/Δ and als3Δ/Δ mutants were so large that it was not possible to detect any further reduction in endocytosis of the ssa1Δ/Δ als3Δ/Δ double mutant ( Figure 5 and data not shown ) . Therefore , we increased the incubation period to 150 min . We found that at this time point , the als3Δ/Δ single mutant had a greater defect in inducing endocytosis by endothelial and epithelial cells than did the ssa1Δ/Δ single mutant ( Figure 8C ) . However , the endocytosis defect of the ssa1Δ/Δ als3Δ/Δ double mutant was similar to that of the als3Δ/Δ single mutant . Collectively , these results indicate that Ssa1 and Als3 function in the same pathway ( s ) to induce endocytosis , perhaps by binding to the same host cell surface proteins and/or forming part of the same multiprotein complex . It seemed paradoxical that Ssa2 was sufficient to induce endocytosis by host cells , yet the ssa2Δ/Δ mutant was endocytosed similarly to the wild-type strain . One potential explanation for these results is that SSA1 is expressed at a higher level than SSA2 . Previously , we had found that , in yeast-phase C . albicans , the SSA1 mRNA transcript levels were significantly greater than those of SSA2 , and the fungal cell wall contained 4- to 5- fold more Ssa1 than Ssa2 [16] . To determine if SSA1 was expressed greater than SSA2 in hyphae that were in contact with host cells , we infected FaDu oral epithelial cells with the various C . albicans strains and allowed them to germinate . SSA1 and SSA2 mRNA expression in these organisms was measured by real-time PCR . We found that in the wild-type strain , SSA1 was expressed 8-fold higher than SSA2 ( Figure 9 ) . As expected , SSA1 mRNA was not detectable in the ssa1Δ/Δ mutant . Importantly , SSA2 expression in this mutant was similar to that of the wild-type strain , indicating that SSA2 was not up-regulated in compensation for the absence of SSA1 . Therefore , even though Ssa2 is capable of stimulating host cell endocytosis , the effects of SSA2 deletion are likely masked by the high level expression of SSA1 . C . albicans secretes phospholipases and aspartyl proteases , both of which contribute to the virulence of this organism , probably by damaging host cells [35] , [40]–[46] . Heat shock proteins can influence protein secretion [47] , [48] . Therefore , we investigated the possibility that the attenuated damage of endothelial and epithelial cells caused by the ssa1Δ/Δ mutant was due to reduced secretion of phospholipases or proteases . We grew the various strain on either egg yolk agar or BSA agar and measured the size of the zones of precipitation or clearance around the colonies to screen for total extracellular phospholipase and protease activity , respectively [43] , [49] . The ssa1Δ/Δ mutant was similar to the wild-type and ssa1Δ/Δ::SSA1 complemented strains in these assays ( Figure 10 ) . These data suggest that the reduced virulence of the ssa1Δ/Δ mutant was unlikely the result of diminished secretion of phospholipases or proteases . Heat shock proteins are important for some microorganisms to resist the stressful conditions they encounter in the host [10]–[12] . Therefore , increased susceptibility to host-induced stress could have contributed to the reduced virulence of the ssa1Δ/Δ mutant . To evaluate this possibility , we tested the susceptibility of the ssa1Δ/Δ and ssa2Δ/Δ mutants to various stressors . Both the ssa1Δ/Δ and ssa2Δ/Δ mutants had wild-type susceptibility to oxidant ( menadione and H2O2 ) , cell wall ( Calcofluor white ) , osmotic ( NaCl ) , and plasma membrane ( SDS ) stress ( Figure 11A ) . To verify that deletion of SSA1 or SSA2 did not affect stress resistance , we tested these stressors at higher concentrations . The growth of the ssa1Δ/Δ and ssa2Δ/Δ mutants was inhibited similarly to the wild-type strain under these conditions ( data not shown ) . In addition , all strains had comparable susceptibility to nitric oxide ( data not shown ) . We also tested the susceptibility of the ssa1Δ/Δ mutant to damage by the human neutrophil-like HL-60 cell line . This mutant was not more susceptible than the wild-type and ssa1Δ/Δ::SSA1 complemented strains ( p≥0 . 12 ) ( Figure 11B ) . Therefore , the reduced virulence of the ssa1Δ/Δ mutant was unlikely to be due to increased susceptibility to host induced stress .
Heat shock proteins play diverse roles in the pathogenicity of many microorganisms , including bacteria , protozoa , and fungi [6]–[12] . However , the contribution of members of the HSP70 family of heat shock proteins to C . albicans virulence has not been reported previously . Our results with the ssa1Δ/Δ mutant demonstrate that Ssa1 is require for maximal C . albicans virulence during both hematogenously disseminated and oropharyngeal candidiasis in mice . The data from our in vitro studies indicated that the ssa1Δ/Δ mutant was defective in its capacity to adhere to , invade , and damage both endothelial cells and oral epithelial cells . The invasion defect of this mutant was likely due in part to its impaired capacity to bind to endothelial cell N-cadherin and epithelial cell E-cadherin , receptors that can mediate the endocytosis of C . albicans [36] , [37] , [39] . Other C . albicans mutants that are defective in invading and damaging endothelial cells or oral epithelial cells frequently have attenuated virulence in mouse models of candidiasis [25] , [26] , [32] , [50] . Therefore , it is highly probable that the host cell interaction defects of the ssa1Δ/Δ mutant contributed to its attenuated virulence in mice . The experiments with latex beads coated with Ssa1 or Ssa2 demonstrated that either protein can induce endocytosis by both endothelial cells and oral epithelial cells . However , in intact organisms , Ssa1 is likely more important than Ssa2 for inducing host cell endocytosis because the transcript levels of SSA1 are much greater . In addition , using GFP-labeled Ssa1 , we verified that Ssa1 is expressed on the surface of hyphae that are in contact with epithelial cells . Thus , Ssa1 is located where it can bind to host cell receptors . Collectively , these results indicate that Ssa1 can function as an invasin and induce host cells to endocytose the organism . Members of the Hsp70 family have also been found to mediate the host cell adherence of bacteria such as Helicobacter pylori , Haemophilus influenza , and Mycobacterium avium [6]–[8] . In addition , Hsp60 is expressed on the surface of Histoplasma capsulatum and is the major ligand for CD11b/CD18 on macrophages [9] . Orthologs of Ssa1 and Ssa2 are also present in species of Candida other than C . albicans . Whether these orthologs also mediate pathogenic host cell interactions remains to be determined . Ssa1 is functionally similar to the C . albicans invasin Als3 , which also binds to cadherins on the surface of endothelial cells and epithelial cells , and thereby induces endocytosis [37] , [39] . It is notable that ssa1Δ/Δ and als3Δ/Δ mutants both have defects in their capacity to invade and damage these cells in vitro . Furthermore , the host cell invasion defect of the ssa1Δ/Δ als3Δ/Δ double mutant was similar that of the als3Δ/Δ single mutant . The most likely explanation for this result is that Ssa1 and Als3 bind to same endothelial and epithelial cell surface proteins , either separately or perhaps as part of multiprotein complex . An alternative explanation for these results is that Ssa1 functions as a molecular chaperone that is required for appropriate folding or trafficking of Als3 [47] , [48] , [51] , [52] . Although we cannot completely exclude the possibility that the folding or post-translational modification of Als3 was aberrant in the ssa1Δ/Δ mutant , we determined that deletion of SSA1 had no effect on the amount and surface distribution of Als3 on C . albicans hyphae . These results suggest that proper Als3 trafficking can occur in the absence of Ssa1 . Furthermore , the localization of Ssa1 on the fungal cell surface and the finding that beads coated with rSsa1 were avidly endocytosed suggests the more likely scenario that Ssa1 and Als3 function cooperatively as invasins . After C . albicans invades host cells , it damages them . Host cell damage requires at least some extent of fungal invasion because blocking invasion by treatment with cytochalasin D or infection with non-invasive mutants of C . albicans reduces the extent of host cell damage [27] , [31] , [32] , [50] . Thus , the defective host cell invasion of the ssa1Δ/Δ mutant likely contributed to its reduced capacity to damage these cells . Although the exact mechanism by which C . albicans causes host cell damage is incompletely understood , it is probable that phospholipases and aspartyl proteases secreted by the organism participate in this process [35] , [40]–[46] . Some heat shock proteins can assist with protein secretion [47] , [48] . However , we found no evidence of impaired phospholipase or protease secretion in the ssa1Δ/Δ mutant . While these results indicate that bulk protein secretion is intact in the ssa1Δ/Δ mutant , we cannot completely rule out the possibility that there was reduced secretion or activity of a single phospholipase or protease isozyme in this strain . During both hematogenously disseminated and oropharyngeal infection , C . albicans is exposed to significant stress induced by the host . These stresses include exposure to reactive oxygen intermediates and antimicrobial peptides . Furthermore , in the oral cavity , C . albicans is likely exposed to toxic secondary metabolites produced by the resident oral flora . In some microbial pathogens , such as Leishmania spp . and Francisella tularensis , members of the Hsp70 and Hsp100 family of heat shock proteins are required for them to tolerate host-induced stress and cause persistent infection in mice [10]–[12] . Here , we found that Ssa1 was dispensable for the resistance of C . albicans to oxidant , cell wall , and cell membrane stress . We also determined that ssa1Δ/Δ mutant cells did not have increased susceptibility to killing by neutrophil-like HL-60 cells . Thus , the reduced virulence of the ssa1Δ/Δ mutant was unlikely due to impaired ability to withstand host-induced stress . Previously we have found that Ssa1 acts as receptor for some antimicrobial proteins such as histatin 5 , and that the ssa1Δ/Δ mutant is more resistant to some defensins [17] . Thus , one might have predicted that the ssa1Δ/Δ mutant would be resistant to neutrophil killing and therefore hypervirulent . However , the ssa1Δ/Δ mutant is not resistant to some toxic products of neutrophils , such as human neutrophil defensin 1 [17] , reactive oxygen intermediates , and nitric oxide . Therefore , it is probable that the wild-type susceptibility of the ssa1Δ/Δ mutant to these stressors was the reason why it was neither resistant to killing by the neutrophil-like HL-60 cells in vitro nor hypervirulent in mice . Although induction of host cell endocytosis is an important virulence function of Ssa1 , it is possible that this protein may also contribute to C . albicans pathogenicity in other ways . For example , in S . cerevisiae Ssa1 regulates the transcription of the multidrug resistant gene , PDR3 [53] . Furthermore , in Cryptococcus neoformans , Ssa1 functions as a transcriptional co-activator of laccase , and is required for normal melanin synthesis and virulence [54] . Thus , Ssa1 may either directly or indirectly govern the transcription of genes that also contribute to C . albicans pathogenicity . This possibility is currently being investigated .
The C . albicans strains used in these studies and their relevant genotypes are listed in Table 1 . All strains were maintained on yeast extract/peptone/dextrose ( YPD; Qbiogene ) agar plates and re-cultured at least monthly from −80°C stock . For use in the experiments , yeast-phase cells of the various strains were grown YPD broth overnight in a rotary shaker at 30°C . The initial ssa1Δ/Δ , ssa2Δ/Δ , ssa1Δ/Δ::SSA1 , and ssa2Δ/Δ::SSA2 strains were constructed previously [16] . To verify that the phenotype of the ssa1Δ/Δ mutant was not the result of URA3 being integrated at the SSA2 locus , a Ura- strain was selected by growth on 5-fluoroorotic acid [55] . This strain was transformed by the lithium acetate procedure with the CIp10 vector that had been linearized with NcoI [56] . The resulting strain , ssa1Δ/Δ-URA3 contained URA3 at the neutral RPS10 locus [21] . To delete the entire protein coding region of ALS3 in the ssa1Δ/Δ mutant , the PCR-product-directed gene deletion approach was used [57] . Briefly , deletion cassettes containing ALS3 flanking regions and the URA3 or NAT1 selection markers were amplified by PCR with primers GATATTTTGAATATGGAAATAAATCGTGCATAAGAAAGTTTTGCTATGCACGTTCATACTTCCAAAAATTGTAATACGACTCACTATAGGGC and AAACTATAGAAACAAACTAATCAAATTAACAACACACCAAATTGGAGGTAATTAATCATACCGAAAATAGCTATGACCATGATTACGCCA , using pGEM-URA3 [57] and pJK795 [58] as templates , respectively . These PCR products were then used to successively transform the Ura- ssa1Δ/Δ strain [57] . The resulting ssa1Δ/Δ als3Δ/Δ mutant was plated on 5-fluoroorotic acid to select for a Ura- strain . This strain was transformed with NcoI-linearized CIp10 [56] to re-integrated URA3 at RPS10 locus . Strains containing C-terminal GFP fusions were generated in strain CAF4-2 ( wt ) by substitution of one SSA1 allele with an SSA1-GFP tagged allele as described previously [16] . Plasmid pGFP-URA3 ( PMG1602 ) was kindly provided by Dr . C . A . Gale , and was used as a template to PCR amplify the transformation cassettes using primers containing sequences flanking the SSA1 stop codon ( 67 bp ) and homologous sequences from the vector ( forward primer: 9 bp glycine linker and 21bp vector sequence; reverse primer: 23 bp vector sequence ) . PCR products were used to transform parental strains using frozen EZ yeast transformation II kit ( Zymo Research ) following the manufacturer's protocol and selected on URA3- YNB agar plates . To identify transformants carrying cassettes correctly integrated into the target gene sequence , genomic DNA was prepared and used as the template for PCR reactions using one primer for annealing within the transformation module and a second primer annealing with the target gene locus outside the altered region . Cells were examined using confocal microscopy and Western blotting with anti-GFP to confirm expression of Ssa1-GFP fusion protein . The virulence of the various strains was tested in the mouse model of hematogenously disseminated candidiasis as described previously [26] . Briefly , 10 male BALB/c mice ( 20 g body weight; National Cancer Institute , Bethesda , MD ) were infected via the lateral tail vein with 5×105 yeast-phase cells of each strain of C . albicans in 500 µl of PBS . All inocula were confirmed by quantitative culture . The mice were monitored at least three times daily , and moribund mice were euthanized . To determine the organ fungal burden , 5 to 7 mice were inoculated with each strain as in the survival experiments . After 4 h , 8 h , 1 day , 2 days , 4 days , 7 days , and 14 days of infection , the brain , liver , and one kidney from each mouse were harvested , weighed , homogenized , and quantitatively cultured . The remaining kidneys from the 1 day time point were fixed in zinc-buffered formalin followed by 70% ethanol and then embedded in paraffin . Thin sections were cut and stained with periodic acid-Schiff . The different strains of C . albicans were also tested for virulence in our previously described mouse model of oropharyngeal candidiasis [25] , [27] , [59] . Briefly , male BALB/c mice were immunosuppressed with cortisone acetate ( 225 mg/kg; Sigma-Aldrich ) administered subcutaneously on days -1 , 1 , and 3 relative to infection . For oral inoculation , each mouse was anaesthetized by intraperitoneal injection with ketamine and xylazine ( both from Phoenix Pharmaceuticals ) , after which a calcium alginate swab ( Type 4 Calgiswab; Puritan Medical Products Company LLC ) saturated with 106 yeast/ml was placed sublingually for 75 min . The mice were subsequently provided food and water ad libitum and then sacrificed after 1 , 2 and 5 days of infection . The tongue and adjacent hypoglossal tissue were excised and cut in half . One half was weighed and homogenized for quantitative culture and the other half was processed for histopathological analysis as described above . Endothelial cells were isolated from human umbilical cord veins by the method of Jaff et al . [60] and cultured as described [32] in M-199 medium supplemented with 10% fetal bovine serum and 10% defined bovine calf serum ( Gemini Bio-Products ) , and containing 2 mM L-glutamine with penicillin and streptomycin ( Irvine Scientific ) . Endothelial cells were used at the third passage . The FaDu oral epithelial cell line ( American Type Culture Collection ) was cultured in Eagle's minimum essential medium with Earle's balanced salt solution ( Irvine Scientific ) supplemented with 10% fetal bovine serum , 1 mM pyruvic acid , 2 mM l-glutamine , and 0 . 1 mM nonessential amino acids , with penicillin and streptomycin [25] , [27] . Both cell types were maintained in a humidified incubator in 5% CO2 at 37°C . The extent of endothelial and epithelial cell damage caused by the different strains of C . albicans was measured using our previously described 51Cr release assay [25]–[27] , [31] , [32] . Briefly , endothelial cells or FaDu oral epithelial cells were grown to 95% confluency in 96 well tissue culture plates with detachable wells ( Corning ) and loaded with 5 µCi/ml Na251CrO4 ( MP Biomedicals ) overnight . After removing the unincorporated 51Cr by rinsing , the cells were then infected with yeast of the various C . albicans strains suspended in RPMI 1640 medium ( Irvine Scientific ) . The inoculum was 4×104 organisms per well of endothelial cells and 1×105 organisms per well of oral epithelial cells . The infected host cells were incubated for 3 h , after which the amount of 51Cr released into the medium and retained by the cells was determined by γ-counting . Wells containing host cells , but no organism , were processed in parallel to determine the spontaneous release of 51Cr . After correcting for well-to-well differences in the incorporation of 51Cr , the percent specific release of 51Cr was calculated using the following formula: ( experimental release - spontaneous release ) / ( total incorporation - spontaneous release ) . Experimental release was the amount of 51Cr released into the medium by cells infected with C . albicans . Spontaneous release was the amount of 51Cr released into the medium by uninfected host cells . Total incorporation was the sum of the amount of 51Cr released into the medium and remaining in the host cells . Each assay was performed in triplicate on three separate occasions . The capacity of the various C . albicans strains to invade a three dimensional model of the oral mucosa in vitro was determined by our previously described method [34] , [61] . OKF6/TERT-2 oral epithelial cells were grown on top of a feeder layer of collagen embedded NIH 3T3 cells in 30 mm diameter cell culture inserts ( Millipore ) . They were infected via their apical surface by adding 105 C . albicans cells in 100 µl of airlift medium ( DMEM with 4 . 5 g/l glucose [Fisher Scientific] and Ham's F-12 medium [Invitrogen] mixed 3∶1 , and supplemented with 5 µg/ml insulin , 0 . 4 µg/ml hydrocortisone , 2×1011 M 3 , 3′ , 5-triiodo-L-thyronine , 1 . 8×10−4 M adenine , 5 µg/ml transferrin , 10−10 M cholera toxin , 2 mM L-glutamine; 5% FBS , and penicillin–streptomycin ) . For histopathology , the cultures were fixed after 2 days with 10% formaldehyde in PBS and embedded in paraffin . Thin sections were stained with periodic acid-Schiff and then evaluated by light microscopy . The extent of epithelial cell damage was quantified after 2 days of infection by measuring the accumulation of LDH in the medium using the CytoTox-96 assay ( Promega ) . The capacity of the various C . albicans strains to adhere to and be endocytosed by endothelial and FaDu epithelial cells was quantified using our previously described differential fluorescence assay [25] , [27] , [32] , [36] . The host cells were grown to 95% confluency on fibronectin-coated glass coverslips in a 24-well tissue culture plate . Each coverslip was infected with 105 organisms in RPMI 1640 medium . After a 90 or 150 min incubation , the medium was aspirated , non-adherent organisms were removed by rinsing the coverslips with Hank's balanced salt solution ( HBSS; Irvine Scientific ) , and cells were fixed with 3% paraformaldehyde . The adherent and non-internalized portions of the organisms were stained with rabbit anti-C . albicans antiserum ( Biodesign International ) conjugated with Alexa 594 ( Molecular Probes ) . Next , the host cells were permeabilized with 1% Triton X-100 for 15 min , and then all of the C . albicans cells were stained with rabbit anti-C . albicans antiserum conjugated with Alexa 488 ( Molecular Probes ) . The coverslips were mounted inverted on a microscope slide and organisms were viewed under epifluorescence . The number of endocytosed organisms was determined by subtracting the number of non-endocytosed organisms ( which fluoresced red ) from the number of cell-associated organisms ( endocytosed plus non-endocytosed organisms , which fluoresced green ) . Organisms that were partially internalized were considered to be endocytosed . At least 100 organisms were counted per coverslip , and each experiment was repeated in triplicate , at least three times . Results were expressed as the number of endocytosed and cell-associated organisms per high powered field . The capacity of each strain to bind to N-cadherin and E-cadherin was determined using our affinity purification procedure as outlined previously [36] , [37] . Germ tubes were prepared by incubating yeast-phase cells for 90 min at 37°C in 150 mm diameter Petri dishes containing RPMI 1640 medium buffered to pH 7 . 5 with 150 mM HEPES . The resulting germ tubes were removed by scraping and rinsed once with PBS containing calcium and magnesium . The germ tubes ( 2×108 cells ) were incubated for 1 h on ice with 250 µg of cell membrane proteins of endothelial cells or FaDu epithelial cells in PBS with calcium and magnesium containing 1 . 5% octyl-glucopyranoside and protease inhibitors . The unbound proteins were removed by rinsing in the same buffer , after which the proteins that remained bound to the germ tube were eluted with 6 M urea . The eluted proteins were separated by SDS-PAGE . N-cadherin and E-cadherin were detected by immunoblotting with anti-N-cadherin murine monoclonal antibody ( clone 32 , Transduction Laboratories ) , and an anti-E-cadherin murine monoclonal antibody ( clone HECD 1 ) , respectively . Recombinant Ssa1 ( rSsa1 ) and rSsa2 were produced in S . cerevisiae as outlined previously [16] . Latex beads were coated with these proteins or biotinylated BSA as described [37] , [62] . Briefly , 5×107 fluorescent , yellow-green , amine-modified , 2 . 0 µm diameter polystyrene latex beads ( Sigma-Aldrich ) were washed with PBS followed by coupling buffer ( 0 . 2 M Na2HCO3 [pH 8 . 5] , 0 . 5 M NaCl ) . The beads were then incubated with rSsa1 , rSsa2 ( 0 . 5 mg/ml ) or coupling buffer at 37°C for 30 min . The beads that had been coated with Ssa proteins were incubated with 1% rabbit serum , while the control beads were incubated with biotinylated BSA ( 1 mg/ml ) . All beads were sonicated briefly , blocked with unlabeled BSA and rinsed PBS-BSA . They were suspended in PBS containing 2 mg BSA per ml . Binding of rSsa1 or rSsa2 to the beads was verified by indirect immunofluorescence with an anti-Xpress monoclonal antibody ( Invitrogen ) directed against the Xpress leader sequence of these recombinant proteins . Binding of biotinylated BSA to the beads was confirmed using Alexa 568 conjugated streptavidin ( Invitrogen ) . The interactions of these beads with endothelial and FaDu epithelial cells were determined by the differential fluorescent assay described above . The relative transcript levels of SSA1 and SSA2 in hyphae of the various strains of C . albicans were determined by our previously described method [63] . Yeast cells were suspended in RPMI 1640 medium and added to FaDu epithelial cells at a final concentration of 5×105 cells/cm2 . After a 90 min incubation , the non-adherent organisms were removed by rinsing with ice-cold distilled water . Next , ice-cold DEPC-treated water was added and the C . albicans and epithelial cells were removed with a cell scraper . The mixture was vortexed for 30 sec to lyse the epithelial cells and then the organisms were collected by a brief centrifugation at 4°C . These organisms were suspended in TES buffer ( 10 mM Tris , 10 mM EDTA , 0 . 5% SDS ) , and then snap frozen in liquid nitrogen . The total time from rinsing the organisms to freezing them in liquid nitrogen was less than 5 min . At a later time , the cells were thawed on ice and the fungal RNA was extracted by the hot phenol method [64] . For real-time PCR , the C . albicans RNA was treated with DNase I ( Ambion ) , after which cDNA was synthesized using MMLV reverse transcriptase ( Ambion ) . Quantitative real-time PCR was carried out using the SYBR green PCR kit ( Applied Biosystems ) and an ABI 7000 Real-Time PCR System ( Applied Biosystems ) following the manufacturer's protocol . The results were analyzed by the 2-ΔΔCT method [65] using ACT1 as the endogenous control . The SSA1 and SSA2 transcript levels were determined in three biological replicates , each tested in duplicate . Flow cytometry was used to quantify the amount of Als3 on the surface of the various strains using a slight modification of our previously described method [37] . Briefly , 3×10 106 yeast-phase cells in 15 ml RPMI 1640 medium buffered to pH 7 . 5 with 10 mM HEPES were added to 100 mm diameter Petri dishes and incubated in 5% CO2 at 37°C . After 90 min , the resulting germ tubes were scraped from the dishes with a cell scraper , fixed with 3% paraformaldehyde , blocked with 1% goat serum , and then stained with a rabbit polyclonal anti-Als3 antiserum [37] followed by an Alexa 488 conjugated goat anti-rabbit antibody ( Molecular Probes ) . An als3Δ/Δ mutant was included in these experiments as a negative control . The amount of Als3 surface staining on 104 germ tubes per strain was analyzed by flow cytometry . To analyze the surface distribution of Als3 on the various strains , 105 yeast-phase cells in 1 ml RPMI 1640 medium buffered to pH 7 . 5 with 10 mM HEPES were added to 12 mm diameter glass cover slips in a 24-well tissue culture plate . After a 90 min incubation in 5% CO2 at 37°C , the resulting germ tubes were fixed with 3% paraformaldehyde , and blocked with 1% goat serum , and then stained with the rabbit polyclonal anti-Als3 antiserum followed by an Alexa 488 conjugated goat anti-rabbit antibody . Next , the germ tubes were counter stained with the Alexa 594 conjugated anti-Candida antibody to label the cell surface . The coverslips were mounted inverted on microscope slides and imaged by confocal microscopy . Stacked images were acquired along the z-axis and then combined for the final images . The surface expression of Ssa1 on C . albicans expressing Ssa1-GFP was determined by a similar process , except that FaDu oral epithelial cells were grown on the coverslips prior to adding the organisms and the anti-Als3 antiserum was omitted . The C . albicans strains were screened for phospholipase activity by the method of Samaranayake et al . [49] . Briefly , egg yolk agar consisting of Sabouraud's dextrose agar , NaCl , CaCl2 and egg yolk emulsion was prepared . A 10 µl of suspension of 107 yeast-phase cells per ml in 10 mM sodium phosphate buffer was plated on the agar and incubated at room temperature for 5 days . The phospholipase activity of each strain was determined by measuring the width of zone of precipitation around the colony . The total extracellular protease activity of the various strains was assessed using BSA agar by the method of Ruchel et al . [66] . BSA agar ( 0 . 2% BSA , 1 . 17% yeast carbon base , and 0 . 01% yeast extract , pH 5 ) was spot inoculated with the various C . albicans strains and then incubated at 37°C . After 5 days the agar was stained with 0 . 5% amido black and the width of the zone of clearance around each colony was measured . The susceptibility of the various C . albicans strains to stressors were tested by spotting dilutions of 104 to 101 yeast-phase cells in a total volume of 7 µl onto YPD agar plates containing menadione ( 12 . 5 and 25 µM ) ; H2O2 ( 1 , 2 and 4 mM ) , Calcafluor white ( 25 and 50 µM ) , NaCl ( 1 and 2 M ) and SDS ( 0 . 002% ) . The plates were incubated at 30°C for 24 hours and then imaged . The susceptibility of the different strains to damage by a neutrophil-like cell line was determined by the 2 , 3-bis ( 2-methoxy-4-nitro-5-sulfophenyl ) -5-[ ( phenylamino ) carbonyl]-2H-tetrazolium hydroxide ( XTT ) assay , as previously described [67] . Briefly , HL-60 cells were cultured in RPMI 1640 medium containing 10% fetal bovine serum and 25 mM HEPES . They were induced to differentiate into neutrophil-like cells by exposure to 1 . 25% dimethyl sulfoxide for 7 days . For the damage assay , C . albicans yeast suspended in Dulbecco's Modified Eagle Medium containing 10% fetal bovine serum were added to 96-well plates at concentrations ranging from 105 to 2×104 cells/well . There was a linear relationship between viable cell number and colorimetric signal ( XTT activity ) in this concentration range with all three strains ( not shown ) . HL-60 cells were added to the C . albicans cells at effector to target cell ratios ( E:T ) ranging from 5∶1 to 1∶2 . After incubation at 37°C in 5% CO2 for 3 hours , the medium was aspirated and the HL-60 cells were lysed with sterile H2O . To each well was added 100 µl of a mixture of XTT and coenzyme Q0 ( 0 . 25 mg/ml XTT and 40 µg/ml coenzyme Q0 ) , after which the plate was incubated at 37°C in 5% CO2 for 2 h . Supernatants were transferred into new plates , and optical densities ( OD ) were measured by an Opsys Microplate Reader ( Thermo Labsystems ) at 450–490 nm , with a 630 nm reference filter . Antifungal activity was calculated according to the following formula: %fungal damage = ( 1−x/n ) *100 , where x is the OD450 of experimental wells ( C . albicans with effectors ) and n is the OD450 of control wells ( C . albicans only ) . Each experiment was performed in triplicate and repeated 3 times . The results of the survival experiments were analyzed with the Log-Rank Test , and the organ fungal burden data were analyzed using the Log Rank Test . The results of the in vitro experiments were analyzed using the Student's T test . P values ≤0 . 05 were considered to be significant . All experiments were approved by the Los Angeles Biomedical Research Institute Animal Care and Use Committee and as outlined in the Guide for the Care and Use of Laboratory Animals of National Institutes of Health . The collection of umbilical cords for the harvesting of endothelial cells used in these studies was approved by the Institutional Review Board of the Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center . | The fungus Candida albicans can proliferate in the mouth , causing oropharyngeal candidiasis . In other patients , it can enter the bloodstream and spread throughout the body , resulting in hematogenously disseminated candidiasis . Fungal invasion of host cells is a key feature of both types of infection . One mechanism by which C . albicans invades both the epithelial cell lining of the oropharynx and the endothelial cell lining of the blood vessels is by inducing its own uptake . This uptake is induced in part by the binding of the C . albicans invasin Als3 to host cell proteins , which include N- and E-cadherin . Here we show that C . albicans Ssa1 , a member of the 70 kDa heat shock protein family , is expressed on the surface of C . albicans where it functions as an invasin . The key role of Ssa1 in host cell invasion is illustrated by the reduced capacity of an ssa1Δ/Δ null mutant to induce its own uptake by epithelial and endothelial cells in vitro , and by the significantly attenuated virulence of this mutant in mouse models of oropharyngeal candidiasis and disseminated candidiasis . Thus , Ssa1 is the second identified invasin of C . albicans . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"microbiology/cellular",
"microbiology",
"and",
"pathogenesis"
] | 2010 | Host Cell Invasion and Virulence Mediated by Candida albicans Ssa1 |
Hepatitis C Virus ( HCV ) is a major public health concern , with no effective vaccines currently available and 3% of the world's population being infected . Despite the existence of both B- and T-cell immunity in HCV-infected patients , chronic viral infection and HCV-related malignancies progress . Here we report the identification of a novel HCV TCR from an HLA-A2-restricted , HCV NS3:1073–1081-reactive CTL clone isolated from a patient with chronic HCV infection . We characterized this HCV TCR by expressing it in human T cells and analyzed the function of the resulting HCV TCR-transduced cells . Our results indicate that both the HCV TCR-transduced CD4+ and CD8+ T cells recognized the HCV NS3:1073–1081 peptide-loaded targets and HCV+ hepatocellular carcinoma cells ( HCC ) in a polyfunctional manner with cytokine ( IFN-γ , IL-2 , and TNF-α ) production as well as cytotoxicity . Tumor cell recognition by HCV TCR transduced CD8− Jurkat cells and CD4+ PBL-derived T cells indicated this TCR was CD8-independent , a property consistent with other high affinity TCRs . HCV TCR-transduced T cells may be promising for the treatment of patients with chronic HCV infections .
Hepatitis C Virus ( HCV ) infection is a major public health concern with approximately 3% of the world's population being infected [1] . Unfortunately , 70–80% of infected individuals are unable to clear the virus , resulting in a chronic infection with the potential for developing severe liver diseases such as cirrhosis and hepatocellular carcinoma ( HCC ) [2] , [3] . These liver diseases are the major indication for liver transplantation in the US and Europe [4] , [5] . The combination therapy of interferon-α and ribavirin is used to treat HCV infections with limited success [6] . The development of preventative and therapeutic vaccines has been hindered by a lack of relevant animal models to study HCV viral replication and disease progression in vivo . Both cellular and humoral immunity exists against HCV proteins in HCV-infected individuals [7] . However , not all HCV-infected patients can mount an effective anti-HCV immune response leading to the reduction of the viral load [8] , [9] . There is evidence that demonstrates that the HCV genome mutates rapidly suggesting that mutations in T-cell and B-cell epitopes lead to immune escape variants which may be the main reason for HCV persistence in chronically infected patients [10]–[12] . Therefore , until better immune-based strategies are developed , immune therapy will have limited benefit for HCV-infected patients . An approach has been described in which retroviral vectors encoding T-cell receptor ( TCR ) genes are used to redirect the specificity of normal peripheral blood lymphocyte ( PBL ) -derived T cells to recognize the melanoma associated antigen , MART-1 [13] , [14] . Subsequently , this approach has been extended to other tumor antigens and viruses [15]–[28] . In fact , three phase I/II clinical trials using this approach to treat melanoma have been reported [29]–[31] . In two of these studies , no serious adverse events were observed and a few objective clinical responses were reported [29] , [30] . However , the third study reported an increase in the frequency of clinical responses but a few patients experienced adverse events [31] . With TCR gene transfer becoming a reality for cancer patients , it opens the possibility for engineering a patient's own T cells to recognize their HCV virus-infected cells , regardless of their immune status . It has also been well known that the HCV genome contains several regions and it is genetically unstable and mutates readily . The high variation of HCV is used to produce escape mutants that can sneak past the immune response of the host . The variants also play a significant role in the progression of virus infection due to resistance to immunotherapy . We have previously identified HCV NS3:1406–1415 reactive T cells that express high-affinity TCRs [24] , [32] . In the current study , we cloned novel HCV TCR genes from an HLA-A2-restricted , HCV NS3:1073–1081-reactive , T-cell clone isolated from a patient with chronic HCV infection . This is an important epitope since it is frequently the immunodominant epitope targeted by anti-HCV CTL in HCV infected patients [33] . It also shares sequence homology with a peptide from the influenza A virus and CTL have been shown to crossreact with both peptides [34] . And finally , it is often mutated in HCV immune escape variants making it likely to be an important target in the anti-HCV immune response [35] . For the first time , we demonstrate that a recombinant retroviral vector encoding the HCV NS3:1073–1081 TCR could efficiently transduced both CD4+ and CD8+ T cells . Functional analysis demonstrates that the HCV TCR-transduced both CD4+ and CD8+ T cells produce interferon-γ , TNF-α , and to a lesser extent , IL-2 when stimulated with the peptide-loaded targets or HCV+ hepatoma cells . Our results indicate that an HCV TCR can engineer antigen reactive CD4+ and CD8+ T cells , raising the possibility that we can provide any HCV patient with a source of autologous HCV-reactive T effector and helper cells which have been implicated in eradicating HCV infections .
An HCV NS3:1073–1081-reactive CTL clone was isolated from the blood of a patient with a chronic HCV infection by limiting dilution cloning . The T cell clone was analyzed for antigen recognition in cytokine release assays . As shown in Figure 1 , the T cell clone secreted significant amounts of interferon-γ when stimulated with T2 cells loaded with the HCV NS3:1073–1081 peptide but not the control CMVpp65 or HCV NS3:1406–1415 peptides . These results indicate that the T cells isolated from the patient with a chronic HCV infection were reactive with the HCV NS3:1073–1081 antigen . The HCV NS3:1073–1081 TCR α and β chains from the HCV NS3:1073–1081-reactive , T cell clone were identified as described [36] . The germline V genes , J regions , and the unique CDR3 region sequences for each TCR chain are shown in Figure 2 . DNA sequence analysis of random 5′ RACE TCR α chain cDNA clones revealed the HCV NS3:1073–1081 T cell clone expressed a single TCR α chain consisting of AV20s1/AJ10/AC . This TCR α chain was in-frame and contained all of the landmarks consistent with a functional TCR α chain . The TCR β chain was identified by RT-PCR using a panel of TCR BV subfamily specific primers as described [37] . The only primer that amplified a fragment of the predicted size was the BV13 primer suggesting that the TCR expressed by the HCV NS3:1073–1081 reactive T cell clone used member of the BV13 subfamily ( data not shown ) . DNA sequence analysis of that PCR fragment revealed that the TCR β chain consisted of BV13s6/BJ2s1/BJ2s7/BC2 . Like the TCR α chain , the DNA sequence of the TCR β chain indicated it was in-frame and contained all of the features consistent with a functional TCR β chain . The identification of the TCR β chain being BV13 was confirmed by immunofluorescence staining with an anti-Vβ13s6 mAb ( data not shown ) . Thus , the HCV NS3:1073–1081 T cell clone expresses an AV20s1/BV13s6 TCR . The critical feature for a TCR gene-modified T cell is its ability to recognize endogenous antigen on the target-cell surface . However , human liver or HCC cells infected with HCV were not available for our experiments . Therefore , we established an HCV expressing HCC cell line to test the ability of our TCR-transduced T cells to recognize HCV+ liver tumor cells . An HCV expression construct was prepared by fusing the HCV NS3:1073–1081 minigene to the EGFP gene which was used as a marker to monitor the level of antigen expression by the HCC cells . The HCV/EGFP fusion construct was inserted into the retroviral vector pMFG ( Figure 3A ) which was used to transduce the HLA-A2+ hepatocellular carcinoma cell line , HepG2 ( HLA-A2 expression is shown in Figure S1 ) . HCV minigene-positive cells were defined by their EGFP expression as measured by flow cytometry ( Figure 3B ) and EGFP expressing cells were sorted by FACS for high and homogeneous antigen expression . The HCV+ HCC cells were then used as stimulators in cytokine release assays to evaluate the function of HCV TCR transduced T cells . We have generated the recombinant retroviral construct containing the TCR α and β chains of the HCV NS3:1073–1081 reactive T cell clone ( Figure 4 ) . To verify the expression and function of this cloned TCR , we used this retroviral vector to transduce Jurkat 76 cells . Jurkat 76 cells are a TCR α− and β− derivative of the CD8− human T cell lymphoma Jurkat cell line . Since Jurkat 76 cells are TCR negative , any introduced TCR would not have to compete with the endogenous TCR and its expression can be monitored by staining with anti-CD3 mAb . Furthermore , Jurkat 76 cells expressing a cloned TCR secrete IL-2 upon antigen stimulation in an antigen-specific fashion . Therefore , Jurkat 76 cells are an excellent model to evaluate the expression and function of any cloned TCR . As shown in Figure 5A , Jurkat 76 cells expressing the HCV NS3:1073–1081 TCR stained with anti-Vβ13 . 6 and anti-CD3 mAb's indicating the TCR could assemble on the surface of the Jurkat 76 cells . When stimulated with antigen , these HCV TCR transduced Jurkat 76 cells secreted significant IL-2 in response to T2 cells loaded with the HCV NS3:1073–1081 peptide but not T2 cells alone or T2 cells loaded with the CMV pp65:495–503 peptide ( Figure 5B ) . The HCV TCR transduced Jurkat 76 cells also recognized HepG2 cells loaded with the HCV NS3:1073–1081 peptide or transfected to express the HCV NS3:1073–1081 epitope . It should be noted that despite the mock transfectants having higher expression of EGFP ( Figure 3B ) , they were not recognized by the Jurkat cells expressing the HCV TCR . More importantly , recognition of HCV+ HepG2 cells by CD8− Jurkat cells indicates our HCV TCR transfers CD8 independent tumor cell recognition to alternate effectors . These results indicate the HCV TCR is functional and has high affinity for antigen . Although Jurkat 76 cells are a good model cell line for verifying the function of a cloned TCR , they can't be used in preclinical animal studies or clinical trials to control HCV infections or the growth of HCV+ HCC cells . Therefore , it is critical to evaluate the function of normal PBL-derived T cells expressing our HCV TCR , particularly with regards to their ability to recognize a physiologically relevant target such as hepatocellular carcinoma cells . To accomplish this goal , we generated populations of HCV TCR transduced PBL-derived T cells from a total of seven normal healthy donors . The level of expression and the percent HCV TCR transduced T cells was measured by anti-Vβ13 . 6 mAb staining . The results from a typical HCV TCR transduced T cell culture is shown in Figure 6A . Compared to the isotype control , mock transduced T cells contained 1 . 1% Vβ13 . 6 staining cells . This represents the frequency of Vβ13 . 6 staining cells present in normal PBL . The HCV TCR transduced T cell cultures contain 36% Vβ13 . 6 staining cells with the level of TCR expression being variable as expected by a TCR-transduced T cell population . These results indicate that our HCV TCR can be efficiently expressed by PBL-derived T cells from normal donors . However , despite having anywhere from 20%–40% Vβ13 expressing T cells in the HCV TCR transduced T cell cultures , only about 0 . 25% of the CD4+ and CD8+ T cells bind HCV 1073 peptide loaded pentamers ( Figure S2 ) . We and others have found that tetramer binding does not always correlate with TCR expression and function so this result was not surprising [38]–[40] . PBL-derived T cells from the normal donors were transduced with our HCV TCR and assessed for their ability to recognize antigen . Using a combination of cytokine release and intracellular cytokine staining , we evaluated the antigen reactivity of each of the HCV TCR transduced T cell cultures . All of the bulk T cell cultures produced significant amounts of interferon-γ when stimulated with T2 cells loaded with the HCV NS3:1073–1081 peptide but not T2 cells alone or T2 cells loaded with the control CMV pp65:495–503 peptide ( Figure 6B , Figure 7 , and Figure S3 ) . These HCV TCR transduced T cells did produce TNF-α and IL-2 upon stimulation with peptide loaded T2 cells ( Figure 7 and Figure S3 ) . The HCV TCR transduced bulk T cells also efficiently recognized HepG2 cells ( which naturally express HLA-A2; Figure S1 ) loaded with HCV NS3:1073–1081 peptide or transfected to express the HCV NS3:1073–1081 epitope ( Figure 6 and Figure S3 ) . It should be noted that despite the mock transfectants having higher expression of EGFP ( Figure 3B ) , they were not recognized by the normal PBL-derived T cells expressing the HCV TCR . Moreover , TCR-transduced PBL demonstrated cytotoxicity as shown by the production of CD107a ( Figure 8 ) . Therefore , our HCV TCR efficiently engineers normal PBL-derived T cells to recognize HCV peptide loaded targets with a polyfunctional response ( production of IFN-γ , IL-2 , TNF-α and CD107a ) . Recognition of peptide loaded targets by HCV TCR-transduced T cells confirms the reactivity of our HCV TCR in normal PBL-derived T cells . However , it is more important to verify the recognition of antigen presented by HCV+ cells such as HCC cells . The recognition of HCV+ HepG2 cells by the HCV TCR transduced Jurkat 76 cells indicated that our HCV TCR transfers CD8-independent tumor cell recognition to alternate effectors ( Figure 5B ) . We have previously reported that CD8 independent TCR's are capable of generating MHC class I restricted CD4+ T cells making it possible to provide patients with a novel source of T cell help [41] , [42] . To determine if our HCV TCR can generate MHC class I restricted CD8+ effector and CD4+ helper T cells , we transduced PBLs derived T cells from three healthy donors and purified the CD4+ and CD8+ T cells to greater than 99% purity using immunomagnetic beads to measure cytokine production by ELISA ( Figure 6 and Figure S3 ) or analyzed each subset for intracellular cytokine production or CD107a expression ( Figure 7 and Figure 8 ) . CD4+ and CD8+ T cells , transduced to express our HCV TCR , produced significant amounts of interferon-γ , TNF-a , IL-2 , and CD107a when stimulated with HCV peptide loaded T2 cells or HepG2 cells but not controls . Importantly , the HCV TCR-transduced CD4+ T cells secreted significant amounts of cytokine when stimulated with HCV+ tumor cells . These results indicate that our HCV TCR can engineer both CD8+ T cells and CD4+ T cells to recognize HCV+ cells . Also , our ability to generate MHC class I restricted CD4+ T cells raises the possibility that we can provide any HCV patient with a source of autologous HCV-reactive T helper cells which has been implicated in eradicating HCV infections [43] . It has been shown that there is a correlation between the functional avidity of a T cell and its ability to recognize tumor cells or virus-infected cells [43]–[47] . Furthermore , T cells expressing a high affinity TCR have been shown to be exquisitely sensitive to low levels of antigen [48] . T cells with identical specificities , but different functional avidities , influence each other during activation and homeostatic proliferation [49] . T cells exhibiting increased sensitivity to stimulation , or a lower threshold , are said to have a relatively high functional avidity [44]–[47] . T-cell responsiveness to peptide is commonly used as a measure of T cell avidity as it provides a measure of the stimulation threshold required to activate T cell effector functions . Relative avidities were evaluated by measuring T-cell interferon-γ production . To test the avidity of the HCV TCR-transduced T cells , we loaded different peptide concentrations on T2 or HepG2 cells and incubated with HCV TCR transduced T cells . The ability of the transduced T cells to produce interferon-γ was measured under conditions of increasing concentrations of peptide stimulation . As shown in Figure 9 , the two representative HCV TCR transduced T cell cultures ( Donors 4 and 5 ) had high avidity for antigen since they secreted significant amounts of interferon-γ when stimulated T2 cells loaded with 5 nM or less of peptide . Similar results were found with three other HCV TCR transduced T cells cultures ( Donors 1–3 ) with the functional avidity of the HCV TCR transduced T cells being approximately half a log lower than the parent T cell clone ( <1 . 0 vs <0 . 5 nM ) ( Figure S4 ) . In contrast , Peptide loaded HepG2 cells were not recognized as well as the T2 cells since it required between 50–500 nM peptide to stimulate the HCV TCR transduced T cells . This was not surprising since our HepG2 cells but not our T2 cells have their MHC class I molecules loaded with peptides requiring peptide exchange on the HepG2 cells for T cell recognition to occur . In fact , the HCV TCR transduced T cells did not recognize the HepG2 cells loaded with 5 µM peptide as well as the HepG2 cells expressing the HCV 1073 minigene ( 576/345 pg/ml vs 717/418 pg/ml respectively for donors 1 and 2 ) further supporting the notion that exogenous peptide loading was less efficient than endogenous peptide loading . Compared to other published studies , our functional avidity measurements using peptide loaded T2 cells indicate our HCV TCR transduced T cells have relatively high avidity for antigen .
The relationship between T cell avidity and the clearance of viral infections and tumor cells has been well documented [44] , [45] . Many studies have been directed at elucidating the relationship between T-cell activity and TCR affinity , dissociation rate , and CD8+ dependence . CD8+ played an essential role in T-cell recognition of low-affinity T-cell reactions [48] . However , we speculated that any TCR that can bind peptide/MHC complexes without CD8 would have higher relative affinity than a TCR that requires CD8+ for binding . To date , only a limited number of CD8-independent TCRs have been cloned and characterized [50] . The novel HCV TCR described herein exhibits CD8-independent target cell-recognition since the HCV TCR-transduced CD4+ T cells could secret interferon-γ and IL-2 when stimulated with peptide-loaded targets or HCV+ HCC cells ( Figure 6B and Figure S3 ) . Based on this study , we conclude that the affinity of this HCV TCR is higher than other TCRs that require the CD8 coreceptor for target-cell recognition , These results indicate that T cells isolated from patients with chronic HCV infection can have high affinity TCRs and these TCRs may be important for developing novel TCR-based gene therapy studies . Thus , we have identified another high affinity TCR that could be used to engineer normal PBL-derived T cells for clinical application . CD4+ T cells are thought to contribute to anti-viral immune responses by secreting cytokines , thereby providing help to CD8+ T cells [51] , [52] . Antigens are taken up by antigen presenting cells which may activate CD4+ T cells to secrete either Th1 or Th2 cytokines . By producing Th1-cytokines like interferon-γ and IL-2 CD4+ T cells contribute to anti-viral immune responses providing help to CD8+ T and B cells . Furthermore , CD4+ T cell lines and clones can display direct cytotoxic effector function [51] , [52] . The identified HCV TCR has been successfully transduced into CD4+ T cells and these TCR-transduced CD4+ T cells may not only provide help to CD8+ T cells , but also directly act on the HCV+ target cells such as HCV-infected cells and HCV+ HCC cells . This is especially important for clearing HCV infection because one of the fundamental problems typical of chronic HCV infection is a weak or absent HCV- specific CD4+ T-cell response [43] , [53] . The instability of the HCV genome makes the identification of this high affinity HCV NS3:1073–1081 TCR particularly important . We and others have previously shown that T cells can express two functional TCRs capable of recognizing both target antigens [54]–[56] . Therefore , when combined with our previously identified HCV NS3:1406–1415 TCR , T cells expressing both TCRs might be effective against HCV immune escape variants for treatment of HCV-associated diseases . Adoptive transfer of HCV TCR-transduced T cells may show promise as a new treatment for patients with chronic HCV infection or HCV-related malignancies , particularly in light of the recent demonstration that HCC express HCV antigens [57] .
T2 and HepG2 cells were obtained from the American Type Culture Collection ( Rockford , MD ) . The TCR-negative Jurkat 76 cell line has been described elsewhere [58] . Unless otherwise indicates , All medium components were obtained from Mediatech ( Herndon , VA ) unless otherwise noted . Jurkat 76 and T2 cell lines were maintained in complete medium consisting of RPMI 1640 medium supplemented with 10% fetal bovine serum ( Tissue Culture Biologicals , CA ) 100 U/mL penicillin , 100 µg/mL streptomycin . Plat-A cells [59] , [60] and HepG2 cells were maintained in Eagle's medium supplemented as described above . TCR-transduced Jurkat 76 cells were maintained in RPMI medium as described above supplemented with 2 mg/mL G418 . To engineer tumor cell lines to express HCV sequences , we first inserted synthetic oligonucleotides encoding the HCV NS3:1073–1031 epitope into the retroviral vector , pMFG-EGFP . Briefly , oligonucleotides encoding the HCV NS3:1073–1031 epitope and containing mutated Nco I restriction sites ( forward: 5′-catgTGCATCAATGGGGTATGCTGGACTGTCgctgcttatgg-3′; reverse 5′- ccataagcagcGACAGTCCAGCATACCCCATTGATGCAcatg-3′ ) were synthesized and annealed . The underlined base pairs indicate the overhang for the ligation of the double stranded oligonucleotides into the pMFG-EGFP vector using a shotgun ligation strategy as described [41] . The recombinant vector was transiently transfected into Plat-A packaging cells and the retrovirus containing supernatant was collected for transduction of HepG2 cells . The expression of the HCV NS3:1073–1081 minigine in the transduced HepG2 cells was confirmed based on the EGFP expression as measured by flow cytometry . The EGFP positive cells were sorted for high and uniform expression and the resulting HCV+ HCC cell line was established . All T cells were maintained in AIM V medium ( Invitrogen , GIBCO ) supplemented with 5% heat-inactivated pooled human AB serum ( Valley Biomedical , Inc ) , 100 U/mL penicillin , 100 µg/mL streptomycin and 300 IU/mL recombinant human IL-2 ( rhIL-2; Novartis Pharmaceuticals Corporation , East Hanover , NJ ) at 37°C in a humidified 5% CO2 incubator . The isolation and characterization of HCV-reactive T-cell clones has been previously described [32] . The HCV NS3:1073–1081-reactive CD8+ T-cell clone used in this study was isolated from a patient with a chronic HCV infection . . All PBMC used in this study came from apheresis products purchased from ( Research Blood Components , L . L . C . , MA ) . Normal PBL-derived T cells were isolated from the PBMC cells of three independent normal healthy donors using Ficoll-Hypaque density gradient centrifugation . The HCV T cell clone and the TCR-transduced T cells were expanded using 30 ng/mL anti-CD3 monoclonal antibody ( Ortho Biotech , Raritan , NJ ) and 300 IU/mL rhIL-2 in the presence of irradiated pooled allogeneic peripheral blood mononuclear cells as feeders as previously described [61] . HCV NS3:1073–1081 ( CINGVCWTV ) , HCV NS3:1406–1415 ( KLVALGINAV ) , CMV pp65:495–503 ( NLVPMVATV ) were obtained from Synthetic Biomolecules ( San Diego , CA ) . T2 or HepG2 cells were loaded with each peptide by incubating 1×106 cells/ml in complete medium containing 5 µg/ml ( unless otherwise noted ) of peptide at 37°c for 2 hours . Peptide-loaded cells were washed with fresh complete medium before coculture with responders . The TCR α chain from the HCV NS3-1073–1081-reactive T-cell clone was identified by 5′ RACE as previously described [37] , [60] . Briefly , total RNA was isolated from 2 . 5×106 cells using TRIzol ( Invitrogen ) , first-strand cDNA was synthesized , and the TCR cDNAs were amplified using the SMART RACE cDNA Amplification kit ( Clontech Laboratories , Inc , Mountain View , CA ) . Fragments containing random TCR α chains were amplified using the Advantage 2 PCR Enzyme system ( Clontech Laboratories , Inc ) using the universal primer A mix and a TCR α constant region ( AC ) specific reverse primer . The random PCR products were ligated into TA PCR2 . 1-Topo cloning vector , and transformed into Escherichia coli TOP 10 competent cells ( Invitrogen ) . Bacterial clones were screened for the presence of TCR α chain cDNA by PCR and random 5′ RACE clones were sequenced using fluorescent dye labeled ddNTPs ( Applied Biosystems Inc , Foster City , CA ) . DNA sequence analysis revealed a single productively rearranged TCR α chain which used the AV20s1 . The full-length α chain was amplified from cDNA using an AV20s1 forward ( 5-AAGTCGACGTTTGCACCTAGAATATGAGGCAAGTGGCG-3 ) and an AC reverse ( 5-AAGTCGACTCAGCTGGACCACAGCCGCAG-3 ) primer containing Sal I restriction sites for subsequent subcloning . The PCR product was ligated into the pCR 2 . 1 TA cloning vector ( Invitrogen ) , and transformed into Escherichia coli TOP 10 competent cells ( Invitrogen ) . Bacterial clones were screened for the presence of the α chain cDNA via PCR and were sequenced to ensure that no errors had occurred during PCR amplification . The TCR β chain from the HCV-reactive T-cell clone was identified via RT-PCR using a panel of TCR β chain V region ( BV ) subfamily specific primers as previously described [37] . Briefly , total RNA was isolated and first strand cDNA was prepared from 2 . 5×106 T cells using the procedure as described above for the α chain identification . A single band was amplified using the BV13 subfamily specific primer and the TCR β chain was identified as BV13s6 based on known TCR BV genomic DNA sequences . The full-length β chain was amplified from cDNA using a BV13s6 forward ( 5′-CTCGAGGCACCTGCCATGAGCATCAGCCTC-3′ ) and a BC2 reverse ( 5′-AACTCGAGCTAGCCTCTGGAATCCTTTCTCTTGACCAT-3′ ) primer that each contained Xho I restriction sites for subsequent subcloning . The PCR fragment was ligated into the pCR 2 . 1 TA cloning vector , and transformed into Escherichia coli TOP 10 competent cells . Bacterial clones were screened for the presence of the β chain gene , and recombinant clones were sequenced to ensure that no errors had occurred during PCR amplification . The SAMEN CMV/SRα retroviral vector has been previously described [14] and was used as the backbone for all retroviral constructs . The TCR α and β chains were linked by a 2A self cleavage peptide . The HCV TCR α chain , 2A linker and β chain fusion gene fragment was inserted into the Xho I and Sal I restriction sites of the retrovirus vector . The configuration of the retroviral vector used in this study is shown in Figure 4 . Retroviral supernatants were prepared using a transient transfection protocol as described [41] . Briefly , 5×106 Plat-A cells were plated in 10 cm poly-D-Lysine coated plates in 10 ml DMEM containing 10% FBS without antibiotics at sufficient density to provide 60% to 70% confluence after 24 hr . Cells were transiently cotransfected with 9 µg of retroviral vector DNA and 4 . 5 µg of plasmid DNA containing the vesicular stomatitis virus envelope gene using Lipofectamine 2000 ( Invitrogen ) . Transfection medium was replaced with 10 ml complete medium after 6 h incubation , and retroviral supernatants were collected after 48 hr . Jurkat 76 were transduced by spinoculation as described [14] . Briefly , Jurkat 76 cells were resuspended at a concentration of 2×106/ml in retroviral supernatant containing 8 µg/mL polybrene . 1 ml of cells was added to each well of a 24-well flat-bottom tissue culture plate then spun for 90 min at 1000×g at 32°C . After centrifugation , the cells were resuspended in their wells , incubated for 4 hr at 37°C , and 1 mL fresh complete medium was added to each well . This spinoculation procedure was repeated the next day using fresh retroviral supernatant . After 24 hr , the transduced cells were resuspended at 5×105/ml in culture medium and transduced cells were selected by the addition of 2 mg/mL of G418 . T cells were transduced by spinoculation as described [14] . Briefly , retrovirus was first loaded onto RetroNectin-coated 24-well flat-bottom non-treated tissue culture plates by adding 1 mL of fresh retroviral supernatant per well and the plates were spun for 2 hr at 2 , 000×g at 32°C . T cells derived from healthy donors were activated using 50 ng/mL anti-CD3 monoclonal antibody and 300 IU/mL rhIL-2 . The activated T cells were resuspended at 1×106 cells/mL with culture medium supplemented with 300 IU/mL of rhIL2 . The T cells were then gently added to the plates and mixed with the viral supernatnant . The plates were continuously centrifuged at 1 , 000×g for 10 min at 32°C . After 24 hours , the transduced T cells were selected by adding 1 mg/mL of G418 . The CD4+ T cells were sorted from the TCR-transduced T cells by positive selection with magnetic beads . The purity of separated CD4+ and CD8+ cells was confirmed by FACS analysis . All T cell and tumor cell lines were stained for immunofluorescence with fluorochrome conjugated anti-CD3 ( APC ) , Vβ13 . 6 ( FITC ) , anti-CD4 ( APC ) , anti-CD8 ( FITC ) , anti-CD107a , ( PE ) and anti-HLA-A2 ( PE ) purchased from BD Biosciences , San Diego , CA . PE conjugated HCV NS3:1073–1081 or HCV Core:132–140 peptide loaded HLA-A2 pentamers were purchased from Proimmune Ltd . , Oxford , United Kingdom . In all experiments , 106 live cells were stained for 30 minutes on ice with individual monoclonal antibodies or pentamers . Cells were washed and stained with a second reagent or analyzed immediately on an Accuri C6 or BD FACSCalibur flow cytometer . The log fluorescence of a minimum of 106 cells was analyzed for each sample . Antigen reactivity by the HCV-reactive T cell clones and HCV TCR transduced cells was measured in cytokine release assays as described [13] . Briefly , 1×105 responder and stimulator cells were cocultured in a 1∶1 ratio in 96-well U-bottom tissue culture plates in 200 µL complete medium . For the Jurkat 76 experiments , 10 ng/mL of PMA ( Sigma-Aldrich , St . Louis , MO ) was added to each well . As a positive control for Jurkat stimulation , maximal cytokine release was obtained by the addition of 1 µg/mL ionomycin ( Sigma-Aldrich ) . Cocultures were incubated at 37°C for 20 hours , and then supernatants were harvested . The amount of cytokine released was measured via ELISA using monoclonal antibodies to interferon-γ ( Pierce , Rockford , IL ) or IL-2 ( R&D Systems , Minneapolis , MN ) . Multiparameter flow cytometry was performed using a BD FACSCanto II instrument ( BD Biosciences , San Jose , CA ) and analyzed using FACSDiva software ( BD ) . Antibodies for cell surface CD3 , CD4 and CD8 and for intracellular IFN-γ , TNF-α , and IL-2 were purchased from BD or eBioscience ( San Diego , CA ) . Transduced T cell cultures were stimulated for 6 hours at 37°C in the presence of brefeldin A ( Sigma-Aldrich ) with equal numbers of T2 cells that had previously been loaded with the TCR-specific antigen ( HCV-1073 peptide , 1 µg/ml ) or with another HCV-derived A2-restricted control peptide ( HCV-132 , 1 µg/ml ) . After stimulation cells were stained for surface antigens , fixed for 30 minutes at 4°C in 100µl Fix and Perm Medium A ( Caltag , Burlingame , CA ) , permeabilized using 100µl Fix and Perm Medium B ( Caltag ) and incubated with anti-cytokine antibodies for 1 hour at 4°C . Cell suspensions were then washed in PBS-BSA-Azide and fixed in 200 µl 1% PFA and acquired after 1 hour . CD107a expression was used as a surrogate marker to assess the cytolytic ability of HCV 1073 TCR transduced T cells . HCV 1073 TCR transduced T were cocultured with a panel of stimulators using methods similar our cytokine release assays described above . Stimulators included T2 cells loaded with the HCV NS3:1073–1081 or CMV pp65;495–503 and tumor targets ( HepG2 and HepG2 expressing the HCV 1073 minigene ) . Briefly , 1×105 responder and stimulator cells were cocultured in a 1∶1 ratio in 96-well U-bottom tissue culture plates in 200 µL complete medium . Cocultures were incubated at 37°C for 20 hours , and then cells were harvested and washed . The cells were stained with anti-CD3 mAb , anti-CD8 mAb and anti-CD107a mAb ( BD Pharmingen , San Diego , CA ) and were analyzed by flow cytometer . Each histogram represents the log fluorescence of 104 live T cells ( gated using CD3 staining ) . | Hepatitis C Virus ( HCV ) is a major public health concern with a large number of individuals infected ( 3% world wide ) . Currently , there is no effective vaccine available to prevent HCV infection and the treatment is effective in less than half of all patients . Therefore , many patients have long term infections that lead to severe liver damage or liver cancer . It has been shown that some HCV infected patients can eliminate the virus and the host immune system is involved . The problem is most people do not have the capacity to fight their HCV infection . We have developed a gene therapy based approach where a patient's own immune cells can be made to recognize cells expressing HCV genes . This can be accomplished regardless of his or her natural capacity to fight their HCV infection . This manuscript describes how normal immune cells can be genetically altered to recognize cells expressing HCV proteins and characterizes their reactivity and sensitivity to antigen stimulation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"gastroenterology",
"and",
"hepatology/gastrointestinal",
"cancers",
"immunology/antigen",
"processing",
"and",
"recognition",
"gastroenterology",
"and",
"hepatology/hepatology",
"virology/new",
"therapies,",
"including",
"antivirals",
"and",
"immunotherapy",
"infectious",
"disea... | 2010 | Transduction of Human T Cells with a Novel T-Cell Receptor Confers Anti-HCV Reactivity |
The haplotype map constructed by the HapMap Project is a valuable resource in the genetic studies of disease genes , population structure , and evolution . In the Project , Caucasian and African haplotypes are fairly accurately inferred , based mainly on the rules of Mendelian inheritance using the genotypes of trios . However , the Asian haplotypes are inferred from the genotypes of unrelated individuals based on population genetics , and are less accurate . Thus , the effects of this inaccuracy on downstream analyses needs to be assessed . We determined true Japanese haplotypes by genotyping 100 complete hydatidiform moles ( CHM ) , each carrying a genome derived from a single sperm , using Affymetrix 500 K Arrays . We then assessed how inferred haplotypes can differ from true haplotypes , by phasing pseudo-individualized true haplotypes using the programs PHASE , fastPHASE , and Beagle . We found that , at various genomic regions , especially the MHC locus , the expansion of extended haplotype homozygosity ( EHH ) , which is a measure of positive selection , is obscured when inferred Asian haplotype data is used to detect the expansion . We then mapped the genome using a new statistic , XDiHH , which directly detects the difference between the true and inferred haplotypes , in the determination of EHH expansion . We also show that the true haplotype data presented here is useful to assess and improve the accuracy of phasing of Asian genotypes .
Chromosomes carry records of mutational and recombinational events that they have experienced during their evolution as haplotypes . Thus , the exact knowledge of the genome-wide haplotype structure of a population can provide information essential to the study of the population history , and to association studies to discover alleles responsible for common diseases that are expected to have a long history since their first emergence . In this spirit , the International HapMap Project has been launched , and has achieved the construction of haplotype maps of extremely high resolution for three major populations: Africans , Asians , and Europeans [1] , [2] . However , the accuracy of the haplotypes was not equal for the three populations , a fact that can affect the results of downstream studies . Resolving haplotype structures from the available genotype data on the level of entire genomes remains a challenging task , both experimentally [3] and computationally [4] . Since the size of the datasets that researchers will want to phase is increasing dramatically , in terms of the numbers of both the loci and the individuals , the statistical estimation of haplotypes by phasing genotypes of randomly sampled individuals is potentially valuable , if it can be done accurately . Estimation by phasing has therefore received much attention in recent years , and several computational and statistical approaches have been developed [4]–[8] . Among these , the coalescence model and haplotype clustering have generally been accepted as reliable methods for inferring phases . However , even with these methods , haplotype estimation is not perfect in the absence of information from genetically related individuals [9] . In the International HapMap project , African ( YRI ) and European ( CEU ) haplotypes were inferred based mainly on family data ( trios of children and their two parents ) . On the other hand , the East Asian ( JPT+CHB ) samples consisted of arbitrarily sampled individuals without any family data . Haplotypes that are inferred without trio information and based on relatively small samples may contain significant phasing errors ( switch error ) [10] . To avoid this issue of phasing uncertainty , we have experimentally determined genome-wide definitive haplotypes using complete hydatidiform moles [11] , [12] . The formation of a CHM is started by a maternal event of rare sporadic occurrence: enucleation of the oocyte , followed by conjugation with a sperm , occurring in 0 . 1–0 . 05% of pregnancies . Thus , CHM genotypes are direct reads of sperm haplotypes that are unbiased and highly unlikely to be related to one another [13] , [14] , such that they can be considered to be randomly sampled from the population to which they belong [11] , comparable to a population of lottery winners . In previous work , we genotyped 74 Japanese CHMs with 280 K Perlegen SNPs [11] , [12] . In the present study , we expanded our definitive haplotyping project of Japanese , by examining 100 CHMs using Affymetrix 500 K Array Sets . Using these data , we performed a genome-wide scan to detect significant differences between the definitive and inferred haplotypes . Our approach is a modification of the statistical method used to evaluate signals of positive selection from the extended haplotype homozygosity ( EHH ) , first introduced by Sabeti et al . and further developed by others [15]–[18] . In this study , we compared integrated EHH of the same allele within a single population using different datasets , i . e . , definitive and inferred haplotypes , and captured significant regional departures of this measure from background levels . We also show that the availability of definitive haplotypes can improve the accuracy of the inference of haplotypes of unrelated Asian genomes .
The haplotype datasets of CHMs ( this study ) and the HapMap Project ( CEU , CHB , JPT , and YRI ) were used for comparison . The SNP allele calls of the Affymetrix Gene Chip Human Mapping 500 k Array Sets were used throughout , except where otherwise stated , because they were available in all the datasets . Two JPT samples , NA18987 and NA18992 , whose estimated cryptic relatedness coefficients were more than 1/32 [2] , were excluded from the analysis in order to avoid inappropriate inflation of the haplotype expansions due to their unusually long haplotype homozygosity [11] . To provide a comprehensive assessment of the haplotypes constructed by a variety of algorithms and samples , we prepared the following datasets , and named them using a “prefix-sample” rule . The prefix “PHASEd+” means that the parental haplotypes were inferred using the PHASE [4] program from the genotype data of trios , supplemented with pedigree information . “PHASEd-” means that the phasings were done by the same program from parental genotype data , but in the absence of pedigree information . “PHASEd+CEU” , “PHASEd+YRI” , and “PHASEd-JPT” were the corresponding haplotype datasets downloaded from the HapMap database ( Release 22 , corrected on June 30 , 2008 ) [1] and extracted as the subsets for Affymetrix 500 K SNPs . In this study , some of the haplotypes of CEU parents were regionally inferred from their genotypes without trio information using PHASE ( thus , “PHASEd-CEU” ) . The prefix “fastPHASEd-” means that the haplotypes were inferred from genotypes using the fastPHASE [8] program without pedigree information . The definitive haplotype dataset of CHMs lacks a prefix , since these haplotypes were determined directly , without inference . The prefixes were followed by the names of diploid genotype datasets , some of them carrying a second prefix , “pi . ” This stands for “pseudo-individuals , ” and indicates that the diploid datasets were made using randomly paired haplotypes that were directly determined ( piCHM ) . The datasets and their names are summarized in Table 1 . We applied principal component analysis [19] , [20] to our CHM datasets , together with those of the HapMap Project , in order to confirm the identity of each set and to detect possible hidden population structure in the samples . Figure S1 shows the plots for the first two eigenvectors calculated to resolve three Asian datasets ( Figure S1A ) , or all five datasets of three ethnicities ( Figure S1B ) , after filtering out SNPs with low call rates . The figure shows that both CHMs and JPTs were in the same cluster and were different from even the closest population , CHB , except for pseudo-individuals with the CHM065 haplotype ( see also Figure S1C , Figure S1D , and Table S1 ) . This haplotype seems to be of recent Asian ancestry , based on its coordinate . We excluded this haplotype from further analyses . After this exclusion , no substructure was discernible in the CHM dataset , and both CHMs and JPTs were seen to belong to the same population . Statistical estimates of LD from inferred haplotypes continue to be important for the purposes of defining a set of tagging SNPs [21] , [22] and imputation-based approaches [23] , [24] in association studies . To assess the accuracy of inferred haplotypes in estimating the LD measure , we compared the r2 calculated from the definitive haplotypes of CHM and phased piCHM using pairs of autosomal common SNPs ( minor allele frequency >5% ) within 200 kb regions . In this case , we employed fastPHASE , although it is known to be less accurate than PHASE , because of the limitations of our computational capacity . As shown in Figure 1A , even by phasing with fastPHASE , we found an extremely high concordance of r2 values between the two datasets ( R = 0 . 991 for SNP pairs with r2>0 . 5 ) , in accordance with earlier results [9] , [25] . Marchini et al . have also assessed the accuracy of estimated r2 by root-mean-square-error ( RMSE ) measure after inferring haplotypes using simulated datasets . The RMSE between haplotypes of CHM and fastPHASEd-piCHM ( 0 . 0254 ) was higher than previous results of PHASE ( 0 . 011 ) for unrelated individuals , which probably is due to the difference of the phasing algorithms , or our real vs . their simulated datasets [9] . However , we also noted that a small but definite fraction of the estimations ( boxed in the figure ) revealed some deviation from strict correlation ( tendency toward underestimation in phased dataset ) even in the high r2 range ( r2>0 . 8 ) . Figure 1B shows the comparison of r2 values obtained from CHM and PHASEd-JPT . The values from the two datasets are again in good correlation ( R = 0 . 952 ) , other than the more scattered nature of the plot compared to Figure 1A ( which is obviously because of sampling error due to the limited number of haplotypes contributing to each dataset ) . We again noticed that a small fraction of the estimates deviate from the strict correlation , similar to the observation in Figure 1A . No further analysis was done on these discrepancies of the r2 values , because they constitute an extremely small fraction and are unlikely to influence downstream analyses , such as imputations in association studies . However , the presence of these discrepant SNPs ( or SNP pairs ) may be indicative of some genomic regions that tend to be wrongly phased without Mendelian information . A haplotype map is a crucial source of information not only for genome-wide association studies , but also for the detection of genomic regions possibly subject to selection . In the latter studies , the conservation of exceptionally long haplotypes is the indicator of positive selection , and long-distance LD structure can have significant effects . Therefore , we compared LD decay curves among these datasets ( Figure 1C ) . As is evident from the figure , the mean r2 values of inferred haplotypes were consistently underestimated at long distances ( >300 kb ) . Thus , we considered it important to reexamine possible positive selection using definitive haplotypes . Voight et al . have introduced a statistic ( integrated haplotype score , iHS ) to identify genomic regions of recent positive selection [17] . Their method is based on the detection of exceptional expansion of extended haplotype homozygosities ( EHH ) , initially proposed by Sabeti et al . as an indicator of genomic region of positive selection [15] . We first carried out iHS analysis using CHM and PHASEd+CEU , both of which are virtually true haplotype datasets . As shown in Figure 2A , prominent iHS peaks were observed at several places , especially at the MHC locus ( black arrows in the figure ) in both CHM and CEU . Another peak , at the lactase locus ( LCT ) , was found only in CEU . These , and other loci that show statistically significant iHS peaks , are listed in Tables S2 and S3 . Voight et al . have examined the haplotypes in HapMap populations by iHS analysis using SNP data of Phase I [17] , and shown that several genomic regions are under positive selection , including the MHC locus at chromosome 6p . Their data indicate that this selection was significant for CEU and YRI , but not for JPT/CHB ( See Table 2 in ref . [17] ) . Exceptional extension of long haplotypes in the MHC locus has also been reported by de Bakker et al . [26] . These authors also found that the extension was much more pronounced for CEU than for CHB or JPT , although this difference was not clearly stated in the text ( see Figure 4 in ref . [26] ) . However , given the central role of this locus in the immune system , it is hard to interpret the population-dependent presence or absence of the selection . Rather , our results above suggest that the difference between populations is explained by the loss of information during the phasing of Asian samples without trio information . We next examined fastPHASEd-piCHM and PHASEd-JPT ( both are inferred Japanese haplotype datasets ) by iHS , and the results are shown in Figure 2B . As is evident from the figure , the peak at the MHC locus was obscured in both cases , supporting the explanation that the information on expansion of EHH at this locus was lost during the inference of haplotypes . Positive selection at the lactase locus has been documented in Caucasians and Africans but not in Asians [27] , [28] . In this case , we also observed strong signal enrichment at this gene region only for CEU and not for CHM ( Figure 2A , red arrow , and Table S2 ) . Thus , the results confirm that the population-specific positive selection observed at the lactase locus is real . Population-dependent positive selection was also examined by cross-population EHH ( XP-EHH ) [16] statistics using CHM vs . CEU and CHM vs . YRI datasets ( Tables S4 and S5 ) . As shown in the tables , many regions that had not been noted previously were newly identified . Some of these may have been missed in previous analyses [16] because of phasing problems . However , no definite conclusions could be reached , because the populations examined were not exactly the same . To more directly focus the lowering of information content in inferred haplotypes relative to definitive haplotypes , we introduce a new test; the cross-dataset integrated extended haplotype homozygosity ( XDiHH ) . This statistic is similar to XP-EHH [16] , except that the comparison is between two determinations of iHH: in the present case , inferred vs . definitive haplotypes . Here , iHH is an integration of EHH [17] against genetic distance , in both directions away from the core SNP , until at least two identical haplotypes are extended . XDiHH is expressed in simple terms by the following equation . Here , allele is either ancestral or derived , depending on whether it is computed with respect to the ancestral or derived core allele , while D1 and D2 are the two datasets for comparison . When the rate of EHH decay is the same between datasets , XDiHH is equal to 0 . Positive values indicate that D1 haplotypes are longer than D2 haplotypes , and negative values indicate the opposite . In principle , phasing errors can result in false elongation or false shortening of extended haplotype homozygosity . Therefore , XDiHH can be both positive and negative . Simulation experiments using a simple two-SNP system and assuming various local r2 values indicate that XDiHH tends to decrease in high LD regions , while it tends to increase in low LD regions ( see Figure S2 and its legend ) . We examined the XDiHH statistics between the haplotypes of CHMs and those obtained by phasing of pseudo-individualized CHMs ( piCHMs ) . Phasing of chromosome 6p was done by PHASE program ( Figure 3 ) . In this figure , we also plotted the switch error rate and recombination rate . The switch error rate was calculated as the proportion of neighboring heterozygous sites that are not correctly phased [29] , [30] . The recombination rate was averaged over values for all HapMap populations , calculated from the data of HapMap II . As shown in Figure 3 , we detected a negative signal at the MHC locus , corroborating the discrepancy of iHS results between fastPHASEd-piCHM and CHM . It can also be seen in the figure that the switch error rate at MHC locus was low compared to the surrounding regions , which was contrary to naïve expectations , since the deviation of XDiHH from zero must be caused by switch errors . However , as described above , a consistent reduction of XDiHH is observed only if the switch error occurs in region of low recombination rate ( i . e . , high r2 regions ) , as demonstrated in Figure S2 . Thus , the limited number of switch errors was sufficient to visibly reduce the XDiHH values on the MHC locus . Essentially the same results were obtained using the CEU dataset of HapMap Phase II ( Figure S3 ) . That is , we used the parental genotype data of CEU [1] and constructed the haplotypes without offspring data using PHASE ( that is , PHASEd-CEU ) for chromosome 6p . We then compared these haplotypes with those from the HapMap database ( that is , PHASEd+CEU ) using XDiHH statistics . A prominent negative peak of XDiHH was also detected at the MHC locus ( Figure S3 ) , supporting the conclusion drawn based on observations of CHMs and piCHMs ( Figure 3 ) , that the negative XDiHH peaks at the MHC locus were the result of phasing in the absence of trio information . Apart from MHC locus , the switch error rate of overall chromosome 6p region by PHASE was 0 . 0503 and 0 . 0633 , for CEU without children and piCHM , respectively . These values are in accordance with the previously reported switch error rate of PHASE ( 0 . 0543 ) , that was estimated by Marchini et al . [9] using randomly sampled genomic regions of HapMap CEU datasets removing the children . We next asked if there are regions of expanded EHH , other than the MHC locus , that escaped detection when analyzed using inferred haplotypes . Because of the limitations of our computational capacity , we inferred the phases of piCHMs using fastPHASE . The loci that gave significant negative XDiHH values are listed in Table S6 . Figure 4 shows the overall distribution of XDiHH when plotted against the recombination rate [31] . As is shown in the figure , XDiHH of derived alleles is generally more negative than that of ancestral alleles ( mean values −0 . 0315 vs . −0 . 0059 ) . This is consistent with the fact that the derived alleles arose by new mutations and are typically associated with longer haplotypes , and are therefore more sensitive to fragmentation by phasing errors than are ancestral alleles [17] , [32] . In addition , the majority of XDiHH values were in weakly positive areas , while clusters of extremely negative XDiHH values were observed in the low recombination rate range , where the MHC locus signal was observed . These are consistent with the expectation from the simulation results ( see legend to Figure S2 ) , and are consistent with the results of the local analysis of chromosome 6p by PHASE . We also examined the genome-wide view of the XDiHH profile for CEU and YRI , which was calculated using the haplotypes inferred by fastPHASE in the absence of trio information . Regions showing exceptionally negative XDiHH values in CEU and YRI are listed in Tables S7 and S8 . It is noteworthy that previously reported positively selected regions are significantly enriched in the regions of excessively negative XDiHH , possibly suggesting that regions under positive selection are more likely to have reduced XDiHH by phasing errors . We also noticed prominent negative values at 17q21 . 31 in CEU , but the corresponding region in YRI or CHM did not reveal such negative values . This site has been reported as a locus under selection specifically in Europeans , due to inversion polymorphism [33] . The recent increase in large-scale association studies is yielding a high volume of high-density genotype data , and presently available haplotype maps are likely to be greatly improved by incorporating these newly collected data [5] , [6] , [9] . The question here is whether and how the availability of definitive haplotype data can improve the accuracy of inference of haplotypes of unrelated individuals . To answer this question , we phased the data of the same sources by two procedures . One was phasing using only genotypes . In this procedure , combined datasets of JPT and/or CHB in HapMap II and all piCHMs ( 49 pairs ) were used ( simple phasing ) . Another was phasing genotypes with definitive haplotypes of CHMs serving as references ( referenced phasing ) . In this procedure , we used the same JPT/CHB and a subset of piCHM ( 5 pairs ) as genotypes to be phased , and the remaining CHM ( 88 haplotypes ) served as a reference haplotype dataset . Phasing accuracy was evaluated by comparing the inferred haplotypes of the five piCHMs ( that were common in all phasings ) with the true haplotypes , employing Beagle program ( ver . 3 . 0 . 1 ) [6] , which is capable of phasing genotypes in the presence or absence of reference ( or definitive ) haplotypes . As shown in Figure 5 , the global switch error rate observed in the simple phasing of 49 piCHMs was approximately 0 . 0878 ( evaluated using 5 piCHMs ) , which was somewhat higher than that expected from the reported value for Beagle [5] . This may be attributable to the difference in the nature of the data source ( pseudo-individuals of real haplotypes vs . phased genotypes generated by population simulation ) . On the other hand , phasing with reference yielded significantly better inference haplotypes , that is , a switch error of 0 . 0715 , when 88 reference haplotypes were used . When the number of unphased genotypes was increased ( 45 JPT and 45 CHB unphased genotypes were added ) , the accuracy of both phasings steadily improved , and the difference in the switch error levels between the two phasings gradually decreased . However , the phasing accuracy of Asian genotypes of unrelated individuals can be significantly improved if true haplotype data are included in the phasing procedures ( Beagle ) , at least at the sample sizes employed by the HapMap Project II ( 0 . 0608 vs . 0 . 0661 ) . Interestingly , the accuracy in the combined datasets of sub-structured populations ( CHM+CHB ) was significantly poorer than that of the non-structured populations ( CHM+JPT ) , indicating that in order to improve the phasing accuracy by increasing the sample size , it is best to use samples of the same population .
We determined high-resolution , genome-wide definitive haplotypes of East Asians ( Japanese ) by genotyping haploid materials ( CHMs ) using Affymetrix 500 K array sets , and assessed the influence of non-Mendelian inference of haplotypes to downstream analyses across the entire genome . We employed a new statistical method , XDiHH , to detect the difference in the extent of haplotype homozygosity between two haplotype datasets . The assessment was made after restricting the analysis to only shared SNPs and matching sample sizes , in order to achieve a strict comparison between the datasets . We examined XDiHH between the haplotypes obtained by phasing pseudo-individuals and the original true haplotypes . We found that , in some regions , notably at the MHC locus , the XDiHH values were considerably lowered , regardless of the datasets of the true haplotypes or the program used for phasing ( PHASE , fastPHASE , or Beagle ) , as long as the allele assignments were done in the absence of pedigree information . The accuracy of phasing genotypes of unrelated individuals can be increased by increasing the number of individuals [6] . We have shown that the switch error rate of phasing using a reference panel of 88 known haplotypes was 0 . 0715 , which is equivalent to the rate when 89 phase-unknown genotypes ( i . e . , genotypes for 178 halotypes ) contributed to the phasing ( Figure 5 ) . Thus , the phased haplotypes were approximately two-times more efficient than the unphased data in the phasing by Beagle at the sample size examined . It is interesting to know how the fixed number of reference panel can contribute to the improvement of phasing accuracy , when the number of unphased individuals to be phased is further increased . However , it is likely that increasing the number of the definitive haplotypes in the reference panel can significantly improve the phasing accuracy even at sample size ranges far exceeding those examined here . Recently , several next-generation technologies for DNA sequencing have become available or are being developed [34]–[37] , and projects on whole genome resequencing of large numbers of individuals in diverse populations have been proposed or are now underway . These projects will bring us knowledge of the entire genetic make-up of humans , including rare forms of genetic variation . However , further laborious tasks , such as discrimination of heterozygous sites and haplotype assembly , remain to be solved . These tasks are especially difficult in the case of rare variations . Perhaps using either haploid cells such as CHMs or trio samples may alleviate the burden of downstream analyses arising from phasing uncertainty . We are presently genotyping another set of SNPs ( Affymetrix Genome-Wide Human SNP Array 6 . 0 and Illumina Human1M-Duo DNA Analysis BeadChip ) using more than 100 CHM samples , which are expected to be useful as reference haplotypes as well as to develop or calibrate models for future accurate haplotype inference . The data generated here are in D-HaploDB [11] and are freely accessible via the internet at http://finch . gen . kyushu-u . ac . jp .
Informed consent was obtained from all donors of the CHM tissues . Use of these samples in the present work was approved by the Ethical Committee of Kyushu University . CHM samples were collected in a nationwide effort supported by the Japan Association of Obstetricians and Gynecologists . Both the female donors and their male partners were self-reported Japanese . Genomic DNA was extracted using a QIAamp DNA Mini Kit ( Qiagen ) . DNA of 100 CHMs confirmed to have no heterozygous sites by microsatellite analyses were genotyped using Affymetrix GeneChip 500 K arrays ( Santa Clara , California , United States; http://www . affymetrix . com ) . The physical coordinates of SNPs referenced to NCBI build36 of the human genome were obtained from the Affymetrix web site . The Dynamic Model at P = 0 . 33 was employed to call the alleles , and the concordance for the 50 SNPs that were common in the Nsp and Sty arrays was over 93 . 9% . The call rate of the worst CHM was 94 . 86% ( Table S9 ) , and the averaged sample call rate was 97 . 59% , while the averaged SNP call rate for all CHMs was 98 . 58% ( Table S10 ) . A subset of the CHMs examined in the present study ( 74 of 100 ) had been genotyped by Perlegen SNP arrays ( 280 K SNPs ) in our previous study [12] , and 56 , 883 SNPs were re-typed in the present work using Affymetrix 500 K Array Sets . Based on the results of these shared typings , the concordance rate was calculated to be 99 . 96% . In principle , all genotype calls of CHM samples should be homozygous , since the materials are haploid . However , a small fraction ( 0 . 67% ) of the calls was heterozygous . We believe that at least some , if not all , of these heterozygous calls were attributable to signals from the paralogous regions , and the rest were typing errors . In subsequent analyses , all heterozygous calls were considered to be “no calls , ” since the changes were unlikely to seriously affect any of the conclusions described here . The raw microarray CEL and CHP files have been submitted to Gene Expression Omnibus database at NCBI ( GEO; http://www . ncbi . nlm . nih . gov/geo ) under accession number GSE12713 . To make it more convenient to compare datasets using various statistical methods , we adjusted the number of samples to 86 , which was the maximum sample number available for JPTs after quality-checking ( see below ) . CHM065 was excluded from the dataset for reasons described in the Results section , and 86 samples were chosen from the remainder based on their high call rate ( >96 . 30% ) . We imputed the missing genotypes ( 3 . 08% , including those converted from heterozygote calls ) using fastPHASE v . 1 . 3 . 0 beta for Linux [8] , so that the analyzed dataset did not contain any missing values . This version of the program was kindly provided by P . Sheet [8] . We also created inferred haplotype datasets by phasing pseudo-individuals ( piCHMs ) that were created by randomly pairing the CHM haplotypes . Haplotypes of chromosome 6p was inferred by PHASE ( v2 . 1 ) using the default settings , after dividing the region into 25 non-overlapping segments , each with 500 SNPs . Neighboring segments of PHASEd-piCHM haplotypes were linked by referring to the phases of CHM at heterozygous sites that were nearest to the segment ends . This method of segment phasing is not as strict as that employed in PHASElink [9] . However , the present method does not overestimate the XDiHH values at the junction , because the phases of the two SNPs that bridge the segments are forced to be the same between PHASEd-piCHM and CHM , and the likelihood that the haplotypes branch at the junction is equal between the two datasets . Genome-wide haplotypings of piCHMs were carried out using either fastPHASE v . 1 . 3 . 0 beta for Linux [8] with ‘-KL6 -KU14 Ki2’ options or Beagle v . 2 . 1 . 3 [5] with nsamples = 25 option . For HapMap JPT samples , two individuals , NA18987 and NA18992 , whose estimated cryptic relatedness coefficients have been shown to be more than 1/32 [2] , were excluded from consideration in order to avoid inappropriate inflation of homozygosity statistics due to their unusually long-range haplotype homozygosity . PHASEd-JPT was a subset of the haplotype datasets in the HapMap database ( http://www . hapmap . org/downloads/phasing/2007-08_rel22/ ) , created by extracting the data for the SNPs that were genotyped using the Affymetrix 500 K Array . Forty-three CEU individuals ( parents ) with high call rates ( >99 . 08% ) were selected for analyses . The cryptic relatedness coefficients for these samples were less than 1/32 [2] . The individual IDs used in this study are listed with their call rates in Supplemental Table S11 . Haplotypes of chromosome 6p for all SNPs in HapMap Phase II using PHASE ( v2 . 1 ) in the absence of offspring information were inferred using the default settings , after dividing the region into 138 non-overlapping segments , each carrying 500 SNPs . Neighboring segments of PHASEd-CEU haplotypes were linked by referring to the HapMap data ( i . e . , phases of PHASEd+CEU ) as described above . Forty-three YRI individuals ( parents ) with high call rates ( >99 . 06% ) were selected for analyses . The cryptic relatedness coefficients for these samples were less than 1/32 [2] . The individual IDs used in this study are listed with their call rates in Table S11 . We applied principal component analysis ( PCA ) to detect hidden population stratification by using the SMARTPCA program of the EIGENSOFT version 2 . 0 package [19] , [20] . For this analysis , the CHM haplotypes were randomly paired to create pseudo-individuals . Figures 1 and S1 were generated using different pairs . PCA was applied using autosomal SNPs , for which more than 99% of the samples were successfully genotyped [1] . To obtain ancestral states for SNPs , we obtained alignment sequence files between human and chimpanzee sequences from the UCSC database ( http://hgdownload . cse . ucsc . edu/goldenPath/hg18/vsPanTro2/axtNet/ ) . We then assumed the chimpanzee allele at the appropriate position to be ancestral . For 97 . 9% of SNPs ( 479 , 864 of 489 , 992 ) , the ancestral states were successfully determined . No ancestral state was inferred for the remainder . The iHH was computed as described in previous work , using the SNPs with minor allele frequencies greater than 5% [17] . For both ancestral and derived chromosomes , we calculated EHH values between the core SNP and every other SNP and integrated with respect to genetic distance over the longest region for which at least two haplotypes were homozygous . The genetic distance data were downloaded from the HapMap database ( corrected on June 30 , 2008 ) [1] . These integrals were denoted as iHHAncestral or iHHDerived . If the region spanned by EHH reached gaps was longer than 500 kb or reached the chromosome ends , no further iHH scoring was reported for the core SNP . The unstandardized integrated haplotype score , which is ln ( iHHAncestral/iHHDerived ) , was then calculated for every SNP . The scores were standardized for the whole genome by normalizing them according to the frequency of the derived allele , in order to obtain iHS . These normalized scores have zero mean and unit variance . The rate is defined as sw/ ( n – 1 ) , where n denotes the number of heterozygous sites and sw is the number of switches between neighboring heterozygous sites needed to recover the original haploid sequence [29] , [30] . In computing these scores for each individual , we ignored sites where one or both alleles were missing . | Precise haplotype maps are preferred for the performance of a variety of genetic studies including identification of disease-associated loci and dissection of evolutionary mechanisms such as selection and recombination . For diploid organisms , the haplotype information appears as the genotypes when we obtain the information using widely used high-throughput techniques . The process of extracting haplotype information from genotypes is called phasing , which can be accurately done if the genotypes are from related individuals , such as parent–child trios , by considering the constraints imposed by the rules of Mendelian inheritance . For the genotype data without family information , phasing is done by one of the methods that are based on haplotype clustering , and the inferred haplotypes are known to be less accurate . Here , we experimentally determined genome-wide definitive haplotypes using a collection of Japanese complete hydatidiform moles ( CHM ) , each of which carries a genome derived from a single sperm . Using these resources , we asked if the definitive haplotype data can detect long-distance information that has been obscured when we rely solely on the haplotypes inferred by clustering . We also show that by introducing definitive haplotypes as references , inference of haplotypes of unrelated individuals is significantly improved . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/genomics",
"computational",
"biology/population",
"genetics",
"computational",
"biology/genomics",
"genetics",
"and",
"genomics/population",
"genetics"
] | 2009 | Evaluation of Haplotype Inference Using Definitive Haplotype Data Obtained from Complete Hydatidiform Moles, and Its Significance for the Analyses of Positively Selected Regions |
Bilirubin is the terminal metabolite in heme catabolism in mammals . After deposition into bile , bilirubin is released in large quantities into the mammalian gastrointestinal ( GI ) tract . We hypothesized that intestinal bilirubin may modulate the function of enteric bacteria . To test this hypothesis , we investigated the effect of bilirubin on two enteric pathogens; enterohemorrhagic E . coli ( EHEC ) , a Gram-negative that causes life-threatening intestinal infections , and E . faecalis , a Gram-positive human commensal bacterium known to be an opportunistic pathogen with broad-spectrum antibiotic resistance . We demonstrate that bilirubin can protect EHEC from exogenous and host-generated reactive oxygen species ( ROS ) through the absorption of free radicals . In contrast , E . faecalis was highly susceptible to bilirubin , which causes significant membrane disruption and uncoupling of respiratory metabolism in this bacterium . Interestingly , similar results were observed for other Gram-positive bacteria , including B . cereus and S . aureus . A model is proposed whereby bilirubin places distinct selective pressure on enteric bacteria , with Gram-negative bacteria being protected from ROS ( positive outcome ) and Gram-positive bacteria being susceptible to membrane disruption ( negative outcome ) . This work suggests bilirubin has differential but biologically relevant effects on bacteria and justifies additional efforts to determine the role of this neglected waste catabolite in disease processes , including animal models .
Heme is a critical co-factor in aerobic respiration and energy production , yet in excess is also highly toxic [1] . The turnover and degradation of heme is a protective pathway , terminating in the production of bilirubin [1] , [2] , [3] . Heme is broken down predominately in the spleen by heme oxygenase 1 and 2 into molecular iron , carbon monoxide , and a green pigment called biliverdin , which is further reduced by biliverdin reductase to yield bilirubin [4] . Lipophilic bilirubin is transported by serum via albumin to the liver and removed from circulation . Biotransformation of bilirubin from a lipophilic molecule to water-soluble or conjugated forms ( bilirubin mono- and di-glucuronide ) is facilitated by the 1A1 isoform of uridine 5′-diphosphate-glucuronosyltransferase ( UGT1A1 ) [4] . After conjugation with glucuronic acid , bilirubin is actively secreted across the canalicular membrane of hepatocytes by MRP2 and into bile [5] . Host and bacterially secreted β-glucuronidases , as well as non-enzymatic hydrolysis , lead to the deconjugation of bilirubin glucuronide after bile is released into the intestine , resulting in unconjugated bilirubin that can be found in the intestinal lumen of humans at submillimolar concentrations [6] , [7] . Approximately 300 mg of bilirubin are produced daily by healthy adults , the vast majority of which is excreted in feces [8] , [9] , [10] . Bilirubin affects mammalian systems in diverse ways . In a series of elegant experiments , Stocker et al . demonstrated bilirubin can scavenge peroxyl radicals and other reactive oxygen species ( ROS ) [2] , [10] . In vivo , bilirubin decreased injury from hyperoxia in neonatal Gunn rats [11] . In population studies , high baseline serum bilirubin concentrations correlated significantly with reduced rates of cancer-related mortality , suggesting bilirubin may alleviate the oxidative stress contributing to carcinogenesis [12] , [13] . In excessive quantities , however , bilirubin can enter the central nervous system , leading to encephalitis . This associated toxicity has been previously described in both neural and non-neural cell lines , leading to decreased neuronal cell viability and hemolysis of circulating erythrocytes [14] . However , none of these effects have any direct connection to the site in the body where bilirubin is heavily deposited , the GI tract . ROS in the GI tract is generated by three main sources , including infiltrating phagocytes , the intestinal epithelium , and the resident commensal bacteria . An essential factor to kill engulfed bacteria , infiltrating phagocytes utilize NADPH-dependent systems of generating superoxide called the NADPH oxidase ( Nox ) family of proteins , which oxidize NADPH to NADP+ in order to transfer electrons to molecular oxygen . The oxidation of NADPH to generate ROS , termed the “respiratory burst” mechanism , is also utilized by dual oxidase ( Duox ) family of proteins , which are expressed on the apical plasma membrane of GI epithelia cells and generate hydrogen peroxide into the luminal environment . The generation of hydrogen peroxide can further assist in the generation of antimicrobial molecules by interacting with lactoperoxidase to form hypothiocyanite ions [15] . In addition , intestinal commensal bacteria can produce ROS . A system for production of extracellular ROS has been described for the Gram-positive bacterium Enterococcus faecalis , which generates extracellular superoxide through the autoxidation of demethylmenaquinone , and is implicated in the damaging of colonic epithelial DNA [16] . Superoxide production by E . faecalis has been postulated to provide a competitive advantage for growth in the intestinal ecosystem; however , this has not been proven . Postulating that this somewhat neglected heme catabolite bilirubin can form functional interactions with intestinal bacteria , we characterized the effect of this pigment on the GI bacteria . We chose to utilize a Gram-negative intestinal pathogen ( enterohemorrhagic Escherichia coli ) which can cause severe diarrhea and life-threatening kidney damage , and a Gram-positive opportunistic pathogen ( E . faecalis ) known for their increasing broad-spectrum antibiotic resistance [17] , to study the effects of bilirubin on two enteric pathogens . Our data reveal , for the first time , that bilirubin can dramatically alter bacterial biology , enhancing on one hand pathogen survival while on the other causing a disruption of cell integrity . These results have wide-spread implications for those who study host-pathogen and host-commensal homeostasis in the human intestine .
We formulated the simple hypothesis that bile , because of its high concentration of bilirubin , may act as an antioxidant and neutralize free radicals . To test this hypothesis , cultures of E . coli serotype O157:H7 ( EHEC ) , an outbreak strain that can cause life-threatening intestinal infections , was supplemented with the quinone plumbagin . Membrane-associated quinones , like plumbagin , generate ROS by shuttling electrons from the electron transport system to molecular oxygen , thereby generating oxygen radicals such as superoxide , which can kill cells [18] . The addition of plumbagin to EHEC cultures increased the time to mid-log phase in a dose-dependent manner , suggesting plumbagin , presumably through the generation of ROS , severely inhibited bacterial growth ( Fig . 1A ) . Addition of bovine serum albumin ( BSA ) to EHEC cultures containing plumbagin decreased the time to mid-log phase in a dose-dependent manner ( Fig . 1B ) . No effect on the culture growth was observed with the addition of BSA in the absence of plumbagin . This is consistent with the notion that BSA , a known antioxidant , was mitigating the negative effects of plumbagin by protecting from ROS . When EHEC was grown in the presence of plumbagin and ox bile , the time to mid-log phase decreased in a dose-dependent manner when compared to the absence of bile , an effect dependent on the presence of the radical generator ( Fig . 1B ) . This effect was also observed for bile from other species , including human ( Fig . 1C ) . Interestingly , rabbit bile seemed to be the most effective , which is consistent with its nearly two-fold greater concentration ( approximately 600 uM ) of bilirubin when compared to bile from ox ( approximately 250 uM bilirubin in 100 mg/mL whole bile ) or human ( approximately 400 uM bilirubin , quantification conducted according to [19] ) . Collectively , this data suggests plumbagin-mediated toxicity can be alleviated by mammalian bile , possibly by the heme catabolite bilirubin . We hypothesized that the protection of EHEC from plumbagin could be due to the bile pigments bilirubin or biliverdin . To test this hypothesis , we grew EHEC in the presence or absence of plumbagin with or without bilirubin . Interestingly , EHEC cultures exposed to ROS supplemented with bilirubin showed a dose-dependent reduction in the time to mid-log phase compared to those in the absence of bilirubin ( Fig . 2A ) . Surprisingly , the addition of either biliverdin ( the metabolic precursor to bilirubin ) or bilirubin ditaurate ( a more soluble synthetic form of bilirubin and substitute for bilirubin glucuronide ) did not reduce the growth time of EHEC that was exposed to ROS , suggesting little to no antioxidant potential for these two bile pigments . Since biliverdin has a reduced capacity to neutralize free radicals ( based on its structure , biliverdin cannot donate its central hydrogen , leading to reduction of ROS , as easily as bilirubin – see Fig . S1 ) and bilirubin ditaurate ( which is soluble and contains the conjugated system similar to bilirubin , and thus retains antioxidant potential ) also lacks the ROS scavenging activity ( both do not protect EHEC from ROS ) , these data suggest unconjugated bilirubin protects by direct association with bacteria . In support of this finding , when we exposed EHEC to bilirubin and fractionated the cells , we found approximately 1/3 of the total bilirubin was recovered in the soluble lysate , whereas the remaining bilirubin was found in the insoluble pellet after clearance of the lysate ( Fig . S2 and data not shown ) . This data suggests that bilirubin associates with EHEC , likely through hydrophobic interactions with membranes , as has been previously observed for the interaction of bilirubin with eukaryotic cells [14] . Interestingly , expression of human biliverdin reductase ( BVR ) in EHEC , which should generate intracellular bilirubin when cultures are fed the precursor biliverdin , failed to rescue the bacteria from ROS generated by plumbagin ( Fig . S3 ) . This may suggest that bilirubin's antioxidant activity may occur at the bacterial surface , possibly the inner and/or outer membrane . To determine if the observed protection from ROS was strain dependent , the growth of the 2011 outbreak strain EAEC O104:H4 , which sickened over 3 , 000 people in Europe ( ∼50 deaths ) , was examined for growth when exposed to plumbagin and bilirubin . Indeed , EAEC was protected from ROS when bilirubin was added to the cultures ( Fig . 2B ) [20] . Similarly , E . coli BL21 , a laboratory strain that is not considered a pathogen , was protected from ROS when supplemented with bilirubin in a dose-dependent manner ( Fig . 2C ) . Further , E . coli isolated from a healthy donor's stool sample were likewise tested and found to be protected from ROS when cultured with bilirubin ( Fig . 2D ) . Each of the strains displayed varying susceptibilities towards the toxic effects of plumbagin , possibly from differential regulation of ROS neutralizing factors . Yet , in each experiment , the addition of bilirubin decreased the amount of toxicity generated by plumbagin , suggesting bilirubin could act as an antioxidant for each strain . Finally , to strengthen the notion that bilirubin was relieving oxidant stress generated by plumbagin , we repeated the above experiments with EHEC using a photosensitive ROS-generating agent called Rose Bengal , which generates superoxide from molecular oxygen under fluorescent lighting [21] . Growth of EHEC strain 86-24 was inhibited in a light-dependent manner with Rose Bengal present , yet growth was rescued when increasing amounts of bilirubin were supplemented into the cultures ( Fig . 2E ) . Collectively , these results strengthen the notion that bilirubin protects diverse strains of E . coli from exogenous ROS , possibly at the level of the bacterial membrane , and justify efforts to examine the utility of this effect in a more biologically relevant context . Macrophages kill bacteria after phagocytosis through the generation of superoxide by NADPH oxidase . This so-called “oxidative burst” is a major component of the host's native immunity against invading bacteria , including enteric pathogens in the intestine [15] . We hypothesized that bilirubin , due to its ability to protect EHEC from exogenous ROS ( Fig . 2 ) , may also mitigate ROS produced by macrophages during engulfment of EHEC . To test this hypothesis , we added EHEC strain 86-24 to J774A . 1 murine macrophages at an MOI of approximately 3 and followed bacterial survival by plating the colony forming units ( CFUs ) of EHEC over time on selective agar post osmotic lysis of macrophages . Macrophages equivalently internalized EHEC cultured with and without bilirubin ( P-value>0 . 05 ) ( Fig . 3A ) . Interestingly , EHEC treated with bilirubin showed several-fold higher levels of live bacteria than the untreated control , suggesting bilirubin reduced the rates of killing by J774A . 1 macrophages ( Fig . 3B ) . These data remained consistent between experiments when macrophages were primed or unprimed with LPS ( Fig . S4A and B ) . Consistent with data from Figure 2 , biliverdin failed to promote survival of EHEC within macrophages , even when supplemented into media during initial exposure ( Fig . 3B ) . Furthermore , the known lipophilic antioxidant vitamin E ( α-tocopherol ) led to an increased survival of EHEC within macrophages , although at a reduced level compared to bilirubin ( Fig . 3B ) [22] . These data strongly support the notion that bilirubin can act as a lipophilic antioxidant capable of protecting EHEC from exogenous and host ( macrophage ) -generated ROS . The finding that bilirubin could promote the survival of EHEC when exposed to ROS prompted us to assess if other bacteria , particularly Gram-positives , might also benefit by interacting with bilirubin . To test this hypothesis , we determined the effect of bile pigments on Enterococci , which are similar to E . coli in that they are part of the aerobic flora within the large intestine [23] . For this purpose , we used E . faecalis , an opportunistic enteric pathogen , and otherwise commensal organism , which has wide-spread antibiotic resistance [17] , [24] . Increasing the concentration of bilirubin led to a decrease in the numbers of E . faecalis , as judged by bacterial lawn formation on agar plates and quantified by densitometry ( Fig . 4A and B ) . Consistent with the trends already noted with EHEC , no effect was observed when the experiment was repeated with biliverdin or α-tocopherol , even at very high concentrations ( Fig . 4A and B ) . Furthermore , three other strains of E . faecalis with different origins of isolation ( feces , blood , or peritoneal fluid ) were tested against bilirubin by this method . Each strain responded similarly to OG1RF , with substantial decreases in viable cells as bilirubin increased in concentration ( Fig . 4C ) . These data suggest , in contrast to E . coli , that bilirubin exerts a toxic effect on E . faecalis , and which is seemingly independent of strain and origin . To provide a greater understanding into how bilirubin may be affecting E . faecalis , we monitored changes in colony forming units over time during planktonic growth while exposed to bilirubin . Exposure to bilirubin reduced the amount of viable bacteria by approximately 4 orders of magnitude within a single hour ( Fig . 4D , compare white triangles to orange diamonds ) , suggesting that bilirubin had a direct and immediate effect on cell viability and survival . Interestingly , given enough time , the remaining cells started to slowly grow and nearly reached the initial inoculum by 8 hours . To determine if this observation could be due to an intrinsic adaptation of E . faecalis to bilirubin as opposed to a change in the physical properties of the pigment , the experiment was repeated but with a secondary addition of bilirubin to growing cultures at 4 hours ( Fig . 4D , see arrows , yellow squares and red circles ) . E . faecalis that was previously exposed to bilirubin decreased in viable CFUs after the new infusion , suggesting the new population was not intrinsically resistant ( i . e . underwent a mutational event that conferred resistance ) but rather that the bilirubin was either titrated out or lost activity with time . This data further supports the idea that bilirubin contains an intrinsic ability to disrupt cell viability . Finally , to determine if this effect was specific to E . faecalis as opposed to a more general effect on bacterial cells with a single membrane , we determined the effect of bilirubin on two medically-significant Gram-positive pathogens , Bacillus cereus and Staphylococcus aureus . As observed in Table 1 , bilirubin also led to a dramatic decrease in cell viability of these two bacteria without affecting the control , E . coli . When taken together , these additional results suggest bilirubin is highly toxic to three Gram-positive bacterial species , including E . faecalis , and likely works through a direct , rapid , and physical mechanism of disruption . The finding that bilirubin was cytotoxic to E . faecalis and that this effect could be titrated out promoted us to hypothesize that unconjugated bilirubin was , by virtue of its lipophilic properties , intercalating into the bacterial membrane and causing disruption of membrane function . Bile salts , amphiphilic detergents capable of disaggregating lipids , are used for selection and enrichment of Gram-negative bacteria , through a similar mechanism [25] . Investigation into bile resistance suggests the outer membrane of Gram-negative bacteria slow the diffusion of bile salts into the inner membrane , and mechanisms of efflux and enzymatic alteration lead to higher levels of resistance when compared to Gram-positive bacteria , yet many of the mechanisms of bile salt resistance remain undetermined . Bile salts may increase membrane permeability or instability , potentially leading to lysis , in both bacteria and erythrocytes [25] . We therefore tested if bilirubin exposure to Gram-positive bacteria caused membrane instability by adding propidium iodide to EHEC and E . faecalis previously incubated with increasing concentrations of heme , biliverdin , bilirubin , and bilirubin ditaurate . Propidium iodide ( MW = 668 ) fluoresces intensely when associated with DNA , and its use in this context would indicate if the membrane became permeable after bile pigment exposure [26] . As expected , significant increases in fluorescent intensity occurred in heme-treated E . faecalis , as compared to solvent-treated samples ( Fig . 5A and B ) . It is thought that heme toxicity may be partially related to its ability to intercalate in membranes [1] . Interestingly , bilirubin-treated E . faecalis also showed an increase in propidium iodide-specific fluorescence compared to solvent-treated cells ( Fig . 5A and C ) . However , this was not true of E . faecalis treated with biliverdin or bilirubin ditaurate , which showed no increase in fluorescence . Both B . cereus and S . aureus also showed increases in propidium iodide fluorescence upon exposure to bilirubin , suggesting the membranes of Gram-positive bacteria are permeabilized by unconjugated bilirubin ( Fig . 5A and C ) . This hypothesis was further tested with the use of DiSC3 ( 5 ) , a probe that binds polarized membranes , yielding fluorescence [27] . The exposure of E . faecalis to the control conditions , solvent or α-tocopherol , yielded strong fluorescence , consistent with intact polarized membranes ( Fig . 5D , open and blue bars ) . In contrast , bilirubin treatment led to more than a two-fold decrease in the fluorescence , which was similar to treatment with the proton gradient uncoupler , CCCP , a known de-polarizing agent ( Fig . 5D , orange and yellow bars ) . Taken together , the decrease in fluorescence induced by bilirubin is consistent with a decrease in membrane polarization , suggesting bilirubin disrupts membrane physiology , presumably via membrane permeabilization . We hypothesized that bilirubin-mediated membrane instability would decrease the respiratory metabolism at the cell membrane , an essential cellular mechanism which can be measured with high sensitivity . This in turn could explain the substantial cytotoxic effect bilirubin has on E . faecalis , S . aureus , and B . cereus . Assays to quantify bacterial and eukaryotic respiratory metabolism have been developed and principally use artificial electron acceptors , such as tetrazolium salts and resazurin based compounds , which change spectral absorbance after cellular metabolic reduction . In this regard , we used resazurin to monitor the metabolic activity in E . faecalis since these bacteria are highly active at reducing artificial electron acceptors compared to other bacteria ( Fig . 6A ) [28] . Using this system , we observed that the reduction of resazurin is dependent on the electron transport system , as demonstrated by a decrease in resazurin reduction when E . faecalis is incubated with the succinate dehydrogenase inhibitor TTF ( Fig . 6B ) . In addition , the abundance of a carbon source is also essential for resazurin reduction , since a decrease in the reduction of resazurin was observed when sucrose is removed from the reaction buffer ( Fig . 6C ) . These controls aside , we hypothesized the reduction of resazurin would be increased by the extracellular production of superoxide , a characteristic of E . faecalis . Production of extracellular superoxide requires the autoxidation of molecular oxygen by demethylmenaquinone , a membrane associated quinone reduced by cellular dehydrogenases [16] . Indeed , cultures supplemented with both resazurin and superoxide dismutase ( SOD ) resulted in an equivalent reduction of resazurin within each culture ( Fig . 6D ) , thereby demonstrating that resazurin reduction must occur at the cellular membrane since resazurin was not reduced by extracellular superoxide . Hypothesizing that those concentrations of bilirubin that increase the membrane permeability of E . faecalis would similarly decrease the amount of resazurin reduced , we tested resazurin reduction in the presence of bilirubin and E . faecalis . As demonstrated in Fig . 6E , as little as 10 µM bilirubin led to a decrease in the reduction of resazurin , which incidentally is the same concentration of bilirubin which increases membrane permeability of E . faecalis ( Fig . 6C ) . Since these results suggest bilirubin inhibits cellular metabolism at concentrations that increase the cellular permeability of E . faecalis , we conclude that bilirubin-mediated cytotoxicity towards E . faecalis likely stems from the physical disruption of membrane permeability , thereby perturbing membrane polarity and aerobic respiration .
Earlier work to characterize bilirubin and other bile pigments , such as those conducted by Stocker et al . , demonstrated the antioxidant potential of these small molecules [1] , [10] . Lipid-soluble bilirubin effectively scavenged peroxyl radicals both in solution and within liposomes in vitro and more so than α-tocopherol , one of the best known antioxidants . Both bilirubin and biliverdin were capable of interacting synergistically with α-tocopherol , acting as chain-breaking antioxidants inhibiting lipid-peroxidation . In vivo , antioxidant models of bilirubin indicate biliverdin reductase , the enzyme responsible for creating bilirubin from biliverdin , interacts with bilirubin upon its oxidation to biliverdin , effectively removing a free radical in circulation and regenerating the cellular pool of bilirubin [29] . Since most species of bacteria ( excluding cyanobacteria ) lack biliverdin reductase activity , the cyclical antioxidant process of biliverdin and bilirubin and back would not likely occur within bacteria , though significant amounts of bilirubin are supplemented into the human intestine daily [30] , [31] . Concentrations of bilirubin within the human body can range dramatically depending on the physiologic location and individual's health . Healthy adult serum bilirubin concentrations are nearly 20 µM , yet can reach concentrations higher than 300 µM during hyperbilirubinemia [1] , [32] . Bilirubin is found in conjugated forms at millimolar concentrations in bile , yet deconjugation occurs through enzymatic and non-enzymatic processes , leading to unconjugated bilirubin in bile isolated from human gall-bladders ranging from 50 to 150 µM , well above the antioxidant potency concentrations observed in human serum [33] , [34] , [35] . Human bile contains β-glucuronidase activity , an enzyme responsible for deconjugating bilirubin from glucuronic acid [7] . The effects of bilirubin observed in this study are within the range of bilirubin concentrations observed at various locations in the body , especially the intestine . We sought to determine if bilirubin could function as an antioxidant for GI-associated bacteria . Little is known about the interaction of this metabolite with commensal or pathogenic bacteria and a bioactive role for bilirubin would be expected to influence bacterial homeostasis in the gut . Using characterized ROS-generating toxin , plumbagin , and a photosensitizing agent , Rose Bengal , with a well-known GI pathogen ( EHEC ) , we developed assays to test the antioxidant potential of compounds with pathogenic bacteria [18] , [21] . These assays allowed us to determine that bilirubin can be a functional antioxidant and protect EHEC from oxygen radicals . Since the bile pigments biliverdin , which has a very similar structure to bilirubin but does not have the capacity to accept free electrons , and bilirubin ditaurate , which can accept free electrons but is soluble , do not protect EHEC from ROS , the data suggest that the mechanism of protection is through the direct neutralization of free radicals after bilirubin associates with the bacterial cell . Previous work has demonstrated rapid cellular uptake of free bilirubin in eukaryotic cells [36] . Indeed , we have observed that all of the bilirubin given to these cells in culture either ends up associated with the bacteria ( ∼30% , being in the membrane , periplasm , and/or cytosol ) or the insoluble pellet after clarification of the lysate ( ∼70% , Fig . S2 ) . Furthermore , expression of BVR in EHEC that were fed biliverdin did not lead to protection from ROS . Although it is possible not enough bilirubin was made in this reaction , it is tempting to speculate that bilirubin's association with membranes after exogenous addition serves to protect from extracellular ROS generated at the bacterial surface by the action of plumbagin ( i . e . an “outside in” rather than an “inside out” mechanism of protection ) . Such a mechanism would protect from radical-mediated lipid peroxidation , a known and potent ROS-generated effect on cells , by possibly neutralizing ROS at membrane [10] . The finding that macrophages , which use oxygen radicals to destroy engulfed pathogens , are less efficient at killing bacteria pre-loaded with bilirubin indicates this effect may be biologically relevant , perhaps mitigating host ROS during the infectious process . Between the rescue of EHEC from ROS-induced growth impairment and the decreased phagocytic killing of EHEC within macrophages , the evidence that bilirubin can act as a potent antioxidant at physiologically relevant concentrations for EHEC appears strong . In addition to what appears to be a direct physical interaction with bacteria , bilirubin may also alter host gene expression . Indeed , a preliminary screen of proteomic changes that occur in EHEC upon exposure to bilirubin yielded several changes ( Fig . S5 and S6 , Table S1 ) . 2D-DIGE analysis identified more than 50 spots modified in abundance upon exposure to bilirubin . After identifying ten proteins found in higher or lower abundance ( though this could not be confirmed at the transcriptional level – Fig . S6 ) , we found four redox enzymes involved in cellular redox related metabolism . In a previous study by Shao et al . , Helicobacter pylori exposed to bile increased the abundance of several metabolically-related redox enzymes . Little investigation was reported as to the reason or mechanism behind the increased abundance other than the hypothesis that the enzymes could possibly regulate bacterial homeostasis after bile exposure [37] . Previous research supports toxic effects of bilirubin leading to uncoupling of oxidative phosphorylation , inhibition of hydrolytic enzymes , dehydrogenases , and enzymes involved in the electron transport system [1] , [38] . Possibly , several redox enzymes increase in abundance after bilirubin exposure to compensate for inhibitory effects , or alternatively , bilirubin may act as a signal to prime the bacterial cells to prepare for host-generated ROS . Given that the products of these genes showed an increase upon analysis of their proteins levels after exposure to bilirubin , but not at the transcriptional level , this data suggests the effects of bilirubin on these proteins is post-transcriptional . This might occur in a number of ways , including that bilirubin may enhance the rates of translation of these genes , or act to stabilize them once made . Interestingly , we observed bilirubin is toxic to the human commensal E . faecalis . At concentrations of bilirubin found to protect EHEC from ROS cytotoxicity , E . faecalis was readily killed . This was also true of at least two other distinct Gram-positive species . Interestingly , biliverdin did not cause a change in the growth of E . faecalis on agar , demonstrating specificity toward bilirubin activity . Furthermore , viability decreases dampened with time , but could be restored with fresh addition of bilirubin . Bilirubin increased the propidium iodide uptake by cells , and caused a disruption of the membrane potential . When taken together , this information led us to conclude that bilirubin decreases the viability of the assayed species of Gram-positive bacteria through direct intercalation into the bacterial plasma membrane , further disrupting essential cellular functions . This is similar to the effects of bile salts and sodium dodecyl sulfate on bacterial cells and argues bile contains several chemically distinct compounds that act on bacterial membrane structures [25] , [39] . Since this susceptibility to bilirubin is not observed with the four distinct strains of E . coli we tested , it is tempting to speculate that Gram-negative bacteria resist the membrane intercalating effects of bilirubin because they contain two membranes ( inner and outer ) , with the outer absorbing most of the molecule , sparing the inner from damage . In this regard , having an outmost layer of membrane that contains a ROS neutralizing molecule , can , under the certain conditions , actually provide a benefit . This model would support that since Gram-positive bacteria contain a single membrane , their susceptibility may be due to bilirubin disrupting this single structure , which serves the same functions as the two membranes of Gram-negative bacteria . In this model , structural differences in the surface of Gram-negative and Gram-positive bacteria may be the determinant as to whether bilirubin is beneficial or harmful . It is also possible that the structure or composition of the outer membrane in Gram-negative bacteria confers resistance to the intercalating effects of bilirubin , as compared to Gram-positive bacteria . For example , changes in the structure of the O-antigen of Salmonella changes the susceptibility of this bacterium to bile acids [40] . Future studies will be needed to address this exact mechanism . Most work on bilirubin has ignored its potential role in modulating intestinal biology , with research largely focused on bilirubin-associated toxicity and metabolism relating to neurological damage during severe hyperbilirubinemia [8] , [36] , [38] , [41] . Bilirubin at high concentrations can enter the central nervous system and cause acute encephalopathy [1] , [32] . Lipid-soluble bilirubin associates quite rapidly with exposed cells , which correlates with an increase in cell death [3] , [36] . Bilirubin may act as an uncoupler of oxidative phosphorylation and an inhibitor of respiration [38] , [42] , and also leads to an increase in the loss of cellular proteins [36] . Taken together , these data suggest bilirubin toxicity may be related to its ability to disrupt membrane integrity in higher organisms and data presented here with bacteria provide experimental support for these ideas [1] , [25] . Bile salts lead to the selection and enrichment of Gram-negative bacteria species , presumably through negative effects on surface membranes [25] . Likewise , heme toxicity may include membrane destabilization [1] . Our data suggests the plasma membranes of E . faecalis become permeabilized above concentrations of bilirubin as low as 10 µM , an observation which is shared with heme exposure . This was also true of B . cereus and S . aureus , which suggests bilirubin-induced membrane permeability may be a universal property common to all Gram-positive bacteria . What are the consequences of bilirubin's action on Gram-positive bacteria ? To correlate decreases in bacterial viability with increased membrane permeability , we investigated the changes in overall metabolic activity of E . faecalis when exposed to bilirubin . Metabolic activity was quantified using artificial electron acceptor resazurin [28] . As expected , exposure to bilirubin decreased the rate at which E . faecalis reduced resazurin at similar concentrations which previously increased permeability of membranes . If these results are extrapolated to other Gram-positives ( indeed , in this report we present consistent data that supports this extension ) , it is possible that bilirubin acts as an antagonist against some bacterial species in the diverse GI tract , a process that may influence the abundance and composition of the GI flora . Unlike bile salts which can be prominently reabsorbed during transit down the GI tract , unconjugated bilirubin has been shown to increase in comparison to gallbladder concentrations as one moves through the intestine [9] , [43] , [44] . β-glucuronidases are well known to be secreted with bile and facilitate the reverse reaction of conjugation performed by the liver to conjugate bilirubin to either one or two glucuronic acid molecules [7] . Though a physiologic function for bile-secreted β-glucuronidases is not well described , secreted enzymes may function to increase the concentration of unconjugated bilirubin during bile transit down the GI tract by utilizing the high concentration of conjugated bilirubins . Further , the concentration of β-glucuronidase can be increased by populations of coliform bacteria and to a much greater extent by Clostridia species , both of which actively secrete β-glucuronidases , implicating these bacteria in initial bilirubin breakdown within the GI tract [6] , [45] . Intestinal epithelia expression of UGT1A1 , the enzyme responsible for biotransformation of bilirubin by conjugating the molecule with glucuronic acid , decreases from the ileum to the cecum , suggesting a driving force toward conjugated bilirubin diminishes as intestinal contents approach the colon [46] . Thus , it is possible that bilirubin exerts differential effects on different communities as it traverses down the intestine , being modified along the way by native species . This information , in combination with data presented here , allows us to propose the hypothesis that bilirubin is simultaneously acting as an antioxidant and membrane-destabilizing agent within the GI tract . This proposed dual function may impose a selective pressure on resident and pathogenic bacteria , allowing bacteria capable of resisting the toxic effects of bilirubin to utilize the antioxidant properties for protection from ROS . In this study we demonstrated both activities of bilirubin with various intestinal-associated bacteria , suggesting the function of bilirubin does not stop until it passes from the GI tract . Future work will define bacterial responses towards bilirubin , determine their relationship to oxidative stress , and be correlated to possible responses observed in the intestinal tract . It will be necessary to test these ideas against a greater number of bacterial species and in relevant models of animal infection . Animal models of infection that center on intestinal bilirubin and allow one to probe these questions , although non-existent , will need to be developed to answer these questions . Such models may include the use of inhibitors or surgery to inhibit bile ( and thus bilirubin ) secretion into the intestine , or the use of knockout ( e . g Nox or Duox-deficient ) mice to further explore certain host components that generate ROS .
Bacterial strains used in this study included Escherichia coli serotype O157:H7 ( EHEC ) strain EDL933 ( ATCC# 700927 ) and 86-24 [47] , [48] , E . coli serotype O104:H4 2011 German outbreak ( EAEC ) ( generously provided by Dr . Alison Obrien , Uniformed Health Services ) , Enterococcus faecalis strain OG1RF ( ATCC# 47077 ) [24] , E . faecalis strain X33 ( ATCC# 27274 ) , E . faecalis strain UWH 1936 ( ATCC# 49533 ) , E . faecalis strain NJ-3 ( ATCC# 51299 ) , Staphylococcus aureus strain MW2 ( ATCC# BAA-1707 ) , and Bacillus cereus strain NRS 248 ( ATCC# 10987 ) . “Commensal” E . coli strains ( CN-5 and CN-7 ) were isolated from healthy donor stool samples and identified by selection on MacConkey agar plates , Gram-staining , and by VITEK analysis ( generously conducted by Dr . Audrey Wanger , University of Texas Health Science Center ) . E . coli strains were grown at 37°C in Luria-broth ( LB ) , while E . faecalis , S . aureus , and B . cereus were grown at 37°C in brain heart infusion broth ( BHI ) . Cultures were started from a single colony selected from LB or BHI agar plates using aseptic techniques . Kanamycin ( 25 ug/mL ) was supplemented to media for the selection of EHEC ( 86-24 ) . Whole ox bile was purchased from Fluka Analytical ( B3883-25G ) ; rabbit bile from Pel-Freez Biologicals ( 41206-1 ) ; hemin ( heme ) from Sigma Life Science ( H9039-100G ) ; biliverdin hydrochloride ( biliverdin ) from Frontier Scientific ( B655-9 ) ; unconjugated bilirubin ( bilirubin ) from Alfa Aesar ( A17522 ) ; bilirubin ditaurate from Frontier Scientific ( B850 ) ; plumbagin from Sigma Life Science ( P7262-100MG ) ; α-tocopherol from Alfa Aesar ( 10191-41-0 ) ; bovine serum albumin ( BSA ) from Fisher Scientific ( 9048-46-8 ) ; resazurin from Acros Organics ( 62758-13-8 ) ; MBTH from Research Organics ( 0133M ) ; kanamycin from EMD ( OmniPur 25389-94-0 ) ; ampicillin from USB corporation ( 69-52-3 ) ; spectinomycin from MP Biomedicals , LLC ( 158993 ) ; Rose Bengal Sodium Salt from Santa Cruz Biotechnology , Inc . ( sc-203757 ) ; 3 , 3′-Dipropylthiadicarbocyanine iodide ( DiSC3 ( 5 ) ) from Santa Cruz Biotechnology , Inc . ( sc-209690 ) . Human bile was generously provided by Dr . Mary Estes ( Baylor College of Medicine ) . Minimal media IDM was formulated as described in [49] . Reagents were solubilized as follows: ox bile , heme , biliverdin , bilirubin , bilirubin ditaurate , and BSA in 0 . 1 M NaOH; α-tocopherol in 100% ethanol , followed by a 1 to 100 dilution for stock concentrations in 0 . 1 M NaOH; Pbm and DiSC3 ( 5 ) in DMSO; and Rose Bengal Sodium Salt in sterile milliQ water . Expression of biliverdin reductase in EHEC 86-24 was conducted through the use of a pUC19 vector system . Briefly , bvr was amplified from ATCC #MGC-14706 using the forward primer ( GATCGATCGTCGACTATGAATGCAGAGCCCGAG ) and reverse primer ( GTAATGGGTACCTTATTATGCATAATCCGGAACATCATACGGATACTTCCTTGAACAGCAATATTTCTG ) followed by restriction enzyme digesting with both SalI and KpnI enzymes before ligating with SalI and KpnI digested pUC19 vector ( Invitrogen , #54357 ) . Ligation products were heat shock transformed into NEB 5 alpha competent E . coli ( NEB , #C2987I ) and positive colonies were selected by blue white screening . Positive colonies were further screened by PCR , restriction enzyme digest , and finally by DNA sequencing . Correctly cloned pUC19-BVR was transformed by electroporation into EHEC 86-24 and selected for on ampicillin LB-agar plates . EHEC 86-24 pUC19-BVR was determined to express BVR by anti-HA Western blot against the HA-tag incorporated into the C-terminus of BVR . EHEC ( EDL933 and 86-24 ) were inoculated into cultures at a final OD600 of 0 . 03 ( 1 cm pathlength , Beckman Coulter DU 800 ) with or without plumbagin , whole bile ( ox , rabbit , or human ) , biliverdin , bilirubin , bilirubin ditaurate , BSA , and α-tocopherol at concentrations indicated in the figures at 37°C with shaking . For experiments including EHEC 86-24 with pUC19 or pUC19-BVR , media was further supplemented with ampicillin ( 100 µg/mL , final concentration ) . Growth was monitored using a Bioscreen C machine , with wideband absorbance from 420–580 nm . Time to mid-log phase was calculated through linear interpolation of data points that were at mid-log densities . Monitoring bacterial quantities ( CFU/mL ) by optical density has been demonstrated in our laboratory as having a strong correlation ( data not shown ) . For cultures supplemented with Rose Bengal , bacteria prepared in the same fashion as above were supplemented with Rose Bengal ( final concentration 750 uM ) and grown in 1 mL cultures under fluorescent lighting for 12–14 hours . Bacterial growth density was quantified using a Tecan Infinite M200 Pro plate reader to measure the absorbance at 600 nm . For the experiments in Figure 4 , bacterial lawn formation was monitored by mixing mid-log phase bacteria with bilirubin or biliverdin in PBS at an OD600 value of 0 . 5 . Mixtures were then diluted 1∶1000 , of which 10 µL were spotted onto a LB agar plate . Plates were incubated at 37°C overnight and imaged using either a transilluminator ( W/M-26XV ) in a UVP BioSpectrum 810 Imaging System or UVP Benchtop Imaging System . Densitometry of agar plates was calculated using the freeware program ImageJ ( version 1 . 45s , Wayne Rasband , NIH , USA ) . Colony forming unit ( CFU ) determination was conducted by serial dilution of cultures and spotting on LB agar plates , grown overnight at 37°C . Murine macrophages ( J774A . 1 , ATCC# TIB67 ) were cultured at 37°C in 5% CO2 in RPMI-1640 with L-glutamine ( Lonza 12-702F; RPMI ) supplemented with 100 µg/mL spectinomycin and ampicillin and 10% heat-inactivated fetal bovine serum ( JR Scientific , Inc . ; 43635 ) . Intercellular amounts of bacteria were determined in a method similar to [50] . Briefly , EHEC 86-24 was cultured for 4 . 5 hours in the presence of biliverdin , bilirubin , or α-tocopherol ( all 250 µM ) , washed extensively with 1× PBS , diluted in RPMI , and added to J774A . 1 cultures at an MOI of 3 . Bacteria were incubated with macrophages for 30 minutes prior to removal and macrophages were washed with 1× PBS to remove unbound bacteria . Growth media containing antibiotics was added to washed macrophages to kill bacteria not phagocytosed . To lyse the macrophages , media was removed and cells washed with 1× PBS before addition of milliQ water . Samples were then serially diluted and plated to determine CFU per well . Mid-log phase bacteria cultured in LB were incubated with heme , biliverdin , bilirubin , and bilirubin ditaurate in milliQ water at an OD600 of 0 . 4 for 30 minutes at 37°C , pelleted by centrixfugation , and washed once with 1× PBS . Propidium iodide ( 10 µM , Invitrogen; B34954 ) was added to cells and incubated at 25°C for 10 minutes in the dark , followed by analysis at excitation and emission wavelengths of 535 and 625 nm , respectively . Similarly , bacteria prepared in by the same method were supplemented with DiSC3 ( 5 ) ( final concentration 1 uM ) and analyzed using 622 nm excitation and 670 nm emission wavelengths for fluorescent quantification . Fluorescent quantification was conducted using a Tecan Infinite M200 Pro fluorescent plate reader . Mid-log phase bacteria were diluted to an OD600 of 0 . 1 in minimal media or 1× PBS with 0 . 5% sucrose and supplemented with resazurin ( 50 µM ) and either heme ( 50 µM ) , biliverdin , bilirubin , bilirubin ditaurate , TTF , superoxide dismutase , and heat-inactivated superoxide dismutase . These cultures were incubated for 30 minutes to 2 hours at 37°C while shaking . After allotted time , cultures were spectrophotometrically measured using a Tecan Infinite M200 Pro for absorbance at 600 nm wavelength . As unreduced resazurin has a strong absorbance at 600 nm , a property which diminishes as the compound is reduced , measurements were taken at A600 and background was subtracted from cultures not supplemented with resazurin . Using reactions not containing reducing agents and those not containing resazurin , the amount of resazurin reduced in each culture could be calculated on a percentage basis . 2D-DIGE was used to investigate the proteomic response of E . coli to bilirubin and was performed by Applied Biomics , Inc . ( Hayward , CA ) . Mid-log phase E . coli str . 86-24 was exposed to bilirubin , biliverdin ( 250 µM ) , or a solvent control ( NaOH ) for 4 . 5 hours , bacteria were washed four times with ice cold PBS , and then flash frozen with dry ice and ethanol . Protein from samples was extracted using 2-D lysis buffer ( 7 M urea , 2 M thiourea , 4% 3- ( ( 3-cholamidopropyl ) -dimethylammonio ) -1-propanesulfonate ( CHAPS ) , 30 mM Tris-HCl , pH 8 . 8 ) and quantified . Equivalent amounts of samples were covalently labeled with CyDye and run on first dimension isoelectric focusing , followed by second dimension SDS-PAGE . Image analysis was conducted with DeCyder software . Prominently changed spots were chosen for analysis by mass spectrometry ( MALDI/TOF/TOF ) and received data was used in peptide fingerprinting for protein identification . Bacteria prepared in the same fashion were subjected to RNA isolation using the Qiagen RNEasy Mini kit ( Cat . no . 74104 ) . Quantitative reverse-transcription PCR was conducted using the Qiagen QuantiFast SYBR Green RT-PCR kit ( Cat . no . 204154 ) with an Applied Biosystems 7500 Real-time PCR system . Analysis of data was conducted without using an efficiency correction as each reaction was observed to have efficiencies of greater than 90% . Solvent and treated samples were compared using the students T-test to determine significance . Values were considered statistically different if the comparison of their groups yielded a p-value less than 0 . 05 . | Bilirubin is the terminal breakdown product of heme , which is deposited at high concentrations in the human intestine , where it can come into contact with host cells , the gastrointestinal ( GI ) microflora , and invading pathogens . Here , we report that bilirubin can act as a protectant for the Gram-negative bacterial pathogen E . coli O157:H7 , which causes severe hemorrhagic diarrhea and life-threatening kidney damage . Paradoxically , bilirubin is highly toxic towards another enteric opportunistic pathogen , the Gram-positive bacterium E . faecalis . Whereas the protection of E . coli stems from the neutralization of host reactive oxygen species , bilirubin's toxicity toward E . faecalis is rooted in its lipophilic properties , which drives the rapid association of bilirubin with bacteria , leading to disrupted cell membranes and concomitant death . These results suggest small molecule metabolites can modulate bacterial communities in the intestine , a finding that may have important implications for diseases caused by enteric bacteria and disrupted flora . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods",
"and",
"Materials"
] | [
"gram",
"negative",
"emerging",
"infectious",
"diseases",
"microbial",
"pathogens",
"host-pathogen",
"interaction",
"biology",
"microbiology",
"bacterial",
"pathogens",
"gram",
"positive"
] | 2013 | A Product of Heme Catabolism Modulates Bacterial Function and Survival |
The Leishmania developmental life cycle within its sand fly vector occurs exclusively in the lumen of the insect’s digestive tract in the presence of symbiotic bacteria . The composition of the gut microbiota and the factors that influence its composition are currently poorly understood . A set of factors , including the host and its environment , may influence this composition . It has been demonstrated that the insect gut microbiota influences the development of several human pathogens , such as Plasmodium falciparum . For sand flies and Leishmania , understanding the interactions between the parasite and the microbial environment of the vector midgut can provide new tools to control Leishmania transmission . The midguts of female Phlebotomus perniciosus from laboratory colonies or from the field were collected during the months of July , September and October 2011 and dissected . The midguts were analyzed by culture-dependent and culture-independent methods . A total of 441 and 115 cultivable isolates were assigned to 30 and 11 phylotypes from field-collected and colonized P . perniciosus , respectively . Analysis of monthly variations in microbiota composition shows a species diversity decline in October , which is to the end of the Leishmania infantum transmission period . In parallel , a compilation and a meta-analysis of all available data concerning the microbiota of two Psychodidae genera , namely Phlebotomus and Lutzomyia , was performed and compared to P . perniciosus , data obtained herein . This integrated analysis did not reveal any substantial divergences between Old and New world sand flies with regards to the midgut bacterial phyla and genera diversity . But clearly , most bacterial species ( >76% ) are sparsely distributed between Phlebotominae species . Our results pinpoint the need for a more exhaustive understanding of the bacterial richness and abundance at the species level in Phlebotominae sand flies in order to capture the role of midgut bacteria during Leishmania development and transmission . The occurrence of Bacillus subtilis in P . perniciosus and at least two other sand fly species studied so far suggests that this bacterial species is a potential candidate for paratransgenic or biolological approaches for the control of sand fly populations in order to prevent Leishmania transmission .
Sand flies are vectors of various pathogens , including arboviruses and bacteria , but are best known as the principal vectors of Leishmania , the etiological agent of leishmaniasis , a neglected tropical disease with clinical symptoms varying in form from cutaneous to visceral [1 , 2] . According to the most recent reports , leishmaniasis affects nearly 12 million people located in tropical , subtropical , and Mediterranean regions [3 , 4] with an estimated 350 million people at risk [5] . Among all vector-borne diseases , visceral leishmaniasis ( VL ) is the second leading cause of death after malaria , with an annual incidence of 500 , 000 cases and 60 , 000 deaths each year [3 , 4] . To date , no effective vaccine is available against leishmaniasis , and treatments mainly rely on chemotherapy using pentavalent drugs . Currently , the effectiveness of the treatment varies because of adverse side effects on patients and the emergence of parasite drug resistance [6 , 7] . Phlebotomus and Lutzomyia are the main sand fly genera involved in the transmission of Leishmania sp . in both the Old and the New World [2 , 8–10] . Sand flies become infected when they blood feed on an infected host . Ingested amastigote parasites undergo a complex developmental cycle within the sand fly and are limited to the midgut of the insect [11] . Thus , the midgut of the vector is the first point of contact between ingested parasites and the apical surface of the intestinal epithelial cells of the vector . Bacteria have been isolated from the midgut of P . papatasi , a vector of Leishmania major , the etiologic agent of zoonotic cutaneous leishmaniasis ( ZCL ) [12] , and studies have suggested a role for these bacteria in the immune response and homeostasis [12–15] . Female sand flies feed on blood for egg laying . In addition to blood , they take sugar meals derived from a number of different sources , including leaves , fruit , and aphid honeydew . Such food sources offer many opportunities to ingest microorganisms [16–18] . The microbiota found in sand fly guts could mirror their diets . In low- and middle-income countries , such as Tunisia , large vector eradication programs are challenging owing to limited resources . New approaches to control vector transmission of Leishmania infantum are of major interest . These programs are needed to control the transmission of L . infantum in Tunisia . Paratransgenesis has been suggested as a feasible strategy for controlling the transmission of pathogens by arthropod vectors . This approach consists of the use of genetically altered symbiotic bacteria that secrete effector molecules that kill the infectious agents . Since these bacteria should co-localize with the pathogen and be transmitted vertically to the next generation , they are introduced into vectors to block pathogen transmission [19–20] . This "Trojan-Horse" approach was initially developed to interfere with the transmission of Trypanosoma cruzi by its triatomine vector [19] . Among possible bacterial species that could be considered as candidates for the development of a paratransgenic approach , Bacillus pumilus and Bacillus flexus were identified as the most frequent cultivable bacteria identified in the midgut of P . papatasi field-collected from Tunisia , Turkey , and India [21] . In addition , Bacillus subtilis isolated from Phlebotomus argentipes is currently being considered as a possible candidate for paratransgenesis aimed at preventing Leishmania donovani transmission [22 , 23] . In North Africa , Phlebotomus perniciosus is the main vector of L . infantum , the etiologic agent of zoonotic visceral leishmaniasis ( ZVL ) [24] . We sought to develop a paratransgenic platform to control the transmission of L . infantum by P . perniciosus . Here , we assessed the richness of bacterial species of laboratory-reared and field-collected sand flies . We investigated the monthly variations of the bacterial diversity carried by sand flies in an endemic area of ZVL in Tunisia , during the period of Leishmania infantum transmission . We analyzed these new data within the context of previously published studies on the microbiota of sand flies .
Sand flies collection: Laboratory-reared P . perniciosus ( Tunisian strain ) was obtained from a colony maintained at the Vector Ecology Laboratory of Pasteur Institute of Tunis [25] . Phlebotomus perniciosus individuals were also collected in a sheep shelter in the village of Utique located in Northern Tunisia ( 37°08’N , 7°74’E ) , with the owner consent , by using CDC traps . Sand fly trapping was performed from dusk to dawn one night per month , from July to October 2011 . This period corresponds to the period of main activity of P . perniciosus in Tunisia [26] . Field-collected sand flies were brought alive to the laboratory . However , as it is difficult to determine the age of field-collected sand flies , we arbitrarily attribute the day of their sampling as the day one . All field-collected sand flies were dissected within three days after collection . Laboratory-reared sand flies were dissected three-to-seven days after their emergence . Prior to dissection , each sand fly was rinsed in 70% ethanol for 3 minutes , followed by three successive rinsings in sterile PBS . Sand flies were then dissected on ice under stereo-microscope , in order to remove the midgut for bacterial identification and the genitalia for morphological identification to species level [26 , 27] . Only P . perniciosus females were used . Gut dissection: Each sand fly gut was individually placed in 1 . 5 ml microcentrifuge tubes containing 200 μl of sterile PBS ( pH 7 . 3 ) , homogenized with a disposable pestle , and diluted from 10−1 to 10−10 in 200 μl PBS . Each homogenate was plated onto individual 1 . 5% agar plates with TSA ( Trypticase Soy Agar ) , PCA ( Plate Count Agar ) , YMA ( Yeast Mannitol Agar ) or Luedemann medium and incubated at 30°C for 2 to 4 days in aerobic conditions . Individual colonies were selected and used for further identification . Chromosomal DNA extraction was performed as previously described [28] . After overnight incubation at 30°C in TSA , PCA , Luedemann or YMA medium , colonies were suspended in 500 μl of TE buffer ( 10 mM Tris-HCl , 0 . 1 mM EDTA , pH 8 ) to which 20 μl of lysozyme ( 35 mg/ml ) was added and incubated at 37°C for 30 min . Then , 40 μl of sodium dodecyl sulfate ( SDS 10% ) and 5 μl of freshly prepared proteinase K ( 10 mg/ml ) were added , and the solution was incubated at 30°C for 30 min . The solution was homogenized after the addition of 100 μl of 5 M NaCl and 80 μl of CTAB/NaCl ( 10%/0 . 7 M ) and incubated at 65°C for 10 min . DNA was purified by the addition of phenol-chloroform-isoamyl alcohol ( 25:24:1 , pH 8 . 0 ) , followed by chloroform-isoamyl alcohol ( 24:1 ) and then precipitated by the addition of 0 . 6 volumes of isopropanol . DNA pellets were washed with 200 μl of 70% ethanol and dried at 37°C before being resuspended in TE buffer ( 10 mM Tris-HCl , 0 . 1 mM EDTA , pH 8 ) and stored at -20°C . Total DNA extraction for the Denaturing Gradient Gel Electrophoresis ( DGGE ) analysis was conducted on whole midguts dissected from sand flies using the same total DNA extraction protocol described above [28] . Fig 1 summarizes the procedure used for the isolation and identification of bacterial species . A total of 180 field-collected and 35 colonized P . perniciosus females were processed . From field-collected sand flies , 135 guts were used for culture-dependent identification and 45 guts were analyzed by DGGE , a culture-independent method . The 35 samples from colonized P . perniciosus were processed only for culture-dependent identification . The length and sequences polymorphisms of the Intergenic Transcribed Spacers ( ITS ) , located between the 16S and 23S rRNA , is quite often due to the presence of tRNA genes . PCR amplification of the 16S-23S intergenic transcribed spacer regions between the rRNA genes ( ITS ) was performed for screening the bacterial phylotype diversity [29–31] . The universal primers , ITSF ( 5’-GTCGTAACAAGGTAGCCGTA-3’ ) and ITSR ( 5’-CAAGGCATCCACCGT-3’ ) , are complementary to nucleotide ( nt ) positions 1423–1443 of the 16S rDNA and nt positions 38–23 of the 23S rDNA of Escherichia coli , respectively [30] . Each reaction tube contains 1X PCR buffer ( Invitrogen ) , 2 mM MgCl2 , 0 . 2 mM deoxynucleoside triphosphate mix , 0 . 1 μM of each primer , 0 . 5 U of Taq polymerase ( Invitrogen ) and 400 ng of DNA extracted from single colonies . The total volume was adjusted to 25 μl . Amplification parameters were as follows: initial denaturation at 94°C for 5 min , followed by 35 cycles at 94°C for 30 s , 50°C for 30 s , 72°C for 45 s , with a final extension step of 10 min at 72°C , using an ABS2720 thermocycler . Amplification of the 16S rDNA gene was carried out with universal primers SD-Bact-0008-a-S-20 and S-D-Bact-1495-a-S-20 [32] . Each reaction tube contained 1x PCR buffer ( Invitrogen ) , 0 . 5 μM of each primer , 2 . 5 mM MgCl2 , 200 ng of purified DNA , 0 . 2 mM dNTPs and 0 . 3 units of Taq polymerase ( Invitrogen ) and the total volume was adjusted to 25 μl . Samples were amplified according to the following cycle: an initial denaturation step at 94°C for 10 min , followed by 35 cycles at 94°C for 1 min , 55°C for 1 min , 72°C for 1 min and a final extension step of 10 min at 72°C , using an ABS2720 thermocycler . PCR amplicons were then purified using the QIAquick PCR Purification Kit ( Qiagen ) and sequenced . Amplification of the V3-V5 region of the 16S rDNA: PCR amplification targeting the 16S rDNA genes was performed using the universal primers specific to the bacterial domain: 907r ( 5’-CCGTCAATTCCTTTGATGTTT-3’ ) and 357f ( 5’-TACGGGAGGCAGCAG-3’ ) [33] . A 40-bp GC-clamp was added to primer 357f to avoid complete denaturation of the DNA and allow the separation of DNA strands during migration in denaturing conditions [34–36] . Each reaction tube contained 1x PCR buffer ( Invitrogen ) , 2 . 5 mM MgCl2 , 0 . 12 mM dNTPs , 0 . 3 mM of each primer , 1 U of Taq DNA polymerase ( Invitrogen ) and 50 ng of DNA in a final volume of 50 μl . Amplification parameters were as follows: an initial denaturation step at 94°C for 4 min , 10 cycles at 94°C for 30 s , 61°C for 1 min and 72°C for 1 min , followed by 20 cycles at 94°C for 30 s , 56°C for 1 min and 72°C for 1 min . At the end of these cycles , a final extension step was performed at 72°C for 10 min . DGGE analysis: PCR products were run on a 7% polyacrylamide gel in a 40%–60% denaturing gradient of urea and formamide for 16S rDNA analysis . DGGE was performed using a BioRad DCode Universal Mutation Detection System at 100 V at 59°C for 17 hr , in 1 . 0 × TAE buffer ( 20 mmol/L Tris , 10 mmol/L acetate , 1 mmol/L EDTA pH 7 . 4 ) . After electrophoresis , gels were stained for 30 min with ethidium bromide . Identification of the DGGE Bands: Excised bands of DGGE gels were washed twice with 1 mL sterilized distilled water in a 1 . 5-mL tube . A portion of the gel piece ( < 1 mm3 ) was used as the direct template for PCR to recover DNA fragments . Amplification conditions for the V3-V5 region were as follows: an initial denaturation step at 94°C for 4 min followed by 35 cycles at 94°C for 30 s , 56°C for 1 min and 72°C for 1 min and a final extension step at 72°C for 10 min . Primers were identical to those described above except that the forward primer had no GC-clamp attached . The amplified products were purified with the QIAquick PCR Purification Kit ( Qiagen ) and then sequenced . The 16S rDNA sequencing was carried out using the BigDye Terminator v3 . 1 Cycle sequencing Kit and the ABI 3130 sequence analyzer . The partial 16S rRNA gene sequences were compared with sequences available in the ribosomal database , release 11 . 4 . Isolates were assigned at the species level on the basis of the 16S rRNA gene sequence similarity of the available sequences in the ribosomal database , measured by using the Seqmatch tool of RDP [37] ( https://rdp . cme . msu . edu/ ) . In addition , the partial 16S rDNA sequences were submitted to the BLASTn server of NCBI , using the 16S ribosomal RNA database ( Bacteria and Archea ) ( http://blast . ncbi . nlm . nih . gov/Blast . cgi ) . The nucleotide similarity thresholds of the 16S rDNA sequences with the nearest neighbor were: ≥ 95% and 97 . 5% [38] applied at the genus and species levels , respectively . All the analyses were conducted with the R-vegan package , v . 2 . 0–10 [39] . α-diversity was calculated using Shannon’s and Simpson’s diversity indices . Correspondance analysis ( CA analysis ) on the monthly data was carried out with the FactomineR package ( https://cran . r-project . org/web/packages/FactoMineR/ ) using the R language ( http://www . R-project . org ) . All the published data concerning bacterial species identification associated with Phlebotomus and Lutzomyia species ( the only two genera for which we have data ) were compiled and analyzed . Studies describing the identification of the midgut bacteria at the family , class or phylum level were not considered . To assess bacterial richness associated with the adult sand fly , data were collected without taking into account the method of bacterial isolation ( culture-dependent vs culture-independent ) and identification ( DNA sequencing of 16S rDNA , bacteriology ) . The overall dataset used in our analyses included ten Phlebotominae ( L . cruzi [40] , L . longipalpis [41 , 42] , L . evansi [43] , P . argentipes [22] , P . duboscqi [44] , P . halepensis [45] , P . papatasi [21 , 45–48] , P . sergenti [45] , P . perfiliewi [45] , P . chinensis [49] and P . perniciosus ) and their associated microbiota for the present study . Bacterial richness is visualized through network analysis using Cytoscape ( http://www . cytoscape . org/ ) [50] . To achieve this goal , data were extracted from our own database ( focused on Phlebotominae ) as CSV files , containing vertices or nodes ( representing hosts and bacteria ) and edges ( representing links ) . These files were loaded into Cytoscape v 3 . 4 . 0 , a tool specializing in graphical representation . This graph was modified to keep only one edge between host and bacteria . Bacterial nodes were colored to show their degrees of interaction with hosts .
First , we determined the number of Colony Forming Units ( CFU ) of each individual midgut using PCA medium; they ranged from 5 to 121 per individual sand fly midgut . A total of 441 and 115 independent colonies were obtained from field-collected and colonized sand flies , respectively ( Fig 1 ) . Examples of the different types of bacterial colonies are shown in Fig 2A . ITS-PCR analysis was used to dereplicate strain diversity among the 556 colonies . Each ITS profile is composed of one to five reproducible bands that display apparent molecular weights ranging from 50 to approximately 1 , 500 bp; these bands represent a phylotype ( Fig 2B ) . Among the 441 colonies isolated from field-collected P . perniciosus , 25 distinct ITS-PCR profiles were identified ( Fig 2B see * ) . Of a total of 115 independent colonies from colonized P . perniciosus , only 6 independent phylotypes were identified ( Fig 2B ) . When possible , the 16S rDNA locus of a sample representative of each ITS-PCR profile was further amplified to attempt identification at the genus and species levels . All the bacterial colonies isolated from sand fly midguts belong to three phyla: Firmicutes , Actinobacteria , and Proteobacteria . For the field-collected P . perniciosus , the calculated midgut bacterial composition was: Firmicutes ( 53 . 5% ) , Actinobacteria ( 15 . 2% ) and Proteobacteria ( 31 . 3% ) . For laboratory-reared sand flies , the midgut bacterial composition was Firmicutes ( 66 . 7% ) and Proteobacteria ( 33 . 3% ) . We did not isolate bacteria belonging to the Actinobacteria phylum from the midgut of laboratory-reared sand flies . Nevertheless , only 35 females were processed and bacterial colonies were isolated solely using the PCA medium , which might have influenced the output of our analysis . The results of isolating bacterial species from the midguts of field-collected and lab-reared P . perniciosus , performed in a culture dependent manner , are shown in Table 1 . Of the six bacterial species identified in laboratory-reared sand flies ( Table 1 ) , three are also found in the midgut of field-collected sand flies ( Stenotrophomonas maltophilia , Bacillus sp . , Lysinibacillus sp . ) ( Table 1 ) . We isolated Veillonella sp . and Burkholderia fungorum only from the laboratory-reared sand flies ( Table 1 ) . Overall , the bacterial richness recorded in field-collected sand flies , at the species level , seems to be more important than in laboratory-reared flies , even if the total number of lab-reared flies studied is small . To further characterize the bacterial richness in field-collected sand flies , a culture-independent method ( DGGE ) was performed on the 45 dissected midguts ( Fig 1 ) . Despite variation in the number and intensity of the bands detected , the observed DGGE profile is composed of at least 12 distinguishable bands . Among these bands , six were successfully sequenced . In addition to bacteria already identified using culture-dependent methods , like Enterococcus sp . ( Accession N° KY303721 and KY303722 ) , we also identified Wolbachia sp . and Ehrlichia sp . ( Fig 3 ) . BLASTing the sequence from the DG5 band ( 459 bp Accession N° KY303723 ) indicated an overall similarity of 99% with the Pel strain of Wolbachia , isolated from Culex quinquefasciatus ( NR-074127 . 1 ) . The same query on the RDP database disclosed 98% similarity with Wolbachia inokumae DQ402518 , which was already found in field collected P . perniciosus from Marseille , France [51] . A search in the RDP database with the sequence obtained from the DG1 band ( 718 bp , Accession N° KY322518 ) produced hits with various species of Ehrlichia , including 96% similarity with Ehrlichia canis-M73226 . A similarity of 96% with Ehrlichia ewingii ( NR-044747 ) was found when BLAST analysis was performed on the 718-bp DNA fragment ( Fig 3 ) . To our knowledge , this is the first report of the presence of Ehrlichia sp . DNA in sand fly midguts . A meta-analysis was conducted to assess the bacterial species diversity of Phlebotomus and Lutzomyia microbiota . This analysis included previously published studies concerning adults of seven phlebotomine sand fly species ( P . argentipes , P . chinensis , P . duboscqi , P . halepensis , P . sergenti , P . papatasi , P . perfiliewi ) our study reported on P . perniciosus and previously published data reported on three Lutzomyia species ( L . cruzi , L . evansi , L . longipalpis ) [22 , 40–49] . Owing to the small number of studies conducted on the microbiota of Phlebotominae and the lack of information about sex in several cases , we chose to not take into account the genera of the specimen in order to highlight trends . This analysis shows that most bacteria identified from Old World sand fly species belong to the Firmicutes phylum , 39 , 8% ( Fig 4A left panel ) ( 41–42% for our study on P . perniciosus ) and the Proteobacteria phylum , 46 , 8% ( Fig 4A right panel ) ( 37% for our study on P . perniciosus ) . Bacteria of the Bacteroides genus are not recorded in the present study and represent only 0 . 5% calculated from the pooled published data ( i . e . , Meta-set ) . Bacteria of the Actinobacteria phylum account for 11 . 9% of the Meta-set ( 20% , in our study on P . perniciosus ) . In Lutzomyia sp . , more than 57% of bacteria currently characterized , belong to the Proteobacteria phylum ( Gram-negative bacteria ) , Firmicutes representing 23 . 9% and Actinobacteria 5 . 6% . Bacteria of the Bacteroidetes phylum account for approximately 6% of the species in Lutzomyia but only 0 . 5% in Old World sand fly species ( Fig 4A ) . Nevertheless , we did not notice significant differences in Bacterial phylum composition between Old World and New World sand flies ( chi-squared = 5 . 8226 , df = 2 , p-value = 0 . 0544 ) ( Fig 4A ) . Within the Proteobacteria phylum , compared with the alpha- , beta- and deltaproteobacteria identified , gammaproteobacteria are by far the most frequently found bacterial class in Lutzomyia and Phlebotomus species ( Fig 4B right panel ) . Within the Firmicutes phylum , a higher number of classes is observed in Lutzomyia , with bacteria belonging to Negativicutes , Bacilli , and Clostridia . The Bacilli class is almost the sole representative of Firmicutes class in the Old World sand fly species ( Fig 4B left panel ) . In the Gammaproteobacteria class , bacterial species of the Enterobacteriaceae family are the most represented ( more than 60% so far isolated ) in both the Old and New World sand fly species , followed by bacteria belonging to the Pseudomonadaceae and Moraxellaceae families ( less than 20% ) and Xanthomonadaceae , with less than 10% ( Fig 4C right and left panel ) . Bacteria of the Coxiellaceae family have only been isolated from Old World sand fly species . Our meta-analysis shows that bacteria of the Serratia genus has been identified in almost all Old World and New World sand fly species so far studied , but Serratia marcescens was characterized only in P . duboscqi . Bacteria of the Enterobacter genus are found in five of the eleven sand fly species studied . Enterobacter cloacae and Enterobacter aerogenes were recorded in three sand fly species , while Enterobacter gergoviae and Enterobacter ludwigii were found in two sand fly species . The most frequently isolated bacteria in sand flies are Stenotrophomonas maltophilia ( Pseudomonadaceae ) , followed by Escherichia coli ( Enterobacteriaceae ) , Klebsiella ozaenae ( Enterobacteriaceae ) , and Staphylococcus epidermidis ( Staphylococcaceae ) . Bacillus subtilis ( Bacillaceae ) and Acinetobacter baumannii ( Moraxellaceae ) were identified in three of the eleven sand fly species currently studied ( Fig 5 ) . Despite that neither statistical nor bioinformatics analysis were performed to test the existence of biological patterns between sand fly species and their corresponding microbiota , the network representation displayed in Fig 6 suggests some relationships between the eleven studied New World and Old World sand fly species and the bacteria inhabiting their guts . As an example , the Bacillus genus is found in almost all Old World sand fly species . Bacillus subtilis was isolated from P . halepensis , P . papatasi and P . perniciosus . Bacillus megaterium was isolated from P . papatasi and P . argentipes . Bacillus oleronius , Bacillus brevis , Bacillus endophyticus , Bacillus pumilus , Bacillus circulans , Bacillus mojavensis , Bacillus firmus , Bacillus licheniformis , Bacillus vallismortis , Bacillus cereus , Bacillus amyloliquefasciens , Bacillus altitudinus and Bacillus flexus were isolated only from P . papatasi . Bacillus closei and Bacillus mycoïdes were isolated only from P . argentipes . Bacillus oleronius , Bacillus galactosidilyticus , and Bacillus casamensis were isolated only from P . perniciosus ( Fig 6 ) . Bacillus thuringiensis is the only species of the Bacillus genus that was isolated from L . evansi and P . chinensis , two sand fly species belonging to the New World and Old World , respectively ( Fig 6 ) . Nevertheless , in the Meta-Set no significant differences in the microbiota composition at the genus level was observed as demonstrated in the Fig 7 that depicts the Shannon ( left ) and Simpson ( right ) indices of diversity for Old World ( i . e Phlebotomus ) and New World sand flies ( Lutzomyia ) . Fig 8A depicts the monthly proportions of each previously characterized bacterial species . To perform this analysis , bacterial species identification was linked to each individual phylotype recorded by PCR-ITS analysis . Then , the number of colonies harboring the same ITS-PCR profile was determined and their monthly proportion calculated . In July , among the nine bacterial species identified in midguts of wild P . perniciosus , only bacteria belonging to two phyla , Proteobacteria and Firmicutes , were found . In September , the bacteria belonged to the Actinobacteria , Proteobacteria and Firmicutes phyla . In October , only bacteria belonging to Firmicutes were isolated ( Fig 8A ) . Shannon and Simpson indices of diversity confirmed a lower diversity in October ( Fig 8B ) . Correspondence analysis performed on data from monthly dynamics highlights the contrasting situation between October and September/July , depicted on the first axis . October is characterized by B . oleronus and unidentified Bacillus and Lysinibacillus species . The differences between July and September/October is depicted by the second axis of the correspondance analysis ( Fig 8C ) . This analysis further reinforces our observation of a monthly evolution of the microbiota within the sand fly gut .
Currently , there are considerable efforts to study arthropod gut microbiota , especially those of medically important vectors . The microbiota is considered in the context of possible extended phenotypes conferred on the insect hosts that allow niche diversification and rapid evolution [52] . As early as 1929 , Adler and Theodor [53] suggested that the presence of microorganisms in the guts of sand flies might impact the development of the parasite Leishmania . In the mid-80s , the team of Shlein and collaborators [12] observed a large number of “germ ( bacteria ) contaminations” in guts of wild-caught female P . papatasi . However , the composition of the sand fly’s gut microbiota was studied much later by Dillon et al . [54] . Ochrobactrum sp . was the first bacterium to be isolated from the midguts of P . duboscqi , a proven vector of L . major in Sub-Saharan Africa [44] , and from other sand fly species [40] , including laboratory-reared Lutzomyia longipalpis [55] and New World L . intermedia [56] . This bacterium , probably ingested by larva , passes to nymphs and up to the adults through transstadial transmission [44] . Recently , several publications were dedicated to the study of the microbial composition associated with the digestive tract of sand flies . Only a few studies concerning biotic and abiotic factors influencing the composition of the bacterial community of the midgut of sand flies were performed . This study brings additional evidence on the microbiota composition in the midgut of P . perniciosus . Our results suggest that lab-reared P . perniciosus display a lower bacterial richness in their midgut than in field-collected sand flies . This difference is likely due in part to the type of food diet ingested by larvae and adults during rearing . In the laboratory , P . perniciosus larvae are fed sterile chaw ( 50% rabbit food plus 50% rabbit feces ) . After emergence , glucose is the main source of carbohydrates for adults [25] . Under natural conditions , larvae , as well as adult P . perniciosus , have a wide variety of diet including various sources of blood meals [18 , 57] . Therefore , the nature of the feeding regimen leads to a striking contrast between field-collected and laboratory-reared sand flies , which might explain the lower bacterial richness observed in colonized sand flies . Among the bacterial genera found associated with P . perniciosus midgut , we identified isolates belonging to the Burkholderia genus and Stenotrophomonas maltophilia , an aerobic non-fermentative and a Gram-negative bacterium . We also identified bacterial species commonly found in the digestive tract of humans or other mammals , but which have not yet been described in the midguts of sand flies , like Veillonella sp . In addition Sporosarcina koreensis , Rhizobium pusense and Nocardia ( a rare endophyte bacterium ) have never been found in association with the sand fly gut . The richness of sand fly-associated bacteria , illustrated by the meta-analysis , point to some interesting outcomes . In Lutzomyia sp . , more than 57% of identified bacteria belong to the Proteobacteria phylum ( Gram-negative bacteria ) , whereas for Old World sand fly species , including P . perniciosus , Proteobacteria ( 47% ) and Firmicutes ( 40% ) are preponderant . Such a difference in the gut microbiota composition might be due to a number of factors , including the long divergence of evolution between the two subgenera [2]; some new studies are required to assess this observation . Another surprising finding is the high richness of Bacillus species found in Old World sand flies , in which the majority of these bacteria are host specific ( Fig 6 ) . Stenotrophomonas maltophilia , that has emerged as an important opportunistic pathogen [58] was found to inhabit the gut of most of the sand fly species so far studied . This bacterial species is a common microorganism found in aqueous habitats , plant rhizosphere , animal food and water sources . Thus , delineating the origin of the colonization of midguts by S . maltophilia and evaluating its role , if any , in the sand fly biology and physiology are of major importance . Our results have , for the first time , disclosed monthly variation in the diversity of the sand fly’s gut microbiota , during the period of transmission of L . infantum . In fact , it appears that the richness of the gut microbiota is related to sand fly seasonal activity . This diversity could reflect the environmental conditions , such as temperature and humidity , but it may also be linked to variations in plant cover , such as flower blooming . At the beginning of the sand fly season ( July ) , Ochrobactrum sp . and Serratia sp . , both affiliated with the Proteobacterium phylum , were the principal bacterial genera isolated . The peak of activity of P . perniciosus occurs in September and October , a period that also corresponds to the L . infantum transmission season [59] . The analysis of the gut bacterial flora of sand flies collected in September reveals a higher diversity ( Fig 8 ) . In particular , we recorded the presence of Microbacterium , Micrococcus , Kocuria , Stenotrophomonas , and Bacillus sp . ( Actinobacteria , Proteobacteria and Firmicutes ) . In July , O . intermedium and Serratia sp . are the dominant bacteria genera in the midgut of P . perniciosus and these bacteria became undetectable towards the main peak of sand fly activity identified in Tunisia , i . e . , during the months of September and October [59] . The prevalence of L . infantum infection in the P . perniciosus population increases over the summer months and reaches a peak of 9% during September-October [60 , 61] . Ochrobactrum intermedium has been found previously to negatively affect Leishmania mexicana infection in L . longipalpis [55] . Certain strains of S . marcescens are capable of producing a pigment called prodigiosin , which ranges in color from dark red to pale pink depending on the age of the colonies . Derivatives of prodigiosin have recently been found to have anti-T . cruzi and anti-Leishmania ( Leishmania mexicana ) activity by promoting mitochondrial dysfunction leading to parasite programmed cell death [62 , 63] . To what extent such interplay between the bacterial colonies that exert toxic effects might interfere with the dynamic of L . infantum transmission awaits further investigation . Sand flies are vectors of medical and veterinary importance . Understanding the establishment of the sand fly microbiota is critical towards clarifying underlying details of sand fly Leishmania-microbiota interactions [64] . Bacteria such as O . intermedium , which has been previously characterized in the guts of larvae , pupae , and adults of P . duboscqi [44] , is an opportunistic pathogen to humans [65] . Serratia sp . , an entomopathogenic bacteria found in this study , has been previously isolated from L . longipalpis [40] and L . intermedia [56] . Bordetella avium , isolated only once from a specimen caught during July , has never been previously isolated from sand fly midgut microflora . Bordetella avium is a highly pathogenic bacterium , causing the avian bordetellosis [66] . Klebsiella ozaenae , known also as a human pathogenic bacterium , has been found in four out of the ten studied sand fly species ( not isolated in this study ) . K . ozaenae was isolated from the midgut of gravid and freshly fed females of P . papatasi and P . halepensis [45] and from some Lutzomyia species ( Fig 5 ) . Klebsiella species are ubiquitous in nature [67 , 68] and are recorded in all habitats where sand flies proliferate . Moreover , the presence of K . ozaenae in the midgut of gravid females [45] will highlight their capacity to survive in the gut of this insect . Nevertheless , as for all bacterial species known to be etiological agents of human diseases , the sole observation of their presence in sand fly gut is not sufficient to incriminate sand flies as a potential vector but gives information on the bacterial dissemination via blood-feeding insects . The data collected are not sufficient to incriminate sand flies as a biological vector of K . ozaenae but are enough to raise suspicion regarding their role in the dissemination of K . ozaenae . Furthermore , whether certain clinical outcomes from leishmaniasis may be linked to bacteria potentially deposited during the Leishmania-infected sand fly bite still remains to be fully investigated [68] . These studies will not only shed light on the effect of the gut bacterial community on the sand fly fitness but also on the establishment and the transmission of Leishmania parasites in endemic areas . This meta-analysis aimed to identify the best bacterial candidate for a paratransgenic approach . Our study is based on data aggregated from various publications that use culture-dependent and culture-independent methodologies and various set of technical approaches used to study the sand fly microbiota . For these reasons , conclusions raised with this study should be taken with caution and analyzed in the light of the limitations and pitfalls inherently associated with the compilation of heterogeneous data . Among limitations , some are linked to the physiological state of the sample . The gut microbiome is highly dynamic [69] and therefore influences the outcome of the analysis . When using a culture-dependent approach , we have to keep in mind that only 20% of environmental bacteria can be grown on a growth medium [70] . Therefore , the composition of the microbiota is not a direct reflection of the bacterial community structure ( abundance and richness ) inside the insect , but an altered version of the ecosystem from where they came . Nucleic acid-based analysis , involving historically used methods ( such as construction and Sanger sequencing of metagenomic clone libraries , automated ribosomal internal transcribed spacer analysis ( ARISA ) , terminal restriction fragment length polymorphism ( T-RFLP ) , denaturing gradient gel electrophoresis ( DGGE ) ) and next generation sequencing technology require a critical step that must combine an efficient cell disruption without DNA degradation and uniform nucleic acid extraction . Unfortunately , no consensus protocol for microbial DNA extraction of insect-associated microbiota is currently available [70] . Although 16S rRNA gene sequencing is highly useful with regards to bacterial classification , it has a low phylogenetic power at the species level for some genera [71 , 72] . Depending on the 16S rRNA variable region targeted and the database used to perform the taxonomic profiling , misassignation of bacterial OTU at the species level could be frequent [73] . Nevertheless , taking into account all the above mentioned limits and pitfalls , we think that an exhaustive approach aimed at collecting a maximum of data on the microbiota of sand flies will give key information on the most commonly identified bacteria in sand fly species and those that are more specific . Our groups are interested in the development of a paratransgenic platform to control the transmission of leishmaniasis . To that end , a strain of the non-pathogenic Bacillus species ( Bacillus subtilis ) , isolated from P . papatasi , is proposed as a possible candidate for paratransgenic approach . In this study , we isolated B . subtilis from P . perniciosus midgut , in addition to other Bacillus species ( Bacillus oleronius , Bacillus casamensis , Bacillus galactosidilyticus and Bacillus sp . ) . Bacteria belonging to the Bacillus genus seem to display a host-specific distribution , with only B . subtilis being isolated in more than one sand fly species ( P . halepensis , P . papatasi , and P . perniciosus ) . In addition , we observed that no bacteria belonging to the Bacillus genus have been characterized to date in adult New World sand fly species . Therefore , even if this bacterium possesses the main advantages of being non-pathogenic , easy to cultivate and to perform genetic manipulation , its use for paratransgenic control of Leishmania can be challenged by its capacity to establish long-term colonies in the gut of various sand fly species . In particular , if a paratransgenic approach is developed using B . subtilis as a host , it will be essential to probe its capacity to efficiently colonize the gut of Lutzomyia species and of other Old World sand fly species in which this bacterium has yet not been found in the gut . Thus , it will be of major epidemiological importance to develop a regional strategy for each endemic area with different bacterial isolates .
The knowledge of interactions between sand flies , Leishmania and nonpathogenic microorganisms that inhabit the gut will help to delineate an appropriate bacterial host recipient that can be used for paratransgenesis designed to prevent Leishmania transmission . The identification at the species level of the midgut’s cultured flora of P . perniciosus , linked to its seasonal variation , is likely to provide new perspectives towards a better understanding of the role of the gut bacterial community on sand fly-pathogen interactions . This knowledge is crucial in order to implement control strategies for sand fly zoonotic visceral leishmaniasis . | The use of conventional microbiological methods gave us the opportunity to investigate the richness of symbiotic bacteria that inhabit the gut of P . perniciosus during its main period of activity . Our results were subsequently analyzed in the framework of what has been done on sand flies microbiota in order to validate our results and to address the question of the definition of the core bacterial microbiota of sand flies . A meta-analysis on the respective gut microbiota of Old and New World sand flies shows that the majority of bacterial species is observed only in one host whereas less than 8% are shared by more than two hosts . Our results pinpoint the need for a more exhaustive understanding of the microbiota composition and dynamic in phlebotominae , with the aim to implement new biological approaches for the control of sand fly populations in order to prevent Leishmania transmission . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion",
"Conclusion"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"microbiome",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"bacillus",
"microbiology",
"sand",
"flies",
"animals",
"diptera",
"prokaryotic",
"models",
"experimental",
"organism",
"systems",
"insect",
... | 2017 | An integrated overview of the midgut bacterial flora composition of Phlebotomus perniciosus, a vector of zoonotic visceral leishmaniasis in the Western Mediterranean Basin |
Plant root border cells have been recently recognized as an important physical defense against soil-borne pathogens . Root border cells produce an extracellular matrix of protein , polysaccharide and DNA that functions like animal neutrophil extracellular traps to immobilize pathogens . Exposing pea root border cells to the root-infecting bacterial wilt pathogen Ralstonia solanacearum triggered release of DNA-containing extracellular traps in a flagellin-dependent manner . These traps rapidly immobilized the pathogen and killed some cells , but most of the entangled bacteria eventually escaped . The R . solanacearum genome encodes two putative extracellular DNases ( exDNases ) that are expressed during pathogenesis , suggesting that these exDNases contribute to bacterial virulence by enabling the bacterium to degrade and escape root border cell traps . We tested this hypothesis with R . solanacearum deletion mutants lacking one or both of these nucleases , named NucA and NucB . Functional studies with purified proteins revealed that NucA and NucB are non-specific endonucleases and that NucA is membrane-associated and cation-dependent . Single ΔnucA and ΔnucB mutants and the ΔnucA/B double mutant all had reduced virulence on wilt-susceptible tomato plants in a naturalistic soil-soak inoculation assay . The ΔnucA/B mutant was out-competed by the wild-type strain in planta and was less able to stunt root growth or colonize plant stems . Further , the double nuclease mutant could not escape from root border cells in vitro and was defective in attachment to pea roots . Taken together , these results demonstrate that extracellular DNases are novel virulence factors that help R . solanacearum successfully overcome plant defenses to infect plant roots and cause bacterial wilt disease .
The growing tip of a plant root is uniquely vulnerable to infection as it moves through the dense microbial community of the soil , unprotected by cuticle or bark . However , roots are defended by tiles of loosely attached secretory cells called root border cells , which produce a matrix of proteins , polysaccharide and DNA [1 , 2] . It has long been known that plants deposit DNA into soil [3–6] , but this extracellular DNA ( exDNA ) was only recently found to contribute to plant defense , possibly by trapping root pathogens [7] . For example , pea root border cells release DNA that limits root infection by the fungal pathogen Nectria hematococca , and treating pea root tips with DNase I accelerates necrosis caused by fungal infection [2] . ExDNA also forms the backbone of neutrophil extracellular traps , which are an important element of the animal immune system [8] . During microbial infection , neutrophils are recruited to the site of infection , where they release extracellular traps comprised of DNA matrices studded with antimicrobial compounds like histone , calprotectin , and serine proteases that can immobilize and/or kill various bacteria and fungi [8–10] . The DNA component is directly bactericidal because it chelates cations and disrupts bacterial membrane integrity [11] . NET release can be triggered by conserved molecules associated with pathogens , such as LPS , flagella , or the protein kinase C activator PMA; these are known as microbe-associated molecular patterns or MAMPs [12 , 13] . In response to this sophisticated physical and chemical defense , microbes have evolved ways to escape from extracellular traps . Streptococcus pneumoniae and Pseudomonas aeruginosa have modified cell surfaces that do not bind antimicrobial peptides or DNA , respectively [11 , 14 , 15] . Most commonly , pathogenic bacteria evade NETs by producing extracellular nucleases ( ex DNases ) that degrade the DNA backbone of the traps . Such nucleases are virulence factors for bacteria such as Group A Streptococcus , Streptococcus suis , Staphylococcus aureus , Vibrio cholerae , Neisseria gonorrhoea and also the eukaryotic parasite Leishmania infantum [16–22] . Indeed , nuclease treatment is enough to abolish the bactericidal activity of neutrophils [8] . Additionally , pathogens can convert nuclease-degraded trap components into counter-weapons that trigger neutrophil death [19] . It has been suggested that the exDNA released by plant border cells forms structures that are functionally analogous to animal NETs [23] . We will refer to these structures as NETs ( Nucleic acid Extracellular Traps ) . Like animal pathogens , many plant pathogenic microbes secrete DNases that may help them overcome NETs . Conidiospores of the plant pathogenic fungi Fusarium solani and Verticilium dahliae release exDNases [24 , 25] . Bioinformatic data suggest that several plant pathogenic bacteria have nucleases with secretory signals [7] . One such pathogen , Ralstonia solanacearum , carries genes putatively encoding two extracellular nucleases . R . solanacearum is a soil-borne Betaproteobacterium that causes the destructive bacterial wilt disease [26] . The pathogen has an exceptionally wide host range spanning more than 50 plant families , including economically important crops like potato , tomato , and banana , and it is notably difficult to control [26 , 27] . R . solanacearum is strongly attracted to root exudates by chemotaxis , and bacterial motility is required for effective root infection [28 , 29] . R . solanacearum enters host roots through wounds or natural openings , then multiplies and spreads rapidly in the water-transporting xylem vessels of the vascular system . The resulting mass of bacterial cells and extracellular polysaccharide obstructs water transport in the xylem and leads to wilting [30 , 31] . In late-stage disease , bacteria actively leave the roots and return to the soil . Many R . solanacearum virulence factors have been identified [32] , but the role of DNases in infection has not been explored . While it has been established that plant exDNA protects plants from root pathogens , only correlative evidence supports the idea that nucleases secreted by pathogens could overcome this defense . The presence of exDNA in the root border cell secretome and the putative secreted DNase genes in R . solanacearum genomes suggested that nucleases produced by this pathogen facilitate bacterial infection by digesting the exDNA in the root cap slime [7] . We directly tested this hypothesis with R . solanacearum mutants that cannot degrade exDNA . These experiments revealed that plants release extracellular traps in response to conserved pathogen factors and that R . solanacearum secretes two extracellular endonucleases that enable it to degrade and escape from these traps . Further , these enzymes contribute to root attachment , competitive fitness , plant colonization , and virulence of the pathogen .
We used pea , a model system for root border cell studies , and tomato , an economically important host of R . solanacearum , to study interactions between border cells and this pathogen . When pea border cell suspensions were inoculated with R . solanacearum , staining with DAPI and SYTOX Green revealed exDNA matrices in the suspension ( Fig 1A and 1B ) . Some GFP-tagged bacterial cells were trapped and immobilized along these DNA strands , while non-trapped bacteria moved freely in the suspension ( Fig 1B ) . SEM visualization of pea roots inoculated with R . solanacearum revealed web-like structures that resembled neutrophil extracellular traps . These were only present when pea roots were exposed to R . solanacearum ( Fig 1C ) . Bacteria were trapped by NETs near border cells within an hour after inoculation and in some cases , these structures could be seen originating from collapsed border cells ( Fig 1D ) . These NETs consisted of threads that varied in size , with the smallest about the diameter of chromatin fibers ( 20 nm ) . The bigger cables may consist of smaller threads bundled together , as has been observed in animal NETs [8] . Although exDNA was not associated with every border cell exposed to bacteria , it was abundant in suspensions with bacteria , while border cells mock-inoculated with water did not contain any noticeable exDNA ( Fig 2A ) . In the water control , SYTOX Green stained only the nuclei of pea border cells . In response to R . solanacearum , border cells released DNA traps quickly . Chromatin decondensation , visible as expanding , brightly stained ovals , was visible approximately 30 min after bacteria were added , as indicated by the fact that SYTOX Green was no longer limited to the nuclei . Release of exDNA followed shortly afterwards . About 45–60 min post inoculation , many border cells collapsed and were associated with SYTOX Green-stained DNA-containing NETs ( Fig 2A ) . Tomato border cells also released exDNA in response to R . solanacearum ( S1 Fig ) . In contrast , non-pathogenic bacteria such as E . coli , Sinorhizobium meliloti , and Pseudomonas aureofaciens did not trigger release of exDNA from pea border cells ( with the exception of Pseudomonas fluorescens ) ( S2 Fig ) . Plant recognition of pathogen signals such as MAMPs or effector proteins is a critical first step in defense responses . Although previous work identified EPS as an elicitor of plant defense against R . solanacearum in resistant tomato , little is known about plant responses that occur early in the interaction [33] . To identify the pathogen signal ( s ) that trigger NETs release , we inoculated pea border cells with mutants of R . solanacearum missing typical plant defense elicitors such as a positive regulator of type III secretion system ( hrpB ) [34] , the acidic exopolysaccharide I that is recognized by some plants ( epsB ) [33] , or flagellin ( fliC ) [29] . The hrpB and epsB mutants triggered release of thread-like extracellular DNA structures indistinguishable from those induced by wild-type strain GMI1000 ( Fig 2B ) . However , the fliC mutant did not induce trap formation , and border cells exposed to the fliC mutant were indistinguishable from those treated with water alone , even though R . solanacearum fliC cells were in close proximity to pea border cells ( S3 Fig ) . Treating border cells with 20 μg/ml of the conserved flagellin-derived peptide Flg22 also triggered some trap release . These results suggest that extracellular trapping is a previously undescribed element of MAMP-triggered immunity . The bactericidal activity of animal NETs is partly the result of histone proteins , which disrupt charges on bacterial surfaces [35] . Interestingly , histone H4 is also found in the plant root cap secretome [36] . Microscopic studies showed that R . solanacearum cells were immobilized in exDNA traps , but it was unclear if the traps were toxic to the pathogen . To test the hypothesis that plant histones contribute to the bactericidal effect of border cells , we first measured R . solanacearum survival following exposure to human histone H4 , which is 97% identical to plant histone H4 ( NIH Histone Sequence Database ) . At 10 μg/ml , histone H4 killed about 50% of the bacteria in 3 h , as determined by Live/Dead staining . Treatment with anti-Histone H4 antibody partially reversed the bactericidal effect of histone H4 on R . solanacearum ( Fig 3A ) . To determine if border cell trapping can kill R . solanacearum , we measured bacterial survival following 3 h incubation with pea border cells . About 25% of the R . solanacearum population was killed compared to a water-treated control population . Freeing the bacteria from the traps prevented bacterial killing , because when the suspension was treated with DNase I there was no difference in survival between bacterial populations exposed to border cells and a water-treated control population ( Fig 3B ) . Adding anti-Histone H4 antibody to the challenging pea border cells also fully protected the pathogen , possibly because the antibody either prevented bacteria from binding to histones in exDNA traps or neutralized the antimicrobial activity of histone H4 . This suggests that both exDNA and histone H4 are required for effective killing of R . solanacearum by border cell traps . The genome of R . solanacearum strain GMI1000 contains two putative endonuclease genes , nucA ( locus tag: RSc0744 ) and nucB ( locus tag: RSc2452 ) , and both genes encode predicted secretion signals ( Signal P 4 . 0 ) ( S4 Fig ) . Expression levels of nucA and nucB are high in two different R . solanacearum strains growing in rich medium and during tomato pathogenesis [37] . Both nucA and nucB belong to the core Ralstonia solanacearum genome and are highly conserved in all 28 sequenced R . solanacearum strains in the MaGE RalstoniaScope database and in the opportunistic human pathogen R . pickettii , but are not present in the related free-living bacterium R . eutropha ( http://www . genoscope . cns . fr/agc/microscope/home/index . php ) [38] . To determine the biological role ( s ) of R . solanacearum’s extracellular nucleases , we constructed deletion mutants lacking nucA , nucB , or both nucA and nucB ( referred to hereafter as ΔnucA/B ) ( S4 Fig ) . All three DNase mutants produced significantly less exDNase activity than the parental strain ( Fig 4A ) . The ΔnucA mutant showed a small reduction in exDNase activity in vitro , but the ΔnucB and ΔnucA/B double mutants produced only about 18% of wildtype DNase activity levels ( in relative fluorescence units ) , indicating that NucB is responsible for most of the exDNase activity in strain GMI1000 Adding a copy of nucB to the ΔnucB mutant restored wild-type nuclease activity levels . Complementation restored the wild-type phenotype to ΔnucB and partially restored ΔnucA mutant . The individual and double nuclease mutants did not express the corresponding genes , and the complemented strains had transcript levels comparable to those of wild-type strain GMI1000 ( S4B and S4C Fig ) . Bacterial culture supernatants contained nuclease activity , indicating that NucA , NucB or both enzymes may be secreted . The Type II secretion system ( T2SS ) exports many cell wall degrading enzymes in R . solanacearum [39] . Cell-free culture supernatants of a GMI1000 gspM mutant , which lacks an essential component of the T2SS , had only 18% of wildtype DNase activity , similar to the activity produced by the ΔnucB mutant . This suggests that R . solanacearum exports at least NucB via the T2SS . Wild-type strain GMI1000 grew on minimal medium supplemented with salmon sperm DNA , but growth was slow and weak , reaching log phase after approximately 24 h and stationary phase after 60 h . The ΔnucA/B mutant grew even less ( S5 Fig ) . These results indicate that R . solanacearum needs exDNases to use exDNA as a carbon source , but DNA is not an ideal carbon source for this organism in vitro . Bioinformatic analysis suggested that NucA and NucB are distinct DNases ( S3 Fig ) . Purified NucA and NucB were approximately 30 kDa and 15 kDa , respectively , in agreement with bioinformatic predictions ( S6A Fig ) . We could purify a soluble form of NucA from E . coli only by removing the gene region encoding the 55 N-terminal residues containing a predicted transmembrane domain; this supports the bioinformatic prediction that NucA localizes on a bacterial membrane . To confirm that NucA is secreted through the inner membrane , we used a NucA::PhoA fusion construct and an alkaline phosphatase plate assay . Colonies of E . coli phoA- strain KS272 expressing the NucA-PhoA fusion were bright blue , similar to those of the FtsI::PhoA positive control ( S6B Fig ) , indicating that the NucA catalytic domain is located in the periplasm or outside the cell . In addition , purified NucA and NucB digested all DNA substrates tested , including salmon sperm DNA , R . solanacearum genomic DNA , pea genomic DNA and supercoiled plasmid DNA , indicating that they are non-specific endonucleases ( Fig 4B ) . NucA required Mg for full activity , but NucB was fully active without a cationic cofactor ( S6C and S6D Fig ) . To determine the effects of purified DNase on root border cell NETs , we measured total DNA from root border cells inoculated with R . solanacearum using SYTOX Green ( Fig 4C ) . Addition of 10 units of commercial DNase I reduced SYTOX Green fluorescence approximately 6-fold . Adding 2 μg of purified overexpressed NucA or NucB also reduced relative fluorescence to 30% of the no enzyme control , demonstrating that R . solanacearum NucA and NucB can degrade the DNA of plant border cell traps . The total fluorescence signal of pea border cells was unchanged 6 h after addition of R . solanacearum cells , suggesting that no de novo DNA synthesis occurred in response to R . solanacearum . For a more dynamic perspective on the effects of nucleases on trapped bacteria , we filmed R . solanacearum interactions with pea border cells using light microscopy . Wild-type bacteria , which were initially highly motile , were immobilized shortly after they were added to a border cell suspension ( S1 Video ) . However , the wild-type cells were able to escape from traps after 1 h and they remained free at 24 h ( S2 Video ) . In contrast , the ΔnucA/B double nuclease mutant was still immobilized in NETs 24 h after incubation with pea border cells ( S3 Video ) . Treatments with purified NucA ( S4 Video ) or NucB ( S5 Video ) released the trapped ΔnucA/B bacteria . Together , these data indicate that R . solanacearum cells immobilized by plant NETs can be freed by either their endogenous exDNases or exogenously applied nucleases . We used the nuclease-deficient mutants to test the hypothesis that exDNases contribute to R . solanacearum bacterial wilt virulence at early stages of infection by helping the pathogen escape from the exDNA traps produced by plant roots . Susceptible tomato plants were inoculated with the wild-type strain or nuclease mutants using a soil-drenching method that introduces bacteria to the soil near unwounded plants . This requires the pathogen to follow its natural route of infection through an intact host root system . All three exDNase mutants were quantitatively reduced in virulence compared to the parental strain ( P<0 . 05 , repeated measures ANOVA ) ( Fig 5A–5C ) . The delayed disease development that we observed in tomato plants inoculated with ΔnucA/B could be the result of reduced plant colonization ability . To evaluate this hypothesis , we quantified bacterial populations in the mid-stems of wilt-susceptible tomato plants following soil-drench inoculation with either wild type or the ΔnucA/B mutant . The population sizes of the wild-type strain were at least two orders of magnitude larger than those of ΔnucA/B at 4 and 6 days post inoculation ( Fig 5D ) ( P<0 . 05 , Student’s t-test ) . To measure the ability of a nuclease-deficient strain to compete effectively with its wild-type parent in a plant host , we inoculated tomato plants with a 1:1 mixture of wild-type and ΔnucA/B mutant cells and measured the population size of each strain in tomato mid-stems when wilt symptoms first appeared . The wild-type bacterium slightly outcompeted the ΔnucA/B mutant with a competitive index of 0 . 88 ( Fig 5E; P<0 . 05 , Wilcoxon signed-ranked test ) . To evaluate the contribution of bacterial exDNases to bacterial virulence on plant roots , we inoculated axenic tomato and pea seedlings in germination pouches with R . solanacearum . Ten days after inoculation , mock-inoculated pea and tomato plants developed healthy root systems , but roots of plants inoculated with wild-type R . solanacearum were severely stunted ( Fig 6A and 6B ) . Pea roots inoculated with the ΔnucA mutant were stunted like those inoculated with the wild-type strain , but peas inoculated with either ΔnucB or ΔnucA/B grew nearly as well as the mock-inoculated control plants ( Fig 6A ) . On tomato seedlings , only the ΔnucA/B double nuclease mutant caused significantly less root stunting than wild-type ( Fig 6B ) , suggesting that both nucleases additively contribute to symptom development on this host . Attachment to root surfaces is important for R . solanacearum infection [40] . If exDNA in the root cap trap matrix blocks pathogen access to the root surface , we would expect that a R . solanacearum strain lacking exDNases would not attach to plant roots as well as the wild type strain . We used a seedling root attachment assay to test the hypothesis that the reduced virulence of nuclease mutants in the soil-drenching assay was caused by poor root attachment . There were no significant differences among the strains in attachment to tomato seedling roots , although the ΔnucB complemented strain trended towards better attachment . In contrast , the nuclease mutants did not attach to pea roots as well as the wild-type strain , and complementation restored wild-type levels of attachment to pea roots ( Fig 6C and 6D ) . This indicates that exDNases can play a critical role in R . solanacearum attachment to seedling root surfaces , at least on pea .
Very soon after the discovery that animal neutrophil cells immobilize and kill pathogens with DNA-containing extracellular traps , it was found that that microbes escape from these traps by means of secreted nucleases [8 , 15] . A few years later Hawes and colleagues showed that plant root border cells , which are quite biologically distinct from the neutrophils of the animal circulating immune system , also release DNA-containing extracellular traps that protect plants from pathogens [2] . Because this effective defense likely put strong selection pressure on pathogen populations , and because enzymatic degradation is a straightforward counter-defense , we hypothesized that the root-infecting plant pathogen R . solanacearum also uses nucleases to escape from DNA-containing traps . NET release appears to be a specific response to pathogen recognition . Consistent with a previous report that E . coli was not trapped by maize border cells [42] , we found that pea border cells did not release NETs in response to the non-pathogenic bacteria P . aureofaciens , S . meliloti , or E . coli . Similarly , non-pathogenic fungi were not aggregated or inhibited by pea roots and a fungal pathogen was not inhibited by non-host plant species [43] . These observations suggest that plant NET release is triggered by a specific signal . Border-like cells of flax and Arabidopsis respond to several MAMPs , including peptidoglycan and Flg22 , with ROS production and other innate defenses collectively known as MAMP-triggered immunity or PTI [44] . Bacterial flagella are potent inducers of PTI; flagellin-mediated responses account for an estimated 90% of PTI in tomato [45] . We found that the Pseudomonas-derived flagellin sub-peptide Flg22 could trigger NET production by itself . Our finding that border cells recognize this well-known MAMP suggests that NET trapping is an element of PTI [46] . Consistent with this idea , an aflagellate fliC mutant of R . solanacearum did not induce NET release . We previously found that R . solanacearum flagellin is not recognized by the Arabidopsis FLS2 receptor , probably because the N-terminal R . solanacearum flagellin sequence that corresponds to Flg22 contains multiple polymorphisms compared to the canonical Pseudomonas flagellin Flg22 sequence [47] . Therefore , the pea root border cells in our experiment recognized both the synthetic Flg22 peptide and an element of R . solanacearum flagellin that is different from Flg22 . This additional MAMP could be the recently-discovered FlgII-28 peptide , which is recognized by the tomato FLS3 receptor [48] . It is not clear why NETs were not triggered by the flagella of non-pathogenic bacteria . E . coli DH5α may not have triggered NET release because E . coli lab strains are often poorly flagellated [49] . Some non-pathogenic rhizosphere-dwelling microbes like P . fluorescens and S . meliloti have developed mechanisms to suppress plant NET release . Additional studies are needed to explore this interesting result . Flagellin can trigger NET release in animal neutrophils [50] , as can calcium [51] , LPS [8] or nitric oxide ( NO ) [52] . Interestingly , NO is also a plant signal molecule that connects multiple defense pathways [53] . Several lines of evidence indicate that ROS leads to animal cell NET release by activating the release and translocation of neutrophil elastase and myeloperoxidase into the nucleus and promoting chromatin decondensation [54–56] . Phorbol-12-myristate-13-acetate ( PMA ) , a widely-used animal NET activator , also activates protein kinase C-like proteins in maize plants [57] . Since plant and animal defense responses are activated via a parallel series of MAPK cascades , with several conserved MAMP receptors [58] , we speculate that similar plant signaling pathways may govern the formation of border cell extracellular traps , although the exact molecular machinery that orchestrates plant NET production still requires further study . We found that histone H4 , the only histone detected in root cap mucilage [1] , is bactericidal to R . solanacearum cells in vitro , and treatment with anti-H4 antibody prevented pea border cells from killing the pathogen . Histones are toxic to a wide range of bacteria , possibly because they bind to and disrupt bacterial membranes [59] . Our results suggested that histone H4 can kill plant pathogens , either by direct contact and/or by binding bacterial cells to DNA strands where they are exposed to concentrated antimicrobial compounds . In addition to histones , root border cells have a distinct metabolic profile that is highly enriched in secondary metabolites like flavonoids and triterpene saponins that could kill or inhibit soil pathogens that contact NETs [60] . It is well established that root border cells contribute to plant health [61] . Although every plant species produces a fixed range of border cells each day [62] , border cell production is also cultivar-dependent [41] . Intriguingly , we observed that seedlings of a bacterial wilt-resistant tomato breeding line , Hawaii 7996 , released almost three times as many border cells as a wilt-susceptible tomato cultivar , Bonny Best ( S7A Fig ) . Further , border cells from the resistant line appeared to be surrounded by a thicker mucilage layer ( S7B Fig ) . Together with our finding that root border cells trap and exclude the bacterial wilt pathogen , this preliminary observation suggests the testable hypothesis that more abundant and more mucoid border cells increase plant resistance to soil-borne diseases . If additional studies find a consistent correlation between number of root border cells and wilt resistance , plant breeders seeking to reduce crop losses to root-infecting pathogens might be able to select for this easily quantifiable trait . Wild-type R . solanacearum cells quickly escaped from pea root border cell NETs , but a mutant missing both nucleases remained immobilized . Adding purified nucleases ( NucA , NucB or DNase I ) reversed bacterial trapping . Together these two findings are direct evidence that , like many animal pathogenic microbes , this plant pathogen uses nucleases to escape extracellular DNA traps from host root border cells . In addition , we demonstrated that exDNases are required for full bacterial wilt virulence . Interestingly , although the disease progress curves of the three mutants differed from those of wild type , the curves of the three mutants were not significantly different from each other . This suggests that although loss of either or both exDNases reduced virulence , some mutant cells escaped or avoided the traps at this relatively high inoculum density . It appears that these escaped cells could then infect normally , although at a reduced rate . To the best of our knowledge , this is the first report that DNase contributes to plant pathogenesis . We suspect this will prove to be a widespread counter-defense mechanism since extracellular DNases are produced by many fungal and bacterial phytopathogens [25 , 63–65] . Interestingly , the ability of R . solanacearum to escape from NETs correlated with its ability to attach to and inhibit pea root development . Although pea is not a natural host of R . solanacearum [66] , in our assay conditions it caused necrosis and stunting on pea root and inhibited growth of the aboveground plant parts . These symptoms did not occur on pea seedlings inoculated with the double nuclease mutant . We also found that nucleases were not required for attachment to roots of tomato , a natural host plant . This difference could be due to the smaller number of border cells released by tomato seedlings . In contrast to pea roots , which make thousands of root border cells per day , tomato roots produced only approximately 200 border cells daily ( S7 Fig ) . Therefore , tomato border cells may not trap R . solanacearum cells very effectively at the seedling stage when plants were collected for the pouch assay . In larger plants , lateral root development with more root tips likely results in more border cells; also , border cells are renewed daily as roots move through the soil [61] . As a consequence , border cell NETs could efficiently limit bacterial entry to whole plants growing in soil , as observed in our plant virulence assays . Both of the R . solanacearum extracellular nucleases contributed to stunting symptoms in tomato roots , but NucB was more important in pea root infection than NucA . This may be because NucA requires a cation cofactor and magnesium availability may differ between pea and tomato roots . In addition , exDNA can sequester cations from bacterial cell surfaces [11] . If the abundant pea root border cells release large amounts of exDNA , the resulting cation sequestration could inhibit NucA . Differential inhibition of these enzymes by natural DNase inhibitors like actin [67] , which is present in the pea root cap secretome [1] , could also affect the relative activity of each DNase in the rhizosphere . The double nuclease mutant grew almost as well as the wild-type strain in a paired-strain competition plant assay intended to detect quantitative differences in strain fitness [68] . This result may appear inconsistent given the significantly lower colonization ability and virulence of the ΔnucA/B double mutant . However , we speculate that when the two strains were inoculated together , nucleases secreted by the wild-type strain functionally complemented the ΔnucA/B mutation This would allow the double mutant to escape NETs and attach to roots , and colonize stems almost as well as its wild-type parent . DNA is rich in carbon , nitrogen and phosphorus and is abundant in soil and marine sediments [70 , 71] . Microbial nucleases can thus be important for nutrient scavenging in those environments . Extracellular nucleases allow many bacteria , including human pathogens like Serratia marcescens , E . coli or Pseudomonas aeruginosa , to grow on DNA as sole carbon or nitrogen source [72–75] . Indeed , NucA and NucB enabled R . solanacearum to use DNA as a sole carbon source . Although in vitro growth on DNA was slow , this ability could confer a valuable advantage in the soil where the bacterium must compete for nutrients with other soil inhabitants . In addition , extracellular nucleases could contribute to R . solanacearum success in other ways . For example , deoxyadenosine and other by-products of NET degradation by Staphylococcal nucleases are toxic to macrophages , which generates a macrophage-free zone around abscesses during S . aureus infections [69] . It would be interesting to determine if R . solanacearum nucleases also release toxic byproducts like deoxyadenosine . In conclusion , we demonstrate for the first time that DNases are virulence factors in a plant pathogen . Their importance derives from the apparently convergent evolution of DNA-based defenses in animal and plants: plant root border cells and animal macrophages both deploy DNA-containing NETs to trap and kill pathogens . In response , animal pathogens and at least one plant pathogen use secreted enzymes to overcome a DNA-based host defense system . We also observed that R . solanacearum cells trigger release of DNA-containing traps from plant root border cells . An aflagellate fliC mutant cannot induce trap release , but the flg22 flagellin peptide alone can do so , suggesting that root border cell traps are a previously unrecognized element of PAMP-triggered plant innate immunity .
Strains and plasmids used in this study are listed in S1 Table . R . solanacearum was grown at 28°C in either Casamino Acids-Peptone-Glucose ( CPG ) broth or solid TZC ( 1 . 8% agar+0 . 05% tetrazolium chloride ) [76] . For DNase activity assays and natural transformation , bacteria were grown in Boucher’s Minimal Medium ( BMM ) supplemented with 0 . 2% glucose [77] . E . coli was grown in Luria-Bertani medium at 37°C . Media were supplemented with antibiotics as needed: kanamycin ( 25 μg/ml ) , gentamycin ( 25 μg/ml ) and tetracycline ( 15 μg/ml ) . Primers used for strain construction are listed in S2 Table . To create deletion mutants of R . solanacearum strain GMI1000 lacking nucA ( locus tag Rsc0744 ) or nucB ( RSc2452 ) , we designed primers to amplify approximately 500 bp upstream and downstream of each target gene . The resulting fragments were annealed to the pSJG GmR cassette ( for ΔnucA construct ) or a the pBYJ-Km1 KmR cassette ( for ΔnucB ) using splicing-by-overlap-extension PCR [78] . The resulting deletion constructs were ligated to pCR-Blunt and transformed into E . coli cells . Plasmids containing correct constructs as verified by PCR were electroporated into R . solanacearum and putative mutants with the expected antibiotic resistances were confirmed by PCR . A double ΔnucA/B mutant was generated by transforming the ΔnucA deletion construct in pCR-Blunt into GMI1000 ΔnucB and selecting for KmR , GmtR transformants . A type II secretion system-deficient mutant was constructed by natural transformation of GMI1000 with genomic DNA from K100 ( K60 gspM::KmR ) [47] . To complement the nuclease mutants we used pRCT-GWY , which integrates into a selectively neutral attTn7 site in the R . solanacearum chromosome [79] . Complementation constructs containing the nucA or nucB ORFs with their putative promoters were ligated into pCR-Blunt , then subcloned into the SpeI/BglII or KpnI/AvrII sites of pRCT-GWY to create pRCT-nucAcom and pRCT-nucBcom , respectively . These constructs were naturally transformed into GMI1000 ΔnucA and ΔnucB [77 , 80] . Transformants with desired antibiotic resistance profiles were screened for complementation by PCR , nuclease activity assay and qRT-PCR; the confirmed complemented strains were named GMI1000 nucAcom and nucBcom . For gene expression analyses , the hot phenol-chloroform method [37] was used to extract total RNA from bacterial cultures grown to mid-log phase in CPG and cDNA was synthesized with SuperScript III VILO ( Invitrogen , Carlsbad , CA ) . Expression of nucA and nucB was measured using quantitative RT-PCR with Power SYBR Green master mix ( Invitrogen ) in an ABI PRISM 7300 real-time PCR system ( Applied Biosystems , Life Technologies , Waltham , MA ) . Relative transcript abundance was normalized using the reference gene rplM [81] . DNase activity was quantified using the DNase Alert Kit ( IDT , Coralville , IA ) . Overnight cultures of R . solanacearum grown in BMM+0 . 2% glucose were centrifuged , passed through a 0 . 2-μm filter , and this cell-free supernatant was incubated with DNase Alert buffer and substrate at 37°C for 3 h according to the manufacturer’s instructions . DNase activity was read as fluorescence signal released by cleavage of a synthetic oligonucleotide with a fluorescent Hex probe at one end and a fluorescence quencher at the other , using a Synergy HT microplate reader ( Biotek Instruments , Winooski , VT ) . Ten units of DNase I ( Ambion , Life Technologies , Carlsbad , CA ) and uninoculated BMM were used as positive and negative controls , respectively . DNA sequences encoding amino acid residues 51–275 of NucA ( the 225 amino acids following a predicted trans-membrane domain ) and for full-length NucB were amplified from GMI1000 genomic DNA by PCR and cloned into pCR-Blunt , then subcloned into in the pET29b expression vector , in frame with a 6X-His tag sequence ( Novagen , EMD Biosciences , Madison , WI ) . The resulting expression constructs , pET29b-EnucA and pET29b-EnucB , were transformed into the expression strain E . coli BL21 Star ( Invitrogen ) and confirmed by colony PCR and sequencing . Overnight cultures of E . coli strains carrying the nucA and nucB overexpression vectors were diluted 1:500 into 1 L of LB+kanamycin , and grown for 3 h at 37°C . Protein expression was induced by adding 1mM IPTG . After an additional 3 h , bacteria were collected by centrifugation and the His-tagged proteins were purified using nickel columns according to the manufacturer’s protocol ( Qiagen , Valencia , CA ) . Purified proteins were analyzed by SDS-PAGE and by Western Blot with anti-His antibody ( Invitrogen ) . Protein concentration was measured using the Bradford assay . The substrate specificity of purified nucleases was determined by incubating 1 μg of purified nuclease protein in DNase I buffer for 30 min at 37°C with 1 μg of either salmon sperm DNA ( Sigma ) , R . solanacearum GMI1000 genomic DNA , pea genomic DNA , or supercoiled plasmid pUCGM [82] . Reactions were stopped by adding 0 . 1 vol 10% SDS and DNA degradation was visualized by electrophoresis in a 1 . 5% agarose gel . Cation preference of NucA and NucB was determined using a modified DNase Alert assay , described above . One microgram of purified nuclease was incubated at 37°C for 3 h in PBS with 10 μl DNase Alert substrate and 10 mM of either Mg2+ , Mn2+ , Zn2+ , or DNase Alert buffer . Buffer amended with 10 mM EDTA to sequester cations was used as a negative control . We used phoA fusions to test for the membrane location of NucA in E . coli . The nucA ORF without its native stop codon was amplified from GMI1000 genomic DNA using PCR primers nucA-phoAF/R and the resulting PCR product was cloned into pCR-Blunt and then subcloned into the phoA fusion vector pDSW438 [83] to create pDSW-nucA-phoA . The resulting construct was confirmed by PCR and sequencing , then transformed into chemically competent cells of E . coli KS272 ( phoA- ) . PhoA activity was determined by streaking E . coli strains carrying pDSW438 ( empty vector ) , pDSW439 [83] ( positive control ) or pDSW-nucA-phoA onto plates amended with arabinose and X-phos ( 5-bromo-4-chloro-3-indolyl phosphate ) ( Sigma-Aldrich , St . Louis , MO ) . Results were read after overnight incubation at 37°C . Colonies were blue if the PhoA domain was facing the periplasm and white if the PhoA domain was in the cytoplasm . Pea seeds ( cv . Little Marvel ) were sterilized in full-strength bleach for 30 min , then 95% ethanol for 10 min , followed by five washes with sterile water [84] . Pea seeds were imbibed in water for 6 h , and allowed to germinate on 1% water agar overlaid with sterile filter paper for 2 days in the dark at room temperature . Root border cells , collected by dipping root tips into sterile water for 5 min , were incubated with 1x108 CFU/ml . R . solanacearum-GFP for 30 min and then stained with SYTOX Green or DAPI ( Life Technologies , Carlsbad , CA ) according to the manufacturer’s instructions . Border cells were visualized under an Olympus BX60F5 fluorescence compound microscope ( Olympus , Japan ) . To determine which R . solanacearum signals trigger release of plant extracellular traps , we inoculated pea border cells with wild-type bacteria or mutants lacking either EPS ( epsB ) , flagellin ( fliC ) , or the type III secretion system ( hrpB ) ( S1 Table ) , or with 20 μg/ml of the synthetic P . aeruginosa flagellin peptide Flg22 ( Genscript , Piscataway Township , NJ ) . The specificity of trap release was tested by inoculating pea root border cells with suspensions of the non-pathogenic bacteria Sinorhizobium melilotii , Pseudomonas aureofaciens , P . fluorescens and E . coli . Approximately 10 , 000 border cells from 4 pea roots were pooled and used for each biological replicate . Each experiment was repeated three times . After staining with SYTOX Green , the suspensions were observed under a Zeiss Elyra PS . 1 LSM 780 confocal laser scanning microscope between 30 and 60 min after inoculation with 107 CFU/ml R . solanacearum . For each replicate , at least 5 images were taken; representative images are shown . Confocal imaging was performed at the Newcomb Imaging Center , Department of Botany , University of Wisconsin-Madison . To visualize border cell extracellular traps by scanning electron microscopy ( SEM ) , axenic pea roots were inoculated by dipping the roots into a 107 CFU/ml suspension of R . solanacearum GMI1000 . One hour after inoculation at room temperature , the roots were excised and fixed for 8 h in primary fixative ( 2 . 5% glutaraldehyde , 2% formaldehyde , 0 . 003 M MgCl2 , 0 . 003 M CaCl2 in 0 . 05 M PIPES buffer , pH 7 ) . The samples were washed twice with 0 . 05M PIPES buffer and fixed with secondary fixative ( 1% OsO4 in 0 . 05M PIPES ) overnight , then rinsed twice with 0 . 05M PIPES , dehydrated in ascending ethanol concentrations ending at 100% and critical point dried . Root samples were sputter coated with gold-palladium ( 1 . 8 kV and 6 mA for 60 s ) and examined in a Zeiss LEO 1530 high-resolution scanning electron microscope at the Material Sciences Center , University of Wisconsin-Madison . To measure the bactericidal activity of histone H4 , R . solanacearum GMI1000 cells were grown overnight in CPG , pelleted , washed , and adjusted to an OD600 of 1 . 5 with sterile water . Recombinant human histone H4 ( New England Biolabs , Ipswitch , MA ) was added to a final concentration of 10 , 20 , 50 or 100 μg/ml and incubated for 3 h at 28°C . Bacterial viability was then measured using the LIVE/DEAD Baclight kit ( Life Technologies ) according to the manufacturer’s protocol . The percentage of live cells was determined by plotting the SYTO9/PI fluorescence ratio against a standard curve based on known mixtures of live and dead cells . The experiment was repeated twice , each with three technical replicates . Border cells from four pea roots ( approximately 10 , 000 cells ) were inoculated with 104 R . solanacearum cells , incubated in water at room temperature for 3 h with or without 10 units of DNase I or 10 ng/ml anti-histone H4 antibody , vortexed vigorously , and serially dilution plated on CPG plates . Colonies were counted after 2 days at 28°C to quantify the surviving bacterial population size . The experiment was repeated three times , each with six technical replicates . Approximately 5000 pea root border cells collected as described above were placed in each well of a black 96-well plate ( Corning Inc . , Corning , NY ) with or without 5x106 CFU R . solanacearum . Ten units of DNase I ( Ambion ) or 10 μg of purified NucA or NucB were added along with 2 nM of SYTOX Green DNA stain and reactions were incubated at 25°C for 6 h and fluorescence was measured with a microplate reader at 485nm excitation/528nm emission . Relative fluorescence units ( RFU ) in each well at 6 h were normalized to RFU of the same well at the beginning of the assay . The experiment was repeated three times , with four technical replicates in each treatment . Attachment of R . solanacearum to pea and tomato roots was measured as described [40] , with some modifications . Pea seeds were sterilized as described above . Tomato seeds ( cv . Bonny Best ) were surface-sterilized by soaking for 10 min in 10% sodium hypochlorite; 5 min in 70% ethanol and washed five times in sterile water . Sterilized seeds were then germinated on 1% water agar plates overlaid with filter paper to avoid dispersal of root border cells from the root tip by free water on the agar surface . After 2 or 3 days , the seedlings were transferred to a water agar plate and root tips were inoculated by gently pipetting 2 , 500 bacteria/cm root and incubating for 2 h at room temperature . Inoculated roots were excised , washed in sterile water to remove unattached bacteria and blotted dry . Four roots were pooled as a technical replicate , homogenized in sterile water using a PowerLyzer ( MoBio ) , and the bacterial population size attached to the root surface was quantified using serial dilution plating . The experiment was repeated three times with at least five technical replicates per treatment in each experiment . Percent cells attached was calculated as the number of CFU recovered from plates over total bacteria inoculated , and normalized to wild-type attachment . Wilt-susceptible tomato plants ( cv . Bonny Best ) were used to evaluate R . solanacearum virulence using a naturalistic soil-drenching method that requires the pathogen to locate and infect intact host roots [40] . Briefly , unwounded 21-day old plants were inoculated by pouring onto the soil a suspension of wild-type GMI1000 , ΔnucA , ΔnucB or ΔnucA/B to a final concentration of 1x107 CFU/g soil . Plants were maintained in a growth chamber with a 12 h light cycle at 28°C and rated daily over 14 days using a 0–4 disease index scale as follows: 0 , no leaf wilted; 1 , 1–25% leaves wilted; 2 , 26–50% leaves wilted; 3 , 51–75% leaves wilted and 4 , more than 75% leaves wilted . The experiment was repeated three times and each experiment included 14 plants per treatment . The ability of bacteria to colonize plants was quantified by sampling the mid-stems of tomato plants soil-drench inoculated as described above with either the wild-type strain or the ΔnucA/B mutant . At 2 , 4 and 6 days after inoculation , a 100-mg transverse slice was harvested from the mid-stem of each plant , homogenized in 900 μl of sterile water using a PowerLyzer ( MoBio , Carlsbad , CA ) , and bacterial population sizes were quantified using serial dilution plating . The experiment was repeated twice with at least 10 plants per biological replicate . A root growth assay was used to quantify the effect of R . solanacearum wild-type strain and exDNase mutants on plant root length . Tomato seeds ( cv . Bonny Best ) and pea seeds ( cv . Little Marvel ) were sterilized as described above and germinated in the dark at 28°C or room temperature , respectively . Three-day old tomato seedlings and two-day old pea seedlings were inoculated by dipping root tips for 15 min into bacterial suspensions containing 107 CFU/ml ( pea ) and 106 CFU/ml ( tomato ) . Inoculated seedlings were placed in germination pouches ( Mega International , Newport , MN ) and kept at 28°C ( tomato ) or 25°C ( pea ) . Root systems were imaged 10 days post inoculation . The experiment was repeated twice with 5–6 pea seedlings or 10 tomato seedlings per germination pouch . Twenty-one day old ‘Bonny Best’ tomato plants were inoculated by pouring into each pot 50 ml of a suspension containing a 1:1 ratio of GMI1000-GFP and the ΔnucA/B mutant at total density of 108 CFU/ml . When the first wilting symptoms appeared , a 100 mg transverse stem section was excised from each wilted tomato plant , homogenized in 900 μl of sterile water by a PowerLyzer ( MoBio , Carlsbad , CA ) and serially dilution plated on CPG plates amended with Tet ( to select for wild-type ) , or Kan + Gen ( to select for the ΔnucA/B strain ) . The experiment was repeated twice with at least 30 plants per replicate . Overnight CPG cultures of R . solanacearum were pelleted by centrifugation , washed three times with sterile water , resuspended in BMM to A600 = 0 . 01 and incubated in 96-well microtiter plates with or without 5 μg/ml of salmon sperm DNA ( Sigma-Aldrich , St . Louis , MO ) as the sole carbon source . Bacterial growth was monitored over 72h as A600 using a microplate reader . The experiment was repeated three times with three technical replicates in each experiment . | Plant root tips are covered by a protective sleeve of loosely attached border cells that can release a matrix containing proteins , polysaccharides , and DNA . In animal immune systems , extracellular DNA forms the backbone of neutrophil extracellular traps ( NETs ) deployed by immune cells to immobilize and kill invading microbes . Some animal pathogens can secrete DNases to degrade NETs and facilitate infection . We found that plant border cells release DNA-containing extracellular traps in response to the high-impact plant pathogenic bacterium Ralstonia solanacearum . R . solanacearum secretes two DNases that free the pathogen from these extracellular traps . The bacterium needs these DNases for full virulence and normal colonization of its host plants . This work reveals that , like animal pathogens , the plant pathogen R . solanacearum can overcome a DNA-based host defense system with secreted enzymes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"nucleases",
"deoxyribonucleases",
"enzymes",
"dna-binding",
"proteins",
"enzymology",
"epithelial",
"cells",
"plant",
"science",
"parietal",
"cells",
"crops",
"plant",
"pathology",
"seedlings",
"plants",
"legumes",
"crop",
"scie... | 2016 | Escaping Underground Nets: Extracellular DNases Degrade Plant Extracellular Traps and Contribute to Virulence of the Plant Pathogenic Bacterium Ralstonia solanacearum |
Dengue virus is an emerging infectious agent that infects an estimated 50–100 million people annually worldwide , yet current diagnostic practices cannot detect an etiologic pathogen in ∼40% of dengue-like illnesses . Metagenomic approaches to pathogen detection , such as viral microarrays and deep sequencing , are promising tools to address emerging and non-diagnosable disease challenges . In this study , we used the Virochip microarray and deep sequencing to characterize the spectrum of viruses present in human sera from 123 Nicaraguan patients presenting with dengue-like symptoms but testing negative for dengue virus . We utilized a barcoding strategy to simultaneously deep sequence multiple serum specimens , generating on average over 1 million reads per sample . We then implemented a stepwise bioinformatic filtering pipeline to remove the majority of human and low-quality sequences to improve the speed and accuracy of subsequent unbiased database searches . By deep sequencing , we were able to detect virus sequence in 37% ( 45/123 ) of previously negative cases . These included 13 cases with Human Herpesvirus 6 sequences . Other samples contained sequences with similarity to sequences from viruses in the Herpesviridae , Flaviviridae , Circoviridae , Anelloviridae , Asfarviridae , and Parvoviridae families . In some cases , the putative viral sequences were virtually identical to known viruses , and in others they diverged , suggesting that they may derive from novel viruses . These results demonstrate the utility of unbiased metagenomic approaches in the detection of known and divergent viruses in the study of tropical febrile illness .
Viral infections pose a significant global health burden , especially in the developing world where most infectious disease deaths occur in children and are commonly due to preventable or treatable agents . Effective diagnostic and surveillance tools are crucial for reducing disability-adjusted-life-years ( DALYs ) due to infectious agents and for bolstering elimination and treatment programs [1] . Previously unrecognized and novel pathogens continually emerge due to globalization , climate change , and environmental encroachment , and pose important diagnostic challenges [2] , [3] . Dengue virus ( DENV ) infection is the most common arthropod-borne viral disease of humans , with an estimated 50–100 million clinical infections occurring annually worldwide [4] . DENV infection manifests clinically as dengue fever or the more severe dengue hemorrhagic fever/dengue shock syndrome ( DHF/DSS ) [4] . The increased spread of dengue virus and its mosquito vectors in many subtropical regions over the past several decades , especially in Latin America and Asia [5] , highlights the need for additional methods of dengue virus surveillance . Diagnosing dengue relies on detecting viral nucleic acid or antigens in the blood or confirming the presence of anti-DENV IgM and IgG antibodies and therefore traditionally depends on RT-PCR , ELISA , and viral cell culture methods [5]–[7] . Dengue diagnostics are of crucial importance due to its broad spectrum of clinical presentations , global emergence and spread , unique disease epidemiology , and possible clinical relation to other as-yet unknown tropical febrile pathogens . Traditional viral detection methods , such as serology , virus isolation , and PCR , are optimized for the detection of known agents [2] . However , novel and highly divergent viruses are not easily detected by approaches that rely on a priori sequence , antigen , or cell tropism knowledge . PCR-based assays that employ degenerate primers may successfully target conserved regions within related virus groups , but unlike bacteria , viruses lack universally conserved genetic regions , such as ribosomal RNA , that can be exploited to amplify all viruses [8] . Metagenomic analysis enables more systemic detection of both known and novel viral pathogens [9]–[12] and is approached through a variety of microarray and sequencing strategies [13] , [14] . The Virochip is a pan-viral microarray platform that has been previously utilized in the detection and discovery of viruses from both human and animal samples [15]–[19] . Deep sequencing and shotgun sequencing of human clinical samples has been used for viral detection [20]–[23] , novel virus discovery [24]–[27] , and divergent virus genome recovery [28] . Viral metagenomic approaches have also been employed as a diagnostic supplement to pathogen detection as part of public health monitoring systems [22] , but have been limited to shotgun sequencing of viral-enriched libraries and have yet to utilize deep sequencing data . Currently available sequencing platforms can generate millions to billions of sequencing reads per run , far exceeding large-scale shotgun sequencing [13] . Deep sequencing of clinical samples , in which hundreds of thousands to millions of sequencing reads are generated per sample , can be incorporated into stepwise virus detection pipelines [29] . Database searches using Basic Local Alignment Search Tool ( BLAST ) and other alignment tools [30] can be used to identify sequences in samples that correspond to known and novel viruses , including those present at low concentrations or deriving from viruses that may be too divergent to be detected with PCR or microarray methods . Deep sequencing represents an unbiased , highly sensitive method for identifying viral nucleic acid in clinical samples . This study describes the use of the Virochip microarray and deep sequencing for the direct viral diagnosis of serum from cases of acute pediatric febrile illness in a tropical urban setting . Patient clinical data and serum samples were collected between 2005 and 2009 as part of an ongoing pediatric dengue study in Managua , Nicaragua [31] . Virochip and deep sequencing were performed on positive control samples and on 123 dengue virus-negative serum samples . Using these methods , viruses were detected in 45 of 123 ( 37% ) previously negative samples . Sequences derived from known and apparently divergent viruses . The viruses identified in some of the cases are known to induce symptoms consistent with those observed , though the definitive causative agent of these infections remains to be determined .
Acute serum samples were collected from suspected dengue cases at the Hospital Infantil Manuel de Jesús Rivera ( HIMJR ) , the National Pediatric Reference Hospital in Managua , Nicaragua , after undergoing informed consent or the informed consent procedure . Patients were enrolled in the study if they presented with fever or history of fever less than 7 days and one or more of the following signs and symptoms: headache , arthralgia , myalgia , retro-orbital pain , positive tourniquet test , petechiae , or signs of bleeding . Patients with a defined diagnosis other than dengue , e . g . pneumonia , were excluded . Suspected dengue cases were tested for dengue virus ( DENV ) infection at the Centro Nacional de Diagnóstico y Referencia ( CNDR ) of the Nicaraguan Ministry of Health and were considered laboratory-confirmed if: 1 ) DENV was isolated , 2 ) DENV RNA was detected by reverse transcriptase-polymerase chain reaction ( RT-PCR ) , 3 ) seroconversion was observed by IgM capture enzyme-linked immunosorbent assay ( ELISA ) of paired acute and convalescent sera , or 4 ) a ≥4-fold increase in DENV-specific antibodies was demonstrated by inhibition ELISA in paired acute and convalescent sera [32] . All patients were aged 6 months to 14 years and presented between August 2005 and January 2009 . Approximately one half of the suspected dengue cases testing negative by all four dengue diagnostic assays were included in the metagenomics analysis described here . 34 cases ( pools 1–4 , see below ) corresponded to the subset of patients who presented within 4 days of symptom onset and who reported both fever or history of fever and rash . 89 of the samples ( pool 5 ) were selected randomly from among the remaining samples . As positive controls , seven samples ( pool 5 ) that had been clinically diagnosed as virus positive were included . The study protocol was reviewed and approved by the Institutional Review Boards ( IRB ) of the University of California , Berkeley , and of the Nicaraguan Ministry of Health . Total nucleic acid from 140 µl of serum was extracted using the QIAamp Viral RNA Isolation Kit ( Qiagen ) , which co-purifies RNA and DNA . End-tagged dsDNA libraries were created essentially as previously described [28] . RNA was reverse transcribed in reactions containing 1× reaction buffer , 5 mM dithiothreitol , 1 . 25 mM dNTPs , 20 pmoles primer ( 5′-CGC TCT TCC GAT CTN NNN NN-3′ ) , 100 U Superscript III ( Invitrogen ) , and ∼20 ng template . Following reverse transcription , Sequenase reaction buffer and 2 U of Sequenase DNA polymerase ( Affymetrix ) were added to samples for second strand synthesis . The Sequenase reactions were performed twice so that starting DNA templates would be converted into end-tagged library molecules . The resulting libraries were amplified by PCR using primer 5′-CGC TCT TCC GAT CT-3′ . PCRs contained 1× reaction buffer , 2 µM primer , 0 . 25 mM dNTPs , 2 U Taq DNA polymerase , and 2 µl library template . Thermocycling conditions were 95°C for 2 min; 25 cycles of 95°C for 30 sec , 40°C for 30 sec , and 72°C for 1 minute , with a final extension of 5 minutes . These libraries were further processed for microarray hybridization and deep sequencing as described below . For microarray hybridization , a fraction of each library was amplified by PCR as above but with a modified dNTP mixture including 5- ( 3-aminoallyl ) -dUTP ( Ambion ) in lieu of 75% of the dTTP normally in the mixture . The resulting amino-allyl-containing DNA was purified using a DNA Clean and Concentrator-5 column ( Zymo Research ) . The eluate was heat denatured at 95°C for 2 min , cooled briefly on ice , then fluorescently labeled in reactions containing 100 mM sodium bicarbonate pH 9 , 10% DMSO , and 667 µM Cy3 mono NHS ester ( GE Healthcare ) for 1 hour at 25°C . Labeled DNA was purified using DNA-CC-5 columns and added to hybridization reactions containing 3×SSC , 25 mM HEPES pH 7 . 4 , and 0 . 25% SDS . Hybridization mixtures were heated at 95°C for 2 minutes , applied to microarrays , and hybridized overnight at 65°C . Following hybridization , arrays were washed twice in 0 . 57× SSC and 0 . 028% SDS and twice in 0 . 057× SSC , then scanned on an Axon GenePix 4000B microarray scanner . Three analysis tools were used to analyze Virochip data: E-predict [33] , Z-score analysis [34] , and cluster analysis [35] . An array was deemed positive for a particular virus if the virus was identified by at least two of these methods . Virochip results were deposited in the NCBI GEO database ( GEO accession series: GSE28142 ) . For deep sequencing , the Illumina paired-end adapter sequences were appended to library molecules using PCR , essentially as previously described [28] . Library generation primers ( Table S1 ) were modified from adapter A and adapter B sequences ( Illumina ) . Samples were reverse transcribed and libraries were created and amplified as described above for the Virochip . Library molecules of approximately 300 bp were purified on a 4% native polyacrylamide gel , ethanol precipitated , and PCR amplified for 17 additional cycles using a 22-nt-long primer consisting of the 3′-end of Illumina adapter A ( primer 2 ) and the full-length 61-bp Illumina adapter B ( primer 4 ) under the following conditions: 2 cycles of 94°C for 30 s , 40°C for 30 s , and 72°C for 1 min , followed by 15 cycles of 94°C for 30 s , 55°C for 30 s , and 72°C for 1 min . Amplicons generated with the correct adapter topology ( one end with adapter A and the other with adapter B ) were approximately 355 bp and were separated by polyacrylamide gel electrophoresis from adapter A/A and adapter B/B amplicons , which migrate differently ( approximately 40 bp smaller or larger than the expected size ) . An additional 10 cycles of PCR were then performed using the full-length adapter sequences as primers ( primers 3 and 4 ) . Libraries were validated by Sanger sequencing before high throughput sequencing . Following validation , samples were combined into five pools for sequencing . For pools one through four reverse transcription primers included a three or four-nucleotide barcode sequence at the 3′-end . For pool five , barcodes were located internally in the adapter sequence . Each pool was sequenced on one lane of a flowcell on the Illumina Genome Analyzer II ( pools 1–4 ) or HiSeq 2000 ( pool 5 ) . Pools' 1–4 molecules were sequenced as 67 nucleotide paired ends , and pool 5 molecules as 97 nucleotide paired ends . Paired-end sequencing was performed for several reasons: ( 1 ) to double the overall amount of data generated , ( 2 ) to double the amount of sequence information per molecule , and ( 3 ) to provide anchors from which additional sequence could be recovered by subsequent PCR . In some cases , PCR and Sanger sequencing was used to confirm Virochip and deep sequencing calls and to recover additional sequence . Primer sequences are listed in Table S1 . PCR conditions were: 95°C for 2 minutes , 35 cycles of 95°C for 30 seconds , 50–60°C for 30 seconds ( primer dependent ) , 72°C for 1 minute , and 72°C for 2 minutes . PCR products were size-selected on an agarose gel , purified with the Purelink gel extraction kit ( Invitrogen ) , cloned , and Sanger sequenced . Full-length poliovirus genomic RNA was transcribed from MluI-linearized plasmid prib ( + ) XpA using T7 RNA polymerase as previously described [36] . Poliovirus RNA was mixed with HeLa total RNA in a dilution series ranging from 10−2 to 10−6 poliovirus gRNA per HeLa RNA . Randomly-primed dsDNA libraries were prepared , hybridized to the Virochip , and analyzed as described above . Predicted circovirus-like replicase sequences were searched against the NCBI non-redundant protein database ( BLASTx , E value 10−2 ) . Aligning sequences were retrieved and consolidated using CD-HIT into a set of representative sequences [37] ( CD-HIT version 4 . 5 . 4; parameters: -c 0 . 7 ) . These sequences were aligned in Geneious [38] as a global alignment with free end gaps and trimmed to the 47 amino acid overlap shared by the two recovered sequences . A neighbor-joining tree was generated by Geneious Tree Builder [38] . The initial FASTQ data from each pool's lane were binned by barcode . The barcode-split reads were trimmed of non-template deriving and potentially error-prone sequence: a randomly incorporated nucleotide ( N ) , the barcode bases , and the sequence corresponding to the random hexamer , leaving 55 ( pools 1 , 2 , and 4 ) , 54 ( pool 3 ) , or 90 ( pool 5 ) bases per read . The lowest complexity fraction was identified by sequences with LZW ratios ( compressed size/uncompressed size ) less than 0 . 45 [39] . Reads were aligned to the human genome ( build hg18 ) first using BLAT [40] with the “–fastMap” flag , and after filtering , the remaining reads were aligned using BLAT without the flag . Paired reads for which at least one of the reads in the pair had at least 80% identity to the database were marked as human and removed from subsequent analyses . After removal of reads identified as human by BLAT , remaining reads were aligned and filtered by mapping to the human transcriptome using nucleotide BLAST ( BLASTn version 2 . 2 . 21 , word size 30 , E value 10−3 ) . Remaining reads were next aligned to the human genome using BLASTn ( word size 30 , E value 10−3 ) , filtered , and again aligned to the human genome by BLASTn ( word size 11 , E value 10 ) . After all human filtering , we reanalyzed the distribution of the complexity of reads and observed a relative enrichment of reads with LZW ratios lower than 0 . 54 ( pools1–4 ) or 0 . 48 ( pool5; different LZW ratio distributions are an inherent property of different read lengths ) , and those reads were removed from further analysis . To look for reads with viral homology , we searched the non-redundant nucleotide database ( nt ) using BLASTn ( word size 20 , E value 10−3 ) . Reads that did not map to nt were aligned to the non-redundant protein database ( nr ) using translated BLAST ( BLASTx , word size 4 , E value 10 ) . In order to make specific virus-positive calls , we implemented a set of rules to minimize false positives while maintaining sensitivity . In order to reduce the number of false positive sequences that may share identity equally with both viral and non-viral genomes , we restricted our analysis to those queries whose best alignments were only to animal viral sequences . In a number of datasets , we detected human klassevirus 1 , a virus identified and studied in our lab [26] , human poliovirus , used in our Virochip sensitivity experiments , sequences from mosquito densoviruses , also studied in the lab , as well as Moloney murine leukemia virus ( MMLV ) , the polymerase of which was used in the sequence library preparation . We believe these reads represent lab contaminants , and others studies that prepared sequence libraries in the same location have reported similar findings [41] . To account for these contaminants , positive calls were only made on viruses for which there were more supporting reads than there were reads to any known contaminant . Finally , in order to avoid making calls based on potentially spurious alignments , we considered only those viruses for which there were at least 10 reads supporting their presence .
We initially screened the serum samples with the Virochip pan viral detection microarray . This was done as a complement to the deep sequencing analysis and in order to compare the sensitivity of the two approaches . We included 7 blinded positive control samples that had been previously diagnosed in the clinic as being positive for DENV-2 ( n = 4 ) , DENV-1 ( n = 1 ) , or hepatitis A virus ( HAV; n = 2 ) . The Virochip successfully identified the correct virus in all of these positive controls , and in the case of the dengue virus positive samples , the correct serotype as well ( Table 1 ) . We also identified ten samples positive for torque teno virus ( TTV ) . We applied in vitro transcribed poliovirus RNA diluted into HeLa cell total RNA to the Virochip as an additional positive control to quantify Virochip sensitivity . Using the E-predict analysis tool , the lowest detectable concentration of poliovirus was 1 viral RNA per 105 HeLa RNA molecules ( approximately 10 polio gRNAs per cell equivalent of HeLa RNA; Figure S1 ) . A total of 130 serum samples were deep sequenced , including 7 positive controls and 123 previously undiagnosed samples . We performed deep sequencing on 34 of the serum samples using the Illumina GAII platform , generating a total of 184 . 6 million 65-nucleotide long paired-end reads ( one flow cell lane each for four sample pools , 12 . 0 billion bases total , median of 3 . 7 million reads per sample ) . We sequenced 96 serum samples ( pool five ) on a HiSeq 2000 instrument , which provides more , longer sequences per run . The HiSeq run generated 196 . 4 million 97-nt sequences ( one flow cell lane , 19 billion bases total; median of 1 . 7 million reads per sample ) . The raw reads were first separated by barcode and analyzed as individual data sets as described in the Methods . The bioinformatic filtering process consisted of removing low complexity and low quality sequences , then filtering sequences of human origin ( Figure 1 ) . After the filtering steps , an average of 1 . 9% of the initial reads remained , with an absolute average of 60 , 000 reads remaining per sample ( Figure 1 and Table 1 ) . A few of the barcode datasets appeared to have a larger non-human fraction . Upon further inspection , the non-human components were accounted for by known library preparation contaminants , such as E . coli and S . cerevisiae . The reads remaining after filtering were then compared to sequences in the NCBI non-redundant nucleotide and protein databases using BLASTn and BLASTx respectively . Virus-derived sequences were detected in all 7 positive control samples and in 45/123 ( 37% ) of previously negative serum samples ( Table 1 ) . In 78/123 ( 63% ) samples , we were unable to identify virus sequence by our detection criteria ( Methods ) . We recovered virus sequences matching the expected viral genomes in all of the positive control samples . The fraction of viral sequences in the controls spanned 4 orders of magnitude , from 0 . 002% to 2 . 8% of total reads . The two HAV positive control samples ( #401 ) were aliquots of the same serum sample and were processed and analyzed independently . The fraction of viral reads in the duplicates was within 4-fold ( 0 . 4% and 1 . 2% ) . This demonstrates that our library preparation , sequencing , and bioinformatics pipeline is capable of reproducibly detecting evidence of clinically relevant infections . In addition to the controls , two non-control samples contained evidence of RNA virus sequence . Both samples had reads deriving from GB Virus C ( GBV-C , also known as Hepatitis G Virus ) and were essentially identical to GBV-C database sequences . We detected no sequences that best aligned to dsRNA viruses or to retroviruses ( except for human endogenous retrovirus and contaminating MLV RT-derived sequences , see Methods ) . Human Herpesvirus 6 ( HHV-6 ) sequence was detected in 13/123 previously negative samples ( 10 . 6% ) . The HHV-6 positive samples had an average normalized read count of 145 HHV-6 reads per sample ( range: 24–411 ) , representing 0 . 002% to 0 . 02% of the datasets ( Table 1 ) , and all of these reads possessed high sequence identity to the HHV-6B reference genome sequence ( gi: 9633069 ) . We generated alignments to the reference genome to investigate the depth and genomic position of the sequence coverage across the HHV-6 genome ( Figure 2 ) . Although the reads only constitute a relatively small fraction of each dataset , there is coverage across the entire genome and over many genes in most of the HHV-6 positive samples . In addition to HHV-6 , we detected Human Herpesvirus 4 ( HHV-4 , also known as Epstein Barr Virus ) sequences in one sample . As with HHV-6 , The HHV-4 sequences were virtually identical to previously reported sequences . One sample also contained reads similar to another dsDNA virus , African Swine Fever Virus ( ASFV ) , which has been previously detected in human serum [42] . In this case , the reads best matched ASFV capsid sequences and were relatively divergent ( 47–51% amino acid identity; no similarity to non-ASFV sequences by BLASTx ) . Attempts to recover additional ASFV sequence by PCR were unsuccessful . We also identified sequences derived from single-stranded DNA viruses in some samples . In one sample we detected Parvovirus B19-derived reads with high identity to database sequences . Sequences related to various members of the Anelloviridae virus family ( TTVs ) were detected in 21 ( 17% ) samples . This frequency of detection is within the range reported previously for human serum [43] , [44] . The TTV sequences ranged from 40–97% amino acid identity to their closest database matches . We did not pursue these sequences further , because TTVs are known to form a divergent family of viruses and are commonly detected in apparently healthy individuals . Sequences similar to members of the Circoviridae family of ssDNA viruses were detected in 13/123 samples ( 10 . 6% ) . All of the sequences aligned to circovirus or circovirus-like replicase protein sequences . The alignments ranged from 36–84% amino acid identity , and appeared to derive from the replicase genes from multiple related species ( Table 1 ) . Circovirus-like replicase sequences have been detected in human stool , animals , and environmental samples [45]–[48] . We detected a range of 12 to 205 circovirus-like reads per positive sample ( Table 1 ) . The low sequence coverage prohibited complete genome sequence assembly but informed sequence-specific primer design , from which we were often able to recover larger continuous regions of the replicase genes by PCR and Sanger sequencing ( GenBank accessions JF781513 , JN837698 , and see Table S2 ) . We termed the extended replicase-like sequences Circovirus-like NI/2007 1–3 ( Cvl-NI 1–3 ) , and compared them to a representative set of other replicase sequences ( Figure 3 ) . The Cvl-NI-1 sequence is most closely related to Circovirus-like virus RW-E ( gi: 254688530 ) , a circular single-stranded DNA virus previously found in reclaimed water samples in Florida [45] . The Cvl-NI-2 sequence is most closely related to a replicase sequence recovered from bat feces in Yunnan Province , China ( gi: 342356307 ) [48] . The Cvl-NI-3 sequence did not overlap with the other sequences enough to be included in the phylogenetic analysis , but was most similar to Circovirus-like CB-A , a circovirus-like genome identified in a Chesapeake Bay environmental sample ( gi: 229562105 ) [45] . A subset of the positive samples ( Table 1 ) contained sequences from more than one virus , which may be evidence of co-infection . Almost all of the cases with multiple viruses involved TTV-derived sequences along with HHV-6 , DENV-2 , or circovirus-like sequences ( samples 282 , 235 , 183 , 270 , 350 , and 377 ) . Two samples contained circovirus-like sequences with ASFV-like ( sample 315 ) or GBV-C sequences ( sample 387 ) .
In this study , we examined the virus diversity in serum samples from Nicaraguan children with unknown acute febrile illness . We performed Virochip microarray and deep sequencing analyses on 7 positive control and 123 undiagnosed samples . Both of these methods succeeded in detecting the expected virus in the positive control samples . Virochip analysis produced putative viral hits in 10/123 ( 8% ) of the previously negative samples , whereas deep sequencing revealed virus or virus-like sequences in 45/123 ( 37% ) . This study demonstrates the utility of these metagenomic strategies to detect virus sequence in multiple human serum samples and is the first to utilize second-generation sequencing to simultaneously investigate many cases of acute unknown tropical illness . Monitoring the emergence and spread of novel human pathogens in tropical regions is a central public health concern . Metagenomic analysis enables more systemic viral detection of both known and novel viral pathogens [42] and can be employed as diagnostic supplements to pathogen detection as part of public health monitoring systems and epidemiologic surveys [9]–[12] , [15]–[17] , [19] , [21] , [23] . Despite the headway , metagenomic virus detection studies will have to confront several remaining difficulties concerning diagnostic accuracy . Foremost concerns include enhancing the sensitivity and specificity of deep sequencing-based diagnostic methods and re-evaluating the evidence for disease causality in light of increasingly sensitive nucleic acid detection and pathogen discovery methods . The former will require improved strategies to biochemically enrich and computationally identify viral sequences while reducing host background sequences . The latter will require a cautious reconsideration of criteria used to establish causal links between microbes and disease , as well as extensive case-by-case follow-up studies employing classical laboratory methods , such as serological analysis and cell culture amplification . It is important to highlight that observing viral sequence in sequencing data is insufficient to establish the role of a virus in disease causality . Like other detection strategies , deep sequencing will serve to inform secondary tests , including seroconversion assays , further nucleic acid testing , cell culture amplification , and additional investigations into plausible disease mechanisms . We detected virus sequence at concentrations as low as ∼2 in 106 reads . Virus sequence detected in a clinical sample at vanishingly low copy numbers may reflect several possible host-microbe scenarios . The sequence detected may be that of a pathogenic virus capable of causing illness at low copy number or through indirect effects , a ubiquitous non-disease causing microbe , a virus outside of its primary replication site , low-level contamination , an artifact of sample collection timing/processing , or remains of incomplete immune clearance . Additional evidence must be considered in each case to define the host-microbe relationship . In this study , we compared the performance of the Virochip and deep sequencing for detecting virus sequence in human serum . The limit of detection of the Virochip was approximately one part in 105 for the poliovirus controls , for which there are microarray probes with perfect sequence complementarity ( Figure S1 ) . The sensitivity of deep sequencing is limited by the number of reads generated per sample , or read depth . In this study , we detected virus sequences down to two parts per million . Nearly every virus that was detected on the microarray was also detected by deep sequencing; additionally , in numerous samples ( n = 44 ) , sequencing revealed viruses not detected by the Virochip ( Table 1 ) . There were two instances where Virochip analysis identified a virus ( TTV ) that was not detected by deep sequencing ( Table 1 ) . Deep sequencing , therefore , is a superior method for novel virus discovery , because it is more sensitive and provides more conclusive genotypic information than the Virochip . Nevertheless , the Virochip is a relatively fast and inexpensive method that is best applied to samples with expected virus copy numbers present at levels greater than 1 in 105 host sequences . We were unable to detect a virus in two thirds of the 123 dengue-like illness samples . These results could reflect true negative status , which would result from a non-viral infection , illness due to non-infectious agent , or complete immunologic clearance . Alternatively , the negative results could reflect failures in our diagnostic approaches due to imperfect sensitivity , unsatisfactory sample preparation , improper sample type , or failure to recognize highly divergent viral sequences . The presence of sequences that lack even remote similarities to known species also highlights the need for further development of de novo assembly methods for metagenomic data . Assembled data , increased depth , and enhanced sequenced comparison methods should enable more sensitive detection of divergent viruses in metagenomic samples . Determining the etiology of human diseases with symptoms that overlap with dengue-like illness is important for understanding the full spectrum of emerging or previously uncharacterized pathogens in tropical populations . In this study , 10% of acute serum samples negative for dengue virus from cases of pediatric dengue-like illness were positive for HHV-6 . Primary HHV-6 infection causes undifferentiated febrile illness and exanthem subitum ( roseola infantum or sixth disease ) , an acute illness with high fever and rash that typically resolves in three to seven days [49] . Exanthem subitum is a common disease of infants worldwide , and HHV-6 infection most frequently occurs between 6 and 12 months of age [50] , with seropositivity estimates of >95% in adult populations in developed countries [51] . The HHV-6 positive patients in this study were between 7–12 months old , and presented with fever and rash ( Table S3 ) . We detected multiple kilobases of HHV-6 sequence in each positive sample , with sequence deriving from multiple viral genomic regions ( Figure 2 ) . After acute infection , HHV-6 can latently persist in the host quiescently , with no production of infectious virions or with low levels of viral replication . Latency is believed to endure in several cell types , including monocytes and bone marrow progenitor cells [52] , [53] , and may undergo chromosomal integration that can be vertically transmitted [54] . The confounding effects of chromosomal integration make differentiating between active and latent HHV-6 infections difficult when detecting HHV-6 sequence in serum DNA [55] , [56] . A previous study detected integrated HHV-6 genomic sequence in ∼1% of healthy blood samples [57] . Since detection of HHV-6 nucleic acid in serum alone does not prove active viral infection , we cannot definitively confirm that the HHV-6 sequences in these samples were not derived from the vertical transmission of chromosomally integrated virus . However , the clinical , epidemiological , and virus sequence data suggest HHV-6 may be the etiologic agent in these febrile illness cases . Primary HHV-6 infection is a major cause ( ∼20% ) of infant hospitalizations in the United States [58] , a clinical burden likely shared throughout the tropical world given similar seroprevalence rates [59] . The results of this study illustrate the importance of administering HHV-6 diagnostic tests to cases of suspected dengue-like illness in infants from dengue-endemic regions to differentiate between cases of exanthem subitum , a ubiquitous self-limiting childhood illness , and dengue fever , which carries a greater risk of severe clinical complications and death . Similarly , the one sample positive for Parvovirus B19 sequence may be a case of acute infection with a commonly acquired childhood virus . Parvovirus B19 can manifest as erythema infectiosum ( fifth disease ) , a condition associated with characteristic “slapped cheek” rash [60] . Infection can also be subclinical or result in mild nonspecific symptoms . It is possible that Parvovirus B19 infection caused the symptoms in this case ( Table S3 ) , though as with HHV-6 , the identification of viral sequences does not definitively demonstrate causality . Epstein Barr Virus ( HHV-4 ) sequences were found in the serum of one patient who presented with relatively severe symptoms , and died during hospitalization ( Table S3 ) . HHV-4 infection is a nearly universal occurrence in the first two decades of life [61] , [62] . Primary infection in adolescents or adults can manifest as infectious mononucleosis , and chronic infection is associated with various malignancies later in life . Primary infection during childhood , however , is usually asymptomatic or produces only mild symptoms . It is not clear that HHV-4 infection or HHV-4 alone caused the illness in this case . In addition to the viruses for which a plausible disease association exists , many samples contained sequences from viruses with no well-established link to human disease . These included the two samples positive for GBV-C and those containing ASFV-like , TTV-like , and circovirus-like sequences . The Circoviridae family is an extraordinarily diverse group of small , single-stranded circular DNA viruses that includes cycloviruses ( genus Cyclovirus ) and circoviruses ( genus Circovirus ) , which are commonly detected in human stool and blood , and also in environmental samples [43]–[48] . Some circovirus species , such as beak and feather disease virus and porcine circovirus 2 , have been associated with disease in bird and pig hosts , respectively , but the pathogenic potential of circoviruses in humans remains unconfirmed [63] , [64] . The circovirus-like sequences reported here were detected in nucleic acid libraries prepared from acute human serum and were most closely related to circovirus-like viruses ( Figure 3 ) , which were first reported in environmental samples and in bats [45] , [48] . We were unsuccessful in recovering a full genome sequence corresponding to any of the circovirus-like sequences , and it has not yet been possible to prove that these sequences were not an environmental artifact introduced during sample preparation . It is also possible that these sequences derive from other organisms , such as Giardia intestinalis or Entamoeba dispar , whose genomes encode proteins that share amino acid similarity with circovirus replicase proteins ( Figure 3 ) . Furthermore , it has yet to be established whether circoviruses are capable of replicating in humans . Pending additional screening and serologic studies , the detection of circovirus-like sequences from human serum should be interpreted with caution . Metagenomic approaches provide an effective high-throughput method to detect uncharacterized virus diversity in a tropical setting from many samples simultaneously . The findings presented in this study further our knowledge of well-characterized and previously unknown viruses present in serum collected from pediatric dengue-like illness patients and advance our understanding of the application of metagenomic approaches to human pathogen detection . Deep sequencing analysis of clinical samples holds tremendous promise as a diagnostic tool by permitting the detection of many different viruses simultaneously , including those present at low-copy numbers and of divergent origin . Major remaining barriers to high-throughput sequencing strategies becoming standard diagnostic practice include prohibitive cost , lengthy sample preparation time , and computationally intensive data analysis requirements . These challenges are magnified in resource-limited settings , such as Nicaragua , but are gradually being addressed . Industry hardware and technical advancements have steadily decreased the per-base cost of deep sequencing , and the results presented here strengthen our expectations of multiplexed sample preparation and bioinformatic data filtering within the framework of current second-generation sequencing platforms . Long-term bi-directional partnerships with developing country collaborators facilitate easier access to techniques not currently available on-site , such as deep sequencing , and are also important in providing training opportunities for local scientists and developing relevant pathogen tests and diagnostic policies . This study expands our understanding of the virus diversity in pediatric dengue-like illness in Nicaragua and the application of genomic detection techniques in a tropical setting , findings that are particularly valuable given the pressing need for improved global emerging pathogen surveillance . | Dengue virus infection is a global health concern , affecting as many as 100 million people annually worldwide . A critical first step to proper treatment and control of any virus infection is a correct diagnosis . Traditional diagnostic tests for viruses depend on amplification of conserved portions of the viral genome , detection of the binding of antibodies to viral proteins , or replication of the virus in cell cultures . These methods have a major shortcoming: they are unable to detect divergent or novel viruses for which a priori sequence , serological , or cellular tropism information is not known . In our study , we use two approaches , microarrays and deep sequencing , to virus identification that are less susceptible to such shortcomings . We used these unbiased tools to search for viruses in blood collected from Nicaraguan children with clinical symptoms indicating dengue virus infection , but for whom current dengue virus detection assays yielded negative results . We were able to identify both known and divergent viruses in about one third of previously negative samples , demonstrating the utility of these approaches to detect viruses in cases of unknown dengue-like illness . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"virology",
"global",
"health",
"neglected",
"tropical",
"diseases",
"biology",
"computational",
"biology",
"microbiology"
] | 2012 | Virus Identification in Unknown Tropical Febrile Illness Cases Using Deep Sequencing |
Dynamic balance of excitation and inhibition is crucial for network stability and cortical processing , but it is unclear how this balance is achieved at different membrane potentials ( Vm ) of cortical neurons , as found during persistent activity or slow Vm oscillation . Here we report that a Vm-dependent modulation of recurrent inhibition between pyramidal cells ( PCs ) contributes to the excitation-inhibition balance . Whole-cell recording from paired layer-5 PCs in rat somatosensory cortical slices revealed that both the slow and the fast disynaptic IPSPs , presumably mediated by low-threshold spiking and fast spiking interneurons , respectively , were modulated by changes in presynaptic Vm . Somatic depolarization ( >5 mV ) of the presynaptic PC substantially increased the amplitude and shortened the onset latency of the slow disynaptic IPSPs in neighboring PCs , leading to a narrowed time window for EPSP integration . A similar increase in the amplitude of the fast disynaptic IPSPs in response to presynaptic depolarization was also observed . Further paired recording from PCs and interneurons revealed that PC depolarization increases EPSP amplitude and thus elevates interneuronal firing and inhibition of neighboring PCs , a reflection of the analog mode of excitatory synaptic transmission between PCs and interneurons . Together , these results revealed an immediate Vm-dependent modulation of cortical inhibition , a key strategy through which the cortex dynamically maintains the balance of excitation and inhibition at different states of cortical activity .
The excitatory and inhibitory inputs received by cortical neurons are normally under dynamic balance during cortical functions [1]–[7] . The interaction among these inputs , together with intrinsic membrane properties of cortical neurons , often results in shifts of membrane potential ( Vm ) between different states [8]–[11] , which could regulate neuronal responsiveness to synaptic and sensory inputs [1] , [12]–[17] . However , it is unclear how the balance of excitation and inhibition is achieved when cortical neurons are at different Vm levels . During global membrane oscillations involving a large number of cortical neurons , excitation and inhibition may be proportionally altered by the Vm shift , but the underlying mechanisms remain unknown , in view of diversity of connectivity and functions of local inhibitory interneurons [18] , [19] . In the case of persistent activities associated with some behaviorally relevant conditions , e . g . , during working memory , a subpopulation of neurons undergoes changes in the Vm [20]–[24] . Microcircuits involving these active neurons also require dynamic control of their excitation-inhibition balance . In this study , we investigated how Vm changes of a cortical neuron may modulate the efficacy of recurrent inhibition within the microcircuit . It has been shown recently that cortical excitatory neurons communicate not only through the generation of all-or-none action potentials ( APs , digital mode ) but also through a presynaptic Vm-dependent modulation of transmitter release ( analog mode ) [25] , [26] . It remains unknown to what extent the analog-mode communication influences the operation of local circuitry and has a functional role in the cortex . Considering that interneurons within the microcircuit are driven by excitatory neurons , leading to recurrent inhibition , we hypothesized that the amount of recurrent inhibition might be subjected to modulation in a manner that depends on the level of depolarization of the excitatory neuron . The Vm changes in the presynaptic excitatory neuron may modulate the size of excitatory postsynaptic potentials ( EPSPs ) in interneurons , leading to changes in their AP firing , which in turn alter the efficacy of their inhibition on neighboring neurons . Cortical inhibitory interneurons show a huge diversity in their biochemical and physiological properties [18] , [19] . Two distinct subtypes of interneurons , low-threshold spiking ( LTS ) neuron and fast spiking ( FS ) neuron , mediate the slow and the fast recurrent inhibition , respectively [27]–[29] . The LTS neuron receives EPSPs that show facilitation in response to a train of high-frequency stimuli of the presynaptic PC [30]–[32] , generating APs with a long onset latency and evoking late-onset ( slow ) disynaptic IPSPs in its neighboring PCs [27] , [29] , [33] . The FS neuron ( and some other inhibitory interneurons ) receives EPSPs that show depression during high-frequency presynaptic stimulation [31] , [34] . However , the high release probability of PC-FS synapses often allows discharges of the FS neuron in response to single APs in the PC , leading to time-locked early onset ( fast ) IPSPs in neighboring PCs [29] , [35] . In this study , we sought to examine whether the slow and the fast recurrent inhibition meditated by these two distinct microcircuits ( PC-LTS-PC and PC-FS-PC ) were subjected to modulation in response to Vm changes of the presynaptic PC . We performed paired whole-cell recording ( PC-PC , PC-LTS , and PC-FS ) in rat somatosensory cortical slices and found that both the slow and the fast recurrent inhibition were indeed modulated by the presynaptic somatic Vm changes , and this modulation resulted from analog-mode signaling in excitatory synapses between PCs and interneurons . These results show an important role of analog communication in controlling the operation of cortical microcircuits and provide new insights into the cellular mechanisms underlying dynamic balance of cortical excitation and inhibition .
We performed paired whole-cell recording from nearby layer-5 pyramidal cells ( PCs , <100 µm apart ) in acutely isolated rat somatosensory cortical slices . In response to stimulation of the PC with a burst of APs ( 70∼200 Hz ) , disynaptic IPSPs were observed in 19% of the pairs successfully tested ( 207/1 , 087 pairs ) , with 2% ( 23/1 , 087 ) exhibiting reciprocal IPSPs ( see Methods , Figure 1A ) . Consistent with previous studies [27] , [29] , these IPSPs had a peak amplitude of 1 . 3±0 . 1 mV ( s . e . m . , n = 38 PC-PC pairs ) and a long but rather precise onset latency ( 111±4 ms ) following PC stimulation ( 15 APs at 100 Hz ) . The IPSPs were detected only when the presynaptic PC fired at a frequency higher than 50 Hz . Bath application of either CNQX ( 10 µM ) or picrotoxin ( 50 µM ) completely abolished these IPSPs ( n = 8/8 PC pairs ) , consistent with the involvement of both excitatory and inhibitory transmission in these disynaptic responses . In our PC-PC paired recordings , a single AP or a burst of APs in one PC could evoke monosynaptic EPSPs but never triggered firing in postsynaptic PCs , suggesting no polysynaptic events involved in generating the disynaptic IPSPs . To examine the Vm-dependence of this slow ( late-onset ) recurrent inhibition , we manipulated Vm by injecting repetitive step-depolarizing currents ( duration ∼45 s , at 90 s intervals ) to induce subthreshold depolarization in the presynaptic PC . The magnitudes of disynaptic IPSPs were measured by applying trains of current pulses ( duration: 1 ms ) to evoke AP bursts ( 15 APs at 100 Hz ) in the presynaptic PC at the resting and depolarized Vm ( Figure 1B and 1C ) . In the majority of PC-PC pairs tested with this protocol ( n = 38 pairs ) , we found that step presynaptic depolarization from a resting Vm ( −64 . 6±0 . 6 mV ) to a level near the firing threshold ( −47 . 2±0 . 6 mV ) significantly increased the average amplitude ( n = 25/38 PC pairs tested , p<0 . 05 ) and integrated voltage area ( mV×s , n = 28/38 , p<0 . 05 ) of the disynaptic IPSPs and decreased the average onset latency and jitter ( n = 23/38 , p<0 . 05 ) . Cumulative frequency distribution of the average amplitude ( or total voltage area , unpublished data ) showed a highly significant difference between the two Vm levels ( n = 38 pairs , Figure 1D , p<0 . 01 , Kolmogorov-Smirnov test ) . In these experiments , we noted that the extent of modulation of these IPSPs varied greatly from pair to pair . Comparison of the IPSP amplitude ( or total voltage area ) and onset latency at two different Vm levels of the same pairs also showed highly significant differences ( Figure 1D and 1E , p<0 . 01 , t test ) . Interestingly , the percentage increase in the peak amplitude ( or total voltage area ) of IPSPs found at the depolarized Vm levels decreased with increasing average peak amplitude . As shown in Figure 1F , the data could be well fitted by a hyperbolic function ( y = 100%+0 . 67/x ) , which predicts a linear relationship between the IPSP amplitudes at depolarized Vm and those at resting Vm ( y = x+0 . 53 , with the slope fixed at 1 , see Figure S1 and Discussion ) . In 12 PC-PC pairs , we varied the time difference between the depolarization onset and the first AP burst during the depolarization in order to examine the time course of the facilitation and found that the IPSP amplitudes progressively increase after depolarization with a time course τ of 3 . 5 s ( single exponential fit , Figure 1G ) , consistent with the slow component of the EPSP facilitation induced by presynaptic depolarization found in monosynaptically connected PC pairs [26] . Further analysis revealed that the disynaptic IPSP facilitation could be attributed in part to a decrease in its failure rate . In 33/35 PC pairs tested , IPSP failures occurred at both depolarized ( by 18 . 0±0 . 8 mV ) and resting presynaptic Vm , but the average failure rate was lower under depolarized Vm ( 0 . 08±0 . 02 ) than that at resting Vm ( 0 . 25±0 . 03 ) . Among all experiments ( n = 123 PC pairs ) , nine showed complete failure of disynaptic IPSPs at resting Vm but detectable IPSPs at depolarized Vm ( Figure 2A–C ) , suggesting that such modulation could not only change but also turn on recurrent inhibition . This abrupt appearance of disynaptic IPSPs and the shortened onset latency associated with presynaptic depolarization may narrow the time window of the integration of EPSPs . Indeed , in PC connections that had both monosynaptic EPSPs and disynaptic IPSPs , the EPSP summation time was shorter at depolarized Vm in comparison with that at resting Vm ( 136±9 ms versus 175±13 ms , p<0 . 01 , n = 11 , Figure 2D–G ) . We next examined a range of presynaptic Vm in PC-PC pairs that showed IPSP facilitation to determine the threshold depolarization for inducing facilitation and the Vm-dependence of facilitation ( Figure 3A–B ) . None of the nine connections tested showed facilitation when the presynaptic PC was depolarized by only 3–5 mV . However , 5–10 mV depolarization resulted in IPSP facilitation in 24% ( n = 4/17 ) of the connections tested . The percentage of pairs exhibiting facilitation increased to 100% for depolarization more than 20 mV ( n = 11/11 , Figure 3C ) . The IPSP amplitudes also increased and their onset latencies decreased with increasing depolarization of the PC , as compared to those observed at the resting Vm . As shown in Figure 3D–E , for pairs that exhibited disynaptic facilitation ( n = 20 ) , the average IPSP amplitude ( including failures ) increased progressively with presynaptic depolarization ( r = 0 . 95 , 5 . 7% per mV ) , whereas the onset latency decreased accordingly ( r = −0 . 80 , 0 . 8% per mV ) . Again , the increase in IPSP amplitude was partially due to the decrease of failure rate ( Figure 3F ) . Together , these results show that a relatively small Vm shift of 5–10 mV of the presynaptic PC can alter the amplitude and the onset latency of disynaptic IPSPs received by its neighboring neurons , indicating a robust Vm-dependent modulation of slow recurrent inhibition . We next investigated the mechanisms underlying this Vm-dependent modulation . The late-onset disynaptic IPSP between excitatory PCs is known to be mediated by LTS interneurons [27] , [29] . In PC-LTS paired recordings , a train of high-frequency APs in the PC results in facilitating EPSPs and AP generation in the LTS neuron , which in turn triggers IPSPs in the PC [27] , [29] . We therefore examined whether the Vm changes in the presynaptic PC could modulate the magnitude of summated EPSPs and discharge probability of its postsynaptic LTS neuron . In PC-LTS paired recordings ( Figure 4A , see also Figure S2A–D ) , a burst of APs ( 15 APs at 100 Hz ) initiated in the presynaptic PC caused significant synaptic facilitation that triggered spiking of the LTS cell ( Figure 4B–C ) . Consistent with previous studies [27] , [29] , the spiking probability and the onset of LTS spiking depended on the number and the frequency of presynaptic APs ( unpublished data ) . To reveal the effect of presynaptic Vm on the summated EPSPs , we hyperpolarized the LTS cell to prevent its spiking . Steady depolarization of the presynaptic PC from the resting Vm to a level near the firing threshold significantly increased the peak amplitude of the summated EPSPs ( from 4 . 4±0 . 7 to 5 . 2±0 . 9 mV , n = 18 , p<0 . 01 ) and the total integrated area associated with the EPSPs ( from 0 . 6±0 . 1 to 0 . 8±0 . 2 mV×s , p<0 . 01 ) . Close examination of the individual EPSPs revealed that the failure rate of the 2nd to 5th EPSPs was significantly lower when presynaptic PC was depolarized , as compared to that found at the resting Vm ( p<0 . 05 , Figure 4D ) . Peak amplitudes ( measured from the baseline before the train ) of individual EPSPs during the train were also significantly increased ( Figure 4E , Figure S3 ) , as reflected by the increased slope in the plot of normalized EPSP amplitude versus the AP number ( from 0 . 12 to 0 . 15 per AP , n = 18 PC-LTS pairs ) . Next , we compared the spiking probability of the postsynaptic LTS cell before and after the presynaptic Vm shift from resting to depolarized levels . Summated EPSPs evoked by trains of presynaptic APs ( 15 APs at 100 Hz ) triggered AP generation in some of the LTS neurons recorded at resting Vm ( n = 6/22 PC-LTS pairs; Figure 4B–C and Figure 5 ) . In these six pairs that exhibited LTS spiking , the shift of Vm in the presynaptic PC from resting to a level near the firing threshold resulted in an increase in numbers of APs per trial in LTS cells ( from 2 . 0±1 . 0 to 2 . 5±1 . 1 , p<0 . 05; Figure 5A–C ) and a decrease in the onset latency ( from 113 . 3±16 . 4 to 99 . 7±13 . 7 ms , p<0 . 05; Figure 5A–B and D ) and jitter of spiking ( from 25 . 2±3 . 7 to 21 . 2±2 . 9 ms , p = 0 . 055 ) . Similar results could be obtained even when presynaptic PC was stimulated at low intensities ( 10 APs at 20 Hz , Figure S4 ) . Plot of the onset latency of LTS spiking and disynaptic IPSPs ( shown in Figure 1E ) at depolarized Vm as a function of that at resting Vm revealed a close correlation between them ( Figure 5E ) , indicating that the decrease in IPSP latency resulted from the early spiking of LTS cells . Taken together , these results correlate well with the findings on the Vm-dependent modulation of slow disynaptic IPSPs described above ( see Figure 1 ) and indicate that PC depolarization may recruit more LTS cells and/or trigger more firing in these cells , thus causing more inhibition in their neighboring PCs . In these experiments , we also found that single APs in LTS cells evoked IPSPs in PCs with a failure rate of 0 . 19±0 . 04 and an average amplitude of −0 . 43±0 . 14 mV ( n = 19 LTS-PC pairs ) at a holding potential of approximately −50 mV . Consistent with their distal input location at the apical dendrite , these IPSPs had a reversal potential of −80 . 2±1 . 9 mV ( n = 6 pairs ) . The rise time and the decay time constant ( τ ) were 7 . 25±0 . 82 and 58 . 6±8 . 3 ms , respectively ( n = 16 pairs ) . These basic kinetics were different from those of FS-PC IPSPs ( rise , 4 . 76±0 . 64 ms; decay , 85 . 2±15 . 5 ms; n = 28 pairs ) , which had a reversal potential of −70 . 8±1 . 4 mV ( n = 4 pairs ) . In 13/52 LTS-PC pairs tested , we observed reciprocal connections ( Figure 4A ) , suggesting that the Vm-dependent modulation of disynaptic IPSPs could directly influence the feedback inhibition of the presynaptic PC , in addition to the inhibition of other downstream PCs . To investigate whether the fast disynaptic inhibition mediated by interneurons that receive depressing excitatory inputs is also Vm-dependent , we analyzed the PC-PC pairs that exhibited fast ( early onset ) IPSPs in response to a single presynaptic AP . Consistent with previous reports [27]–[29] , we found that the probability of success in detecting fast disynaptic IPSPs was very low . Only 7/1 , 103 PC-PC pairs tested bi-directionally exhibited fast disynaptic IPSPs , and three out of these seven pairs showed Vm-dependent modulation . Presynaptic depolarization of ∼18 mV from the resting Vm ( −63 mV ) substantially reduced the failure rate of evoking IPSPs in these three pairs ( from 0 . 78 to 0 . 67 , 0 . 85 to 0 . 78 , and 0 . 63 to 0 . 36 , respectively; Figure 6A–B ) . This reduced failure rate is consistent with our hypothesis that the depolarized Vm in the PC elevated the EPSP amplitude and increased the spiking probability of the FS cell . When the average amplitude of IPSPs ( failure excluded ) evoked by single APs was measured , we found that it was unchanged in the first two pairs ( from 1 . 42±0 . 05 to 1 . 44±0 . 04 mV , p = 0 . 4; from 2 . 0±0 . 1 to 2 . 1±0 . 1 mV , p = 0 . 2 ) . This supports the notion that disynaptic modulation was mainly due to changes in the firing probability of the interneurons . However , we found surprisingly that the average IPSP amplitude observed in the third pair was significantly increased from 0 . 62±0 . 02 to 0 . 75±0 . 02 mV ( p<0 . 01 ) even with failure excluded . Given the low probability of observing fast disynaptic IPSPs ( n = 7/1 , 103 PC-PC pairs ) , the recruitment of two interneurons in this case is unlikely . However , it is possible that triggering of two APs ( instead of one ) in the interneuron during presynaptic depolarization could account for the remaining increase in IPSP amplitude after failure exclusion . The precise mechanism for these observations remains to be further examined . The interneurons that mediate these fast disynaptic IPSPs are most likely FS neurons , which receive depressing EPSPs in response to high-frequency presynaptic stimulation [31] , [34] . We therefore recorded PC-FS pairs to examine whether EPSPs are indeed subjected to modulation by the Vm levels of PCs . Consistent with previous findings [25] , [26] , presynaptic depolarization of 15–20 mV significantly increased the average amplitude of the single AP-triggered EPSPs ( single AP at 1 Hz ) in about one-third of the pairs tested ( n = 12/35 PC-FS pairs , Figure 6C–D ) . These elevated EPSPs may increase the spiking probability of FS cells , thus reducing the failure rate in evoking IPSPs in their postsynaptic PCs . Together , these results show that fast recurrent inhibition is also Vm-dependent , resulting from the Vm-dependent analog signaling in excitatory synapses between PCs and interneurons . Since cortical states affect both PCs and inhibitory interneurons [8] , [11] , we next examined whether this Vm-dependent analog signaling occurs at inhibitory synapses . In the FS-PC and LTS-PC pairs that showed monosynaptic inhibitory connections , we depolarized the presynaptic FS or LTS cells from resting Vm ( ∼−70 mV ) to a level near the firing threshold and found that monosynaptic IPSPs evoked by single APs ( using similar protocol as that shown in Figure 6A , also see [26] ) showed no significant change in basic kinetics of IPSPs ( FS-PC: rise time from 4 . 76±0 . 64 to 4 . 16±0 . 50 ms and decay τ from 85 . 2±15 . 5 to 62 . 0±3 . 6 ms , n = 28 pairs; LTS-PC: rise time from 7 . 25±0 . 82 to 7 . 04±0 . 72 ms and decay τ from 58 . 6±8 . 3 to 60 . 9±7 . 4 ms , n = 16 pairs; p>0 . 05 ) ; however , the average amplitude of IPSPs was significantly enhanced in a small subpopulation of tested pairs . The percentages of FS-PC and LTS-PC pairs that showed IPSP facilitation in response to presynaptic depolarization were 17 . 2% ( n = 5/29 ) and 10 . 5% ( 2/19 ) , respectively , which were smaller than that for PC-PC pairs ( 37 . 0% , n = 10/27 ) exhibiting EPSP facilitation ( Figure 7 ) . The lower probability of finding Vm modulation in LTS-PC pairs than in FS-PC pairs may result from the differences in the location of the inhibitory synapses . The LTS cells send their axons to superficial layers and form synapses onto the distal apical dendrites of PCs , while FS cells mainly target the perisomatic region of PCs . Thus , Vm changes at the soma may decay more substantially when arriving at axon terminals in LTS cells than FS cells to influence synaptic transmission . Taken together , these results demonstrate that monosynaptic IPSPs are also subjected to modulation by Vm changes in presynaptic interneurons in a small subpopulation of inhibitory connections . The above experiments showed that both fast and slow recurrent inhibition were subjected to modulation by Vm changes in PCs , resulting from the Vm-dependent analog-mode signaling in PC-interneuron excitatory synapses . A rapidly activating but slowly inactivating axonal K+ current , known as D-current [36] , has been shown to regulate the axonal AP duration and potentially contribute to the Vm-dependent modulation of the EPSP amplitude [37] , [38] . We thus further investigated the role of axonal D-currents in Vm-dependent modulation of recurrent inhibition in cortical microcircuits . Consistent with previous findings , bath application of α-dendrotoxin ( α-DTX , 100 nM ) blocked the PC depolarization-induced facilitation of the summated EPSPs in LTS cells . In six PC-LTS pairs , depolarization of ∼20 mV in the PC caused a significant increase in the peak amplitude ( 126 . 9%±6 . 2% of the control , p<0 . 05 ) and the integrated area ( 132 . 3%±7 . 0% , p<0 . 05 ) of the summated EPSPs evoked by 15 APs at 100 Hz , and this increase was blocked in the presence of α-DTX ( peak amplitude , 95 . 0%±3 . 5%; voltage area , 100 . 0%±3 . 4%; Figure 8A–B ) . In PC-PC pairs , we found that α-DTX application significantly increased the amplitude and integrated area of the disynaptic IPSP to 187%±14% ( p<0 . 01 ) and 211%±24% ( p<0 . 01 ) , respectively , and shortened the onset latency to 86 . 1%±8 . 1% ( p<0 . 05 , n = 6 ) . These effects suggest that D-current inhibition is sufficient to facilitate disynaptic IPSPs . Furthermore , after α-DTX application , 20 mV depolarization of the presynaptic PC had no significant additional facilitation in the amplitude ( 89 . 7%±7 . 1% , p = 0 . 31 ) , total integrated area ( 73 . 7%±6 . 9% , p = 0 . 9 ) , and onset latency ( 99 . 2%±3 . 4% , p = 0 . 46 , Figure 8C–D ) of disynaptic IPSPs , indicating that the effect of α-DTX had occluded that of Vm changes . Similar results were obtained with the application of a low concentration of 4-AP ( 50 µM , Figure S5 ) . These results support the hypothesis that inhibition of axonal D-current in the presynaptic PC mediates the Vm-dependent modulation of recurrent inhibition .
Recent studies have discovered that the PC-LTS-PC microcircuit mediates the slow ( late-onset ) recurrent inhibition in somatosensory and other cortices [27] , [29] , [39] . The somatostatin-positive LTS interneuron is the key player in this microcircuit . It receives EPSPs that show facilitation in response to a train of presynaptic stimuli that may initiate firing of APs , which in turn evoke IPSPs in its postsynaptic PCs [27] , [29] , [33] . Generation of these disynaptic IPSPs depends on the number and frequency of APs in PCs . Our results show that the presynaptic Vm is also a powerful determinant for controlling the strength and timing of disynaptic IPSPs—a few mV depolarization ( >5 mV ) can cause substantial IPSP facilitation ( Figure 3 ) . More importantly , stronger depolarization could not only modulate the amplitude of existing disynaptic IPSPs but also turn on silent recurrent connections ( Figure 2 ) . Further analysis showed a close relationship between the magnitude of IPSP facilitation and the extent of presynaptic depolarization in PCs , consistent with the requirement of excitation-inhibition balance during elevated network activity [1] , [3] . The abrupt occurrence and the facilitation of disynaptic IPSPs may result from the increases in the spiking probability or the number of APs in LTS interneurons ( Figures 4 and 5 ) . Since the inhibitory synaptic strength at synapses between the newly recruited LTS interneurons and the postsynaptic PC should not depend on the amplitude of existing IPSPs , we expect that the net increase does not depend on the baseline IPSP amplitude . This was supported by the finding that the data shown in Figure 1F were well fitted by a hyperbolic function , which predicts a linear function between IPSP amplitudes at depolarized Vm and those at resting Vm ( see Figure S1 ) . The PC-FS-PC microcircuit is a potential candidate mediating the fast ( early-onset ) recurrent inhibition [27] , [29] . In contrast to LTS cells , FS neurons receive EPSPs that show depression in response to presynaptic high-frequency APs [31] , [34] . A single presynaptic AP can trigger the FS neuron to discharge once and subsequently evoke a unitary IPSP at neighboring PCs [28] , [29] . Consistent with the findings reported previously [27] , [29] , we observed fast disynaptic IPSPs in PC pairs at a very low frequency . In the PC pairs that showed depolarization-induced facilitation , we found that the facilitation resulted mainly from a reduced IPSP failure rate ( Figure 6B ) , consistent with an increased EPSP amplitude and spiking probability of FS neurons . Thus , the same mechanism underlies Vm-dependent modulation of disynaptic IPSPs in both types of microcircuits . Consistent with previous studies [25] , [26] , our results showed that Vm changes in presynaptic PC modulated the sizes of EPSPs in both the LTS and the FS interneurons ( Figure 4 and Figure 6 ) . The underlying mechanism may depend on unique intrinsic properties of axonal ion channel subtypes . Recently , a low-threshold , slowly inactivating K+ current ( known as D-current [36] , mediated by Kv1 alpha subunits ) was recorded at the PC axons and was shown to selectively control the duration of axonal APs ( instead of somatic APs ) and the depolarization-induced facilitation of EPSPs [37] , [38] . Consistently , our results demonstrated that the inhibition of Kv1 channels was by itself sufficient to increase the size of summated EPSPs in PC-LTS and of disynaptic IPSPs in PC-PC pairs . Furthermore , it also occluded the effects of Vm at PC-LTS synapses and disynaptic transmission between PCs . Depolarization in the presynaptic cell could prolong axonal APs as well as activate presynaptic Ca2+ channels , thereby enhancing synaptic transmission by increasing the presynaptic background Ca2+ concentration and/or AP-triggered Ca2+ influx [40]–[42] . Indeed , high concentrations of EGTA could drastically reduce the success rate of EPSP facilitation induced by presynaptic somatic-depolarization [26] ( but see [25] ) , suggesting that background Ca2+ may also contribute to the modulation of disynaptic IPSP . Inhibitory interneurons contain many kinds of Ca2+-binding proteins , such as parvalbumin , calretinin , and calbindin; they could function as Ca2+ buffer and thus may prevent the Vm-dependent modulation of inhibitory transmission . Indeed , our results showed that the percentages of inhibitory connections ( including FS-PC and LTS-PC pairs , Figure 7 ) showing Vm modulation are far less than those of excitatory connections ( including PC-PC and PC-interneuron pairs , Figures 6 and 7 ) . Whether the intracellular Ca2+-binding proteins may be responsible for regulating analog-mode signaling at inhibitory synapses remains to be further examined . Subthreshold Vm changes in the soma spread down the axon with a length constant of 400–800 µm [25] , [26] , [37] , [43] . Over 150 putative synaptic boutons are distributed at axon collaterals within 500 µm of the cell body of layer-5 PCs [26] . Boutons in remote axon terminals may not be affected by somatic Vm changes . This may explain why not all recurrent inhibitory connections were subjected to the Vm-dependent modulation and why the percentages of LTS-PC pairs showing Vm modulation are smaller than those of FS-PC pairs . In comparison with FS cells that target the perisomatic region of their neighboring PCs , LTS cells in layer 5 send their axons to superficial layers and innervate distal apical dendrites of nearby PCs . Therefore , Vm modulation should be weaker in LTS-PC than in FS-PC synapses . Whether such modulation is indeed spatially confined to local circuits within the range of axonal spread of somatic depolarization , or alternatively only specific cortical microcircuits are modulated , remains to be further determined . The Vm-dependent modulation of recurrent inhibition described here may serve several distinct functions . First , it may contribute to maintaining a dynamic excitation-inhibition balance at different cortical activity states appropriate for diverse behavioral conditions [8] , [10] , [11] , [44] . For example , when Vm depolarizes during an active but relatively stable cortical state , e . g . , the “Up” state , the inhibitory conductances due to recurrent connections increase to match the elevated excitatory conductances [1] , [3] , [4] . A recent work revealed that excitation-inhibition balance is also instantaneously controlled with a millisecond precision during spontaneous and sensory-evoked activities [2] and disruption of this balance causes dysfunction of the network [3] , [9] , leading to various disorders such as epileptic seizures [45] , [46] and schizophrenia [47] . Second , the Vm-dependent modulation of recurrent inhibition may also contribute to rapid transitions between “Up” and “Down” states [1] , [3] that are important for gain modulation of synaptic and sensory inputs [1] , [12] , [13] , [16] , [17] , [48] , [49] . For example , transient excitation-inhibition imbalance caused by abrupt changes in Vm-dependent recurrent inhibition , e . g . , unsilencing of recurrent connections ( Figure 2A–C ) induced by Vm fluctuation at the depolarized Vm levels , could cause a switch from “Up” to “Down” state . Third , the shortening of IPSP latency associated with Vm-dependent facilitation of recurrent inhibition ( Figure 2D–G ) may regulate the time window of integration of excitatory inputs , therefore providing a mechanism for Vm-dependent feedback control of the timing of spike initiation in PCs [50]–[52] . To conclude , we have shown that both slow and fast recurrent inhibition is susceptible to modulation by the Vm changes of PCs . These results demonstrate a circuit function of Vm-dependent modulation of excitatory transmission ( analog-mode signaling ) . Whether such Vm-dependent modulation is universal among all cortical circuits and whether it plays an important function in regulating circuit dynamics in behaviorally relevant conditions remain to be examined .
The use and care of animals complied with the guidelines of the Animal Advisory Committee at the Shanghai Institutes for Biological Sciences . We anesthetized the animal ( 15∼18-d-old Sprague-Dawley rats ) with sodium pentobarbital ( 30 mg kg−1 ) before decapitation . The brain was quickly dissected out and immersed in an ice-cold oxygenated ( 95% O2 and 5% CO2 ) slicing solution in which the NaCl was substituted with sucrose ( 213 mM ) and dextrose was reduced to 10 mM . We cut parasagittal slices ( 350 µm ) of somatosensory cortex in this solution with a Leica microtome ( VT-1000S ) and immediately transferred to an incubation beaker filled with aerated normal artificial cerebrospinal fluid ( ACSF ) containing ( in mM ) : NaCl 126 , KCl 2 . 5 , MgSO4 2 , CaCl2 2 , NaHCO3 26 , NaH2PO4 1 . 25 , and dextrose 25 ( 315∼325 mOsm , pH 7 . 3 ) . Slices were incubated at 34 . 5°C for at least 45 min , then at room temperature until use . Visualization of cortical layers and neurons were made with an upright infrared-DIC microscope ( BX51WI , Olympus ) equipped with an infrared camera ( OLY-150 ) . All experiments were done at a temperature of 35 . 5–37°C . We obtained dual whole-cell recordings from layer-5 PCs and inhibitory interneurons ( LTS and FS neurons ) using Multiclamp 700B amplifiers ( Molecular Devices ) . Patch pipettes were prepared with a P-97 microelectrode puller ( Sutter Instruments ) and filled with an internal solution containing ( in mM ) KGluconate 140 , KCl 3 , MgCl2 2 , Na2ATP 2 , BAPTA 0 . 025 , and HEPES 10 ( pH 7 . 2 with KOH , 280∼290 mOsm ) . Patch electrodes had an impedance of 3–6 MΩ . For tracing and labeling the recorded neurons , we added Alexa Fluo 488 ( 100 µM ) and biocytin ( 0 . 2% ) to the pipette solution . We identified the recorded layer-5 neurons through their unique somatic and dendritic morphology under the DIC and fluorescent microscope and their distinct firing patterns . The total time the cell was exposed to fluorescence was kept to less than 10 s to minimize cell damage . The PCs could be easily distinguished from interneurons because of their thick apical dendrite and large pyramid-shaped somata . The FS neurons were identified through their non-adapting and fast-spiking ( 300–500 Hz in response to current stimulation , Figure 6 and Figure S2 ) firing properties and their “noisy” resting Vm constantly bombarded with large-amplitude EPSPs . The LTS neurons ( Figure 4 ) were classified through their low-threshold regular firing patterns with initial accelerating then decelerating discharges in response to step current injection ( see also Figure S2 ) . After electrophysiological recording , the neurons were further identified using DAB-staining . The intrinsic properties of individual neurons and the properties of synaptic connections between PCs and FS and LTS neurons ( <100 µm apart ) were examined as soon as a dual or triple recording was achieved . We injected negative ( −0 . 1∼−0 . 5 nA ) and positive current pulses ( 0 . 1∼1 . 5 nA , 500 ms ) to examine the input resistance and firing patterns of each neuron . To test for synaptic connections , we injected 1-ms current pulses ( 10∼20 pulses ) at a frequency of 20∼200 Hz to each PC to evoke a train of APs every 15 s while monitoring the Vm changes in other PCs or interneurons . Unless otherwise stated , for data analysis and figures , we normally evoked a train of 15 APs at 100 Hz through current injection in the presynaptic PC . In most experiments , Cl− concentration in the recording pipette was 7 mM , and the calculated reversal potential for Cl− was −74 mV . Disynaptic IPSPs recorded between PCs with this pipette solution were hyperpolarizing potentials at a depolarized postsynaptic Vm ( normally depolarized from resting to ∼−46 mV with DC current injection ) . We injected constant DC current to evoke intermittent depolarizing and hyperpolarizing Vm levels ( ∼10–20 mV , ∼45 s each ) in the presynaptic PC while keeping the postsynaptic Vm constant . In the recordings examining the monosynaptic connections between PC and FS neurons , we used a high concentration of Cl− ( 75 mM ) in the pipette solution . To test for the presynaptic Vm-dependent modulation of monosynaptic EPSPs in PC-FS pairs and fast disynaptic IPSPs in PC pairs , we only injected brief pulses to presynaptic PC to evoke single APs at 1 Hz on top of the intermittent depolarizations and hyperpolarizations . In most of our recordings , synaptic responses were stable and could be recorded up to 1–2 h after obtaining whole-cell recordings without any apparent rundown . Data were discarded if the evoked IPSPs and EPSPs showed significant rundown , as shown by a statistically significant change in the amplitude of the IPSP or EPSP between the first and last third of the interspersed control periods ( when the presynaptic PC was at resting Vm ) . The Vm values were not corrected for the liquid junction potential ( 15 mV ) . During the whole period of recording , access resistance was monitored frequently; recordings with access resistance higher than 25 MΩ were discarded . Bridge balance and capacitance neutralization were carefully adjusted before and after every experimental protocol . We collected the electrophysiological data using a Micro 1401 digitizer and Spike 2 software ( Cambridge Electronic Design , Cambridge , UK ) . After a recording was completed , the slice was transferred to 4% paraformaldehyde in 0 . 1 M phosphate buffer for subsequent immunostaining and visualization . CNQX ( AMPA receptor antagonist ) , picrotoxin ( PTX , GABAA receptor antagonist ) , α–dendrotoxin ( α–DTX , Kv1 channel blocker ) , and 4-aminopyridine ( 4-AP , D-current blocker when applied at low concentrations ) were applied through bath perfusion . Their concentrations were indicated in the text . We performed all computations using Spike 2 and MATLAB ( MathWorks , Bethesda , MD ) . The significance of differences between the cumulative frequency distributions was determined by Kolmogorov-Smirnov test using original disynaptic IPSPs . We used Student's t test to test the significance of differences in peak amplitude , integrated voltage area , and onset latency between resting and depolarized presynaptic Vm in individual pairs . Values were presented as mean ± standard error in the figures as well as in the main text . For disynaptic IPSPs , the peak amplitude was the difference between the peak value and the average baseline Vm ( 2 s prior to the stimuli onset ) , whereas the integral voltage area was the curve area underlying the responses . The onset latency was the time difference between the response onset and the beginning of presynaptic stimulation . For comparison of these values at different Vm , we normalized the values to those obtained at the baseline Vm for each pair and then performed the statistical tests . To identify the EPSP failures in PC-LTS pairs , we first averaged the EPSPs evoked by presynaptic AP trains and selected an EPSP template from the average trace , then performed a correlation test between the voltage trace after each AP and the template EPSP . A failure was identified if the correlation coefficient was lower than 0 . 8 . Considering that the first five EPSPs during the train had a high failure rate and most of them showed significant differences in the failure rates at different Vm ( Figure 4D ) , we therefore chose the 6th EPSP amplitude ( measured from the baseline before the train , see Figure S3 ) as a reference for normalization . The peak amplitude of each EPSP during the train was normalized to the 6th EPSP for each PC-LTS pair and then averaged for group data presentation ( Figure 4 , Figure S3 ) . For the monosynaptic connections from PC to FS neurons ( or from interneurons to PCs ) , we calculated the average EPSP ( or IPSP ) and determined the time of the peak . The amplitude of each evoked EPSP ( or IPSP ) on single trials was taken as the difference between the postsynaptic Vm at the peak time of the average EPSP ( or IPSP ) after the AP and the Vm before onset of the current pulse evoking the AP . We measured the baseline activity as the difference in Vm over the same time delay , but without a presynaptic AP . The rise time of the monosynaptic IPSP was measured as the time from 20% to 80% of the peak amplitude , and the decay time constant was obtained through a single exponential fit to the decay phase . | Proper functioning of the neocortex requires a balance between excitation and inhibition . This balance can be achieved through the operation of cortical microcircuits interweaved by excitatory and inhibitory neurons . Since the membrane potentials ( Vm ) of cortical neurons fluctuate at different levels during cortical activities , it is important to know how the balance of excitation and inhibition is dynamically maintained at different Vm . Recurrent inhibition between excitatory pyramidal cells is mediated by two distinct types of inhibitory interneurons . Here , we show that the amount of recurrent inhibition depends on the Vm levels of presynaptic pyramidal cells . Modest depolarization of a pyramidal cell substantially increases , and sometimes turns on , disynaptic inhibition on its neighboring pyramidal cells . We find that this effect is due to an increase in the strength of synaptic connections from the pyramidal cell to inhibitory interneurons and a consequent elevation of interneuronal firing . The depolarization-induced increase in synaptic strength from the pyramidal cell therefore reflects “analog-mode” signaling in cortical excitatory synapses . We thus reveal a profound impact of analog-mode signaling on the operation of cortical microcircuits and provide a new mechanism for dynamic control of the balance of cortical excitation and inhibition . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cellular",
"neuroscience",
"central",
"nervous",
"system",
"synapses",
"biology",
"neuroscience",
"neurophysiology"
] | 2011 | Membrane Potential-Dependent Modulation of Recurrent Inhibition in Rat Neocortex |
Despite substantial progress in the study of diabetes , important questions remain about its comorbidities and clinical heterogeneity . To explore these issues , we develop a framework allowing for the first time to quantify nation-wide risks and their age- and sex-dependence for each diabetic comorbidity , and whether the association may be consequential or causal , in a sample of almost two million patients . This study is equivalent to nearly 40 , 000 single clinical measurements . We confirm the highly controversial relation of increased risk for Parkinson’s disease in diabetics , using a 10 times larger cohort than previous studies on this relation . Detection of type 1 diabetes leads detection of depressions , whereas there is a strong comorbidity relation between type 2 diabetes and schizophrenia , suggesting similar pathogenic or medication-related mechanisms . We find significant sex differences in the progression of , for instance , sleep disorders and congestive heart failure in diabetic patients . Hypertension is a highly sex-sensitive comorbidity with females being at lower risk during fertile age , but at higher risk otherwise . These results may be useful to improve screening practices in the general population . Clinical management of diabetes must address age- and sex-dependence of multiple comorbid conditions .
Diabetes is a global pandemic disease . The world-wide number of adult diabetes patients doubled over the last three decades to approximately 350 million as of 2010 , and is expected to double again until 2030 as a result of population ageing and a shift to western lifestyle patterns in developing countries [1] . Diabetes comprises a heterogeneous group of disorders with the most prominent types being type 1 ( DM1 ) and 2 diabetes ( DM2 ) . These disorders have different pathophysiology and phenotype; the exact underlying mechanisms , their interplay finally leading to manifestation , progressions of the diseases , and their complications are still unclear . Diabetes is related to a large number of comorbid diseases , including but not limited to vascular complications [2] , renal failures [2] , neuropathy [2] , heart diseases [3 , 4] , cognitive disorders [5 , 6] , retinopathy [7] , and hypertension [8] . Each of these comorbidities opens up a unique direction of research . Following the methodological approach developed in this work , thousands of such relations can be investigated in parallel . Besides studying the individual diabetic comorbidities and how they depend on patient age and gender , this allows to compare the strength of these relations among each other and to rank them according to their significance . Nation-wide collections of physician and hospital claims data allow to explore the health state of an entire country’s population with unprecedented precision and scale [9] . To exploit the full potential of ‘big data’ for medical sciences the development of novel , quantitative methods to extract clinically relevant features from large datasets of electronic health records ( EHR ) is necessary . First efforts in this direction have proven to be extremely fruitful by developing or improving data-driven comorbidity indices to predict mortality rates [10] , or by studying healthcare utilization and outcome measures of specific patient cohorts [11] . Large-scale analyses of comorbidities using EHR data have demonstrated that human disease phenotypes can be related to each other in highly connected networks with strong pairwise correlations between diseases [12 , 13 , 14 , 15] . In this work we develop a new quantitative framework to measure age- and gender-dependent relative risks for all possible comorbidity relations for DM1 and DM2 using medical claims data from almost two million people . We introduce tests to assess the significance of the comorbidity relations , the influence of sex , and whether diabetes is more likely to be a diagnosed before or after the other disease .
A research database of the Main Association of Austrian Social Security Institutions containing pseudonymised claims data of all persons receiving care in Austria between January 1st , 2006 and December 31st , 2007 is used [16] . The data gives a comprehensive , nation-wide picture of the medical condition of most of the approximately 8 . 3 million Austrians . The patient collective was formed by extracting all persons receiving inpatient care in 2006 or 2007 . We identified patients being diagnosed with DM1 or DM2 ( ICD10 codes E10 and E11 ) . Patients who died in 2006 or 2007 were removed . In this way 16 667 DM1 patients ( 8 355 males and 8 312 females ) and 105 904 with DM2 ( 50 596 males and 55 308 females ) were selected . The total sample of inpatients used in this study consists of 1 862 258 patients ( 1 064 952 females and 797 306 males ) . From these patients we know their year of birth , sex , ATC codes of all their prescriptions , and the ICD codes of all their diagnoses ( main- and side-diagnoses ) . Co-occurrence analysis/ relative risks for comorbidities . For the occurrences of each diagnosis x ( ICD10 , three-digit-level ) a patient-age-resolved cross tabulation with the occurrences of DM1 and DM2 is performed . Symptoms , injuries , pregnancies , and external causes and factors of morbidity were excluded . We therefore test 1 051 diagnosis ( ICD10 codes ranging from A01 to N99 ) for their co-occurrence with diabetes . The patients are grouped by their age in five-year intervals and by their gender . Patients older than 95 have been excluded . We test 1 051 possible comorbidities for 19 age groups for DM1 and DM2 , giving 39 938 tests . For each diagnosis and age interval a contingency table is built . If each entry in the table is greater than 10 , relative risks RR1 ( 2 ) ( x , t ) are computed , a chi-squared test is performed and p-values are calculated for rejecting the null hypothesis that co-occurrence of the diagnosis with DM1 or DM2 is independent . This leads to a multiple hypothesis testing problem for each age group where 1 051 hypotheses are tested in parallel . To correct for these multiple comparisons we apply the Benjamini-Hochberg procedure [17] to control for the false discovery rate α . This procedure is a multiple comparison correction where the value of α gives the expected probability that a null hypothesis is incorrectly rejected . For example , if 100 comorbidities are identified with a false discovery rate α of α = 0 . 01 , the expected number of false positives among these comorbidities is one . If there are less than ten co-occurrences or the results are not significant , the relative risk is set to one . For the co-occurrence analysis we use both the main and the side diagnoses of each patient . Validation of the co-occurrence analysis . To validate the results of the co-occurrence analysis we compile a list of major known diabetic complications from different literature sources [18 , 19 , 20] . These lists are based on hand curated collections of diabetic comorbidities , some of them validated using EHR data [19 , 20] . These studies disagree on the exact list of ICD codes for diabetic complications , but each list focusses on cardiovascular , renal , and ophthalmic comorbidities . The ICD codes that are listed as diabetic complications in each of these studies are therefore used to validate our co-occurrence analysis , see Table 1 . Note that , for example , mental disorders like depression or pancreatic cancer , both well-known diabetic comorbidities [5 , 6 , 21] , are not included in any of these studies . Nevertheless , a valid method to detect comorbidities is supposed to pick up a substantial number of the diagnoses listed in Table 1 , among other comorbidities . We will therefore be interested in the recall R ( α ) as a function of the false discovery rate α . R ( α ) is the probability that a diabetic comorbidity listed in Table 1 is also identified by our co-occurrence analysis at a given level of α . Sex ratio . The sex ratio SR ( x , t ) is related to the quotient of the percentage of female and male diabetes patients in age group t that also have diagnoses x or are prescribed a medication x . Denote the number of male ( female ) DM1 and DM2 patients in age group t by Dm ( f ) ( t ) and the number of male ( female ) diabetes patients who also have diagnoses or medication x by Dm ( f ) ( x , t ) . The sex ratio SR ( x , t ) is then related to the logarithmic quotient of the percentage of female and male diabetes patients who also have diagnoses x , SR ( x , t ) =log[1+Df ( t ) Df ( x , t ) 1+Dm ( t ) Dm ( x , t ) ] . ( 1 ) A value of SR ( x , t ) that is close to zero indicates that the co-occurrence of the diagnosis or medication x with diabetes is equally likely for males and females . Positive ( negative ) values of SR ( x , t ) indicate that the co-occurrence is more likely for females ( males ) . To assert the statistical significance of nonzero SR ( x , t ) values we build a contingency table for all diabetes patients of a given age group t . The table contains the two variables sex and co-occurrence with diagnosis/medication x . If the null hypothesis of statistical independence of these two variables cannot be rejected in a chi-squared test using a p-value of p = 0 . 05 the sex ratio is set to zero , SR ( x , t ) = 0 . Lead/lag indicator . The lead/lag indicators assess whether patients with diagnoses di are more likely to be later diagnosed with another disease x , the lead indicator Ilead ( di , x ) , or whether it is more likely that people having diagnoses x will be diagnosed with diabetes , the lag indicator Ilag ( di , x ) . There exist several known biases in EHR data that need to be addressed in the definition of these indicators [22] . ( i ) The first occurrence of a coding of a diagnosis in the EHR data will typically not correspond to the true initial diagnosis of the disease . ( ii ) The data only spans two years , which may not be enough to observe the manifestation of diabetic complications directly . We use the following methodology to measure the lead/lag indicators and adjust for these known biases . Let us consider the lead indicator Ilead ( di , x ) that measures if the diagnosis x is typically made after the diabetes diagnosis . Given the limitations of our data , we cannot observe the typical time between the manifestations of the two diseases . We can , however , measure whether there is a tendency that x will be diagnosed in a patient that already had a prior diabetes diagnosis . As opposed to the co-occurrence analysis , it is crucial for the lead/lag analysis to distinguish between main- and side diagnoses . To this end we consider the probability that a male ( female ) patient has a diabetes diagnosis ( main or side diagnosis ) in year t1 , and a main-diagnosis x in year t2 , but no diagnosis of x in t1 ( main or side diagnosis ) . Denote this probability by pm ( f ) ( x , t2|di , ¬x , t1 ) for males ( females ) . This number over-estimates the true effect size , since some cases where a patient does not have diagnosis x in year t1 might be due to inaccuracies in the coding or incompleteness of the data , in particular with respect to unknown pre-existing conditions . However , we assume that these errors are not systematic in the sense that they are equally likely to influence the data for year t1 and t2 . If there is no true temporal ordering in the onsets of di and x , the value of pm ( f ) ( x , t2|di , ¬x , t1 ) just measures noise due to incomplete or inaccurate data . But this is equally true for the probability that diagnosis x does not occur for a patient in year t2 , given that she ( he ) has both diagnosis di and x in t1 , the probability pm ( f ) ( x , t1|di , ¬x , t2 ) . If there is a substantial tendency that x is diagnosed after the onset of di , however , these two probabilities are likely to differ . The lead indicator Ilead ( di , x ) is therefore given by Ilead ( di , x ) =p ( x , t2|di , ¬x , t1 ) -p ( x , t1|di , ¬x , t2 ) . ( 2 ) The lag indicator Ilag ( di , x ) is constructed in analogy to the lead indicator Ilead ( di , x ) and by exchanging the roles of di and x , Ilag ( di , x ) =p ( di , t2|¬di , x , t1 ) -p ( di , t1|¬di , x , t2 ) . ( 3 ) If the frequency of the diagnosis x itself is very small already a very small number of events might lead to comparably large indicator values for Ilead ( di , x ) and Ilag ( di , x ) . We therefore exclude diagnoses x from the analysis if they have less than a threshold of z male or female patients that also have di in t2 . In the following we set t1 = 2006 and t2 = 2007 . For the lag indicator for DM1 we exclude all patients older than 30 . Finally , a statistical test is developed to assess the significance of positive values for Ilead ( di , x ) and Ilag ( di , x ) . Surrogate data is created by keeping the list of diagnoses for each patient fixed and by shuffling the information about the year when the diagnoses were made . Assume that patient p has np diagnosis {xi} made in the years {τi} with i ∈ {1 , … , np} . The surrogate data is constructed by replacing {τi} by a random permutation of itself . This procedure is repeated 1 000 times and the lead and lag indicators are computed for each surrogate dataset . We test the null hypothesis that the values for the lead and lag indicators observed in the data are as large as one would expect for indicator values taken from the surrogate data , where the temporal information has been randomly shuffled . The p-value for each lead and lag indicator is the probability of obtaining the observed values for Ilead ( di , x ) and Ilag ( di , x ) from the surrogate data . The null hypothesis is rejected if p<0 . 01 , that is if out of 1 000 surrogate datasets less than ten give indicator values that are larger than the observed values . A significant value of the lead indicator Ilead ( di , x ) suggests that the incidence of disease x is more likely in patients with pre-existing diabetes compared to the incidence of diabetes in patients with pre-existing disease x . A significant value of the lag indicator Ilag ( di , x ) , on the other hand , suggests that diabetes is typically incident in patients already diagnosed with x . A similar approach to study lead/lag behavior between diseases , but without a test for statistical significance of the results , was proposed for networks of comorbid diseases [12] .
Parkinson’s disease . In the literature there is no consensus on whether diabetes patients have a higher risk for Parkinson’s disease ( PD ) , or if there is actually a lower risk or no relation at all . There are two large prospective studies finding an increased risk for PD in diabetes patients , one study finding no relation , and one study reporting lower risk of diabetes [23] . We find that PD is comorbid ( 2 . 3 , CI 1 . 9–2 . 7 for DM1 and 1 . 5 , CI 1 . 4–1 . 6 for DM2 ) with an excess of male patients . It has been suggested that surveillance bias may lead to the reporting of spurious positive correlations between PD and diabetes [23] . Given our patient cohort we can exclude this kind of bias . Note that the size of our patient cohort ( 1 . 8 million patients ) is at least 10 times larger than the largest cohorts in previous studies on the relation between PD and diabetes [23 , 24] . As potential mechanism of this association the involvement of insulin in the regulation of brain dopanergic activity has been proposed [25 , 26] . Animal and in vitro studies have shown that insulin and dopamine may exert reciprocal regulation [26] . Mental disorders . Depression , schizophrenia , and schizo-affective disorders are also comorbid . While the relative risks for DM1 patients are highest in the age group 65–70 with values from 1 . 9–2 . 3 for these diseases , we find higher risks for DM2 patients at younger ages , e . g . a relative risk of 4 . 8 , CI 3 . 3–7 . 0 , for recurrent depressive disorders at age 35–40 . We find that depression is usually incident in DM1 patients . From these results one may speculate that DM1 patients develop depressions because of the burden of the disease and the psychological distress of maintaining a good level of glycemic control . Depression in diabetic patients in general , DM1 and DM2 , is dominated by females [5] , so is the association between depression and overweight [27] . Indeed it is remarkable that depression and overweight as diabetic comorbidities show nearly the same age and sex dependence . A possible biological mechanism is that obesity increases the risk of increased insulin resistance , which may induce alterations in the brain which in turn increase the risk of depression [28] . Of importance are also psychological pathways , since the perception of being overweight increases psychological distress [29] . Diabetes has also been associated with the use of atypical neuroleptics in the treatment of schizophrenia [30] . The sex ratios for antipsychotics show a strong excess of female patients , see Fig . 2 , which compares well with the female excess in the sex ratios for depression and schizophrenia . It is interesting to note that the comorbidity relations with schizophrenia and schizo-affective disorders stand out as much weaker for DM1 than for DM2 patients , when compared to all other results of the comorbidity analysis . Endocrine and metabolic disorders . While patients with thyroiditis , hypothyroidism , thyrotoxicosis , and obesity are predominantly female , disorders of the lipoprotein , purine , and pyrimidine metabolism tend to be found in males . Diabetic patients feature a two to three times higher increased risk of disorders of the thyroid gland , particularly those with autoimmune diabetes , a comorbidity relation that is strongly influenced by gender [31] . For volume depletion and disorders of fluid , electrolyte and acid-base balance there appears to be an age switch , from an excess of male patients for ages 20–40 to an excess of females in older age . Primary hypertension is a comorbidity with relative risks of 5 . 3 ( CI 4 . 8–5 . 9 ) for DM1 and 9 . 5 ( CI 8 . 8–10 ) for DM2 . These switches may indicate an important impact of sexual hormones and of potential pregnancies but may also point to social factors related to sex-specific phases of life . The prescriptions of beta and calcium channel blocker , as well as ACE inhibitor show a sex-dependence very similar to hypertension , suggesting that these drugs are commonly used to treat hypertension , see Fig . 2 . There is a strong excess of female patients in the prescriptions of diuretics , especially in elderly patients , whereas there is a strong excess of younger males being prescribed statins or other lipid modifying agents , the latter matching the sex ratio observed for hypercholesterolemia and hyperlipidemia . Note that our results make no statements about the combinations of antihypertensive drugs which are actually used in the treatment of individual patients . Infections and sepsis . Bacterial and viral infections ( gastroenteritis , erysipelas , pneumonia , osteomyelitis , hepatitis , dermatophytosis , candidiasis ) show an excess of male patients with the exception of gastroenteritis and candidiasis , which are dominated by female patients . We find an excess of sepsis comorbidity which is strongest in male DM1 patients at the age around 50 , with higher relative risks for DM1 ( 12 , CI 8 . 2–18 ) than DM2 ( 2 . 7 , CI 2 . 4–2 . 9 ) . Epilepsy . The increased risk for epilepsy ( 4 . 6 , CI 3 . 1–6 . 9 , for DM1 and 1 . 6 , CI 1 . 4–1 . 7 , for DM2 ) in young type 1 diabetics [32] may be linked to ketoacidosis as a two times higher risk of epilepsy was found in children and adolescents with metabolic acidosis [33] . A four times greater risk of DM1 was also described in young adults with epilepsy [34] . Both metabolic extremes , hypoglycemia and diabetic ketoacidosis , relate to EEG abnormalities in diabetic children which may increase risk of epilepsy . Congestive heart failure . The Framingham heart study reported that diabetic women are more vulnerable to congestive heart failure ( CHF , RR of 5 . 2 , CI 4 . 7–5 . 9 , for DM1 , 3 . 8 , CI 3 . 6–3 . 9 , for DM2 ) than men [35] . However , subsequent cohort studies found no such sex differences [35 , 36] . We find an excess of male patients and that diabetes is typically detected before CHF in females with DM1 and in males with DM2 . Sleep disorders . Sleep disorders are comorbid in DM1 ( 1 . 9 , CI 1 . 5–2 . 4 ) and DM2 patients ( 2 . 3 , CI 2 . 1–2 . 6 ) . We find support for sex specific progression routes . It is known that DM2 and obstructive sleep apnea ( OSA ) present a vicious circle , with OSA exerting adverse effects on glucose metabolism and thereby increasing the risk for DM2 [37] . In patients with already existing DM2 , on the other hand , there is a significant relationship between sleep-disordered breathing ( SDB ) and insulin resistance independent of obesity [38] . The fact that there is an excess of male patients in the comorbidity relation may be related to the higher prevalence of central adiposity and therefore OSA in men [37] . Pancreatic cancer . There are higher relative risks for DM1 patients ( 8 . 6 , CI 5 . 6–13 ) than DM2 patients ( 2 . 5 , CI 2 . 1–2 . 8 ) . The risks peak in the age range 50–70 with a balanced sex ratio . It has been shown that diabetic patients are at increased risk of pancreatic cancer with a pooled RR of approximately two compared to non-diabetics in a meta-analysis [21] with at least one year diabetes duration prior to diagnosis of pancreatic cancer [39] . Diabetes also leads the diagnosis of pancreatic and lung cancer . Behavioral and related disorders . Nicotine dependence ( 3 . 3 , CI 2 . 7–4 . 1 , for DM1 and 2 . 8 , CI 2 . 6–3 . 0 , for DM2 ) and alcohol related disorders dependence ( 2 . 3 , CI 1 . 7–3 . 2 and 2 . 1 , CI 1 . 9–2 . 4 ) are comorbidities with relative risks peaking at ages 30–45 , dominated by male patients . Alcoholic liver disease dependence ( 4 . 0 , CI 2 . 7–5 . 7 and 2 . 6 , CI 2 . 3–2 . 9 ) is also a male-dominated comorbidity . Toxic liver disease ( 2 . 8 , CI 1 . 6–4 . 9 and 14 , CI 8 . 5–23 ) and fibrosis and cirrhosis of liver ( 5 . 0 , CI 3 . 7–6 . 6 and 2 . 4 , CI 2 . 2–2 . 7 ) show also an excess of male patients . There tend to be higher risks for DM1 than DM2 patients , potentially outlining greater impact of chronic hyperglycemia than of overweight-related parameters of the metabolic syndrome . The relationship between alcohol consumption and DM2 has been shown to be dosage dependent . While moderate alcohol consumption is protective , dosages of more than 60g/day increase diabetes risk [40] . It is not possible to establish an alcohol-dosage dependent diabetes risk from our data . Cardiovascular diseases . Identified comorbid diseases of the circulatory system include ischemic and pulmonary heart disease , cardiomyopathy , valvular disorders , tachycardia , as well as cerebrovascular diseases and diseases of the arteries and veins [2 , 4 , 41] . Comorbid diseases of the circulatory system show a consistent excess of male patients , including ischemic , pulmonary , and other heart diseases ( cardiomyopathy , valvular disorders , tachycardia ) , as well as cerebrovascular diseases and diseases of the arteries and veins . The highest relative risks among cardiovascular diseases are found for acute ischemic heart diseases for DM1 patients ( 6 . 6 , CI 5 . 2–8 . 3 , compared to 3 . 1 , CI 2 . 8–3 . 4 , for DM2 patients ) at ages higher than 60 . Pulmonary diseases . Pneumonia and acute bronchitis show increased relative risks for older ages ( e . g . for pneumonia 2 . 7 , CI 2 . 4–3 . 0 , for DM1 , 2 . 3 , CI 2 . 1–2 . 4 , for DM2 ) . Chronic obstructive pulmonary disease ( COPD ) is led by diabetes ( 2 . 9 , CI 2 . 5–3 . 5 and 2 . 2 , CI 2 . 1–2 . 3 ) . Diabetes is often identified as independent risk factor for lower respiratory tract infections [42] . Individuals with COPD are substantially more likely to have pre-existing DM [43] , on the other hand lung function impairment in COPD is a risk factor for developing diabetes and insulin resistance [44] . Benign pleural effusion ( 3 . 4 , CI 2 . 1–5 . 6 and 3 . 1 , CI 2 . 5–3 . 9 ) , representing a symptom of various underlying diseases , is dominated by males . In diabetic patients pleural effusion may be related to left ventricular dysfunction as described previously [45] . Other comorbidities . Iron-deficiency and anemia in chronic diseases show higher relative risks for DM1 ( 3 . 7 , CI 3 . 0–4 . 6 , and 6 . 3 , CI 4 . 9–8 . 1 ) than DM2 ( 2 . 7 , CI 2 . 4–2 . 9 and 2 . 8 , CI 2 . 5–3 . 2 ) patients . Cataracts , retinal detachments , glaucoma , disorders of the vitreous body , and blindness are identified here with relative risks up to 200 . The higher relative risks for DM1 compared to DM2 patients for retinopathies [7] at older age suggest a higher lifespan for type 1 diabetics . Chronic and acute kidney diseases , the nephrotic syndrome , and glomerular disorders are identified as comorbidities with an excess of male patients; relative risks range up to 128 for DM1 patients and 8 . 6 for DM2 . There is an excess of female patients in the age range 20–40 . Intestinal malabsorption ( including celiac disease ) shows elevated risks for ages 10–25 for DM1 ( 10 , CI 6 . 3–17 ) with a weak female excess; there are no significant results for DM2 . Cholelithiasis is a female dominated comorbidity ( 1 . 7 , CI 1 . 5–2 . 0 and 1 . 5 , CI 1 . 4–1 . 6 ) . Cholecystitis is typically followed by DM2 in males . Pressure and non-pressure ulcers exhibit higher risks for DM1 ( 7 . 2 , CI 5 . 2–9 . 9 , and 7 . 4 , CI 5 . 8–9 . 4 ) than DM2 patients ( 2 . 2 , CI 2 . 0–2 . 4 and 4 . 2 , CI 3 . 9–4 . 6 ) . For males there are increased risks for disorders of prepuce ( 6 . 0 , CI 3 . 5–10 and 3 . 1 , CI 2 . 5–3 . 8 ) , while for females there is increased risk for disorders of the urinary system ( 2 . 5 , CI 2 . 2–2 . 8 and 1 . 8 , CI 1 . 7–1 . 9 ) . Evidence from epidemiological studies suggests that asymptomatic bacteriuria and symptomatic urinary tract infections occur more commonly in women with DM compared to non-diabetic controls [41 , 46] . Increased prevalence of urinary incontinence and urge incontinence among women with DM2 [47 , 48] has been reported . Limitations . Only persons with inpatient stays were included in the study . To test if this pre-selection introduces a bias in our results , we repeated the study with a sample of all patients having been prescribed at least once a drug used in diabetes ( ATC code starting with ‘A10’ ) in 2006 or 2007 . We compare the frequencies of their diseases with those in the rest of the population , roughly 8 . 3 million patients . This assumes that DM patients with no hospital stay in the study period have no diagnosis and therefore no comorbidities . Although this is a highly incorrect assumption , it serves as a conservative test-assumption , which allows to test if the comorbidities are simply significant as a consequence of our limited sample that contains only inpatients . Results are shown in the supplement in S1 Fig . In the enlarged sample only one out of the 123 comorbidities using the inpatient sample has a p-value greater than 0 . 05 ( M23 ) , all other remain significant ( p<0 . 05 ) . Significance of comorbidity in the inpatient sample is therefore highly representative of comorbidity in the entire population . However , our approach might miss diabetic comorbidities that are typically not related to hospitalizations and that are most prevalent in younger patients , where the inpatient sample contains a lesser amount of the entire population , compare Fig . 1 ( a ) . Unknown pre-existing conditions may also affect the observed temporal order of the diseases , which has been addressed by applying a series of corrections to the lead/lag indicators , equations ( 2 ) and ( 3 ) . Other limitations relate to the coding quality of disorders in the medical claims data , which has been shown to lead to an under-reporting of comorbidities [49] and may cause false negatives in our testing procedure . This work shows the enormous potential that large-scale analyses of EHR data offer for the medical sciences . For the first time we develop a standardized testing procedure to obtain a complete comorbidity profile for DM1 and DM2 using medical claims data . This analysis is equivalent to 39 938 individual tests , each with the maximum number of patients available in a country . We identified 123 highly significant disorders with increased or decreased risks , strongly depending on patient age and sex . The comorbidities are investigated by a lead/lag analysis to inquire whether the relation between the diseases is more likely causal or consequential . Taken together , these results underscore that there is a substantial number of disorders that are related to diabetes , besides the well-known long-term complications . Diabetic comorbidities are rule rather than exception and their treatment must address their high degree of age and sex dependence . Despite being a risk factor for certain diabetic complications , sex may also influence and to a certain degree even determine the mechanisms underlying the disease progressions . Our results may be of immediate use to improve screening practices and therapy of diabetic patients to increase their quality of life and potentially contribute to longer life expectancy due to early detection and treatment of important comorbidities . In particular we propose to screen and , where applicable , treat diabetes patients for comorbid depressions , since this allows a more efficient treatment of diabetes itself . Depressive patients should be screened for diabetes to detect it at an early stage and perform lifestyle interventions that focus on weight control . It is also important to treat depressive patients with drugs that have a minimum of side effects on weight gain , and lipid and glucose metabolism . Our results emphasize that physicians must be aware of non-traditional diabetic comorbidities and risk factors during anamnesis and that , for example , screening for diabetes may be appropriate in patients with cardiovascular diseases , CHF , or fatty liver , whereas diabetes patients should be screened for pancreatic cancer . | We quantify for the first time age- and gender-dependent relative risks for each possible comorbidity of type 1 and 2 diabetes in a nation-wide claims dataset containing almost two million patients , and test whether the association may be consequential or causal . This study therefore contains almost 40 , 000 single clinical measurements , all with the maximum patient number available in an entire country . We confirm the relation between diabetes and Parkinson's disease , and find different progression routes of mental disorders in type 1 and type 2 diabetics . Among many other results , we also report significant gender differences in the progression of congestive heart failure , sleep disorders , hypertension , and hyperlipidemia . This work provides the first complete statistical description of all diabetic comorbidities and their dependence on patient age and sex . These results may be of immediate use to improve screening practices and therapy of diabetic patients due to more accurate diagnosis and treatment of important comorbidities . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results/Discussion"
] | [] | 2015 | Quantification of Diabetes Comorbidity Risks across Life Using Nation-Wide Big Claims Data |
Symptomatic acute schistosomiasis mansoni is a systemic hypersensitivity reaction against the migrating schistosomula and mature eggs after a primary infection . The mechanisms involved in the pathogenesis of acute schistosomiasis are not fully elucidated . Osteopontin has been implicated in granulomatous reactions and in acute hepatic injury . Our aims were to evaluate if osteopontin plays a role in acute Schistosoma mansoni infection in both human and experimentally infected mice and if circulating OPN levels could be a novel biomarker of this infection . Serum/plasma osteopontin levels were measured by ELISA in patients with acute ( n = 28 ) , hepatointestinal ( n = 26 ) , hepatosplenic ( n = 39 ) schistosomiasis and in uninfected controls ( n = 21 ) . Liver osteopontin was assessed by immunohistochemistry in needle biopsies of 5 patients . Sera and hepatic osteopontin were quantified in the murine model of schistosomiasis mansoni during acute ( 7 and 8 weeks post infection , n = 10 ) and chronic ( 30 weeks post infection , n = 8 ) phase . Circulating osteopontin levels are increased in patients with acute schistosomiasis ( p = 0 . 0001 ) . The highest levels of OPN were observed during the peak of clinical symptoms ( 7–11 weeks post infection ) , returning to baseline level once the granulomas were modulated ( >12 weeks post infection ) . The plasma levels in acute schistosomiasis were even higher than in hepatosplenic patients . The murine model mirrored the human disease . Macrophages were the major source of OPN in human and murine acute schistosomiasis , while the ductular reaction maintains OPN production in hepatosplenic disease . Soluble egg antigens from S . mansoni induced OPN expression in primary human kupffer cells . S . mansoni egg antigens induce the production of OPN by macrophages in the necrotic-exudative granulomas characteristic of acute schistosomiasis mansoni . Circulating OPN levels are upregulated in human and murine acute schistosomiasis and could be a non-invasive biomarker of this form of disease .
Schistosomiasis is a severe tropical disease caused by Schistosoma spp . flatworms that affects over 200 million of people from 76 countries and territories [1] . S . mansoni is the only species in the Americas and Brazil holds the majority of infected individuals with 25 million living in endemic areas and 4–6 million infected [2] . Infected individuals have various clinical manifestations that generally cluster into three distinct forms of the disease: acute , hepatointestinal and hepatosplenic schistosomiasis [2–5] . In patients from endemic areas , the acute phase of schistosomiasis is rarely symptomatic ( 0 . 3% ) due to infection early in life ( 3–4 years-old ) and exposure to schistosoma antigens/antibodies against antigens in-utero and/or in breast milk [3] . The majority of chronically infected patients from endemic areas ( 90–96% ) develop the hepatointestinal form of the disease , which is asymtomatic or oligosymptomatic in most cases and characterized by granulomatous inflammation in the liver and intestines , little or no hepatosplenomegaly , and minimal liver fibrosis without any sign of portal hypertension [2 , 4–7] . A small proportion ( 4–10% ) of infected individuals from endemic areas develops the hepatosplenic form of disease characterized by hepatosplenomegaly , severe liver fibrosis and portal hypertension [2 , 4–7] . Among individuals from non-endemic areas , the acute form of schistosomiasis mansoni is a systemic hypersensitivity reaction against the migrating schistosomula ( pre-postural phase of infection ) and mature eggs ( post-postural phase of infection ) . This typically develops within 16–90 days after a primary infection [2] . The burden of infection ( and probably host genetic background ) dictates the severity of the clinical manifestations: more worm couples produce more eggs and consequently , trigger an exacerbated host immune response [2 , 8] . The pre-postural phase occurs during the initial 35 days after infection and is caused by immune modifications induced by the schistosomules , immature and adult worms before laying eggs [2] . Cercarial dermatitis may occur soon after infection , but symptoms are more evident when schistosomules/immature worms arrive/grow/mature in the hepatoportal veins ( peak 15–21 days post infection ) [2] . High fever ( 38–39°C ) , cough , abdominal pain , discrete hepatosplenomegaly and nonspecific symptoms such as muscular pain , arthralgia and headache , are observed [2 , 9] . Blood eosinophilia ( 10–75% of eosinophils ) is frequent [2 , 9] . Liver biopsy reveals discrete inflammatory infiltrate consisting of lymphocytes , eosinophils , neutrophils and macrophages surrounding schistosomules/immature worms , non-specific portal hepatitis , and sparse focal intralobular necrosis [2] . During this phase a Th1 response is predominant and an increase in pro-inflammatory cytokines such as IL-2 , gamma Interferon and TNF alpha is frequently observed [2 , 3] . Other less frequent clinical manifestations may be present: transverse myelitis or pseudotumoral lesions in encephalon ( neural schistosomiasis ) [2 , 9] . The post-postural phase is initiated by egg laying ( approximately 35 days post-infection ) and egg maturation ( which begins about 6 days later ) [2] . Symptoms are aggravated , episodes of diarrhea increase , and the patient experience severe weight loss [2 , 4 , 8] . Clinical symptoms can continue until 90 days after infection [2 , 4 , 8] . Severe , toxemic forms of acute disease in which there is massive dissemination of eggs throughout the intestines and lungs may be fatal [2] . Moderate to mild disease spontaneously resolves two to three months after infection [2] . During acute schistosomiasis intense miliary distribution of eggs occurs in the liver , colons , small intestines , visceral peritoneum , abdominal lymph nodes , pancreas and lungs [2] . Periovular granulomas localize on the serosal surface of affected organs and macroscopically appear as translucent granule or nodules [2] . Microscopically , the granulomas are large ( over 100 times the size of the egg ) , necrotic-exudative , and enriched with eosinophils [2] , due to the naïve hosts’ hyperergic reaction to novel parasite antigens . In the liver , granulomas are irregularly distributed through the parenchyma and portal tracts and non-specific inflammatory cells frequently surround portal tracts . Because hepatocellular lesions are relatively mild ( loss of basophilia , hydropic degeneration and rare focal necrosis ) , the serum aminotransferases are usually normal or slightly elevated [2 , 9] . An important feature of acute schistosomiasisis is that all the granulomas are uniformly in the same necrotic-exudative phase of formation , with prominent central necrosis [2] . This finding in liver biopsies is pathognomonic of acute infection . With egg-laying the Th2 immune response starts to suppress the initial Th1 response and IL4 , IL5 , IL10 and IL13 are the most predominant cytokines [2 , 3] . The hyperergic , massive granulomas are modulated as the infection evolves to the chronic phase . By around 90 days post-infection , liver granulomas are smaller [2 , 3 , 10–12] and progressively heal by fibrosis [2 , 7 , 10 , 12] . The symptoms usually disappear due to the modulation of the immune response to the eggs [2] . Because the signs and symptoms of acute schistosomiasis are nonspecific and diagnosis is established by presence of eggs in stools that occurs only six weeks after infection , acute schistosomiasis mansoni is frequently misdiagnosed , under diagnosed or has delayed diagnosis [2 , 9] . Efforts to develop tests for earlier diagnosis of the disease have been challenging . Unfortunately , lesions similar to those observed in pre-postural phase of human acute schistosomiasis are not observed in mouse models of schistosomiasis mansoni [11 , 13 , 14] , likely because the granulomas that form in mice are generally less necrotic than those that occur in acutely infected humans [14] . Osteopontin ( OPN ) , a pro-inflammatory cytokine and pro-fibrogenic molecule [15–17] , was recently associated with hepatosplenic schistosomiasis mansoni [18] . Soluble egg antigens ( SEA ) directly induce liver cells to produce OPN . Moreover , serum and hepatic osteopontin levels correlate with the degree of liver fibrosis and the level of portal hypertension , suggesting that this molecule could be a novel biomarker for hepatosplenic schistosomiais mansoni [18] . The authors observed that macrophages , stellate cells and bile ductular cells in/around the granulomatous reaction are the major sources of OPN in schistosomiasis [18] . Osteopontin was also demonstrated to play a role in recruitment and activation of macrophages/Kupffer cells , neutrophils and lymphocytes [15–17 , 19] . OPN-/- mice injected with S . mansoni eggs develop abnormal granuloma formation in the lung due to reduced macrophage accumulation [20] . Since in acute schistosomiasis the liver is enriched with necrotic-exudative granulomas and there is an exacerbated immune response , our aims were to evaluate if OPN increases in acute Schistosoma mansoni infection of both humans and mice , and to determine if circulating OPN levels might be a novel biomarker of this infection .
This was a comparative cross-sectional study . A total of 28 patients with acute schistosomiasis mansoni diagnosed at Tropical Diseases Outpatient Clinic of the University Hospital of Universidade Federal de Minas Gerais ( Belo Horizonte , Brazil ) from January 2014 to December 2015 were included in the study . Serum samples from acute patients ( n = 28; age 19 . 8±11 . 8 years; 21 males/7 females ) and uninfected controls ( n = 21; age 27 . 86±9 . 45 years; 14 males/7 females ) were collected for analysis . Formalin-fixed , paraffin-embedded liver needle biopsies were available in a subgroup of patients ( n = 5 ) . Plasma samples from uninfected controls ( n = 21 ) and from patients with Hepatointestinal ( n = 27; age 35 . 66±12 . 09 years; 16 males/7 females ) , Hepatosplenic ( n = 39; age 38 . 25±9 . 4 years; 30 males/9 females ) and Acute ( n = 3; age 39±25 . 23; 3 males ) schistosomiasis were also included in the analysis . Diagnosis of acute schistosomiasis was based on epidemiological data ( recent contact with stream water in an endemic area ) , clinical data ( cercarial dermatitis , acute enterocolitis , fever , cough , malaise , paraplegia , pulmonary involvement , hepatomegaly and or splenomegaly ) , laboratory assays ( eosinophilia , IgG antibodies against SWAP , S . mansoni eggs in stools or rectal biopsy fragments ) , and imaging techniques ( Ultrasound to observe liver , spleen and intra abdominal lymph node enlargement; MRI to demonstrate spinal cord injury ) . To be considered as having acute schistosomiasis in the present study the participants had to have more than 1 or more symptoms/signs described above , evidence of infection ( parasitologic or serologic ) and reported contact with contaminated waters . All patients included in the study were residents of the metropolitan region of Belo Horizonte ( capital of Minas Gerais state ) , a non-endemic area for schistosomiasis mansoni . No previous history of contact with S . mansoni was reported by the patients or parents/guardians . The present study was conducted in accordance with the Declaration of Helsinki ( 2013 ) of the World Medical Association and was approved by the Ethics Committee of Universidade Federal de Minas Gerais , Belo Horizonte , Minas Gerais , Brazil ( UFMG ) ( Protocol ETIC 204/06 ) . Written informed consent was obtained from all participating subjects or their parents/guardians ( on behalf of child participant ) . All data regarding human participants was anonymized . Female Swiss Webster outbred mice were infected with 50 cercariae of S . mansoni ( Feira de Santana strain , CPqGM/FIOCRUZ ) for 6 , 7 , 8 weeks ( acute phase , n = 15 ) and 30 weeks ( chronic phase , severe fibrosis , n = 8 ) . Uninfected , age- and strain-matched animals were used as controls ( n = 8 ) . Liver tissue and serum were collected for analysis . The present study protocol meet the regulation and guidelines of Brazil’s National Animal Experimentation Control Board ( CONCEA ) and was approved ( Protocol 003/2010 ) by the Ethical Committee for Animal Research of Centro de Pesquisas Gonçalo Moniz , Oswaldo Cruz Foundation , Salvador , Bahia , Brazil ( CPqGM/FIOCRUZ ) . OPN was quantified in the serum ( humans and mice ) or plasma ( humans ) using OPN Quantikine ELISA kit ( R&D Systems ) according to the manufacturer’s protocol . Liver sections were stained with H&E ( haematoxylin and eosin ) for general histology . Immunohistochemistry ( IHC ) analysis was performed to evaluate the expression of osteopontin ( R&D Systems; Antigen retrieval: 3% pepsin digestion for 10 min at 37°C; 5ug/mL of primary antibody , incubation overnight at 4°C ) . To confirm that Macrophages produce osteopontin , double IHC was performed using the chromagen DAB ( 3 , 3_-diaminobenzidine ) for OPN and the chromagen Vina Green for CD68 ( a macrophage marker ) . OPN staining was quantified in 15 x200 fields/sample by computer-assisted morphometry using MetaMorph ( Universal Imaging Corp . ) . OPN ( + ) bile ducts were counted in 15 x200 fields/sample by three independent observers . The SEA was prepared at Centers for Disease Control and Prevention ( CDC ) as previously described [21] . The amount of Gram-negative bacterial endotoxin present in the SEA preparation was quantified using the end-point chromogenic limulus amebocyte lysate assay ( Lonza ) . To investigate if macrophages produce osteopontin , primary human Kupffer cells ( from Thermo Fisher Scientific ) were incubated with 10 μg/ml SEA or 0 . 0001 μg/ml LPS ( lipopolysaccharide; control , same amount of endotoxin present in the SEA preparation ) for 3 , 6 , 12 and 24 hours . RNA was collected for analysis . RNA was extracted using RNeasy mini kit ( Qiagen ) according to the manufacture’s protocol . Reverse transcription was performed using the First Strand Superscript III kit ( Life Technologies ) using the random hexamers protocol . Osteopontin mRNA expression was evaluated by real-time PCR ( Taqman , Thermo Fisher Scientific ) . Each sample was analysed in duplicate and target gene levels in treated cells are shown as a ratio to levels detected in corresponding control samples , according to the ΔCT method , relative to the housekeeping gene ( 18s ) . The probes were designed by Thermo Fisher Scientific . Results are expressed as means ± S . E . M . ( Standard Error of the Mean; for normal distribution variables ) or as medians ( for non-normal distribution variables ) . Comparisons between groups were performed using the oneway ANOVA and Student’s t test ( parametric ) or Kruskal–Wallis one-way ANOVA and Mann–Whitney U test ( non-parametric ) . Significance was accepted at the 0 . 05 level; Bonferroni correction was applied when comparing more than two groups . Receiver operating characteristics ( ROC ) curve analysis was used to investigate if sera OPN levels could be a good biomarker for symptomatic acute schistosomiasis . All statistical analyses were performed using SPSS Statistics 22 ( IBM ) and Prism 6 ( GraphPad ) .
Patients with acute schistosomiasis have increased circulating levels of osteopontin in the plasma ( p = 0 . 0005 vs non-infected; p = 0 . 0005 vs HI and p = 0 . 0012 vs HS ) ( Fig 1A ) and serum ( p = 0 . 0001 ) ( Fig 1B ) . The plasma OPN levels in acute schistosomiasis are even higher than in patients with hepatosplenic form of the disease ( p = 0 . 0012 ) ( Fig 1A ) . We observe that OPN starts to increase in the beginning of the post-postural phase ( 5–6 weeks post-infection , p = 0 . 0005 vs non-infected ) and OPN levels peaked 7–11 weeks post-infection ( p = 0 . 0001 vs uninfected; p = 0 . 04 vc 5–6 weeks; p = 0 . 001 vs 12 weeks and p = 0 . 0001 vs 24 weeks ) , when the livers are enriched with necrotic-exudative granulomas ( Fig 1C ) . Twelve weeks after infection the symptoms start to disappear , the granulomas reach a modulated state and circulating OPN levels start to fall , reaching levels comparable to uninfected individuals 24 weeks post-infection ( Fig 1C ) . Receiver operating characteristics ( ROC ) curve analysis demonstrated that serum OPN measurement could be a good biomarker to identify patients with symptomatic acute schistosomiasis mansoni ( Area under the curve = 0 . 9959; p<0 . 0001; 95% confidence interval 0 . 9848–1 . 007; S1 Fig ) . In our study population OPN serum test >23 . 34 can detect a symptomatic acute patient with 95 . 65% sensitivity and 95 . 24% specificity ( Likelihood ratio = 20 . 09 ) . Immunohistochemistry demonstrated that the inflammatory cells in the necrotic-exudative liver granulomas express OPN , especially in the macrophage ( epithelioid cells ) enriched area around the egg and central necrosis ( Fig 1D and S2 Fig ) . Similar to humans , mice in the acute phase of infection also have more circulating and hepatic OPN levels than mice in the chronic phase of infection where there is severe fibrosis ( p = 0 . 001 vs Non-infected; p = 0 . 0124 vs chronic phase ) ( Fig 2A , 2C and 2D ) . OPN levels in mice also peaked in the liver ( p = 0 . 0001 vs non-infected; p = 0 . 0245 vs 6 weeks and p = 0 . 0104 vs 30 weeks ) and serum ( p = 0 . 0286 vs non-infected; p = 0 . 0286 vs 6 weeks; p = 0 . 0286 vs 8 weeks and p = 0 . 004 vs 30 weeks ) 7 weeks post-infection , at a time when the livers were enriched with necrotic-exudative granulomas and inflammatory cells ( Fig 2B , 2C and 2D ) . During the acute phase of infection in both mice and humans , the majority of liver OPN producing cells are inflammatory cells ( Figs 1D and 2C; S2 Fig ) , while the ductular reaction is the most important source of OPN in chronic schistosomiasis ( Fig 2C and 2E ) . OPN expression in both human and murine acute schistosomiasis is enriched in the macrophage area of the necrotic-exudative granulomas . Double immunohistochemistry for OPN and CD68 ( a macrophage marker ) confirmed that the macrophages in acute schistosomiasis express this pro-inflammatory cytokine ( Fig 3A ) . Since the macrophages are in contact with egg antigens , we investigated if soluble egg antigens could stimulate OPN production in vitro . Primary human Kupffer cells incubated with SEA for 3 hours upregulated OPN mRNA ( p = 0 . 0082 ) ( Fig 3B ) , indicating that infection per se can directly increase macrophage expression of this proinflammatory and profibrogenic molecule .
We demonstrated for the first time that circulating osteopontin levels are increased in human acute schistosomiasis mansoni . Our results also suggest that serum OPN measurement could be a good biomarker to diagnose symptomatic acute schistosomiasis . The highest levels of OPN were observed in patients during the peak of clinical symptoms ( 7–11 weeks post infection ) . Once the granulomas were modulated ( >12 weeks post infection ) the OPN levels decrease significantly . Circulating and hepatic OPN levels were also elevated in the acute phase of experimental murine schistosomiasis mansoni . Chen et al . ( 2011 ) demonstrated that liver OPN levels peaked at the acute phase of S . japonicum infection . As previously mentioned , the murine model has some limitations in regard to acute schistosomiasis [14] . However the model may be helpful to identify the factors related to the onset of the generalized reactive changes during the early course of a primary schistosomal infection [14] . Importantly , our new data in humans demonstrate that the mouse model mirrored the human disease with regards to the pattern of OPN expression , reinforcing that this model could be useful to understand the mechanisms related to the acute phase of schistosomiasis in humans . Macrophages are the major OPN producing cell in acute schistosomiasis and SEA induces OPN expression in primary human Kupffer cells . Pereira et al . ( 2015 ) also observed that macrophages are one of the major sources of OPN in the early phases of infection in mice and in patients with hepatointestinal schistosomiasis , while bile ducts are the main producers of OPN in patients with hepatosplenic disease . We confirm that osteopontin is mostly expressed by the ductular reaction in mice in the late chronic phase of infection . Pereira et al . ( 2015 ) also observed that SEA stimulates primary mouse Kupffer cells , stellate cells and cholangiocytes to produce OPN , demonstrating that egg antigens directly induce the expression of this pro-inflammatory and pro-fibrogenic molecule by multiple types of cells that localize in schistosoma-infected livers . Osteopontin has been previously associated with acute hepatic injury [16 , 17 , 22 , 23] . Patients with acute liver failure of different etiologies such as acetominophen toxicity , ischemia ( shock ) , idiosyncratic drug-induced liver injury , autoimmune hepatitis and viral hepatitis A and B , have increased OPN plasma levels [22 , 23] . Recent findings indicate that OPN plays a central role in liver diseases associated with necrosis [16 , 17 , 23] . Liver injury triggers OPN production in Kupffer cells and NKT cells that attract neutrophils , lymphocytes and macrophages to affected areas [16 , 17 , 19 , 24] . The recruited cells become activated and produce OPN and Th1 cytokines , exacerbating liver necrosis [16 , 17 , 19 , 24] . In acute liver failure patients , OPN was particularly associated with hyperactute injury [23] . The role of OPN has been described in granulomatous reactions , especially Th1-mediated , [16 , 19] . OPN is essential for Th1 polarization [25] and OPN from dendritic cells mediates granuloma formation against bacterial antigens [26] . OPN expression in sarcoidosis , tuberculosis and other Th1-mediated granulomas is more associated with macrophages than extracellular matrix [27] . Using the B-glucan model , Morimoto et al . ( 2004 ) demonstrated that OPN-/- mice have a reduction in granuloma size and number and a 2-fold decrease in macrophage accumulation [28] . Overexpression of OPN increased granuloma formation and delayed its resolution , promoting an exacerbated fibrotic response [28] . Similar findings were observed by O’Regan and coworkers ( 2008 ) in S . mansoni egg-induced lung granulomas , a typical Th2-mediated granuloma [20] . Our results confirm the pivotal role of OPN in the Th1 and Th2 mediated granulomas and demonstrate that pathogen antigens directly induce OPN production by macrophages . Acute schistosomiasis is a systemic hypersensitivity reaction against S . mansoni and it is characterized by miliary distribution of hyperergic necrotic-exudative granulomas [2] . The live miracidia inside the egg secrete a series of antigens and lytic substances that can trigger OPN production , recruiting inflammatory cells and inducing the granulomatous reaction to prevent further liver damage ( Th1 over Th2 response ) [2 , 3 , 10 , 18] . As disease progress ( Th2 over Th1 response ) , the granulomas are modulated ( decrease in IFN-gamma and increase in IL10 ) , the antigens and lytic substances are sequestered , necrosis is no longer observed and OPN is down regulated [2 , 3 , 10 , 12] . Patients that will develop hepatosplenic schistosomiasis continue to produce OPN , especially by the ductular reaction , promoting fibrosis and portal hypertension [18] . The plasma levels in acute schistosomiasis were even higher than observed in hepatosplenic patients . Although OPN was demonstrated to be stable in both serum and plasma , OPN levels in the serum are 3 . 8–4 . 8 times lower than in plasma [29] . The authors speculate that this phenomenon may reflect OPN sequestration by the clot or its cleavage by thrombin , leading to loss of immunoreactivity [29] . In our cohort of acute patients only a small number of individuals had both plasma and serum samples collected and we also observed a 4–4 . 5 times reduction of OPN levels in serum compared to plasma ( S1 Table ) . Ideally , future studies should use plasma samples in order to measure the total amount of circulating osteopontin . In conclusion , S . mansoni egg antigens induce the production of OPN by macrophages in the necrotic-exudative granulomas characteristic of acute schistosomiasis mansoni . Circulating OPN levels are upregulated in human and murine acute schistosomiasis and could be a non-invasive biomarker of this form of disease . | Schistosomiasis is a major health problem that affects over 200 million people . Symptomatic acute schistosomiasis is a systemic reaction to the worms and eggs in individuals from non-endemic areas after a primary infection . Tourists , military personnel and people who practice water sports are at risk . Although most cases resolve 90 days post infection , severe cases with massive distribution of eggs can be fatal . It is frequently misdiagnosed , under diagnosed or has delayed diagnosis because the signs and symptoms are nonspecific and eggs are usually present in stool only 6 weeks post-infection . The mechanisms underlying the pathogenesis of acute schistosomiasis are not fully elucidated and currently there is a lack of noninvasive biomarkers to diagnose this form of disease . We report that serum osteopontin levels are increased in patients with acute schistosomiasis and parallel the clinical symptoms , returning to baseline level once the granulomas were modulated and the symptoms resolve . Soluble egg antigens provoke macrophages to produce osteopontin , recruiting more macrophages to the site of injury and inducing the granulomatous reaction . This observation suggests that osteopontin plays an important role in acute schistosomiasis mansoni and could be a novel non-invasive biomarker for this form of the disease . | [
"Abstract",
"Introduction",
"Material",
"and",
"Methods",
"Results",
"Discussion"
] | [
"blood",
"cells",
"kupffer",
"cells",
"schistosoma",
"invertebrates",
"schistosoma",
"mansoni",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"enzyme-linked",
"immunoassays",
"helminths",
"granulomas",
"immunology",
"tropical",
"diseases",
"biomarkers",
"par... | 2016 | Osteopontin Is Upregulated in Human and Murine Acute Schistosomiasis Mansoni |
The World Health Organization , the World Organization for Animal Health , and the Food and Agriculture Organization have resolved to eliminate human rabies deaths due to dog bites by 2030 , and the Vaccine Alliance ( Gavi ) has added human rabies vaccines to their investments for 2021–2025 . Implementing these goals cost-effectively and sustainably requires understanding the complex connections between dog rabies vaccination and human risk and response . The objective of this paper is to estimate how dog rabies vaccinations affect human rabies deaths , mediated through dog rabies cases , dog bite reporting , and post-exposure human rabies vaccination . To approach this objective , we apply multivariate regression analysis over five rabies-related outcomes: ( a ) dog vaccinations , ( b ) dog rabies cases , ( c ) reported human exposures , ( d ) human post-exposure prophylaxis ( PEP ) use , and ( e ) human rabies cases . Analysis uses aggregate annual data over 1995–2005 for seven Latin American countries that experienced dramatic declines in canine and human rabies . Among other results , we estimate the following . ( i ) A 10% increase in dog vaccinations decreases dog rabies cases by 2 . 3% . ( ii ) Reported exposures decline as concurrent dog rabies cases decline , but these declines are more than offset by increases in reported exposures per dog rabies case , which may result from higher rabies awareness due to anti-rabies campaigns . ( iii ) A 10% increase in PEP use decreases human deaths by 7% , but a 10% increase in dog vaccination induces a 2 . 8% decrease in PEP use . The net effect is that a 10% increase in dog vaccination reduces human deaths by 12 . 4% overall , although marginal effectiveness declines as dog rabies incidence declines . ( iv ) Increases in income and public health expenditures increase PEP demand . The findings highlight the importance of mass dog vaccination , heightened awareness , treatment access , and clinical algorithms to reduce both false negatives leading to death and false positives leading to costly unnecessary PEP prescriptions .
In November 2018 , The Vaccine Alliance ( Gavi ) added human rabies vaccines to investments for 2021–2025 [1] . This investment in human vaccine for post-exposure prophylaxis aligns Gavi with the tripartite goal of the World Health Organization , World Animal Health Organization , and Food and Agriculture Organization to eliminate rabies as a cause of human death [2] . While rabies was ranked high for its public health impact in Gavi’s analysis for vaccine prioritization , concern was raised regarding the operational complexity of human post-exposure prophylaxis ( PEP ) as cost effective control requires progressive reduction in the zoonotic transmission of rabies and thus decreased demand for PEP over time . Given that bites from rabid dogs are responsible for over 95% of the annual 59 , 000 human deaths [3 , 4] , sustainable elimination of human deaths will likely require both Mass Dog Vaccination ( MDV ) and improved access to PEP for bite victims [5 , 6] . With sufficient coverage , MDV is effective in eliminating transmission source and is significantly more cost effective than relying solely on PEP [7 , 8] . Nonetheless , access to PEP is necessary following bites of unvaccinated dogs or dogs for which the vaccination and disease status is uncertain [9 , 10] . Understanding the budgetary commitment required for eliminating and maintaining elimination on a national and regional scale is essential , especially given that the remaining burden of dog mediated rabies falls exclusively on low- and middle-income countries ( LMIC ) [4] . Reliance on PEP alone represents a continuous cost burden [3 , 8] . In contrast , when PEP is combined with MDV reaching consistent 70% coverage , models based on studies in sub-Saharan Africa suggest that the total cost per averted death would decrease relatively rapidly [8] . The most compelling evidence for this has been witnessed in eight Latin American countries where national rabies control programs initiated a regional elimination effort in 1983 incorporating both MDV and improved access to and subsidization of PEP [12 , 13] . This effort , although still short of complete elimination [11 , 14–16] , resulted in a sustained 95% reduction in human rabies deaths by 2012 [11 , 14–16] . However , the evidence from these sustained national campaigns illustrates not only the effectiveness of MDV , but also suggests a cautionary note: While the number of rabid dogs dramatically decreased , consistent with the efficacy of MDV , the number of doses of PEP has increased in recent years [16] . The objective of this paper is to quantitatively estimate the relationships among dog vaccinations and human rabies deaths , mediated through dog rabies cases , while controlling for dog bite reporting and PEP use . To do so , we use an integrated econometric model of the relationships among i ) dog vaccinations , ii ) dog rabies cases , iii ) suspected human exposure; iv ) PEP use , and v ) human rabies cases . The model accounts for intertemporal dynamics that relate current and past vaccinations and dog rabies cases to be propagated over time , and accounts for PEP demand , reflecting both the perceived risk of acquiring rabies given an exposure , and the reduction in its likelihood . The model allows estimation of an extensive set of of indirect and conditional epidemiological relationships and human decisions that drive the connection between dog vaccinations and human rabies death .
Our data represent seven countries ( Brazil , Colombia , Peru , Venezuela , Nicaragua , Dominican Republic , Mexico ) , which were substantially affected Latin American countries between 1995 and 2005 , and for which we have the most reliable data . The data were compiled from a series of reports published by the Reunión de Directores de los Programas de Rabia de las Américas ( REDIPRA ) from 1995–2006 , based on surveys implemented by the Pan American Health Organization in tandem with semiannual REDIPRA meetings [13] . Table 1 summarizes data descriptions and sources , and Table 2 provides summary statistics . The data span 1995 through 2005 for seven countries , for a potential sample frame of N ( max ) = 77 . However , dog population estimates are missing for 1995 and 1996 , and vaccination data are missing for 1995–1997 . Further , regression sample sizes are also limited because of the use of lagged values in the regressions . Table 3 includes summary statistics for all observations for each variable that are used in any of the regressions reported in results . Fig 2 shows time trends for the five focal variables , by country and in aggregate . Fig 3 provides six panels , including dog rabies cases per dog vaccination , reported exposures per dog rabies case , human cases per dog case , PEP completions per dog case , PEP completions per exposure , and PEP completions per human rabies case . Additional factors hypothesized to affect Reported Exposures and PEP Completions ( Zi , t-jHandZi , t-jP in Fig 1 ) were collected from the World Bank Open Data website [23] , including human Urban Population ( Hi , tu ) and Rural Population ( Hi , tr ) , [per capita] Income ( Ii , t ) , [per capita] Health Expenditures ( Xi , t ) , and Out of Pocket [health expenditures] ( Oi , t ) . Additional data issues , including missing dog population data , two potential anomalies in the Brazilian data , and that reported exposures are not limited to dog bites are discussed in S1 File . Freire de Carvalho et al . ( 2018 ) [16] , whose data overlap with ours to some degree , discuss some additional limitations of the data , mostly having to do with potential shortcomings of the reporting process for PEP . The limited sample size is among most pressing limitations of our study , but the overall trends in mass dog vaccination , dog and human rabies cases , and their relationship to reported exposures and PEP completions tend to be relatively robust . We estimate five regressions: one each for Dog Vaccinations ( Vt ) , Dog Rabies Cases ( Ri , tD ) , Reported Exposures ( Ei , t ) , PEP Completions ( Pi , t ) , and Human Rabies Cases ( RtH ) . The data are in the form of a panel of country-specific time series , so we use panel regression models for each . We hypothesize that Dog Vaccinations , Dog Rabies Cases , and Reported Exposures ( Vi , t , Ri , tD , Ei , t ) exhibit some dynamic autoregressive relationship . We therefore use an Arellano-Bond panel estimator for these regressions for statistical consistency [24 , 25] . Reported Exposures and PEP Completions accumulate based on decisions driven by subjective rabies risk assessment and other underlying supply and demand factors . Further , Reported Exposures may be an endogenous determinant of PEP Completions , and PEP Completions may be an endogenous determinant of Human Rabies Cases . We therefore test for endogeneity of these variables for use as regressors in other regressions using a Durbin-Wu-Hausman Test . We fail to reject exogeneity of Reported exposures ( Ei , t ) in the PEP Completions regression , and reject exogeneity of PEP Completions ( Pi , t ) in the Human Rabies Cases regression . We therefore treat Reported Exposures as exogenous and use the original values for Ei , t in the PEP Completions regression , but use predicted values from the PEP Completions regression as an instrument in the Human Rabies Cases regression . All dependent variables are non-negative count data and wide domains allow most to be treated as continuous variables . Exploratory analysis suggested that regression error distributions approximate a lognormal distribution for Vi , t , Ri , tD , Ei , t , and Pi , t , so we transformed each of these dependent variables into natural logarithms for estimation . Human rabies cases Ri , tH are smaller in number for each country/year than the rest . Specification search and tests lead us to use a Negative Binomial regression in this case . Consequently , parameter estimates of all continuous variables in each regression ( including the Ri , tH regression ) can be interpreted as elasticities ( i . e . the percent changes in the dependent variable in response to a one percent change in the associated explanatory variable ) . Robust standard errors are applied throughout . Analysis was performed using Stata version 14 . 2 . Our theoretical model is built on concepts of causality ( e . g . dog vaccination is a treatment that systematically leads to fewer dog rabies cases , ceteris paribus; a higher rate of PEP use per dog bite reduces the proportion of human deaths among dog bite victims ) . Further , the endogeneity of decision variables such as self-reporting and PEP uptake are accounted for in the empirical estimation in part to try to identify causality where our theoretical model suggests it . However , limitations of the sample frame and data ( omitted variables , small sample , data collection errors ) imperfect instruments for testing and addressing endogeneity , and any remaining specification errors suggest caution in inferring strict quantitative causality among the estimated relationships . While we use language of causation in the presentation of some of the results for expediency and clarity , especially in terms of treatment effects , implications of causality in our results should be inferred with caution . Additional methodological details and data limitations are provided in S1 File .
Factors that drive dog rabies cases include current rabies prevalence in the host population , vaccination coverage , and transmission rates . Dog Rabies Cases regression results ( Table 3 , Regression 1 ) suggest that the number of dog rabies cases is positively related to the number of rabies cases in prior years , and negatively affected by prior vaccination activity . The results show that a 10% increase ( decrease ) in a year’s Dog Rabies Cases is associated with 5 . 6% more ( fewer ) cases in the following year ( indicated by the coefficient of 0 . 56 associated with ln ( Rt-1D ) ) . The estimated effects of current year and last year’s vaccinations ( ln ( Vt ) and ln ( Vt−1 ) ) are negative , but statistically weak . Vaccinations conducted two years prior ( ln ( Vt−2 ) ) has a significant but slightly smaller direct effect , with 10% more vaccinations resulting in a 3 . 1% decrease in dog rabies cases ( corresponding to a parameter estimate of -0 . 31 associated with ln ( Vt−2 ) ) . As expected , current and prior year Dog Vaccinations negatively affect current Dog Rabies Cases . A χ2 test suggests that this set of vaccination variables are jointly significant at the 10% level ( χ2 ( 3 ) = 6 . 64 , p = 0 . 084 ) . The cumulative effect of vaccination over a three-year period is -2 . 30: a 10% increase in vaccinations decreases dog rabies over three years by 23% ( see Equation ( S8 ) in S1 File for details ) . In contrast to declines in Dog Rabies Cases and Human Rabies Cases , there was an increase in Reported Exposures ( Fig 2 ) in our sample period , resulting in a marked increase in reported human exposures per dog rabies case ( Fig 3 panel ( b ) ) . Table 3 , Regression 2 provides estimates of the effect on Reported Exposures of past exposure reports , current and recent rabies cases , a time trend , and dog and human populations . Correlation across years in Reported Exposures could be expected because of stable reporting infrastructure including clinics/hospitals , health care providers , and available rabies-related information . There is a statistically strong correlation between current Reported Exposures and past Dog Rabies Cases: 10% fewer Dog Rabies Cases in the current year or prior year is associated with 1% fewer current Reported Exposures , suggesting that if all other factors were held constant , Reported Exposures would have declined along with declines in Dog Rabies Cases . When the effects of current Dog Rabies Cases are controlled for in our regression framework , past Dog Rabies Cases should not change current exposures , and neither should last year’s Human Rabies Cases ( as humans are terminal hosts ) . However , even while statistically controlling for dog rabies cases in Regression 2 , the results show a positive relationship between current Reported Exposures and both Dog and Human Rabies Cases in the previous year . This suggests that past human and dog rabies cases are salient events that inform people’s subjective rabies risk assessment , and they act on the information by reporting current exposure events more often when there have been recent publicized dog and human rabies cases . A time trend ( measured in years from 1994 ) in logarithmic form ( ln ( t ) ) is included in Regression 2 to control for unobserved factors that may be driving exposure reporting . Its associated parameter estimate is strongly positive . Reported Exposures result from a complex mix of private decisions to seek care , including individuals’ assessment of risk given both case-specific ( e . g . wound severity , behavior of and familiarity with the dog , knowledge of dog vaccination status ) and non-case-specific factors ( e . g . formal and informal information about rabies circulating in the community ) . We hypothesize that the increase in Reported Exposures is at least partly due to increasing awareness of rabies during this period of active dog vaccination campaigns [26] . Although the actual risk of human rabies decreased during the period due to the declines in dog rabies , there was no clear reduction in PEP Completions . Consequently , given the marked reduction in both Dog and Human Rabies Cases during the period , there was a substantial increase in both the number of PEP Completions per Dog Rabies Case and per Human rabies Case ( Fig 3d and 3f , respectively ) . PEP Completions ( Table 3 , Regression 3 ) is a reduced-form regression representing both demand for and supply of PEP ( further discussion in S1 File ) . As with exposure reporting , investing in PEP entails a series of complex and often subjective risk and cost-benefit assessments . Contemporaneous Reported Exposures had a positive but weak relationship to PEP Completions . The mean number of Reported Exposures was two and a half times higher than that of PEP Completions ( Table 2 ) , suggesting that a clinical decision to forgo PEP occurred for many of the reported exposures . There is not a significant relationship between current PEP Completions and current Dog Rabies Cases , but there is a positive relationship between current PEP Completions and prior years’ Dog Rabies Cases: a 1 . 3% increase in PEP Completions per 10% increase in lagged Dog Rabies Cases . Similarly , PEP Completions is positively related to lagged human rabies cases , with an estimated 3% increase in a current year’s PEP use in response to 10% more human rabies cases in the previous year . In comparison with the Reported Exposures regression results , we hypothesize that increased PEP demand is responding to changes in subjective risk assessments which are informed by prior dog and human rabies outbreaks . Per capita Income ( Ii , t ) is strongly associated with PEP Completions ( Table 3 , Regression 3 ) . A 1% percent increase in Income is associated with a 2% increase in PEP Completions , conditional on other factors . PEP Completions is also positively associated with public per capita Health Expenditures ( Xi , t ) , and negatively associated with the proportion of healthcare borne Out of Pocket ( Oi , t ) . The latter is consistent with the income effect in relation to PEP costs , and both results are consistent with access and availability to PEP being perceived to be within the remit of the public health care systems in Latin America . The incidence of human rabies cases ( deaths ) is determined jointly by exposure to a rabid animal and PEP completion , mediated through exposure reporting . PEP use is an endogenous factor affecting human rabies cases because PEP will be completed when risk is high , which in itself would lead to a positive correlation between PEP use and human rabies cases in aggregate data even though PEP use reduces the incidence of death in the exposed population . Using the two-stage instrumental variable method described , a 10% increase in PEP Completions decreases Human Rabies Cases ( death ) by 7% , all else constant ( Table 3 , Regression 4 ) . Without implementing the instrumental variable approach , the estimated impact is spuriously positive ( 0 . 42 , p = 0 . 016 ) , reflecting the use of PEP when risk would be highest . Conditional on PEP Completions , Human Rabies Cases are positively related to Dog Rabies Cases , with elasticity of 0 . 63 . Thus , a 10% increase in Dog Rabies Cases is associated with a 6 . 3% increase in Human Rabies Cases , holding PEP Completions constant . The effect of Dog Vaccinations on Human Rabies Cases can be estimated through its effects on Dog Rabies Cases while accounting for changes in PEP Completions as %ΔRH%ΔV= ( ∂ln ( RH ) ∂ln ( RD ) +∂ln ( RH ) ∂ln ( P ) ∂ln ( P ) ∂ln ( RD ) ) ×∂ln ( RD ) ∂ln ( V ) = ( 0 . 63+ ( −0 . 71× ( 0 . 13−0 . 01 ) ) ) × ( −2 . 30 ) =−1 . 25 . ( 1 ) The numerical values are based on regression parameter estimates ( Table 3 , Regressions 3 and 4 ) , where 0 . 12 = ( 0 . 13 − 0 . 01 ) is the total impact over two years of Dog Rabies Cases ( RD ) on PEP Completions , and -2 . 3 is the three-period ( long-run ) effect of Dog Vaccinations on Dog Rabies Cases . Thus , a 10% increase in Dog Vaccinations reduces human deaths by 12 . 5% over the subsequent three years . The marginal effect of Dog Vaccinations on Human Rabies Cases is ∂RH∂V=%ΔRH%ΔVVRD=∂lnRH∂lnVVRD . Consider the conditions at the beginning and end of our sample: In 1995 , about 100 Human Rabies Cases were reported in these seven countries . By 2005 , this number was under 20 . While we do not have vaccination data for 1995–1997 , the subsequent years’ data indicate that an average of 3 . 5 million dogs were vaccinated annually in these countries combined; an average of 33 million per year prior to 2000 and about 37 million from 2000 to 2005 . Consequently , the marginal effectiveness of vaccination for reducing human deaths for early and late in the sample were: ∂RH∂V|early=∂lnRH∂lnV×V ( early ) RH ( early ) =−1 . 25×10033=3 . 8⇒3 . 8humanlivessavedpermillionvaccinationsperyear , ( 2 ) ∂RH∂V|late=∂lnRH∂lnV×V ( late ) RH ( late ) =−1 . 25×2037=0 . 68⇒0 . 68humanlivessavedpermillionvaccinationsperyear . ( 3 ) The difference between effectiveness early and late in the sample statistically illustrates the “tribulations of the last mile” described in Del Rio Vilas et al . [14] . The percentage change in PEP Completions in response to a percentage change in Dog Vaccinations is embedded in Eq ( 1 ) as: %∂P%∂V=∂ln ( P ) ∂ln ( RD ) ×∂ln ( RD ) ∂ln ( V ) =0 . 12×−2 . 3=−0 . 276 ( 4 ) This implies that a 10% increase in dog vaccinations leads to a 2 . 8% decrease in PEP use . For Mexico , the average annual number of dog vaccinations in the period was approximately 14 million and PEP Completions averaged approximately 17 , 000 per year . One thousand dog vaccinations reduces PEP completions by 0 . 34 units , or equivalently , one PEP completion is avoided per approximately 3 , 000 dog vaccinations , all else constant . This amounts to a net effect of dog vaccination , which decreases the risk of rabies given an exposure , and we speculate also increases public awareness of rabies as a threat to health and of PEP as an option for alleviating the rabies risk following a potential exposure . Income affects the number of human deaths through its effect on PEP use . The percent change in Human Rabies Cases with a percent change in Income is %ΔRH%ΔI=∂ln ( RH ) ∂ln ( P ) ×∂ln ( P ) ∂ln ( I ) =−0 . 71×2 . 05=−1 . 46 , ( 5 ) suggesting that a 1% increase in Income is associated with 1 . 46% fewer human deaths ( 14 . 6% for a 10% increase in income ) , an effect slightly larger in magnitude than the estimated effect of dog vaccination on human rabies cases ( -1 . 25% , or -12 . 5% for a 10% increase in vaccination ) . Per capita annual income ( 2011 base ) ranges from approximately US$3 , 000 ( Nicaragua ) to US$15 , 000 ( Mexico ) , and the mean number of deaths ranges from 0 . 3 ( Nicaragua ) to 19 . 5 ( Brazil ) , respectively . Taking Brazil as an example , with a mean per capita income of about $11 , 000 and an average of 19 . 5 deaths per year in the sample , an increase in per capita income of about $1 , 000 per year would reduce human deaths by about 2 . 6 people per year ( -1 . 46× ( 19 . 511 ) =2 . 6 ) . Given an average number of PEP Completions for Brazil of 22 , 300/year , an increase of per capita income of $1 , 000 would also be associated with about 40 more PEP completions , all else constant .
Elimination of human death due to rabies by 2030 requires reduction of the primary transmission risk ( rabid dogs ) through vaccination , prompt exposure reporting , and universal PEP access . The Latin American example illustrates how coordinated multi-national investment can substantially reduce dog and human rabies over time [27] . However , there has been justifiable concern over an apparent positive connection between successful dog vaccination campaigns and increased expenditures on PEP , because if this connection holds the costs of rabies elimination that result from increased PEP use may be high . PEP demand is driven in part by awareness , salience , and perceptions of rabies risk . Our results suggest that as dog and human rabies cases and risk decline , reported exposures and PEP use tends to decline as well , all else constant . The overall increases in exposures and PEP use appear instead to be driven by other factors , such as awareness and salience of rabies as a risk , which are likely to increase with rabies control efforts and real income and public health investments . If lessons from Latin America hold , potential decrease in dog and human rabies in Africa and Asia through national and regional MDV may not necessarily generate a proportional reduction in PEP demand . These findings about exposure reporting and PEP demand underscore the importance of promoting and supporting integrated bite case management approaches [28 , 29 , 30] , such as clinical algorithms , which allow systematic clinical assessment of potential exposures PEP and treatment determinations following suspect dog bite exposures . Effective , continuously improving algorithms , based on confirmed dog vaccination and health history , quarantine , and laboratory testing , have the potential to reduce costly false-positive PEP prescriptions and false negatives leading to human death . The lessons from Latin America highlighted in our study emphasize the need for ongoing evaluation of human exposure data and PEP use relative to reduction in canine rabies due to MDV and understanding of the drivers of risk perception in addition to actual risk . These factors will be critical in sustaining the national investments , predominantly by low- and middle-income countries , required to meet the 2030 goal of no human deaths due to rabies . | Several global health organizations have prioritized investment for global elimination of human death from canine rabies . Cost-effective deployment requires understanding complex connections between rabies risk and healthcare seeking behavior . During 1995–2005 , there was a rapid decline in dog and human rabies cases in several Latin American countries following concerted investment in mass dog vaccinations and human post-exposure vaccination . Using data from this period , we econometrically estimate relationships between dog vaccinations , dog rabies cases , human exposure reports , human post-exposure prophylaxis ( PEP ) use , and human rabies deaths . We estimate quantitatively how dog vaccinations indirectly reduce human rabies deaths , mediated through dog rabies cases , dog bite reporting , and PEP use . We find that reported human exposures decline as dog rabies cases decline , but these declines are offset by increases in reported exposures per dog rabies case , which may be driven by increased rabies awareness through anti-rabies campaigns . Furthermore , while PEP demand declines as dog rabies cases decline , increases in income and public health expenditures concurrently increase demand for PEP . These findings provide better understanding of the underlying factors driving PEP demand and exposure ( under ) reporting to inform cost-effective policy design to reduce unnecessary PEP prescription and false negatives leading to human death . | [
"Abstract",
"Introduction",
"Data",
"and",
"methods",
"Results",
"and",
"discussion",
"Conclusion"
] | [
"medicine",
"and",
"health",
"sciences",
"post-exposure",
"prophylaxis",
"immunology",
"tropical",
"diseases",
"vertebrates",
"social",
"sciences",
"dogs",
"animals",
"mammals",
"health",
"care",
"vaccines",
"preventive",
"medicine",
"rabies",
"global",
"health",
"negle... | 2019 | Healthcare demand in response to rabies elimination campaigns in Latin America |
Neuronal oscillations are ubiquitous in the human brain and are implicated in virtually all brain functions . Although they can be described by a prominent peak in the power spectrum , their waveform is not necessarily sinusoidal and shows rather complex morphology . Both frequency and temporal descriptions of such non-sinusoidal neuronal oscillations can be utilized . However , in non-invasive EEG/MEG recordings the waveform of oscillations often takes a sinusoidal shape which in turn leads to a rather oversimplified view on oscillatory processes . In this study , we show in simulations how spatial synchronization can mask non-sinusoidal features of the underlying rhythmic neuronal processes . Consequently , the degree of non-sinusoidality can serve as a measure of spatial synchronization . To confirm this empirically , we show that a mixture of EEG components is indeed associated with more sinusoidal oscillations compared to the waveform of oscillations in each constituent component . Using simulations , we also show that the spatial mixing of the non-sinusoidal neuronal signals strongly affects the amplitude ratio of the spectral harmonics constituting the waveform . Finally , our simulations show how spatial mixing can affect the strength and even the direction of the amplitude coupling between constituent neuronal harmonics at different frequencies . Validating these simulations , we also demonstrate these effects in real EEG recordings . Our findings have far reaching implications for the neurophysiological interpretation of spectral profiles , cross-frequency interactions , as well as for the unequivocal determination of oscillatory phase .
Neuronal oscillations are ubiquitous in the human brain , being present in both cortical and subcortical structures . Moreover , they have been shown to be relevant for sensory [1 , 2] , motor [3 , 4] and cognitive [5 , 6] functions . Traditionally , neuronal oscillations as recorded by EEG/MEG are considered to be sinusoidal . This observation is particularly driven by the analysis tools frequently used in neuroscience . These often include Fourier , Morlet wavelet and Gabor transforms , all of which use sinusoids as a basis function [7] . There is no a-priori reason why exactly these basis functions would be most relevant for describing neuronal oscillations ( “Fourier fallacy” [8] ) . Many nonlinear periodic processes in nature are in fact quasi-sinusoidal [9] . For instance , the non-sinusoidal nature of ocean waves has for long time been recognized [10] , where it was emphasized that conventional spectral analysis is not sensitive to the non-sinusoidal nature of periodic processes . Due to the complexity of such waves , analysis in time domain is often suggested and elaborate measures of horizontal and vertical asymmetries have been presented [11] . A similar claim has been recently voiced for large scale neuronal oscillations [12] , which represent a particularly good example where many nonlinearities are present including thresholds , exponential decays and non-linear coupling between neuronal elements . It is therefore not surprising that often neuronal recordings only approximately resemble sinusoidal processes especially when they are obtained with invasive techniques [13 , 14] . This in turn indicates that other concepts and analysis tools are needed for a more adequate description of periodic neuronal processes recorded with EEG/MEG . Waveform was largely neglected in large scale EEG/MEG analysis up until recently [15 , 16] . However , the reasons why non-invasive neuronal recordings rather show sinusoidal oscillations in contrast to invasive recordings have not yet been clearly identified . Some evidence for non-sinusoidality is also visible in the spectral domain , as non-sinusoidal processes are manifested through the presence of additional peaks being usually integer multiples of the base frequency . Spectral harmonic peaks are often observed in LFP and EEG/MEG recordings . For instance , a spectral peak in β-frequency range has been found to be exactly twice the individual α-frequency peak [17–19] . The waveform of oscillations is also important for the understanding of non-linear neuronal interactions . which can be carried out not only within the same frequency band ( e . g . , α , β , γ ) but also across different bands . In this case they are referred to as cross-frequency interactions and describe a mechanism through which spatially and spectrally distributed information can be integrated in the brain [20] . The extent to which the presence of such cross-frequency interactions can be due to spurious effects , particularly due to non-sinusoidal waveform of oscillations is being debated [21–23] . Furthermore , a description of oscillations which takes into account their non-sinusoidal waveform has implications for the understanding of oscillatory phase . Oscillatory phase is important in theories of neuronal processing [24–26] , reflecting a change in membrane potential for many synchronous neurons . This in turn results in changes in cortical excitability , which has been associated with periodic inhibition . A non-sinusoidal waveform is associated with a deviation from a 50% duty cycle and with a non-uniform phase velocity [27] . This in turn would lead to non-uniform changes in cortical excitability and subthreshold stimulus detection rates along the oscillation cycle . Here , we investigate measures for quantifying non-sinusoidality in the time domain , with simulation and analysis primarily focused on α- and β-oscillations in EEG recordings . The aim of the present study is to show that the degree of non-sinusoidality in oscillations may depend on the spatial mixing of the neuronal sources reflected in EEG/MEG/LFP recordings . Depending on synchronization strength and the temporal delay between neuronal populations , the resulting waveform of oscillations can vary from strongly non-sinusoidal to sinusoidal . Spatial mixing will influence measures such as amplitude envelope correlations and α/β-ratio , as different temporal delays will cancel or enhance different frequency components of the non-sinusoidal waveform . Moreover , this might lead to spurious inferences about cross-frequency interactions , which may rather relate to changes in the waveform reflecting in turn changes in spatial synchronization .
The study protocol conformed to the Declaration of Helsinki and by the ethics committee at the medical faculty of the University of Leipzig ( reference number 154/13-ff ) . The EEG data were previously collected as part of the “Leipzig Cohort for Mind-Body-Emotion Interactions” data set ( LEMON ) [28] . Written informed consent was obtained prior to the experiment from all participants . Data from 13 participants were excluded due to missing event information , different sampling rate , mismatching header files or insufficient data quality . Additionally , data from 17 participants was excluded for insufficient signal-to-noise ratio ( see section Data analysis and Statistics ) . This resulted in data sets from 186 participants ( 117 male , 69 female , age range: 20–70 years ) with no history of neurological disease and usage of CNS drugs . Scalp EEG was recorded from a 62-channel active electrode cap ( ActiCAP , Brain Products GmbH , Germany ) , with 61 channels in the international 10-20 system arrangement and one additional electrode below the right eye recording vertical eye movements . The reference electrode was located at electrode position FCz , the ground was located at the sternum . Electrode impedance was kept below 5 kΩ . Data were acquired with a BrainAmp MR plus amplifier ( Brain Products GmbH , Germany ) at an amplitude resolution of 0 . 1 μV with a bandpass filter between 0 . 015 Hz and 1 kHz and with a sample rate of 2500 Hz . The recordings were performed in a sound attenuated EEG booth . The experimental session was divided into 16 blocks , each lasting 60 s , with two conditions interleaved , eyes closed ( EC ) and eyes open ( EO ) , starting in the EC condition . Participants were instructed to fixate on a digital fixation cross during EO blocks . Changes between blocks were announced with the software Presentation ( v16 . 5 , Neurobehavioral Systems Inc . , USA ) . Only data from the EC condition were used for analysis . α/β-ratios and amplitude envelope correlations were evaluated using the compound signal X ( t ) . The compound signal was composed of 20 sources , mixed with temporal delays drawn from a normal distribution with mean 0 and varying values for the standard deviation σ . α/β-ratios were calculated as the ratios of α- and β-SNR values of the compound signal , evaluated over time series segments of varying length . Here , α-SNR was taken as oscillatory power at the α-peak frequency and β-SNR as the oscillatory power at twice that frequency . Power was computed by FFT , Hann window , 1 s window length , 50% overlap . To investigate time courses between α- and β-components of the compound signal , we calculate correlations between their amplitude envelopes . Amplitude envelopes were calculated by individually bandpass-filtering the compound signal in the base frequency range and first harmonic frequency range ±2 Hz , respectively ( Butterworth , filter order = 9 ) . Amplitudes envelopes were determined for each frequency band by the means of the Hilbert transform . Then , the Spearman rank correlation coefficient was calculated between α- and β- amplitude envelopes . The calculation was repeated 1000 times , every time using a new instantiation of the compound signal , sampling new temporal shifts and 1/f-noise for amplitude modulation .
Spatial mixing of non-sinusoidal sources results in more sinusoidal compound signals . Considering the example in Fig 2A , seven basis signals are added with temporal delays drawn from a normal distribution . The compound mean signal has lost its asymmetrical shape and shows no difference between crest and trough periods ( shown for one oscillation cycle in Fig 2B ) , compared to the basis functions . Note that the disappearance of the non-sinusoidal waveform is not due to the changes in SNR but due to the time delay between individual sources . As a temporal delay of e . g . 10 ms is equivalent to 1 5 π for the α-component , but twice as large , 2 5 π for the β-component , this leads to faster attenuation of the β-component . To quantify the attenuation of the faster component , we computed the power spectrum of the compound signal X ( t ) by the Fourier transform as a function of standard deviation of the temporal delays σ . The analytical solution is proportional to exp ( −2 ( π ⋅ σ ⋅ f ) 2 ) , as obtained by Fourier analysis of the compound signal as a function of σ . The quadratic dependency on the frequency term results in a faster attenuation of higher frequencies , as seen in Fig 2C . This results in a more sinusoidal signal for larger values of σ . Not only spectral power , but also the proposed measure for non-sinusoidality in the temporal domain is able to detect non-sinusoidality in the compound signal as a deviation from 0 ( see Fig 2D ) . In our unconstrained simulations , the spectral peak of the β-component can be higher than the spectral peak of the α-component ( see also later sections in the Results ) . In this case , extreme ΔCT values are observed , leading to an increased standard deviation for large temporal delays . The implication is that the degree of non-sinusoidality present in the waveform can serve as an indicator of spatial synchronization . It can also constrain the mixing coefficients , which are known in simulations , but are not known for real EEG recordings . As an example for the ΔCT-measure , we illustrate the Tc and Tt-distributions for different types of EEG oscillations in the 8–13 Hz frequency band for an exemplary participant . After SSD decomposition , one motor and one visual component was identified from the associated activation patterns . We compute Tc and Tt for motor and posterior oscillations , shown in Fig 3 . In this participant , a more non-sinusoidal oscillation can be found for the motor-component . The motor component is typically described as arc- or comb-shaped [33 , 34] and the posterior α-component as symmetric [35] . Using a measure such as ΔCT can quantify waveform shape and shows that also posterior α-oscillations can also be of non-sinusoidal shape in line with [29] . We provide more evidence to demonstrate the non-sinusoidal nature of α-oscillations , see supplementary material , S2 Fig . We quantified the extent to which ΔCT is affected by a demixing procedure , which brings sensor signals closer to their sources . For this , ΔCT was computed in sensor space recordings for all included participants , as well as for SSD-extracted components . SSD components have a higher ΔCT indicating a higher degree of non-sinusoidality across participants ( p = 5 . 7 ⋅ 10−15 , two-sided Wilcoxon signed rank test ) , as illustrated in Fig 4 . We additionally ran our analyses using fastICA instead of SSD . The results are comparable to SSD , with fastICA achieving a higher ΔCT compared to sensor space ( p = 8 . 86 ⋅ 10−4 , two-sided Wilcoxon signed rank test ) . The dependence of the ΔCT of SSD-components to SNR was assessed by computing the SNR in the α-frequency band via 1/f-corrected spectrum and absolute value of ΔCT for all SSD-components with α-SNR > 5 dB . We found a correlation of . 242 ( Spearman’s rho , p < 6 . 99 ⋅ 10−31 ) of absolute ΔCT with α-SNR , with more non-sinusoidal signals as measured by ΔCT for higher SNR . Resorting the absolute ΔCT-values according to their associated β-SNR values shows that a β-SNR-level of the same magnitude as α-SNR of e . g ∼8 dB is associated with higher ΔCT values . In other terms , a pronounced β-peak in the 1/f-corrected spectrum corresponds to a higher degree of non-sinusoidality than an α-peak of the same magnitude . This observation is in agreement with our simulations presented above indicating that the presence of β-oscillations defines non-sinusoidality of the waveform . The topographic distribution of ΔCT-values can be seen in Fig 4D , which shows considerable variation across participants . Although the group average in Fig 4A shows increased values for both central-motor and occipital channels , on a single subject level either a central-motor or an occipital maximum is rather visible . In sum , the non-sinusoidality of EEG recordings is affected by spatial mixing of oscillatory sources and also by SNR in relation to 1/f-noise . A spatial summation of basis signals with the same spectral content but different temporal delays can have differential consequences for the respective constituent frequencies , enhancing or diminishing respective oscillatory power . We provide three examples for this phenomenon .
In the present study , we realized the mixing of signals from individual neuronal populations with unitary weights in the simulations . For empirical recordings , data B recorded with EEG/MEG can in general be represented as B = L ⋅ J , where L is a lead-field matrix and J contains dipole currents at different locations . In our simulations , the sources can be assumed to be located close to each other ( e . g . <5 mm ) and in practical terms their location and orientation could be considered to be approximately the same thus having the same gain in L matrix . In this way the same gain ( unitary or not ) for all sources is justified and would lead to similar results . Already on this spatial scale , sources display great dynamical variety [42] , with diverse temporal delays [43] . Of course , EEG activity reflects the superposition of a large number of other remote sources , where the mixing of signals at the sensor level would occur with different weights . This , however , would not change one of the main findings of the study qualitatively , namely that the mixing of many non-sinusoidal sources results in more sinusoidal signals . From our simulations , it follows that if the amplitude weight from one of the sources would be very large ( far larger than the weight from other sources ) , then the signal would remain strongly non-sinusoidal . Only when weights of other multiple sources have sufficient strength and these sources are not synchronized at exactly zero-lag delay [44] , only then the superposition of the signals results in more sinusoidal signals . At the level of the remote neuronal populations recorded with EEG , this observation has been confirmed in our study . We showed that ΔCT deviated stronger from 0 for SSD components compared to sensor space data since in the latter case effects of the spatial mixing are more pronounced . Consequently , introducing simulations with different spatial weights would only result in superimposed signals having more non-sinusoidal waveform . Even despite relatively simple but neurophysiologically plausible simulations , we are still capable to show the effects of spatial mixing on waveforms and on complex cross-frequency interactions . The model should be sufficiently complex ( but not too complex ) to capture the phenomenon under study . Nikulin and Brismar ( 2006 ) [29] showed that two sinusoids at different frequencies with a specific phase shift and amplitude ratio capture accurately the prototypical shape of non-sinusoidal oscillations recorded in actual EEG experiments . This shape is reproducible across a majority of subjects ( n = 176 ) in that study and for central and occipito-parietal regions . Similar simple parametric models with superposition of trigonometric functions have been used to learn the morphology of the μ-rhythm for monitoring mental states in brain-computer interfaces [45 , 46] or investigating the effects of non-sinusoidal shape on phase-amplitude coupling [47] . A biophysical model could further improve the understanding of how exactly transmembrane currents and kinetic of ion channels lead to the generation of a given wave shape . Typically , neural mass models , where the output macroscale signal is produced through an interplay of excitatory and inhibitory populations , are able to produce signals of non-sinusoidal waveforms [48 , 49] . Specifically , for the generation of the sensorimotor μ-rhythm , a model on the level of a cortical column with spatially extended neuronal morphologies is able to generate non-sinusoidal source signals through integration of thalamic driving input via basal and apical dendrites of pyramidal neurons [50] . However , once we have an accurate description of the waveform , we can then proceed with the question of what would be the consequences for EEG/MEG/LFP signals when processes with such waveform are mixed with variable time-delays . Although these time delays are not accurately known in advance , we provide a wide range of simulations covering a relatively broad distribution of delays . Therefore , investigating waveform of oscillations can aid in constraining empirical mixing temporal delays . If we assume that the underlying source signals are non-sinusoidal , as evidence from LFP and invasive recordings suggests , the degree of non-sinusoidality present in macroscale EEG and MEG recordings relates to spatial synchronization with small time lags . Non-sinusoidality in EEG/MEG recordings should be present to a higher degree in signals which demonstrate less spatial mixing . This is the case for instance for many LFP recordings where spatial mixing is restricted to local neuronal populations located in the proximity to the recording electrode [13 , 14] . Therefore , non-sinusoidality of the oscillations can be used as a proxy for demixing of neuronal signals recorded with EEG/MEG . SSD is based on covariance matrices of narrow band processes . Utilizing a broader spectrum of information content is possible with other methods , for instance by learning a dictionary of canonical waveforms and associated spatial patterns [16] or using bicoherence for localizing non-sinusoidal waveform shape generators [51 , 52] . Note that the synchronization index can be 1 for sources having phase lag of 0 or phase lag of π 2 . In our study we are not measuring the synchronization strength per se , but rather state that the wide distribution of time delays between sources translates to the degree of non-sinusoidality in the measured neuronal signals . Improved methodology will aid in determining functional properties of oscillations with increased sensitivity ( not affected by narrow band-pass filtering ) when relating oscillatory component to behavioral and stimulation outputs . Investigating waveform in the temporal domain may aid in an improved determination of phase . A shortcoming of current methods for the computation of spatial filters which are based on linear decompositions ( SSD , CSP , ICA ) is that their solutions are invariant with respective to sign/polarity of the extracted signals . It has been shown that brain states associated with specific phases have differential functional consequences for cortical excitability and plasticity [24 , 53 , 54] . For instance , magnetic stimulation at the trough of the sensorimotor rhythm elicits a higher response compared to stimulation at the peak of the sensorimotor rhythm [53] . In this study , the rhythm of interest was extracted with a local spatial filter using a fixed electrode set , which is agnostic to physiological state . In subjects where the spatial filter would extract an inverted source signal , for instance due to EEG cap positioning , this functional relationship would be inverted , with higher response at peak states compared to trough states . Therefore , it is important to be able to uniquely define positive and negative peaks of an ongoing rhythm , which is possible when considering measures such as ΔCT . Additionally , the concept of a protophase [27] may aid in describing non-uniform phase velocity and the resulting relationships between cognitive functions and the evolution of oscillations . In fact , as indicated in previous studies [55] duty cycle in neuronal oscillations relates to windows of opportunity for spike transfer between distinct neuronal populations . While 50% duty cycle relates to the same duration of excitatory and inhibitory phases , a deviation from this number ( e . g . 30% ) can introduce significantly shorter duration of excitatory phase thus providing more precise tuning for the neuronal communication , effectively blocking effects of spikes arriving at the considerably longer inhibitory phase . Spatial mixing in EEG/MEG , leading to more sinusoidal signals , might create an illusion of oscillations with 50% duty cycles while at the source level the duty cycle can be considerably different . When using band-pass filtering non-sinusoidality is removed since only one Fourier component is preserved effectively representing only one frequency and its immediate neighborhood . Behavioral and stimulation effects of such band-pass filtered signal will still be present yet neurophysiological interpretation can be different . It has been debated whether α/β-rhythms have a common or separate origin [29 , 38 , 56] . One of the arguments in favor of both rhythms originating from the same source is that if α- and β-oscillations are generated by the same neuronal source , producing rhythmic but non-sinusoidal waveform , then one should observe a strong positive amplitude correlation between the two oscillations [38] . This argumentation is based on the linearity of the Fourier transform , as briefly illustrated in the following: As shown above , our non-sinusoidal signal can be represented as S = α + b ⋅ β , with the corresponding Fourier transform of S being F ( S ) . When the amplitude of S is changing in different time segments ( multiplied by Ai ) , the corresponding Fourier transform at segment i , can be written as: F ( Ai ⋅ ( α + b ⋅ β ) ) = Ai ( F ( α ) ) + Ai ⋅ b ( F ( β ) ) , which in turns shows that the amplitude of α- and β-oscillations should covary linearly when the amplitude of S changes by Ai . The amplitude of oscillations in different frequency bands can covary for different neuronal sources , but the presence of strong correlations between oscillations at different frequencies with similar spatial topographies is consistent with the idea of them originating from the same neuronal source . Yet , our simulations show that even when a comodulation between α- and β-oscillations is certainly known to originate from the non-sinusoidal waveform of oscillations , due to the peculiarities of the spatial mixing , it is possible not to observe such positive comodulation . Moreover , surprisingly it is even possible to detect anticorrelation between the amplitudes of α- and β-oscillations . However , this is entirely due to the effects of spatial mixing of individual signals each of which by itself has only positive correlations between α- and β-oscillations . Yet , a spatial summation may lead to the occurrence of negative correlations at the sensor level . Importantly , even when using sophisticated spatial filtering techniques such as ICA , SSD , etc . it is unlikely to disentangle such spatial mixing effects originating from the local cortical patches since the resolution of EEG/MEG and even LFP recordings is not sufficient . This also applies to the argument supporting a separate origin of oscillatory components requiring independence of the corresponding temporal dynamics . We have shown that seemingly separate amplitude time courses may not be an indication for the independence of the rhythms , but can also occur when the coupling between different sources changes in the span of only a few hundreds of milliseconds . Whether β-events can arise through decoupling of oscillators as in the presented simulations , is a topic for further research . This can reveal insights about mesoscopic brain organization and the interplay of different local rhythms , as extracted by EEG/MEG . Regarding cross-frequency interactions , our study shows that the amplitude-to-amplitude cross-frequency coupling can also be affected by the non-sinusoidal waveform of the oscillations . For all three types of cross-frequency interactions ( phase-to-phase , phase-to-amplitude , amplitude-to-amplitude ) , spatial synchronization can lead to either very strong or weak indices characterizing cross-frequency interactions , corresponding respectively to a small or rather large jitter in the time delays between neuronal sources ( see Fig 10 ) . This again requires careful interpretation of the obtained data and discussion about the possible effects of spatial synchronization among neuronal populations generating EEG/MEG/LFP signals . | The electrical activity in the human brain demonstrates oscillations of intricate complexity . Interestingly , such complex waveforms are primarily visible in invasive recordings but not so much when neuronal activity is recorded with non-invasive methods such as electroencephalography . Yet a specific waveform is informative about the postsynaptic processes which are at the core of our understanding of cortical excitability and information transfer in neuronal networks . In our study , we show with simulations and real EEG data , how temporal delays between different cortical sources can contribute to a more sinusoidal or non-sinusoidal shape of neuronal oscillations . We illustrate how , depending on the temporal delays , low- and high-frequency components of oscillations can be enhanced or attenuated to a different degree thus affecting the shape of oscillations and corresponding spectra which are often associated with specific functional consequences . We further show how this phenomenon can challenge our understanding of the link between neuronal oscillations and motor function , cognition and perception . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"fourier",
"analysis",
"medicine",
"and",
"health",
"sciences",
"engineering",
"and",
"technology",
"signal",
"processing",
"brain",
"electrophysiology",
"electrophysiology",
"neuroscience",
"signal",
"filtering",
"butterworth",
"filters",
"clinical",
"medicine",
"brain",
... | 2019 | Spatial neuronal synchronization and the waveform of oscillations: Implications for EEG and MEG |
Campylobacteriosis is a leading foodborne zoonosis worldwide , and is frequently associated with handling and consumption of poultry meat . Various studies indicate that Campylobacter causes a substantial human disease burden in low to middle-income countries , but data regarding the organism’s epidemiology in countries like Kenya are scarce . In sub-Saharan Africa , 3 . 8 million deaths of children under-5 years of age are reported annually . Of those , 25% are caused by diarrheal diseases , and Campylobacter is one of the most frequently isolated bacteria from diarrheic children . With the growth of urban conglomerates , such as Kenya’s capital , Nairobi , changes in diets , food production systems , and retailing dynamics , it is likely that exposure and susceptibility to this pathogen will change . Therefore , the importance of Campylobacter disease burden in Kenya may increase further . The objectives of this study were: 1 ) to determine the prevalence of Campylobacter spp . in Nairobi’s small-scale chicken farms and meat retailers , and 2 ) to identify potential risk factors associated with its presence in those sites . The prevalence data provides the first detailed baseline for this pathogen in the urban Kenyan context . The risk factors provide context-specific insights for disease managers . A cross-sectional study of broiler , indigenous chicken farms , and chicken meat retailers , was conducted in a peri-urban , low to middle-income area ( Dagoretti ) , and a very-low income informal settlement ( Kibera ) of Nairobi . Chicken faeces were collected using one pair of boot socks per farm , and 3 raw chicken meat samples were purchased per retailer . Samples were cultured for viable Campylobacter spp . using mCCDA , followed by blood agar plates in aerobic/microaerobic conditions for prevalence calculations . A questionnaire-based survey on sanitary , sourcing and selling practices was conducted at each site for risk factor identification using logistic regression analyses . A total of 171 farm premises and 53 retailers were sampled and interviewed . The prevalence results for Campylobacter spp . were between 33 to 44% for broiler and indigenous chicken farms , 60% and 64% for retailers , in Dagoretti and Kibera , respectively . Univariable logistic regression showed an association between Campylobacter spp . presence and the easiness of cleaning the display material used by the retailer . Restricting access to the flock was also associated with the pathogen’s presence . Multivariable logistic regression identified the selling of defrosted meat as a retailer risk factor ( OR: 4 . 69; 95% CI: 1 . 31–19 . 97 ) , calling for more investigation of the reported repetitive freezing-thawing processes and cold chain improvement options . At the farm-level , having a pen floor of material not easy to clean was found to increase the risk ( OR: 2 . 31; 95%CI: 1 . 06–5 . 37 ) . The relatively high prevalence of Campylobacter spp . across different areas and value chain nodes indicates a clear human exposure risk . The open nature of both small-scale broiler and indigenous chicken production practices with low biosecurity , hygiene and informal transactions , likely plays a role in this . While gradual improvement of farm biosecurity is recommended , risk factors identified suggest that consumer education and enforcement of basic food safety principles at the retailer end of the food continuum represent key targets for risk reduction in informal settings .
Campylobacteriosis is one of the leading bacterial foodborne zoonosis globally [1] , with handling and consumption of chicken meat identified as a major risk factor in high-income countries [2] . The estimated public health impact of Campylobacter-induced enteric disease is around 0 . 35 million disability-adjusted life years per year for EU-27 , with annual costs estimated at about 2 . 4 billion euros [3] . Despite intensive research on the pathogen and testing of a range of control measures , campylobacteriosis has been the most frequently reported gastronintestinal disease in Europe since 2005 [4 , 5] . An overall rate of 59 . 8 cases of campylobacteriosis per 100 , 000 population was reported in 2014 for the European Union and two European Economic Area countries , ranging from 1 . 3 to 197 . 4 by country [4] . While differences in reporting , climatic conditions , and chicken production systems may explain differences in incidence , the epidemiology of the bacteria remains poorly understood and other factors may be involved . In low and middle-income countries ( LMIC ) , surveillance for Campylobacter seldom exists in people and chickens , and data regarding the organism’s presence , risk factors and impacts are scarce . Yet , the disease burden of Campylobacter in the global South should not be underestimated . In Sub-Saharan Africa alone , 3 . 8 million deaths in children under 5 years are reported annually , 25% of which are caused by diarrheal diseases [6] . This bacterium is among the most common pathogens found in diarrheic children in LMIC [7] . In a multisite birth cohort study from 2009 to 2012 ( MAL-ED study ) in Asian , Latin American and African countries , Campylobacter spp . were the most frequently detected pathogens , occurring in 84 . 9% of 1892 children , and contributed the highest burden of diarrhoea in the first year of life . Campylobacter infection in children was associated with growth deficits across sites [8 , 9] . A Campylobacter isolation rate of 8% was reported in all-age diarrheic patients in Ethiopia [10] , compared to 12% ( higher than for Salmonella and Shigella ) in Kenya [11] . A Campylobacter prevalence of 19% was reported in children under 5 from Morogoro , Tanzania [12] , whereas a study in Western Kenya health centers isolated Campylobacter spp . from 42% of diarrheic children under 5 [13] . Hence , a better understanding of the sources of Campylobacter is needed to reduce diarrhoea-related child mortality . With the aim to address these data gaps , this study focused on Nairobi , Kenya , to investigate the epidemiology of Campylobacter spp . in a likely source , namely the chicken meat production system , in this setting . Kenya’s capital , Nairobi , illustrates the global trend of fast urbanisation in LMIC countries . The human population has grown from 350 , 000 in 1962 to 3 , 375 , 000 in 2009 , whilst the spread of informal settlements has led to over 60% of city’s population residing in conditions of significant poverty [14] . In parallel , the middle-class has been growing rapidly , with increasing demands in terms of food quality , and a surge in supermarkets and fast food outlets [15] . To meet the increasing demand in poultry meat , poultry production systems have been intensifying in Kenya [16] . An increase in commercial chicken farming , generally using imported fast growing broiler breeds such as the Cobb 500 , is observed in and around Kenyan urban centres such as Nairobi , Mombasa , Nakuru , Kisumu and Nyeri , where the demand for poultry meat and market access for chicken producers are greater in comparison to rural areas . Outside of urban areas , indigenous chickens ( i . e . local breeds which grow slowly and are used for egg and meat production ) are the main chicken species kept [17] . These changes in retailing dynamics , diets , and poultry production systems , are altering the epidemiological setting for campylobacteriosis . While indications of protective immunity against the bacteria in adults [18] may have led to the disease been seen as low priority in LMIC , the evolving Nairobi setting may lead to significant changes in exposure and susceptibility to the disease in the population , and calls for a better understanding of the pathogen’s epidemiology . While poultry is recognised as a major source of Campylobacter spp . [2] , western studies have identified consumption of poultry meat , undercooked red meat , raw milk , untreated water , contaminated raw foods like salads , contact with pets and farm animals , and international travels as risk factors for disease in humans [2 , 16 , 17] . Studies investigating the disease in the global South are sparse . In the MAL-ED study covering 8 low-resource sites in Asia , Latin America and Africa , factors associated with a reduced risk of Campylobacter detection in children regular surveillance stools included treatment of drinking water , exclusive breastfeeding , access to an improved latrine , and recent macrolide antibiotic use [8] . C . jejuni and C . coli have been isolated from chickens , goats and sheep in Nigeria and similarities between strains isolated in chickens and humans suggest that poultry is an important source of human campylobacteriosis [7] . Risk factors identified for Campylobacter infection in people include home slaughtering and eating undercooked meat in Cambodia [19] , the presence of animals or uncovered garbage in the cooking area , and lack of piped water in Egypt [7] , contact with animals and HIV infection in Burkina Faso [20] , young age , consumption of chicken meat and prepared salad in Tanzania [21] , and poor hygienic conditions in LMIC in general [7] . The most important source of Campylobacter infection in chickens is thought to be the external environment [2] . Risk factors for Campylobacter presence reported for intensive commercial production systems in high-income countries include the use of contaminated water [22]; flock thinning ( partial depopulation ) , carry-over from a previous flock following inadequate cleaning and disinfection , increasing bird age at slaughter and number of birds reared per year on farm [23 , 19]; organic rearing [24] , broiler houses older than 15 years old , and long downtime between flocks [25] . A 2004 study in Senegalese broiler chickens found a 63% Campylobacter prevalence . On-farm presence of laying hens , cattle and sheep , lack of exclusive clothing for poultry workers , and use of chick transport cartons as feeders were found to increase the risk of infection in chickens , whereas thorough cleaning and disinfection of the poultry house were protective [26] . Two studies in South African broiler flocks found Campylobacter prevalence in chickens to be higher in rural areas ( 68% ) , compared to commercial indoor broiler flocks ( 47% ) or layer flocks and ( 94% ) [27] . To the authors’ knowledge , only one risk factor study has been published so far for chicken meat production systems in Nairobi , Kenya , which identified cleaning of the poultry house before restocking as a risk factor [28] . To mitigate carcass contamination by the intestinal tract of positive birds during the slaughter process [29] , industrial slaughter and processing facilities in high-income countries use a variety of strategies such as chemical treatment , irradiation or freezing of carcasses to reduce the bacterial count [2 , 25] . At retailer level , general hygiene measures to prevent cross-contamination between the meat , retailer’s hands , contact surfaces and utensils , are recommended to minimise Campylobacter spread [30] . National prevalence in chickens and chicken meat have been found to vary greatly worldwide , from 4 . 9 to 100% in EU broiler carcasses [31] , with a mean prevalence in broiler meat across Europe in 2015 of 46 . 7% [5] , and from 8 to 100% in poultry meat at retail level across 32 different countries globally [32] . The mean prevalence reported for poultry meat in Senegal and South Africa in the latter study was 73 . 1% . In the sub- Saharan Africa context , Campylobacter prevalence in chicken meat was found to be 81 . 9% in poultry processing plants [33] and 100% in retail outlets in Nigeria [34] , and varied between 11 . 1% and 100% in South African supermarkets [35] . One study found a prevalence for thermophilic Campylobacters of 77% ( C . jejuni 59% , C . coli 39% and C . laridis 2% ) in raw chicken sourced from butcheries , markets and supermarkets in Nairobi , Kenya [11] , while studies in Ghana and Ethiopia found a prevalence close to 22% [36 , 37] . Risk factors identified for industrial commercial chicken production in high-income countries are highly context-specific and cannot be applied directly to informal meat production systems , such as small-scale Nairobi chicken farms , where biosecurity is limited , even in commercial broiler operations . Except for a few large integrated broiler companies and high-end supermarkets chains , informal production and retailing still dominate [31 , 14] . The lack of literature on Campylobacter risk factors in food animals and food animal products in LMIC , where rearing systems and level of hygiene may differ greatly from Western settings , represents a major gap [38] . Considering the public health importance of Campylobacter , especially for vulnerable groups in LMIC , poultry’s predominant role in the global North as a risk factor , and the scarcity of epidemiological data in the context of rapid African urbanisation , the objectives of this study were: 1 ) to determine the prevalence of Campylobacter spp . in Nairobi’s small-scale chicken farms and chicken meat retailers , and 2 ) to identify potential risk factors associated with the presence of Campylobacter spp . in those same sites . The data provide a system-wide picture of the risks of exposure to Campylobacter at farm and retailer levels , and the first detailed baseline for this pathogen in the urban Kenyan context , whilst the identified risk factors help understand its epidemiology and provide insights for Kenyan disease managers .
The selection of Nairobi was based on the following criteria: representativeness of growing urban centers in East Africa , transitioning urban landscape and evolving chicken production systems . Nairobi , one of the major fast-growing urban centers in East Africa , with both a growing middle class and expanding informal settlements , is a prime candidate to investigate the epidemiology of Campylobacter in the context of transitioning urban landscape and chicken production systems . The study design was informed by previous work on the chicken meat value chains in Nairobi [39] . In the latter study , small-scale broiler and indigenous chicken farms , and small-scale broiler meat retailers were identified as key nodes , and were therefore targeted for the understanding of the risk of exposure to Campylobacter . Small-scale chicken farms were defined as a flock of 2 to 100 birds for indigenous chicken flocks , and 2 to 800 birds for broiler flocks , and small-scale broiler meat retailers were defined as any premise selling raw broiler meat ( butchery ) or a mix of raw and cooked meat ( combination of butchery and small restaurant ) , not belonging to a franchise; they were found to be the most numerous chicken meat value chain actors in Nairobi . Poultry abattoirs and indigenous chicken meat retailers were found to be rare in Nairobi ( indigenous chickens are commonly sold on-farm directly to consumers ) , and were therefore excluded . Large integrated broiler companies could not be sampled due to the sensitivity of the business information . A cross-sectional survey of small-scale broiler and indigenous chicken farms as well as broiler meat retailers was conducted ( layer chickens were excluded ) . In order to provide a representative picture of Nairobi’s food system and major types of urban landscape found in the city , two areas of different wealth levels and production systems were purposely selected . Dagoretti , a low to middle-income , peri-urban area , characterised by a rural-like landscape with pockets of residential areas and moderate population density , was selected as a major livestock raising area within Nairobi , and due to its easy accessibility for the research team . Kibera , characterised by high population density , fully urban landscape and lower livestock activity , was selected as it represented the largest very low-income informal settlement ( slum ) in Nairobi . As major differences in the value chain structure and risky practices had been identified by Carron et al . , 2017 [39] , these two areas were targeted to test the hypothesis whether socio-economic status could affect the presence and survival of Campylobacter at farm and retailer level . Sample sizes were calculated for independent populations ( see S1 Appendix for more information on sample size calculation ) , namely two types of chicken production systems per area , and one retailer group per area , using an expected Campylobacter spp . prevalence of 50% , a 10% confidence limit and 90% confidence interval . No regular records of farms were available to guide the selection of farms to be sampled . The team worked with community elders that had been recruited to participate in the project to create a census of all broiler farms in each area . Since the number of broiler farms was limited ( close to or below the calculated sample size ) , all were targeted for sampling . Because it is a common practice in the study sites to own indigenous chickens , it was not realistic to undertake a census of indigenous chicken farms . This resulted in an overall sample size for Dagoretti and Kibera , respectively , of 42 and 8 small-scale broiler farms , 67 and 63 small-scale indigenous chicken farms , and 21 to 40 small-scale broiler meat retailers per area . Using the target sample size for indigenous farms , a corresponding number of random GPS coordinates within each area was computer-generated using ArcGIS . The first farm found North of each GPS point by the sampling team was targeted for sampling . A census approach was used for broiler meat retailers , as these were reported by the elders to be few , and located along a few main streets in each area . In Dagoretti , the census of all butcheries selling chicken meat was performed by walking or driving along the main streets and asking employees whether they sold chicken . GPS coordinates for each retailer selling chicken were recorded . Due to the small number of retailers ( close to or below the calculated sample size ) , it was decided to sample all retailers willing to participate . In Kibera , due to security issues , a key informant was asked to perform a similar retailer census with support from the local elders . On chicken farms , one sample each of chicken faeces and/or housing litter was collected using boot socks dampened with sterile saline [40] . Three meat samples ( 100g or more ) from different chicken carcasses were bought from each retailer . Each boot sock pair or meat sample was put in a sterile ziplock bag and stored in a cool box with ice packs , until testing at the laboratory within 5 hours of collection . All samples were cultured for viable Campylobacter spp , using a protocol based on the ENIGMA consortium 2017 study [40] . Boot sock samples were enriched using 50 ml of Exeter broth and incubated at 42°C under aerobic conditions with a minimal air space for 24 hours before sub-culturing . A 50g piece of each meat sample was cut aseptically , added to 200ml of saline and subject to stomaching for 1 minute; 5ml of the stomacher content was added to 5ml of double strength Exeter broth , and 10ml of the enriched sample incubated similarly to boot sock samples . All samples were then cultured for viable Campylobacter spp . at the Kenya Medical Research Institute ( KEMRI ) of Nairobi . Samples were first plated onto mCCDA and incubated for 48 hours at 42°C under microaerobic conditions ( using CampyGen microaerobic gas pack in jars ) . Plates were visually examined for suspect Campylobacter colonies using colony size , shape and surface colour , and other key characteristic: C . jejuni 2 . 0–3 . 0 mm , flat/entire/glossy , grey/white , can be efflorescent ( spreading moist ) , and C . coli 1 . 0–2 . 5 mm , convex/entire/glossy , creamy grey moist . For each mCCDA plate that showed growth for Campylobacter spp . , four suspect colonies were subcultured on two different Columbia blood agar plates . One plate was incubated under microaerobic conditions for 48 hours at 42°C , and the other under aerobic conditions at 37°C for 48 hours . Growth in microaerobic conditions only was considered as positive for Campylobacter spp . A subset of isolates ( 428/560 ) was confirmed by LPX-PCR [41] . In order to evaluate risk factors for Campylobacter spp . exposure , a questionnaire was used to collect data from each site visited . The farmers’ and retailers’ questionnaire ( S2 and S3 Appendices ) covered the following categories of variables/themes ( Table 1 ) : 1 ) Farm or retailer’s environment and characteristics , 2 ) Management practices , 3 ) Biosecurity , health or sanitary practices , and 4 ) Sourcing and selling of chickens/chicken products . Questionnaires were written in English and conducted using Open Data Kit ( ODK , https://opendatakit . org/about/tools/ ) software on electronic tablets . Sites and samples were identified by scanning unique barcodes . Enumerators were Kenyan citizens familiar with the city , bilingual in English and Kiswahili . Pre-sampling training of the enumerators on the questionnaires took place . Prior to data collection , ethical approvals were sought from the ILRI-IREC ( International Livestock Research Institute—Institutional Ethical Research Committee , project reference ILRI-IREC2016-01 ) . ILRI-IREC is accredited by the National Commission for Science , Technology and Innovation ( NACOSTI ) in Kenya . Approval from the Royal Veterinary College ( RVC ) ethical committee was also received ( project reference: URN 2015 1453 ) . Permission to interview people was obtained from the Ministry of Agriculture and the local Veterinary Authorities . The study’s objectives and participants’ rights were explained in Kiswahili to farmers and retailers upon arrival at the site . Verbal and written consent to participate in the study were obtained before initiating data collection . Variables in the survey data with too many missing observations ( >25% ) and variables with no substantial variability ( >95% responses identical ) were not kept for analysis . This first variable screening lead to a total of 45 farm exposure variables and 43 retailer variables for inclusion in the risk factor analysis . Using Excel and R version 3 . 3 . 2 ( 2016-10-31 ) , each site ( farm or retailer ) barcode and meta data was linked to the corresponding sample barcodes and laboratory results . Inconsistencies and data gaps were reviewed with the field coordinator and discussed with laboratory technician to clean the database . Using a chicken farm or retailer as a sampling unit , a site with one or more positive samples on culture classified as positive for Campylobacter spp . A sample was considered positive if at least one isolate was obtained and comfirmed by culture . Culture prevalences were calculated using QuickCalcs ( GraphPad , http://www . graphpad . com ) . A Fisher’s exact test was used to compare prevalence between groups . The LPX-PCR results obtained for a subset of the samples were used to confirm the culture prevalences . A two-step statistical analysis using univariable logistic regression followed by multivariable logistic regression , was performed to identify risk factors for the presence of Campylobacter spp . at farm , or retailer level , respectively . No derivative analysis was made for farms or retailers from a specific area ( Dagoretti or Kibera ) , or for a specific type of farm ( broiler vs . indigenous chicken ) due to the limited size of each subgroup . Rather , area and type of farms were included as confounders in the models . A univariable analysis was performed in order to identify possible associations between the 88 selected exposure variables and the the presence of Campylobacter spp . using univariable logistic regression for each of the predictors . Odds ratio ( OR ) and 95% confidence intervals ( CI ) were calculated . All variables with a p-value ( calculated with a likelihood ratio test ) lower than 0 . 2 were retained for assessment in the multivariable analyses , except if the variable belonged to a nested question not applicable to the whole “farm” or “retailer” population . Two multivariable logistic regression analyses were conducted independently , one for retailers , one for farmers . In each analysis , variables selected during the respective univariable analysis were included in the initial model . A stepwise backward selection procedure was used to refine models until all variables remaining in each model met the criterion of a p-value ≤0 . 05 . Two-way interactions between predictors were assessed using a likelihood ratio test and considered significant if p ≤0 . 05 . In order to evaluate potential collinearity effect between predictors , the levels of association between risk factors identified during the univariable analysis were assessed using a Fisher test; risk factors with more than two-fold changes in the logistic regression coefficients were also checked during the selection process . As data collection took place following a sampling frame designed for investigating Campylobacter spp . prevalence in two Nairobi areas and two types of chicken farms , multiple logistic regression models were built to account for the potential confounding effect of the study design . One farm model included “farm area” and “farm type” variables , despite their non-significance in the univariable analysis , whilst a second model did not include them . Similarly , one retailer model included the “retailer area” , and another did not include this variable . The predicted probability was calculated for each observation based on the final model and the fit was assessed using the distribution of the model’s residuals , residuals close to zero suggesting a good fit [42] . Finally , an R-squared value was calculated [43] . All statistical analyses were performed using the statistical software R version 3 . 3 . 2 ( 2016-10-31 ) .
The culture prevalence of Campylobacter spp . in small-scale farms varied between 33 and 44% across types of production systems and areas , whereas the prevalence in retailers was 60% in Dagoretti and 64% in Kibera ( Table 2 ) . While Campylobacter spp . prevalence at retailer level was higher than at farm level , no statistically significant difference was found between the types of site . Out of the 429 isolates tested by LPX-PCR , only 1 was not confirmed as Campylobacter , suggesting reliable culture resuts . A total of 63% of the 428 Campylobacter isolates were C . Jejuni .
This study is the first to document Campylobacter spp . prevalence in both small-scale chicken farm and chicken meat retailer levels in Nairobi and to investigate factors determining the heterogeneity of Campylobacter presence in these settings . The results provide valuable insights into the potential risks of human exposure in an otherwise undocumented context . The great variability found in Campylobacter prevalence across broiler batches or carcasses in the EU , the limited number of similar studies in the East African context , and the differences in epidemiological units used in the literature ( e . g . retailer versus carcass-level prevalence ) , make it difficult to compare these results directly with other studies . However , the relatively high Campylobacter prevalence results found in Nairobi retailers echoes some of the prevalence reported ( 73 . 1% or higher ) in retail poultry meat in sub-Saharan Africa [28 , 36 , 37] . In Nairobi , Kenya , isolation rates of 59% for C . jejuni , 39% for C . coli , and 2% for C . laridis were found in raw chicken sourced from butcheries , markets and supermarkets [11] , with chicken meat tested less than 24 hours after slaughter showing a higher prevalence ( 85 . 3% ) . Time since slaughter might aso have influenced results in our study , since meat samples were not collected at the slaughter plant . Few studies identified much lower Campylobacter prevalence values , such as 21 . 7% in retail raw chicken meat tested in Ethiopia [36] , and 21 . 9% of commercial chicken carcasses swabbed in Ghana [37] . Broiler flock prevalence in our study are moderately lower than in other sub-Saharan African studies ( 47% to 68% Campylobacter prevalence overall ) [10 , 44] , which might be due to the small number of broiler farms sampled , to a difference in size of commercial flocks , or a difference in sampling unit and testing methods . Few studies found a prevalence lower than 30% . A Ghanaian study found Campylobacter in 22 . 5% of ceacal samples [37] , a Tanzanian study in 42 . 5% of chickens ( various breeds ) using cloalcal swabs [12] and an Ethiopian study , in 28 . 9% of chickens ( various breeds ) [45] . A study from 1988 found a prevalence of 51 . 5% in Kenyan broilers [46] , whereas a 2018 study found an overall prevalence of 69 . 5% in Nairobi chickens [28] . The prevalence results in our study are indicative of a relatively uniform distribution of the pathogen across the chicken meat system studied . This can most likely be explained by the informal nature and overall lack of biosecurity in these systems , which is unlikely to limit the introduction of Campylobacter into either indigenous or broiler flocks . Unlike in Europe and North America , practices used in broiler versus “backyard” indigenous chicken farms in Nairobi share more similarity . Due to a lack of resources , small-scale Nairobi broiler rearing infrastructure is heterogeneous , using suboptimal materials , and often in proximity to other livestock . Flock management is often lead by irregular market access , with limited sanitary considerations . This is exacerbated in informal settlements where space is lacking , and resources are further limited . In such areas , a broiler flock can be found under a vegetable shop stall or staircase . The limited number of broiler farms observations , as fewer broiler farms than expected were identified , also limits the power of this study to identify differences between management systems . Indeed , studies in Ethiopia and Tanzania have identified marked differences in prevalence between broiler and indigenous chicken flocks , with conflicting results . Two studies in Tanzania found a higher Campylobacter prevalence in indigenous chickens ( 76 . 49% and 75% ) compared to broilers ( 26 . 4% and 50% ) [47 , 12] . Another Tanzanian study found no significant difference between broilers and indigenous chickens , but rather a higher prevalence in local chickens from rural areas compared to those in urban areas [48] , while an Ethiopian study found significantly higher Campylobacter isolation rates in animals ( chicken , sheep , cattle and pigs ) in urban areas ( 56 . 7% ) compared to rural areas ( 26 . 7% ) [49] . Finally , a 2018 study found a prevalence of Campylobacter of 91 . 07% in broilers , 70 . 96% in layers , and 61 . 04% in indigenous chickens in peri-urban areas of Nairobi , Kenya [28] . The higher prevalence found in meat sellers compared to farms in our study may be explained by the risk of cross-contamination between chicken meat products of mixed sources during meat handling , cutting , storage and display . A Ugandan study found Campylobacter survived much better on wooden cutting boards than plastic or metal ones [50] , wooden boards being widely used in Nairobi retailers . Nairobi-specific factors that may affect Campylobacter’s survival include the average temperature , which is constantly above 16°C , or the precipitation which is high ( 80 to 191 mm ) during the two rainy seasons . Indeed , unlike the reported summer and autumn peaks of campylobacteriosis in Europe and North America , seasonality of Campylobacter has not been reported in LMIC , potentially due to a lack of study in tis setting [51] . The common practice of freezing and defrosting chicken meat in Nairobi , further discussed below , could also influence the bacteria’s presence . In addition to investigating the prevalence of Campylobacter in the meat system , determining the level of contamination of the chicken meat sampled in Nairobi retailers would have brought an additional key indication of the risk of human exposure , but was not feasible due to resource limitations . A higher load of Campylobacter on meat increases the risk of contamination of the direct meat environment and spread within a household , or site . The European Food Safety Authority ( EFSA ) has estimated a public health risk reduction of 50%–90% could be achieved , if all broiler batches complied with the critical limit of <1000 and <500 CFU/g of neck and breast skin , respectively [52] . However , the infectious dose for Campylobacter being low at a few hundred cells ( 500 or less ) [53] , prevalence of the bacteria at retailer-level was considered an appropriate indicator of the risk of exposure in this study . Few explanatory variables were found to have a significant association with the presence of Campylobacter spp . in the univariable analysis or were identified as risk factors in the multivariable analysis . Retailers using a display material “not easy to clean” ( e . g . made of wood or porous material ) were shown to have higher odds of Campylobacter spp . presence , compared to those using a display material easy to clean . This is in line with literature describing lower levels of hygiene at retail-level as a risk factor [53] . A risk assessment of Campylobacteriosis linked to chicken meals prepared by households in Dakar , Senegal , determined that washing of cooking utensils during food preparation was not sufficient to significantly reduce the risk of Campylobacteriosis , whereas changing knife , board and dishes between pre and post-cooking was [54] . “Selling defrosted meat” increased the odds of Campylobacter spp . presence in both steps of the analysis . This finding is surprising given that freezing can be used as a strategy to reduce numbers of Campylobacters present on the meat [55 , 56 , 57 , 58] . However , freezing-thawing of chicken meat was found to be a common retailer practice in Nairobi and could favour re-contamination of the meat . Multiple retailers interviewed described how they turned off their freezer during the day to soften the meat for cutting , and turned it back on at night to preserve unsold meat until the next day . Freezing fresh chicken meat for 24 hours has been shown to reduce the log number of viable Campylobacters by up to 2 . 5 [55 , 56] , and a 2–3 day freezing period to diminish the risk by 50–90% [52] . However , freezing temperatures in Nairobi are not verified , and incomplete freezing may be common . The repetitive freezing-thawing-refreezing practice observed in chicken retailers is driven by resource scarcity , and the demand from consumers for small quantities of chicken meat . The latter has led to a selling culture of cutting small pieces of meat from a whole carcass in the presence of the customer . Hardly any retailers were found to freeze small meat pieces in individual packaging . This may be due to customers wanting to see the carcass of origin . Where cold chain infrastructure is more affordable and food hygiene is strictly regulated and enforced , multiple freezing-thawing cycles are not allowed . Studies have found that the refrigeration prior to freezing , as well as the type of meat surface ( e . g . skin versus meat muscle , or ground chicken ) will affect the number of Campylobacter cells surviving freezing [32 , 36] . A 2013 study by the UK Food Standards Agency determined that the freezing temperature and length of time taken to freeze chicken livers influenced the bacteria’s survival [55] . Another study found lower Campylobacter prevalence in chicken meat from Malaysia wet markets compared to supermarkets , hypothesising that the chilling infrastructure in supermarkets favours survival of the bacteria whereas the ambient temperature of 29 . 6°C in wet markets is not favourable for growth [59] . On the other hand , a Kenyan study found lower levels of E . coli contamination in raw chicken meat sold in supermarkets compared to smaller-scale retailers [60] . This illustrates how specific freezing processes can influence risk reduction , and have a different effect depending on the pathogen . It highlights how significant the challenges linked to the cold chain can be in a context of limited resources , especially for a highly perishable product like chicken meat . Further research will need to investigate if the Campylobacter presence related to defrosted meat identified in this study originates from the freezing-thawing process with sub-optimal cold chain conditions , or from cross-contamination post-freezing . Since chicken meat freezing in Nairobi was well accepted by consumers , food safety interventions could capitalise on this practice and its potential for Campylobacter risk reduction . Awareness trainings regarding sanitary practices to avoid cross-contamination and promoting the freezing of small chicken pieces wrapped individually to minimise the handling and repetitive thawing-freezing could be considered . In the farm univariable analysis , three variables were identified as having a significant association with the outcome of interest . The predictor “restricting access to the flock” was found protective . Arsenault et al . [61] specifically assessed the permanent locking of the broiler house , which was associated with a reduced risk of Campylobacter colonization in chickens . This practice is not readily applicable for indigenous free-ranging chickens . Even in the case of broilers , while greater access restriction could be encouraged by providing training to farmers , it is unlikely to result in any significant risk reduction without the general on-farm biosecurity being improved . This would require substantial investment , which in turn would demand external support or simultaneous improvement of small-producers’ market access and business profitability . Using a pen material “not easy to clean” and cleaning the pen without disinfectant were also found to increase the odds of Campylobacter presence , in line with literature citing inadequate cleaning and disinfection between flocks as risk factors [2] . Of interest , is a similar result from a 2018 Nairobi study , which only identified cleaning and disinfection of the chicken house before restocking as a risk factor ( p<0 . 05 ) in the multivariable analysis [28] . Many of the risk factors identified in the literature for Campylobacter at farm-level ( e . g . water source , thinning , biosecurity measures ) [2 , 23 , 61 , 62] , and retailer ( or carcass ) level ( e . g . contact between different carcass parts ( e . g . liver and meat ) , cross contamination via handling practices ) [63] , were tested in the univariable analysis , yet did not show any significant association with the presence of Campylobacter . The identification of few risk factors may be linked to the cross-sectional sampling design , less suited for risk factor analyses compared to case-control or cohort studies , and selected for the Campylobacter prevalence estimation objective . The limited number of observations , as well as the high number of potential risk factors , may have also limited the power of the study . In addition , the specificity of the Nairobi context and scarcity of similar studies in informal settings , are likely to explain some of the discrepancies with Western studies . The extreme variability in production practices and in the level of implementation of sanitary practices in this context is difficult to analyse accurately and illustrates the challenges of capturing risk factor data in messy settings . Overall , we can hypothesize that the minimal biosecurity and sanitary measures observed in small-scale Nairobi farms and retailers create an open system , with numerous sources of contamination , making individual risk factors hard to identify and isolate from the general environment . Still , the study , despite not following a risk assessment structure , provides useful risk indicators to be further investigated . While Roesel and Grace [15] have found that formal retailing settings in sub-Saharan Africa do not necessarily translate into a lower risk for consumers , repeating the analysis made for small-scale retailers in Nairobi’s high-end supermarkets , where stricter sanitary standards are applied , would enhance our understanding of the broad risk context . While the prevalence and risk factor analyses were designed to provide a system-wide picture of the risks of exposure to Campylobacter at farm and retailer levels , it should be noted that the lack of information regarding the origin of the carcasses at retailer-level limits our understanding of transmission dynamics in the chains . Indeed , retailers in Dagoretti have been found to source their carcasses locally , whereas Kibera retailers have reported selling low-value cuts from from major integrated broiler companies outside the informal settlement ( see S4 Appendix for more information on the chicken meat supply ) . In terms of recommendations arising from this study , the risk factors identified support training initiatives on biosecurity and food safety practices . Group feedback sessions are planned for farmers and retailer having participated in the study , including education on basic biosecurity principles , sanitation measures and safe handling of chicken meat . While gradual improvement of biosecurity measures ( via appropriate cleaning and disinfection , better farming infrastructure and flock management ) targeted at commercial farms should be supported , initiatives focusing on consumer education and enforcement of basic food safety principles seem more easily manageable , and with potentially greater impact as a first step , in informal settings . By using a risk-based sampling approach based on a value chain analysis to design the prevalence and risk factor analyses , this study presents methodological novelty . Substantial economic value chain studies in Africa can be found , but the combination of value chain analysis and risk identification , or disease investigation , remains limited . Finally , this study is the first to describe Campylobacter prevalence and risk factors both in chicken farms and chicken meat retailers at this level of detail in a peri-urban and informal settlement Kenyan setting , providing key insights into the specificities of Campylobacter epidemiology in quickly urbanising areas of East Africa . | Gastrointestinal disease following food-poisoning can cause severe clinical signs in humans and represent high costs for society . Examples of bacteria causing foodborne diseases include Salmonella and Campylobacter . In low to middle income countries , where resources are limited and a significant part of the population cannot always afford treatment , foodborne diseases such as Campylobacteriosis can play an important role in child mortality . Chickens and undercooked chicken meat have been found to commonly harbour this bacterium . In countries like Kenya , where fast urbanisation is occurring and chicken farming systems are intensifying , diets and food retailing infrastructure are also changing . Scientific research has not yet well documented how widely distributed Campylobacter is in such changing contexts , and which risk factors can favour its presence . In this study , the researchers have investigated small chicken farms and chicken meat sellers in Nairobi , Kenya’s capital , to better understand the risk that Campylobacter could represent for human health . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"livestock",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"microbiology",
"vertebrates",
"diet",
"animals",
"animal",
"products",
"bacterial",
"diseases",
"farms",
"nutrition",
"meat",
"bacteria",
"campylobacter",
"... | 2018 | Campylobacter, a zoonotic pathogen of global importance: Prevalence and risk factors in the fast-evolving chicken meat system of Nairobi, Kenya |
We are remarkably adept at inferring the consequences of our actions , yet the neuronal mechanisms that allow us to plan a sequence of novel choices remain unclear . We used functional magnetic resonance imaging ( fMRI ) to investigate how the human brain plans the shortest path to a goal in novel mazes with one ( shallow maze ) or two ( deep maze ) choice points . We observed two distinct anterior prefrontal responses to demanding choices at the second choice point: one in rostrodorsal medial prefrontal cortex ( rd-mPFC ) /superior frontal gyrus ( SFG ) that was also sensitive to ( deactivated by ) demanding initial choices and another in lateral frontopolar cortex ( lFPC ) , which was only engaged by demanding choices at the second choice point . Furthermore , we identified hippocampal responses during planning that correlated with subsequent choice accuracy and response time , particularly in mazes affording sequential choices . Psychophysiological interaction ( PPI ) analyses showed that coupling between the hippocampus and rd-mPFC increases during sequential ( deep versus shallow ) planning and is higher before correct versus incorrect choices . In short , using a naturalistic spatial planning paradigm , we reveal how the human brain represents sequential choices during planning without extensive training . Our data highlight a network centred on the cortical midline and hippocampus that allows us to make prospective choices while maintaining initial choices during planning in novel environments .
Goal-directed behaviour rests on being able to rapidly evaluate the potential consequences of future actions . For example , consider the neuronal processing required for planning a new route home when a road you normally take is closed . Although previous studies have implicated anterior prefrontal regions in planning [1–5] , it has been difficult to tease apart the relative contributions of different prefrontal cortex ( PFC ) regions ( i . e . , rostral versus caudal or lateral versus medial PFC ) that respond to choices later in a sequence [6–7] . Moreover , the neural representation of how we rapidly make a series of novel choices remains unclear , because planning studies generally rely on extensive learning about the outcomes of alternative choices [2–5 , 7] . Here , we ascertained whether different anterior PFC regions signal uncertainty about novel sequential choices in a distinct manner during plan formation . Specifically , we were interested whether rostrodorsal medial PFC ( rd-mPFC ) , a brain region associated with imagining/simulating potential choices [8–10] , might be biased towards responding to choices later in a sequence , even in the absence of prior learning about the consequences of choices . We created a spatial planning task that would require little to no learning in which participants could call on an internal model of space deployed during exploration of the physical world [11] . Our task required participants to choose the shortest route between a start and goal location during functional magnetic resonance imaging ( fMRI ) scanning: participants viewed one of 220 mazes with either two routes ( shallow mazes ) or four routes ( deep mazes ) to the goal . Shallow mazes only had one choice point at the start location , whereas deep mazes also offered a second choice point deeper into the maze . This design enabled us to see how responses to plan formation were modified by the depth of prospection ( i . e . , the number of choice points ) and the uncertainty about those choices ( i . e . , the difference in lengths between the two available paths from each choice point ) . After planning their route , participants were asked to make a decision—at a specified choice point in a given maze—about the direction of the shortest path ( i . e . , optimal choice ) to the goal location . This gave us an additional measure ( reaction time [RT] ) to quantify the uncertainty about a choice beyond the difference in available path lengths ( Fig 1A ) . As with shallow mazes , participants were only prompted to make one choice after seeing a deep maze , but until the choice point was highlighted , they did not know which choice point ( starting point or the choice point further in the maze ) would be probed .
Participants made correct choices 84 . 0% of the time ( standard deviation [SD] = 5 . 13%; n = 29 ) during the fMRI experiment , with an average RT of 492 ms ( SD = 150 ms ) . In deep mazes , when participants were prompted with choices that were at junctions deeper in the maze ( i . e . , the second/prospective choice point of a two choice sequence ) , they made correct choices 84 . 9% ( SD = 9 . 89% ) of the time , with an average RT of 671 ms ( SD = 172 ms ) . There was no significant difference ( t ( 28 ) = 1 . 84; p = 0 . 077; SD = 6 . 62% ) in behavioural accuracy ( percentage of correct choices ) between deep ( mean = 85 . 2%; SD = 6 . 33% ) and shallow trials ( mean = 82 . 9%; SD = 5 . 88% ) . In contrast , there was a significant difference in RT ( t ( 28 ) = 14 . 3; p < 0 . 001; SD = 39 . 3 ms ) , with greater RTs for deep ( mean = 545 ms; SD = 148 ms ) versus shallow trials ( mean = 440 ms; SD = 152 ms ) . Notably , mean RTs were not correlated with accuracy across participants ( r = 0 . 257; p = 0 . 178 ) . Investigating the effect of path length differences on participant choice accuracy and RT in deep mazes , we observed a significant interaction between initial ( i . e . , first choice point ) and prospective ( i . e . , at the second choice point ) path length differences for both accuracy ( F ( 2 , 27 ) : 25 . 6; p < 0 . 001; Fig 2A ) and RT ( F ( 2 , 27 ) : 11 . 4; p < 0 . 001; Fig 2B ) . There was a significant positive linear trend for accuracy and initial path length differences ( F: 19 . 4; p < 0 . 001 ) but no similar linear trend for RT ( F: 0 . 18; p = 0 . 674 ) . Notably , we observed positive and negative significant linear trends with prospective path length differences for accuracy ( F: 13 . 9; p = 0 . 001 ) and RT ( F: 7 . 5; p = 0 . 011 ) , respectively . In shallow mazes , we observed a significant main effect of path length difference for both accuracy ( F ( 2 , 27 ) : 173 . 1; p < 0 . 001; Fig 2A ) and RT ( F ( 2 , 27 ) : 52; p < 0 . 001; Fig 2B ) . As expected , there was a significant positive linear trend for accuracy ( F: 354 . 6; p < 0 . 001 ) with larger path length differences , whereas there was a significant negative trend with RT ( F: 81; p < 0 . 001; Fig 2 ) . In deep mazes , accuracy was much lower when there were both small initial and prospective path length differences ( Fig 2A ) . We then investigated the influence of path length differences at the initial choice point on prospective choice behaviour . Unsurprisingly , when participants were prompted with the prospective choice point , we observed a significant ( p < 0 . 05 ) main effect of the prospective choice path length difference ( F ( 2 , 27 ) : 6 . 57; p = 0 . 005; S1 Fig ) on these choices and a linear increase in accuracy with larger path length differences ( F ( 2 , 27 ) : 8 . 47; p = 0 . 007 ) . However , we found no significant difference in prospective choice accuracy when split by the initial path length difference ( F ( 2 , 27 ) = 0 . 887; p = 0 . 424; S1 Fig ) . Investigating prospective choice RT , we observed a main effect of prospective choice RT based on the path length difference of the prospective choice point ( F ( 2 , 27 ) = 6 . 40; p = 0 . 005; S1 Fig ) and also when split by the ( unprobed ) initial path length difference ( F ( 2 , 27 ) = 5 . 70; p = 0 . 009; S1 Fig ) . Similar to choice performance , there was a negative linear trend for higher prospective choice point RT with smaller path length differences at the prospective choice point ( F ( 2 , 27 ) = 6 . 0; p = 0 . 021 ) . However , we did not observe a significant linear decrease in RT when we split prospective choice RTs by the path length difference of the initial choice/starting point ( F ( 2 , 27 ) = 3 . 1; p = 0 . 089 ) . Taken together , these results suggest that the path length difference of the initial choice did not affect performance on prospective choices but did influence deliberation time ( i . e . , RT ) . To assess the impact of planning sequential choices with varying processing demands , we classified deep maze trials by the path length difference between the shortest path and the other paths separately ( i . e . , initial , prospective , and unchosen path length differences; see Fig 1D for schematic of each path length comparison and S4 Table for list of regressors ) . Additionally , we asked whether subsequent choice behaviour ( RT and accuracy ) as well as other aspects of the planning task ( e . g . , the length of the shortest available path and whether the first or second choice was prompted during deep maze trials ) also explained brain activity during the planning phase . To summarize , we included the following parametric modulators for deep maze trials: the path length difference between the two shortest paths present at the starting point ( Initial Path Length Difference ) , the path length difference at the optimal second choice point ( Prospective Path Length Difference ) , the path length difference between the longest/least viable path in the initially unchosen direction and the shortest path ( Unchosen Path Length Difference ) , participants’ subsequent log RT during the choice phase ( Log RT ) , the length of the shortest available path , whether participants answered the subsequent choice trial correctly or not ( Accuracy ) , and whether participants were prompted to make an initial or prospective choice ( Prompted Choice ) . Importantly , the same parametric modulators were included for shallow maze trials except for Prospective Path Length Difference , Unchosen Path Length Difference , and Prompted Choice . We only report clusters that survive family-wise error ( FWE ) correction for multiple comparisons ( p < 0 . 05 ) at the statistical threshold of p < 0 . 005 uncorrected . The only exception is in the hippocampus , where all reported activations contain a peak-voxel that survives ( p < 0 . 05 ) small-volume correction ( SVC ) for the bilateral hippocampus . We first asked whether , during deep maze trials , there were fMRI responses specifically related to inferences about the prospective choice point , i . e . , blood-oxygen-level dependent ( BOLD ) changes related to choosing between the two paths at the second choice point that were not fully explained by path length differences at the initial choice point . We observed a very large cluster peaking in dorsal anterior cingulate cortex/pre-supplementary motor area ( dACC/pSMA; x = 6; y = 23; z = 37; Z-score: 5 . 08; Fig 3 ) with a sub-peak extending into rd-mPFC; x = −15; y = 38; z = 34; Z-score: 2 . 8; Fig 3 ) that responded to smaller prospective path length differences . Notably , there were also significant clusters in lateral frontopolar cortex ( lFPC; x = −27; y = 53; z = 4; Z-score: 3 . 86; Fig 3 ) , posterior parietal cortex ( PPC; x = 3; y = −73; z = 55; Z-score: 4 . 41 ) , left inferior temporal cortex ( x = −57; y = −43; z = −17; Z-score: 3 . 68 ) , and right cerebellum ( x = 30; y = −55; z = −26; Z-score: 3 . 73; S8 Table ) . Given that the rd-mPFC activation was a small sub-peak in a very large cluster centred on dACC/pSMA , we wanted to confirm whether there was truly a robust rd-mPFC signal selectively related to planning prospective choices and whether this signal differed from the other prefrontal responses observed in dACC/pSMA and lFPC . We therefore conducted a paired t test comparing responses to prospective path length differences versus initial path length differences in shallow mazes . We observed a significant rd-mPFC sub-peak ( x = −15; y = 38; z = 28; Z-score: 3 . 61; Fig 3D ) that responded to smaller prospective versus initial path length differences . The cluster covered the peak rd-mPFC voxel from the previous contrast and was centred on left dorsolateral prefrontal cortex ( dlPFC; x = −18; y = 17; z = 43; Z-score: 4 . 33; see S2 Fig for images of dlPFC peak ) . Crucially , this cluster was much smaller than the previous rd-mPFC result and did not include the dACC/pSMA region that responded to prospective path length differences . Likewise , we observed no significant difference in lFPC responses to smaller prospective versus initial path length differences . Of particular interest , the significant effect in rd-mPFC was driven by its significant response to both large initial and smaller prospective path length differences ( Fig 3C ) —a pattern that was not observed in lFPC or dACC/pSMA . The null result suggesting that lFPC does not respond to smaller prospective versus initial path length differences in shallow mazes should be interpreted with caution . Our general linear model ( GLM ) based on path length differences did not distinguish whether these fMRI results were due to the number of paths or the depth of planning . Indeed , when using a Shannon entropy model that compared RT-fitted uncertainty for prospective path length differences versus the absolute value of the difference between all four available paths lengths ( see S1 Text for details ) , we found that both rd-mPFC and lFPC selectively responded to prospective uncertainty ( S1 Text ) . In the reverse contrast , larger path length differences at the prospective choice point elicited responses in pregenual anterior cingulate cortex/ventromedial PFC ( pgACC/vmPFC; x = −3; y = 38; z = −11; Z-score: 3 . 69; Fig 3B ) . Notably , this finding is in contrast to a model-based analysis ( see Supplemental Results , S1 Text ) in which no parallel activation in pgACC/vmPFC related to decreasing prospective uncertainty was observed . This is possibly due to the inclusion of all path length differences and not just the two shortest paths available at either choice point . We also examined whether in both deep and shallow planning trials there were regions that responded to the difference between the two shortest path lengths available at the initial/first choice point ( See Fig 1D for illustration ) . We found that larger path length differences at the initial choice point elicited responses in the temporoparietal junction ( TPJ ) /angular gyrus , vmPFC ( S3 Fig ) , and posterior cingulate cortex ( PCC; see S3 Fig and S7 Table ) . Notably , rd-mPFC ( t ( 28 ) : 2 . 41; p = 0 . 023 ) but not lFPC ( t ( 28 ) : −0 . 468; p = 0 . 644; Fig 3C ) significantly responded to increasing initial path length differences ( see Table 1 for rd-mPFC and lFPC t-values related to other parametric regressor of interest ) . It is important to note that this vmPFC cluster only responding to large initial path length differences ( S3 Fig ) is rostral and superior to the pgACC/vmPFC cluster responding to both large initial and prospective path length differences . Following our results related to larger path length differences , smaller path length differences at the first choice point elicited responses in the dACC/pSMA , along with right dlPFC , anterior insula , and PPC ( see S7 Table ) . The dissociation between regional responses increasing and decreasing with initial path length differences reflects similar responses to larger versus smaller reward prediction errors observed during value-guided choice [12–14] . In a separate comparison , we examined responses to the difference between the shortest and the least viable counterfactual/unchosen path ( i . e . , what regions corresponded to an exhaustive search or pruning of all potential paths ) . We found that larger unchosen path length differences engaged the right angular gyrus/TPJ ( x = 51; y = −61; z = 25; Z-score: 5 . 98; Fig 4A and S9 Table ) , which was the strongest response we observed in any region relating to a path length difference regressor . Additionally , we found PCC ( x = 12; y = −46; z = 37; Z-score: 4 . 93; Fig 4A ) and right striatum ( x = 27; y = 8; z = 1; Z-score: 4 . 58; Fig 4A ) responses related to larger unchosen path length differences . Notably , in our Shannon Entropy model analysis , which did not have a specific parametric regressor accounting for unchosen path length differences , we found that right angular gyrus/TPJ and PCC both significantly related to increasing prospective uncertainty ( Supplemental Results , S1 Text ) . Taken together , these analyses suggest that angular gyrus and PCC prune unviable paths in deep mazes that afford demanding prospective choices . In contrast , smaller unchosen path length differences engaged dACC ( x = 9; y = 17; z = 34; Z-score: 4 . 41; Fig 4B ) and bilateral lateral occipital cortex ( LOC; left: x = −24; y = −88; z = −4; Z-score: 4 . 71; right: x = 27; y = −91; z = 10; Z-score: 5 . 1; S9 Table ) . Our post hoc region of interest ( ROI ) analyses revealed that neither rd-mPFC ( t ( 28 ) : −0 . 19; p = 0 . 85 ) nor lFPC ( t ( 28 ) : 0 . 399; p = 0 . 693 ) significantly encoded the unchosen path , further suggesting that these regions corresponded to rapid sequential inference but not necessarily an exhaustive search of all possible paths . Asking whether other aspects of mazes ( beyond path length differences ) influenced neural responses during planning , we investigated whether any fMRI signals during planning correlated with subsequent RT during the choice phase . During planning , fMRI signals in an extremely large portion of cortex—peaking in visual cortex—positively correlated with subsequent RT ( S10 Table ) . The large visual cortical cluster also encompassed ventral temporal regions extending into the bilateral posterior hippocampus ( left: x = −27; y = −37; z = −11; Z-score: 4 . 97; small-volume corrected ( SVC ) p < 0 . 001 ) , peaking in the right hippocampus ( x = 24; y = −37; z = −8; Z-score: 5 . 0; SVC p < 0 . 001; Fig 5 ) . Notably , the right posterior hippocampus peak showed a significantly stronger relationship with subsequent RT in deep versus shallow maze trials ( t ( 28 ) = 2 . 71; p = 0 . 011; Fig 5C ) . We also observed similar significant responses in smaller clusters in middle temporal gyrus and dACC ( see S10 Table ) . Likewise , we observed significant ( p < 0 . 05 ) positive correlations with increased subsequent RT in right angular gyrus/TPJ ( t ( 28 ) = 3 . 53; p = 0 . 002 ) and PCC ( t ( 28 ) = 3 . 01; p = 0 . 006 ) regions relating to larger unchosen path length differences , which provides additional evidence that these regions prune unviable paths during deep planning trials . The only negative correlation with subsequent RT was in the insula extending into a large portion of white matter ( x = 27; y = −10; z = 10; Z-score: 4 . 72 ) . We investigated which regions responded to the distance of the shortest available path length ( i . e . , whether the optimal path was distal or proximal to the goal location , irrespective of the other available paths ) . We observed responses in inferior occipital cortex extending into right posterior hippocampus ( x = 33; y = −37; z = −8; Z-score; 4 . 79; SVC p < 0 . 001; Fig 5B ) that correlated with increasing length of the shortest available path to the goal , along with dACC ( S11 Table ) . Conversely , bilateral TPJ , pgACC/vmPFC , rd-mPFC , precuneous , posterior superior temporal sulcus , and lateral PFC ( see S4 Fig and S11 Table ) correlated with decreasing distance of the shortest available path length . Further characterizing the functional contribution of different brain regions , we asked if the responses of different regions during the planning phase related to whether participants subsequently made a correct or incorrect choice . We observed a left hippocampal activation ( x = −18; y = −13; z = −17; Z-score: 3 . 87; SVC p = 0 . 044; Fig 5A ) that preceded correct choices with a subthreshold activation in right anterior hippocampus . Additionally , bilateral cerebellum and motor cortex activations during the planning phase related to correct choices ( S12 Table ) . However , the spatial extent of these performance results should be interpreted with caution , because the hippocampal cluster extended into a large portion of white matter . Conversely , there was a significant dACC/pSMA cluster ( x = 6; y = 17; z = 49; Z-score: 6 . 89; S5 Fig ) that preceded subsequently incorrect choices , which was the strongest activation observed in any contrast . We then tried to determine whether this response was feedback related , because it could have been due to an unobserved choice point . However , we found no significant difference between deep and shallow planning ( t ( 28 ) = 1 . 32; p = 0 . 198; S5 Fig ) . Likewise , adding a regressor encoding whether the initial or prospective choice point was highlighted in deep mazes ( Prompted Choice ) did not modify the robustness of the dACC/pSMA activation . Notably , we also observed significant clusters in bilateral anterior insula and intraparietal sulcus ( IPS ) preceding incorrect choices ( S12 Table ) . Investigating whether any regions responded differently to initial path length differences in deep versus shallow mazes , we found that a large cluster in PPC responded more strongly to smaller initial path length differences in shallow versus deep mazes . Likewise , we also observed smaller but significant clusters in premotor cortex ( PMC ) and dlPFC ( S6 Fig and S13 Table ) . We did not observe any other significant clusters responding to initial path length differences in deep versus shallow mazes . Next , we investigated whether during the planning phase there were any regions outside of the hippocampus whose responses correlated with subsequent RT for deep versus shallow planning trials differently . We found that visual cortex and right PMC correlated with increasing RT more strongly during shallow planning trials ( S6 Fig and S13 Table ) but did not find any other significant responses . When splitting responses to the length of the shortest path , we observed that inferior temporal cortex and superior frontal gyrus ( SFG ) responded to longer optimal path lengths more during deep versus shallow planning trials . Lastly , left LOC , left PPC , and right IPS responses to incorrect choices were higher for shallow planning trials ( see S6 Fig and S13 Table ) . We conducted a psychophysiological interaction ( PPI ) [15] analysis of whether the right posterior hippocampal region ( Fig 5C ) relating to longer subsequent RT was coupled with rd-mPFC as a function of planning depth ( mazes affording single versus sequential choices ) . We tested which regions exhibited increased coupling with hippocampus for deep versus shallow maze planning trials . Taking an 8-mm sphere around the rd-mPFC peak that selectively responded to smaller prospective path length differences ( x = −15; y = 38; z = 34 ) , we observed significantly increased coupling between the hippocampus and rd-mPFC for deep versus shallow planning ( t ( 28 ) = 2 . 69; p = 0 . 012; Fig 5D ) . Notably , the hippocampus coupled more strongly with rd-mPFC than any brain region ( peak voxel , x = 12; y = 47; z = 28; Z-score: 3 . 95; in a separate cluster that did not survive FWE cluster correction p < 0 . 05 at the whole-brain level ) . We did not observe any other significant clusters that coupled with the hippocampus anywhere else in the brain for deep versus shallow planning . To assess the functional relevance of hippocampal coupling with rd-mPFC during deep planning , we conducted a separate GLM analysis splitting deep planning trials based on whether the subsequent choice trial was answered correctly or not ( see Supplemental Methods in S1 Text for details of the GLM ) . We found that hippocampal coupling with rd-mPFC was significantly higher for correct versus incorrect deep planning trials ( t ( 28 ) = 3 . 04; p = 0 . 005; Fig 5D ) .
Highlighting distinguishable contributions to prospective planning for medial versus lateral anterior prefrontal regions , we find that rd-mPFC responds to difficult prospective choices while maintaining easier initial choices , whereas lFPC responds to prospective path length differences without being significantly modulated by initial path length differences . These findings are in line with the perceived capacity of anterior PFC to exploit recent reward trends during value-guided choice [16] and spatial navigation [2] . More specifically , our findings suggest that rd-mPFC might be guiding computations related to chaining the whole sequence of choices , whereas lFPC more exclusively relates to robust planning at the second , more prospective choice point independent of the initial choice . Alternatively , when there are increased computational demands at the initial choice point , rd-mPFC might deactivate when it is not clear what the first step should be , allowing lFPC or dACC to take over more robust prospective planning . The ability of lFPC and dACC to help robustly compute second-step choices is in line with previous findings related to counterfactual signals in FPC [6 , 17–18] and dACC signals related to strategic shifts in decision-making [19–20] , along with the smaller unchosen path length dACC signals presented here . Notably , our behavioural results showed initial choice path length differences modulate subsequent RT during prospective choices but not whether the choice was correct or not , which suggests more than one underlying computation occurring related to prospective planning . Taken together with our anterior PFC findings , these data broadly implicate at least two distinct anterior prefrontal computations when planning next-step choices in novel environments—one rapid and another more deliberative computation related to prospective planning . This lateral versus medial distinction parallels previous research on anterior PFC , where lateral areas are believed to process stimulus-independent ( i . e . , counterfactual ) information , whereas medial areas are engaged by stimulus-oriented information [21] . Furthermore , prospective choices responses in rostral mPFC were primarily dorsal , but the exact location of responses was highly variable over participants , which may relate to the high anatomical variability between individuals in an evolutionarily complex region [1] . Still , our observation of prospective planning responses throughout rostral mPFC is consistent with recent findings showing that different populations in mPFC contribute to internal strategy shifts ( see [22–24] for human evidence and [25–28] for rodent evidence ) and persistent activity in order to reevaluate sequential choices [29] . Our result showing increased lateral FPC responses to prospective path length differences might relate to the perceived function of FPC as a simultaneous evaluator of multiple options , perhaps due to a higher sampling capacity ( i . e . , capable of maintaining more information ) than rd-mPFC . Simultaneous evaluation of multiple options is necessary whether a decision is a sequential choice problem or not and is supported by the putative role of FPC in the rapid learning of novel abstract rules [30] and counterfactual choice [6 , 17 , 31] . Further work could focus on the influence of working memory load or cognitive control on types of planning [32–34] and how or whether different cognitive demands determine how a plan is formed or implemented and which prefrontal structures ( e . g . , dACC versus lFPC or rd-mPFC ) are engaged . Decisions often rely on prospection during multi-step events in order to anticipate a potential outcome , which is a process commonly linked with hippocampal-based memory ( [7 , 35–37]; see [38] for review ) . Furthermore , spatial planning in novel environments is usually associated with the use of a hippocampal-based internal model formed by exploration of the physical world [11] , yet corresponding evidence of hippocampal involvement during on the fly planning without extensive prior learning has been lacking . Here , we present evidence of posterior hippocampal responses related to increased deliberation for novel sequential choices and anterior hippocampal responses that relate to choice accuracy . Although our experiment is more akin to a perceptual decision-making task than virtual navigation , our results are still consistent with the role of the hippocampus during navigational planning [5 , 39–41] . More specifically , posterior hippocampal activity related to increasing distance between the start and goal locations—along with higher right posterior hippocampal activity prior to longer choice RT in deep mazes—helps link our spatial decision-making results to the putative role of the right posterior hippocampus , which is thought to encode memory related to the spatial layout of an environment [42–44] . In novel environments , posterior hippocampal functional connectivity with rd-mPFC increased during deliberative planning for deep mazes and was highest before choice trials that were answered correctly . Likewise , a recent fMRI study has shown increased anterior prefrontal coupling with the hippocampus during remembering and planning upcoming trajectories to goal locations [5] . Oscillatory coupling between the posterior medial temporal lobe and rostrodorsal portions of mPFC has been observed during dynamic spatial imagery [45] , and our data add further support that coupling between these regions could relate to comparison of novel choices with previous experience [38] . Notably , the hippocampus is also thought to play a key role in rapid incidental learning [46–47] . Our anterior hippocampus activation related to spatial planning performance illustrates how the hippocampus can contribute to quick model-based inferences during tasks with little to no learning . Yet it is still unclear how one-shot episodic learning might contribute to hierarchical planning . Investigating the neural representations of novel decisions might help uncover contextualization processes important for decision-making ( e . g . , chaining together sequential choices as a single decision outcome ) and episodic memory ( e . g . , chaining together individual learned representations into a cohesive episode ) . We have elaborated on the distributed neural responses that relate to rapid prospective planning , but the precise computations required for our task are unclear . One disadvantage of our task is the inability to probe the time scale of plan formation and implementation in novel environments , particularly when choice accuracy and RT are influenced differently by path length differences . Most planning studies test after extensive training and are biased towards action-by-action evaluation without the need to maintain prior choices [3–4 , 48–50] . With extensively trained choices , the neural computations leading to increased decision implementation/RT are well studied [51–52] . On the other hand , the anterior prefrontal regions selectively responding to prospective uncertainty make evaluations that are more akin to rapid approximation of the best looking trajectory or jumping ahead to the most important sub-goal , which are neural computations that have not been as well explored . Interestingly , this “jumping ahead” process resembles computations that facilitate generalization between similar sequential states ( successor representations ) during episodic learning [53–55] and also best-first forward search models [56] . Exploring the interactions between the successor representation , time scales , and heuristic pruning during plan formation could potentially help disclose the computations underlying rapid and efficient multi-step planning in novel environments [57–59] . Given that our experiment does not separate responses related to plan formation and implementation , the role of the vmPFC and dACC in our task is unclear . We observed dACC/pSMA responses related to an exhaustive comparison of path lengths ( comparing the shortest path with every other available path ) , with additional responses related to increased deliberation , longer distance between starting and goal locations , and , most prominently , subsequently incorrect choices . Taking into account the importance of the dACC in model updating [60–61] , it is not surprising that dACC responses would relate to uncertainty about potential trajectories at different choice points . However , due to the poor temporal resolution of our task , it is unclear whether dACC/pSMA responses are related to checking back on an uncertain initial choice point [62] , focusing on one choice point for an extended period of time [63] , performance monitoring [64] , or increased cognitive control caused by difficult choices ( see [61 , 65] for an in-depth discussion of the potential role of dACC in these behaviours ) . In contrast with dACC , vmPFC responses did not relate to comparisons of all available path lengths . Although subgenual portions of vmPFC responded to larger path length differences at both initial and prospective choice points , we did not observe any vmPFC signals that correlated with subsequently correct choices or quicker subsequent RT . A potential explanation for this result could be that vmPFC initially helps locate task-relevant sub-goals and signals an update of the current state [19 , 66] . Our findings also uncovered parietal responses that parallel activations observed in dACC and vmPFC . Smaller path length differences at both initial and prospective choice points engaged structures like PPC that have previously been implicated in value-guided decision-making when there is surprise and/or time pressure [60 , 67] . Notably , in other areas of the parietal lobe , right angular gyrus/TPJ and PCC responses during planning related to large initial and unchosen path length differences but also correlated with increased subsequent choice RT . One way to reconcile these seemingly contradictory results is that angular gyrus and PCC might be responding to irrelevant paths that need to be pruned/ignored [68] , which could then help us suddenly proceed/shift [69–71] to a subsequent decision during planning . Planning studies informed by recent work investigating divisive normalization during multi-alternative choice [72] and dACC–PCC interactions when pursuing unlikely choices [20] can potentially isolate the biophysical mechanism underlying pruning irrelevant alternatives during sequential decision-making . Notably , vmPFC , TPJ , and PCC responses to larger initial path length differences ( i . e . , certainty ) overlap with a brain network commonly observed during value-guided choice [14 , 73] . Specifically , regions that increased with the precision of beliefs about choices overlap with regions that respond to reward differentials , i . e . , greater value differences between chosen and unchosen options during value-guided decision-making [12–13 , 74] . Likewise , PPC and dACC/pSMA responses are also observed both during difficult value-guided choices ( i . e . , smaller value differences between chosen and unchosen options ) [12 , 14 , 75] and smaller initial path length differences . This suggests a similar mechanism guiding probabilistic choice in both spatial and value-guided decision-making , regardless of whether an explicit reward , like food or monetary gain , is present . We observed increased coupling between the hippocampus and rd-mPFC during sequential plan formation that also predicted subsequent performance . Notably , resting-state fluctuations in these same regions—along with angular gyrus and PCC—are also correlated and form the default network [76–78] . Promising clues relating internal models of the physical world to resting default network fluctuations might come from investigating hippocampal sharp-wave ripples: spontaneous oscillations that co-occur with the reactivation ( and pre-activation ) of hippocampal place cell ensembles [79–83] . Indeed , a recent study in macaques has shown that ripples selectively influence ongoing activity in the default network but not other resting-state networks [84] . Additionally , reactivation of hippocampal representations of previously learned goal locations has been observed during pre-navigational planning in familiar environments in humans [5] . Despite these promising findings , further research is still necessary to determine whether endogenous hippocampal interactions with cortical midline regions reflect reactivation/exploration of internal states in order to prepare decision-making networks for upcoming novel choices [59 , 70 , 85–86] . We present a task adapted from rodent spatial navigation that enabled us to elucidate core neural computations underlying our ability to make fast and robust multi-step inferences in the absence of prior learning [85–87] . Our findings highlight a unique contribution of brain regions that do not respond to an exhaustive search of possible options during planning like caudal PFC and premotor regions but rather maintain current choices while planning subsequent choices . These data offer preliminary evidence of rapid heuristic-based computations in rd-mPFC and the hippocampus during sequential planning that might further elucidate how we make inferences about states beyond a current subjective state [88] .
Thirty-four healthy adult participants performing the fMRI experiment gave informed written consent and were studied and compensated ( as approved by the local research ethics committee at University College London and in accordance with Declaration of Helsinki protocols ) . Due to poor participant performance ( answering less than 75% of trials correctly ) in the fMRI experiment , we removed five participants , leaving a final sample of 29 participants ( 14 female; 23 . 4 mean age in y; SD of 4 . 09 y ) . All participants were right-handed had normal or corrected-to-normal vision and reported good health with no prior history of neurological disease . Stimuli were presented using the Cogent ( http://www . vislab . ucl . ac . uk/cogent . php ) toolbox running in MATLAB ( Mathworks , Natick , MA , USA ) . Over the course of 220 trials , participants viewed 220 different mazes from a slightly tilted ( overhead ) viewpoint and later chose from first-person viewpoints within mazes generated using Blender ( http://www . blender . org ) . All mazes had a starting location ( a red square ) towards the bottom of the maze and a goal location ( a green square ) further into the maze . Mazes differed by hierarchical depth ( number of paths to a goal location ) : there were 110 mazes with two possible routes ( shallow mazes ) and 110 mazes with four possible routes ( deep mazes ) . In the scanner , participants were first presented with pictures of mazes of varying difficulty ( from our overhead viewpoint ) and then asked to determine the shortest path from a starting location ( a red square ) at the bottom of the screen to the goal location ( a green square ) . The overhead view appeared on the screen for 3 . 25 s , after which a location ( choice point ) along the path was highlighted briefly for 250 ms with an orange circle . The choice point location could either be the starting location or , if there were four paths to the goal location , a second choice point . Crucially , participants would only have to make a decision about one choice point for each trial . At any choice point , it was necessary to choose between two different directions , which could be left , forward , or right , with an additional option to select equal , if both routes were the same distance . No second choice points with two incorrect choices were ever chosen , only a second choice point along the optimal path after the starting location could be chosen ( due to viewpoint constraints , only 47 choice points further were chosen versus the initial starting point/red square , which was chosen 53 times ) . After the choice point was highlighted , a “zoomed in” viewpoint of this location ( always one square back and facing the same direction as the overhead viewpoint ) was presented . Depending on the possible direction at the location , participants had less than 1 , 500 ms to decide whether to go left , forward , right , or occasionally either direction . If no button press was made within 1 , 500 ms , the trial counted as an incorrect trial and the experiment moved on to the 1500-ms intertrial interval ( ITI ) phase . Participants never received any feedback or reward for making the correct choice . As soon as participants chose a direction , the ITI phase of a trial began . Participants repeated this trial sequence 110 times per session , for a total of two sessions . Sessions lasted approximately 10–15 min . Session order was counterbalanced between participants . All participants completed a brief practice session consisting of 40 mazes/trials before the experiment ( on a laptop outside of the scanner ) . Deep mazes contained another branch/choice between routes further in the maze , and the path length to reach the two choice points further in the maze was always equal . Mazes had square tiled floors and were 8 x 8 , 9 x 9 , or 10 x 10 squares in total area . In shallow mazes , path length differences were split between 2 , 4 , and 6 , with one catch trial per session having equal path lengths . In deep mazes , path length differences were split between 2 ( small difference ) , 4 ( medium difference ) , or 6 ( large difference ) squares ( for an example , see square tiles in the mazes presented in Fig 1 ) for the two paths at the starting location and a path length difference of 2 , 4 , or 6 squares at the optimal choice point in the maze . There was one catch trial for deep and shallow mazes in each session , each containing all equal path lengths ( path length differences of 0 ) . In sum , shallow trials could either have path length difference of 2 , 4 , and 6 , while deep maze trials could be 2 , 2; 2 , 4; 2 , 6; 4 , 2; 4 , 4; 4 , 6; 6 , 2; 6 , 4; 6 , 6; ( e . g . 4 , 2 would have a medium path length difference of 4 at the starting location , whereas the second choice point would have a small path length difference of 2; see Fig 1C for examples ) . Functional images were acquired on a 3T Siemens Trio scanner . BOLD T2*-weighted functional images were acquired using a gradient-echo EPI pulse sequence acquired obliquely at 45° with the following parameters: repetition time , 3 , 360 ms; echo time , 30 ms; slice thickness , 2 mm; inter-slice gap , 1 mm; in-plane resolution , 3 × 3 mm; field of view , 64 × 72 mm2; 48 slices per volume . A field-map using a double echo FLASH sequence was recorded for distortion correction of the acquired EPI [89] . After the functional scans , a T1-weighted 3-D MDEFT structural image ( 1 mm3 ) was acquired to co-register and display the functional data . Functional images were processed and analysed using SPM8 ( www . fil . ion . ucl . uk/spm ) . The first five volumes were discarded to allow for T1 equilibration . Standard preprocessing included correction for differences in slice acquisition timing , realignment/unwarping to correct for inter-scan movement , and normalization of the images to an EPI template ( specific to our sequence and scanner ) that was aligned to the T1 Montreal Neurological Institute ( MNI ) template . Finally , the normalized functional images were spatially smoothed with an isotropic 8-mm full-width half maximum Gaussian kernel . For the model described below , all regressors , with the exception of six movement parameters of no interest , were convolved with the SPM hemodynamic response function . Data were also high-pass filtered ( cut-off period = 128 s ) . Statistical analyses were performed using a univariate GLM with a rapid event-related experimental design . GLM1 was based on path length differences ( see task description for possible path length differences ) : for the two shortest paths present at the starting point ( Initial Path Length Difference ) , the path length difference between the shortest path and the longest unchosen path length that was not available at the second choice point ( Unchosen Path Length Difference ) , the path length difference at the second choice point ( Prospective Path Length Difference ) , log RT for the subsequent decision phase ( Log RT ) , length of the shortest available path ( Length of the Shortest Path ) , whether the participant made a correct choice during the subsequent choice phase ( Performance ) , and whether the first or second choice point was prompted for deep maze trials ( see Fig 1D for schematic showing the paths contributing to Initial Path Length Difference , Prospective Path Length Difference , and Unchosen Differences ) . For shallow trial regressors , there were only parametric regressors for Initial Path Length Difference , Log RT , Length of the Shortest Path , and Performance . Inferences about the effects of uncertainty were based upon t tests using the standard summary statistic approach for second-level random effects analysis ( see S1 Text for additional follow-up GLMs and corresponding results and S5 Table for a complete table of conditions and parametric regressors for each GLM ) . We conducted a PPI analysis [15] to examine hippocampal coupling with rd-mPFC and the rest of the brain during deep versus shallow planning trials . The group-level right posterior hippocampus peak ( x = 24 , y = −37 , z = −8 ) that correlated with increased RT served as a centre for the spherical region of interest ( 8-mm radius ) . The first eigenvariate from these ROIs constituted the physiological variable . The psychological variable was the contrast vector representing the task effect of deep versus shallow mazes . These regressors and their interaction term were estimated at the first level . Contrast images associated with the PPI regressor were then entered into a one-sample t test . Post hoc statistical analyses were conducted using 8-mm radius spheres in MarsBar [90] toolbox within SPM8 around the respective peak voxel specified in the GLM analysis . This allowed us to compare the effects of different parametric regressors of interest ( e . g . , to determine whether a length of the shortest available path effect was present in a region defined by an orthogonal main effect of prospective path length difference ) . This ensured we did not make any biased inferences in our post hoc analyses . Given the previously hypothesized role of the hippocampus in spatial planning , we report whether hippocampal peak-voxels survive SVC for multiple comparisons ( p < 0 . 05 ) based on a bilateral ROI of the hippocampus constructed using the SPM Anatomy toolbox [91–92] . For all analyses outside of the hippocampus , we report activations surviving an uncorrected statistical threshold of p = 0 . 005 and cluster-level correction for multiple comparisons ( FWE p < 0 . 05 ) , unless indicated otherwise . We also mention whether any significant clusters had a very large cluster extent ( k > 2 , 000 ) , and the cluster extent for every significant effect is reported in S7–13 Tables . Coordinates of brain regions are reported in MNI space . BOLD signal time courses in S5 Fig were plotted using the rfxplot toolbox [93] . | We are remarkably adept at inferring the consequences of our actions , even in novel situations . However , the neuronal mechanisms that allow us to plan a sequence of novel choices remain a mystery . One hypothesis is that anterior prefrontal brain regions can jump ahead from an initial decision to evaluate subsequent choices . Here , we examine how the brain represents initial versus subsequent choices of varying difficulty during spatial planning in novel environments . Specifically , participants visually searched for the shortest path to a goal in pictures of novel mazes that contained one or two path junctions . We monitored the participants’ brain activity during the task with functional magnetic resonance imaging ( fMRI ) . We observed , in the anterior prefrontal brain , two distinct responses to demanding choices at the second junction: one in the rostrodorsal medial prefrontal cortex ( rd-mPFC ) , which also signalled less demanding initial choices , and another one in the lateral frontopolar cortex ( lFPC ) , which was only engaged by demanding choices at the second junction . Notably , interactions of the rd-mPFC with the hippocampus , a region associated with memory , increased when planning required extensive deliberation and particularly when planning led to accurate choices . Our findings show how humans can rapidly formulate a plan in novel environments . More broadly , these data uncover potential neural mechanisms underlying how we make inferences about states beyond a current subjective state . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"learning",
"medicine",
"and",
"health",
"sciences",
"diagnostic",
"radiology",
"functional",
"magnetic",
"resonance",
"imaging",
"decision",
"making",
"prefrontal",
"cortex",
"brain",
"social",
"sciences",
"neuroscience",
"learning",
"and",
"memory",
"magnetic",
"reson... | 2017 | The Neural Representation of Prospective Choice during Spatial Planning and Decisions |
Virus populations can display high genetic diversity within individual hosts . The intra-host collection of viral haplotypes , called viral quasispecies , is an important determinant of virulence , pathogenesis , and treatment outcome . We present HaploClique , a computational approach to reconstruct the structure of a viral quasispecies from next-generation sequencing data as obtained from bulk sequencing of mixed virus samples . We develop a statistical model for paired-end reads accounting for mutations , insertions , and deletions . Using an iterative maximal clique enumeration approach , read pairs are assembled into haplotypes of increasing length , eventually enabling global haplotype assembly . The performance of our quasispecies assembly method is assessed on simulated data for varying population characteristics and sequencing technology parameters . Owing to its paired-end handling , HaploClique compares favorably to state-of-the-art haplotype inference methods . It can reconstruct error-free full-length haplotypes from low coverage samples and detect large insertions and deletions at low frequencies . We applied HaploClique to sequencing data derived from a clinical hepatitis C virus population of an infected patient and discovered a novel deletion of length 357±167 bp that was validated by two independent long-read sequencing experiments . HaploClique is available at https://github . com/armintoepfer/haploclique . A summary of this paper appears in the proceedings of the RECOMB 2014 conference , April 2-5 .
Genetic diversity is an important characteristic of evolving populations and it affects the chances of survival in changing environments . Assessing the genetic diversity of a population experimentally is generally labor-intensive and difficult . Populations of individual cells or viruses , however , can be analyzed efficiently using next-generation sequencing ( NGS ) . Although single-cell approaches are still immature , direct NGS of mixed samples at deep coverage allows for probing populations in great detail . The challenges with this bulk sequencing approach are ( i ) to separate sequencing errors from genetic variation , ( ii ) to assemble the short NGS reads into an unknown number of different , unknown , longer haplotype sequences , and ( iii ) to estimate their frequency distribution . Viruses such as human immunodeficiency virus ( HIV-1 ) and hepatitis C virus ( HCV ) populate their hosts as swarms of related but genetically different mutant strains , each defined by its haplotype sequence . The structure of such a mutant cloud , which is often referred to as a viral quasispecies [1] , is of clinical importance , because it has been shown to affect virulence [2] and pathogenesis [3] . In addition , low-frequency genetic variants may harbor resistance mutations that are capable of evolutionary escape from the selective pressure of host immune responses [4] and of medical interventions , such as anti-viral drug treatment [5] . NGS is currently introduced into clinical diagnostics , but the de facto standard procedure for assessing the quasispecies structure is simply based on single-nucleotide variant ( SNV ) calling . This approach allows only for estimating the per-site allele frequency spectrum of the virus population and it ignores patterns of co-occurrence among mutations . This limitation is critical , because epistatic interactions are abundant in RNA viruses [6] . Hence , one cannot predict viral phenotypes without knowing the underlying mix of haplotypes . Here , we address this challenge and present a computational approach for the viral quasispecies assembly problem . The viral haplotype reconstruction problem is related to the human haplotype reconstruction problem , but it differs in several key aspects and faces different challenges . First , the number of unique haplotypes in a viral quasispecies is unknown unlike in the case of human diploid genomes . Second , viral populations typically exhibit more than two variants at each polymorphic locus and often all four different nucleotides . Hence , viral haplotypes cannot be described by binary sequences . Third , in a viral quasispecies , low-frequency variants are abundant and of clinical importance , yet they are difficult to distinguish from technical sequencing errors . Finally , RNA virus genomes are orders of magnitude shorter than the human genome , but exhibit more diversity within one host than the ∼0 . 1% diversity between the two parental human haplotypes [7] . Several methods for viral haplotype reconstruction have been developed in recent years , specialized for different NGS technologies , experimental designs , and quasispecies structures . In general , reconstruction can be performed either locally , in a genomic region that can be covered by the average read length , or globally , over longer regions such that overlapping reads are necessary for assembly . Local reconstruction means estimating the number of locally unique haplotype sequences and , at the same time , correcting sequencing errors . Probabilistic clustering [8]–[11] and k-mer statistics [12] have been proposed for this task . Global reconstruction is more challenging , as it requires computational solutions for assembling NGS reads , which has proven itself to be demanding even in settings without poly-ploidy [13] . For quasispecies assembly , approaches from different domains have been developed: ( i ) probabilistic mixture models [14] , ( ii ) hidden Markov models [15] , ( iii ) sampling schemes [16] , ( iv ) combinatorial approaches based on analyzing the read overlap graph [8] , [17]–[19] , ( v ) coloring of overlap and conflict graphs by constraint programming [20] , and ( vi ) exploiting the “identical by descent” information [21] in the HapCompass framework [22] , originally designed for diploid single nucleotide polymorphism data . The performance of global haplotype reconstruction depends on several factors , including the true underlying diversity of the population , the distribution of amplification and sequencing errors , the read length , and the distribution of the read coverage along the genome [23]–[25] . A major shortcoming of all existing methods is that they are unable to handle large insertions or deletions ( indels ) . For example , large deletions can result from erroneous replication or , as observed recently in HIV-1 , they may occur as alternative splice variants [26] . In the context of analyzing structural variation in the human genome , such as indels of varying sizes , the use of paired-end reads has been instrumental . For viral haplotype reconstruction , however , approaches that systematically exploit paired-end information are lacking . In this paper , we present a new quasipecies assembly method for paired-end reads , called HaploClique , based on enumeration of maximal cliques ( max-cliques ) as a general approach to clustering NGS paired-end reads . Although , in general , the runtime of enumerating all max-cliques in a graph is exponential , it has recently been shown that the graphs induced by overlapping NGS reads can be handled efficiently [27] , [28] . Here , we exploit this fact for the quasispecies assembly problem and develop a probabilistic model of sequence and structural similarity between reads . Using max-clique enumeration for reference-based read assembly is orthogonal to combinatorial approaches for de novo assembly that rely on path finding in de Bruijn or similar graphs [29]–[31] . Instead of computing paths , we iteratively transform max-cliques into super-reads and then seek max-cliques of super-reads , thereby obtaining haplotype segments of increasing length . The haplotype segments can eventually be extended to global haplotypes if the degree of heterogeneity of the viral quasispecies is high enough . HaploClique is related to max-cut-driven approaches in human haplotype reconstruction [32] , but the computational complexity of those approaches is prohibitive for virus populations of high and unknown ploidy . While HaploCliques enumerates all max-cliques , a max-cut approach seeks an optimal cut of the overlap graph . HaploClique explicitly incorporates paired-end information for assembling viral haplotypes . We define the insert as the unsequenced fragment between the two ends of a paired-end read . We use linkage information among variant alleles in the distant pairs to identify reads that stem from the same haplotypes and generate error-corrected paired-end super-reads . Paired-end reads allow to bridge homogeneous , and hence ambiguous , genomic regions if the insert size is sufficiently large . They also increase the statistical power to distinguish local haplotypes from sequencing errors in homogeneous regions if the paired read is located in a more heterogeneous region . Employing our iterative clique enumeration procedure , we show that error-free full-length HIV-1 viral haplotypes can be reconstructed in a heterogeneous mix of five viral strains in silico from a data set with mean coverage of 600× . Furthermore , we demonstrate that , unlike existing methods , HaploClique can detect large indels in mixed virus populations in silico and in vivo . Finally , we apply HaploClique to a HCV Illumina paired-end NGS data set and predict a novel deletion of length bp that has been confirmed independently by two long-read NGS platforms .
HaploClique integrates paired-end and base quality information for improved sequencing error correction and haplotype frequency estimation , which we assess first . Second , we evaluate HaploClique's behaviour when confronted with low heterogeneity among the different haplotype strains . Third , we demonstrate HaploClique's ability to detect large insertions and deletions in the quasispecies by making use of paired-end information . Fourth , we evaluate the quality of the local and global haplotypes that HaploClique predicts . Lastly , we compare HaploClique to state-of-the-art tools ShoRAH [33] , PredictHaplo [14] , and QuRe [16] in quasispecies reconstruction of a simulated five virus mix of well-known HIV-1 lab-strains . In all of the following experiments , we simulated Illumina 2×250 bp paired-end reads using SimSeq [34] with fragment size 600 bp . To make the simulated data as realistic as possible , we estimated the required error profiles from an in-house MiSeq data set of a mixture of known HIV-1 strains . The average error-rate was 0 . 33% per base . For the application of HaploClique to a clinical sample , HCV RNA was extracted from the plasma collected from a subject isolated 135 days post infection and the NS5 region RT-PCR amplified as previously described [35] . In this subject there was experimental evidence of antigen-specific CD8+ T cell responses targeting two epitopes in the NS5 region ( K2629SKRTPMGF and W2820LGNIIMFA ) . The NS5B region encodes for the RNA-dependent RNA polymerase and is essential for the replication of the virus . This amplicon was sequenced to a coverage of 80 , 000× on a MiSeq instrument using a 2×250 bp read kit . The resulting reads were aligned using BWA-MEM [36] . We found the insert size distribution to have a mean of 155 bp and a standard deviation of 167 as estimated by HaploClique . Despite this large standard deviation , HaploClique was able to discover a bp deletion . No other indels were reported by HaploClique . In two independent sequencing runs of the same amplicon , once on a 454/Roche GS FLX+ system and once on a PacBio instrument , the presence of the deletion was confirmed . Both technologies yield longer reads than MiSeq that could successfully be aligned across the deletion breakpoint , allowing to determine breakpoint coordinates at base-pair resolution . For the alignment of the longer reads , extreme affine gap costs have been used to find the deletion . In general , this leads to alignment artifacts in other regions , causing false positive haplotype calls . With a read length of 250 bp , we did not succeed to align reads across the large deletion . Comparing coordinates , we found that the start positions predicted by HaploClique was 15 bp off the true position and the true length amounted to 444 bp . That is , the length difference between true and predicted deletion amounted to 87 bp , or 0 . 52 standard deviations .
We have presented HaploClique , a method for local haplotype reconstruction , structural variant detection of large insertions and deletions , and global haplotype assembly , which represents a principled approach to viral quasispecies assembly from NGS paired-end reads . HaploClique builds on a read alignment graph as underlying combinatorial model , where nodes correspond to single-end or paired-end alignments of reads . Edges are modeled in a probabilistic fashion . They are based on sequence similarity of the read overlap by incorporating phred-style quality scores in combination with a position-wise prior for the non-overlapping parts of the reads , and on a criterion that measures insert size compatibility of the two alignments . While the sequence similarity criterion accounts for correct assembly of reads , the insert size criterion allows for detecting insertions and deletions in viral haplotypes that cannot be detected from single-end read alignments alone . We suggest a model that unifies sequencing error correction , clustering reads into haplotype groups , as well as assembling reads into longer fragments , all of which naturally emerge from the model . In the read alignment graph , max-cliques represent maximal read sets that overlap and represent ( locally ) identical haplotype sequence . The advantage of the max-clique computation is twofold . First , it clusters reads , thereby separating reads stemming from different haplotypes . Second , it enables sequencing error correction in a way that can make full use of co-occurrence , that is , statistical correlation of variant alleles within reach of the reads participating in a max-clique . In particular , the error correction exploits paired-end information if provided . The improved error correction is important , as it gives rise to improved frequency estimates and allows for distinguishing between haplotypes whose pairwise distance is below 1% . HaploClique allows for reconstructing full-length global haplotypes using a read assembly procedure that is orthogonal to all existing assembly methods . In our iterative approach , we alternate between transforming max-cliques into super-reads , which form the nodes of a new alignment graph , and finding max-cliques in the new graph . We repeat this process until convergence , which is established when super-reads do not grow any longer . HaploClique depends on three parameters to be adjusted manually: minimal read overlap , , a threshold for the probability that two overlapping reads stem from locally identical haplotypes , , and the minimal coverage to call the super-read sequence . In general , if one of the parameters is decreased , the number and size of cliques will increase . If and are too small , the purity of cliques will decrease , meaning reads from different but very similar haplotypes cluster . If is too large , cliques will grow slower and less frequent haplotypes may be missed . If is too large , reads are more likely to cluster not only if they stem from the same haplotype but also if they have technical errors in common; this leads to lower error correction efficiency . If is too small , there might not be enough statistical power to correct for sequencing errors . If is too large , the false negative rate will rise , as low-frequency haplotypes do not provide enough reads to form cliques . We used two different parameter sets for HaploClique . In the first iteration , local haplotype reconstruction with error correction is performed and we chose and . In practice , the results are insensitive to the parameter choice ( Figure S2 ) . For the following iterations , the quasispecies assembly , we assume that haplotypes are error-corrected and must match perfectly . We set to account only for the stochasticity of the Phred scores . We evaluated HaploClique by extensive simulation studies . The simulated haplotypes were well-known and much analyzed HIV-1 virus strains . We kept coverage in the simulation study rather low , so as to evaluate our tool in the presence of only weak signals . We did this also in comparison with extant state-of-the-art tools . We demonstrated that our approach has superior error correction capabilities . This , in turn , yields accurate haplotype frequency estimates , even at the rather low coverage of 120× per haplotype . The tools we compared to were not able to provide similarly accurate frequency estimates . HaploClique proved to be insensitive to a coverage reduction of one order of magnitude with respect to prevalent sequencing experiments , which commonly operate at 5000× or higher . Beyond improved frequency estimates , we also improve haplotype sequence reconstruction . In all experiments , more than 99% of the haplotype segments we predict perfectly matched true haplotype sequences . None of the other tools generated even only one such perfectly matching segment , possibly because they require much higher coverage . This improvement in terms of accuracy may be due to the probabilistic model that treats error correction and assembly within one unifying framework . Our simulations also indicated that the degree of heterogeneity required in order to reconstruct large enough haplotype segments can be lower than 1% . We also ran HaploClique on a real , Illumina MiSeq dataset of coverage 80 , 000× , which was found to consist of two HCV strains one of which had a frequency of only approximately 3% and contained a deletion of size 444 bp , as conformed by independent 454/Roche and PacBio sequencing experiments . In the MiSeq dataset , the deletion in question could not be detected by state-of-the-art read alignment tools . HaploClique successfully predicts this deletion , despite the large standard deviation of the fragment size distribution ( 167 bp ) . These experiments document that our method can detect large deletions also in Illumina paired-end datasets that otherwise would be difficult to identify . Despite these improvements over previous methods , there are limitations of this approach . For example , the runtime of HaploClique is exponential in the read coverage . This feature is critical in the first two iterations of the procedure , before the number of reads is decreased . We observed that , approximately , the runtime doubles for each additional 250 reads of coverage . The baseline runtime was ∼4 minutes for a data set with coverage 1000× , on a single 3 GHz core . To overcome this computational bottleneck , one may perform the first iterations of haplotype reconstruction on subsets of the data and then assemble the merged results . Another extension that may decrease the runtime is to employ improved clustering techniques [37] . In the future , we also plan to explore on human whole-genome data , including polyploid cancer genomes , to perform error correction of the paired-end reads by local haplotype reconstruction and to assemble diploid haplotypes . This problem is more challenging due to the larger genome size and smaller levels of diversity , but several ideas presented here and implemented in HaploClique may prove useful for this task .
Let be the set of all reads from a viral quasispecies sequencing experiment and the set of their alignments to a reference genome as computed by a read aligner . In this paper , we assume that each read can be uniquely mapped , which is a reasonable assumption for short , non-repetitive viral genomes . In our experiments , we use the Illumina MiSeq technology for sequencing and BWA-MEM [36] as a read aligner . However , HaploClique depends on the sequencing technology and read mapper only insofar that it expects reads to be equipped with quality scores and that the reads can be properly aligned . We construct a graph where the read alignments are the vertices . An edge indicates that the two alignments and overlap sufficiently and that the corresponding reads are likely to originate from ( locally ) identical haplotypes . More precisely , we draw an edge between and if they satisfy two criteria based on sequence similarity and insert sizes , respectively . While the sequence-based criterion ensures that the reads and do not exhibit mutually contradictory sequences , the insert size-based criterion guarantees that and do not contradict each other in terms of their fragment sizes . We allow alignments and to be any combination of single- and paired-end reads , but the size-based criterion applies only if both alignment are based on paired-end reads . Cliques are fully connected subgraphs of the read alignment graph . They indicate groups of reads that are all likely to stem from locally identical haplotypes . Hence max-cliques form maximal groups of reads that originate from locally identical haplotypes . The algorithm proceeds by first sorting all nodes from left to right , in ascending order of the alignment coordinates such that starts left of . The algorithm then computes maximal cliques by processing all nodes in this order . Let be the induced subgraph of with vertices and let be all max-cliques in . For a node , let be all nodes that are connected to by an edge . If the rightmost coordinates of all alignments in a clique are smaller than the leftmost coordinate of , this clique cannot be further affected by any node — such cliques are maximal in and can be output if only nodes are left to be considered . After having processed all nodes , we declare all cliques that can still be affected by nodes to be active . When processing node , we compute its neighborhood and add a new clique if intersecting with each active clique yields the empty set . Otherwise , for each active clique , we set if , and we add a new clique if . Among all new cliques to be added , we eliminate duplicates . Max-clique enumeration is related to the problem of finding a minimum clique cover [38] , where a minimal set of non-overlapping max-cliques is sought , whose vertices cover the graph . Each max-clique is a set of reads with mutually compatible alignments . Therefore , we can construct consensus sequences for max-cliques , which we refer to as super-reads . The purpose of super-read construction is two-fold . First , super-reads represent haplotype segments . Second , super-reads can be used as input to further iterations of max-clique enumeration , with the goal of global haplotype reconstruction , which is discussed in the next section . To construct super-reads , let be the alignments participating in a max-clique and let be the set of positions where at least of these alignments contain non-gap characters . We recall that denotes the probability that nucleotide gave rise to position in read , although might differ from due to sequencing errors . We determine the nucleotide sequence of the super-read by means of a weighted position-wise majority vote . We set , for each position , where is defined to be zero when does not cover position . The parameter ensures that the super-reads have sufficient coverage of high quality . For later frequency estimation , we keep track of which original reads gave rise to which super-read . For global haplotype reconstruction , or quasispecies assembly , we iterate the clique enumeration procedure . We align reads against the reference sequence , construct the read alignment graph , find max-cliques , and merge them into larger super-reads with updated phred scores that reflect corrected error profiles . In the next iteration , we use these super-reads as reads and restart the procedure until number and length of super-reads have converged . For the assembly step , we assume that reads have already been error-corrected in the first iteration and set . Reads have to match perfectly and only allows for stochasticity of the Phred scores . We start the iterations with a relative overlap of . Once the length and number of super-reads converged , we decrease by down to a minimum of . We estimate haplotype abundance by counting the number of ( original ) reads that participate in the super-reads giving rise to the haplotypes . Original reads may participate in several super-reads and thereby contribute to abundance counts for several haplotypes . We resolve this issue by keeping track of the original read in each iteration , such that each read can be assigned to the final haplotypes after convergence . Reads contributing to several haplotypes abundances are then taken into account by weighting them accordingly . The MiSeq raw read data set is available through the Sequence Read Archive under the BioProject accession number SRP034655 . The MiSeq 2×250 bp error profiles for SimSeq [34] used in the simulations are available at https://github . com/armintoepfer/haploclique under data . A summary of this paper appears in the proceedings of the RECOMB 2014 conference , April 2-5 [42] . | Humans infected with a virus , such as the human immunodeficiency virus ( HIV-1 ) or hepatitis C virus ( HCV ) , host a population of billions of virus particles . Among these , there is an unknown number of genetically different strains , some of which can harbor drug resistance and immune escape mutations . It is of clinical importance to know the DNA sequences and abundances of these variants , as they can affect treatment outcome . Here , we present HaploClique , a computational approach to reconstruct these sequences and to predict large insertions and deletions from paired-end next-generation sequencing data . Using simulations , we demonstrate that HaploClique can reconstruct full-length HIV-1 variants from low-coverage samples . Using real-world clinical data , we predict a novel deletion of 357±167 bp in a HCV patient sample that has been validated by two independent long-read sequencing experiments . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"genome",
"sequencing",
"genome",
"analysis",
"tools",
"structural",
"genomics",
"haplotypes",
"virology",
"population",
"genetics",
"biology",
"genomics",
"microbiology",
"computational",
"biology",
"sequence",
"assembly",
"tools"
] | 2014 | Viral Quasispecies Assembly via Maximal Clique Enumeration |
Graph-based representations are considered to be the future for reference genomes , as they allow integrated representation of the steadily increasing data on individual variation . Currently available tools allow de novo assembly of graph-based reference genomes , alignment of new read sets to the graph representation as well as certain analyses like variant calling and haplotyping . We here present a first method for calling ChIP-Seq peaks on read data aligned to a graph-based reference genome . The method is a graph generalization of the peak caller MACS2 , and is implemented in an open source tool , Graph Peak Caller . By using the existing tool vg to build a pan-genome of Arabidopsis thaliana , we validate our approach by showing that Graph Peak Caller with a pan-genome reference graph can trace variants within peaks that are not part of the linear reference genome , and find peaks that in general are more motif-enriched than those found by MACS2 .
Transcription factors are known to play a key role in gene regulation , and detecting regions associated with transcription factor binding is an important step in understanding their function . The most common technique used to detect transcription factor binding sites is ChIP-seq , combining chromatin immunoprecipitation ( ChIP ) assays with sequencing ( seq ) . A ChiP-seq experiment involves obtaining DNA fragments that bind to the transcription factor of interest and sequencing arbitrary ends of these fragments , yielding short reads . Obtaining putative binding regions from these reads is done using computational techniques known collectively as performing peak calling . Several peak callers , programs to perform peak calling , have been developed for this purpose , for example MACS2 [1] and SPP [2] [3] . Common for all current peak callers is that they take reads mapped to a linear reference genome , such as GRCh38 , as input . Graph-based reference genomes offer a way to include known variants within a population in the reference structure [4] . The software package vg supports mapping reads to a graph-based reference genome with potentially increased accuracy [5 , 6] as compared to mapping reads to a standard linear reference genome using tools like BWA [7] or Bowtie [8] . Several types of genomic analyses , such as variant calling and haplotyping , can now be performed using graph-based references [5 , 6] . However , no tool currently exists for performing peak calling on graph-based references .
A problem with using motif enrichment to compare the peaks detected by MACS2 and Graph Peak Caller is that the two peak callers use different sets of alignments . Thus , we checked whether vg tends to align reads to the graph reference genome that more often end up in alignments with motif match than what BWA does . We found no such tendency . In aggregate for the peaks detected on A . thaliana , we found on average 0 . 0208 motif matches per graph alignment , and and on average 0 . 0214 motif matches per linear reference genome alignment . We also checked whether the higher motif match among the peaks detected by Graph Peak Caller might be a result of peaks coming from artificially clustered alignments , something which can be the case if there are complex regions in the graph reference containing the motif . Such regions could allow for many different reads to align to the same place , resulting in high-scoring peaks with motif match . We found that 99 . 6% of the motif matches among the peaks detected by Graph Peak Caller are consistent with a single haplotype . For the unique peaks detected by Graph Peak Caller , 98 . 6% of the motif matches were consistent with a single haplotype . For each detected peak , we also checked what percentage of the reads among those aligned to the peak were aligned to one or two haplotypes ( allowing for diploidy ) . In aggregate , we found that 55 . 6% of the unique peaks detected by Graph Peak Caller had all reads compatible with maximum two haplotypes . We found that the peaks having alignments from more than two haplotypes were more often enriched for motif , which might indicate that peaks coming from mismapped reads to high complexity areas inflate the motif match percentage . However , when we removed those peaks from the analysis , the ratio of motif peaks matching motif among the peaks detected by Graph Peak Caller is still significantly higher than the ratio of peaks matching motif among the peaks detected by MACS2 . In addition , the peaks found by MACS2 are also more motif enriched when the vg alignments overlapping those peaks are consistent with more than two haplotypes , indicating that the correspondence between multiple haplotypes and motif enrichment is not solely attributable to mismapping of reads containing motifs to high-complexity regions . The full analysis results of motif enriched alignments and haplotypes can be found in S3 Appendix .
We have developed Graph Peak Caller , a tool for performing peak calling from ChIP-seq reads mapped to a graph-based reference genome . Graph Peak Caller is based on the same principles as MACS2 . We have validated our approach by using both Graph Peak Caller and MACS2 to call peaks using ChIP-seq datasets on A . thaliana , showing that the peaks found by Graph Peak Caller in general are more enriched for DNA-binding motifs than those found by MACS2 on a linear reference genome . Graph Peak Caller is also able to provide candidates for differentially expressed peaks , and together with vg it provides a first method for doing peak calling on graph-based reference genomes .
Our approach to graph-based peak calling is implemented in an open source Python 3 package , Graph Peak Caller . Graph Peak Caller was developed by extending the methodologies and concepts from MACS2 to directed acyclic graphs ( DAGs ) . The MACS2 algorithm can be divided into five steps: estimating the fragment length , creating a fragment pileup by extending input reads to match the estimated fragment length , calculating a background track based on local and global average number of reads , calculation of p/q scores based on the fragment pileup and background track , and finding peaks based on thresholded scores . We have adopted each of these steps to work on DAGs . Fig 1 illustrates the method on a graph-based reference genome , and the following describes the details of each step . Graph Peak Caller uses the linear estimation algorithm from MACS2 to estimate the fragment length f by using the linear path through the graph with the highest number of aligned reads as reference . Graph Peak Caller generates the fragment pileup by extending each read to the estimated fragment length f , and counting the number of extended reads that cover each base pair in the graph . For a single read with length r , the extension is done by including all possible paths of length f − r in the graph that start at the read’s end position , using a breadth first search . The background track is an estimate of the expected number of reads mapping to each position in the reference . This is , for a given position in the reference , estimated by measuring the amount of reads mapping in the “neighbourhood” of that position . The reads can either be the input reads or a set of control reads . On a linear reference genome , the background track is simply estimated by taking the average pileup count in a local window around each base pair . This is less trivial to do on a graph-based reference genome , since the concept of neighborhood is not as well defined . We solve this problem by projecting the graph onto a single linear path where parallel paths are projected to the same position on the linear path . This allows us to perform background track estimation much the same way as MACS2 does , using a linear reference , and then projecting the resulting track back to the graph again . If control reads are used to generate the background track , the background track is scaled with the ratio of control reads to input reads . The fragment pileup and background track are then treated as counts and rates in Poisson distributions , and p-values are computed for each position for the observed count , given the corresponding rate . Since one test is performed for each position in the graph , we compute q-values ( adjusted p-values ) to control the false discovery rate . The q-values are thresholded at a user specified threshold , yielding a binary track of potential binding regions . Graph Peak Caller then removes small gaps ( similarly to MACS2 ) between these potential binding regions . On a graph , this is done by joining regions that are connected by a path shorter than the read length . If a gap consists of several paths , all paths of length shorter than the read length are included in the joined region . Then , the resulting regions are grouped into connected subgraphs , representing areas of potential binding events . The final peaks are selected by finding the path through each subgraph that has the highest number of input reads mapped to it . Similarly to MACS2 , peaks that are shorter than the estimated fragment length are removed . For each subgraph , Graph Peak Caller can also report an “alternative” peak in addition to the main peak . This is done by using Fimo [12] to estimate the exact location within the peak subgraph that matches the binding motif , and looking for an alternative path through this area which is covered by at least one input read . Such alternative peaks can be used to infer differential binding . To test our peak caller , we used vg [6] to create a whole genome Arabidopsis thaliana reference graph by using variants from the 1001 Genomes Project [13] . We selected all transcription factors listed in the transcription factor database of Expresso [14] that also had a motif in the Jaspar database of transcription factor binding profiles [15] , resulting in a set of 5 transcription factors: ERF115 , SEP3 , AP1 , SOC1 , and PI . ( Two transcription factors , SVP and ATAF1 , were omitted due to invalid fastq files . AP2 and AP3 were omitted based on their close relatedness to AP1 . Also , PIF3 was omitted since neither the detected binding events by Graph Peak Caller nor MACS2 had any association with the motif we found in the Jaspar database ) . Raw ChiP-seq reads were downloaded from the NCBI Sequence Read Archive ( SRA ) ( SRA accession numbers in S1 Appendix ) and trimmed using Trim Galore ! v0 . 4 . 4 [16] ( default parameters ) . Reads were mapped both to our graph-based reference genome using vg and to the Tair10 [10] reference genome using BWA v0 . 7 . 12 ( bwa aln followed by bwa samsse , default parameters ) . In the linear case we filtered out low-quality alignments using SAMtools v0 . 1 . 19 [17] with the command samtools view -F 1804 -q 37 , and for the graph alignments we used vg filter -q 37 . MACS2 v2 . 1 . 0 was used to call peaks on the linear reference genome , using default parameters . We created DNA-binding motif enrichment plots ( Fig 5 ) for each set of detected peaks ( URLs to the motif models that were used are in S1 Appendix ) . We have created a Docker repository with the A . thaliana graph-based reference genome , Graph Peak Caller , vg and all other software and scripts used to generate the results in this article . A simple guide on how to re-run the experiments can be found in the wiki in the Github repository for Graph Peak Caller . To investigate how MACS2 performed when using vg alignments projected to the linear reference genome , we projected each vg alignment by finding a corresponding start and end position for the alignment on the linear reference genome . Given a start or end position on the graph , the new corresponding position on the linear reference genome was found by finding the shortest distance in the graph going from the original position backwards through the graph to a node shared with the linear reference genome , and then going this distance forward through the graph by following only nodes shared by the linear reference genome . The analysis of motif enrichment alignments was performed by extracting the sequences from the alignments to the linear and graph-based reference genomes . We extracted the sequences from the linear alignments by first using Bedtools version 2 . 26 . 0 to convert the alignments from BAM to BED format by running bamtools bamtobed ( default parameters ) , and then running bedtools getfasta ( default parameters ) on the BED files . We extracted sequences from the graph alignments by running graph_peak_caller vg_json_alignments_to_fasta using graphs created by graph_peak_caller create_ob_graph . We then ran Fimo with default parameters using these sequences as input . The haplotype analysis was performed by finding all alignments overlapping a peak , and finding which variants each alignment contained , then finding the haplotypes from the VCF file that contained each variant . An alignment was decided to be compatible with a haplotype if all the variants included in the alignment were present in the haplotype . A set of alignments was decided to be compatible with two haplotypes if we could find two haplotypes in the vcf such that all the alignments in the set were compatible with at least one of them . | The expression of genes is a tightly regulated process . A key regulatory mechanism is the modulation of transcription by a class of proteins called transcription factors that bind to DNA in the spatial proximity of regulated genes . Determining the binding locations of transcription factors for specific cell types and settings is thus a key step in understanding the dynamics of normal cells as well as disease states . Binding sites for a given transcription factor are typically obtained through an experimental technique called CHiP-seq , in which DNA binding locations are obtained by sequencing DNA fragments attached to the transcription factor and aligning these sequences to a reference genome . A computational technique known as peak calling is then used to separate signal from noise and predict where the protein binds . Current peak callers are based on linear reference genomes that do not contain known genetic variants from the population . They thus potentially miss cases where proteins bind to such alternative genome sequences . Recently , a new type of reference genomes based on graph representations have become popular , as they are able to also incorporate alternative genome sequences . We here present Graph Peak Caller , the first peak caller that is able to exploit such graph representations for the detection of transcription factor binding locations . Using a graph-based reference genome for Arabidopsis thaliana , we show that our peak caller can lead to better detection of transcription factor binding locations as compared to a similar existing peak caller that uses a linear reference genome representation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"biotechnology",
"engineering",
"and",
"technology",
"gene",
"regulation",
"regulatory",
"proteins",
"brassica",
"dna-binding",
"proteins",
"invertebrate",
"genomics",
"genetic",
"mapping",
"plant",
"science",
"model",
"organisms",
"genome",
"analysis",
"transcription",
"... | 2019 | Graph Peak Caller: Calling ChIP-seq peaks on graph-based reference genomes |
A prophylactic vaccine against human immunodeficiency virus ( HIV ) remains a top priority in biomedical research . Given the failure of conventional immunization protocols to confer robust protection against HIV , new and unconventional approaches may be needed to generate protective anti-HIV immunity . Here we vaccinated rhesus macaques ( RMs ) with a recombinant ( r ) DNA prime ( without any exogenous adjuvant ) , followed by a booster with rhesus monkey rhadinovirus ( RRV ) −a herpesvirus that establishes persistent infection in RMs ( Group 1 ) . Both the rDNA and rRRV vectors encoded a near-full-length simian immunodeficiency virus ( SIVnfl ) genome that assembles noninfectious SIV particles and expresses all nine SIV gene products . This rDNA/rRRV-SIVnfl vaccine regimen induced persistent anti-Env antibodies and CD8+ T-cell responses against the entire SIV proteome . Vaccine efficacy was assessed by repeated , marginal-dose , intrarectal challenges with SIVmac239 . Encouragingly , vaccinees in Group 1 acquired SIVmac239 infection at a significantly delayed rate compared to unvaccinated controls ( Group 3 ) . In an attempt to improve upon this outcome , a separate group of rDNA/rRRV-SIVnfl-vaccinated RMs ( Group 2 ) was treated with a cytotoxic T-lymphocyte antigen-4 ( CTLA-4 ) -blocking monoclonal antibody during the vaccine phase and then challenged in parallel with Groups 1 and 3 . Surprisingly , Group 2 was not significantly protected against SIVmac239 infection . In sum , SIVnfl vaccination can protect RMs against rigorous mucosal challenges with SIVmac239 , a feat that until now had only been accomplished by live-attenuated strains of SIV . Further work is needed to identify the minimal requirements for this protection and whether SIVnfl vaccine efficacy can be improved by means other than anti-CTLA-4 adjuvant therapy .
Human immunodeficiency virus ( HIV ) continues to infect thousands of new people every day , despite advances in prevention modalities and antiretroviral therapy coverage [1] . Mathematical models have suggested that combining current HIV prevention and treatment strategies with a prophylactic HIV vaccine could significantly restrict the growth of the HIV pandemic [2] . Unfortunately , however , developing such a vaccine has been exceedingly difficult , as seen by the failure of most HIV vaccine trials conducted to date [3–7] . Although the RV144 trial remains the only report of vaccine-mediated reduction in HIV infection rates [8] , the observed results were modest , short-lived , and continue to be questioned [9 , 10] . Given the refractoriness of HIV to immune responses induced by conventional immunization protocols , new or unorthodox approaches may be needed to generate protective anti-HIV immunity . One non-conventional strategy that holds great promise is the use of live herpesviruses to deliver HIV antigens . Because herpesviruses establish persistent infections that remain largely subclinical in their hosts , a herpesvirus-based HIV vaccine could promote chronic low-level exposure to HIV antigens , a feature that might facilitate the induction of long-term protective anti-HIV immunity . We have recently generated recombinant ( r ) forms of the gamma-herpesvirus rhesus monkey rhadinovirus ( RRV ) containing a near-full-length simian immunodeficiency virus ( SIVnfl ) genome [11] . SIVnfl expresses all nine SIV gene products and assembles noninfectious SIV particles . Importantly , rhesus macaques ( RMs ) inoculated with a rRRV-SIVnfl vector became persistently infected , mounted durable anti-Env antibody ( Ab ) responses , and developed effector-differentiated SIV-specific CD8+ T-cells [11] . Of note , rRRV-SIVnfl vaccination induced CD8+ T-cell responses against all nine SIV proteins , consistent with the ability of the SIVnfl insert to express the entire SIVnfl proteome [11] . Crucially , these features are also observed following inoculation with live-attenuated SIV strains , the most effective vaccine modality against pathogenic SIV challenge in nonhuman primates . Despite these encouraging results , the magnitude of SIV-specific immune responses induced by rRRV-SIVnfl vaccination was low compared to what is seen after live-attenuated SIV inoculation . This prompted us to test whether booster immunizations with SIVnfl-expressing rDNA plasmids [delivered by intramuscular ( IM ) electroporation ( EP ) ] could amplify SIV-specific immune responses in rRRV-SIVnfl-primed RMs . We evaluated this possibility in a recent study , where rRRV-SIVnfl-primed RMs received a series of four rDNA-SIVnfl boosters given three weeks apart . The first two rDNA-SIVnfl boosters were administered without any exogenous adjuvants . However , in an attempt to augment vaccine immunogenicity , the third and fourth rDNA-SIVnfl boosters were followed by an infusion of the monoclonal ( m ) Ab Ipilimumab ( Ipi ) , which blocks the immune checkpoint receptor cytotoxic T-lymphocyte antigen-4 ( CTLA-4 ) [12] . CTLA-4 is upregulated on T-cells shortly after activation and suppresses immune responses by interfering with CD28-mediated signaling , a key step for T-cell activation [13] . Consequently , blocking CTLA-4 in vivo can enhance adaptive immune responses . Indeed , the ability of anti-CTLA-4 therapy to enhance anti-tumor immunosurveillance has led to the approval of Ipi as a drug against advanced melanoma [12 , 14] . Of note , previous studies have shown that anti-CTLA-4 can also amplify adaptive immune responses induced by prophylactic vaccination [15–17] . The aforementioned rDNA-SIVnfl booster vaccinations significantly expanded SIV-specific T-cell responses and Env-specific Ab responses in the rRRV-SIVnfl-primed RMs . Importantly , this rRRV/rDNA-SIVnfl vaccine regimen afforded significant protection against repeated , marginal-dose , intravenous ( IV ) challenges with SIVmac239 [18] . However , because that study lacked an Ipi-untreated vaccine arm , we could not delineate the contribution of the Ipi infusions to the outcome of the rRRV/rDNA-SIVnfl vaccine trial . Here we sought to expand upon our recent findings by assessing if SIVnfl vaccination can protect RMs against rectal challenges with SIVmac239 . Because DNA vaccines are typically used to prime immune responses in mixed-modality immunization protocols , we used the same SIVnfl-expressing vectors described above in a rDNA-prime/rRRV-boost configuration . Given that our previous study left some unanswered questions regarding the role of CTLA-4 blockade on vaccine performance , we characterized the immunogenicity and efficacy of the rDNA/rRRV-SIVnfl vaccine regimen in the absence ( Group 1 ) or presence ( Group 2 ) of anti-CTLA-4 therapy during the priming phase .
The sixteen RMs in Groups 1 and 2 were each primed with two rDNA-SIVnfl plasmids ( administered by IM EP ) that differed only in the Env protein expressed by each plasmid ( S1A and S1B Fig ) . The SIVnfl insert in vector 1 expressed a truncated version of SIVmac239 Env ( E767Stop ) intended to increase Env incorporation into the non-infectious SIVnfl virions ( S1A Fig ) [19] , while the SIVnfl insert present in vector 2 expressed an intact SIVmac316 Env protein ( S1B Fig ) . SIVmac316 is a neutralization-sensitive derivative of SIVmac239 [20] . The Env proteins of these two SIV clones differ from each other in only eight amino acids [20] . Given the poor immunogenicity of SIVmac239 Env ( closed conformation ) , the use of SIVmac316 Env ( open conformation ) was intended to expose potential neutralizing epitopes that would normally be occluded in the SIVmac239 Env spike . Furthermore , both vectors 1 and 2 contained a 6-base pair ( bp ) deletion in nef ( S1A and S1B Fig ) , corresponding to amino acids 239–240 , that abrogates Nef-mediated major histocompatibility complex class-I ( MHC-I ) down-regulation [21] . Additionally , the tat gene in vectors 1 and 2 encoded a L35Q substitution within the Mamu-A*01-restricted Tat28-35SL8 epitope that severely reduces its affinity for the Mamu-A*01 molecule ( S1A and S1B Fig ) [22] . Following SIVmac239 infection , Mamu-A*01+ RMs tend to mount a vigorous Tat28-35SL8-specific CD8+ T-cell response that is rapidly rendered ineffective by the selection of SIV “escape” variants [22] . This selection occurs with no apparent fitness cost to the virus [22 , 23] . Because subdominant CD8+ T cell responses can be actively suppressed by dominant CD8+ T cell responses in the context of DNA immunization [24] , the rationale for the Tat L35Q change in the rDNA-SIVnfl vectors was to prevent the priming of Tat28-35SL8-specific CD8+ T-cells . Our hope was that this strategy would broaden the repertoire of vaccine-induced SIV-specific CD8+ T cells in the Mamu-A*01+ RMs used in the present study . All sixteen RMs were primed four times with vectors 1 and 2 . Those in Group 1 ( n = 8 ) did not receive any exogenous adjuvant during the rDNA-SIVnfl immunizations , whereas those in Group 2 ( n = 8 ) were infused with Ipi ( 3 . 0 mg/kg of body weight ) on the day after each rDNA-SIVnfl prime ( Fig 1 ) . This rDNA-SIVnfl+Ipi priming schedule was intended to match the clinically-approved Ipi therapy regimen , which consists of four infusions of 3 . 0 mg/kg given every 3 weeks [12] . Six weeks after the 4th priming immunization , animals in both groups were boosted with a mixture of five rRRV constructs referred to as the “rRRV pentamix” ( vectors 3–7; S1C–S1G Fig; Fig 1 ) . Three of these rRRV vectors expressed SIVnfl ( vectors 3–5 ) , albeit under the control of different promoters ( S1C–S1E Fig ) . The SIVnfl inserts present in vectors 3–5 lacked the tat and nef modifications described above ( S1C–S1E Fig ) . In an attempt to augment vaccine-induced anti-Env Ab responses , the rRRV pentamix also included two constructs that expressed SIV env alone ( S1F and S1G Fig ) . Vector 6 encoded SIVmac239 env and vector 7 encoded SIVmac316 env ( S1F and S1G Fig ) . Both SIV env inserts contained the aforementioned E767Stop truncation ( S1F and S1G Fig ) . Because Ipi therapy in humans is associated with immune-related adverse events [25] , the safety of the rDNA-SIVnfl+Ipi immunizations was evaluated in two Group 2 RMs ( Group 2a ) first , before the remaining six Group 2 vaccinees ( Group 2b ) received their rDNA-SIVnfl+Ipi doses ( Fig 1 ) . For the sake of comparison , the RMs in Group 1 were also subdivided in Groups 1a ( n = 2 ) and 1b ( n = 6 ) and vaccinated with rDNA-SIVnfl ( without any exogenous adjuvants ) in parallel with their Group 2a and Group 2b counterparts . ( Fig 1 ) . The rDNA-SIVnfl+Ipi immunizations were well tolerated by the Group 2a vaccinees; neither animal experienced adverse events commonly associated with Ipi therapy ( e . g . , skin rashes and diarrhea ) during the rDNA-SIVnfl+Ipi priming phase . The pilot Group 1a/2a vaccinations also enabled us to track vaccine-induced SIV-specific CD8+ T-cells in peripheral blood mononuclear cells ( PBMCs ) by fluorochrome-labeled MHC-I tetramer staining , as the four animals in those groups expressed the MHC-I allele Mamu-A*01 . Five Mamu-A*01-restricted SIV epitopes were evaluated: Vif100-109VL10 , Env620-628TL9 , Env233-241CL9 , Tat28-35SL8 , and Gag181-189CM9 . Vaccine-elicited CD8+ T-cells directed against the first two epitopes were largely undetectable in both groups ( S2A and S2B Fig ) . Env233-241CL9-specific CD8+ T-cells reached modest frequencies in the Group 2a monkey r13053 , but were present at low levels in the other three vaccinees ( S2C Fig ) . Tat28-35SL8-specific CD8+ T-cells remained at baseline levels during the rDNA-SIVnfl immunizations but then expanded in two animals after the first rRRV-SIVnfl boost ( S2D Fig ) , consistent with this epitope being inactivated in the rDNA-SIVnfl constructs but intact in the rRRV pentamix . Vaccine-induced CD8+ T-cells against Gag181-189CM9 were detected in all four animals , especially those in Group 2a ( Fig 2A ) . Indeed , up to 16% of peripheral CD8+ T-cells in Group 2a stained positive for the Mamu-A*01/Gag181-189CM9 tetramer two weeks after the 4th DNA prime , compared to only 3 . 6% in Group 1a ( Fig 2C ) . Boosting with rRRV-SIVnfl at week 15 further expanded these Gag-specific CD8+ T-cells , especially in the Group 2a vaccinee r13053 ( Fig 2A and 2C ) . However , a second rRRV-SIVnfl boost given at week 25 had little or no effect on the frequencies of CD8+ T-cells targeting Gag181-189CM9 and the other aforementioned epitopes ( Fig 2A and 2C; S2 Fig ) , probably as a result of anti-RRV immunity generated by the first rRRV-SIVnfl vaccination . The frequencies of vaccine-elicited Gag181-189CM9-specific CD8+ T-cells contracted over time in both groups but were still detectable at study week 53 , the last time point before the challenge phase ( Fig 2A and 2C ) . A memory phenotype analysis conducted at this time point revealed that the fraction of Mamu-A*01/Gag181-189CM9 tetramer+ CD8+ T-cells displaying a fully-differentiated effector memory T-cell ( TEM2 ) signature ( CD28−CCR7− ) was 40–80% in Group 1a and 91–98% in Group 2a ( Fig 2B and 2D ) . The durability of these vaccine-induced CD8+ T-cell responses and their TEM-biased phenotype are consistent with the ability of rRRV-SIVnfl vectors to provide chronic low-level exposure to SIV antigens . Since the Ipi infusions were well tolerated by the Group 2a vaccinees , we proceeded with the immunization schedule of the Group 1b and Group 2b animals and delivered their first rDNA-SIVnfl or rDNA-SIVnfl+Ipi prime at study week 12 ( Fig 1 ) . Similar to the Group 1a/2a immunization schedule , the first three rDNA-SIVnfl primes were given at 3-week intervals ( Fig 1 ) . However , the 4th rDNA-SIVnfl prime had to be delayed until week 24 because of an error handling the rDNA-SIVnfl plasmids ( Fig 1 ) . Again , none of the animals in Group 2b experienced Ipi-associated irAEs during the rDNA-SIVnfl+Ipi immunizations . Six weeks after the 4th rDNA-SIVnfl prime , all vaccinees in Groups 1b and 2b were boosted with the rRRV pentamix ( Fig 1 ) . Since the 2nd rRRV-SIVnfl administration resulted in little or no boosting of SIV-specific CD8+ T-cells in Groups 1a and 2a , we chose not to give the Group 1b and Group 2b vaccinees a second dose of the rRRV-SIVnfl pentamix ( Fig 1 ) . Even though the immunization schedules of the Groups 1a/2a and Groups 1b/2b monkeys were offset by 12 weeks , all vaccinated animals were challenged with SIVmac239 around the same time , that is , week 53 for the vaccinees in Groups 1b and 2b , and week 54 for those in Groups 1a and 2a ( Fig 1 ) . We used ELISA to monitor the plasma concentrations of Ipi in the Group 2b vaccinees after each rDNA-SIVnfl+Ipi immunization . Ipi was readily detectable after the first rDNA-SIVnfl+Ipi immunization and its concentrations shot up after each subsequent infusion ( Fig 3 ) . Following the fourth rDNA-SIVnfl+Ipi immunization , an increase in plasma Ipi concentrations was observed in all Group 2b RMs , except for r14070 ( Fig 3 ) . The lack of detectable Ipi in r14070 was likely due to the robust anti-drug Ab ( ADA ) response developed by this animal . Indeed , the endpoint titer of anti-Ipi Abs in r14070 at study week 28 was 1:2 , 560 ( Fig 3 ) . By comparison , contemporaneous endpoint titers of ADAs were 1:20 in r14123 and less than 1:10 in the remaining Group 2b vaccinees ( Fig 3 ) . These responses were also detected in the Group 2a vaccinee r13053 , whose endpoint titer of ADAs on day 14 post 4th rDNA-SIVnfl+Ipi prime was 1:40 . The Ipi infusions were also associated with transient increases in the absolute counts of several peripheral lymphocyte subsets , including CD3+ T-cells , CD4+ T-cells , CD25+ FoxP3+ T regulatory cells ( Tregs ) , CD8+ T-cells , and CD20+ B-cells ( S3 Fig ) . Compared to Group 1b , the peripheral numbers of all these lymphocyte subsets ( except for Tregs ) were significantly elevated in Group 2b in at least one time point following the third rDNA-SIVnfl+Ipi priming immunization ( S3 Fig ) . We carried out intracellular cytokine staining ( ICS ) at multiple time points during the vaccine phase in order to characterize vaccine-induced SIV-specific T-cell responses in Groups 1b and 2b . The stimuli for these assays consisted of pools of peptides ( 15mers overlapping by 11 amino acids ) spanning the entire SIVmac239 proteome . Although Gag , Env , and Nef were the primary targets of vaccine-induced CD8+ T-cells in Groups 1b and 2b ( Fig 4 ) , CD8+ T-cell responses against Pol , Vif , Vpx , Vpr , Tat , and Rev were also detected ( Fig 5 ) , consistent with the ability of SIVnfl to express the entire SIV proteome . Except for a modest , but statistically significant , rise in Pol-specific CD8+ T-cells in Group 1b versus Group 2b at week 53 ( Fig 5A ) , vaccinees in these groups developed similar levels of CD8+ T-cell responses against each SIV protein ( Figs 4 and 5 ) . CD4+ T-cell responses followed a similar pattern but were either undetectable for some of the proteins or present at much lower frequencies ( S4 Fig; S5 Fig ) . Gag-specific CD4+ T-cells were an exception , as these responses were significantly higher in Group 2b than in Group 1b at several time points during the vaccine phase ( S4A Fig ) . Of note , we could not detect vaccine-elicited Gag-specific T-cells in disaggregated lymphocyte suspensions from pooled colon and rectal biopsies obtained from the Group 1b and Group 2b animals at week 3 post rRRV-SIVnfl boost . We also assessed the breadth and magnitude of vaccine-elicited SIV-specific CD8+ T-cell responses at week 53 , the last time point before the SIV challenge phase . The vaccinees in Groups 1a and 2a were also included in this analysis since they were challenged with SIVmac239 around the same time as their Group 1b and Group 2b counterparts . There was considerable animal-to-animal variability in the magnitude of vaccine-induced CD8+ T-cell responses targeting each protein ( Fig 6A and 6B ) . In contrast to a previous study reporting that Ipi therapy broadens the repertoire of melanoma-reactive CD8+ T-cells in cancer patients [26] , the rDNA-SIVnfl+Ipi immunizations appeared to restrict the breadth of vaccine-induced SIV-specific CD8+ T-cells in Group 2 ( Fig 6C ) . Indeed , vaccine-elicited CD8+ T-cells in Group 2 recognized significantly fewer SIV proteins than those in Group 1 ( P = 0 . 039; Fig 6C ) . Notwithstanding this difference , the sum of vaccine-elicited CD8+ T-cell responses against all nine SIV proteins was not significantly different between Groups 1 and 2 ( Fig 6D ) . No significant sex differences were observed in the breadth or magnitude of vaccine-induced SIV-specific CD8+ T-cell responses within Group 1 ( 1a+1b ) and Group 2 ( 2a+2b ) ( S6A and S6B Fig ) . To directly assess the impact of priming anti-Env Ab responses in the context of CTLA-4 blockade , we tracked the expansion of gp140-specific B-cells in PBMC from vaccinees in Groups 1 and 2 following the 4th rDNA-SIVnfl prime . For this analysis , freshly-sorted plasmablasts were used in IgG enzyme-linked immune absorbed spot ( ELISpot ) assays for the detection of cells producing gp140-specific IgG . These Env-specific plasmablasts were present at low frequencies in Groups 1b and 2b on the day of the 4th DNA prime , but underwent considerable expansion in the ensuing days , particularly in the Group 2b animals ( S7A Fig ) . A similar pattern of Env-specific plasmablast expansion was observed in Groups 1a and 2a ( S7B Fig ) . A comparison of gp140-specific plasmablast frequencies on day 7 post 4th DNA prime revealed significantly higher frequencies of these cells in Group 2b than in Group 1b ( S7C Fig ) , as well as in Group 2a than in Group 1a ( S7D Fig ) . A time-course analysis of vaccine-induced Env-binding Abs in Groups 1b and 2b showed that anti-gp140 Abs appeared with faster kinetics and reached greater levels in plasma than did anti-gp120 Abs ( Fig 7A and 7B ) . These responses followed a similar pattern in Groups 1a and 2a ( Fig 7C and 7D ) . At the time of SIV challenge , vaccinees in Group 1 ( 1a+1b ) and Group 2 ( 2a+2b ) had significantly higher titers of IgG Abs capable of binding gp140 than gp120 ( Fig 7E ) , suggesting that gp41 was heavily targeted by vaccine-induced Env-specific Abs . Consistent with the plasmablast ELISpot results , this analysis also revealed that vaccinees in Group 2 ( 2a+2b ) had greater titers of gp140-binding Abs than their Group 1 ( 1a+1b ) counterparts ( Fig 7F ) . However , the levels of gp120-binding Abs were not significantly different between the two groups ( Fig 7F ) . Vaccine-induced gp140-binding Abs were also detected in rectal secretions at study week 48 , that is , 5–6 weeks prior to the first SIV challenge ( S8 Fig ) . There was no significant difference in the levels of these rectal anti-gp140 Abs between Group 1 ( 1a+1b ) and Group 2 ( 2a+2b ) ( S8 Fig ) . Next , we characterized antiviral functions of vaccine-induced anti-Env Abs at the time of the first SIV challenge . While little or no serological neutralizing activity against SIVmac239 was detected , the 50% inhibitory dilution ( ID50 ) titers of anti-SIVmac316 neutralizing ( n ) Abs in Group 1 and Group 2 ranged from 1 , 011 to 12 , 811 ( Fig 8A ) . As a reference , these titers were 2 . 5- to 32-fold lower than the mean titer of anti-SIVmac316 nAbs present in sera from monkeys that had been infected with SIVmac239Δnef for 23 weeks ( Fig 8A ) . There was no significant difference in the ID50 titers of anti-SIVmac316 nAbs between Group 1 ( 1a+1b ) and Group 2 ( 2a+2b ) ( Fig 8A ) . We also evaluated the ability of contemporaneous plasma samples to direct natural killer ( NK ) cell-mediated , Ab-dependent cellular cytotoxicity ( ADCC ) against SIVmac239-infected cells . ADCC activity was calculated based on the ability of serially diluted plasma from each animal to kill target cells infected with SIVmac239 or SHIV-AD8 ( internal control ) . The extent that serially diluted plasma resulted in killing of target cells infected with each virus allowed us to calculate the area under the curve ( AUC ) for each condition . The area under the SHIV-AD8 curve was then subtracted from the area under the SIVmac239 curve , yielding a relative ( r ) AUC value for each animal . Based on these rAUC values , ADCC was detected in all samples analyzed and , in some cases , at levels that approximated those seen in pooled plasma from SIVmac239-infected RMs ( Fig 8B ) . Similar to the anti-SIVmac316 nAb titers , there was no significant difference in ADCC activity between Group 1 ( 1a+1b ) and Group 2 ( 2a+2b ) ( Fig 8B ) . Furthermore , no significant sex differences were observed within Group 1 ( 1a+1b ) and Group 2 ( 2a+2b ) in midpoint titers of Env-binding Abs , anti-SIVmac316 nAb titers , or ADCC activity ( S6C–S6F Fig ) . Vaccine efficacy was assessed by subjecting all vaccinees in Group 1 ( 1a+1b ) and Group 2 ( 2a+2b ) , as well as eight unvaccinated control RMs ( Group 3 ) , to repeated IR challenges with SIVmac239 . Animals were exposed to a marginal dose ( 200 TCID50 ) of SIVmac239 every two weeks . Viral loads ( VLs ) determined on plasma samples collected on days 6 and 10 after each SIVmac239 exposure were used to determine which monkeys were re-challenged on day 14 . Only animals that remained aviremic on days 6 and 10 after each SIV challenge were re-exposed to SIV; those with positive VLs in either sample were considered to be infected and continued to be bled at regular time points until week 12 post infection ( PI ) . Following six IR challenges with SIVmac239 , all eight control animals in Group 3 became infected , compared to only three vaccinees in Group 1 and five vaccinees in Group 2 . This difference was statistically significant for Group 1 ( P = 0 . 044 , Fisher’s exact test ) but not for Group 2 ( P = 0 . 352 , Fisher’s exact test ) . The rate of SIV acquisition in Group 1 was also significantly slower than that in Group 3 ( P = 0 . 028 , Cox proportional hazard model; Fig 9A ) , indicating that the rDNA/rRRV-SIVnfl vaccine regimen afforded significant protection against IR challenge with SIVmac239 . Specifically , vaccination with the rDNA/rRRV-SIVnfl regimen reduced the per-exposure probability of SIV infection by 78% in Group 1 . In contrast , the rDNA+Ipi/rRRV-SIVnfl vaccine regimen did not confer significant protection against SIVmac239 acquisition , as the pace of SIV infection in Group 2 was not significantly different than that in Group 3 ( P = 0 . 199 , Cox proportional hazard model; Fig 9B ) . The per-exposure probability of infection in Group 2 was reduced by 52% but it was not significantly different than that in the control group . Since there was no significant difference in infection rates between Groups 1 and 2 ( P = 0 . 506 , Cox proportional hazard model ) , we asked whether a vaccine-mediated effect on SIV acquisition could still be detected if all 16 vaccinees in Groups 1 and 2 were grouped together and compared to Group 3 . This analysis revealed that the pace of SIV infection in all 16 vaccinees ( Groups 1 + 2 ) was still significantly slower than that of the control group ( P = 0 . 031 , Cox proportional hazard model; Fig 9C ) . In order to validate these findings , we compared the rates of SIV acquisition in Group 1 , Group 2 , and Groups 1 + 2 to that of an expanded control group consisting of the 8 contemporaneous RMs in Group 3 plus 52 historical control RMs ( i . e . , 60 control animals total ) . These historical RMs were part of previous experiments conducted by our group and were subjected to the same IR challenge regimen described above , utilizing the same dose of the same stock of SIVmac239 used here . This analysis confirmed the findings described above , with partial ( albeit statistically significant ) vaccine-mediated protection being detected in Group 1 ( P = 0 . 043 , Cox proportional hazard model; Fig 9D ) and Group 1 + 2 ( P = 0 . 038 , Cox proportional hazard model; Fig 9F ) , but not in Group 2 ( P = 0 . 356 , Cox proportional hazard model; Fig 9E ) . Of note , immunological measurements performed in the eight vaccinees that remained uninfected two weeks after the 6th SIV exposure revealed no significant enhancement in vaccine-induced SIV-specific immune responses compared to the levels measured just prior to the first SIV challenge ( S9 Fig ) , thereby confirming their protected status . Vaccine-mediated reductions in viremia were evident by visual inspection of the VL profiles of control and vaccinated RMs ( Fig 9G–9I ) . While there was no difference in peak and setpoint VLs between the SIV-infected vaccinees in Groups 1 and 2 , their VLs were significantly lower than those in the Group 3 controls ( Fig 9J and 9K ) . Lastly , we searched for immune attributes that could distinguish the vaccinees that were protected from SIVmac239 infection versus those that were not . We selected 7 pre-challenge attributes of vaccine-induced immune responses for this analysis: i ) the total breadth of SIV-specific CD8+ T-cell responses; ii ) the total magnitude of SIV-specific CD8+ T-cell responses; iii ) the serum midpoint titers of gp140- and ( iv ) gp120-binding IgG Abs; ( v ) the serum titers of anti-SIVmac316 nAbs; ( vi ) the levels of ADCC activity against SIVmac239-infected cells ( rAUC values ) ; and the ( vii ) rectal levels of gp140-binding IgG Abs . All vaccinees in Groups 1 and 2 were included in this analysis . None of these immunological variables were associated with resistance to SIVmac239 infection ( Table 1 ) . We also searched for correlations between these immunological variables and either peak or setpoint VLs in the vaccinees that acquired SIV infection , but again found no significant association ( Table 2 ) . Since one of the hallmarks of anti-CTLA-4 therapy in humans is the expansion of effector-like CD4+ T-cells [27] , we also investigated whether the Ipi infusions could have undermined the efficacy of the Group 2 vaccine regimen by increasing the availability of SIV target cells prior to the first SIVmac239 challenge . To do that , we compared the pre-challenge frequencies of peripheral CD4+ T-cells expressing ki-67 and/or CCR5 between Groups 1 and 2 ( Fig 10A ) . This analysis revealed that the proportions of ki-67+CCR5− ( Fig 10B ) , ki-67+CCR5+ ( Fig 10C ) , and ki-67−CCR5+ ( Fig 10D ) CD4+ T-cell subsets in PBMC were not significantly different between Groups 1 and 2 , suggesting that increased SIV target cell availability did not underlie the failure of the Group 2 vaccine regimen to protect against SIV infection . Thus , the immunological basis by which the rDNA/rRRV-SIVnfl vaccine regimen protected a subset of Group 1 animals against rectal acquisition of SIVmac239 remains unclear .
Experimental challenge of RMs with SIV provides a valuable system for identifying immune correlates of protection and selecting the most promising vaccine approaches for human testing [28] . Because a successful outcome in this model can be used to justify costly clinical trials , it is critical that the efficacy of pre-clinical AIDS vaccine concepts be assessed against stringent challenge viruses . The SIVmac239 molecular clone is well suited for this purpose . Consistent with its neutralization resistance and high replicative capacity in Indian-origin RMs , vaccine protection against SIVmac239 acquisition is exceedingly difficult to achieve , even when there is a complete match between vaccine-encoded sequences and the challenge virus . Several vaccine trials have reported significant reductions in post-acquisition SIVmac239 viremia in RMs but no protection from infection [29–34] . Even the rhesus cytomegalovirus-based vaccine platform developed by Picker and colleagues , which results in profound control and eventual clearance of SIVmac239 infection in half of vaccinees [35–37] , does not block acquisition of SIVmac239 . In fact , live-attenuated strains of SIVmac239 ( e . g . , SIVmac239Δnef ) remain the only vaccine modality to consistently afford significant levels of apparent sterilizing immunity against SIVmac239 challenge in Indian-origin RMs [38–40] . Given the stringency of this challenge model , it is remarkable that following repeated IR challenges with SIVmac239 , the Group 1 ( rDNA/rRRV-SIVnfl ) vaccine regimen significantly decreased the per-exposure probability of infection by 78% . This result is reinforced by our recent demonstration that delivery of the same vectors in reverse order ( i . e . , rRRV-SIVnfl prime/rDNA-SIVnfl boost ) also significantly reduced the per-exposure probability of IV SIVmac239 infection by 79% in RMs [18] . While the mechanisms of this protection remain unknown , we posit that the use of SIVnfl to prime and , in the case of the rRRV-SIVnfl vectors , continually boost SIV-specific immune responses facilitated the induction of protective anti-SIVmac239 immunity . From the standpoint of HIV vaccination , immunization with near full-length ( nfl ) viral genomes has both practical and biological advantages . For example , delivering the entire HIV proteome in a single rDNA plasmid or viral vector , as opposed to different genes in separate constructs , would likely reduce manufacturing costs and face fewer regulatory barriers down the translational pipeline . Additionally , since B-cells tend to respond with high avidity to virus-sized antigens [41] , a properly adjuvanted HIVnfl-based vaccine could be expected to elicit high titers of anti-Env Ab responses , especially if this vaccine contains modifications to stabilize Env trimers and increase their surface expression [42] . Furthermore , based on the ability of the SIVnfl insert to induce CD8+ T-cell responses against the entire SIV proteome , an HIVnfl-based vaccine could also be useful for generating broadly-targeted HIV-specific CD8+ T-cell responses , thereby providing an extra line of defense in the event of HIV transmission . We also evaluated whether treating RMs with the CTLA-4-blocking mAb Ipi during antigen priming could enhance the SIV-specific immune response induced by the rDNA/rRRV-SIVnfl vaccine regimen . In contrast to the powerful immunological effects of anti-CTLA-4 therapy in humans [12] , the impact of the rDNA-SIVnfl+Ipi priming immunizations on vaccine immunogenicity was modest and , at least in one case , detrimental . Although the Ipi infusions likely contributed to the moderately elevated CD4+ T-cell responses and Env-binding Abs detected in Group 2 , the anti-CTLA-4 therapy was also associated with a narrower repertoire of vaccine-induced SIV-specific CD8+ T-cell responses in Group 2 compared to Group 1 . We propose two explanations for this worse-than-expected adjuvant activity of Ipi therapy . The first one relates to the dose of Ipi . The Group 2 vaccinees received a series of four infusions of 3 . 0 mg/kg/dose of Ipi , a regimen that was clinically approved based on its ability to enhance anti-tumor immunosurveillance in patients with advanced melanoma [12] . This dose might not , however , be sufficient to amplify pathogen-specific immune responses induced by prophylactic vaccination . While previous studies have reported increased vaccine- or infection-induced SIV-specific T-cell responses in nonhuman primates treated with 10 . 0 mg/kg/dose of Ipi [15 , 16 , 43] , rashes or death were reported in two of those studies [16 , 44] , suggesting that any gains in vaccine immunogenicity achieved by raising the dose of Ipi might be counterbalanced by an increased incidence of immune-related adverse events . The second possibility is that ADAs triggered by the Ipi infusions interfered with the ability of this mAb to block CTLA-4 in vivo . Consistent with this idea , ADAs were detected in 3/8 Group 2 vaccinees ( ~38% ) and appeared to accelerate the clearance of Ipi in r14070 , the monkey with the highest ADA response . The fact that clinical grade Ipi ( a human IgG1 molecule ) was used in this study likely favored the induction of ADAs in Group 2 . It is noteworthy , however , that Ipi-treated cancer patients also develop ADAs at a rate ( 26% ) that is not too distant than the one observed here [45] . Thus , ADAs might compromise the efficacy of mAb-based therapies even when there is complete species match between the mAb and its recipients . Although CTLA-4 is known to regulate the primary immune response to antigen exposure , there is still no consensus on whether Ipi therapy enhances anti-tumor immunosurveillance by boosting pre-existing neoantigen-specific CD8+ T-cells or priming CD8+ T-cells against new tumor epitopes [26 , 46 , 47] . We actually used rDNA-SIVnfl+Ipi to boost SIV-specific immune responses in the rRRV-SIVnfl-primed RMs in our recent SIV vaccine trial [18] . Those animals mounted robust anamnestic SIV-specific cellular and humoral immune responses following the rDNA-SIVnfl+Ipi boosters and were ultimately protected against IV challenges with SIVmac239 . However , because of the lack of a group of Ipi-untreated , rRRV-SIVnfl/rDNA-SIVnfl-vaccinated RMs in that study , the precise contribution of anti-CTLA-4 therapy to the efficacy of the rRRV-SIVnfl/rDNA-SIVnfl vaccine regimen is unknown . Since CTLA-4 has been shown to have distinct roles in the regulation of naïve and memory T-cells [48–51] , it is possible that anti-CTLA-4 therapy might have had greater adjuvant activity had it been co-administered with rDNA-SIVnfl booster immunizations . In sum , here we show that a rDNA-SIVnfl/rRRV-SIVnfl vaccine regimen provided significant , albeit partial , protection against rectal acquisition of SIVmac239 in RMs . No protection was observed in a group of RMs vaccinated in parallel with the same immunization protocol in which infusions of the CTLA-4-blocking mAb Ipi were given after each rDNA-SIVnfl immunization . Future studies should focus on the identification of the immune responses that are critical for the antiviral properties of SIVnfl vaccination and whether SIVnfl vaccine efficacy can be improved by methods other than anti-CTLA-4 adjuvant therapy .
Twenty-four RMs ( Macaca mulatta ) of Indian origin were used in this study . All animals were housed at the Wisconsin National Primate Research Center ( WNPRC ) and cared for under a protocol approved by the University of Wisconsin Graduate School Animal Care and Use Committee ( animal welfare assurance no . A3368-01; protocol no . G005563 ) . The macaques in this study were managed according to the animal husbandry program of the WNPRC , the guidelines of the Weatherall Report and the principles described in the National Research Council’s Guide for the Care and Use of Laboratory Animals . The animal husbandry program of the WNPRC aims at providing consistent and excellent care to nonhuman primates at the center . This program is employed by the Colony Management Unit and is based on the laws , regulations , and guidelines promulgated by the United States Department of Agriculture ( e . g . , the Animal Welfare Act and its regulations , and the Animal Care Policy Manual ) , Institute for Laboratory Animal Research ( e . g . , Guide for the Care and Use of Laboratory Animals , 8th edition ) , Public Health Service , National Research Council , Centers for Disease Control , and the Association for Assessment and Accreditation of Laboratory Animal Care International . The nutritional plan utilized by the WNPRC is based on recommendations published by the National Research Council . Specifically , macaques were fed twice daily with 2050 Teklad Global 20% Protein Primate Diet and food intake was closely monitored by Animal Research Technicians . This diet was also supplemented with a variety of fruits , vegetables , and other edible objects as part of the environmental enrichment program established by the Behavioral Management Unit . Paired/grouped animals exhibiting stereotypical and/or incompatible behaviors were reported to the Behavioral Management staff and managed accordingly . All primary enclosures ( i . e . , stationary cages , mobile racks , and pens ) and animal rooms were cleaned daily with water and sanitized at least once every two weeks . Lights were on a 12:12 diurnal schedule . Vaccinations were performed under anesthesia ( Ketamine administered at 5–12 mg/kg depending on the animal ) and all efforts were made to minimize suffering . Euthanasia was performed whenever an animal experienced conditions deemed distressful by one of the veterinarians at the WNPRC . All euthanasia were performed in accordance with the recommendations of the Panel on Euthanasia of the American Veterinary Medical Association and consisted of an IV overdose ( greater than or equal to 50 mg/kg or to effect ) of sodium pentobarbital or equivalent , as approved by a clinical veterinarian , preceded by ketamine ( at least 15 mg/kg body weight ) given by the IM route . The MHC-I genotype , sex , and age of each monkey at the beginning of the SIV challenge phase are shown in Table 3 . All animals in Groups 1 and 2 were RRV seronegative in the beginning of the vaccine phase . The RMs in Groups 1 and 2 were primed four times with vectors 1 and 2 ( S1 Fig ) , which utilized the pCMVkan backbone . These rDNA vectors contained SIVnfl inserts that differed in the env genes they encoded ( S1 Fig ) . Each animal was vaccinated with 2 . 0 mg of each rDNA-SIVnfl vector per occasion ( total: 4 . 0 mg per occasion ) . This dose was split into two separate syringes , so that a total of four syringes , each containing 1 . 0 mg of rDNA-SIVnfl , were used to deliver vectors 1 and 2 in each immunization . PBS was used to adjust the volume of each syringe to 0 . 5 ml . These rDNA-SIVnfl formulations were administered intramuscularly by the TriGrid in vivo electroporation system ( Ichor Medical Systems , Inc . , San Diego , CA ) . Muscles in the thighs and forearms were used for these vaccinations . On the day after each rDNA-SIVnfl immunization , the Group 1 vaccinees were treated with 3 . 0 mg/kg of clinical grade Ipilimumab ( Bristol-Myers Squibb ) . After the second rDNA-SIVnfl immunization , the Ipi infusion that was supposed to be given to r14129 ( Group 2a ) was , instead , given to r14019 ( Group 1a ) by mistake . Six weeks after the 4th rDNA-SIVnfl prime , all animals in Groups 1 and 2 were boosted with a mixture of five rRRV vectors ( termed “rRRV pentamix” ) . A dose of 1 . 0×1010 genome copies of each rRRV vector was administered intravenously in 1 . 0 ml of PBS . Three of the vectors in the rRRV pentamix expressed SIVnfl , albeit under the control of different promoters . Vector 3 contained the CMV enhancer/promoter ( pCMV ) placed upstream of the SIVnfl insert ( S1 Fig ) . In vector 4 , a hybrid early/late promoter construct consisting of the late promoter for RRV ORF26 ( p26 ) and the early promoter for the RRV Poly Adenylated Nuclear RNA ( PAN ) was inserted just upstream of the SIVnfl insert ( S1 Fig ) . This artificial promoter was termed pDUAL ( S1 Fig ) . In vector 5 , the SIV promoter/enhancer region was used by restoring nucleotides 1–521 of the 5’ long terminal repeat ( pLTR ) that had been deleted in the original SIVnfl insert ( S1 Fig ) . We used this combination of three promoters because the extent to which one may be better than another in terms of the ability to express in vivo in monkeys , the ability to persist , and the magnitude , quality , and persistence of resultant immune responses is unknown . We therefore hoped that these three promoters would promote stable expression of SIVnfl during all stages of the RRV life cycle . In order to maximize the generation of anti-Env Abs , the rRRV pentamix also contained two additional constructs that encoded SIV env alone . Vector 6 encoded SIVmac239 env and vector 7 encoded SIVmac316 env ( S1 Fig ) . Both vectors expressed a truncated version of SIV Env ( E767Stop ) intended to increase Env incorporation into non-infectious SIV virions [19] . The codon usage of these env inserts was modified to reflect the codon usage of RRV glycoprotein in order to allow adequate expression in the monkeys [52] . The monkeys in Groups 1a and 2a received a second dose of the rRRV pentamix at study 25 , that is , ten weeks after the first rRRV pentamix boost . However , since this second rRRV boost resulted in little or no expansion of SIV-specific immune responses , we decided to omit it in the Group 1b and Group 2b immunization schedules . At study weeks 53 or 54 ( Fig 1 ) , all vaccinated and control RMs were subjected to repeated IR inoculations of 200 TCID50 ( 4 . 8×105 vRNA copies ) of SIVmac239 . This challenge inoculum was administered in 1 . 0 ml of PBS . IR challenges occurred every two weeks . Plasma VLs were assessed six and ten days after each exposure . Once an animal experienced a positive VL at either one of these time points , it was no longer challenged . Only RMs that remained aviremic at both time points were re-challenged on day 14 . VLs were measured using 0 . 5 ml of EDTA-anticoagulated RM plasma based on a modification of a previously published [53] . Total RNA was extracted from plasma samples using QIAgen DSP virus/pathogen Midi kits , on a QIASymphonyXP laboratory automation instrument platform . Six replicate two-step RT-PCR reactions were performed per sample using a random primed reverse transcription reaction , followed by 45 cycles of PCR using the following primers and probe: forward primer: SGAG21: 5’-GTCTGCGTCAT ( dP ) TGGTGCA TTC-3’; reverse primer SGAG22: 5’-CACTAG ( dK ) TGTCTCTGCACTAT ( dP ) TGTTTTG-3’; probe: PSGAG23: 5’-FAM-CTTC ( dP ) TCAGT ( dK ) TGTTTCACTTTCTCTTCTGCG-BHQ1- 3’ . The limit of reliable quantitation on an input volume of 0 . 5 ml of plasma was 15 vRNA copies/ml . We isolated peripheral blood mononuclear cells ( PBMC ) from EDTA-treated blood by Ficoll-Paque Plus ( GE Health Sciences ) density centrifugation . Cells were subsequently washed in R10 medium [RPMI 1640 medium supplemented with GlutaMAX ( Life Technologies ) , 10% FBS , and 1% antibiotic/antimycotic] and then resuspended at various concentrations depending on the application . In order to isolate lymphocytes from rectal and colon biopsies from the same animal , these specimens were pooled and resuspended in 10 mls of RPMI 1640 medium supplemented with GlutaMAX ( Life Technologies ) containing Liberase ( Sigma-Aldrich; 40 μg/ml ) , DNase I ( Akron Biotechnology; 4 . 0 μg/ml ) , and 1% antibiotic/antimycotic . Conical tubes containing these samples were rocked at 200 rpm for 1 hour ( hr ) at 37°C on a table-top shaker . At the end of this step , the cell suspensions were passed through 70-μm cell strainers , washed once in R10 , and then used in immunological assays . Weck-cel sponges containing rectal secretions were frozen at −80°C on the day of collection . Prior to use in immunological assays , these sponges were thawed and eluted using a fresh mixture of Protease Inhibitor Cocktail ( P8340; Sigma-Aldrich ) , PBS with 0 . 25% bovine serum albumin , and Igepal ( Sigma-Aldrich ) . Sponges were first spun using Spin-X columns ( 0 . 22-μm filters ) and then concentrated with Vivaspin protein concentrator columns ( 50 kDa , 500 μl , GE Healthcare Life Sciences ) . These rectal secretions were eluted in a final volume of ~90 μl . The kinetics of anti-Env Abs during the vaccine phase were measured by a semi-quantitative ELISA . To begin , ELISA plates were coated with 100 μl of purified SIVmac239 gp140 or gp120 protein ( Immune Technology Corp . ) at a concentration of 1 . 0 μg/ml . These plates were incubated overnight at room temperature . On the following day , the plates were washed with PBS-Tween20 and wells were blocked with 300 μl of 5% powdered milk in PBS for 1 hr at 37°C . Subsequently , the plates were washed and plasma samples diluted 1:1 , 000 were added in 100 μl to the corresponding wells . After a 1-hr incubation at room temperature , the plates were washed and 100 μl of a 1:10 , 000 dilution of goat anti-human IgG HRP-conjugated detection antibody ( Southern Biotech ) was added to all wells and incubated for 1 hr at 37°C . Finally , the plates were washed before being developed with 100 μl of 3 , 3' , 5 , 5'-Tetramethylbenzidine ( EMD Millipore ) . After a short incubation , the reaction was stopped with TMB Stop Solution ( Southern Biotech ) and the plates were read ( Biotek Synergy 2 ) at 450 nm . The same ELISA setup was used to determine the midpoint titers of anti-gp140 and anti-gp120 Abs , except that the biological specimen used for this analysis was serially diluted serum collected at the time of the first SIV challenge . The midpoint titers of anti-Env Abs were determined using the “Sigmoidal , 4PL , X is log ( concentration ) ” function in Prism ( version 7 . 0e , GraphPad Software , Inc . ) . The relative levels of anti-gp140 Abs in rectal secretions were determined by coating half-well ELISA plates ( Corning Inc . ) overnight at 4°C with either SIVmac239 gp140 ( 2 . 0 μg/ml ) or purified mouse anti-human IgG ( 5 . 0 μg/ml ) . The plates were then washed and blocked with powdered milk as described above . After an additional wash , serial dilutions of the eluted fractions of the rectal secretions were added to the plates . To do these serial dilutions , the 85-μl eluate from each rectal weck was diluted 1:3 followed by serial 3-fold dilutions . Rectal wecks from animals chronically infected with SIVmac239 were included as internal controls . The rhesus macaque IgG1 mAb 5L7 and purified rhesus IgG , both produced in house , were used as standards at a starting concentration of 2 . 0 μg/ml , followed by 4-fold serial dilutions . After a 1-hr incubation at room temperature , the plates were washed and the same detection antibody described above was added to the plates for 1 hr at 37°C . The plates were developed and read as described above . The concentrations of anti-gp140 Abs in rectal secretions were normalized to the total IgG content in each sample . The pharmacokinetics of Ipi was determined in plasma from the Group 2 animals . Half-well ELISA plates ( Corning Inc . ) were coated overnight at 4°C with 1 . 0 μg/ml of a chimeric CTLA-4-Ig protein consisting of human CTLA-4 ( CD152 ) and a mouse IgG2a Fc tag at its C-terminus ( Acro Biosystems ) . The plates were washed with PBS-Tween20 and blocked with 5% bovine serum albumin ( BSA; Spectrum Chemical ) in PBS for 1 hr at 37°C . The plates were then washed with PBS-Tween20 and the serially diluted plasma samples were added . The same Ipi batch ( Bristol-Myers Squibb ) that was infused into the Group 2 animals was used as the standard for this assay . The starting concentration of the Ipi standard was 5 . 0 μg/ml , which was serially diluted 2-fold . Bound antibodies were tagged with a goat anti-human IgG HRP-conjugated detection antibody ( 1:10 , 000 dilution; Southern Biotech ) and incubated for 1 hr at 37°C . The plates were developed and read as described above . The concentration of Ipi in plasma was determined using the “Sigmoidal , 4PL , X is log ( concentration ) ” function in Prism ( version 7 . 0e , GraphPad Software , Inc . ) . Anti-Ipi antibodies were quantified by coating half-well ELISA plates with 50 μl of Ipilimumab ( Bristol Myers-Squibb ) at a concentration of 10 . 0 μg/ml . These plates were covered with a plastic film and incubated overnight at 4°C . On the next day , the plates were washed with 1x PBS-Tween20 and blocked with 150 μl of Superblock reagent ( Thermo Fischer ) in PBS for 15 min at room temperature . The plates were washed again , and 50 μl of 2-fold serially diluted plasma was added to the corresponding wells . The plates were incubated for 1 hr at room temperature and then washed . Next , 50 μl of Anti-Ig human lambda light chain-biotin ( Miltenyi Biotec , Inc . ) diluted 1:100 in dilution buffer was added to each well . Plates were incubated for 1 hr at room temperature and washed generously . Then , 50 μl of Streptavidin HRP-conjugated detection antibody ( Invitrogen ) diluted 1:10 , 000 in dilution buffer was added to each well , followed by a 1-hr incubation at room temperature . Plates were then thoroughly washed before being developed with 50 μl of 3 , 3' , 5 , 5'-Tetramethylbenzidine ( EMD Millipore ) at room temperature . This reaction was stopped with 50 μl of TMB Stop Solution ( Southern Biotech ) after a brief incubation . Endpoint titer was determined to be the highest dilution at which the optical density of the post-treatment plasma was greater than two times that of the pre-treatment plasma . Sera from the research animals were screened for neutralization of SIVmac239 and SIVmac316 utilizing the luciferase-based , TZM-bl assay , as described previously [54] . Stocks of replication-competent SIVmac239 and SIVmac316 were produced by transfecting HEK293T cells ( ATCC ) with full-length DNA using the jetPRIME technology ( Polyplus transfection ) . Supernatant was harvested after 72 hrs and stored at −80°C until use . Neutralization was tested by incubating SIVmac239 or SIVmac316 and monkey sera for 1 hr at 37°C before transferring them onto TZM-bl cells ( AIDS Research and Reference Reagent Program , Division of AIDS , NIAID , NIH ) . Neutralization was measured in duplicate wells within each experiment . Neutralization was tested starting at 1:4 serum dilutions followed by eight serial 2-fold dilutions for SIVmac239 or eight 4-fold serial dilutions for SIVmac316 . The ID50 titer was defined by the Sigmoidal , 4PL , X is log ( concentration ) equation in Prism7 ( GraphPad Software ) . The SIVmac239Δvif and SHIVAD8-EOΔvif stocks used in the ADCC assays were produced by transfection of infectious molecular clones into HEK293T cells using GenJet transfection reagent ( SignaGen ) . Virus-containing supernatants were collected 48 and 72 hr post-transfection and stored at −80°C . The original SHIVAD8-EO clone was provided by Dr . Malcom Martin ( NIAID , Bethesda , MD ) . After heat inactivation for 30 min at 56°C , RM plasma samples were tested for non-specific ADCC due to the presence of antibodies to human cellular antigens by co-incubating uninfected CEM . NKR-CCR5-sLTR-Luc target cells ( AIDS Research and Reference Reagent Program , Division of AIDS , NIAID , NIH ) with an NK cell line ( KHYG-1 cells ) expressing rhesus CD16 at a 10:1 effector-to-target ratio in the presence of serial dilutions of plasma [55] . This NK cell line was developed in house , as described previously [55] . Non-specific lysis was detected as a reduction in background luciferase activity ( % RLU ) for target cells incubated with NK cells in the presence compared to the absence of plasma . Plasma samples that directed ADCC against uninfected cells were depleted of anti-human Abs by repeated cycles of incubation with CEM . NKR-CCR5-sLTR-Luc cells , followed by centrifugation and plasma transfer , until ADCC responses to uninfected cells were no longer detectable . To measure ADCC activity in plasma of vaccinated animals , CEM . NKR-CCR5-sLTR-Luc target cells were infected with SIVmac239Δvif or SHIVAD8-EOΔvif ( internal negative control ) by spinoculation for 3 hr at 1200 × g in the presence of 40 μg/ml polybrene ( EMD Millipore ) . Four days after infection , target cells were incubated with the NK cell line KHYG-1 at a 10:1 effector-to-target ratio in the presence of serial plasma dilutions . Luciferase activity was measured after 8 hr using the britelite plus luciferase assay system ( PerkinElmer ) . Triplicate wells were tested at each plasma dilution , and wells containing effector cells incubated with uninfected or infected target cells in the absence of plasma were used to determine background and maximal luciferase activity , respectively . Changes in RLUs under each plasma dilution were used to calculate AUC values for SIVmac239- and SHIV-AD8-infected cells . The area under the SHIV-AD8 curve was subtracted from the area under the SIVmac239 curve , thereby yielding the rAUC value for each animal . These rAUC values were used as a measure of ADCC activity . Mamu-A*01 tetramers bound to peptides corresponding to the Tat28-35SL8 , Env233-241CL9 , Env620-628TL9 , and Vif100-109VL10 epitopes were obtained from the NIH Tetramer Core Facility . These tetramers were labelled with phycoerythrin ( PE ) , allophycocyanin ( APC ) , or Brilliant Violet ( BV ) 421 . The Mamu-A*01/Gag181-189CM9 tetramer was obtained from MBL International . These reagents were used to quantify SIV-specific CD8+ T-cells in PBMC , as described previously [56] . For the time-course analysis of vaccine-induced SIV epitope-specific CD8+ T-cell responses in Groups 1a and 2a , approximately 800 , 000 PBMC were incubated in R10 medium with titrated amounts of each tetramer at room temperature for 45 min . The cells were then stained with fluorochrome-labeled mAbs directed against the surface molecules CD3 ( clone SP34-2; PerCP Cy5 . 5 ) , CD8α ( clone RPA-T8; BV785 ) , CD14 ( clone M5E2; BV510 ) , CD16 ( clone 3G8; BV510 ) , and CD20 ( clone 2H7; BV510 ) for 25 min at room temperature . This surface staining mAb cocktail also included an amine-reactive dye ( ARD; Live/DEAD Fixable Aqua Dead Cell Stain; Life Technologies ) . The cells were then washed with Wash Buffer ( Dulbecco's PBS with 0 . 1% bovine serum albumin and 0 . 45 g/L NaN3 ) and fixed with 1× BD FACS Lysing Solution ( BD Biosciences ) for 10 min at room temperature . The cells were washed one more time before they were acquired on a BD LSR II cytometer equipped with a 50-mW 405-nm violet laser , a 100-mW 488-nm blue laser , and a 30-mW 635-nm red laser using the FACSDIVA ( version 6 ) software . To determine the memory phenotype of vaccine-induced SIV-specific CD8+ T-cells in Groups 1a and 2a at the time of the first SIV challenge , 2 . 4×106 PBMCs were incubated with an APC-conjugated Mamu-A*01/Gag181-189CM9 tetramer at room temperature for 45 min . The cells were then stained with the same mAbs specific for CD8α , CD14 , CD16 , and CD20 described above , plus mAbs against CD28 ( clone 28 . 2; PE Cy7 ) and CCR7 ( clone 150503; FITC ) . ARD Aqua was also included in this surface staining mAb cocktail . After a 25-min incubation at room temperature , cells were treated with 1× BD FACS Lysing Solution ( BD Biosciences ) for 10 min and subsequently washed with “Wash Buffer” ( Dulbecco’s PBS with 0 . 1% BSA and 0 . 45 g/L NaN3 ) . Cells were then permeabilized by treatment with “Perm buffer” [1X BD FACS Lysing Solution 2 ( Beckton Dickinson ) and 0 . 05% of Tween-20 ( Sigma-Aldrich ) ] for 10 min , washed , and stained with a mAb against CD3 ( clone SP34-2 ) . After a 30-min incubation in the dark at room temperature , cells were washed and stored at 4°C until acquisition . Samples were acquired in the same flow cytometer described above . FlowJo 9 . 9 ( FlowJo , LLC ) was used to analyze the data . Doublets were excluded by gating out events with disproportionally high width first in a forward scatter ( FS ) area vs . FS width plot , and then in a side scatter ( SC ) area vs . SC width plot . Next , a time gate was created that included only those events that were recorded within the 15th and 85th percentiles of acquisition time . The resulting cells were then gated on "dump channel" ( CD14 , CD16 , CD20 , ARD ) negative , CD3+ cells . Because MHC-I tetramer binding to the T-cell receptor ( TCR ) can lead to TCR internalization , the CD3 gate at this stage included cells that expressed intermediate levels of CD3 . Next , the lymphocyte population was delineated based on its FS and SS properties and subsequent analyses were conducted within CD8+ cells . After outlining MHC-I tetramer+ cells , the memory phenotyping analysis was performed within this gate . Rhesus macaque memory T-cells can be classified into three subsets based on the differential expression of CD28 and CCR7: central memory ( TCM; CD28+CCR7+ ) , transitional memory ( TEM1; CD28+CCR7− ) , and fully-differentiated effector memory ( TEM2; CD28−CCR7− ) . Cells stained with fluorochrome-labeled mAbs of the same isotypes as the anti-CD28 and anti-CCR7 mAbs guided the identification of the memory subsets within the MHC-I tetramer+ population . Based on this gating strategy , all tetramer frequencies mentioned in this manuscript correspond to percentages of live CD14− CD16− CD20− CD3+ CD8+ tetramer+ lymphocytes ( S10 Fig ) . The antigen stimuli for the ICS assays consisted of 16 pools of SIVmac239 15mer peptides overlapping by 11 amino acids corresponding to Gag amino acids 1–263 and 253–510; Pol amino acids 1–354 , 344–700 , and 690–1060; and Env amino acids 1–175 , 161–355 , 340–531 , 516–707 , and 692–879 . Individual pools were used containing peptides that covered the entire ORFs of Vif , Vpx , Vpr , Tat , Rev , and Nef . The final concentration of each 15mer in the ICS tubes was 0 . 1–1 . 0 μM , depending on the peptide pool . Leukocyte Activation Cocktail ( LAC; BD Pharmingen ) and tissue culture medium devoid of stimulatory peptides were used as the positive and negative controls , respectively . The assay was set up in 12 × 75-mm polypropylene tubes ( Corning , Inc . ) each containing 1 . 6×106 freshly isolated PBMC in a final volume of 1 . 0 ml of R10 . The medium also contained unlabeled co-stimulatory mAbs against CD28 and CD49d , and a PE-conjugated mAb specific for CD107a . After adding the appropriate stimulation cocktails to each tube , the cells were placed in a 5 . 0% CO2 incubator at 37°C for 9 hr and then refrigerated to 4°C until staining with fluorochrome-labeled mAbs . To inhibit protein transport , Brefeldin A ( Biolegend , Inc . ) and GolgiStop ( BD Biosciences ) were added to all tubes 1 hr into this incubation period at concentrations of 5 . 0 μg/ml and 0 . 7 μg/ml , respectively . The surface staining master mix included mAbs against CD4 ( clone OKT4; BV605 ) and CD8 ( clone RPA-T8; BV785 ) , in addition to the same BV510-conjugated mAbs against CD14 , CD16 , and CD20 , and the ARD AQUA reagent described above . For the ICS step , the cells were fixed with BD FACS Lyse buffer and subsequently permeabilized with Perm Buffer , as described above . The cells were then incubated for 1 hr in the dark at room temperature with mAbs against CD3 ( clone SP34-2; PerCP Cy5 . 5 ) , IFN-γ ( clone 4S . B3; BV421 ) , TNF-α ( clone Mab11; APC ) , and CD69 ( clone FN50; PE Cy7 ) . Once this incubation was completed , the cells were washed and stored at 4°C until acquisition in the same flow cytometer described above . FlowJo 9 . 9 ( FlowJo , LLC ) was used to analyze the data . The same gating strategy described above was used to exclude doublets , and gate on live CD14− CD16− CD20− CD3+ lymphocytes acquired within the 15th and 85th percentiles of time . Next , T-cell subsets were analyzed based on their expression of either CD4 or CD8 , but not both markers . Functional analyses were conducted within these two compartments by creating gates for each function ( IFN-γ , TNF-α , and CD107a ) . The Boolean gate platform was used to generate a full array of possible combinations , equating to 8 response patterns when testing 3 functions ( 23 = 8 ) . Cells were considered positive for IFN-γ , TNF-α , or CD107a if these molecules were co-expressed with CD69 , a marker of recent activation . At time points of robust SIV-specific CD8+ T-cell activation , a fraction of the CD8+ T-cells that expressed high levels of the degranulation marker CD107a exhibited low to intermediate levels of CD69 . In those cases , all CD107a-expressing cells were considered to be positive based on the lack of background CD107a staining in the negative control tests . LAC-stimulated cells stained with fluorochrome-labeled control mAbs of the same isotypes as those against IFN-γ , TNF-α , and CD107a guided the identification of positive populations . Two criteria were used to determine whether a response was positive . First , the frequency of gated events had to be ≥2-fold higher than their corresponding values in background-subtracted negative control tests . Second , the gates for each response had to contain ≥10 events . These calculations were performed with Microsoft Excel and results were presented as the percentages of responding CD4+ or CD8+ T-cells , that is , live CD14− CD16− CD20− CD3+ lymphocytes of either subset producing any combination of IFN-γ , TNF-α , or CD107a ( S11 Fig ) . Cryopreserved PBMC samples collected at study week 50 ( i . e . , 3–4 weeks prior to the first SIV challenge ) were used in this assay . First , cell suspensions from each animal were stained with LIVE/DEAD Fixable Aqua Dead Cell Stain Kit ( Life Technologies ) for 30 min at room temperature . Next , a cocktail of fluorochrome-labeled mAbs directed against CD3 ( PerCP Cy5 . 5 ) , CD4 ( BV605 ) , CD8α ( BV785 ) , CD14 ( BV510 ) , CD16 ( BV510 ) , CD20 ( BV510 ) , CCR5 ( clone 3A9; PE ) , and HLA-DR ( clone G46-6; APC ) was added to the cells . Except for the latter two mAbs , the clones for all mAbs were the same as the ones described above for the tetramer staining and ICS assays . After a 30-min incubation at room temperature , the cells were fixed and permeabilized utilizing the Transcription Factor Buffer Set kit ( BD Biosciences ) according to the manufacturer’s instructions . The intracellular staining step included a mAb against ki-67 ( clone B56; BV421 ) . Data was collected using the BD LSR II flow cytometer described above . FlowJo 10 . 5 . 3 ( FlowJo , LLC ) was used to analyze the data . The same gating strategy described above was used to exclude doublets , and gate on live CD14− CD16− CD20− CD3+ lymphocytes acquired within the 15th and 85th percentiles of time . Because CD4+ T-cells showed little or no staining with the anti-HLA-DR mAb , HLA-DR expression was not considered for this analysis . The percentages of cells expressing ki-67 and/or CCR5 were determined within a CD4+ CD8− gate . ELISpot assays were performed based on a previous protocol [57] . The assays were carried out according to the manufacturer’s instructions ( Mabtech , Inc . ) and used 120 , 000 purified plasmablasts per well . To measure total IgG-secreting cells and SIV Env-specific IgG-secreting cells , the plates were coated with either an anti-human IgG mAb ( Mabtech , Inc . , 15 . 0 μg/ml ) or purified SIVmac239 gp140 ( Immune-Tech , 1 . 0 μg/ml ) , respectively . RM plasmablasts were sorted from freshly isolated PBMC utilizing a FACSJazz ( BD Biosciences , San Jose , CA ) cell sorter equipped with a 488-nm blue laser and a 640-nm red laser . The cells were sorted based on the following phenotype: lymphocytes / singlets / live / CD3− CD16− CD20 ( neg/int ) / HLA-DR+ / CD14− CD11c− CD123− / CD80+ . Wells were imaged and spots were enumerated with an AID ELISpot reader ( AID ) . Assay results are shown as spot-forming cells ( SFC ) per 106 cells . To obtain positive and negative controls for this assay , SIV-infected and SIV naïve RMs were bled 4–7 days prior to each ELISpot assay and their PBMCs were cultured in R10 in the presence of R-848 ( 1 . 0 μg/ml ) and rhesus interleukin-2 ( 10 . 0 ng/ml ) . On the day of the ELISpot assays , these cultures were washed extensively and then subjected to the same cell sorting procedure as the freshly isolated PBMC from the Group 1 and Group 2 animals . To determine whether the rate of SIVmac239 acquisition differed between Groups 1 and 3 , and between Groups 2 and 3 , the time to productive infection was analyzed using the Cox proportional hazard model . Groups 1 and 3 , and Groups 2 and 3 , were also compared in terms of the number of infected versus uninfected animals after six SIV challenges using Fisher’s exact test . These same comparisons were performed using an expanded control group consisting of the eight contemporaneous monkeys in Group 3 plus 52 historical control animals . For comparing immunological variables between Groups 1 and 2 , median regression was used when the variable was expressed as a proportion or ratio [58] . In all other cases , Welch’s t-test was used for these comparisons . Welch’s t-test was also used for comparing peak and setpoint viral loads between Groups 1–3 . To quantitatively assess the association between infection rate and pre-challenge attributes of vaccine-induced SIV-specific responses , we performed Cox proportional hazard regression using the number of SIV challenges as the outcome and immunological responses as primary predictors . As the scale of immunological responses varies , we also performed the same analysis using standardized values ( also known as Z-score ) , which results in different hazard ratios ( reported in Table 1 ) but identical P values as the classical analysis . We searched for associations between pre-challenge attributes of vaccine-induced SIV-specific responses and control of viral replication in SIV-infected vaccinees using the Spearman rank correlation method . To test the difference in immunological responses over time , we performed mixed effect median regression , using time and group-by-time interaction as fixed effects , and individual difference as a random effect . The type I error rate for all the tests was fixed at 0 . 05 ( 2-tailed ) . | Given the remarkable immune evasion properties of human immunodeficiency virus ( HIV ) , unorthodox vaccine approaches might be needed to elicit protective anti-HIV immunity . Here we investigated whether vaccination with a near-full-length simian immunodeficiency virus ( SIVnfl ) genome could protect rhesus macaques ( RMs ) against pathogenic SIV challenge . To deliver SIVnfl , we used a DNA prime followed by a booster with a herpesvirus that establishes persistent infection in RMs . SIVnfl vaccinees developed long-lasting antibodies against the SIV Envelope protein and T-cell responses against the entire SIV proteome . Encouragingly , SIVnfl vaccinees were significantly protected against intrarectal challenge with SIVmac239 , a feat that until now had only been accomplished by live-attenuated strains of SIV . Curiously , combining the aforementioned vaccine regimen with immune checkpoint blockade did not result in protection against SIVmac239 infection in a separate group of RMs . Vaccination with inactivated immunodeficiency virus genomes should , therefore , be considered as a potential strategy for eliciting anti-HIV immunity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"enzyme-linked",
"immunoassays",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"microbiology",
"retroviruses",
"viruses",
"immunodeficiency",
"viruses",
"vaccines",
"p... | 2019 | Vaccine protection against rectal acquisition of SIVmac239 in rhesus macaques |
Positive-strand RNA viruses genome replication invariably is associated with vesicles or other rearranged cellular membranes . Brome mosaic virus ( BMV ) RNA replication occurs on perinuclear endoplasmic reticulum ( ER ) membranes in ~70 nm vesicular invaginations ( spherules ) . BMV RNA replication vesicles show multiple parallels with membrane-enveloped , budding retrovirus virions , whose envelopment and release depend on the host ESCRT ( endosomal sorting complexes required for transport ) membrane-remodeling machinery . We now find that deleting components of the ESCRT pathway results in at least two distinct BMV phenotypes . One group of genes regulate RNA replication and the frequency of viral replication complex formation , but had no effect on spherule size , while a second group of genes regulate RNA replication in a way or ways independent of spherule formation . In particular , deleting SNF7 inhibits BMV RNA replication > 25-fold and abolishes detectable BMV spherule formation , even though the BMV RNA replication proteins accumulate and localize normally on perinuclear ER membranes . Moreover , BMV ESCRT recruitment and spherule assembly depend on different sets of protein-protein interactions from those used by multivesicular body vesicles , HIV-1 virion budding , or tomato bushy stunt virus ( TBSV ) spherule formation . These and other data demonstrate that BMV requires cellular ESCRT components for proper formation and function of its vesicular RNA replication compartments . The results highlight growing but diverse interactions of ESCRT factors with many viruses and viral processes , and potential value of the ESCRT pathway as a target for broad-spectrum antiviral resistance .
A universal feature of positive-strand RNA ( ( + ) RNA ) viruses is that they multiply their RNA on intracellular membranes , usually in association with vesiculation or other membrane rearrangements [1–3] . Cellular membranes play crucial roles in RNA replication by providing necessary host factors [4–11] , serving as a scaffold to localize viral components necessary for viral replication [12] , and protecting viral RNA from cellular defense mechanisms [13] . One ( + ) RNA virus that has been studied as a model for RNA replication is brome mosaic virus ( BMV ) , a member of the alphavirus-like superfamily of human , animal , and plant viruses . BMV is the type member of the bromoviruses , a group of icosahedral , tripartite , ( + ) RNA viruses that infect plants [14] . BMV can also replicate in the yeast Saccharomyces cerevisiae [15] . The techniques of yeast genetics and molecular biology have greatly facilitated investigation of BMV replication and host-virus interactions [5 , 16 , 17] . In both yeast and plant cells , BMV RNA replication depends on the viral 1a and 2apol proteins and specific cis-acting RNA signals [18] , generates a considerable excess of positive- to negative-strand RNA [15] , and efficiently directs subgenomic mRNA synthesis [15] . Additionally , in yeast as in plants , the 1a and 2apol proteins properly localize to the ER [19 , 20] and viral RNA is selectively encapsidated into progeny virions [21] . In yeast , replication factor 1a localizes to the outer perinuclear ER membranes and induces 50–75 nm vesicular invaginations or spherules that , in the presence of 2apol and a replicable RNA template , serve as compartments or mini-organelles for RNA replication [22] . Highly similar spherular invaginations of the perinuclear ER membrane are prominent features of natural plant infections by BMV and its close relatives CCMV and BBMV [23 , 24] , and equivalent invaginations of other membranes are associated with genome replication in plant cells by other Bromoviridae [25] , Tymoviridae [26 , 27] , Tombusviridae [28] and other plant viruses , and in animal cells by human- and animal-infecting alphaviruses [29–31] . The endosomal sorting complex required for transport ( ESCRT ) machinery is conserved from Archaea to animals and is responsible for the formation of intralumenal vesicles ( ILVs ) in the biogenesis of multivesicular bodies ( MVBs ) [32–38] . MVBs are specialized organelles of the endosomal system that are required for lysosomal or vacuolar degradation of membrane proteins [32] . Five ESCRT complexes are recruited sequentially and perform distinct functions in MVB formation . ESCRT-0 clusters ubiquitinated cargo and recruits ESCRT-I , ESCRT-I and ESCRT-II help remodel the membranes and recruit ESCRT-III , which in turn helps mediate membrane fission ( Table 1 ) [35–40] . Finally , through interactions with ESCRT-III components , the AAA ATPase Vps4p and accessory proteins Vta1p , Vps60p , and Did2p help disassemble the ESCRT complexes from the membranes [33 , 34 , 41] . Parts of the ESCRT machinery also function during cell abscission at the final stage of cytokinesis [35 , 42] and during budding of enveloped viruses from the plasma membrane [43–46] . In each case , the ESCRT machinery functions from the cytoplasmic face of the bilayer and invaginates the membrane away from the cytoplasm , thus inducing “negative” membrane curvature . The spherular replication vesicles induced by BMV and many other ( + ) RNA viruses share topological similarities with ESCRT-dependent viral budding and MVB vesicles , in that the membrane is invaginated away from the cytoplasm , with a major difference being that such spherules do not pinch off and release from the membrane [22] . The assembly and function of BMV replication complexes further share similarities to the replicative cores of retroviruses and double-stranded RNA viruses [22 , 47] . Previous results showed that key components of the ESCRT / MVB pathway , DOA4 and BRO1 , were crucial for proper BMV RNA replication [11] . However , the involvement of DOA4 and BRO1 in BMV RNA replication was not dependent on the ESCRT pathway's membrane-shaping functions , but rather on their regulating expression of OLE1 , which encodes Δ9 fatty acid desaturase , and other lipid synthesis genes required for BMV RNA replication [11] . ESCRT components are recruited by tomato bushy stunt virus , another ( + ) RNA virus , to peroxisome membranes , where they promote RNA replication and the assembly of the replicase complex [7 , 48] . We had previously shown that the reticulons , a family of morphogenic proteins that partition into and stabilize highly curved ER tubules , regulate spherule size and play crucial roles in forming and/or maintaining the proper structure and function of the BMV RNA replication compartments [4] . To determine if other host factors are required for this process , we systematically examined if components of the various ESCRT complexes are required for proper BMV RNA replication and replication complex formation . We show that knockout of several ESCRT components resulted in parallel defects in RNA replication and spherule formation , whereas other ESCRT components affected RNA replication independently of spherule formation . Unlike the reticulons , which regulate spherule size , deleting any of the four subunits of the ESCRT-III complex only altered spherule frequency and resulted in at least a 5-fold inhibition of BMV RNA replication . In particular , deleting SNF7 inhibited BMV RNA replication by > 25-fold and blocked detectable spherule formation . Moreover , BMV’s interaction with the ESCRT machinery is distinct from that of TBSV and HIV-1 , in that BMV 1a interacts strongly with major ESCRT-III effector Snf7p without the necessity for a bridging ESCRT-I protein .
Previously , Kushner et al . screened a yeast single-gene deletion library and identified ~100 genes whose deletion inhibited or enhanced BMV RNA replication by >3-fold [16] . Since many of the ESCRT deletion strains either did not transform or did not grow well in the BY4743 yeast background used for the screen , we deleted individual ESCRT components in the S . cerevisiae YPH500 background . YPH500 has been used in many past studies of the assembly , structure and function of BMV RNA replication compartments [4 , 22 , 49–52] . Among the strains generated were deletions of components of ESCRT-0 ( hse1 Δ ) , ESCRT-I ( vps23 Δ ) , ESCRT-II ( vps36 Δ ) , ESCRT-III ( vps20 Δ , snf7 Δ , vps24 Δ , vps2 Δ ) as well as accessory proteins ( vps4Δ , did2 Δ , vta1 Δ , vps60 ) ( Table 1 ) . In yeast expressing BMV 1a and 2apol proteins , plasmid-launched positive-strand RNA3 transcripts serve as templates for synthesis of negative-strand RNA3 , which in turn is copied to amplify high levels of positive-strand RNA3 and to produce an additional mRNA species , subgenomic RNA4 ( Fig . 1A ) [15] . To determine the effects of these ESCRT gene knockouts on BMV RNA replication , we measured the accumulation of negative-strand RNA3 and positive-strand subgenomic RNA4 , which are produced solely by RNA-dependent RNA synthesis , with no background from plasmid DNA-directed transcription . Deleting components of ESCRT-0 ( hse1 Δ ) , ESCRT-II ( vps36 Δ ) , or the Vps4 accessory proteins Vta1 and Vps60 did not significantly affect either negative-strand RNA3 or positive-strand RNA4 synthesis ( Figs . 1B and S1 ) . Since they had at most minor effects on BMV RNA replication , the vta1 Δ and vps60 Δ strains were used as controls in experiments described in the sections below . A more pronounced effect was observed in the ESCRT-I deletion strain vps23 Δ , resulting in six- and four-fold reductions in positive-strand RNA4 and negative-strand RNA3 levels , respectively ( Fig . 1B and Table 2 ) . In turn , deleting Vps4 , the AAA ATPase that catalyzes the disassembly of the ESCRT machinery and the recycling of its subunits from membranes [53] , reduced positive-strand RNA4 and negative-strand RNA3 production by three- and two-fold , respectively ( Fig . 1B ) . Interestingly , deleting any of the four subunits of the ESCRT-III complex resulted in a 3-fold or greater inhibition of BMV RNA replication . In particular , deleting Vps20 and Snf7 inhibited positive-strand RNA4 replication by 10- and 25-fold , respectively ( Fig . 1B ) . Thus , while multiple ESCRT components are required for efficient BMV RNA replication , the key ESCRT III effector Snf7 in particular appears to be essential for this process . To rule out that the reduced BMV RNA replication in the ESCRT mutant strains might merely be due to destabilizing effects on the viral RNA replication proteins , we measured 1a and 2apol levels in wt and ESCRT deletion yeast . Western blotting of yeast cells expressing 1a plus 2apol showed that 1a and 2apol levels were not significantly changed in the relevant strains ( Fig . 1C and Table 2 ) , indicating that the dramatic effects of multiple ESCRT factor deletions on BMV RNA replication were not due to altered levels of BMV replicase proteins . To test whether the ESCRT proteins influenced 1a’s ability to associate with membranes , we performed a membrane affinity assay . Lysates of yeast cells expressing 1a in the various ESCRT deletion strains were loaded under flotation gradients , which after centrifugation were fractionated and analyzed by SDS PAGE and western blotting using an anti-1a antibody . As a measure of membrane association , flotation efficiency was determined as the percentage of total 1a protein in the gradient that was present in the top two fractions , where membranes and membrane-associated proteins fractionate . In these assays , 1a floated to the top of the gradient with the membrane fraction for all tested deletions , showing that these ESCRT factors are not required for 1a-membrane association ( Fig . 2A ) . In plant cells and yeast , 1a localizes predominantly at the perinuclear ER [19 , 20] . To more precisely determine if 1a still localized to the perinuclear ER in the absence of the relevant ESCRT proteins , we used immunofluorescence confocal miscroscopy . In wt yeast expressing 1a and 2apol , 1a was mainly associated with the perinuclear and to a lesser degree the peripheral ER membrane , and this typical localization of the 1a protein was not altered in any of the ESCRT deletion strains ( Fig . 2B and Table 2 ) . To determine if the ESCRT proteins might be involved in forming and/or maintaining the BMV-induced membrane rearrangements , we used confocal microscopy to examine the localization of C-terminally HA-tagged versions of Vps23p , Vps20p , Snf7p , Vps4p and Vta1p expressed from their endogenous promoters from centromeric , low-copy number plasmids . Since Vps4 catalyzes ESCRT complex disassembly and recycling from membranes , we performed these confocal microscopy experiments in vps4Δ yeast to help stabilize any potentially transient interactions . In the absence of BMV 1a , consistent with previous reports [54 , 55] , ESCRT proteins were found in cytoplasmic punctate structures and were absent from the nuclear envelope ( Fig . 3A ) . In contrast , in yeast expressing BMV 1a ( Fig . 3B ) , the vast majority of Vps23p , Snf7p , Vps4p , and Vta1p colocalized with 1a in the perinuclear ER , the site of viral RNA replication in both plant and yeast cells [19 , 20] . Interestingly , only a small percentage of Vps20p colocalized with 1a in the perinuclear ER , while the rest was found in punctate cytoplasmic structures ( Fig . 3B ) . Since expressing BMV 1a without other viral factors strikingly relocalized multiple ESCRT components ( Fig . 3B ) , we tested whether 1a interacted with these ESCRT proteins . Accordingly , we performed coimmunoprecipitation ( co-IP ) assays in vps4Δ yeast transformed with an empty plasmid or plasmids expressing HA-tagged Vps23p , Vps20p , Snf7p , Vps4p , and Vta1p in the absence or presence of 1a . The yeast cells were lysed in buffer containing ionic ( 0 . 1% SDS ) and nonionic ( 1% NP-40 ) detergents to disrupt membranes , and anti-1a or anti-HA antibodies and protein A sepharose beads were used to precipitate protein complexes . Total cell lysates and the immunoprecipitates eluted from protein A sepharose beads were subjected to SDS-PAGE and immunoblotted using anti-1a or anti-HA antibodies . As expected , on Western blots of lysates that were immunoprecipitated with anti-1a antibodies , a 1a signal was detected for all samples expressing 1a ( Fig . 4A , top panel ) . After immunoprecipitating with anti-HA antibodies , anti-1a antibodies detected 1a only in cells co-expressing Snf7p-HA and 1a ( Fig . 4A ) . Similarly , after immunoprecipitating with anti-1a antibodies , Snf7p-HA was the only ESCRT protein strongly detected by anti-HA antibodies ( Fig . 4A ) . Immunoprecipitating with anti-HA antibody and immunoblotting for HA confirmed that all of the HA-tagged proteins were expressed , although Vps23-HA and Vps20-HA pulled down at lower levels in the presence of 1a ( Fig . 4A , bottom panel ) . The results show that although several ESCRT proteins relocalize to the perinuclear ER in the presence of 1a , only the 1a interaction with Snf7 was preserved under the stringent immunoprecipitation conditions used . Since 1a immunoprecipitates with Snf7p and we had previously shown that 1a can also interact with and relocalize the reticulons to the perinuclear ER [4] , we tested if Rtn1p and Snf7p might interact . Yeast expressing GFP-tagged Rtn1 were transformed with plasmids encoding Snf7-HA and either BMV 1a or no added open reading frame . Although there was some variation in the amount of Rtn1-GFP that pulled down in the IPs , Rtn1p immunoprecipitated with 1a but no signal for Snf7p-HA was detected even in cells co-expressing BMV 1a ( Fig . 4B ) . Since Rtn1p and Snf7p do not co-immunoprecipitate with one another , but each co-immunoprecipites with 1a , these results suggest that Snf7p and the reticulons are independently recruited by 1a to the sites of BMV RNA replication . Although only Snf7 immunoprecipitated with BMV 1a , several other ESCRT components relocalized to the perinuclear ER in the presence of 1a ( Fig . 3B ) . This observation suggested a possible role for Snf7p in mediating the recruitment of other ESCRT components to the perinuclear ER by 1a . To address this , we repeated the co-localization studies in snf7Δ yeast . As seen in Fig . 3C , Vps23p-HA , Vps20p-HA , and Vta1p-HA failed to co-localize with 1a in the perinuclear ER in the absence of Snf7p , showing that recruitment of other ESCRT components to the site of viral replication is dependent on the presence of Snf7p . In MVB formation and virus budding , Snf7p is recruited to viral or cellular proteins through interactions with adaptor proteins such as Vps23p and Bro1p ( yeast homologs of human TSG101 and ALIX , respectively ) [36 , 43] . Previous results show that Bro1p is required for BMV RNA replication but this involvement is not dependent on the ESCRT pathway’s membrane-shaping functions but rather on regulating expression of lipid synthesis genes required for BMV RNA replication [11] . To test for a possible role of Bro1p in bridging the Snf7p-1a interaction , two mutations were made in Snf7p ( Snf7L231A/L234A ) that are known to disrupt its interaction with Bro1p [56] . Additionally , since the interactions between 1a and ESCRT components are transient , the experiments were performed in the presence of Vps4E233Q , a Vps4p mutant that is incapable of mediating ESCRT-III disassembly [57] . In snf7 Δ yeast expressing BMV 1a , Snf7L231A/L234A-HA and Vps4E233Q-FLAG , Snf7L231A/L234A-HA co-localized with 1a in the perinuclear ER , showing that Bro1p is not required for Snf7p to interact with 1a ( Fig . 3D ) . This result is consistent with our previous conclusion that Bro1p’s contribution to BMV RNA replication is independent of the ESCRT pathway’s membrane-shaping functions [11] . As noted earlier , in wt yeast , expressing 1a alone induces the invagination of the outer , perinuclear ER membrane to form spherular compartments associated with BMV RNA protection and replication [22] . Intriguingly , the frequency and average diameter of these RNA replication vesicles can be changed dramatically by mutating one face of a membrane-interacting 1a amphipathic helix [50] , or by knocking out the membrane-shaping host reticulon proteins [4] , or a host factor promoting long chain fatty acid accumulation [52] . To determine whether the formation or structure of such spherules was affected by deleting the tested ESCRT proteins , we examined wt and single-knockout yeast by electron microscopy ( EM ) and measured the abundance and diameter of spherules in the subset of cells that were sectioned through the nucleus among a total of 250 cells for each strain . In wt yeast , the average diameter of spherules was 67 ± 9 nm ( Fig . 5B and Table 2 ) , similar to previous results for BMV and other Bromoviridae in yeast [22] and plants [23–25] . Notably , and unlike the results of depleting membrane-shaping reticulons [4] , none of the tested ESCRT deletions significantly altered the distribution of spherule diameters ( Fig . 5C–F and Table 2 ) . However , deleting certain ESCRT factors reduced the number of spherules formed , which mirrors their inhibitory effects on RNA replication ( Fig . 1 ) . Most significantly , deleting SNF7 abolished detectable spherule formation ( Fig . 5D and Table 2 ) . Similarly , deleting the Snf7p oligomer capping protein Vps24p or the Snf7p recycling ATPase Vps4p , which reduced RNA4 production to 10–30% of wt , reduced spherule frequency to ~40% of that in wt cells ( Fig . 5E–F ) . The remaining ESCRT deletions ( vps2 Δ , 20 Δ , 23 Δ and did2 Δ ) , which inhibited RNA4 production to 10–33% of wt , produced spherules at 77–90% of normal frequency ( Table 2 ) . Thus , these ESCRT genes either alter the replication compartment in ways not visible at the resolution of standard electron microscopy , or are required for other step ( s ) of RNA replication distinct from forming the replication vesicle . Finally , vta1 Δ , and vps60 Δ yeast , which supported 90% of wt RNA4 production , also supported wt levels of spherule production ( Table 2 ) . Thus , while at least some ESCRT-III components and Vps4p are necessary to efficiently form spherules , they are not involved in regulating spherule size . Moreover , Snf7p was the only tested ESCRT component required for this process since detectable spherule formation was abolished in snf7 Δ yeast . To further confirm that ESCRT components were necessary for proper BMV RNA replication , we exogenously expressed in each relevant ESCRT deletion strain a C-terminally HA-tagged version of its corresponding deleted protein from its own promoter in a low copy number plasmid , as in Fig . 3 . As seen in Fig . 6A , expressing Vps4p-HA and Vps23p-HA in their respective deletion strains restored BMV RNA replication to levels similar to those obtained in wt yeast . Likewise , BMV RNA replication was substantially restored when HA-tagged Vps20p , Vps24p , and Did2p were expressed in their respective yeast deletion strains ( Fig . 6A ) . Any residual reductions in BMV RNA replication from wt yeast might be due to impaired function of the HA-tagged protein , since in vta1 Δ yeast , expressing Vta1p-HA reduced BMV RNA replication levels by two fold from their initially wt level . Most importantly , in snf7 Δ yeast , expressing Snf7p-HA not only increased BMV RNA replication by ~20-fold ( Fig . 6A ) but also restored spherule formation in the perinuclear ER ( Fig . 6B ) . Spherules were 62 nm in diameter and were present at ~75% frequency compared to those in wt yeast . These results support the observations above that Snf7p is crucial for both spherule formation and BMV RNA replication . To complement and provide a foundation for testing the effects of SNF7 on BMV RNA replication in plants , we first cloned AtSNF7–2 , the Arabidopsis thaliana homolog of yeast Snf7p , into a yeast vector and expressed it in snf7 Δ yeast . Wild type or snf7 Δ yeast were transformed with plasmids expressing BMV 1a , 2apol , RNA3 and either AtSNF7–2 or an empty plasmid . As shown earlier , positive-strand RNA4 and negative-strand RNA3 accumulation were severely inhibited in the absence of Snf7 ( Fig . 7A ) . However , expressing AtSNF7–2 in snf7 Δ yeast restored RNA4 positive-strand and negative-strand accumulation to 80% and 69% of that in wt yeast ( Fig . 7A ) . Since AtSNF7–2 restored BMV RNA replication , we next looked at spherule formation . While spherules were not detected in snf7 Δ yeast among thousands of cells examined by EM ( Fig . 5D ) , in snf7 Δ yeast expressing AtSNF7–2 , spherules with an average diameter of 66 nm were found in the perinuclear ER at ~70% of their frequency in wt yeast ( Fig . 7B ) . Thus , A . thaliana SNF7–2 is fully compatible with BMV in restoring both BMV RNA replication and spherule formation in snf7 Δ yeast . To test the importance of SNF7 for BMV RNA replication in plants , we turned to N . benthamiana , a systemic host for BMV infection . Agrobacterium infiltration of N . benthamiana plants was used to express Arabidopsis SNF7–2 derivatives C terminally tagged with GFP or the HZZ domain , which contains 6xHis , HA and the ZZ domain of protein A [58] . We used these tagged versions because GFP-tagged SNF7 has been shown to act as a dominant-negative inhibitor of wt SNF7 [7] . Under these conditions , BMV RNA replication was reduced by 72% in plants expressing AtSNF7-HZZ ( Fig . 8A ) . In parallel , we expressed HZZ-tagged yeast RFU1 ( regulator of free ubiquitin 1 ) as a control since RFU1 does not have a homolog in plants . In contrast to the inhibitory effect of AtSNF7-HZZ and AtSNF7-GFP , expressing RFU1-HZZ increased BMV RNA accumulation by ~50% compared to the empty vector . To determine whether this inhibition by AtSNF7 mutants was specific for BMV or more generally suppresses RNA replication by other plant ( + ) RNA viruses , we tested tobacco mosaic virus ( TMV ) , which replicates in the perinuclear ER [59] , and tobacco rattle virus ( TRV ) , which replicates in mitochondria [60] . Expressing RFU1-HZZ modestly increased TMV RNA replication ( Fig . 8B ) , but overexpressing AtSN7-HZZ had no effect on either TMV or TRV RNA replication in the infiltrated leaves ( Fig . 8B–C ) . Thus , overall , these results show that expressing AtSNF7-HZZ selectively and specifically interfered with BMV RNA replication in plants .
Our results show that ESCRT requirements for MVB vesicle formation and BMV spherule formation differ in important ways . MVB formation requires sequential recruitment of ESCRT-0 to cluster ubiquitinated cargo , the ESCRT-I and-II supercomplex to generate membrane invaginations into the MVB lumen , and ESCRT-III to direct membrane scission and vesicle release [36 , 37] . In contrast , BMV spherule formation required ESCRT-III factors and Vps4p , but not ESCRT-0 , -I or-II components ( Figs . 5 and S1 ) . This also contrasts with HIV-1 and TBSV , which require ESCRT-I factors for virion budding and spherule formation , respectively [7 , 61] . HIV-1 Gag interacts with ESCRT-I factor TSG101 and ALIX ( human homologs of yeast Vps23p and Bro1p ) to recruit ESCRT-III factor CHMP4 ( human homolog of yeast Snf7p ) to HIV-1 budding sites in a process that bypasses ESCRT-II [44 , 61] . Similarly , TBSV p33 interacts with Vps23p ( TSG101 ) and Bro1p ( ALIX ) , and these interactions appear to be required to recruit other ESCRT components to form the RNA replication compartments [7] . BMV spherule formation did not require ESCRT-0 , I or II , but was sensitive to ESCRT-III components and Vps4p ( Figs . 5 and S1 ) . Most notable were the lack of BMV spherule formation ( Fig . 5 ) and ~25-fold reduction in RNA replication ( Fig . 1B ) in snf7 Δ yeast . Both spherule formation and BMV RNA replication in snf7 Δ yeast were restored upon ectopic expression of HA-tagged Snf7p ( Fig . 6 ) , confirming the origin of these striking defects . Interestingly , in the absence of Snf7p there was a less pronounced effect on negative-strand than on positive-strand synthesis . Among many possible reasons for this , negative-strand synthesis might begin concomitantly with or soon after 1a initiates membrane invagination , but positive-strand synthesis might not proceed until the vesicle is fully enclosed . Importantly , the critical importance of Snf7p for BMV RNA replication is not restricted to yeast , as overexpressing a dominant-negative mutant of Arabidopsis homolog SNF7–2 in N . benthamiana plants inhibited BMV RNA replication , but had no effect on two other plant positive-strand RNA viruses ( Fig . 8 ) . In keeping with these results , wt Arabidopsis SNF7–2 restored BMV spherule formation and RNA replication in snf7 Δ yeast ( Fig . 7 ) . Consistent with a direct role in spherule formation , upon BMV 1a expression , Snf7p re-localized to the perinuclear ER and was the only ESCRT component to co-immunoprecipitate with 1a ( Figs . 3 and 4 ) . Unlike HIV-1 or TBSV , loss of TSG101 homolog Vps23p did not affect BMV spherule formation ( Fig . 5C and Table 2 ) . Moreover , while HIV-1 and TBSV recruit CHMP4/Snf7p and possibly other ESCRT-III factors through bridging interactions with the ESCRT-I factor TSG101/Vps23p , Vps23p did not pull down any detectable 1a under the same conditions that showed strong Snf7p-1a co-immunoprecipitation ( Fig . 4A ) . Thus , while Vps23p also re-localized to the perinuclear ER in the presence of 1a ( Fig . 3A , B ) , any Vps23p-1a interaction is indirect , transient and/or weak ( Fig . 4A ) . Absence of demonstrable 1a-Vps23p interaction is not surprising since 1a lacks any of the L-domains previously characterized in HIV-1 and other retroviruses . Consistent with the notion that 1a does not interact with Vps23p directly and/or stably , our data further shows that Vps23p recruitment to the perinuclear ER and co-localization with 1a is dependent on Snf7p ( Fig . 3C ) . Thus , BMV must initiate ESCRT-III recruitment and spherule vesicle assembly using different protein-protein interactions from those of MVB vesicles , HIV-1 virion budding and TBSV spherule formation . Such variations in the nature of ESCRT interactions might explain in part why BMV spherules remain connected to their parent membrane , unlike MVB vesicles and HIV-1 virions . Yeast ESCRT-III factors Vps20p , Snf7p , Vps24p , and Vps2p are sequentially recruited to membranes and in vitro function together to deform membranes [40] , including forming intralumenal vesicles [37] . Thus , in addition to the block to spherule formation in snf7 Δ yeast , the 2 . 5-fold reduced frequency of spherules in vps24 Δ and vps4Δ yeast ( Fig . 5 ) argues for the active roles of these proteins in spherule formation , and the lesser reductions in vps20 Δ , vps2 Δ , and did2 Δ yeast suggest accessory roles . In keeping with this , BMV RNA replication was reduced ~ 3- to 5-fold in these strains ( Fig . 1B and Table 2 ) . Overall , the variable importance of different ESCRT-III components for BMV spherule formation is similar to HIV-1 virion budding , for which only Snf7p ( CHMP4 ) and Vps2p ( CHMP2 ) are strongly required , while other ESCRT-III proteins have much lesser effects [43] . The likely roles of Vps2p , Vps20p , Vps24p and Vps4p in BMV spherule formation are suggested by their known functions and interactions with Snf7p: Vps20p triggers polymerization of coiled , membrane-deforming Snf7p filaments that are capped by Vps24p and Vps2p , which in turn recruit Vps4p for ESCRT-III disassembly and recycling [62] . Thus , Vps24p capping of Snf7p filaments may enhance their function in spherule formation . Similarly , Vps20p may promote BMV spherule formation by helping to activate Snf7p filament assembly , although reduced but continuing spherule formation in vps20 Δ yeast implies an alternate trigger for Snf7p assembly , possibly through a 1a-Snf7p interaction ( Fig . 4A ) . Vps4p might contribute to BMV spherule formation in at least two ways . First , Vps4p-mediated disassembly is presumably required to recycle ESCRT-III factors to form new spherules . Residual spherule formation in vps4Δ yeast may occur because these dividing cells actively produce new ESCRT-III factors . Second , multiple results suggest that the Vps4p ATPase plays more active functions than simply recycling ESCRT factors , as in contributing energy for membrane deformation or scission [63 , 64] . In turn , decreased spherule formation in the absence of Vps24p , Vps2p and Did2p may be due to their contributions to recruiting Vps4p ( Table 1 ) . Interestingly , the roles of Vps4p in TBSV and BMV spherule formation differ in that for TBSV vps4Δ yeast only formed vestigial , crescent-shaped membrane deformations [48] , while for BMV vps4Δ yeast supported a reduced frequency of spherules of normal shape and size ( Fig . 5F ) . Although Vps23p is dispensable for BMV spherule formation ( unlike TBSV ) , VPS23 deletion reduced BMV RNA replication ~6-fold ( Fig . 1B ) . Similarly , VPS20 and DID2 deletion induced much greater reductions in BMV RNA replication than spherule formation ( Figs . 1B , 5 and Table 2 ) . While the specific roles of these ESCRT factors in RNA replication are not yet clear , one potentially unifying explanation might involve their interactions with the Doa4p deubiquitinase . We previously showed that deleting DOA4 or its activator BRO1 suppresses BMV RNA replication not due to their links to ESCRT membrane-shaping functions , but rather to depleting free ubiquitin and thereby inhibiting transcriptional induction of certain lipid synthesis genes required for functional lipid composition of the membranes surrounding spherules [11 , 65] . Both Vps23p and Vps20p bind Doa4p [66 , 67] , and mutating Vps20p’s Doa4 binding site exacerbates the deubiquitination defect in cells lacking Bro1 [67] . Thus Vps20p and perhaps Vps23p may help to activate Doa4p’s deubiquitinase activity to support BMV RNA replication . Similarly , while a detailed mechanism awaits further details , Did2p was originally identified as the site of a mutation that suppressed Doa4p mutational defects [68] . Based on insights from electron microscope tomography , a new model for HIV budding was proposed in which HIV-1 Gag accumulates at the membrane , forming a partial shell that leads to plasma membrane invagination . Subsequently , Gag recruits ESCRT-I and—III factors , which constrict the initially wide membrane rim induced by Gag assembly , contributing further membrane bending to complete formation and budding of a full , vesicular membrane envelope lined by a partial Gag shell [69] . Similarly , our recent results show that 1a parallels HIV-1 Gag in achieving high-level multimerization on the ER membrane , and implicate this 1a multimerization as one critical driver of membrane invagination [49] , perhaps explaining why ESCRT-I and ESCRT-II are dispensable for BMV spherule formation ( Fig . 5 ) . A model combining these results on 1a multimerization [49] and reticulon interaction [4] with the present ESCRT results is shown in Fig . 9 . Accordingly , 1a multimerization would initiate perinuclear ER membrane invagination away from the cytoplasm ( Fig . 9A ) , followed by recruitment of Snf7p and other ESCRT-III components to constrict the wide membrane rim induced by 1a , further forming the vesicular spherule body and narrowing the vesicle neck to its final dimensions ( Fig . 9B ) . Snf7p assembles coiled filaments that constrict membranes [70] , and most mechanistic models envision that membrane constriction occurs as the filaments slide past each other to tighten and close the coil [71] . While ESCRT-III constriction continues to cleave the vesicle neck to release MVB intralumenal vesicles and HIV-1 virions , BMV avoids this step , leaving the spherule neck stably attached to the ER membrane as neck scission might be blocked by 1a , host factors or both . One possibility shown in Fig . 9C is that neck scission is blocked by host reticulons , another class of host membrane-shaping proteins that are independently recruited by 1a ( Fig . 4B ) , stabilize positive membrane curvature as in spherule necks , and are required for BMV spherule formation and RNA replication [4] . Consistent with this model , protein scaffolds that stabilize high membrane curvature [72] , like those formed by the reticulons ( Rtn ) , inhibit membrane fission [73] . We had previously shown that the reticulons coimmunoprecipitate with 1a and are recruited to the interior of the spherules . Moreover , since the reticulons regulate spherule size and play crucial roles in forming and/or maintaining the proper structure and function of the BMV replication compartments , they remain stably associated within the spherules ( Fig . 9D and [4] ) . In contrast , the association of ESCRT components with spherules is transient , since visualizing these associations required deleting VPS4 ( Figs . 3 and 4 ) . Thus , in the presence of active Vps4p , ESCRT components are released from spherules and recycled as for MVBs . Interestingly , while modulating reticulon levels alters the diameter of BMV spherules [4] , deleting ESCRT genes only modulated the number of spherules formed , with no significant effect on their size ( Fig . 5 and Table 2 ) . This responsiveness of spherule size correlates and is likely mechanistically linked to the stable association of the curvature-modulating reticulons with BMV spherules , compared to the transient association of ESCRT components during spherule formation . Further understanding of the detailed nature and ultrastructure of these interactions should provide additional means to control and potentially to use the remarkable abilities of positive-strand RNA viruses to remodel membranes and to selectively capture and replicate RNAs .
Yeast strain YPH500 and culture conditions were as described previously [15] . The following strains were generated for this work: hse1Δ ( YPH500 hse1::kanMX4 ) , vps23Δ ( YPH500 vps23::kanMX4 ) , vps36Δ ( YPH500 vps36::kanMX4 ) , vps20Δ ( YPH500 vps20::kanMX4 ) , snf7Δ ( YPH500 snf7::kanMX4 ) , vps24Δ ( YPH500 vps24::kanMX4 ) , vps2Δ ( YPH500 vps2::kanMX4 ) , vps4Δ ( YPH500 vps4::kanMX4 ) , vps60Δ ( YPH500 vps60::kanMX4 ) , did2Δ ( YPH500 did2::kanMX4 ) , vta1Δ ( YPH500 vta1::kanMX4 ) . Genomic replacements were made using amplified KanMX4 cassettes flanked by 5’ and 3’ homologous recombination regions . The strain expressing the chromosomal alleles of RTN1 as a GFP fusion was obtained from Invitrogen ( Carlsbad , CA ) . For endogenous expression of ESCRT proteins , the coding region of the full-length protein plus the ~500 base-pair upstream sequences were PCR amplified from wt yeast DNA with appropriate primers and inserted into a CEN plasmid . An HA- or a FLAG-tag was added at the C-terminus of the ESCRT ORFs . Snf7L231A/L234A-HA and Vps4E233Q-FLAG were created by site directed mutagenesis . To express Arabidopsis thaliana SNF7–2 in the yeast snf7Δ mutant strain , AtSNF7–2 cDNA was amplified by PCR and inserted into ycpLAC33 , a CEN plasmid . An HA tag sequence is attached to the 3’ end of AtSNF7–2 coding sequence and expression is under the control of the GAL1 promoter . Plasmids expressing the dominant negative mutant of AtSNF7–2 , SNF7-HZZ , was made by adding the sequence of the HZZ multitag from pBG1805 at the 3’ end of AtSNF7–2 using PCR . The multitag from pBG1805 contains 6×His , HA , and the ZZ domain of protein A [58] . AtSNF7-HZZ was cloned into agrobacterium binary vector pCAMBIA1300 under the control of an enhanced CaMV 35S promoter to make pAG2P-S7-HZZ . To make pMDC-RFU1-HZZ to overexpress yeast RFU1 in plants , the RFU1 gene in pBG1805 ( a Gateway destination vector ) was mobilized to a Gateway entry vector first and then to the Gateway-based binary vector pMDC32 [74] , which expresses target genes using the enhanced CaMV 35S promoter . BMV 1a was expressed under control of the GAL1 promoter using pB1YT3L , a pB3YT3 [75] derivative with a LEU marker , 2apol was expressed from pB2CT15 ( ADH1 promoter ) [15] and BMV RNA3 was expressed under control of a CUP1 promoter from pB3VG128H , or under control of a GAL1 promoter from pB3MS82 , both RNA3 derivatives have a four-nucleotide insertion in the coat protein gene that abolishes expression of the coat protein [76] . To make plasmids launching BMV RNAs mediated by agrobacterium infiltration , overlapping PCR was used to add a CaMV 35S promoter sequence to the 5’ end and a ribozyme from hepatitis delta virus to the 3’ end of the cDNAs of BMV . The BMV launching cassettes were then cloned into pCAMBIA1300 via EcoRI and PstI digestions . Plasmids launching tobacco mosaic virus ( TMV ) and tobacco rattle virus ( TRV ) by agroinfiltration were pTRBO [77] and pTRV-1 and-2 [78] . N . benthamiana plants were grown in a growth chamber under long day conditions ( 16 h light and 8 h dark ) at 26°C . Agrobacterium infiltrations were performed following a protocol described by Bendahmane et al . 2000 [79] . Agrobacteria cultures were grown at 30°C in medium with kanamycin ( 50 mg/L ) , 10 mM MES ( morpholine ethanesulfonic acid , ph 5 . 9 ) , and 50 uM acetosynringone . These were subcultured once and harvested when the OD600 reached 0 . 8–1 . 0 . Agrobacterial cells were brought to OD600 = 1 . 0 in the infiltration solution ( 10 mM MgCl2 , 10 mM MES , pH 5 . 9 , and 150uM acetosynringone ) and incubated at room temperature for at least 3 hours . These cultures were then mixed for co-infiltration suspensions with final concentrations listed as followings: BMV RNA1 , 2 , or 3 at OD600 = 0 . 05; pTRBO , pTRV-1 and pTRV-2 at OD600 = 0 . 25; pMD32-RFU1 , or pAG2P-HZZ at OD600 = 0 . 5 . The aforementioned culture combinations were co-infiltrated into 8-week-old plants . After 48 hours , the infiltrated leaves were harvested for RNA extraction . Plants infiltrated with BMV only or co-infiltrated with pMDC32-RFU1-HZZ were used as controls . For the Northern blot in Fig . 8B , the first lane of the SNF7-HZZ samples was excluded from quantitative analysis since the RNAs migrated more slowly through the gel due to some salt contamination . Ten OD600 units of yeast cells grown to mid-logarithmic phase were spheroplasted and resuspended in 350 μl buffer TNE ( 50 mM Tris-HCl ( pH 7 . 4 ) , 150 mM NaCl , 5 mM EDTA , 5 mM benzamidine , 1 mM PMSF , and 10 μg/ml each aprotinin , leupeptin , and pepstatin A ) . Spheroplasts were lysed via 25 passes through a 22 gauge , 4 cm long needle . Total lysates were centrifuged for 5 minutes at 4°C at 500×g to remove cell debris , and 250 μl of supernatants were mixed with 500 μl of 60% Optiprep ( Axis-Shield , Oslo , Norway ) . Density gradient centrifugation was performed for 2 hours at 55 , 000 rpm in a Beckman TLS55 rotor using 600 μl of each sample overlaid by 1 . 4 ml of 30% Optiprep and 100 μl of lysis buffer [80] . After centrifugation , 6 fractions were collected from top to bottom of the gradient . For protein detection , samples were boiled in SDS loading buffer prior to SDS-PAGE and western blotting . Total yeast RNA isolation by the hot phenol method , Northern blot analysis , total protein extractions and Western blot analysis , and anti-1a , anti-2apol , anti-Dpm1 , anti-FLAG and anti-Pgk1 antibodies were as described [4 , 15 , 81 , 82] . Mouse anti-PDI was acquired from Abcam , rabbit and mouse anti-HA antibodies were purchased from Santa Cruz and Roche , respectively . Mouse anti-GFP was purchased from Covance . Northern blots were imaged on a Typhoon 9200 ( Amersham Biosciences , Piscataway , NJ ) . Band intensities were analyzed by using ImageQuant software ( Molecular Dynamics , Piscataway , NJ ) . Yeast cells were lysed in RIPA buffer ( 1% NP-40 , 0 . 1% SDS , 50 mM Tris pH 8 . 0 , 150 mM NaCl , 0 . 5% Sodium deoxycholate , 5 mM EDTA , 10 mM NaF , 10 mM NaPPi , 2 mM phenylmethylsulfonyl , 5mM benzamidine , and 10 ug/ml each of chymostatin , pepstatin A , leupeptin , bestatin ) using glass beads and a bead beater and the supernatant was collected after centrifugation . For immunoprecipitation , yeast lysates were mixed with Protein A Sepharose beads ( GE Healthcare . Piscataway , NJ ) and anti-1a , anti-HA , or anti-GFP antibodies overnight at 4°C . Beads were pelleted and washed with RIPA buffer before boiling in 1x SDS gel loading buffer and running samples in 4–15% SDS-PAGE gels . Confocal microscopy was as described [80] . Briefly , to detect the subcellular localization of BMV proteins in ESCRT deletion strains , yeast were transformed with plasmids expressing BMV 1a and 2apol proteins . To detect the localization of the ESCRT proteins , HA-tagged ESCRTs were transformed with empty plasmids or plasmids encoding BMV 1a . Cells were fixed with 4% formaldehyde , spheroplasted with lyticase , and permeabilized with 0 . 1% Triton X-100 . Spheroplasts were then stained either by using rabbit anti-1a serum , mouse anti-PDI , mouse anti-HA antibodies or combinations thereof followed by anti-rabbit or anti-mouse secondary antibodies conjugated to Alexa-488 , or Alexa-568 . For nuclear staining , a 10-minute incubation with 300 nM DAPI ( Invitrogen ) was added after secondary antibody incubation . Fluorescent images were acquired with a Nikon A1R Bio-Rad inverted confocal microscope system . To avoid spectral bleed-through in the multi-color imaging , images were acquired by sequentially scanning with the individual lasers and detecting fluorescence in each channel to coincide with laser illumination . Samples were prepared for electron microscopy as described [22] . In brief , yeast cells were fixed for 1 hr with 2% glutaraldehyde and 4% paraformaldehyde , washed , and post-fixed for 1 hr with 1% OsO4 and 1% uranyl acetate . Cells then were dehydrated via a series of step-wise increasing ethanol concentrations ranging from 50% to 100% , and infiltrated and embedded with Spurr’s resin . Samples were sectioned and placed on nickel grids , washed , and incubated for 15 min in 2% glutaraldehyde , post-stained with 8% uranyl acetate and Reynold's lead citrate , and viewed with a Philips CM120 microscope . For each deletion strain , the diameter of >50 spherules were measured with the imaging program ITEM Analysis ( Soft Imaging Systems , Lakewood , Colo . ) . | Positive-strand RNA { ( + ) RNA} viruses cause numerous human , animal , and plant diseases . ( + ) RNA viruses reorganize host intracellular membranes to assemble their RNA replication compartments , which are mini-organelles featuring the close association of both viral and host components . To further understand the role of host components in forming such RNA replication compartments , we used brome mosaic virus ( BMV ) , a well characterized model virus , to study some common features of ( + ) RNA virus RNA replication . We show that knocking out several components of the cellular Endosomal Complex Required for Transport ( ESCRT ) machinery resulted in parallel defects in BMV RNA replication and replication compartment formation , whereas other ESCRT components affected RNA replication independently of replication compartment formation . Deleting a subset of ESCRT proteins altered the frequency of replication compartment formation but had no effect on the size of these compartments , whereas a second subset affected RNA replication independently of replication compartment formation . Moreover , BMV’s interaction with the ESCRT machinery appears to be distinct from that reported for other viruses and from the ESCRT requirements for forming vesicles in cellular multivesicular bodies . These findings further illuminate the remarkable abilities of positive-strand RNA viruses to integrate viral and host protein functions to remodel membranes , and suggest potentially potent new ways to control such viruses . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Host ESCRT Proteins Are Required for Bromovirus RNA Replication Compartment Assembly and Function |
A majority infections caused by dengue virus ( DENV ) are asymptomatic , but a higher incidence of severe illness , such as dengue hemorrhagic fever , is associated with secondary infections , suggesting that pre-existing immunity plays a central role in dengue pathogenesis . Primary infections are typically associated with a largely serotype-specific antibody response , while secondary infections show a shift to a broadly cross-reactive antibody response . We hypothesized that the basis for the shift in serotype-specificity between primary and secondary infections can be found in a change in the antibody fine-specificity . To investigate the link between epitope- and serotype-specificity , we assembled the Dengue Virus Antibody Database , an online repository containing over 400 DENV-specific mAbs , each annotated with information on 1 ) its origin , including the immunogen , host immune history , and selection methods , 2 ) binding/neutralization data against all four DENV serotypes , and 3 ) epitope mapping at the domain or residue level to the DENV E protein . We combined epitope mapping and activity information to determine a residue-level index of epitope propensity and cross-reactivity and generated detailed composite epitope maps of primary and secondary antibody responses . We found differing patterns of epitope-specificity between primary and secondary infections , where secondary responses target a distinct subset of epitopes found in the primary response . We found that secondary infections were marked with an enhanced response to cross-reactive epitopes , such as the fusion-loop and E-dimer region , as well as increased cross-reactivity in what are typically more serotype-specific epitope regions , such as the domain I-II interface and domain III . Our results support the theory that pre-existing cross-reactive memory B cells form the basis for the secondary antibody response , resulting in a broadening of the response in terms of cross-reactivity , and a focusing of the response to a subset of epitopes , including some , such as the fusion-loop region , that are implicated in poor neutralization and antibody-dependent enhancement of infection .
Dengue virus ( DENV ) , an arthropod-borne virus of the Flaviviridae family , infects an estimated 400 million people each year [1] . There are four antigenically related DENV serotypes , DENV 1–4 , each capable of causing disease . DENV infections are often asymptomatic or result in an uncomplicated fever and can elicit life-long immunity to the infecting serotype and short-term cross-protection against heterotypic DENV infections [2–5] . Although , recent studies have demonstrated that homotypic DENV reinfection is possible [6] . Secondary infection with a heterotypic DENV serotype results in a higher incidence of more severe disease and cross-reactive antibodies are thought to contribute to this by a mechanism termed antibody-dependent enhancement ( ADE ) of infection [7–11] . The antibody response following secondary infection is broadly cross-reactive among DENV serotypes and longer periods of cross-protection are observed [3 , 12] . Further characterizing differences in the antibody response to primary and secondary heterotypic DENV infections , and how these differences are associated with serotype-specificity and neutralization , is critical to understanding DHF pathogenesis and developing dengue vaccines . The DENV virion consists of 180 copies of the envelope ( E ) protein , arranged in 90 dimers in an icosahedral ‘herring-bone’ geometry [13] and is the primary target of DENV neutralizing antibodies [14] . The soluble portion of the E protein consists of three distinct domains [15] , termed Domain I ( DI ) , Domain II ( DII ) , and Domain III ( DIII ) . Neutralizing antibodies ( Abs ) targeting E , reviewed in [16] , are the main focus of current DENV vaccine development efforts . Not all E protein-specific Abs contribute equally to virus neutralization and neutralizing Ab potency is related to its epitope . Early work with mouse mAbs indicated that DIII was a major target of potently neutralizing DENV mAbs [17–27] . However , a low fraction of DIII-specific neutralizing Abs are found in human sera post-DENV infection and they only appear to make a minor contribution to DENV neutralization [28–31] . The human neutralizing Ab response appears to preferentially target the DI/DII hinge region of E protein monomers [32–34] and quaternary E protein epitopes that are only present in the context of intact virions [32 , 35 , 36] . Finally , DENV Abs can vary with respect to serotype cross-reactivity . Complex Abs bind or neutralize all four serotypes , type-specific Abs bind or neutralize only a single serotype , and sub-complex Abs bind or neutralize more than one , but not all four serotypes . It is important to note that there are significant strain and genotype-level differences in antibody neutralization within a serotype as well [27 , 37–39] . The antibody response to dengue infection is a polyclonal response that is thought to consists of a repertoire of >103 unique monoclonal antibodies ( mAbs ) [40] . Previous studies from polyclonal sera [41 , 42] , and panels of monoclonal antibodies [36 , 43–45] , have shown that the serotype-specificity of the antibody response shifts between primary and secondary infections . Primary infections are characterized by a largely type-specific antibody response while secondary infections result in a broadly cross-reactive response . We hypothesize that the basis for the shift in serotype-specificity between primary and secondary antibody responses can be found in a change in the fine-specificity—the relative response to different epitopes on the E protein . To investigate the link between epitope fine-specificity and serotype-specificity , we assembled the Dengue Virus Antibody Database ( http://denvabdb . bhsai . org ) , an online repository containing 410 DENV-specific mAbs , each annotated with information on 1 ) its origin , including the immunogen , host immune history , and selection methods , 2 ) binding or neutralization data against four DENV serotypes , and 3 ) epitope mapping at the domain or residue level . Because the database contains information linking infection type ( primary vs . secondary ) and serotype-specificity ( type-specific , sub-complex , complex ) , with residue or domain-level epitope mapping , it allows us to identify the epitope-level determinants of observed shifts in type-specificity associated with secondary infections . While analysis of any single study or panel of DENV mabs may reveal only a limited understanding of the polyclonal diversity of the DENV antibody response , we hypothesize that a large-scale analysis of hundreds of mAbs collected across dozens of diverse studies may be able to identify systematic trends in the DENV antibody repertoire and provide key insights into DENV antibody cross-reactivity and fine specificity . Finally , it is important to note that this study represents a meta-analysis of decades of previous studies on dengue mAbs , and as such seeks to provide quantitative basis for previously observed differences in cross-reactivity and fine-specificity in antibody responses to primary and secondary DENV infections .
We assembled a database of DENV mAbs described in literature that fulfill the following criteria: 1 ) has information on how the mAb was isolated , in terms of the immunogen , the host organism , and immune history; 2 ) has in vitro binding or neutralization data against all four DENV serotypes; and 3 ) binds to E protein , and has epitope mapping information with at least a domain-level resolution . Overall , we found 410 mAbs that matched these criteria . There are three linked sections of the database: mAb , Activity , and Epitope ( Fig 1 ) . The mAb section contains a single record for each mAb in the database that includes the mAb name and information about its origin , including the host organism , the isotype , the immunogen , the exposure event that lead to the antibody , the host immune history , the selection criteria used to isolate or select that mAb for study , and the PubMedID of the reference that first reported the mAb . The Activity section can contain multiple records for each mAb and includes information on the activity assay details and data against all four DENV serotypes , and the corresponding PubMedID for its reference . This data could be qualitative , such as in the case of a Western Blot , or quantitative , as in the case of titers from a neutralization assay . The Epitope section also contains multiple records for each mAb and includes information on the epitope resolution or type ( ‘residue’ or ‘domain’ ) , the epitope mapping method , the domain of E that the mAb was mapped to , the E amino acid sequence that describes that epitope , and the PubMedID of the corresponding reference . The Dengue Antibody Database is freely accessible online at http://denvabdb . bhsai . org . Through the web-based interface , users are able to search for particular mAbs , sort mAbs by various properties , and download any information on the database directly as a spreadsheet . Based on the available activity information , we classified the serotype-specificity of each mAb in the database as ‘type , ’ ‘sub-complex , ’ and ‘complex’ based on the number of serotypes it was reactive towards . In the instance that different assays for the same mAb showed differing patterns of specificity , we classified the mAb based on the most serotype-restricted assay result . Among the cases where assays had conflicting serotype-specificity results , we found that , in general , neutralization assays tended to be more serotype-restricted than binding assays . We treated all activity methods equally to maximize the sample size for our analyses , however users are able to download the activity data from the database in spreadsheet form , and select subsets of mAbs that have data from certain activity methods . Finally , some studies also identify group-specific mAbs , antibodies that can bind to multiple different flaviviruses . However , since such characterization is available for only a limited number of mAbs in the database , we restricted our study to type , sub-complex , and complex mAbs . For many of the measurements , we focused our analysis on comparing type and complex mAbs . This was for two reasons: 1 ) the number of sub-complex mAbs was insufficient to generate statistically significant results for many measures , and 2 ) we found that sub-complex mAbs , in particular , were sensitive to assay-level variation in classification of their type-specificity , meaning they could not always be reliably distinguished from complex mAbs . We developed a measure of epitope-level antigenicity to describe a set of mAbs in terms of ‘epitope propensity , ’ or the probability that a given E protein residue is found within the epitope definitions that make up the set of mAbs . Briefly , a set of mAbs consists of Nr mAbs with residue-level epitope definitions and Nd mAbs with domain-level epitope definitions . Note that the same mAb may have both residue and domain-level definitions . We define the residue-level epitope definition of mAb j , as a set of residues , Ejres={x1 , x2 , x3…xn} and the domain-level epitope definition of mAb j as Ejdom="DI/DII" or "DIII" . Q ( i ) is the number of times residue i is found among all Nr residue-level epitope definitions in the data set ( Eq 1 ) : Q ( i ) = ∑jNrN ( i∈Ejres ) ( 1 ) The epitope propensity at residue i , P ( i ) , defined in Eq 2 , is the probability of finding residue i among the Nr residue-level epitopes that lie within the domain of residue i ( di ) , multiplied by the probability of finding a epitope in di , as determined by Nd domain-level epitope definitions . P ( i | di ) is calculated by dividing the total count of residue i , by the total count across all epitope residues that fall within the domain di , as shown in Eq 3 . P ( di ) is determined by the total count of domain level epitopes that are the same domain as di divided by the total number of domain-level epitopes in the data set , as shown in Eq 4 . By defining epitope propensity as a function of P ( i | di ) and P ( di ) we can use the residue-level epitope definition to determine the high-resolution details of mAbs in the database , while using domain-level epitope definitions to determine the relative immunogenicity of larger segments of the antigen . In any empirical measure there is the risk of observation bias—that researchers intentionally studying antibodies to a particular epitope region will bias a propensity measure calculated from those observations . By including a separate domain-level term , we can , to some degree , account for this potential observation bias . In addition to the epitope-level measure of propensity , we calculated aggregate epitope-level cross-reactivity from mAbs derived from primary and secondary infection . For each residue , we identified every mAb within the mAb subset ( primary or secondary infection ) that listed that residue as an epitope . We then determined the percentage of mAbs associated with that epitope residue that were classified as ‘complex’ . If a residue had 60% or greater cross-reactivity , meaning at least 60% of all mAbs associated with that residue were classified as ‘complex’ , we defined it as being of ‘high’ cross-reactivity . Residues with cross reactivity 30–60% cross-reactivity were defined as having ‘medium’ cross-reactivity; residues with <30% cross-reactivity were defined as having ‘low’ cross-reactivity . We carried out two separate , but related , analyses to assess the degree of sequence variation in DENV epitopes on E: a residue-level measure of sequence conservation and an epitope-level measure of antigenic mismatch , known as pepitope [46 , 47] . We carried out multiple sequence alignment using a set of 47 sequences of DENV across all four serotypes that were studied by Katzelnick et al . [37] for serum cross-reactivity and neutralization . We downloaded the E protein sequences for each strain from Genbank ( see S2 Table ) . We then carried out a structure-based multiple sequence alignment using the Consurf algorithm [48] using the crystal structure of the DENV E protein [15] ( PDB code: 1OKE ) . We defined sequence conservation at each residue position as the percentage of sequences that had the most common amino acid at that position . pepitope is an epitope-level measure of antigenic mismatch between two viral strains and can be used to predict the likelihood that immunity to one strain would provide protection against the other . We used one representative strain of DENV for each of the four serotypes , which was the consensus sequence for the E protein for each of the four serotypes as determined by a previous bioinformatics study by Danecek et al . [49] . For a residue-level epitope definition for mAb j , Ejres , pepitope is calculated by dividing the number of mismatches between two strains ( S1 and S2 ) along the residues defined by Ejres with the total number of residues that defines Ejres—describing a mismatch percentage along a defined set of epitope residues ( Eq 5 ) . We used ClustalW [50] to carry out the alignment and defined a mismatch as identified non-conserved or semi-conserved substitutions ( denoted by ‘ ‘ and ‘ . ’ , respectively , in the ClustalW sequence alignment file ) in the alignment . For each mAb in the dataset with residue-level epitope definitions of five residues or more , we calculated pepitope for all pairwise comparisons between the four serotypes and report the average pepitope value as the pepitope for that mAb . In addition to calculating pepitope values between representative sequences of DENV1-4 , we also calculated pairwise pepitope values for each strain within each serotype for the 47 sequences in the Kaetzelnick data set in order to compare intra-serotype and inter-serotype sequence variation at the epitope level .
There are three linked sections of the database: mAb , Activity , and Epitope ( Fig 1 ) . The mAb section contains a single record for each mAb in the database that includes the mAb name and information about its origin , including the host organism , the isotype , the immunogen , the exposure event that lead to the antibody , the host immune history , the selection criteria used to isolate or select that mAb for study , and the PubMedID of the reference that first reported the mAb . The Activity section can contain multiple records for each mAb and includes information on the activity assay details and data against all four DENV serotypes , and the corresponding PubMedID for its reference . The breakdown of the database is shown in Table 1 . Overall , the database is approximately evenly divided between mouse mAbs and human mAbs , from both primary and secondary infection . ELISA and neutralization assays are the most common activity records , accounting for almost 75% of the activity information . In Epitope records , there are only eight cryo-EM structures and eighteen x-ray crystallographic structures , representing <4% of the mAbs in the database which reflects the relative paucity of high-resolution epitope information on DENV mAbs . Most mAbs in the database have either mutagenesis or yeast-display data , underscoring the value of synthesizing this sparse epitope information to build a more comprehensive picture of DENV E protein epitopes . It is important to note that factors such as the antigen used for B cell selection , and even the time point , post-infection , when cell samples were collected to isolate antigen-specific B cells , can play a major role in epitope fine-specificity and cross-reactivity . We analyzed the database for all human antibodies , and for human and mouse antibodies , irrespective of these factors , in order to maximize the sample size for the analysis . However , this information is included in the database , and attached in S2 Table , for use in any further study . In certain cases , outlined in their respective sections , we did look for biases that selection methods may have had on the results . Overall , 45% of published mAbs from human primary infection were type-specific , 21% were sub-complex , and 34% were complex , while for mAbs from human secondary infection , 4% were type-specific , 10% were sub-complex , and 86% were complex ( Fig 2a ) . These results reflect findings from polyclonal sera reported elsewhere [41] and underscore the profound shift from a type-specific primary immune response to an almost entirely cross-reactive secondary immune response . We next sought to map the epitopes from primary and secondary responses and determine the link between the epitope fine-specificity of the antibody response and its serotype-specificity . We calculated epitope propensity across all residues in the DENV E protein from published human mAbs from primary and secondary DENV infections ( Fig 2b ) . Overall , the results show that mAbs from secondary infection bind to a subset of the epitope residues for mAbs from primary infections . We also calculated epitope cross-reactivity as the percentage of mAbs associated with a particular epitope residue that have a serotype-specificity classification of ‘complex . ’ We found that epitope residues from primary infections had a mix of cross-reactivity , ranging from low ( <30% ) , medium ( 30%-60% ) , and high ( >60% ) , while epitope residues from secondary infections were almost entirely of high cross-reactivity . We generated a ‘composite’ epitope map of DENV antibody responses by mapping the epitope propensity and cross-reactivity of human mAbs from primary and secondary infections to the structure of the DENV E protein dimer [15] ( Fig 2c ) . We found three overall trends . First , the fusion loop region , which showed high cross-reactivity in both primary and secondary responses , show an enhanced immunogenicity in secondary infections . Second , the DIII region , which displayed comparable immunogenicity in primary and secondary infections , showed a marked shift in both cross-reactivity and fine-specificity . The DIII response in secondary infections , unlike in primary infections , was highly cross-reactive , and shifted away from the lateral-ridge epitopes and towards the A-strand and DII/DIII interface epitopes . Finally , we found that two other epitope regions , the dimer interface , and the DI/DII interface , showed moderate immunogenicity in primary infections , with a medium level of cross-reactivity . In secondary infections , both of these regions showed a shift to high cross-reactivity . Katzelnick et al . found that although DENV strains cluster into discrete serotypes with respect to sequence similarity within the E protein , they do not cluster into discrete serotypes in terms of antigenic similarity [37] . We hypothesized that , among epitope residues identified in the database , the concordance between antigenic and sequence similarity might be greater . We used the set of 47 DENV strains studied by Katzelnick et al . , to quantify the degree of sequence variation between the four serotypes to determine if epitopes associated with type-specific mAbs could be distinguished from epitopes associated with cross-reactive mAbs . We carried out a multiple sequence alignment using the Consurf algorithm [48] to align E protein sequences from all 47 DENV strains . For each residue in E , we calculated a sequence conservation measure as the percentage of the aligned sequences that had the most common amino acid at that position . In S1 Fig , we show the sequence diversity ( defined as 1 − sequence conservation ) , across the E protein for all four serotypes . It is important to note that we sought specifically to compare intra-serotype variation from our epitope-based analysis with the results from the Katzelnick study , not characterize intra-serotype variation more generally . We generated histograms with respect to sequence conservation among DENV 1–4 E protein epitope residues from type-specific and complex mAbs from primary and secondary infections ( Fig 3A ) . Our results show that epitope residues for type-specific mAbs show a bimodal distribution with a majority of residues falling between 40% to 80% sequence conservation . By contrast , most epitope residues from complex-specific mAbs fall in the 80% to 100% sequence conservation range . For comparison , the sequence conservation for all residues shows that distribution if the respective type-specific and complex antibody responses were to target residues at random . These results show that primary type-specific antibodies preferentially target residues with lower sequence conservation . Furthermore , they show that complex antibodies do not have strong preference for conserved residues , as might be expected . Previous research in antigenic variation in influenza virus identified a sequence-based measure , pepitope that was found to be correlated with antibody cross-neutralization [46 , 47] between strains . pepitope is calculated by determining the proportion of sequence mismatches across a defined epitope region , between two strains . In order to calculate a reliable pepitope value , an epitope definition of a sufficient size is necessary . For each mAb in the database with a residue-level epitope definition of at least five residues , we calculated a pepitope value for that epitope based on the average pairwise pepitope across all possible pairs of the four DENV serotypes ( Fig 3B ) . For reference , a distribution of defined epitope sizes is shown in S3 Fig . Our results show that pepitope successfully distinguishes between type-specific and complex mAbs and suggests that a threshold mismatch of 20% of an epitope is sufficient to result in type-specificity . Interestingly , Gupta et al . found that a similar threshold of pepitope = 20% corresponds to a loss of vaccine efficacy between two influenza strains [46] . For a typical mAb epitope of 25–35 residues in size , this would correspond to at least five amino acid mismatches among the epitope residues . It is possible that antigenic variation among only a subset of epitope residues is responsible for type-specificity . For example , previous mutagenesis experiments have shown that only mutations at certain key epitope residues abrogate antibody binding , while mutations at other epitope residues have no effect [19 , 24] . We extracted a subset of human mAbs in the database whose epitopes were defined exclusively by cell passaging or mutagenesis , and would thus reflect not just structural , but functionally significant epitope residues . When we generated a composite map of these epitopes on the E protein structure ( S2 Fig ) , however , we found that the overall epitope map was similar to the epitope map using the entire database ( Fig 2C ) , albeit more sparse . Likewise , when we looked at sequence conservation among this subset of functionally-significant epitope residues ( S4 Fig ) , we saw similar results as when all epitope residues were considered ( Fig 3A ) . Finally , we compared pepitope values between serotypes ( inter-serotype ) with pepitope values within a serotype ( intra-serotype ) , to determine if the high intra-serotype antigenic variation observed by Kaetzelnick et al . could be reflected in higher intra-serotype pepitope values . However , as shown in S5 Fig , this was not the case . pepitope values within each serotype ( S5 Fig ) were significantly lower than pepitope values between serotypes ( Fig 3B ) , typically below 5% . Although we sought to determine whether type-specific epitopes might have more mismatches than complex epitopes , the number of significant mismatches within a serotype ( see Methods ) was too small to reliably determine average pepitope values . A more extensive analysis of intra-serotype variation was outside the scope of this study . Antibody responses to DENV have been most extensively characterized in humans and in mice . We sought to determine the degree to which there are systematic differences in epitope fine-specificity between mouse and human mAbs in the dataset . Towards that end , we calculated epitope propensity for mAbs from mouse and human separately ( Fig 4 ) . Our results show that the published mAbs from mice preferentially target DIII . Whereas , published human mAbs target epitopes on the DI/DII interface and the dimer interface . These findings suggest that there are systematic differences in epitope fine-specificity between published mouse and human mAbs—in particular that the type-specific mAbs in mice predominantly target the DIII region ( >75% ) , while published type-specific mAbs from humans target the DI/DII interface . It is important to note , however , that DIII remains a major target in the human Ab response as 30–50% of the type-specific human mAbs in our database target this region . It is important to note that many mouse mAbs were collected in the 1990’s and 2000’s while many human mAbs were collected more recently . As such , any systematic differences in the methodologies used to select hybridomas in these studies may confound an analysis of host-level differences in the antibody responses . Indeed , 85% of human mAbs in the database were initially selected based on binding to whole-virus preparations , compared to 32% of mouse mAbs; most mouse mAbs were selected based on binding to recombinant E proteins . However , even among mouse mAbs selected based on binding to whole virus , 70% ( 33 of 51 ) targeted DIII .
We hypothesized that the basis for the shift in cross-reactivity in the antibody response between primary and secondary infections could be found in the epitope-level fine-specificity of the respective antibody repertoires . We based our hypothesis on the theory of original antigenic sin that during secondary infection , pre-existing memory B cells specific to cross-reactive epitopes from the primary infection , would be selectively expanded to form the B cell and antibody repertoires in secondary infection [41 , 51 , 52] . Our comprehensive analysis of published epitopes showed that this was partially true . For example after secondary infection , we found that Abs to type-specific DIII epitopes , such as in the DIII-lateral ridge , are significantly diminished , while Abs to cross-reactive fusion loop epitopes are increased . However , we also found that epitopes , such as the DI/DII hinge region and the dimer interface , targeted by less cross-reactive Abs following primary infection were targeted by a higher proportion of cross-reactive Abs following secondary infection . It is important to note that an epitope region encompasses many overlapping epitopes , some of which are more conserved than others . During secondary infection it is these conserved epitopes within the epitope region that appear to be selected for , increasing the apparent cross-reactivity of that region compared to the primary infection . Overall , these results indicate that secondary DENV infections increase the number of cross-reactive Abs that target regions of the E protein recognized by both cross-reactive and type-specific Abs elicited by primary infections . This result is reminiscent of previous modeling work done in our group on a polyvalent malaria vaccine which showed that polyvalent formulations not only enhance the Ab response to shared or cross-reactive epitopes within the vaccine , but enhance the cross-reactivity of what were considered type-specific epitopes as well [53 , 54] . Either serially , as in the case of secondary DENV infection , or in parallel , as in the case with polyvalent vaccine formulations , this increased cross-reactivity results from a selective advantage of cross-reactive B cells over type-specific B cells . Whether similar effects on fine-specificity and cross-reactivity are present for polyvalent dengue vaccines remains to be seen . Furthermore , the immunological consequences of the shift in epitope-level fine-specificity in the antibody response to secondary DENV infection are still unclear . We analyzed sequence variation across DENV sequences and found that epitopes from type-specific antibodies have significantly more variation than epitopes from complex antibodies . Furthermore , we showed that the epitope-level measure , pepitope , successfully distinguished between type-specific and complex antibodies . In a landmark study , Katzelnick et al . [37] showed that even though DENV strains cluster into discrete serotypes in terms of sequence similarity along the E protein , they do not cluster by discrete serotypes in terms of antigenic similarity—in many cases sera raised against one serotype shows higher neutralization to a genetically distant strain from a different serotype , than to genetically similar strain from the original infecting serotype . We hypothesized that , among epitope residues identified in the database , there might be greater concordance between sequence similarity and antigenic similarity . However , this did not turn out to be the case . When we calculated pepitope values across the 47 DENV strains tested in that study , we found that intra-serotype pepitope values were relatively low ( <5% ) and far exceeded by inter-serotype pepitope values , for epitopes from both type-specific and complex antibodies . Previous studies have found that a small number of mutations can result in significant strain-specific differences in neutralization [27 , 39 , 55] , that sequence variation alone failed to predict strain-specific differences in neutralization [56] , and that genotype differences in viral conformational dynamics and epitope accessibility may be responsible [57] . As of yet , the structural and immunological basis for why DENV strains do not cluster antigenically into discrete serotypes is still largely unknown . Many previously published type-specific mouse mAbs predominantly target DIII of the E protein , while a lower proportion of characterized human mAbs ( 30–50% in the database ) target this region [30 , 31 , 58] . Furthermore , type-specific human Abs predominately target other E protein regions such as the DI/DII hinge region that have not been described for mouse mAbs . Our findings suggest this might be the result of systematic differences in epitope fine specificity between the mouse and human antibody responses but may also reflect differences in experimental methods used to produce mAbs from mice versus humans . We developed the DENV antibody database to provide a repository for activity and epitope information for DENV-specific mAbs in order to better characterize repertoire-level properties of the DENV antibody response . Here we provide an overview of the database and demonstrate how it can be used to analyze the relationship between epitope fine-specificity and serotype cross-reactivity in primary and secondary infections . We invite readers to explore the DENV antibody database ( http://denvabdb . bhsai . org ) , use it both as repository for storing and accessing information on DENV mAbs , and as a means to systematically analyze and characterize DENV antibody responses . | Dengue virus ( DENV ) infections are typically asymptomatic , but severe and potentially lethal disease symptoms , such as dengue hemorrhagic fever , are associated with secondary infections . This suggests that pre-existing immunity from primary infection plays a central role in DENV pathogenesis . In order to characterize the antibody response in primary and secondary infections , we assembled the Dengue Virus Antibody Database , a freely accessible online repository ( http://denvabdb . bhsai . org ) storing over 400 unique monoclonal dengue-specific antibodies annotated by their 1 ) origin and host immune history , 2 ) activity information against all four dengue serotypes , and 3 ) epitope mapping information . Here we demonstrate the utility of the database by carrying out a large-scale analysis to characterize shifts in epitope fine-specificity and serotype cross-reactivity in primary and secondary infections . In particular , we show how the antibody response in secondary infections displays a systematic shift towards increased serotype cross-reactivity by focusing on a subset of cross-reactive epitopes on the dengue E protein . Our findings suggest a mechanistic basis for this shift in epitope and serotype specificity and demonstrate how a detailed understanding of the antibody response can provide insight into the mechanisms of dengue pathogenesis . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"dengue",
"virus",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"split-decomposition",
"method",
"epitope",
"mapping",
"immunology",
"pathogens",
"microbiology",
"viruses",
"multiple",
"alignment",
"calcula... | 2017 | Dengue virus antibody database: Systematically linking serotype-specificity with epitope mapping in dengue virus |
Streptococcus pneumoniae is an opportunistic human bacterial pathogen that usually colonizes the upper respiratory tract , but the invasion and survival mechanism in respiratory epithelial cells remains elusive . Previously , we described that acidic stress-induced lysis ( ASIL ) and intracellular survival are controlled by ComE through a yet unknown activation mechanism under acidic conditions , which is independent of the ComD histidine kinase that activates this response regulator for competence development at pH 7 . 8 . Here , we demonstrate that the serine/threonine kinase StkP is essential for ASIL , and show that StkP phosphorylates ComE at Thr128 . Molecular dynamic simulations predicted that Thr128-phosphorylation induces conformational changes on ComE’s DNA-binding domain . Using nonphosphorylatable ( ComET128A ) and phosphomimetic ( ComET128E ) proteins , we confirmed that Thr128-phosphorylation increased the DNA-binding affinity of ComE . The non-phosphorylated form of ComE interacted more strongly with StkP than the phosphomimetic form at acidic pH , suggesting that pH facilitated crosstalk . To identify the ComE-regulated genes under acidic conditions , a comparative transcriptomic analysis was performed between the comET128A and wt strains , and differential expression of 104 genes involved in different cellular processes was detected , suggesting that the StkP/ComE pathway induced global changes in response to acidic stress . In the comET128A mutant , the repression of spxB and sodA correlated with decreased H2O2 production , whereas the reduced expression of murN correlated with an increased resistance to cell wall antibiotic-induced lysis , compatible with cell wall alterations . In the comET128A mutant , ASIL was blocked and acid tolerance response was higher compared to the wt strain . These phenotypes , accompanied with low H2O2 production , are likely responsible for the increased survival in pneumocytes of the comET128A mutant . We propose that the StkP/ComE pathway controls the stress response , thus affecting the intracellular survival of S . pneumoniae in pneumocytes , one of the first barriers that this pathogen must cross to establish an infection .
Sensing and transducing external ( or internal ) signals into an appropriate physiological response is part of a microorganism strategy to survive in a constantly changing environment . Signal transduction is mainly carried out by protein kinases , which autophosphorylate upon sensing stimuli and then catalyze the phosphorylation of a specific substrate that initiates an adaptive cellular response . In prokaryotes , signaling pathways are mainly mediated by two-component systems ( TCS ) consisting of sensor histidine kinases ( HK ) that phosphorylate response regulators ( RR ) on a receiver domain , thereby activating the effector domains of these regulators to induce a physiological event in bacterial cells . Generally , the RR effector domains bind regions of DNA that control gene expression in response to environmental changes [1 , 2] . Each particular HK presents a remarkable specificity for its cognate RR and is capable of identifying particular RRs . Eukaryotic-like Ser/Thr protein kinases ( STKs ) are also present in prokaryotes , where they play key roles in several cellular processes , including the central or secondary metabolism , developmental processes , cell division and virulence [3] . The major human bacterial pathogen Streptococcus pneumoniae ( S . pneumoniae or the pneumococcus ) encodes a single copy of StkP , a eukaryotic-like serine/threonine protein kinase gene [4] . StkP is a membrane protein composed of an N-terminal kinase domain facing the cytoplasm , a short transmembrane region , and an extracellular C-terminal region containing four PASTA ( Penicillin-binding protein and Ser/Thr protein kinase Associated ) domains [4–6] . Comparison of the global expression profile of the wild-type and ΔstkP strains has revealed that the transcription of genes involved in the cell wall metabolism , pyrimidine biosynthesis , DNA repair , iron uptake , and oxidative stress response are controlled by StkP , which explain why stkP mutations have pleiotropic effects [7] . It has also been described that StkP phosphorylates several target proteins , mainly on threonine residues , with PASTA domains being essential for kinase activity [8–10] . However , phosphorylation on serine residues seems to be independent of StkP [11] . Immunofluorescence microscopy of pneumococcal cells localized StkP to the cell division apparatus [12] , with phenotypic studies having demonstrated its impact on several cellular functions [13 , 14] . In fact , the stkP mutant displayed morphological and growth defects , cell division alterations , increased LytA-dependent autolysis ( induced by either antibiotics or growth at an alkaline pH of 7 . 8 ) , reduced tolerance to stress conditions ( including acidic stress ) , and pilus-mediated adherence in endothelial cells [4 , 7 , 11 , 15–17] . StkP is also essential for virulence , being necessary for lung infection and for invading and growing in the bloodstream of intranasally infected mice [4] . In S . pneumoniae , transient competence development ( ability to take up exogenous DNA ) in exponentially growing cells is considered a stress response to alkaline pH [2] , with its core regulatory circuit being controlled by the TCS ComDE . In this quorum sensing system , the membrane-integrated HK ComD senses the extracellular accumulation of a 17 amino acid competence stimulating peptide ( CSP ) [2] . Upon activation by a critical concentration of CSP , ComD phosphorylates the response regulator ComE at Asp58 [18] , which consequently initiates the transcription of comCDE , comAB , and comX ( a gene encoding an alternative sigma factor ) [19 , 20] . ComX turns on the transcription of genes whose products are involved in DNA binding , uptake , and recombination [21] . In this sense , the competence development is considered to be a type of stress response to alkaline pH [2] . It has been reported that StkP can also regulate competence at pH 7 . 8 . Cells lacking StkP do not develop natural competence [4] and show severely reduced CSP-induced competence [7 , 22] , despite having increased expression of many genes of the CSP-regulated competence regulon [7] . To invade tissues , S . pneumoniae must overcome a variety of stress situations , such as acidic pH , as a consequence of host inflammatory responses against the invading pathogen [23] . This characteristic local acidosis is caused by infiltration of neutrophils and activation of inflammatory cells , which leads to increased energy and oxygen demand , accelerated glucose consumption via glycolysis and thus increased lactic acid secretion [24–26] . For instance , pH values obtained from pleural fluids from patients with acute bacterial pneumonia caused by S . pneumoniae showed an acidic pH close to 6 . 80 [27] . Interestingly , the lowest pH value that S . pneumoniae has been shown to be tolerant to is around 4 . 4 in phagosomal vesicles during the first few minutes after phagocytosis [28] . Although S . pneumoniae is considered a typical extracellular pathogen , a transient intracellular life was described , suggesting that it can survive inside eukaryotic cells . S . pneumoniae can cross brain microvascular endothelial cells inside vesicles derived from early and/or late endosomes [29] [30] . It is well accepted that acidification is essential to endosome/lysosome maturation , with early endosomes having a pH in the 6 . 8–6 . 1 range , late endosomes in the 6 . 0–4 . 8 range , whereas lysosomal pH values can drop to 4 . 5 [31] . In the case of a putative endosomal survival , S . pneumoniae must survive acidic conditions . Martin-Galiano et al [32] described that S . pneumoniae is able to induce an acid tolerance response ( ATR ) mechanism . Previously , we also showed that F0 . F1-ATPase , a proton pump that controls intracellular pH , is relevant for ATR induction in S . pneumoniae . In addition , we demonstrated that the F0 . F1-ATPase and ATR are necessary for the intracellular survival of the pneumococcus in macrophages [33] . As part of the acidic stress response , we have reported that exposure of S . pneumoniae to acidic culture conditions triggers a lytic response by the major autolysin LytA . The acidic-stress induced lysis ( ASIL ) response is promoted by ComE and repressed by the CiaRH TCS . Despite requiring ComE , ASIL does not depend on CSP or ComD . Curiously , the comE gene is induced by acidic stress , but the competence-related ComX sigma factor , whose expression is regulated by ComE , does not participate in this signaling pathway [34] . We have also reported that ComDE and CiaRH control pneumococcal survival in pneumocytes in contrasting ways , CiaRH was essential for ATR and intracellular survival , whereas ComE repressed its activation . Moreover , ComE in a CSP-independent manner , was necessary for ASIL , whereas CiaRH protected against its induction by modulating LytA autolysin expression on the pneumococcal surface . These results suggest that both TCSs control the acidic stress response and establish either a survival or a suicidal response by independent pathways , either in acidified culture media or in pneumocyte cultures [33] . These findings indicate that ComE is activated under acidic conditions by an alternative signaling pathway that differs from the quorum sensing mechanism reported during competence development at alkaline pH . Alternatively , it was proposed that StkP is involved in competence at pH 7 . 8 , by the fact that cells lacking StkP do not develop natural competence [4 , 7 , 22] . StkP is also essential for virulence , to establish infections in the lung and for invading and growing in the bloodstream of intranasally infected mice [4] . The main aim of this work was to elucidate whether ComE is part of a novel activation pathway used by S . pneumoniae to induce the acidic stress response and to control its intracellular survival mechanism in pneumocytes . Here , we demonstrate that StkP controls ComE activation by phosphorylation of the Thr128 residue of the latter , increasing both its dimerization capacity and its DNA-binding affinity . Under acidic conditions , the StkP/ComE HK-independent pathway regulated 104 genes involved in different cellular processes , such as H2O2 production and oxidative stress tolerance . The StkP/ComE pathway is independent of the HK-dependent ComD/ComE system , which regulates more than 180 genes at pH 7 . 8 [35] . The participation in HK-independent and HK-dependent stress regulatory systems places ComE as a global regulator . This newly discovered StkP/ComE signaling pathway triggered the acidic stress response by inducing ASIL and inhibiting ATR and the intracellular survival of S . pneumoniae in pneumocytes , one of the first barriers that this pathogen must overcome to establish an infection .
We previously reported that ComE was required in the acidic stress induced lysis ( ASIL ) mechanism , which was independent of its cognate histidine kinase ComD at pH 6 . 0 [34] . This is in contrast to the ComD/quorum-sensing dependence on ComE activation at pH 7 . 8 necessary for competence development [19] . This initial observation led us to investigate whether other pneumococcal TCS-associated HKs could be activating ComE by a crosstalk mechanism , as described for other bacteria [36] . We hypothesized that if other HKs were involved in ComE activation by phosphorylation , the corresponding hk mutant should display alterations in the ASIL induction . Thus , the lytic phenotype was determined under acidic stress conditions for all the pneumococcal hk mutants that we had previously constructed in the background of the R801 strain by insertion-duplication mutagenesis ( S1 Table ) [33] . We observed that all the hk mutants showed the same lytic response as the parental R801strain , indicating that none of the tested HKs was involved in ASIL ( S2 Table ) . In addition , the comDF183X mutant was used because it lacks the HK domain due to a stop codon at residue 163 ( S2 Table ) and therefore it is unable to activate ComE [33] . This mutant was constructed to avoid putative alterations in the comCDE operon expression , and it showed the same ASIL phenotype than the ΔcomD mutant ( S2 Table ) , indicating that the truncated ComD protein expressed by the comDF183X mutant has not impact on the ASIL activation . In order to avoid a putative residual effect of comDF183X on the ComE activation , we determined ASIL in the double comDF183X hk mutants , which were constructed by transforming the comDF183X mutant with individual plasmids containing the different hk mutations . Despite the low transformability of the comDF183X mutant , all the double comDF183X hk mutants displayed the same ASIL phenotype as those obtained for each individual hk mutant ( S2 Table ) . Taken together , these results indicated that HKs were not responsible for activation of ComE and the resulting ASIL . Although ASIL is controlled by ComE activation without HK participation , this finding does not exclude the potential that ComE could have been phosphorylated by another phosphodonor at the Asp58 residue , typically the target of ComD phosphorylation [18] . Prokaryotic response regulators can be phosphorylated in vivo by acetyl-phosphate at the conserved aspartate residue of the receiver domain , resulting in similar activation to that exerted by the cognate HK [37] . In addition , phosphorylation crosstalk between HK and RR that belong to different TCSs has been reported [38] . Therefore , we first analyzed the possibility that Asp58 phosphorylation could be required for ASIL , and tested the comED58A mutant that encodes for the ComED58A protein , in which the phosphorylatable Asp58 residue is replaced by alanine [18 , 33] . The presence of this mutation was phenotypically corroborated under competence development conditions , confirming that CSP-induced competence was eliminated in the comED58A mutant ( S1 Fig and [18] ) . Like the wt R801 strain , the comED58A mutant autolysed under acidic conditions ( Fig 1A ) indicating that Asp58 phosphorylation is not necessary for ASIL . The comED58A phenotype is similar to the phenotype displayed by the hk mutants . Taken together , these results suggest ComE activation under acidic conditions is independent of both Asp58 phosphorylation and HK activity . Since it has been shown that StkP is involved in comCDE expression during competence development at pH 7 . 8 [4] , and that ComE is indispensable for the induction of ASIL [34] , we evaluated whether StkP participated in ASIL development . Thus , the ΔstkP mutant strain did not autolyze when cultured under acidic conditions ( Fig 1A ) , strongly suggesting that StkP is necessary for ASIL induction . To further confirm whether StkP kinase activity was required for ASIL , we also constructed the stkPK42R mutant , which encodes for StkP with reduced enzymatic activity , as previously described [39] . This reduced kinase activity produces multiple septa , peripheral peptidoglycan biosynthesis and elongated cells [8] , and these alterations were confirmed in our mutant and compared with the wt strain ( S2A and S2B Fig ) . As expected , the stkPK42R mutant strain showed ASIL with a degree of autolysis inferior to the wt strains but higher than the ΔstkP strain , indicating that the residual kinase activity in the stkPK42R mutant [39] was likely responsible for ASIL induction . These observations strongly suggest that StkP kinase activity is essential for ASIL activation . To exclude the possibility that the ASIL blockage observed in the ΔstkP mutant is a side effect due to hampered cell division and/or compromised cell wall structure [13] , we constructed another mutant that presents cell division alterations , such as the ΔmapZ mutant [9] . We verified these alterations by Van-Fl staining ( S2C Fig ) and found that the ΔmapZ mutant showed an ASIL phenotype similar to the wt strain ( S2D Fig ) , indicating that the ASIL effect showed by the ΔstkP mutant is independent of cell division alterations . We have previously demonstrated that ASIL is controlled by two independent signaling pathways , CiaRH and ComE . While CiaRH plays a protective function , the ComE acts by promoting ASIL [34] . To determine whether StkP participated in the CiaRH-controlled ASIL pathway , the ΔstkP and ΔciaR mutations were backcrossed , and the lytic phenotype of the resulting double mutants was analyzed . The ΔstkP ΔciaR strain showed enhanced autolysis , similar to the ΔciaR single mutant , demonstrating that the ciaR mutation had an epistatic effect on the non-autolytic stkP phenotype and that StkP activity did not participate in the signaling events of the CiaRH-controlled ASIL pathway ( Fig 1B ) . It has been previously described that the comDT233I mutation produces a hyperactive ComD HK , which in turn hyperphosphorylates ComE , activates comCDE transcription , and results in high intracellular levels of ComE [40] . Although our results suggested that ComD was not involved in ComE-mediated ASIL , we constructed the comDT233I mutant to artificially produce high levels of ComE and competence , as described [34 , 40] . The comDT233I mutant showed constitutive high levels of comE transcripts , at 1000-fold higher compared to the wt strain . This mutant displayed accelerated ASIL compared to the wt strain , however , we previously reported that ASIL was blocked in the double comDT233I ΔcomE double mutant [34] . These results demonstrated that ASIL induction was ComE-dependent and that increase expression of the latter leads to accelerated autolysis [34] . More importantly , when the stkP gene was disrupted in the comDT233I mutant , ASIL was also completely blocked ( Fig 1C ) , suggesting that StkP is essential for ASIL activation despite the high ComE levels expressed in the comDT233I mutant . It is known that comE is part of the comCDE operon , with the transcription of this operon initiated at the pcomC promoter ( bp 2035421–2035806 , S . pneumoniae R6 genome , NCBI reference: NC_003098 . 1 ) during competence development at alkaline pH [41] . To test whether pcomC was responsible for the increase in comE observed under acidic conditions , we constructed the pcomC-lacZ reporter fusion ( S1 Table ) , which was integrated via a single crossover event upstream of comC in a bgaA mutant ( deficient in β-galactosidase activity ) . When the bgaA mutant strain carrying the pcomC-lacZ fusion was incubated at pH 6 . 0 for 30 min . , a 1 . 7-fold increase in β-galactosidase activity was observed . No such increase was detected in a ΔcomE knocked out mutant ( S3 Fig ) . To further confirm this observation , the levels of comE transcript in the wt strain were determined by qPCR , which showed a 4-fold increase in cells exposed for 30 min at pH 6 . 0 ( Fig 1D ) . These results indicate that the increased number of comE transcripts was caused by acidic stress , with this activation being dependent on ComE . Similarly , the levels of comE transcript in the comED58A mutant were increased 4 . 5-fold after 30 min incubation at pH 6 . 0 . This result indicates that ComED58A was able to induce ASIL under acidic stress conditions ( Fig 1D ) . Consequently , these results suggest that ComE is activated by an alternative signaling pathway that does not require phosphorylation of Asp58 . We previously reported that induction of comE transcripts by acidic stress is a characteristic of the ComE-mediated pathway that controls ASIL [34] . To examine whether StkP could be involved in this pathway , we analyzed the comE transcript levels by qPCR in the ΔstkP and stkPK42R mutants constructed in a lytA background to avoid autolysis ( S1 Table ) . After incubation of ΔstkP cells for 30 min at acidic pH 6 . 0 , we observed a five-fold reduction in the levels of comE transcripts in the ΔstkP mutant ( Fig 1D ) . In contrast , the stkPK42R mutant showed a 2-fold decrease in comE transcripts , likely due to reduced kinase activity of StkPK42R . In addition , we observed that StkP was capable of controlling pcomC activation by acidic stress since increased β-galactosidase activity was observed from the pcomC-lacZ reporter fusion presence in the ΔstkP mutant in the bgaA background ( S3 Fig ) . Taken together , these results indicate that StkP kinase activity was required for comE induction under acidic conditions . In the comDT233I ΔstkP double mutant , the levels of comE transcript increased 50 times over those in the ΔstkP strain and 10 times over those in the wt strain ( Fig 1D ) . These results suggest that StkP kinase activity is required for full activation of ComE in order to induce ASIL under acidic conditions , regardless of the presence of high levels of ComE unnaturally induced by the ComDT233I kinase . Such observations led us to speculate that StkP activates ComE by an alternative mechanism other than the classical ComD HK-mediated Asp58 phosphorylation . We hypothesized that StkP controls ComE by a crosstalk phosphorylation event . To test this hypothesis , we carried out an in vitro phosphorylation assay using purified recombinant Hisx6-ComE fusion protein , in the presence or absence of purified recombinant GST-StkP . The phosphorylation reactions were examined by immunoblotting using either anti-phosphoserine or anti-phospho-threonine antibodies . No signal was detected with the anti-phosphoserine antibody , in contrast , positive reactions were detected with the anti-phosphothreonine antibody ( Fig 2A ) , with the phosphorylation reaction occurring at a molar ratio range of GST-StkP/Hisx6-ComE between 1:2 and 1:20 ( Fig 2B ) . The GST-GFP or Hisx6-GFP were included as controls of reaction specificity , and Hisx6-DivIVA was used as a positive control of a StkP target , as previously described [8 , 11] ( Fig 2A ) . We also evaluated the possibility that StkP could trigger ASIL by phosphorylating the major pneumococcal autolysin LytA . As overexpression of the full-length LytA protein was toxic in E . coli cells , the N-terminal region of LytA , which contains the catalytic domain was expressed instead fused to a Hisx6-tag ( N-LytA-Hisx6 ) [42] ) . Incubation with or without GST-StkP , resulted in no evidence of N-LytA-Hisx6 phosphorylation ( Fig 2A ) , suggesting that LytA was not phosphorylated by StkP , at least under the experimental conditions described here . To determine whether StkP-mediated ComE phosphorylation occurs in vivo and because response regulators are usually expressed at low level in bacteria , we constructed by insertion-duplication mutagenesis wt and ΔstkP strain derivatives that express ComE fused to the Hisx6-epitope tag at the C-terminus ( ComE-His6x ) to improve ComE detection . Cells were incubated at either pH 7 . 8 or pH 6 . 0 and ComE-Hisx6 was purified from protein lysates as described in Material and Methods and separated by SDS-PAGE . Phosphoproteins were detected by ProQ Diamond staining while total proteins were detected by SYPRO Ruby staining . We observed that phosphorylated form of ComE in wt cells grown at pH 7 . 8 , that increases 2 . 3 times when cells are grown at pH 6 . 0 ( Fig 2C ) . In contrast , ComE remained unphosphorylated in the ΔstkP mutant , confirming that ComE is phosphorylated by StkP in vivo . To determine the amino acid residues in ComE that are targeted for phosphorylation by StkP , we performed HPLC-MS/MS analysis of the in vitro StkP-phosphorylated ComE-His6x recombinant protein . A single amino acid was identified as a target for StkP-mediated phosphorylation in His6x-ComE , Thr128 , located inside the trypsin-digested 121IEQNIFYTK129 ComE peptide ( Fig 3A ) . To further confirm this observation , we created the non-phosphorylatable ComET128A-His6x recombinant mutant protein that remained unphosphorylated in the presence StkP in vitro ( Fig 3B ) . To evaluate the role of Thr128 phosphorylation on ComE activity in vivo , we constructed the comET128A mutant , which showed significantly blocked autolysis compared to the wt ( Fig 3C ) . Using the comET128A mutant , we produced the revertant comEA128T strain , which showed an ASIL phenotype similar to the wt strain ( Fig 3C ) . To further support the role of ComE Thr128 phosphorylation in ASIL activation , we attempted to replace Thr128 by Glu128 to construct the phosphomimetic comET128E protein , which is typically used to mimic the phosphorylated form of Thr residues [43] . In vitro , the phosphomimetic ComET128E-Hisx6 protein was hyper-activated , as demonstrated by EMSA assays ( see next ) , which may explain our inability to produce a viable comET128E mutant . These assays confirmed that Thr128 phosphorylation is essential for the StkP-mediated ComE activation that controls ASIL and that ComE hyper-activation is likely lethal to S . pneumoniae . StkP is involved in competence in response to stress conditions such as pH 7 . 8 [2 , 4 , 7] . Since , our results indicated that a signaling pathway that involves StkP and ComE controls autolysis in response to acidic stress at pH 6 , we investigated whether this crosstalk mechanism could also regulate competence development at pH 7 . 8 . We have previously described that the stkP mutant showed no competence development at pH 7 . 8 [4] . In contrast , the comDT233I mutant shows a hypercompetent phenotype , accompanied by constitutively high levels of ComE expression [34] . We observed that the competence phenotype of the comDT233I ΔstkP double mutant was similar to the comDT233I single mutant ( S1 Fig ) , indicating an epistatic effect of the comDT233I mutation on the ΔstkP mutation . We also observed that the competence of the non-phosphorylatable comET128A and revertant comEA128T mutants were similar to the wt strain ( S1 Fig ) . These observations suggest that StkP regulates competence at an early stage , which is independent of ComE Thr128 phosphorylation . Furthermore , StkP was not essential for competence once ComE was activated by ComD at pH 7 . 8 . In contrast , StkP is necessary to activate ComE and to trigger autolysis under acidic conditions . Response regulators are composed by a conserved receiver domain , which is phosphorylated on an aspartate residue by their cognate histidine kinase , and DNA-binding domains [44] . In ComE , the receiver domain corresponds to the first 130 residues [43] . Thr128 is located at the end of the α-5 region ( Asp114-Ser130 ) of the receiver domain of ComE , next to the α-4 region ( Ala94-Gln101 ) , and near the loop between α-4 and β-5 ( Val102-Leu105 ) that is involved in ComE dimerization and considered as a dimerization interface [43] ( Fig 4A ) . The proximity between the dimerization interface and the two phosphorylated residues ( Asp58 and Thr128 ) suggest a putative influence on the dimerization capacity of ComE , which was confirmed by in vitro dimerization assays using the phosphomimetic mutants . The ComED58E-His6x , ComET128E-His6x , and ComET128A-His6x mutants showed dimer steady-state levels , which were 70 , 72 and 2 . 9 times higher , respectively , than ComEwt-His6x ( reference level ) . In addition , when the ComEwt-His6x protein was incubated with StkP , the dimerization rate increased 8 . 9 times , whereas ComET128A-His6x showed only a 3 . 3-fold increase ( Fig 4B ) . The different dimerization capabilities found for ComEwt/StkP ( 8 . 9 times ) compared to the phosphomimetic ComET128E-His6x ( 70 times ) suggest that ComEwt-His6x is partially phosphorylated by StkP ( ~12% ) . These observations suggest that Thr128 phosphorylation modifies ComE in a manner that strongly affects its dimerization interface , which is a condition sine qua non for the response regulator activation . In order to determine how Thr128 phosphorylation affected ComE's conformation , we performed molecular dynamic simulations at 40–150 ns comparing ComEwt ( PDB ID: 4CBV , [43] ) and the in silico phosphomimetic ComET128E-His6x protein ( S4 Fig , video ) . The simulations clearly indicated that 3 loops spanning the DNA-binding domain ( Lys218-Asn219-Leu220 , Thr164-Gly165-Val166-Ser167-His168 , and Ser200-Pro201-His202-Lys203 ) presented different dynamics in the ComET128 mutant compared to ComEwt ( Fig 4B and accompanying video shown in S4 Fig ) . ComE is a member of the AgrA/LytTR family of bacterial response regulators , which present certain structural homologies . Coincidently , two of these three loops have been described for the AgrA RR in S . aureus [45] , as key residues that determine the DNA binding affinity for promoters in the phosphorylated form of AgrA ( Fig 4B ) . ComE also revealed positively charged or polar residues ( His168 and Lys169 in loop 1; His202 and Lys203 in loop 2; Arg217 and Lys218 in loop 3 ) , which are shown in AgrA to have a direct contact with DNA bases [45] . These results suggest that after Thr128 phosphorylation ComE may undergo conformational changes . Consistent with this notion , limited proteolysis assays revealed structural differences between ComE-His6x and ComET128E-His6x ( S5 Fig ) . Treatment with trypsin showed more contrasting proteolytic patterns than proteinase K treatment . These findings confirmed that Thr128 phosphorylation causes evident changes in ComE conformation , likely in the DNA-binding domain of ComE , that may modify its DNA-binding affinity , and we explored this possibility with electrophoretic mobility shift assays ( EMSAs ) . During competence development at alkaline pH , Asp58 phosphorylation by ComD results in increased binding of ComE to pcomC and transcriptional activation of the comCDE operon [2 , 46 , 47] . We have observed that at acidic pH , pcomC activation and induction of comE transcription depended both on ComE and on StkP ( Fig 1D and S3 Fig ) suggesting that Thr128 phosphorylation by StkP influences the binding of ComE to pcomC . Electrophoretic mobility shift assays ( EMSAs ) proved that the phosphomimetic ComET128E-His6x protein bound pcomC 5-fold stronger than ComEwt-His6x ( Kd 74 nM vs Kd 371 nM , respectively ) . The DNA binding affinity of the non-phosphorylatable ComET128A-His6x mutant was unaffected ( Kd 375 nM ) whereas ComED58E-His6x affinity for pcomC was 17- fold greater that ComEwt-His6x ( Fig 5 and Table 1 ) . Curiously , when ComEwt-His6x was pre-incubated with StkP in phosphorylation buffer at pH 7 . 8 no binding to pcomC was observed ( S6 Fig , Table 1 ) . As ASIL is regulated by StkP-mediated phosphorylation of ComE Thr128 residue under acidic conditions , we tested if ComE DNA-binding affinity could be affected by pH . When ComEwt-His6x was previously incubated with StkP at pH 6 . 0 , its affinity for pcomC increased ( Kd 64 nM ) and was similar to that shown by ComET128E-His6x ( S6 Fig , Table 1 ) . To determine which of these contrasting effects actually depended on StkP phosphorylation , similar assays were performed with an inactive StkP enzyme ( StkPK42M ) [8] and ComEwt-His6x . Binding to pcomC was comparable in ComEwt-His6x pre-treated with StkPK42M at pH 6 . 0 and untreated ComEwt ( Kd 300 nM vs Kd 371 nM ) but was still absent after incubation at pH 7 . 8 ( S7 Fig , Table 1 ) . These results indicate that StkP phosphorylation at pH 6 . 0 underlied the enhanced pcomC-binding affinity of ComE , but not the blocking of ComE-pcomC interaction observed at pH 7 . 8 , which suggests that at a slightly alkaline pH , StkP makes a complex with ComE masking its DNA-binding sites . The DNA-binding affinity of the phosphomimetic ComET128E-His6x and the non-phosphorylatable ComET128A-His6x proteins were not affected when preincubated with StkP at either pH 6 . 0 or pH 7 . 8 ( S7 Fig , Table 1 ) , indicating that Thr128 mediated the observed StkP effects on ComE: ( 1 ) at pH 6 . 0 , Thr128 phosphorylation by StkP increases ComE DNA binding affinity; ( 2 ) at pH 7 . 8 Thr128 mediates the StkP-ComE interaction that blocks DNA binding . Thus , these experiments indicate that pH modulates the interplay between StkP and ComE . To test for a putative protein-protein interaction between StkP and ComE , a sandwich fluorescence-linked immunosorbent assay ( FLISA ) was utilized , in which the StkP-coated surface of a microtiter plate was incubated with increasing amounts of His-tagged ComE at either pH 6 . 0 or pH 7 . 8 . This assay clearly showed that acidic pH augmented the number of binding sites between ComE and StkP , as reflected by a 3-fold increase in Fmax ( maximum fluorescence when binding is saturated ) at pH 6 . 0 compared to pH 7 . 8 ( Table 2 , S8 Fig ) . The interaction between ComEwt and the StkPK42M mutant revealed the same Fmax as the ComE/StkP at different pH values , indicating that kinase activity did not alter the saturation of this protein complex . However , for the StkPK42M mutant , the K1/2 values were lower , indicating that StkPK42M bound more tightly to ComE at any pH value . In comparison with ComEwt , the interaction between non-phosphorylatable ComET128A mutant and StkP or StkPK42M was 3-fold stronger and was not affected by pH . In contrast , the phosphomimetic ComET128E protein produced a 3-fold increment in K1/2 at pH 7 . 8 , which further raised to 9-fold at pH 6 . 0 , demonstrating that the phosphorylated form of ComE had a lower affinity for StkP ( Table 2 , S8 Fig ) . These data confirm that the StkP/ComE interactions are mediated by ComE residue Thr128 and favored by acidic conditions , which may facilitate ComE phosphorylation . To understand the effect of the StkP/ComE signaling pathway on pneumococcal physiology , we compared the transcriptomes of the comET128A mutant and wt by RNAseq analysis . Three replicates of each strain strains were grown in ABM ( pH 6 . 0 ) for 1 hr at the exponential growth phase ( OD620nm 0 . 3 ) and analyzed . In total , the differential expression of 104 genes was detected , 51 were down-regulated genes and 53 were up-regulated , considering relevant genes to be those with expressions higher than 2 fold and p values <0 . 05 ( Fig 6A ) . The full list of these genes is shown ( S3 Table ) . Based on this differential gene expression analysis , we observed that the StkP/ComE pathway affected , directly or indirectly , the expression of genes involved in oxidative stress , and the purine/pyrimidine , amino-acid and central metabolisms , as well as the ribosomal and translation structures , metabolite transport , molecular chaperones and cell wall biosynthesis , among others ( Fig 6B ) . The list of genes regulated by Thr128-phosphorylated ComE indicated that this new signaling pathway induces global changes in the pneumococcal transcriptome , such as the physiological response to acidic stress . Bioinformatic analysis of the promoter regions ( 240 bp upstream of the start codon ) of 22 ComE-regulated genes obtained from RNAseq assays predicted a putative DNA binding motif ( S9A Fig ) . Martin et al [47] described a potential ComED58~P binding site ( CEbs , 32 bp ) in the comC promoter constituted by two repeats ( DR1 and DR2 ) separated by 12 bp . In this report , we established that the putative ComET128~P binding site ( 26 bp ) only partially overlaps with theses repeats suggesting a different consensus binding sequence to that described for ComED58~P ( S9B Fig ) . Using RT-qPCR we confirmed decreased expression of oxidative stress genes spxB , sodA [48] and tpxD [49] in comET128A mutant compared to wt ( Fig 6C ) . The spxB gene encodes the pyruvate oxidase that produces H2O2 from O2 , sodA encodes the superoxide dismutase that produces H2O2 from superoxide , and tpxD encodes the thiol peroxidase that catalyzes the H2O2 oxidation and contributes to the oxidative stress response . In the comET128A mutant , we found that the spxB , sodA , and tpxD transcripts were downregulated 18 , 2 . 8 and 2 . 7 times , respectively ( Fig 6C ) . These findings were also corroborated by H2O2 production and H2O2 susceptibility assays . The comET128A , ΔcomE , and ΔstkP mutants showed a 4-fold decrease in their H2O2 production compared to the wt and comEA128Tstrains ( Fig 7A ) , which is likely caused by the reduced expression of spxB and sodA protein products . In addition , we observed a 10-fold reduction in the susceptibility to H2O2 by the comET128A , ΔcomE , and ΔstkP mutants compared to the wt strain ( Fig 7B ) , likely due to reduced expression of the TpxD peroxidase . These findings support the notion that the StkP/ComE pathway is essential for the control of H2O2 production and for H2O2 tolerance . Although RNAseq analysis showed that the murN gene was overexpressed in the comET128A mutant , its expression by qPCR was actually found to be 4-fold lower than in wt in three independent assays ( Fig 6C ) , suggesting a typical case of false positive that is commonly found in RNAseq studies . Regarding the physiological impact of the murN mutation , Filipe et al [50] described that a murMN mutant had cell wall alterations and presented increased susceptibility to lysis when exposed to cell wall antibiotics . To test whether an altered murN expression in the comET128A mutant could modify the susceptibility to cell wall antibiotics , we determined the MIC values of the comET128A and wt strains in the presence of either fosfomycin , vancomycin , penicillin , cefotaxime , cefazolin , or piperacillin . The fosfomycin MIC in the comET128A ( 170 μg/ml ) was higher than the wt strain ( 50 μg/ml ) , whereas the MICs for vancomycin , penicillin , cefotaxime , cefazolin , and piperacillin were similar between these strains . The typical lytic effect of fosfomycin ( 50 μg/ml , 1xMIC; Fig 7C ) and vancomycin ( 0 . 4 μg/ml , 1xMIC; Fig 7D ) on the wt strain was inhibited in the comET128A strain . The diminished susceptibility to cell wall antibiotics in the comET128A strain suggests cell wall alterations consistent with the ASIL repression showed by this mutant . We previously described that ComE is involved in the acidic stress response and in the pneumococcal intracellular survival mechanism in pneumocytes [33] . Here , we demonstrate that StkP phosphorylates ComE , and in order to determine whether this crosstalk affects the pneumococcal survival , we measured the intracellular survival capacities in A549 pneumocyte cells of the ΔstkP , stkPK42R ( reduced kinase activity ) , ΔcomE , comET128A and comEA128T ( revertant ) strains compared to the wt in A549 pneumocyte cells [33] . Mutations in either the stkP or comE genes conferred increased survival compared to comEA128T or wt ( Fig 8A ) , indicating that the StkP/ComE pathway controlled pneumococcal survival in pneumocytes . To discriminate whether this phenotype could result from increased ATR or decreased ASIL , we tested the ΔlytA mutant , which lacks autolysin and presented a blocked ASIL [33] , but its intracellular survival was similar to the wt strain ( Fig 8A ) . Thus , a blocked ASIL is not enough to increase intracellular survival of S . pneumoniae in pneumocytes . Consequently , the increased survival showed by either the ΔstkP , ΔcomE , or comET128A mutants ( Fig 8A ) is likely due to higher ATR capacity . To test this hypothesis , we determined the ATR phenotype of the comET128A mutant , but in a lytA deficient background in order to discard residual autolysis . As expected , ATR of the ΔlytA strain increased 2-fold at pH 6 . 0 compared with cells cultured at pH 7 . 8 , whereas the comET128A ΔlytA cells showed a 20-fold increase under the same conditions ( Fig 8B ) , supporting the notion that increased ATR explains the increased survival rate displayed by the comET128A mutant in pneumocytes .
Two-component systems ( TCSs ) represent one of the most important mechanisms of gene regulation in bacteria . Alternatively , eukaryotic-like serine-threonine kinases ( STKs ) constitute another signaling mechanism that bacteria utilize to regulate different cellular functions , such as stress response and pathogenesis . STKs are more promiscuous than the TCS-associated kinases and can phosphorylate different protein substrates producing pleiotropic effects [16 , 51 , 52] . STKs are also able to interact with TCSs by direct phosphorylation of RRs , as reviewed in [3 , 53] . STK-mediated RR activation takes place on either serine or threonine residues , instead of aspartate , which is the typical residue target for HK phosphorylation . STK-mediated phosphorylation on DNA-binding domains of RR have been reported , as described for GraR in S . aureus [54] , YvcK in Listeria monocytogenes [55] and RitR in S . pneumoniae [56] . STKs may also phosphorylate on receiver domains , as observed for CovR in S . pyogenes [57] , WalR in B . subtillis [58] , DosR in M . tuberculosis [59] , or in both domains , as demonstrated for VraR in S . aureus [60] . ComE is the most studied RR in S . pneumoniae and belongs to the AlgR/AgrA/LytTR transcription factor subfamily , showing a typical receiver domain and a DNA-binding ( or LytTR ) domain . When phosphorylated by the ComD histidine kinase at the Asp58 residue , ComE undergoes conformational changes that increase their DNA affinity and modify transcription regulation of competence genes by binding to their promoter regions [2 , 47 , 61] . In the present work , we show that S . pneumoniae utilizes an alternative signal transduction pathway to control acidic stress response ( ASIL and ATR ) , oxidative stress , cell wall biosynthesis , and intracellular survival in pneumocytes . Under acidic conditions , a phosphorylation crosstalk between StkP and ComE involving phosphorylation at Thr128 in the receiver domain resulted in activation of this RR . Using the crystal structure of ComE [43] , a molecular dynamic simulation of ComE permitted a comparison with the phosphomimetic ComET128E protein , predicting that the Thr128 phosphorylation produces structural changes in the DNA-binding domain . These putative conformational changes were confirmed by limited proteolysis assays that revealed differences between ComE and ComET128E . In this sense , we observed that phosphorylation at either Asp58 or Thr128 increases the dimerization and DNA-binding capacity of ComE . These results were coincident with the activation model proposed by Boudes et al [43] , where the most plausible activation mechanism of ComE is first a phosphorylation reaction to induce its dimerization , which occurs at the canonical receiver domain of ComE , followed by binding to DNA via the LytTR domain . It remains to be elucidated how phosphorylation at alternative sites in the receiver domain ( Asp58 or Thr128 ) modifies the DNA-binding domain of ComE . Because ComE Thr128 activation is independent of CSP/ComD activation by a quorum sensing mechanism , this RR requires alternative factors to act as a sensor and/or an environmental signal to trigger an adaptive stress response . We propose that StkP senses an alternative environmental signal , acidic pH . Our results are consistent with such notion: the level of comE transcripts induced under acidic conditions is the indicator of ComE activation due to the comCDE operon is autoregulated . An additional aspect to consider in the StkP/ComE crosstalk phosphorylation is the effect of pH on protein-protein interactions . We observed that the non-phosphorylated forms of both proteins show strong interaction at acidic pH . Once StkP auto-phosphorylates , it becomes metastable complex until it dissociates from the phosphorylated ComE form . Such cycle is favored at pH 6 . 0 and evidence for the outcome of such cycle is shown by the fact that phosphorylation of ComE at Thr128 by StkP prevents further interaction between these two proteins . We propose the following cycle for StkP/ComE interaction: StkP+ComE→StkP/ComE→StkP‑P/ComE→StkP+ComE‑P ( stablecomplex ) ( transientcomplex ) Pneumococcal H2O2 production is one of the most significant among bacterial pathogens , and S . pneumoniae utilizes this intermediate metabolite to compete with the respiratory tract microbiota and to produce cytotoxic effects on the host tissues [62–64] . In this investigation , the transcriptome analysis of the comET128A mutant revealed a marked decrease in the expression of spxB and sodA when cells were grown under acidic conditions . These results correlated with very low H2O2 production by the comET128A mutant associated with low expression of SpxB and SodA . Considering that H2O2 is toxic for eukaryotic cells , we propose that reduction H2O2 production in the comET128A mutant facilitates its intracellular survival in pneumocytes . S . pneumoniae generates hyper-virulent mutants with defective spxB during infection [65] , supporting the hypothesis that the H2O2 levels are controlled during pneumococcal pathogenesis . Because S . pneumoniae lacks catalase , and H2O2 overproduction must be controlled for this pathogen to survive , this pathogen induces an oxidative stress resistance that is induced by endogenous H2O2 [66] . In this sense , the transcriptome analyses of the comET128A mutant revealed a decreased expression of tpxD , which encodes the thiol peroxidase . These findings correlate with the increased H2O2 susceptibility by the comET128A mutant . Similarly , a previous study showed that a tpxD ( or psaD ) mutant eliminated the H2O2-mediated response to high H2O2 levels [67] . Previously , a microarray analysis of the stkP mutant revealed that tpxD expression is repressed , which is correlated with a low H2O2 resistance of this mutant , but the putative regulatory mechanism was not mentioned [7] . In the present study , we have shown for the first time that the StkP/ComE pathway controls oxidative stress resistance and H2O2 production under acidic conditions , which are probably responsible for the intracellular survival of S . pneumoniae in pneumocytes . In a previous study , we proposed that ASIL may be activated under acidic conditions by a translocation of LytA from an intracellular to an extracellular compartment probably due to cell wall alterations by an unknown ComE-dependent mechanism [68] . Here , the comET128A mutant displayed a decreased expression of murN , which encodes one of the first enzymes involved in the cell wall biosynthesis of S . pneumoniae [69] . Filipe et al [70] described that the murMN mutant showed an increased susceptibility to lysis when murMN cells were exposed to cell wall antibiotics , such as fosfomycin and vancomycin , which are involved in the inhibition of both the early and late stages of cell wall biosynthesis , respectively . Following this line of thinking , an altered expression of murN in the comET128A mutant should cause a misbalance in the peptidoglycan biosynthesis and modify susceptibility to cell wall antibiotics . Accordingly , we observed that this mutant had an increased MIC of fosfomycin compared with wt , as well as a greater tolerance to autolysis induced by fosfomycin or vancomycin . The putative cell wall alterations indicated by antibiotic susceptibility tests may explain the autolysis inhibition shown by the comET128A mutant under acidic stress , which probably interfered with LytA activation . Work is in progress to try to determine the nature of such cell wall alterations . We also demonstrated that Thr128 phosphorylation is not involved in competence . Regarding this topic , Guiral et al [71] described a phenomenon of lysis of non-competent cells triggered by competent cells , named allolysis , which involves bacteriocins and the autolysins LytA , LytC , and CbpD . Allolysis is considered to be a competence-induced mechanism of predation of non-competent cells that contributes to virulence by releasing pneumolysin [72] . Here , we show that the StkP/ComE signaling pathway can also trigger autolysis of noncompetent cells in acidic biological niches , such as inflammatory foci or endosomal compartments . This phenomenon occurred without the activation of a quorum sensing mechanism , a situation that allows bacterial cells to lyse even under low population density conditions . StkP and ComE have already been shown to be involved in pneumococcal pathogenesis in different studies using animal models , with StkP appearing to be involved in bacterial survival in vivo [4 , 73] . On the other hand , ComE-mediated competence for DNA transformation has been also associated with virulence [74 , 75] . As mentioned above , pneumolysin release by competence-mediated autolysis was considered to be essential for pneumococcal pathogenesis [71] . Concerning the impact of StkP/ComE pathway regulation on pneumococcal pathogenesis , we propose that two different scenarios should be considered . In extracellular niches , a subpopulation of pneumococci exposed to acidic stress may cause tissue damage by overproduction of H2O2 and induction of ASIL to release pneumolysin , with this suicidal situation being promoted by StkP-mediated phosphorylation of ComE . On the other hand , the Thr128-nonphosphorylated form of ComE might facilitate pneumococcal survival at either the extracellular or the intracellular level in host tissues . We focused our attention on the first barrier that this pathogen must cross to establish an infection , and we hypothesized that intracellular survival in pneumocytes should be important for S . pneumoniae . The ΔstkP , ΔcomE , and comET128A mutants were tested in the pneumococcal infection model in A549 pneumocytes , and they revealed an increased survival compared with wt . Thus , we conclude that this survival could have been caused by increasing their capacity of ATR , decreasing the H2O2 production and modifying the cell wall biosynthesis to repress ASIL ( Fig 9 ) . Establishing how the balance between H2O2 resistance mechanism and H2O2 production affects intracellular survival is beyond the scope of this report , but this work is in progress . Finally , we propose that the StkP/ComE pathway is relevant in the genetic regulation of physiological adaptation to environmental stress , which is necessary for pneumococcal survival in pneumocytes . This is one of the first steps in the pathogenic process that S . pneumoniae must overcome to produce infection .
All strains , plasmids , and oligonucleotides used in this study , as well as cloning and mutagenesis procedures , are listed in the supplementary material ( S1 Table ) . The growth conditions and stock preparation for the pneumococcal and Escherichia coli strains have been reported elsewhere [34] , and the transformation assays have also been previously described [76 , 77] . ASIL was performed as described previously [33] . Firstly , bacterial cells were grown in Todd-Hewitt/yeast extract medium . When cultures reached OD600nm ~0 . 3 , cells were centrifuged at 10 , 000 g for 5 min , the pellet was resuspended in ABM pH 6 . 0 and cultures were re-incubated at 37°C . Autolysis was measured as a change in OD600nm at different time points over 6 h . ATR was performed as described previously [24] . For non-acid-induced conditions , bacterial cells were first grown in THYE ( pH 7 . 8 ) at 37°C , and when cultures reached OD600nm~ 0 . 3 , 100 μl aliquots were taken and added to 900 μl of THYE ( pH 4 . 4 ) and incubated for 2 h at 37°C . Then , serial dilutions were made in THYE ( pH 7 . 8 ) and plated onto 5% of sheep blood tryptic-soy agar ( TSA ) plates . After 24 h of incubation at 37°C , colonies were counted to determine the number of survivors , with the total CFU being obtained by plating serial dilutions of cells grown THYE pH 7 . 8 onto 5% sheep blood TSA , made just before cells were switched to pH 4 . 4 . In parallel , to determine survival under acidic-induced conditions , bacterial cells were grown in THYE ( pH 7 . 8 ) until OD600nm ~ 0 . 3 , centrifuged at 10 , 000 g for 5 min , resuspended in THYE ( pH 6 . 0 ) and incubated for 2 h at 37°C . Culture aliquots were taken and serially diluted in THYE pH 7 . 8 for total cell counting , while other aliquots were diluted ten times in THYE ( pH 4 . 4 ) and incubated for 2 h at 37°C to determine the survival percentage as described above . For both assays ( acid-induced and non acid-induced conditions ) , this was calculated by dividing the number of survivors at pH 4 . 4 by the number of total cells at time zero ( before incubation at pH 4 . 4 ) . For the ASIL and ATR assays , data were expressed as the mean percentage ± standard deviation ( SD ) of independent experiments performed in triplicate . The A549 cell line ( human lung epithelial carcinoma , pneumocytes type II; ATCC CCL-185 ) was cultured at 37°C , 5% CO2 in Dulbecco’s modified Eagle medium ( DMEM ) with 4 . 5 g/l of glucose and 10% of heat-inactivated fetal bovine serum ( FBS ) ( Gibco BRL , Gaithersburg , Md . ) . Fully confluent A549 cells were split once every two or three days via trypsin/EDTA treatment and diluted in fresh media before being cultivated in Filter cap cell flasks of 75 cm2 ( Greiner Bio-one no . 658175 ) until passage 6 . In vitro phosphorylation was carried out with 0 . 5 μg of purified recombinant substrate protein and 0 . 5 μg of purified GST-StkP in 30 μl of kinase buffer ( 50 mM Tris-HCl , 5 mM MgCl2 , 100 μM ATP , 1mM DTT , pH 7 . 5 ) . The reaction was started by the addition of ATP and stopped after 60 min of incubation at 37°C by the addition of 5x Laemmli SDS sample buffer . Samples were separated by standard Tris-glycine-SDS polyacrylamide gel electrophoresis ( PAGE ) gels and electroblotted onto a nitrocellulose membrane . Phosphorylated proteins were detected with an anti-phosphothreonine polyclonal antibody ( 1∶1 , 000; Cell Signaling ) and a goat anti-rabbit immunoglobulin G secondary antibody conjugated to horseradish peroxidase ( 1∶2 , 500; Invitrogen ) . Detection was performed with an enhanced chemiluminescence substrate ( SuperSignal West Pico Chemiluminescent Substrate; Pierce ) and Hyperfilm CL film ( GE ) using exposures of between 1 and 10 min . The pRSET-divIVAspn plasmid was generously provided by Dr . Orietta Massidda ( Università degli studi di Cagliari , Italy ) [78] . The RC838 ( comE-his ) and RC839 ( ΔstkP comE-his ) strains ( S1 Table ) were grown in 2 l of ABM ( pH 7 . 8 ) at 37°C until OD600nm 0 . 3 . Half of the cultures ( 1 l ) were centrifuged for 10 min at 5 , 000 x g , snap-frozen in liquid-air and stored at -80°C . The remaining 1 l was centrifuged as before , resuspended in ABM ( pH 6 . 0 ) , incubated at 37°C for 10 min , and finally centrifuged for 10 min at 5 , 000 x g , snap-frozen in liquid-air and stored at -80°C . Cell pellets were thawed in ice and resuspended in 10 ml of L8 buffer ( 100 mM NaH2PO4/10 mM Tris·HCl , pH 8 . 0/150 mM NaCl/20 mM imidazole/20mM PMSF/8 M urea ) supplemented with MS-SAFE protease/phosphatase inhibitor cocktail ( Sigma-Aldrich ) and lysed by stirring for 1 h followed by sonication . The lysate was cleared by centrifugation at 15 , 000 g for 20 min , and the supernatant was added to a column packed with to 0 . 5 ml of Ni-NTA resin ( Qiagen ) equilibrated in L8 buffer . The column was washed sequentially with 10 ml of LX buffer ( L buffer with X = 8 , 6 , 4 , 2 , and 0 M urea ) , and bound ComE-His6x protein was eluted with 2 ml of L0 buffer containing 500 mM imidazole . Protein samples ( approximately 0 . 5 μg ComE-His ) were separated by SDS-PAGE , and the gels were stained with ProQ Diamond ( Invitrogen ) to detect phosphorylated ComE-His , followed by SYPRO Ruby ( Invitrogen ) total protein staining . Gels were imaged under fluorescence mode in a Typhoon FLA 9500 scanner ( GE ) and protein bands were quantified using ImageQuant software ( GE ) . Identification of the phosphorylation site was carried out by nano-LC-MS/MS analysis as previously described [79] . Protein bands were in-gel-digested overnight with sequencing grade trypsin ( Promega ) at 37°C , desalted using micro-reverse phase columns ( C18 Omix tips , Varian ) , vacuum dried and resuspended in 0 . 1% formic acid ( v/v ) in water . Tryptic peptides were injected into a nano-HPLC system ( Proxeon Easy nLC , Thermo ) fitted with a trap column ( Easy-column C18 2 cm x 100 um ID ) . Posteriorly , the samples were separated on a reverse phase nano-column ( Easy-Column C18 10 cm x 75 um ID; Thermo ) using a linear gradient of acetonitrile 0 . 1% formic acid ( 0–45% in 70 min ) at a flow rate of 400 nL/min . Mass analysis was performed using a linear ion trap mass spectrometer ( LTQ Velos , Thermo ) in a data-dependent mode ( full scan followed by MS/MS of the top 5 peaks ) [79] . Raw data was analyzed using the Proteome Discoverer software package ( v . 1 . 3 . 0 . 339 , Thermo ) , and Sequest search engine , with the following parameters: enzyme: trypsin; maximum missed cleavage: 2; precursor mass tolerance: 1 Da; fragment mass tolerance: 0 . 8 Da; Ser/Thr/Tyr phosphorylation and methionine oxidation as dynamic modifications . Searches were performed using a Streptococcus pneumoniae database downloaded from UniProt ( 17/5/2017 ) and including the His tag-ComE sequence . For phosphosite localization , the phosphoRS algorithm was used and the spectra of phosphorylated peptides were manually inspected to corroborate the phosphosite assignment [80] . The promoter region of comCDE ( 255 bp; pcomC ) was PCR-amplified using the 5’-Cy5 labeled oligonucleotides NGEP516 and NGEP517 ( IDT ) and purified using the QIAquick PCR Purification Kit ( QIAGEN ) . DNA-binding assays were performed in a total volume of 10 μl containing 50 mM NaCl , 50 mM Tris/HCl pH 7 . 5 , 5% ( v/v ) glycerol , 7 . 5 nM Cy5-labeled PCR fragments , 1 mM MgCl2 , 0 . 15 mg Poly ( dI-dC ) ( as the non-specific competitor ) , and varying concentrations of untreated or StkP-treated ComE proteins . In the latter case , phosphorylation of ComE-Hisx6 by StkP was carried out with an equimolar amount of StkP in kinase buffer ( 50 mM Bis-Tris propane , 5 mM MgCl2 , 1mM DTT , 0 . 1 μM ATP , pH 7 . 8 or pH 6 . 0 ) for 60 min at 37°C . Protein-DNA binding reactions were incubated at room temperature for 30 min , and “frozen” with an equal volume of 2X Stop Solution [40% ( v/v ) Triethylene Glycol , 10 mM Tris , pH 7 . 5] . DNA-protein complexes were resolved by electrophoresis in native Tris-Borate-EDTA polyacrylamide gels [10% ( w/v ) ] containing 30% Triethylene Glycol . Gels were run at 4°C for 120 min at constant voltage ( 25 V cm-1 ) in a 0 . 5X TBE buffer and scanned in a Typhoon FLA 9500 biomolecular imager ( GE ) under fluorescence mode . Free and protein-bound DNA were quantified using ImageQuant ( GE ) . The fraction of DNA bound ( FB ) at each ComE concentration was fit with a standard binding isotherm using Kaleidagraph ( Synergy Software ) , according to the equation: FB = [ComE]/ ( Kd +[ComE] ) , where Kd is the apparent equilibrium dissociation constant and reflects the protein concentration required to shift 50% of the labeled DNA fragment . MD simulations were carried out with the NAMD program [81] , using the CHARMM27 force field [82] . Spherical boundary conditions and a non-bonded cut-off of 12 . 0 Å with a switching function of 10 . 0 Å were used . All systems were submitted to structural minimization in vacuum , and then embed in a water sphere for the MD . The temperature was set to 310 K by a Langevin thermostat . MD simulations were run for 40 ns with an integration step of 2 fs . Analysis of the trajectories was performed using VMD software [83] . Crystal structure images were analyzed using PyMOL [84] . StkP and ComE binding interactions were assessed by a sandwich fluorescence-linked immunosorbent assay ( FLISA ) in black 96-well high binding capacity microplates with clear flat-bottoms ( Corning #3601 ) . Each well was filled with 50 μL of 10 μg/ml GST-StkP ( 500 ng of GST-StkP ) dissolved in a 0 . 1 M coating buffer ( 0 . 1 M NaHCO3/Na2CO3 , pH 9 . 4 ) and incubated overnight at 4°C to allow protein adsorption . Wells were rinsed five times with Tris-buffered saline , 0 . 05% Tween 20 , pH 7 . 4 ( TBS-T ) and the reactive sites were blocked with 2% w/v bovine serum albumin dissolved in TBS-T for 2 h at room temperature . Wells were washed three times with TBS-T . Different amounts of ComE-Hisx6 ( 200–1200 ng ) were dissolved in 50 μL of 50 mM Bis-Tris-Propane-HCl , 1 mM MgCl2 , pH 7 . 8 , or in the same buffer but at pH 6 . 0 , and added to StkP-coated wells and incubated for 1h at 37°C . Wells were washed 5 times with TBS-T and incubated for 1h at RT with 50 μL of a Dylight 650-conjugated anti-6X His antibody ( Invitrogen MA1-21315-D650 ) diluted 100-fold in TBS-T . After 5 washes with TBS-T , plates were read in a Typhoon FLA 9500 scanner ( GE ) under fluorescence mode . Fluorescence ( F ) was fit using a Kaleidagraph to a standard binding isotherm with the form F = Fmax [ng ComE/ ( K1/2 + ng ComE ) ] , where Fmax is the maximum fluorescence at binding saturation and reflects the maximum binding capacity ( Bmax ) , and K1/2 is the amount of ComE ( ng ) required to reach half Fmax . The inverse of K1/2 represents an estimate of ComE affinity for the StkP binding sites . The comE gene was amplified from R801 genomic DNA with the primer pair FhkE/RhkE and cloned into the BamHI/EcoRI sites of the pRSET-A expression plasmid ( Invitrogen ) , yielding pRSET-ComE . stkP and stkP-KD ( kinase domain , amino acids 1–282 ) were amplified with primer pairs FstkP-ex/Rstk-ex and FstkP-ex/ Rstk-kd , respectively , and cloned into BamHI/EcoRI sites of pGEX-4T1 expression plasmid ( GE ) to generate pGEX-StkP and pGEX-StkP-KD . Plasmid pTrc-LytA ( N ) expressing the amino-terminal region of LytA ( 1–206 ) was obtained by cloning a lytA fragment generated by PCR amplification with primers FlytA1/RlytA2 into the BamHI/EcoRI sites of pTrcHis2A ( Invitrogen ) . Mutations were introduced in pRSET-ComE by Quickchange site-directed mutagenesis ( Agilent ) , employing primer pairs NGEP514/NGEP515 , NGEP75/NGEP76 and NGEP77/NGEP78 to obtain pRSET-ComE ( D58E ) , pRSET-ComE ( T128A ) , and pRSET-ComE ( T128E ) , respectively . In the same way , the K42M mutation was introduced in pGEX-StkP with primers NGEP770/771 to give pGEX-StkP ( K42M ) . Soluble His6X-tagged LytA ( N ) and ComE-Hisx6 proteins were purified from the E . coli BL21 ( DE3 ) strain co-transformed with either pTrc-LytA ( N ) or pRSet-ComE derivatives and chaperone expression plasmids pBB540 and pBB550 [2 , 85] . E . coli cells were grown on 800 ml of Terrific broth and induced with 100 μM IPTG according to de Marco [85] . His-tagged proteins were purified from protein lysates obtained by sonication using an NTA-Ni2+ resin ( Qiagen ) following the manufacturer’s protocol . Eluted protein was further purified by gel filtration using a HiPrep 16/60 Sephacryl S-200 HR column mounted in a ÄKTA purifier system ( GE ) . ComE containing fractions were pooled , concentrated with an Amicon Ultra-4 centrifugal filter ( Millipore ) , and dialyzed against the storage buffer [50 mM Tris , 200 mM NaCl , 1mM DTT , 50% v/v glycerol , pH 7 . 5 ) . Samples were snap frozen and stored at -80°C until use . Following exactly the same protocol as above , GST-tagged StkP proteins were purified from the soluble protein fraction of BL21 ( DE3 ) cells bearing pGEX-StkP derivatives and plasmids pBB540 and pBB50 . In this case , a Glutathione Sepharose resin ( GE ) was used to retain GST-tagged StkP . DivIVA was purified from BL21 ( DE3 ) cells transformed with pRSET-divIVA according to Fadda et al . [78] . Purified recombinant Aequorea victoria His-tagged GFP protein was purchased from SIGMA . Native PAGE was used to assess the ComE monomer/dimer ratio . Purified ComE-Hisx6 proteins were diluted in 2x Laemmli sample buffer without 2-mercaptoethanol and SDS and loaded in a 4–20% gradient Bis-Tris precast polyacrylamide gel ( GenScript ) . Electrophoresis was performed at 4°C using Tris-MOPS running buffer without SDS ( GenScript ) at a constant electric field of 15 V cm-1 . Proteins were electroblotted onto a PVDF membrane and probed with Dylight 650-conjugated anti-6X His antibody to detect His-tagged ComE . The membranes were imaged under fluorescence mode in a Typhoon FLA 9500 scanner ( GE Healthcare ) , and bands were quantified with ImageQuant software ( GE Healthcare ) . Cells were initially grown in THYE medium at pH 7 . 8 until OD600nm ~0 . 3 ( log phase ) , centrifuged at 14 , 000 g for 10 min at 4°C , resuspended in the same volume in ABM at pH 6 . 0 ( Piñas et al , 2008 ) and incubated a 37°C for 1h . Then , cells were centrifuged at 14 , 000 x g for 10 min at 4°C , resuspended in a 1/10 vol of lysis buffer ( DOC 1% in 0 . 9% Na Cl ) and incubated 3 min a 37°C until complete lysis . Total RNA was purified by TRIzol reagent according to the manufacturer's instructions ( Fisher Scientific ) from three biological replicates for wt and the comET128A mutant . Posteriorly , we used the Ribopure Bacterial RNA Purification Kit ( Ambion ) following the manufacturer's protocol , with the contaminant DNA being removed using the provided Dnase . rRNA was depleted from 8μg of total RNA using the MICROBExpress Bacterial mRNA Enrichment Kit ( Ambion ) , and then the transcriptome libraries were prepared with TruSeq Stranded RNA Library Preparation Kit ( Illumina ) following the manufacturer's instructions . Briefly , enriched mRNA was fragmented using reagents provided with the kit , and this was followed by first-strand cDNA synthesis and second-strand generation . The libraries were tagged with unique indexes and amplified for a limited number of PCR cycles followed by quantification and qualification using the DNA High Sensitivity Assay Kit . Samples were sequenced using PE150bp chemistry and the Illumina HiSeq . Reads were trimmed by Trimmomatic 0 . 36 [86] to generate high-quality reads . Subsequently , these reads of wt and the comET128A samples were separately aligned to the Streptococcus pneumoniae R6 genome using BWA -version 0 . 7 . 12-r1039 ( bio-bwa . sourceforge . net ) at default parameters . The software package SAMtools ( http://samtools . sourceforge . net/ ) was used to convert the sequence alignment/map ( SAM ) file to a sorted binary alignment/map ( BAM ) file . The mapped reads ratio ( MRR ) to the reference in each dataset was calculated by applying the flagstat command of SAMtools software to the BAM file . The aligned reads were assembled by Cufflinks ( version-2 . 2 . 1 ) , and then the differentially expressed genes were detected and quantified by Cuffdiff , which is included in the Cufflinks package , using a rigorous sophisticated statistical analysis . The expression of the genes was calculated in terms of FPKM ( Fragment per kilobase per million mapped reads ) . Differential gene expression analysis was carried out between wt and the comET128A samples . cDNA was synthesized from 2 μg RNA using the ProtoScript II First Strand cDNA Synthesis Kit ( NEB ) following the manufacture's protocol . cDNA was cleaned using the QIAquick PCR Purification Kit ( Qiagen ) . Genes were amplified using the oligos listed in the S2 Table and FastStart Essential DNA Green Master Mix ( Roche ) following the manufacturer's protocol . Expression was determined relative to AU0158 normalized by gyrA ( spr1099 ) expression using the ΔΔCt method [87] . The gyrA had a similar expression by RNA-Seq for wt and the comET128A mutant , and this had been used to normalize the expression in S . pneumoniae in other studies [88] . The assays to determine the intracellular survival of pneumococci were performed as reported previously [33] , but with modifications . Briefly , 3 . 0 × 105 of A549 cells per well were seeded in 6 well plates and cultured in DMEM supplemented with 10% of fetal bovine serum ( FBS ) and incubated for 12 h . Pneumococci were grown in THYE to the mid-log phase ( OD600nm 0 . 3 ) and resuspended in DMEM ( with 10% FBS ) . Infection of cell monolayers was carried out using a multiplicity of infection ( MOI ) 20:1 . Bacterial internalization after incubation and washes with extracellular antibiotics was approximately 1% , and the occurrence of apoptosis/necrosis caused by pneumococcal infection quantified by flow cytometry ( Annexin V/propidium iodide labeling kit; Invitrogen ) was approximately 5–10% for all time points analyzed . A549 cells were incubated 3 h with pneumococcal strains and cells were washed three times with phosphate-buffered saline ( PBS ) and fresh DMEM ( without FBS ) containing 150 μg/ml potassium penicillin G ( Sigma P7794 ) and 900 μg/ml gentamicin sulfate ( US Biological G2030 ) . After a 20 min rest period , cells were washed three times with PBS . The eukaryotic cells were lysed by centrifugation for 5 min at 10 , 000 rpm and the bacterial pellet was resuspended in THYE medium . The number of internalized bacteria at different time points was quantified after serial dilutions and plating on BHI 5% sheep blood agar plates with incubation for 16 h at 37°C . The time scale referred to the time after elimination of the extracellular bacteria by antibiotic treatment . A 100% survival was defined after 20 min of antibiotic treatment ( S1 Fig ) , and all the samples were referred to this point to calculate the respective percentages . For the detection of H2O2 released by bacterial cells , the phenol red oxidation microassay was used . Briefly , cells were grown in BHI to the mid-log phase ( OD600nm 0 . 3 ) . Posteriorly , cells were centrifuged at 10 , 000 x g for 5 min , resuspended in Todd Hewitt broth THB ( pH 6 . 0 ) and incubated by 1 h at 37°C . Aliquots were taken and serially diluted to determine viable cells by plating in BHI-blood agar . Other aliquots were centrifuged at 10 , 000 x g for 5 min , and 100 μl of supernatants were transferred to multiwell plates and mixed with the same volume of PRS buffer ( NaCl 140 mM , dextrose 5 . 5 mM , phenol red 280 μM , and horseradish peroxidase 8 . 5 U/ml in phosphate-buffered saline , pH7 . 0 ) . Reactions were incubated for 90 min at 37°C and the reaction was stopped with 10 μl of 1 N NaOH , and the reactive wells were read in a microplate reader ( Bio-Rad ) with a 595-nm filter . Assays were performed in triplicate and results are expressed as mmoles of H2O2 released by 106 cells . Bacterial strains were grown until OD600nm in BHI and aliquots were treated with H2O2 20 mM ( final concentration ) . Every 30 min , aliquots were taken and serially diluted to determine viable cells by plating in BHI-blood agar . The percent survival was calculated by dividing the CFU of cultures after exposure to H2O2 by the CFU of the control tube without H2O2 . Assays were performed in triplicate and results are shown survival percentage at different time points . Limited proteolysis with proteinase K was carried out in a 30 μl reaction volume with 3 μg of ComE or ComET128E and 6 ng of the proteinase K in 10 mM Tris , 1 mM CaCl2 , pH 7 . 5 , for 30 min at room temperature . Reactions were stopped with 5 mM PMSF , 5 mM EDTA and 1X Laemmli loading buffer , and immediately boiled for 5 min . Digestions with trypsin were performed under the same conditions as before but in 100 mM Tris , pH 8 . 5 . Trypsinized samples were boiled immediately for 5 min after stopping the reactions with 5 mM PMSF and 1X Laemmli loading buffer . The extension of protein digestion was verified by 12% SDS-PAGE followed by SYPRO Ruby staining . The RNA-seq data generated from this study are deposited at the NCBI SRA under the accession numbers SAMN08473835 and SAMN08473836 . | Streptococcus pneumoniae is a major human pathogen and is the causal agent of otitis ( media ) and sinusitis . It is also responsible for severe infections such as bacteremia , pneumonia , and meningitis , associated with 2 million annual deaths . Although this bacterium is part of the human nasopharynx commensal microbiota , it can become a pathogen and cross the epithelial cell barrier to establishing infections of varying intensity . Although S . pneumoniae is considered to be a typical extracellular pathogen , transient intracellular life forms have been found in eukaryotic cells , suggesting a putative survival mechanism . Here , we report that the serine-threonine kinase StkP was able to phosphorylate the response regulator ComE to control different cellular processes in response to environmental stress . Moreover , the phosphorylation of ComE on Thr128 , and the consequent conformational and functional changes resulting from this event , extended the current knowledge of molecular activation mechanisms of response regulators . In this report , we provide evidence for the regulatory control exerted by the StkP/ComE pathway on acid-induced autolysis ( associated with pneumolysin release ) , the acid tolerance response , and H2O2 production to modulate tissue damage and intracellular survival , which are ultimately linked to pneumococcal pathogenesis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"phosphorylation",
"cell",
"walls",
"medicine",
"and",
"health",
"sciences",
"cellular",
"stress",
"responses",
"pathology",
"and",
"laboratory",
"medicine",
"pneumococcus",
"intracellular",
"pathogens",
"pathogens",
"cell",
"processes",
"microbiology",
"dna-binding",
"pr... | 2018 | Crosstalk between the serine/threonine kinase StkP and the response regulator ComE controls the stress response and intracellular survival of Streptococcus pneumoniae |
We have used two different live-cell fluorescent protein markers to monitor the formation and localization of double-strand breaks ( DSBs ) in budding yeast . Using GFP derivatives of the Rad51 recombination protein or the Ddc2 checkpoint protein , we find that cells with three site-specific DSBs , on different chromosomes , usually display 2 or 3 foci that may coalesce and dissociate . This motion is independent of Rad52 and microtubules . Rad51-GFP , by itself , is unable to repair DSBs by homologous recombination in mitotic cells , but is able to form foci and allow repair when heterozygous with a wild type Rad51 protein . The kinetics of formation and disappearance of a Rad51-GFP focus parallels the completion of site-specific DSB repair . However , Rad51-GFP is proficient during meiosis when homozygous , similar to rad51 “site II” mutants that can bind single-stranded DNA but not complete strand exchange . Rad52-RFP and Rad51-GFP co-localize to the same DSB , but a significant minority of foci have Rad51-GFP without visible Rad52-RFP . We conclude that co-localization of foci in cells with 3 DSBs does not represent formation of a homologous recombination “repair center , ” as the same distribution of Ddc2-GFP foci was found in the absence of the Rad52 protein .
The process of repairing a chromosomal double-strand break by Rad51- and Rad52-mediated homologous recombination in budding yeast has been defined by a combination of in vitro analysis of purified recombination proteins [1–3] and from “in vivo biochemistry” analyses of the kinetics of repair of site-specific DSBs [4] . Cleaved DNA ends are attacked by several 5’ to 3’ exonucleases to produce long 3’-ended single-strand DNA ( ssDNA ) tails , which are initially coated by the single-strand binding complex , RPA [5 , 6] . RPA is displaced by Rad51 recombinase through the action of mediator proteins , including Rad52 , creating a nucleoprotein filament composed primarily of Rad51 but also its paralogs , the Rad55-Rad57 heterodimer [7–9] . The Rad51 filament engages in a genome-wide search for a homologous sequence that could be on a sister chromatid , a homologous chromosome or at an ectopic location . Once the donor sequence is encountered , Rad51 catalyzes strand exchange to form a D-loop intermediate , the initial step in repair . The 3’ end of the invading strand then acts as a primer to initiate new DNA synthesis that leads to repair of the DSB via several pathways including gene conversion via synthesis-dependent strand annealing or by a double Holliday junction pathway [4] . A combination of Southern blot , PCR and chromatin immunoprecipitation ( ChIP ) experiments have shown that DSB repair proceeds by a series of kinetically slow steps , taking more than an hour to complete ( reviewed in [4] ) . In haploid cells , successful recombination between a nuclease cleaved site with an ectopic homologous donor sequence is strongly dictated by the prior proximity of the donor with the region in which the cleavage site has been inserted [10–13] , where proximity was determined by their contact probability of sequences , as measured by chromosome conformation capture methods [14–16] . The creation of a DSB results in increased chromatin movement , which may increase the likelihood of contact between two loci [12 , 17–22] . Single particle tracking of fluorescently tagged loci adjacent to DSBs has shown that repair through homologous recombination causes an increase in chromatin movement dependent on the number of DSBs present [17 , 18 , 22] . This increased movement in response to DSBs has been shown to be dependent on the DNA damage checkpoint [12 , 18] , DNA repair factors [17 , 18] and chromatin remodelers [12] . Recently , a role for microtubules in controlling chromatin mobility after DNA damage in budding yeast has been proposed [20 , 23]; but others have found DSB-associated movement to be independent of microtubules [24] . There is also evidence that nuclear actin , in association with the chromatin remodeler , Ino80 , may also play a role in chromatin dynamics [25] . When the ends of the DSB fail to encounter a homologous donor sequence , or when there is no donor , an unrepaired break eventually enters a different pathway , where it associates with the nuclear envelope through its association with the spindle-pole body and nuclear envelope protein Mps3 [26 , 27] and to the nuclear pore [28] . One approach to the study of DSB repair in budding yeast has been the use of live and fixed-cell microscopy to monitor the behavior of different fluorescently tagged repair-associated proteins [29] . The most thoroughly studied is Rad52 , the key mediator for the assembly of the Rad51 filament , but which is also critical in later strand-annealing steps [30] . Strikingly , when there are multiple DSBs , created by ionizing radiation or by site-specific endonucleases , there often appears to be a single fluorescent Rad52 focus . The recruitment of more than one DSB to a common focus has also been studied by creating fluorescent chromosome tags ( arrays of LacO or TetO sequences ) near different DSBs . Lisby et al . [29] found that DSBs created by two different site-specific endonucleases co-localized in about 50% of haploid cells . These observations have led to the idea that there could be a “repair center” where recombination proteins might accumulate to facilitate DSB repair [29] . However , immunofluorescent staining of spread nuclei with multiple DSBs found that the number of foci directly correlated with the number of DSBs [31 , 32] . To directly test the requirement for Rad52 in organizing DSBs , we monitored the localization of the Rad52-indpendent DSB binding protein , Ddc2 , yeast’s homolog of the mammalian ATRIP protein that has been previously shown to bind near a DSB and to recruit Mec1ATR kinase [33 , 34] . With Ddc2-GFP , we show that cells which have three site-specific DSBs form multiple , highly dynamic foci that often coalesce and separate , but most cells do not form a single fluorescent focus . The number of foci and their motion are independent of Rad52 and microtubules . We also constructed and characterized a Rad51-GFP fusion protein . Previously , a Rad51-GFP fusion was characterized in Arabidopsis , where it proved to be defective in mitotic DSB repair , but competent in meiosis [35] . This phenotype resembles the “site II” mutation of Saccharomyces cerevisiae Rad51 , which can bind ssDNA but prevents ternary complex formation with Rad51 bound to ssDNA and thus fails to complete strand invasion and DSB repair in mitotic cells [36 , 37] . Similar results were obtained using an isoform of Rad51-GFP from rice and humans in vitro [38] . In fission yeast , Rad51’s homolog Rhp51 , when fused with CFP , proved to be UV-sensitive and incapable of carrying out repair on its own , but this defect was complimented by expression of wild type Rhp51 [39] . In budding yeast a YFP-Rad51 fusion forms DSB-dependent foci despite its inability to participate in HR in mitotic cells [40] . This loss of function was suppressed by introducing a gain-of-function Rad51-I345T mutation , which largely restored viability upon irradiation [41] . Here we show that budding yeast Rad51-GFP binds to site-specific DSBs in mitotic cells but cannot catalyze homologous recombination when it is the only allele present; however , unlike in Arabidopsis , it is not dominant-negative [35] . Consequently , Rad51-GFP can be used to follow fluorescently-labeled filaments that are engaged in functional recombination . When using Rad51-GFP to examine localizations of 3 site-specific DSBs , we found the distribution of foci to be nearly identical to the distribution found with Ddc2 . When Rad51-GFP and Rad52-RFP foci are co-expressed , they co-localize to multiple DSBs , although some limitation in Rad52-RFP expression or a propensity for self-aggregation appears to restrict the number of Rad52 foci .
Standard yeast genome manipulation procedures were used for all strain constructions [42] . Linear DNA and plasmids were introduced by the standard lithium acetate transformation procedure [43] . To C-terminally tag Rad51 and Ddc2 with eGFP , PCR primers were used to amplify the eGFP fragment from pFA6a-GFP ( S65T ) and the TRP1 or KAN selectable marker in the Longtine collection [44] and introduced to the appropriate parent strain by lithium acetate transformation . To create monomeric emGFP , alanine 206 was mutated to lysine by single-stranded template repair using Cas9 as previously described [45] . Briefly , oligos DW549 and DW550 were duplexed and ligated into plasmid bRA89 after digestion with BplI to create plasmid pDW54 . The plasmid together with oligo DW548 were co-transformed and selected for by plating YPD + hygromycin B . Isolates were screened for the correct insertion by PCR and sequencing . Strain genotypes are listed in ( S1 Table ) . Primer sequences are listed in ( S2 Table ) . Plasmids are listed in ( S3 Table ) . Strain YCSL004 [46] carries three HO cleavage sites , at MAT ( position 200 kb ) on chromosome 3 , at position 98 kb on chromosome 6 and position 252 kb on chromosome 2 . These sites are located approximately 100 , 50 and 15 kb from their respective centromeres . To visualize the chromosomally integrated fluorescent tags ( Rad51-GFP and Ddc2-GFP ) after DNA damage , cells from a single colony were grown overnight in 5ml YEP + 3% lactic acid ( YPLac ) . Cells were diluted to OD600 = 0 . 2 and grown for 4 h in 5 ml of fresh YPLac before addition of galactose to a final concentration of 2% to induce GAL::HO expression . For experiments that require the retention of an autonomously replicating plasmid , the same growth procedure except that cells were grown in SD-leucine media supplemented with 2% raffinose . For nocodazole treatment experiments , cells were first exposed to 2% galactose for 3 h then treated with 15 μg/ml nocodazole in DMSO for 10 minutes or the equivalent volume of only DMSO , and then imaged using the protocol detailed below . The efficiency of DSB repair by homologous recombination was determined as described previously for strain YJK17 [47] . Briefly , cells were selected from a single colony on YPD plates and grown overnight in 5 ml of YPLac . Cells were diluted to OD600 = 0 . 2 and allowed to grow until OD600 = 0 . 5–1 . 0 . Approximately 100 cells from each culture were then plated on YPGal ( 2% v/v ) and YPD in triplicate and incubated at 30°C . Viability was calculated by dividing the number of colonies on YPGal by the number of colonies on YPD . Adaptation assays in strain JKM179 were performed as previously described [48] . Briefly , cells were grown in YPLac or SD- media supplemented with 2% raffinose overnight then individual unbudded ( G1 ) cells were plated on YPGal and observed microscopically for 24 h to determine the percent that remained arrested in the G2/M stage of the cell cycle . Viability on MMS media was determined by as described previously [49] . Cells of the appropriate strain were selected from a single colony on YPD plates and grown overnight in 5 ml of selective media to near saturation . The following day , cultures were diluted to OD600 = 0 . 2 and left to grow at 30°C for 3–5 doublings . Cells were then diluted in 200 μl sterile water to OD600 = 0 . 2 in a 96-well plate and subsequently 10-fold serially diluted six times . Cell dilutions were then plated on YPD , -leu , and -leu +0 . 002% MMS plates and left to grow at 30°C for three days . Prior to imaging , cells were washed twice in imaging media SC supplemented with 2% galactose or 2% raffinose and mounted on a glass depression slide coated with agarose supplemented with all amino acids . GFP , RFP and mCherry signals were visualized on a Zeiss AxioObserver spinning disk microscope with a 63x objective and an Andor Revolution spinning disk system consisting of a Nikon Ni-E upright microscope , equipped with a 100X ( n . a . 1 . 45 ) oil immersion objectives , a Yokogawa CSU-W1 spinning disk head , and an Andor iXon 897U EMCCD camera . 10–12 z-stack images spaced at 0 . 5 μm were taken for each image . For live-cell time courses , z-stacked spaced at 0 . 5 μm were taken every 10 – 60s as indicated . Z-stacks were imported into FIJI and max-projected for image presentation or sum projected for foci intensity quantification . Foci were counted by adjusting the image color threshold to the average nuclear signal intensity for a given image and counting spherical regions that gave pixel intensity above the threshold . Foci and nuclear intensities were quantified by measuring the integrated intensities of circular regions from sum-projecting relevant z-stack slices . For colocalization analysis , z-stacks were imported into FIJI and split into the red and green channels . For each image , individual cells were selected and single corresponding z-stacks from the green and red channel were duplicated for analysis . The nucleus was selected as the region of interest ROI and the signal outside the ROI was cleared using the Clear Outside function in FIJI . To isolate the GFP and mCherry foci in the nucleus , the mean signal in the ROI was subtracted from the total nucleus signal . The plugin JACoP was used to calculate the Pearson’s Correlation Coefficient between the red and green channels . Chromatin immunoprecipitation ( ChIP ) was carried out as described previously [37] . In brief , cells were harvested from log-phase population . 45 ml of culture were fixed and crosslinked with 1% formaldehyde for 10 minutes after which 2 . 5 ml of 2 . 5 M glycine was added for 5 minutes to quench the reaction . Cells were pelleted and washed 3 times with 4°C TBS . Cell walls were disrupted by 1 min bead beating in lysis buffer , after which cells sonicated for 2 minutes . Debris was then pelleted and discarded , and equal volume of lysate was immunoprecipitated using α-ScRad51 antibody for 1 hour in 4°C , followed by addition of protein-A agarose beads for 1 h at 4°C . The immunoprecipitate was then salt washed 5 times , and crosslinking was reversed at 65°C overnight followed by proteinase-K addition for 2 h . Protein and nucleic acids where separated by phenol extraction . Chromatin association with Rad51 was assessed by qPCR . Individual timepoints were normalized to the antibody binding efficiency as determined by immunoprecipitation of Rad51 or Rad51-GFP from clarified whole cell extracts prepared by bead beating . α-ScRad51 antibodies were generous gifts from Akira Shinohara ( University of Osaka , Osaka , Japan ) and from Douglas Bishop ( University of Chicago , Chicago , IL ) . Monitoring repair kinetics by qPCR was performed as described previously [50] . Single colonies were inoculated in 5 ml of media lacking leucine with 2% dextrose and grown overnight at 30°C . Overnight cultures were then diluted into 600 ml of YPLac and grown into log phase . DSBs were induced by adding 20% galactose to a final concentration of 2% . To track the dynamics of DSB repair 50 ml aliquots of each culture was collected every hour over 9 h . DNA was isolated using a MasterPureTM Yeast DNA Purification Kit ( Epicentre cat . MPY80200 ) . The repair product , MATa-inc , was amplified using primers MATp13 and MATYp4 with a SYBR Green Master Mix using a Qiagen Rotor-Gene Q real-time PCR machine . To quantify the relative amount of MATa-inc in each sample , SLX4 was used as a reference gene and was amplified using primers NS047-Slx4p7 and Slx4p1 . Primer sequences are shown in ( S2 Table ) .
Ddc2 localizes to a broken DNA end , either directly or by binding to RPA [33 , 34 , 51] and previous studies have shown strong localization of Ddc2-GFP at DSB sites [52–54] . Cells suffering a single DSB arrest for 9–12 h , dependent on Ddc2 , but then switch off the checkpoint and adapt to damage without completing DNA repair [55–57] . Cells lacking Ddc2 , like those lacking its partner , Mec1 , fail to arrest in response to a single DSB created by a galactose-inducible HO endonuclease in strain JKM179 , where homologous recombination has been eliminated [58] . Appending eGFP ( GFP-S65T ) to the C-terminus of Ddc2 did not alter cell cycle arrest or adaptation following a DSB , indicating that the GFP moiety does not inhibit Ddc2’s checkpoint function ( S1E Fig ) , confirming an earlier report [53] . We monitored the localization of Ddc2-eGFP in strain YCSL004 , where three HO cleavage sites on different chromosomes are each efficiently cut within 60 min of HO expression [46] . Three h after HO induction , we observed cells with 1 , 2 , or 3 foci with an average of 2 foci per cell ( Figs 1A and 1B and S1A ) . Because the eGFP moiety is known to occasionally form dimers [59–61] , which could promote colocalization of DSBs , we repeated our analysis in the monomeric eGFP-A206K ( herein , emGFP ) [59] . The proportion of cells with a single focus remain unaltered , although there was a larger percentage of cells containing 3 foci and fewer cells with 2 foci ( Fig 1B ) . This distribution was unchanged in a rad52Δ derivative ( Figs 1B and S1B ) . In both wild type and rad52Δ cells suffering three DSBs , a small portion of cells contained more than 3 foci ( WT = 3 . 5% , rad52Δ = 10 . 7% ) ; rarely , even 5 foci were evident ( Figs 1B and S1B ) . In cycling cells without HO-induced DNA damage , Ddc2-GFP foci were apparent in 4 . 9% of wild type cells and 17% of rad52Δ ( Figs 1C and S1C ) . We conclude that cells which display more than 3 foci after HO induction likely reflect unrepaired spontaneous DNA damage arising during replication and independent of HO induction . Even in the absence of Rad52 , ~25% of cells displayed a single focus . It is evident from Fig 1A that the intensity of the single focus was much greater than the average intensities of each focus in cells displaying 2 or 3 foci . By measuring the fluorescence intensities of individual Ddc2-emGFP foci we determined that the signal intensity of the single focus in cells with one focus is equal to the sum of the signal intensities of 3 individual foci ( Fig 1D ) . Thus , cells with a single focus likely have 3 DSBs that are co-localized . These intensities were unchanged in rad52Δ ( Fig 1D ) . Chromosomal mobility and chromatin persistence length are radically altered after the induction of a DSB [20 , 62–64] . We examined the stability of foci with 3 DSBs by observing cells over a ten-minute period , 3 h after HO induction , using spinning disk confocal microscopy . 75% of cells at this time displayed large buds indicative of G2/M arrest . In ~70% of cells with large buds , the number of foci in a given cell remained constant over 10 min ( Fig 2A and S4–S7 Movies ) . However , in ~30% of cells , the number and position of Ddc2-GFP foci were highly dynamic: the number of foci sometimes diminished , from three to two , or from two to one ( Fig 2A ) . In other examples , a single focus split into two or three foci ( Fig 2A and S8–S14 Movies ) . This behavior was unchanged in rad52Δ ( Fig 2D ) , with the exception that a few cells displayed >3 foci , as described above . We conclude that DSBs are dynamic and can coalesce or dissociate in a Rad52-independent fashion . Microtubules have previously been implicated controlling chromatin dynamics in budding yeast [65 , 66] and recent evidence has directly implicated microtubules in controlling localization and movement of DSBs [20 , 23] . However , others have found that microtubules had no effect on DSB movement [24] . We tested whether the association of DSBs in our system was dependent on microtubules by treating cells with nocodazole 3 h after HO induction and monitoring Ddc2-emGFP foci . As a landmark , we included the spindle pole body ( SPB ) -associated Msp3-mCherry [67 , 68] . Before nocodazole treatment , Mps3-mCherry was frequently localized to two well-separated foci , indicative of the position of SPBs in cells arrested prior to anaphase ( Figs 2A and S2A ) ; however , 10 min after nocodazole addition , the Mps3-mCherry puncta collapsed into a single dot , or two very closely spaced dots , as expected [69] ( Figs 2B and S2B ) . Despite the absence of microtubules , the Ddc2-emGFP foci distribution and dynamics were unaltered ( Fig 2B–2D ) . Therefore , the coalescence and separation of HO-induced DSBs is apparently independent of microtubules . An ideal tool for monitoring DSB formation and repair would be a fluorescent protein that performs a central role in homologous recombination . We created a Rad51-eGFP ( GFP-S65T ) fusion construct connected by a SSGSSG linker , which we have previously used to increase the functionality of other fusion proteins [48] . We integrated this fluorescent domain at the C-terminus of the genomic copy of RAD51 in strain JKM179 in which a single , galactose-induced irreparable DSB , created by HO endonuclease , is induced within 30 minutes upon addition of galactose [55] . Rad51-eGFP is competent for adaptation whereas rad51Δ is adaptation-defective ( S1E Fig ) , in agreement with previous findings [70] . In a strain suffering a single HO-induced DSB , more than 70% of cells displayed a single eGFP focus 3 h after inducing HO cleavage , increasing to >90% by 5 h ( Figs 3A and 3D and S3A ) . Because Rad51 filament formation is dependent on the Rad52 mediator , we confirmed that Rad51-eGFP foci were absent in rad52Δ cells ( Figs 3B and 3D and S3B ) , as well as in cells lacking an HO cleavage site ( Figs 3C and 3D and S3C ) . When Rad51-eGFP was co-expressed with Rad52-RFP , green and red foci colocalized ( Figs 3F and S3D ) , as suggested from previous studies using chromosome spreads [32] . Rad51 abundance has been shown to increase after DNA damage [71 , 72] . This increase is evident comparing the total nuclear intensity of Rad51-eGFP in cells with a DSB ( with or without Rad52 ) compared to cells lacking the HO cleavage site ( Fig 3E ) . In the assays described thus far , DSBs were not repaired by HR because of the lack of a homologous donor template . To investigate the ability of Rad51-eGFP to participate in HR , we turned to strain YJK17 , in which there is a DSB at MATα on Chr3 and a single ectopic MATa-inc donor sequence inserted in Chr5 ( Fig 4A ) [47] . An HO break is repaired in roughly 80% of cells over the course of 6–9 h . YJK17 carrying Rad51-eGFP failed to repair the DSB ( Fig 4B ) . Given the multimeric nature of the Rad51 filament and that many Rad51 mutations are dominant-negative [73 , 74] we asked if Rad51-eGFP is dominant negative . We found that HO-induced recombination in strain YJK17 with Rad51-eGFP became repair-proficient after introducing wild type Rad51 on a centromeric plasmid , expressed from its own promoter ( Fig 4B ) . The kinetics of repair , monitored by qPCR , were very similar for Rad51-GFP complemented by RAD51 compared to wild type ( Fig 4D ) . In parallel with repair , the percent of cells displaying a eGFP focus decreased from 80% at 4 h to ~50% by 7 h and fewer than 30% by 9 h , whereas without complementing Rad51 , foci persisted ( Fig 4C ) . This decrease correlated with the timing of repair , as monitored by qPCR ( Fig 4D ) . The ability of wild type Rad51 to complement Rad51-eGFP could also be seen by analyzing sensitivity to the DNA damaging agent , MMS . While Rad51-eGFP was indeed sensitive to MMS , this sensitivity was rescued by providing wild type RAD51 , expressed from its own promoter on a centromere-containing plasmid ( Fig 4H ) . To test directly if Rad51-eGFP was bound to the DNA around the DSB , we performed chromatin immunoprecipitation using an antibody recognizing Rad51 to monitor Rad51-eGFP accumulation at the DSB induced at MATα , as described previously [75] . Rad51-eGFP binding 250 bp from the DSB end increased over 2 h , reaching a plateau thereafter ( Fig 4E ) . The rate and extent of Rad51-eGFP binding was very similar to wild type Rad51 for the first two h but leveled off at a lower value . That the extent of Rad51 binding was somewhat diminished when both Rad51-eGFP and Rad51 were expressed may indicate that the GFP derivative slightly impairs Rad51 filament formation , although recombination was proficient . As a further measure of Rad51-eGFP binding , we compared the total nuclear fluorescence intensity and the accumulation of Rad51-eGFP in an HO endonuclease-induced focus . The total nuclear Rad51-eGFP signal was unaltered when wild type Rad51 was co-expressed ( Fig 4F ) , but the intensity of fluorescence in the focus was reduced to about 50% of that seen in the absence of wild type protein ( Fig 4G ) . This result suggests that Rad51-GFP is not strongly out-competed by wild type Rad51 protein in forming the filament that contains both wild type and Rad51-GFP monomers . Therefore , Rad51-eGFP effectively binds to resected DNA around a DSB and , when complemented with Rad51 , will permit a detailed analysis of several steps in DSB repair . Arabidopsis Rad51-GFP proved to be meiosis-competent even though it failed to carry out mitotic recombination [35] . As noted above , this phenotype resembles a Rad51 “site II” mutation in budding yeast [36] . In meiosis , the critical functions of strand exchange depend on Rad51’s homolog , Dmc1 , with Rad51 acting in an apparently allosteric fashion [36] . Nevertheless , Rad51 is required for efficient completion of interhomolog meiotic recombination; spore viability is only a few percent in the absence of Rad51 [71 , 76] . We found that budding yeast Rad51-eGFP is meiosis-proficient . Diploids homozygous or heterozygous for Rad51-eGFP produced the same percentage of spores of as wild type . ( Fig 5A ) . After tetrad dissection , spores resulting from diploids homozygous for Rad51-eGFP exhibited a 40% reduction in spore viability , but nevertheless 60% of spores were viable ( Fig 5B ) . Thus , S . cerevisiae Rad51-GFP strongly resembles a site II mutation [36] . We extended our analysis to monitor the localizations of several site-specific DSBs , to determine whether multiple DSBs would appear as a single Rad51-eGFP focus or as distinct foci , as we observed with Ddc2-GFP . We inserted Rad51-eGFP into strain YCSL004 carrying 3 HO cleavage sites . This strain also expressed Rad52-RFP from a centromere-containing plasmid , in addition to the wild type genomic RAD52 [77–82] . Three h after HO induction , ~75% of cells were dumbbell-shaped , indicative of G2/M arrest ( S4A Fig ) . We observed an average of two Rad51-eGFP foci ( Fig 6A and 6B ) . This distribution was unchanged in lig4Δ cells ( S4B Fig ) , in which repair by end-joining is blocked [83–85] . There were a number of instances where cells displayed a single Rad52-RFP focus but multiple Rad51-GFP foci ( Figs 5A and 5B and S4C ) . In these cells , the single Rad52-RFP focus was typically large and always colocalized with one Rad51-GFP focus . Therefore , monitoring the number and locations of DSBs via Rad52 may underestimate the number of Rad51 foci in response to DSBs . To avoid possible self-aggregation of Rad51-eGFP , we created a monomeric GFP ( Rad51-emGFP ) derivative . Three h after HO induction , we found that the percent of cells exhibiting a single focus was unchanged ( Fig 6C ) . However , 15% fewer cells displayed 2 foci while 15% more displayed 3 foci when compared to Rad51-eGFP ( Fig 6C ) . This distribution was unchanged after expressing a wild type copy of Rad51 from a centromere-containing plasmid ( Fig 6C ) We then monitored the localization of Ddc2-mCherry and Rad51-emGFP co-expressed in our 3-break system , 3 h after HO induction , to compare the localization profiles of both DSB markers . Nearly 100% of Rad51-emGFP foci colocalized with Ddc2-mCherry foci ( Fig 6D and 6E ) . In contrast , about 30% of Ddc2-mCherry foci lacked Rad51-emGFP ( Fig 6D and 6E ) . When we monitored Ddc2-mCherry and Rad51-emGFP foci simultaneously at 30-min intervals after HO induction in cycling cells we found that , within 60 min , nearly 40% of cells displayed at least one Ddc2-mCherry focus while Rad51-emGFP was present in only ~5% of cells . By 120 min , this difference was still apparent , in that ~60% of cells displayed ≥ one Ddc2-mCherry focus while only ~30% displayed Rad51-emGFP foci ( Fig 6F ) . However , three h after HO induction , ≥1 Ddc2 and Rad51 foci were present in an equal number of cells , although–as noted above–some Ddc2 foci did not exhibit a Rad51 focus . These data support the proposal that checkpoint activation precedes the loading of recombination machinery [40] .
DSB repair must be coordinated in space and time in order to faithfully repair lesions to the genome . The roles of many proteins involved in DSB repair have been elucidated through in vitro and in vivo biochemistry , but the lack of suitable live-cell markers in budding yeast has provided a barrier to studying DSB repair in real-time . Here , we report DSB dynamics in single- and multiple-break conditions using two different fluorescently tagged proteins that carry out different functions in response to DNA damage; the recombinase Rad51-GFP and the checkpoint-related protein Ddc2-GFP . In both cases , multiple DSBs usually resulted in multiple fluorescent foci . Using Rad51-GFP or Ddc2-GFP in our 3-DSB system , the majority of cells exhibit two or three foci . Rad51-GFP foci usually colocalize with Rad52-RFP , but there are a significant number of cells with more GFP foci than RFP foci . Previous studies have looked specifically at the role of Rad52 in organizing a “repair center” yeast [29 , 86] . Our data suggest that monitoring Rad52 focus formation may underestimate the number of DSBs throughout the nucleus . This difference may in part reflect the temporal recruitment of DSB repair proteins to the site of DSBs such that the continued presence of Rad52 at a DSB may not be necessary once a Rad51 filament has been established; however our previous analysis of nuclear spreads by immunofluorescence suggested that Rad52 which plays a Rad51-dependent role in completing gene conversion [30] persisted longer than Rad51 [32] . To test whether Rad52 recruits multiple DSBs into a common locus , we used Ddc2-GFP , which forms foci independent of Rad52 . In our 3-DSB strain , we see an average of 2 Ddc2-GFP foci per cell , but still about 25% of cells display a single focus , as with Rad51-GFP . However , this distribution remains unchanged in a rad52Δ derivative . Furthermore , using live cell imaging in a strain with 3 DSBs , we found that Ddc2-coated DSBs are often highly dynamic; a single focus can split into multiple foci while multiple foci may coalesce into one focus . Again , this behavior was unchanged in a rad52Δ derivative . We also examined the role of microtubules in organizing DSBs as previous studies have implicated microtubules in promoting chromatin motion [20 , 23 , 65 , 66] , but found that depolymerizing microtubules did not alter the behavior or number of apparent DSBs . Therefore , we conclude that Rad52 and microtubules are not required for organizing multiple DSBs into one specific nuclear location . While our Rad51-GFP construct is not able , by itself , to repair DSBs by homologous recombination , it is not dominant-negative in mitotic cells and supports recombination in meiosis . Biochemical analysis of human Rad51 fused to GFP determined that the fluorescent tag prevented Rad51 from engaging in the pairing of homologous sequences by inhibiting double-stranded DNA from binding in Rad51’s secondary DNA binding site ( site II ) [38] . We envision the same to be true of our Rad51-GFP construct because our ChIP experiments and microscopy suggest that Rad51-GFP can efficiently bind to ssDNA and form a filament , its first step in homologous recombination . However , when Rad51-GFP is the sole copy of Rad51 in cells , DSB repair by homologous recombination is incomplete , presumably at the strand exchange step . Rad51-GFP’s defect in ectopic gene conversion and in MMS-resistance is suppressed by addition of a single second copy of wild type Rad51 expressed from its endogenous promoter . However , it is not that wild type Rad51 simply excludes Rad51-GFP from binding ssDNA , since Rad51-GFP readily forms a DSB-dependent focus in the presence of wild type Rad51 , similar to a Rad51-CFP in fission yeast [39] . About half of the Rad51 monomers in the focus appear to be GFP-tagged . Our data suggest either that a functional Rad51 filament does not require every Rad51 molecule to be functional or that subunit-subunit interactions between wild type and GFP-tagged Rad51 correct the defect . The exact stoichiometry for a functional filament cannot be determined from these experiments , but from previous analysis of the minimum requirements Rad51-mediated strand exchange in vitro [87 , 88] and in vivo [45] , it is likely that there need to be at least two to three functional Rad51 molecules in tandem to facilitate minimal Rad51-mediated strand exchange . Increased chromatin motion in response to a DSB is believed to aid in DNA repair through facilitating in homology search throughout the genome [12 , 62–64] . but the precise mechanism for this motion is unclear [89] . Our characterizations of these live-cell markers of DSBs will facilitate a more detailed study of the the motions of broken chromosomes . | Double strand breaks ( DSBs ) pose the greatest threat to the fidelity of an organism’s genome . While much work has been done on the mechanisms of DSB repair , the arrangement and interaction of multiple DSBs within a single cell remain unclear . Using two live-cell fluorescent DSB markers , we show that cells with 3 site-specific DSBs usually form 2 or 3 foci that can may coalesce into fewer foci but also dissociate . The aggregation and mobility of DSBs into a single focus does not depend on the Rad52 recombination protein that is required for various mechanisms of homologous recombination , suggesting that merging of DSBs does not reflect formation of a homologous recombination repair center . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"microtubules",
"dna",
"damage",
"fungi",
"model",
"organisms",
"experimental",
"organism",
"systems",
"dna",
"epigenetics",
"cellular",
"structures",
"and",
"organelles",
"chromatin",
"cytoskeleton",
"homologous",
"recombination",
"research",
"and",
"analysis",
"methods"... | 2019 | Live cell monitoring of double strand breaks in S. cerevisiae |
Plasmacytoid dendritic cells ( pDCs ) are the major producers of type I IFN in response to viral infection and have been shown to direct both innate and adaptive immune responses in vitro . However , in vivo evidence for their role in viral infection is lacking . We evaluated the contribution of pDCs to acute and chronic virus infection using the feeble mouse model of pDC functional deficiency . We have previously demonstrated that feeble mice have a defect in TLR ligand sensing . Although pDCs were found to influence early cytokine secretion , they were not required for control of viremia in the acute phase of the infection . However , T cell priming was deficient in the absence of functional pDCs and the virus-specific immune response was hampered . Ultimately , infection persisted in feeble mice . We conclude that pDCs are likely required for efficient T cell priming and subsequent viral clearance . Our data suggest that reduced pDC functionality may lead to chronic infection .
Plasmacytoid dendritic cells ( pDCs ) are a rare subset of DCs first appreciated in the blood and peripheral lymphoid tissues of humans and notable for producing extremely high amounts of type I interferon [1] . Thereafter , three studies reported the identification and characterization of the murine counterpart [2]–[4] which was paired with an increased interest and ability to understand pDC function in vivo . pDCs generate ∼1000× more type I interferon than any other cell type which accounts for the majority of circulating type I interferon during acute viral infections ( reviewed in [5] ) . Therefore , it has been postulated that pDCs play a crucial role in disease and particularly viral infections [5] . Despite this tremendous biochemical capacity , a potent physiological role for pDCs in vivo has remained more difficult to identify . Part of this problem can be attributed to the lack of suitable tools to study pDC function in vivo . Several models have been recently created to address this issue including E2-2 deficient mice [6] , BDCA-2-DTR [7] and Siglec-H-DTR [8] transgenic mice . In addition to the numerous activating functions demonstrated previously , some models have also demonstrated that pDCs can exert significant inhibitory function by controlling the homeostasis of regulatory CD4+ T cells [8] . Therefore , despite the classic immune stimulating functions first intimated for pDCs given their potent production of type I interferon , the ultimate role of pDCs in specific infections remains unclear . Using a novel genetic screen we reported the utility of a mutant allele of Slc15a4 ( solute carrier family 15 , member 4 ) named “feeble” to study pDC function in vivo . This ENU-induced single base pair transition resulted in pDCs unable to sense TLR ligands due to a defect in the SLC15A4 histidine transporter [9] . Importantly , TLR responses are specifically ablated in pDCs from feeble mice ( a defect seen in the traditional Slc15a4 knockout as well [10] ) , but intact in other cell types . Also , pDC number is maintained in mutants . Thus feeble mice provide a unique condition where pDCs are present but largely deficient for their function as producers of inflammatory cytokines [9] . This is in contrast to previous murine models involving deletion of pDCs , resulting in compensatory mechanisms within other immune effectors that may obscure the intrinsic role of pDCs [8] . To address these questions of pDC function in vivo , we used the natural mouse pathogen lymphocytic choriomeningitis virus ( LCMV ) in an acute and chronic infection model . feeble mutants displayed clearance kinetics similar to wild type mice after an acute LCMV Armstrong infection . In contrast , LCMV Clone 13 ( Cl13 ) could not be contained by virus-specific T cells , thereby suggesting a role for pDCs in controlling persistent viral infections .
In order to understand the functional role of pDCs in vivo , we challenged feeble mice with murine pathogens . We observed a striking deficit in the ability to control a persistent LCMV Cl13 infection without any significant change in the course of an acute infection from the parent LCMV Armstrong strain as measured by viremia over time ( Figure 1A ) . To ensure whether this defect was specific to persistent ( as opposed to acute infection ) , we harvested organs from LCMV Armstrong infected mice at several time points . LCMV Armstrong showed mild differences in infectious titer 5 days post infection ( dpi ) which were not present at 10 dpi ( Figure S1 ) or anytime thereafter ( for which we were unable to detect infectious virus ) , consistent with our serum titer data . As pDCs are known to respond rapidly to an infectious challenge [5] , [7] , we reasoned that this defect in immunity may be due to a difference in Cl13 establishment during the initial phase of the infection . To address this possibility , we harvested organs from WT and feeble mice 5 and 8 days after Cl13 infection . However , we found no difference in viral titer or tropism at this stage of the infection ( Figure 1BC ) . By 2 months post infection , and consistent with our serum titer data ( Figure 1A ) , we observed that most organs known to harbor a persistent LCMV infection sustained very high levels of virus in feeble in contrast to WT mice ( Figure 1D ) . Cognate T cell responses are critical in controlling persistent viral infections . This is well documented for LCMV Cl13 infection [11] but also for other , clinically relevant persistent infections such as HIV , HBV and HCV [12] . Therefore , we evaluated T cell function 8 days post LCMV Cl13 infection during the peak response . In comparison to WT infection , feeble displayed hypofunctional CD4+ and CD8+ T cell responses against immunodominant LCMV epitopes ( Figure 2AB ) . Consistent with our findings in Figure 1 , this defect was specific to a persistent infection and not observed during LCMV Armstrong infection ( Figure S2 . ) CD8+ T cell responses are instructed via both direct and cross-presentation . LCMV Cl13 like many other persistent pathogens directly inhibits APC function during infection [13] . In order to avoid viral immune evasion strategies , cross-presentation may be used by the host [14] . Thus , we reasoned that the feeble mutation may attenuate cross-priming . To address this we used a sterile model of inflammation depending on cross-presentation and type I IFN signaling [15] . Seven days after immunization , we observed a hypofunctional CD8+ T cell response in feeble as compared to WT animals , indicating a defect in priming ( Figure 2C ) . As the magnitude of this defect was proportionally less than during Cl13virus infection ( Figure 2C vs . Figure 2AB ) , we hypothesized that both pathways of antigen presentation require fully functional pDCs . Consistent with this idea , we observed that mice mutant in IRF7 ( IRF7inept/inept ) , which is required for the type I interferon response in pDCs , was similarly hypofunctional after sterile immunization ( Figure 2C ) . Previously , we reported that the feeble mutation acted in pDCs [9] . However , it was formally possible that this mutation in the histidine transporter SLC15A4 could compromise T cell function intrinsically . To test this hypothesis , we generated mixed bone marrow chimeras in which the irradiated recipient was reconstituted with an equal portion of WT and feeble bone marrow progenitor cells . In this setting we could determine whether there was a competitive advantage between WT and mutant cells . We did not detect a difference in the expansion nor functional activity of CD4+ and CD8+ T cells 8 days post LCMV Cl13 infection between WT and feeble cells ( Figure 3AB ) . Similarly , when mixed bone marrow chimeras were challenged with an antigen that must be cross-presented , we did not observe any differences between WT and feeble ( Figure 3C ) . Therefore , the defect in antigen specific feeble T cell responses was extrinsic to the T cells themselves . To further address the location where the feeble mutation was acting , we made use of OT-I TCR transgenic mice bearing CD8+ T cells specific for the model antigen ovalbumin [16] . We generated OT-I; feeble double mutant mice and cultured naïve splenocytes ex vivo for 3 days in the presence of a greater than 1000-fold range of the cognate immunodominant epitope SIINFEKL . We chose to assess proliferation as it is known to be extremely sensitive to inhibition in CD8+ T cells [11] . We also focused our efforts on CD8+ T cells given their established importance in controlling persistent LCMV Cl13 infection [11] in contrast to other cell types ( e . g . B cells , NK cells [17] , [18] ) . No differences were observed between OT-I and OT-I; feeble double mutant mice with regards to proliferation ( Figure S3 ) . We concluded that the feeble mutation was not required for T cell function in this assay . In addition , we considered whether the feeble mutation may affect other cell types intrinsically and therefore have a contributing role towards the observed in vivo phenotype ( Figure 1 , 2 ) . Similarly to prior studies in which cDC function was intact in feeble mice [9] , we did not observe a defect in NK cell cytotoxicity ( Figure S4 ) or TLR responses of B cells and thioglycolate-elicited macrophages ( personal observations ) . Given the intact intrinsic lymphocyte function observed ( Figures 3 , S2 , S3 ) , we sought to evaluate other mechanisms which could compromise the adaptive T cell response in vivo . pDC are best recognized for their abundant production of type I interferon and other cytokines [5] . Therefore we reasoned that a defect in these secreted molecules may contribute to virus susceptibility and a defect in antigen specific T cell activation ( e . g . as shown via pDC derived type I interferon [19]–[23] ) . To address this possibility , we assessed cytokine/chemokine production 1 and 8 days post infection in the circulation as a potential contributing mechanism for our findings . We observed a reduction in several cytokines with an almost complete loss in the serum concentration of IFNγ and MCP-1 ( Figure 4 ) . These two factors are well known for their ability to both recruit and activate professional antigen presenting cells ( pAPCs ) and may be expressed by pDC in addition to other cell types . Therefore , we reasoned that a defect may lie in the recruitment and/or activation of pAPCs which could explain the in vivo specific defect in T cell activation in feeble . Splenocytes were harvested at 0 , 1 , 3 and 5 days post infection to quantitate both the numbers and activation of key immune effector cells early during LCMV Cl13 infection ( Figure S5 ) . We observed a mild difference in both the recruitment and activation of pAPCs as measured by the total number of viable cells and their functional markers MHC class I & class II in addition to the costimulatory molecules CD80 and CD86 on the three predominant pAPC ( B cells , macrophages and dendritic cells ) early during infection ( Figure S5BC ) . Defects in both the total number and mean fluorescent intensity of functional markers were most pronounced 1 dpi although some deficits were observed 3 dpi . However , these differences were largely not statistically significant . Given the differences observed in cytokine/chemokine serum levels observed acutely post infection ( Figure 4 ) in combination with a mild defect in the recruitment and activation of pAPC to the spleen , we hypothesize that these defects contribute in part to a more subtle and/or complex mechanism present early on in feeble resulting in attenuation of T cell responses , and that may involve other immune effectors . Additionally , our findings may underlie an underappreciated mechanism for pDC action . Given the dramatic requirement we observed for pDCs in controlling a persistent viral infection , we postulated that manipulation of their function may be of therapeutic benefit . To test this hypothesis , we used the TLR9 agonist type A cytosine guanine oligodeoxynucleotide ( CpG-ODN ) . WT mice were given either vehicle or 2 ug CpG2216-ODN ( a CpG motif specific for pDC [24] ) in dinucleotide-1 , 2-dioleoyl-3-trimethylammonium-propane ( DOTAP ) 4 hours before and after LCMV Cl13 infection as well as 3 days post infection . We observed a window in which CpG-ODN was effective in preventing viremia when administered immediately prior to infection ( Figure 5 ) . Our findings demonstrate the limitations and a potential starting point for using pDC specific TLR agonists as an approach in the treatment of persistent viral infections . This may be due to the already chronic stimulation of pDCs under these conditions [5] . Nevertheless , our data also suggest a role for pDCs for the priming of sterile adaptive immunity , which may be translated into cancer therapy . Along this line , a recent study has shown that pDCs may be the key effectors in eliminating malignant melanoma with the clinically approved and prevalently used TLR7 agonist imiquimod [25] . Very recently , a manuscript has been published , showing that constitutive ablation of pDCs leads to virus persistence in mice [26] . While these results complement our data , they also illustrate differences . Contrary to pedigrees genetically lacking pDCs , feeble mutants have normal pDC numbers and are on a pure C57BL/6J background . Infection with 105 pfu of LCMV Docile resulted in increased virus titers as early as 5 days post infection in pDC lacking mice [26] , while we observed prominently higher viremia in feeble mutants only 2 months after challenge with 2×106 pfu LCMV Clone 13 ( Figure 1D ) , and not during the early stage of the infection . Together with the reduction in priming of feeble CD8+ T cell upon sterile immunization ( Figure 2C ) our data demonstrate that the impaired T cell function in absence of functional pDCs is not secondary to failure in early virus control , as it may occur for viral strains with high replication kinetics [27] . We also show that the feeble mutation negatively impacts on the production of several pro-inflammatory cytokines ( Figure 4 ) , yet without severely affecting the recruitment and activation of pAPCs ( Figure S5 ) . Furthermore , another group used the transgenic BDCA2-DTR mouse [7] to deplete pDC during the revision of this manuscript [28] . Interestingly , this study using LCMV Clone13 did not see a difference in viremia upon depletion of pDC with diphtheria toxin . As this model of pDC depletion results in ∼90% loss of pDCs , we suspect that this discrepancy in findings may be due to residual function of the remaining pDC in the BDCA2-DTR mouse model . Regarding the timing of effects in our model , we note that significant defects in CD4+ and CD8+ T cell function are observed 8 days post infection . At this time , the amount of infectious virus in several organs and the blood stream of feeble mice are very similar to WT animals . However , subsequent to this timepoint ( 8 dpi ) we observe higher systemic viral loads in feeble versus WT mice . As pDC activity is seen early by ourselves ( Figure 5 ) and others ( reviewed in [5] ) one may expect even higher viral loads during the first week of infection in feeble mice , which we do not observe . This may reflect a physiological “ceiling” of maximum viral load allowed by the mouse at this time . For instance , when analyzing mice deficient in type I , type II or both interferon receptors as compared to WT mice [29] , splenic viral loads were very similar between all four strains of mice during the first week of infection . Differences between these strains were only noticed subsequent to this time despite our knowledge of the seminal role of interferons in the control of viral infection . Likewise , in our studies we observe very early defects in cytokine production ( 1 dpi ) and early ( 8 dpi ) defects in T cell function . However , the differences in viral loads are not observed until after 8 dpi . CD4+ and CD8+ T cells are required for control of persistent LCMV infection and adoptive transfer of these cells is sufficient to eliminate a persistent LCMV infection [30] , [31] . Therefore , we suspect that like interferon receptor deficient animals a maximum level of viremia is established during the first 8 days of infection which is ultimately cleared by the combined effort of CD4+ and CD8+ T cells in WT mice . However , feeble mice lack effective CD4+ and CD8+ T cell responses ( although the total # of splenocytes at this time is unchanged , data not shown ) and are unable to clear systemic infection during our studies . We suspect that the defect in T cell function accounts for the inability of feeble mice to control persistent viral infection . Interestingly , the gene mutated in feeble mice , slc15a4 , has been implicated in NOD1 signaling in vivo [10] . Therefore , it is possible that incomplete NOD1 signaling may contribute to defects in controlling persistent viral infection in the presence of mutant slc15a4 . However , we believe this to be less likely than a role for pDC function given the body of literature supporting NOD1 ( a receptor for peptidoglycan derivatives ) function in bacterial infections [32] versus the role of pDC in viral infections [5] . PDCs have also been implicated in the control of persistent infections in humans . This is largely due to their potential function for type I interferon production which directly augments T cell function [19]–[23] and the observation that persistent infectious viruses ( e . g . HIV , HBV , HCV ) actively suppress pDC function [5] , [33] , [34] . Moreover , it is known that patients with higher HIV viral loads and lower CD4+ T cell counts have a deficit in pDC numbers which are not recovered with highly active anti-retroviral therapy ( HAART ) ( reviewed in [5] ) . Therefore , our work gives a functional framework to better understand these observations . An important extension of this work will be to address whether a subset of HIV , HBV or HCV patients who are more susceptible to disease progression have a functional compromise in pDC function encoded genetically or otherwise that can be modulated therapeutically . Our studies may provide a foundation to address these and other clinically relevant questions pertaining to pDCs and human disease .
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 protocols were approved by the Institutional Animal Care and Use Committee of The Scripps Research Institute ( Protocol#:09-0098 ) as well as the Institutional Animal Care and Use Committee of Case Western Reserve University ( Protocol#:2011-0200 ) . C57BL/6J mice were bred locally or ordered from Jackson Laboratories . Slc15a4feeble/feeble ( MGI: 4835997 ) and Irf7inept/inept ( MGI: 4442855 ) mutants have been described previously [9] . OT-I mice have been described previously [16] . Compound mutants were generated by intercrossing F1 progeny . The Armstrong strain and Clone 13 variant of LCMV was injected intravenously at a dose of 2×106 PFU per mouse . Viral titers were determined from serial dilutions of serum and organ homogenate by a focus-forming assay on VeroE6 cells as described previously [35] . The following antibodies were used for flow cytometry , to stain splenocytes: CD3ε ( 145-2C11 , eBioscience ) , CD4 ( L3T4 , eBioscience ) , CD8α ( 53-6 . 7 , eBioscience ) , CD19 ( MB19-1 , eBioscience ) , F4/80 ( BM8 , Biolegend ) , Mac-1 ( M1/70 , Biolegend ) , CD80 ( B7-1 , Biolegend ) , CD86 ( GL1 , Biolegend ) , MHC Class I ( 28-14-8 , Biolegend ) , MHC Class II ( M5/114 . 15 . 2 , Biolegend ) , IFNγ ( XMG1 . 2 , Biolegend ) , TNFα ( MP6-XT22 , Biolegend ) . Cell populations were defined as follows: T cell populations , CD3ε+ and either CD4+ or CD8+; B cells , CD19+ , CD3− , CD11c−; macrophages , F4/80hi , Mac-1hi; dendritic cells CD11chi . Specific T-cell responses were determined ex vivo by intracellular IFNγ and TNFα staining after a 5 hour stimulation with 10−7 M peptide in the presence of brefeldin A ( BD Biosciences ) as described before [35] unless otherwise indicated . Tap1−/− 5E1 mouse embryonic fibroblast cells were used to access for a feeble dependent role in cross presentation . 107 5E1 cells were given i . p . per mouse . These cells express the Adeno E1B 192-200 immunodominant Db-restricted peptide [15] . Cells were irradiated with 3000 rad prior to injection . Seven days after injection , splenocytes were harvested and cultured ex vivo with 10−7 M VNIRNCCYI for 5 hours . Then CD8+ T cell intracellular cytokine production was quantitated by flow cytometry . Blood samples were taken from the retro-orbital plexus of mice . Cytokines ( IFNγ , MCP-1 , TNFα , IL-6 , IL-12p70 and IL-10 ) were enumerated with the BD Mouse Inflammatory Cytometric Bead Array per manufacturer's instructions . IFNα was enumerated by VeriKine Mouse Interferon-Alpha ELISA Kit per manufacturer's instructions . OT-1 transgenic mice [16] were bred onto the feeble allele to produce double mutants . Splenocytes were harvested from OT-1; feeble and OT-1 mice and incubated ex vivo for 3 days with the immunodominant CD8+ ovalbumin epitope SIINFEKL under a range of concentrations . CD8+ transgenic T cell proliferation was quantitated with carboxyfluorescein succinimidyl ester ( CFSE ) staining detected by flow cytometry and analyzed with FlowJo software . For bone marrow transplantation , bone marrow cells were extracted from femurs and tibias and were placed in PBS , 0 . 1% BSA ( vol/vol ) . Equal numbers of bone marrow cells from congenic WT ( C57BL/6 . SJL ) ( PtprcaPep3b; Ly5 . 1+ ) and homozygous feeble mice ( Ly5 . 2+ ) were injected intravenously into the lateral tail veins of recipient feeble mice irradiated ( 1000 rad ) 24 h earlier . A total of 107 control C57BL/6J splenocytes or TAP1-deficient splenocytes were resuspended in 1 mL PBS and labeled with low ( 0 . 5 µM ) and high ( 5 µM ) concentrations of carboxyfluorescein diacetate succinimidyl ester ( CFSE ) ( Sigma-Aldrich ) , respectively , at room temperature for 10 min . The labeling was stopped by addition of cold FCS . Cells were washed twice , counted , and resuspended at a concentration of 5×107 cells/mL . The two populations were mixed at a 1∶1 ratio and injected i . v . into recipient mice . Recipients were bled 24 hours later , and PBMCs were analyzed for CFSE staining by flow cytometry . The statistical significance of differences was determined by unpaired Student's two-tailed t-test . Differences with a P value of less than 0 . 05 were considered statistically significant . For all figures: *P<0 . 05; **P≤0 . 01; ***P≤0 . 001 . Unless marked , p>0 . 05 between WT and feeble and data groups were not statistically significantly different . Error bars show standard error of the mean ( SEM ) . | The immune system consists of two arms aimed at fighting infection . Innate immunity represents the first barrier of defense to swiftly react – within minutes – following intrusion by a pathogen . Adaptive immunity is activated a few days later . Cross-talk between these two systems is critical but the means of communication are not yet fully understood . Plasmacytoid dendritic cells ( pDCs ) are innate immune cells best recognized for their ability to produce type I interferon ( e . g . in response to viral infection . ) Evidence for pDCs to modulate the adaptive system in vivo is only recent and still elusive . Using a newly described mouse model named feeble that is characterized by functional deficiency of pDCs , we analysed the role of feeble in the context of acute and chronic viral infection . We found that the feeble mutation affecting pDCs is dispensable for immunity during an acute infection . However our data show that feeble mice failed to control a chronic infection . This was likely due to a reduction in early cytokine secretion and improper activation of adaptive T cells , resulting in virus persistence . Therefore we propose that pDCs are critical for the resolution of chronic infection by linking both arms of immunity . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"immune",
"cells",
"lymphocytic",
"choriomeningitis",
"immunology",
"microbiology",
"immunodeficiency",
"viruses",
"hepatitis",
"immunomodulation",
"hepatitis",
"c",
"animal",
"models",
"of",
"infection",
"infectious",
"diseases",
"viral",
"immune",
"evasion",
... | 2012 | Slc15a4, a Gene Required for pDC Sensing of TLR Ligands, Is Required to Control Persistent Viral Infection |
Centrioles are microtubule-based organelles important for the formation of cilia , flagella and centrosomes . Despite progress in understanding the underlying assembly mechanisms , how centriole integrity is ensured is incompletely understood , including in sperm cells , where such integrity is particularly critical . We identified C . elegans sas-1 in a genetic screen as a locus required for bipolar spindle assembly in the early embryo . Our analysis reveals that sperm-derived sas-1 mutant centrioles lose their integrity shortly after fertilization , and that a related defect occurs when maternal sas-1 function is lacking . We establish that sas-1 encodes a C2 domain containing protein that localizes to centrioles in C . elegans , and which can bind and stabilize microtubules when expressed in human cells . Moreover , we uncover that SAS-1 is related to C2CD3 , a protein required for complete centriole formation in human cells and affected in a type of oral-facial-digital ( OFD ) syndrome .
Centrioles are small microtubule-based organelles that are critical for the formation of cilia and flagella across eukaryotic evolution , as well as for that of centrosomes in animal cells . Centriolar microtubules exhibit unusual stability , which is thought to contribute to the integrity of the entire organelle . That centrioles retain such integrity is probably key not only to withstand mechanical stresses generated by cilia , flagella and centrosomes , but also to ensure proper assembly of new centrioles in proliferating cells . Several components important for centriole assembly have been uncovered in the last decade ( reviewed in [1] , [2] ) . In C . elegans , genetic and functional genomic screens have led to the identification of five core components that act sequentially during centriole biogenesis [3]–[10] . The first protein to be recruited to centrioles is SPD-2 , which is required for the subsequent presence of the kinase ZYG-1 [11] , [12] . The next components to be loaded onto centrioles are the interacting proteins SAS-6 and SAS-5 . At this stage , a central tube can be observed by electron microscopy as the first emerging structure during centriole biogenesis [12] . Next , SAS-4 is incorporated , after which centriolar microtubules are added onto the forming organelle . Centrioles in C . elegans are only ∼100 nm in both length and width , and are comprised of microtubule singlets [5] , [13] . Due to their minute size , pairs of centrioles in C . elegans cannot be resolved by immunofluorescence analysis , except after their disengagement from one another at the end of mitosis [8] , [10] . Some features of C . elegans centrioles differ from those in most other systems , where centrioles are typically ∼450 nm in length and ∼250 nm in diameter and harbor microtubule triplets [14] , [15] . Nevertheless , homologs of the core components initially identified in worms turned out to be critical for centriole formation from algae to humans [16] , [17] . This indicates that C . elegans can serve as a model to discover fundamental and conserved features of centriole biology . In contrast to most cytoplasmic microtubules that exhibit dynamic instability [18] , [19] , centriolar microtubules are very stable , resisting cold- and nocodazole-induced microtubule depolymerization [20] . Accordingly , microtubules of centrioles from human cells purified at 4°C are comparable by electron microscopy to those of centrioles in cells [15] , [21] , [22] . Moreover , pulse-chase experiments with labeled α- and β-tubulin subunits demonstrated that centriolar microtubules exhibit little turnover over one cell cycle in vertebrate tissue culture cells [23] . The α-tubulin subunit of centriolar microtubules undergoes substantial post-translational modifications , including acetylation and de-tyrosinatation [20] , [24] , [25] , as well as polyglutamylation , which also occurs on the β-tubulin subunit of centriolar microtubules [26]–[28] . Some of these modifications , in particular polyglutamylation , appear to not only mark stable microtubules , but also to contribute to their stability [29] , [30] . Accordingly , injection of antibodies against polyglutamylated tubulin leads to loss of centriole integrity in human cells [31] . Several proteins have been identified as being important for the stabilization of centriolar microtubules in human cells , including hPoc1 , CAP350 and centrobin [32]–[35] . Poc1 is an evolutionarily conserved protein that associates with microtubules in vitro and localizes notably to centriolar microtubules in vivo [33] , [36] , [37] . Depletion of Poc1 from human cells or its inactivation in Drosophila results in shorter centrioles , whereas overexpression leads to overly long centriole-like structures [38] , [39] . Moreover , depletion of Poc1 in Tetrahymena leads to the formation of basal bodies that have compromised integrity , being more sensitive to nocodazole [33] . Likewise , centrioles in human cells exhibit nocodazole sensitivity when depleted of CAP350 , a microtubule-associated protein that localizes to centrioles [32] . Centrobin , a component present solely in newly formed centrioles , interacts with tubulin in vitro and in vivo , and this interaction is needed for centriole stabilization [34] , [40] . These studies notwithstanding , how the unusual long-term stability of centrioles is achieved remains incompletely understood , especially in the context of a developing organism . A particularly acute need for centriole stability is encountered during spermatogenesis in many species , since the centrioles that are formed during the meiotic divisions of male germ cells must be retained during the entire course of spermiogenesis to then be contributed to the zygote ( reviewed in [41] ) . By contrast , centrioles are eliminated or inactivated during oogenesis . As a result , the newly fertilized embryo is endowed strictly with paternally contributed centrioles . Next to these original centrioles , new centrioles are then assembled during the first cell cycle with maternally provided components . For these events to occur faithfully , paternally contributed centrioles must retain their integrity throughout spermatogenesis and after fertilization . Despite such retention of centriole integrity being critical for embryonic development , the underlying mechanisms remain to be discovered .
The sas-1 ( spindle assembly abnormal 1 ) locus was identified in a screen for parental-effect embryonic lethal mutations [42] . Using time-lapse DIC microscopy , we found that embryos derived from sas-1 ( t1476 ) or sas-1 ( t1521 ) homozygote mutant animals raised at 24°C almost invariably assemble a monopolar spindle in the first cell cycle ( Fig . 1A–E , Movies S1–S4 , Table S1 ) . In the second cell cycle , while most of the embryos then assemble a bipolar spindle , some exhibit monopolar or tripolar spindle assembly ( Table S1 ) . Both mutant alleles are temperature-sensitive , as evidenced by spindle assembly in the first cell cycle being bipolar in the majority of cases at 16°C ( Fig . 1C ) . To test whether sas-1 ( t1476 ) behaves as a null allele at 24°C , we crossed it to a strain carrying the deficiency eDf2 , in which the large region of chromosome III to where sas-1 had been located is missing [42] . We again observed monopolar spindle assembly in the resulting embryos in the first cell cycle ( Movie S5 , Table S1 ) , as well as sterility in some of the animals , which is never observed in sas-1 ( t1476 ) homozygous animals . Together , these findings indicate that sas-1 ( t1476 ) is a severe reduction-of-function allele , but not a null . For historical reasons , we focused further analysis on sas-1 ( t1476 ) , but found similar results to the ones reported below with sas-1 ( t1521 ) ( Table S1 ) . Unless stated otherwise , we will use the term “sas-1 mutant” hereafter to refer to sas-1 ( t1476 ) homozygous animals . We tested whether centriolar and pericentriolar material ( PCM ) components are present in the monopolar spindle assembled in sas-1 mutant embryos . To this end , we conducted immunofluorescence analysis with antibodies against the centriolar proteins SAS-4 , SAS-5 , SAS-6 and IFA , as well as the PCM proteins SPD-5 and SPD-2 , the latter also marking centrioles . We also labeled microtubules in these experiments using antibodies against α-tubulin to determine the number of microtubule organizing centers ( MTOCs ) . As anticipated from the results with time-lapse DIC microscopy , this analysis revealed that a large majority of sas-1 mutant embryos at pronuclear migration/pronuclear meeting ( 35/40 ) and during mitosis ( 42/55 ) harbor a single MTOC that contains centriolar and PCM components ( Fig . S1 and Table S2 ) . In addition , we found embryos in which MTOCs were devoid of centriolar proteins ( Table S2 ) . Furthermore , in agreement with the occasional tripolar spindles observed by time-lapse microscopy in the second cycle , we also observed some tripolar configurations by immunofluorescence analysis , with the three spindle poles usually exhibiting different sizes ( Table S2 ) . To investigate the origin of tripolar spindles , we generated a sas-1 mutant strain expressing the centriolar marker GFP-SAS-6 and the microtubule marker mCherry-β-tubulin . Of the nine embryos analyzed in which a tripolar spindle assembled in the second cell cycle , we found that in four cases , all three spindle poles harbored GFP-SAS-6 throughout the recording ( see Fig . S2 ) . In the remaining five embryos , at least one of the three spindle poles marked by mCherry-β-tubulin did not harbor GFP-SAS-6 by the time of mitosis . Accordingly , immunofluorescence analysis confirmed the presence of occasional MTOCs devoid of centriolar markers ( Table S2 ) . We conclude that tripolar spindle assembly is often directed by spindle poles that lack the centriolar marker GFP-SAS-6 , but retain MTOC activity . To test whether monopolar spindle assembly in the first cell cycle reflects a paternal and/or a maternal requirement , sas-1 mutant males were mated with wild type hermaphrodites . Time-lapse DIC microscopy revealed monopolar spindle assembly in the first cell cycle in ∼80% of the resulting embryos ( Table S1 ) . Moreover , we found that ∼95% of the progeny of sas-1 mutant males mated with control hermaphrodites do not hatch ( Fig . 1E ) . We conclude that sas-1 has a strong paternal requirement . One possible explanation for monopolar spindle assembly in the first cell cycle followed by bipolar spindle assembly in the majority of embryos in the second cell cycle is that sas-1 mutant sperm harbors one centriole instead of the usual two , as in embryos derived from males mutant for zyg-1 or sas-5 [5] , [6] . To test this possibility , we performed serial section electron microscopy of high pressure frozen sperm cells . Although we cannot exclude that more subtle defects have gone unnoticed , this analysis revealed that sas-1 mutant sperm contain two centrioles with detectable microtubule blades , analogous to the situation in the wild type ( Fig S3A–B ) . Moreover , immunofluorescence analysis demonstrated that sas-1 mutant sperm harbor SAS-4 , SAS-5 and SAS-6 ( Fig . S3C–I ) . Together , these results establish that although there is a paternal requirement for sas-1 function , mutant sperm cells contain two centrioles that do not seem different from the wild type . What happens to the seemingly normal centrioles contributed by sas-1 mutant sperm once in the embryo ? To address this question , we followed the fate of paternal centrioles labeled by GFP-SAS-6 , GFP-SAS-4 or GFP-β-tubulin ( Fig . 2A–N ) . To this end , males expressing these fusion proteins were mated with control hermaphrodites , and the resulting embryos analyzed shortly after fertilization , as well as at the end of the first cell cycle . Shortly after fertilization , we found that whereas 100% of control embryos retain paternal GFP-SAS-6 , the signal is present in only ∼20% sas-1 mutant embryos ( Fig . 2A–B , I ) . The same holds for endogenous SAS-6 ( Fig . S4A ) . We found a similar trend with GFP-SAS-5 , although the outcome is less telling in this case because SAS-5 exchanges readily with the cytoplasmic pool shortly after fertilization even in the wild type [6] ( Fig . 2C–D , I ) . There is also a slight diminution in the case of GFP-β-tubulin , with ∼80% of centrioles contributed by sas-1 mutant sperm exhibiting a GFP focus , as opposed to 100% in the control condition ( Fig . 2G–H , I ) . In the case of GFP-SAS-4 , all embryos in both control and sas-1 mutant condition retain focused signals shortly after fertilization ( Fig . 2E–F , I ) . This experiment also allowed us to conclude that there is no defect in centriole disengagement in sas-1 mutants , since two GFP-SAS-4 foci can be distinguished shortly after fertilization in ∼30% of embryos fertilized by either control or sas-1 mutant sperm ( see also Fig . S4B–D for endogenous SAS-4 ) . Next , we examined the fate of paternal centrioles marked by GFP-SAS-6 or GFP-SAS-4 during mitosis , at the end of the first cell cycle . Whereas both proteins are invariably present as two foci in control conditions , we found that GFP-SAS-6 is never present when originating from sas-1 mutant sperm ( Fig . 2J–K ) . Moreover , we found that a single focus of paternal GFP-SAS-4 is detected during mitosis in almost all sas-1 mutant embryos analyzed ( Fig . 2L–N ) . We conclude that one of the two paternally contributed centrioles disappears by the end of the first cell cycle , whereas the second looses GFP-SAS-6 but still harbors GFP-SAS-4 . This conclusion is in line with the live imaging analysis of tripolar configurations in the second cell cycle that revealed loss of GFP-SAS-6 from some spindle poles ( see Fig . S2 ) . The above results suggest that centrioles contributed by sas-1 mutant sperm somehow loose stability after fertilization . To address whether this is accompanied by a detectable ultra-structural defect , we performed electron microscopy following high pressure freezing of sas-1 mutant embryos in the first cell cycle . Intriguingly , we found that the characteristic microtubule-based structure of centrioles observed in the wild type ( Fig . 2O–P ) is no longer recognizable in sas-1 mutant embryos ( Fig . 2Q–R ) . Instead , we observed electron dense material in the area where centrioles should reside ( Fig . 2R ) ; microtubules can be seen emanating from this area ( Fig . 2R , arrowheads ) , which presumably contains the focus of SAS-4 and α-tubulin detected by immunofluorescence . Overall , we conclude that centrioles loose their integrity and lack organized centriolar microtubules in sas-1 mutant embryos . Because sas-1 mutant embryos typically form a bipolar spindle in the second cell cycle , we hypothesized that the centriolar structure contributed by sas-1 mutant sperm that retains GFP-SAS-4 signal up to mitosis is sufficient to foster the formation of a centriole-like structure in its vicinity , even if it is not a canonical centriole as evidenced by electron microscopy analysis ( see Fig . 2Q–R ) . To test this hypothesis , we crossed sas-1 mutant males to control hermaphrodites expressing GFP-SAS-4 or GFP-SAS-6 ( Fig . S5A–F ) . We found that GFP-SAS-4 and GFP-SAS-6 are recruited in the vicinity of the single paternally contributed centriolar structure that remains as embryos progress through the first cell cycle ( Fig . S5A–F ) . The same holds for GFP-SAS-6 when both males and hermaphrodites are mutant for sas-1 ( Fig . S5G–J ) . We conclude that both SAS-6 and SAS-4 are recruited to the paternally contributed centriolar structure contributed by sas-1 mutant sperm . This likely explains why sas-1 mutant embryos usually undergo bipolar spindle assembly in the second cell cycle . The fact that sas-1 mutant males mated to control hermaphrodites give rise to 5% viable progeny whereas self fertilized sas-1 mutants yield none indicates that there is also a maternal requirement for sas-1 . In order to uncover the nature of this requirement , control males were mated to sas-1 mutant hermaphrodites and the resulting progeny analyzed by time-lapse DIC microscopy . Interestingly , these embryos always assemble a bipolar spindle in the first cell cycle , but exhibit monopolar or tripolar spindle assembly in some blastomeres starting typically in the third cell cycle or thereafter ( Fig . 3A–E , G Movie S6 ) . We noted also that ∼7% of such embryos hatch ( Fig . 3I ) , suggesting that in some cases the majority of divisions must have happened normally , perhaps reflecting the onset of zygotic transcription in these sas-1 ( t1476 ) heterozygous embryos . Extrapolating from the phenotypic analysis of embryos endowed with centrioles from sas-1 mutant sperm , these observations suggest that impairment of maternal sas-1 function results in the loss of centriole integrity with some probability . To estimate this probability , we first developed a simple mathematical model where centrioles would disintegrate after their formation with a single probability inferred from the data . This model predicted a loss of centriolar integrity for 12 . 5% of the centrioles over the course of the experiment ( Materials and Methods , model 1 ) . However , this predicted percentage is higher than the rate of failure observed experimentally at the two cell stage ( 8 . 5% ) . Thus , we developed a second model in which the probability of loosing centriole integrity is allowed to differ depending on the age of the centriole ( Materials and Methods , model 2 ) . Using a maximum-likelihood optimization procedure to identify the most probable values given the experimental data , we found the probability of losing centriole integrity one cell cycle after its formation to be 6 . 3% and two cell cycles after its formation to be 30 . 3% ( Fig . 3F–H , Materials and Methods , model 2 ) . Importantly , this model fit our data significantly better than the first model ( p = 0 . 02 , likelihood ratio test ) . As can be seen in Fig . 3F ( right-most box ) , this model predicts that in embryos lacking solely maternal sas-1 function , those blastomeres in four-cell stage embryos that inherit one of the two centrioles contributed by wild type sperm should invariably exhibit bipolar spindle assembly . By contrast , those blastomeres that do not have paternally provided wild type centrioles could exhibit abnormal spindle assembly . To test this prediction , we fertilized sas-1 mutant oocytes with wild type sperm harboring GFP-SAS-6 positive centrioles and analyzed the resulting embryos at the four-cell stage . This revealed that paternally contributed centrioles are never present in those cells that exhibit abnormal spindle assembly ( Fig . 3J , 0/8 embryos ) . Overall , we conclude that upon compromised maternal sas-1 function , centriole formation is initiated but the resulting structure looses integrity over time . Using SNP mapping and whole genome sequencing , we mapped the sas-1 locus to the ORF Y111B2a . 24 ( Materials and Methods ) . This locus encodes a 570 amino acid ( aa ) -long protein predicted to contain a C2 domain ( Fig . 4A ) . C2 domains have been implicated in membrane targeting , calcium binding and protein-protein interactions ( reviewed in [43] ) . The two sas-1 alleles harbor single amino acid substitutions at conserved residues within this C2 domain: sas-1 ( t1476 ) a P419S alteration and sas-1 ( t1521 ) a G452E change ( Fig . 4A ) . Importantly , both residues are conserved in a variety of C2 domains from different phyla ( Fig . 4B–C ) , indicating their importance for function . To verify that the correct locus has been identified , we generated a strain expressing GFP-SAS-1 and found that this fusion protein can rescue sas-1 ( t1476 ) mutant embryos ( Fig . S6A ) . The rescue is of ∼64% likely because the gfp-sas-1 transgene is driven from the maternal pie-1 promoter instead of the endogenous one ( Fig . S6A ) . Mating sas-1 ( t1476 ) GFP-SAS-1 males to control hermaphrodites results in ∼15% viability , indicating partial paternal rescue , whereas mating sas-1 homozygote males to sas-1 ( t1476 ) GFP-SAS-1 hermaphrodites results in ∼32% viability , indicating partial maternal rescue ( Fig . S6A ) . Overall , we conclude that the ORF Y111B2a . 24 indeed encodes SAS-1 . Having ascertained the identity of the sas-1 locus , we performed RNAi to investigate whether a more severe phenotype could be revealed . When injecting animals with sas-1 dsRNA to deplete the maternal pool using RNAi , we observed by immunofluorescence the presence of multipolar spindles , starting typically at the four-cell stage , just like for embryos maternally mutant for sas-1 ( Fig . S6B–C , compare to Fig . 3A–E , 3J ) . Furthermore , we sought to deplete both paternal and maternal pools by subjecting animals to RNAi from the first instar larval stage onwards ( Materials and Methods ) . We found that some of the resulting sas-1 ( RNAi ) animals are sterile , in line with the data acquired using the eDf2 deficiency ( see Table S1 ) . Moreover , we found that ∼65% of sas-1 ( RNAi ) embryos assemble a monopolar spindle during the first cell cycle ( Fig . 4D–F , N = 17 , Movie S7 ) . From this subset , ∼73% exhibit a more severe phenotype than that usually observed in the first cell cycle of sas-1 mutant embryos , with the two pronuclei migrating only very slowly towards each other , and typically no bipolar spindle assembly occurring in either first or second cycle ( Movie S8 ) . To test whether this phenotype may reflect a stronger centriolar defect , we performed immunofluorescence analysis and found that , indeed , ∼56% of first cell cycle embryos that exhibit a clear phenotype do not have any MTOC nor do they harbor a focus of centriolar or PCM signal ( Fig . S6D–K ) . We confirmed that the RNAi phenotype in the first cell cycle is of paternal origin , since mating control males with hermaphrodites subjected to sas-1 RNAi rescues bipolar spindle assembly ( Fig . 4F ) , but not viability ( Fig . 4G ) , as for sas-1 mutant animals ( see Fig . 3I ) . Taken together , these results reinforce the notion that sas-1 is needed for centriole integrity . We raised antibodies against SAS-1 , and found them to colocalize with the centriolar protein IFA ( Fig . 4H ) , indicating that SAS-1 is a centriolar component . We found also that the centriolar signal is diminished in the progeny of animals injected with dsRNA directed against sas-1 , attesting to the specificity of the antibodies ( Fig . S6B–C ) . Intriguingly , in addition , we found that the signal detected by these antibodies is often present primarily on just one of the two centrioles in each spindle pole during anaphase of the first mitotic division following their disengagement ( see Fig . 4H; 11/22 embryos ) . Centriolar localization was confirmed using GFP antibodies to label embryos expressing GFP-SAS-1 ( Fig . 4I ) . We also determined that SAS-1 localizes to centrioles in sperm cells ( Fig . 4K ) . Moreover , we found that SAS-1 distribution is not altered in sas-1 mutant embryos ( Fig . 4J ) , but that a fraction of centrioles in sas-1 mutant and sas-1 ( RNAi ) sperm cells lacks detectable SAS-1 signal ( Fig . 4L–O ) , in line with the fully penetrant phenotype that they exhibit shortly after fertilization . Next , we tested whether SAS-1 is a stable component of centrioles by monitoring the fate of paternal centrioles carrying GFP-SAS-1 after fertilization . We found that paternally contributed GFP-SAS-1 is present in one focus in most embryos examined up to the four-cell stage ( Fig . 5A–C ) . These findings establish that SAS-1 is a stable component of centrioles and suggests that the two paternally contribute centrioles may harbor different levels of the protein . We were interested in addressing when during centriole biogenesis SAS-1 is recruited to the forming organelle . Thus , we performed fluorescent recovery after photobleaching ( FRAP ) experiments in embryos in the first cell cycle to address when SAS-1 is recruited onto centrioles . These experiments established that , irrespective of whether centriolar GFP-SAS-1 is bleached at pronuclear migration , at pronuclear meeting or just after cytokinesis , signal recovery is gradual thereafter , with a maximal pace of recruitment after cytokinesis ( Fig . 5D–E ) . These findings lend support to the view that SAS-1 is recruited gradually onto the forming organelle and is stably associated with centrioles thereafter . Next , we addressed how early after fertilization GFP-SAS-1 is recruited onto centrioles . To this end , we crossed control males with GFP-SAS-1 hermaphrodites and found that most centrioles are GFP-positive already shortly after fertilization ( Fig . 5F–H ) . This result could be interpreted in two ways . First , SAS-1 may be a very early component recruited at the onset of centriole formation . Second , SAS-1 could be a very late component that associates with centrioles that are fully assembled . We favor the second possibility because of the gradual recovery following FRAP and because immunofluorescence analysis shows that some centrioles that are positive for IFA are negative for SAS-1 ( Fig . 4H ) . The latter result suggests that SAS-1 has not yet been recruited to this subset of centrioles and hence is unlikely to be an early component . Furthermore , if SAS-1 were a component required for the initiation of procentriole formation , monopolar spindle formation would be expected in the second cell cycle upon maternal depletion , which is usually not the case ( see also Discussion ) . Next , we tested whether SAS-1 recruitment to centrioles shortly after fertilization depends upon other centriolar components . To this end , we depleted each of the five core centriolar components from GFP-SAS-1 hermaphrodites and mated them with control males . As shown in Fig . 5I–N , these experiments established that GFP-SAS-1 can localize to centrioles in embryos depleted of SPD-2 , ZYG-1 , SAS-6 or SAS-5 , in which central tube formation is lacking , or depleted of SAS-4 , in which the subsequent step of microtubule addition does not occur . Note that SAS-6 , SAS-5 and SAS-4 remain present on sperm centrioles in these experiments , such that it is formally possible that SAS-1 recruitment shortly after fertilization depends upon the presence of these proteins on paternal centrioles . In summary , we conclude that SAS-1 is a stable centriolar protein that tends to exhibit an asymmetric distribution on centriole pairs , and that can associate with existing centrioles without the need for the presence in the embryo of the five core centriolar components . We set out to investigate the mechanisms by which SAS-1 stabilizes centrioles . Because proteins cannot be readily overexpressed in C . elegans embryos , we overexpressed SAS-1 in human cells instead . Intriguingly , we found that SAS-1-GFP colocalizes with microtubules in U2OS cells ( Fig . 6A–B ) . Moreover , we noticed that in some cells , SAS-1-GFP is found adjacent to the centriolar protein Centrin , suggesting that SAS-1 can localize to centrioles also in human cells ( Fig . 6Q ) . Furthermore , we addressed whether SAS-1 can associate with stable microtubules by staining cells for acetylated tubulin , a hallmark of such microtubules . Strikingly , we found that SAS-1-GFP exhibits extensive colocalization with acetylated microtubules ( Fig . 6E–F ) . Together , these results indicate that SAS-1 can recognize stable microtubule configurations in this heterologous system , and raises the possibility that it could likewise associate with centriolar microtubules , which are also extremely stable . We noted that cells expressing high amounts of SAS-1-GFP typically harbor more and thicker microtubule bundles ( Fig . 6A–B ) . This raises the possibility that SAS-1 not only associates with stable microtubules but perhaps also promotes their stability . To address this possibility , we depolymerized microtubules using cold shock or nocodazole and found that under these conditions cells expressing SAS-1-GFP harbor more stable microtubules marked by acetylated tubulin than control cells ( Fig . 6I–J , M–N ) . The conclusions reached using immunofluorescence analysis were confirmed in co-immunoprecipitation experiments , whereby SAS-1-GFP immunoprecipitates α-tubulin as well as acetylated tubulin ( Fig . 6S ) . Taken together , these results establish that SAS-1 can bind and stabilize microtubules in human cells . We next addressed whether the mutant versions of SAS-1 are impaired in their binding or stabilization activities in human cells . Importantly , we found that although the P419S and G452E mutant versions of SAS-1 localize to microtubules to some extent , they do not result in the generation of the numerous thick microtubule bundles observed with the wild type protein ( Fig . 6C–D ) . In addition , the mutant proteins localize less extensively with acetylated tubulin ( Fig . 6G–H ) , and SAS-1 P419S fails to localize to centrioles ( Fig . 6R ) . Furthermore , we found that the mutant versions are not able to protect microtubules against cold- or nocodazole-induced depolymerization ( Fig . 6K–L , O–P ) . Together , these observations suggest that the C2 domain is important for the microtubule binding and stabilization activities of SAS-1 . This conclusion is supported by co-immunoprecipitation experiments , which revealed decreased association with α-tubulin and acetylated tubulin for the P419S mutant version of SAS-1 ( Fig . 6S ) . Taken together , these results indicate that SAS-1 binds and stabilizes microtubules in human cells . By extension , we propose that SAS-1 does so with centriolar microtubules in C . elegans ( see Discussion ) . We set out to address whether SAS-1 is evolutionarily conserved . BLAST analysis indicates that SAS-1 is well conserved amongst nematodes , and that proteins related to SAS-1 can be identified outside the nematode phylum . These include the potential SAS-1 human homolog C2CD3 ( Fig . 7A–B ) . C2CD3 exhibits highest homology within the C2 domain of SAS-1 , including conservation of the residues altered in the sas-1 mutant alleles ( Fig . 7A ) . Aligning SAS-1 and C2CD3 enabled us to identify a second domain in the N-terminal region of SAS-1 that bears resemblance to another C2 domain at the N-terminus of C2CD3 ( Fig . 7A ) . Interestingly , C2CD3 is important for the assembly and/or maintenance of distal elements of centrioles in human and mouse cells [44]–[46] . Moreover , C2CD3 is important for primary cilium formation during mouse embryogenesis [47] and is mutated in an oral-facial-digital ( OFD ) syndrome [45] . Endogenous C2CD3 localizes notably to the distal end of human centrioles [44] , and we found in addition here that C2CD3-GFP colocalizes with acetylated microtubules in some cells , reminiscent of the observations with SAS-1-GFP . Such colocalization has also been observed when overexpressing mouse C2cd3 in human cells [45] . Although we found that SAS-1-GFP cannot rescue C2CD3 depletion in human cells ( Fig . 7C ) , we wanted to test if mutations analogous to those that impair SAS-1 function in C . elegans also impact C2CD3 function in human cells . To this end , we expressed the two corresponding mutant versions of C2CD3 , P1688S and G1725E , and analyzed their capacity to rescue depletion of endogenous C2CD3 . We found that P1688S can rescue C2CD3 depletion to some extent , while G1725E cannot ( Fig . 7C ) , in line with sas-1 ( t1521 ) being a more severe allele than sas-1 ( t1476 ) in C . elegans ( see Table S1 ) . Moreover , we found that whereas wild type C2CD3-GFP localizes to centrioles , G1725E does not ( Fig . 7D , F ) . In an attempt to test if this lack of centriolar localization is responsible for the lack of function , we fused the G1725E mutant to a PACT domain to target it to centrosomes [48] . We found that whereas the PACT domain directs the protein to centrosomes in some cells ( Fig . 7G ) , there is no rescue of the phenotype incurred upon C2CD3 depletion ( Fig 7C ) . This may be because the PACT domain targets the protein to the incorrect centrosomal location or else reflect the fact that the G1725E mutation impairs activity in addition to centriolar targeting . Regardless , these results together suggest that C2CD3 may be a bona fide SAS-1 functional homolog in human cells .
Our analysis uncovered both a paternal and a maternal requirement for sas-1 function ( Fig . S7 ) . When paternal function is lacking , centrioles contributed by sas-1 mutant sperm loose their integrity following fertilization , and a monopolar spindle assembles in the first cell cycle . When maternal function is lacking , centrioles loose their integrity with a given probability later during embryogenesis , leading to cell division defects usually from the third cell cycle onwards . In contrast , essentially all centrioles contributed by sas-1 mutant sperm are affected , indicating that SAS-1 plays a particularly important role during spermatogenesis to ensure the maintenance of paternally contributed centrioles after fertilization . Perhaps the centrioles made during male gametogenesis are particularly sensitive because of their extreme persistence compared to those made in rapidly proliferating cells . Alternatively , SAS-1 may protect centrioles from the mechanism that eliminates centrioles during oogenesis , which may conceivably remain somewhat active after fertilization . The phenotypes exhibited by embryos lacking either paternal or maternal SAS-1 function support the notion that SAS-1 is recruited to centrioles after centriole assembly has occurred . Indeed , if SAS-1 were required early during centriole biogenesis , a failure in centriole formation would be expected already during the meiotic divisions of spermatogenesis , leaving mature sperm with just one centriole , as in zyg-1 or sas-5 mutant animals [5] , [6] . Instead , our analysis demonstrates that there are two paternally contributed centrioles in sas-1 mutant sperm , which loose integrity following fertilization . Similarly , a requirement early during centriole biogenesis would be expected to give rise to monopolar spindle assembly in the second cell cycle upon depletion of the maternal contribution of SAS-1 , as upon that of zyg-1 , sas-5 , sas-6 or sas-4 [5]–[10] . Instead , the phenotype upon maternal depletion of sas-1 becomes apparent only in later cell cycles , as expected from a requirement to stabilize centrioles generated in the embryo . We note , however , that SAS-1 function seems to be required for the generation of structurally normal centrioles , as evidenced by the absence of centriolar microtubules visible by ultrastructural analysis of sas-1 mutant embryos during the first cell cycle ( Fig . 2Q–R ) . Overall , we conclude that SAS-1 is a component that acts to ensure the integrity of centrioles both during their assembly and thereafter . Our work established that SAS-1 stabilizes paternally contributed centriolar SAS-6 shortly after fertilization , whereas the requirement for the stabilization of centriolar SAS-4 becomes apparent only later in the cell cycle . In principle , SAS-1 could exert a dual function , one in stabilizing microtubules and one in maintaining centriolar SAS-6 . A similar dual function has been proposed for Tetrahymena Poc1 , which localizes not only to centriolar microtubules , but also to the cartwheel region where SAS-6 proteins reside [33] . Intriguingly , in addition , a fraction of human cells depleted of Poc1A and Poc1B assemble tripolar spindles [35] , reminiscent of the situation in sas-1 homozygous mutant embryos and embryos maternally mutant for sas-1 ( see Fig . S7 ) . A simpler alternative to invoking a dual function for SAS-1 is that impaired centriolar microtubule stability could result in the loss of the central tube , and thus of its constituent protein SAS-6 . Regardless of the actual mechanism , the focus of β-tubulin and SAS-4 that is left in the mutant condition likely reflects trapping of these proteins in the location of what used to be a centriole . In this context , it is interesting to note that SAS-4 can persist in a focus without SAS-6 ( see Fig . 2 ) , indicating that SAS-6 maintenance is not required for SAS-4 maintenance , in contrast to the relationship during centriole biogenesis [11] , [12] . Importantly , these observations also demonstrate that , similarly to what is observed in human cells where HsSAS-6 is absent from the parental centriole [49] , SAS-6 is not essential on the old centriole to foster the formation of a new centriole-related structure in C . elegans . Analyzing the fate of the two centrioles contributed by sas-1 mutant sperm revealed that one of them retains integrity for a longer time than the other ( see also Fig . S7 ) , suggesting that the two sperm centrioles are not identical . Perhaps the age difference between the two sperm centrioles , one having been formed between meiosis I and meiosis II , and the other one being older , is what determines sensitivity to sas-1 impairment . Differential stability of centrioles has also been proposed in human cells based on the observation that single centrioles were observed in some cells upon injection with polyglutamylated antibodies [31] , [50] . The differential requirement of the two sperm centrioles for SAS-1 function ties with the observation that GFP-SAS-1 is most often detected on only one of the two paternally contributed centrioles ( see Fig . 5A–C ) , perhaps the one that relies most on its function . The two sperm centrioles are not identical in most metazoan organisms , including the majority of vertebrate species . For instance , in human sperm , one centriole seeds the formation of the axoneme of the flagellum and is degenerate , while the other centriole seems to remain intact [51]–[53] . By contrast , in C . elegans , in which sperm cells are amoeboid , ultrastructural analysis failed to reveal any difference between the two centrioles [5] . The unequal distribution of GFP-SAS-1 may provide the first molecular means to distinguish between these two entities . We showed that SAS-1 can bind and stabilize microtubules in human cells . Since SAS-1 mutant proteins do not bind or stabilize microtubules as efficiently , the C2 domain must be important for these activities . How do findings in such a heterologous system relate to SAS-1 function in C . elegans ? Centriolar microtubules are not known to be post-translationally modified in C . elegans; in fact , the α-tubulin isoforms expressed in the early embryo do not have the K40 residue required for their acetylation [54] . Nevertheless , centriolar microtubules are also stable in C . elegans embryos , as evidenced by their resistance to cold treatment [55] . One possibility is that SAS-1 recognizes a conformation characteristic of stable microtubules , rather than a specific modification on α- or β-tubulin , and that such a conformation characterizes both stable cytoplasmic microtubules in human cells and centriolar microtubules in both human cells and C . elegans embryos . Another , more radical , hypothesis is that in the absence of tubulin post-translational modifications in the early C . elegans embryo , there must be another system to stabilize centriolar microtubules . SAS-1 could be part of such a hypothetical system and thus guarantee centriolar microtubule stability . Bioinformatic analysis indicates that C2CD3 is a putative human homolog of SAS-1 . C2CD3 also acts late , being needed for the presence of hPOC5 and Centrin in the distal part of centrioles [44] . Interestingly , upon hPOC5 depletion in human cells , centrioles are short and comprise only microtubule doublets instead of the usual triplets [56] . We speculate that human cells depleted of C2CD3 may have impaired centriolar microtubules and/or centriole integrity . In agreement with this suggestion , it has been reported that centrioles in fibroblasts of C2CD3 mutant mice are shorter and lack appendages [45] , although another study reported no such defect [46] . Whether centrioles in such mutant C2CD3 cells harbor microtubule triplets is unclear [45] . Our findings with SAS-1 also have potential implications for human disease-causing genes . Suggestively , patients suffering from an oral-facial-digital ( OFD ) syndrome harbor a single amino acid substitution in a C2 domain of the SAS-1 homolog C2CD3 [45] . Moreover , the two sas-1 mutations lie in evolutionarily conserved residues amongst C2 domains , together suggesting that C2 domains are important for function in this protein family . Intriguingly , the equivalent of the P419S mutation has been implicated in Joubert syndrome ciliopathy , with a P1122S alteration in the C2 domain containing protein CC2D2A [57] . Moreover , several proteins that localize to the transition zone between the centriole and the primary cilium contain C2 domains , including NPHP1/4/8 . Mutations in the genes encoding these three proteins can lead to nephronophtisis , a cystic kidney ciliopathy . Hence , mutations in the residues we describe here may be uncovered in upcoming investigations of ciliopathies ( reviewed in [58] ) . In conclusion , we discovered SAS-1 as a novel C2 domain containing protein that can bind and stabilize microtubules . SAS-1 is required most critically during spermatogenesis to maintain centriole integrity in the newly fertilized embryo , and thus plays an essential role in ensuring that intact centrioles are passed on from one generation to the next .
Nematode culture was according to standard procedures [59] . The parental sas-1 ( t1476 ) unc-32 ( e189 ) /qC1 dpy-19 ( e1259 ) glp-1 ( q339 ) and sas-1 ( t1521 ) unc-32 ( e189 ) /qC1 dpy-19 ( e1259 ) glp-1 ( q339 ) strains [42] were crossed to different strains expressing fluorescent fusion proteins to obtain suitable transgenic animals . When males were needed , we used sas-1 ( t1476 ) simply balanced by hT2 . fog-2 ( q71 ) [60] , plg-1 ( e2001 ) [61] , CB4856 [62] , CB4118 ( unc-32 ( e189 ) , ooc-4 ( e2078 ) ; eDf2 , GFP-SAS-6 [63] , GFP-SAS-4 [64] , rrrSi212[Psas-5::gfp:sas-5 ( synthetic ) ::sas-5 3′UTR; unc-119 ( + ) ] ( II ) , unc-119 ( ed3 ) III; ijmSi8 [pJD362/pSW077; monII_5′mex-5_GFP::tbb-2; mCherry::his-11; cb-unc-119 ( + ) ]I ( a gift from Julien Dumont ) . mCherry-SAS-4 [65] , mCherry-H2B was obtained by outcrossing OD57 ( mCherry-H2B GFP-α-tubulin ) [66] to N2 animals and segregating away GFP-α-tubulin . For generating worms expressing GFP-SAS-1 , the cDNA of Y111B2A . 24 was cloned in frame into pSU25 , which expresses GFP under pie-1 regulatory elements and carries also unc-119 ( + ) . The GFP-SAS-1 fusion construct was bombarded into unc-119 ( ed3 ) worms using a gene-gun ( Biodrad ) [67] , and resulting animals with wild type locomotion were crossed into the sas-1 ( t1476 ) mutant background , resulting in GZ1006 . RNAi by feeding was performed according to standard procedures . L4 worms were subjected to sas-1 ( RNAi ) for 96 h at 20°C and the F1 ( when analyzing sperm ) or F2 ( when analyzing embryos ) of these animals were analyzed; to sas-5 ( RNAi ) for 24 h at 20°C; to sas-6 ( RNAi ) for>20 h at 20°C; to sas-4 ( RNAi ) , spd-2 ( RNAi ) and zyg-1 ( RNAi ) for 48 h at 20°C . To ensure that cross-progeny was examined , in some experiments we utilized otherwise wild type feminized fog-2 ( q71 ) hermaphrodites that do not produce sperm , or plg-1 ( e2001 ) males that leave a clearly visible mating plug in the vulva of the hermaphrodite , but that are otherwise wild type . For simplicity , we refer to fog-2 ( q71 ) hermaphrodites as “control hermaphrodites” and to plg-1 ( e2001 ) males as “control males” throughout the text . sas-1 dsRNA ( nucleotides 15–1252 ) was generated by in vitro transcription ( MEGAscript , Lifetechnologies ) from the SP6 or T7 promoters followed by RNeasy ( Qiagen ) clean up and double stranding . dsRNA was injected into the gonads of young adult hermaphrodite that were then allowed to recover for ∼24 h at 20°C , after which analysis by immunofluorescence analysis or DIC time-lapse microscopy was conducted . To release embryos , worms were dissected in ∼5 µl dH2O on slides coated with 2 µg/µl poly-L-lysine in PBS , then fixed and stained as described for indirect immunofluorescence of centriolar proteins [6] . Briefly , embryos were fixed in ice-cold methanol for <2 min and blocked in 2% bovine serum albumin ( BSA ) for 15–20 min prior to incubation with primary antibodies overnight at 4°C . Human cells were fixed for 7 min in ice-cold methanol , washed in PBS and blocked for 1 h with 1% BSA 0 . 05% Tween-20 in PBS , followed by incubation with primary antibodies overnight at 4°C . The following primary antibodies raised in rabbits were used at the indicated concentrations: 1∶800 SAS-4 [8] , 1∶50 SAS-5 [6] , 1∶1000 SAS-6 [7] , 1∶1000 ZYG-1 [8] , 1∶1000 SPD-2 ( [11] , gift from Michael Glotzer ) , 1∶5000 SPD-5 ( [68] gift from Bruce Bowerman ) , 1∶1000 GFP ( gift from Viesturs Simanis ) . The following primary antibodies raised in mouse were utilized: 1∶500 α-tubulin ( DM1α , Sigma ) , 1∶200 α-tubulin-FITC ( Invitrogen ) , 1∶1000 acetylated tubulin ( T6793; Sigma ) , 1∶50 IFA [69] and 1∶3000 centrin-2 ( 20H5; gift from Jeffrey L . Salisbury ) . Secondary antibodies were goat anti-mouse coupled to Alexa 488 , goat anti-rabbit coupled to Alexa 568 , donkey anti-rabbit coupled to Alexa 594 , and goat anti-mouse coupled to Cy5 , all used at 1∶500 for C . elegans and 1∶1000 for human cells ( Molecular Probes ) . Slides were counterstained with ∼1 µg/ml Hoechst 33258 ( Sigma ) to visualize DNA . For mapping , unc-32 ( e189 ) sas-1 ( t1476 ) or unc-32 ( e189 ) sas-1 ( t1521 ) hermaphrodites were crossed to CB4856 males . Unc-Emb and Unc-non-Emb F2 recombinants were recovered from the heterozygous F1s . SNP mapping [70] was then used to narrow down the locus to a 100 kbp region between SNPs WBVar00567469 and WBVar00182099 . For sequencing , genomic DNA was extracted from>1000 homozygous unc-32 ( e189 ) sas-1 ( t1476 ) worms . Next generation Illumina Technology Sequencing ( FASTERIS ) revealed 5 SNPs and Indels in the region of interest . A C1255T nucleotide change was identified in the sequence of Y111B2A . 24 , translating into a P419S substitution in the C2 domain . A G1355A nucleotide change was found by Sanger sequencing in unc-32 ( e189 ) sas-1 ( t1521 ) , resulting in a G452E change in the C2 domain . Sanger sequencing of the third sas-1 allele , sas-1 ( t1538 ) [42] , showed that it harbors the same single nucleotide change as sas-1 ( t1521 ) . Hence , experiments conducted with sas-1 ( t1538 ) are not reported in the manuscript . The full-length amino acid sequence , the C2 domain and the N-terminal 220 amino acids of SAS-1 were each used to identify homologous sequences using BLAST and PSI-BLAST . Keeping an E-value of <10−5 , after ∼7 PSI-BLAST iterations not more than 7 new proteins were detected for any of the queries . While we did not detect C2CD3 using the C2 domain as input , C2CD3 was the best hit when using both the N-ter and the full length . A single BLAST analysis of SAS-1 against the human proteome identified C2CD3 as the second best hit , with Rabphilin 3A being first due to higher similarity in the C2 domain . However , Rabphilin 3A has another closer homolog in C . elegans ( RBF-1 ) and quickly disappears with PSI-BLAST iterations . Conversely , blasting human C2CD3 against the C . elegans proteome identified SAS-1 as the second hit , with C07A12 . 7 being first . This protein has another closer homolog in human cells , TOM-1 . The multiple sequence alignment was calculated with MAFFT [71] and the phylogenetic tree generated using RAxML [72] and visualized in Dendroscope [73] . The following are the accession numbers of the proteins shown in Fig . 7B: G . gallus XP_004939073 . 1 ( C2CD3 ) and XP_004938682 . 1 ( MYCBP2 ) , X . tropicalis NP_001072727 . 1 , M . musculus NP_001273506 . 1 ( C2CD3 ) and XP_006518463 . 1 ( MYCBP2 ) , H . sapiens NP_001273506 . 1 ( C2CD3 ) and XP_005266356 . 1 ( MYCBP2 ) , C . briggsae XP_002647047 . 1 . Molecular graphics and analyses were performed with the UCSF Chimera package , developed by the Resource for Biocomputing , Visualization , and Informatics at the University of California , San Francisco ( supported by NIGMS P41-GM103311 ) [74] . The N-terminal 441 nucleotides of the sas-1 coding sequence were cloned C-terminally into pMGWA or pHGWA to express MBP-SAS-1N or His-SAS-1N , respectively , which were purified using standard procedures . MBP-SAS-1 was injected into rabbits according to standard protocols ( Eurogentec ) . For affinity purification , His-SAS-1N was blotted on a membrane and the clarified serum incubated with the membrane . Antibodies were eluted using 0 . 1 M glycine at pH 2 . 3 and neutralized with 1 . 5 M Tris-HCl , pH 8 . 8 . Antibodies were dialysed against PBS overnight at 4°C and stored at −80°C or at −20°C in 50% glycerol . SAS-1 antibodies were used at a dilution of 1∶50 or 1∶100 . Time-lapse differential interference contrast ( DIC ) microscopy of early embryos was performed as described [42] . Dual time-lapse and fluorescence DIC imaging was performed on a Zeiss Axioplan 2 [75] . The motorized filter wheel , two external shutters , and the 1 , 392×1 , 040 pixels 12-bit Photometrics CoolSNAP ES2 were controlled by µManager . Images were acquired with an exposure time of 10–100 ms for the DIC and 250 ms for the fluorescence channels using the Zeiss Filter Set 10 ( GFP ) and 400 ms for 43HE ( mCherry ) . Indirect immunofluorescence was imaged on an LSM700 Zeiss confocal microscope , using 0 . 5 µm–1 µm optical slices . Images were processed using ImageJ . U2OS cells were cultured at 37°C and 5% CO2 in DMEM supplemented with GlutaMAX ( Invitrogen , Carlsbad , CA ) and 10% fetal calf serum . Inducible cell lines were generated using a tet-on system [76] . Briefly , cells grown to ∼80% confluency in a 10 cm dish were transfected with ∼7 µg DNA and after 4–6 h the medium was changed . The next day , cells were exposed to medium containing 1 µg/ml puromycin and selected for 1–2 weeks . In experiments with C2CD3-GFP , we used isoform 2 of the protein ( XP_056346 . 3 ) , and it should be noted that there is also a longer isoform 1 ( XP_001273506 . 1 ) . For depletion of C2CD3 , cells were transfected with 20 nM siC2CD3 at 0 h as well as 48 h , and cells fixed at 96 h . For rescue experiments , doxycylin was typically added 24 h after the initial siRNA transfection . To depolymerize microtubules , cells grown in 6-well plates were incubated with 1 µM nocodazole for 1 h or 30 min on lead blocks placed on ice and fixed directly after washing in cold PBS with ice cold methanol . Cells were then washed and incubated with antibodies as described above . Immunoprecipitation experiments were performed using GFP-nanotrap ( Chromotek ) . Briefly , cells were collected following trypsinization , washed with PBS , lysed in MT-buffer ( 50 mM Tris-HCl , pH 7 . 5 , 0 . 5 mM EDTA , 0 . 5% NP40 , 150 mM NaCl ) and incubated with ∼30 µl beads overnight . The following day , beads were washed 3× with MT-buffer and sample buffer was added to the beads . Quantification of western blots was performed in Fiji , using the GelAnalyzer toolkit . For C . elegans embryos , the samples were cryo-immobilized using an EMPACT2+RTS high-pressure freezer ( Leica Microsystems ) , and freeze-substituted at −90°C for 20 h in acetone containing 1% osmium tetroxide and 0 . 1% uranyl acetate . The temperature was raised progressively to room temperature over 22 h in an automatic freeze-substitution machine ( Leica EM AFS ) and samples were thin-layer embedded in Epon/Araldite as published [12] . Thin sections ( 70 nm ) were cut using a Leica Ultracut UCT microtome and collected on Formvar-coated copper grids . Sections were post-stained with 2% uranyl acetate in 70% methanol followed by aqueous lead citrate and viewed in a TECNAI 12 ( FEI ) transmission electron microscope operated at 80 kV . For sperm analysis , worms were preserved for transmission electron microscopy using high pressure freezing and freeze substitution . Briefly , worms were loaded in the 3 mm aluminum carrier along with E . coli food paste and frozen with a high-pressure freezer ( Leica HPM100 , Leica Microsystems ) . The samples were then placed into tubes held in liquid nitrogen and containing frozen acetone with 1% osmium tetroxide , 0 . 5% uranyl acetate , 5% water . The tubes were then warmed inside an ice bucket containing dry ice , and left overnight agitating gently on a shaker table . The following day , the dry ice was removed and the ice bucket and tubes allowed to warm to 0°C over the next 2 hours . The samples were then washed three times with fresh acetone at 5°C and then embedded in a graded series of epon resin-acetone mix , until pure resin . Once completely infiltrated , the samples were arranged in silicon molds and the resin cured for 48 hours at 65°C . Serial sections of the gonad region were cut using a diamond knife and ultramicrotome ( Leica UC7 ) . Sections were collected on single slot copper grids with a formvar support film , stained with lead citrate and uranyl acetate and imaged inside a transmission electron microscope at 80 kV ( Tecnai Spirit , FEI Company ) , using a CCD camera ( Eagle , FEI Company ) . sas-1 ( t1476 ) mutant embryos expressing GFP-SAS-1 were filmed on a LSM700 Zeiss confocal with a 40× objective , acquiring 10–20 µm optical sections at each time point with a time frame of 30 s at a resolution of 1024×1024 pixels . A square of varying size was drawn around a centrosome and bleached with 50 iterations and 100% laser power . For quantification , a square of 10×10 pixels was drawn around the centriole . The outer 64 pixels were measured for background subtraction while in the inner region of the 36 remaining pixels , the brightest 3×3 pixels were quantified as the centriolar signal . Then , the mean pixel intensity of the outer background region was subtracted from the bright inner region . For tracking of tripolar divisions , embryos were imaged on a Yokogawa Spinning Disk Scanning Unit , using a 63× lens and 20 s frame rate over a time of 20–50 min . We define the probability of a sequence of cellular divisions Di given a parameter set termed θ ( i . e . {PD1 , PD2} ) as:where PD1 is the probability of a C1 centriole to disintegrate right away , and PD2 the probability of a C1 centriole to disintegrate after one cell cycle , Pd that of a daughter centriole to disintegrate ( i . e . either after one or two cell cycles ) : θ is either ( PC1 = PC2 ) in the simple model 1 or ( PC1 , PC2 ) in model 2 and M ( x ) and B ( x ) are functions defining the type of division cell x has undergone ) : thus divisions that could not be assessed visually are ignored ( i . e . both B ( x ) and M ( x ) are 0 ) . Using this framework , we define the probability of the division dataset D given a parameter set θ as: To obtain the marginal posterior probability distributions displayed in Fig . 3 , we sampled P ( D|θ ) using a custom Matlab implementation of the Delayed Rejection Adapted Metropolis ( DRAM , [77] ) algorithm , based on their published code . The p-value computed to determine which model to favor was based on a maximum log-likelihood ratio test . | Centrioles are microtubule-based organelles critical for forming cilia , flagella and centrosomes . Centrioles are very stable , but how such stability is ensured is poorly understood . We identified sas-1 as a component that contributes to centriole stability in C . elegans . Centrioles that lack sas-1 function loose their integrity , and our analysis reveals that sas-1 is particularly important for sperm-derived centrioles . Moreover , we show that SAS-1 binds and stabilizes microtubules in human cells , together leading us to propose that SAS-1 acts by stabilizing centriolar microtubules . We identify C2CD3 as a human homolog of SAS-1 . C2CD3 is needed for the presence of the distal part of centrioles in human cells , and we thus propose that this protein family is broadly needed to maintain centriole structure . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"mutant",
"genotypes",
"cell",
"biology",
"heredity",
"cell",
"cycle",
"and",
"cell",
"division",
"genetics",
"biology",
"and",
"life",
"sciences",
"cell",
"processes",
"centrioles",
"cellular",
"structures",
"and",
"organelles",
"genetic",
"mapping"
] | 2014 | SAS-1 Is a C2 Domain Protein Critical for Centriole Integrity in C. elegans |
Plant shoot systems derive from the shoot apical meristems ( SAMs ) , pools of stems cells that are regulated by a feedback between the WUSCHEL ( WUS ) homeobox protein and CLAVATA ( CLV ) peptides and receptors . The maize heterotrimeric G protein α subunit COMPACT PLANT2 ( CT2 ) functions with CLV receptors to regulate meristem development . In addition to the sole canonical Gα CT2 , maize also contains three eXtra Large GTP-binding proteins ( XLGs ) , which have a domain with homology to Gα as well as additional domains . By either forcing CT2 to be constitutively active , or by depleting XLGs using CRISPR-Cas9 , here we show that both CT2 and XLGs play important roles in maize meristem regulation , and their manipulation improved agronomic traits . For example , we show that expression of a constitutively active CT2 resulted in higher spikelet density and kernel row number , larger ear inflorescence meristems ( IMs ) and more upright leaves , all beneficial traits selected during maize improvement . Our findings suggest that both the canonical Gα , CT2 and the non-canonical XLGs play important roles in maize meristem regulation and further demonstrate that weak alleles of plant stem cell regulatory genes have the capacity to improve agronomic traits .
The plant shoot system is derived from the SAMs , pools of stems cells that have the ability of self-renewal , while initiating new leaves and axillary meristems [1] . The CLV-WUS negative feedback loop has been identified as the key pathway to regulate SAM proliferation and differentiation in Arabidopsis , and is widely conserved in other species [2] . This pathway relies on the communication between a battery of receptors , peptides and transcription factors . WUS , a homeodomain transcription factor expressed in the organizing center , promotes stem cell fate [2] , while CLV3 , a small peptide ligand that is secreted from stem cells at the tip of the SAM , is perceived by leucine-rich repeat ( LRR ) receptor kinases , such as CLV1 , and receptor-like protein CLV2 , resulting in the repression of WUS transcription [3–5] . The CLV pathway is conserved in crops , for example maize CLV1 and CLV2 receptor orthologs THICK TASSEL DWARF1 ( TD1 ) and FASCIATED EAR2 ( FEA2 ) function in meristem regulation , and both td1 and fea2 mutants show enlarged meristems , or fasciated , phenotypes [6 , 7] . However , the signaling players and mechanisms downstream of the CLV receptors are poorly understood . A common class of proteins that signal directly downstream of cell surface receptors in mammalian systems is the heterotrimeric G proteins . These proteins , consisting Gα , Gβ , and Gγ subunits , are also key regulators in the transduction of extracellular signals in plants [8] . The classical model established in animals suggests that in the inactive state , the GDP-bound Gα associates with the Gβγ dimer . Ligand activation of an associated 7-transmembrane domain ( 7-TM ) G-protein-coupled receptor ( GPCR ) induces the exchange of GDP for GTP on Gα , promoting dissociation of Gα from the receptor and Gβγ dimer . The activated Gα and Gβγ subunits then interact with downstream effectors to transduce signaling [9] . Emerging evidence suggests that instead of interacting with 7-TM GPCRs as in animals , the plant G proteins interact with single-TM receptors to regulate plant development and disease resistance [10–13] . Recent genetic screens in maize and Arabidopsis identified roles for heterotrimeric G protein α and β subunits in meristem regulation , by interacting with CLV related receptors [10 , 12] . In maize , the Gα subunit COMPACT PLANT2 ( CT2 ) interacts in vivo with the LRR receptor-like protein FEA2 , to control shoot and inflorescence meristem development . ct2 mutants have enlarged SAMs , fasciated ears with enlarged ear inflorescence meristems and more rows of kernels [10] . In contrast , in Arabidopsis the Gβ subunit , AGB1 , interacts with another CLV-related receptor RECEPTOR-LIKE PROTEIN KINASE2 ( RPK2 ) , to transmit the stem cell restricting signal , and agb1 mutants develop bigger SAMs [12] . In addition to interacting with a different class of receptors , the regulatory mechanism of Gα function in plants appears to be fundamentally different from that in animals , since plant Gα subunits spontaneously exchange GDP for GTP in vitro , without requiring GPCR activation [14 , 15] . This novel mechanism of regulation involves a non-canonical Regulator of G-protein Signaling ( RGS ) protein in Arabidopsis , which contains a 7-TM domain coupled to an RGS domain [16] , and promotes conversion of Gα-GTP back to Gα-GDP [16] . However , RGS homologs are missing from many grass species , including maize [15 , 17–19]; therefore , the mechanism of plant G protein regulation , particularly the transition between the active and inactive states , remain largely unknown in these species . Expression of constitutively active Gα subunits that have lost GTPase activity , disrupting the balance between active and inactive Gα , results in distinct phenotypes , supporting the idea that Gα activity needs to be carefully controlled [16 , 20 , 21] . However , the implication of Gα constitutive activity on meristem development has not been addressed . Plants also differ from animals in containing only a relatively small number of heterotrimeric G protein genes . Most plants have only one canonical Gα [15] , however they also encode non-canonical Gα subunits , extra-large GTP binding proteins ( XLGs ) , which contain a Gα domain at the C-terminus [22–28] . Arabidopsis has 3 XLGs , and they function either redundantly or independently , depending on the biological process [22–28] . Arabidopsis xlg1/2/3 triple mutants do not have an obvious shoot meristem phenotype , however knocking out the 3 XLGs in a Gα ( gpa1 ) background leads to a significant increase in shoot meristem size [25] , suggesting they function redundantly with the canonical Gα in meristem regulation; however , the importance of G protein signaling in diverse plant species remains obscure . Taking advantage of the strong developmental phenotypes of maize Gα mutant ct2 , here we explore the roles of G proteins in maize development by either making Gα constitutively active or mutating all maize XLGs using multiplex CRISPR-Cas9 . We demonstrate that CT2 and XLGs have both redundant and specialized functions in regulating meristem development , and importantly , manipulation of maize Gα subunits introduced desirable agronomic traits .
Our previous study showed that the maize heterotrimeric G protein α subunit CT2 plays an important role in shoot meristem regulation , by associating with a maize CLV receptor FEA2 [10] . However , the underlying signaling mechanism remains obscure , and the implication of Gα activity on meristem development has not been addressed . We took the opportunity of the strong maize phenotype to investigate the effect of forcing Gα to be constitutively active in vivo . We hypothesize that the GTPase activity and the GDP-GTP exchange cycle are required for full Gα function in transmitting the CLV signaling to regulate maize meristem development , and thus mutants that are defective in GTPase activity may act as a weak allele of ct2 . Exchange of a single amino acid in mammalian , Arabidopsis , or rice Gα proteins is sufficient to block GTP hydrolysis , resulting in a constitutively active ( GTPase-dead ) form [16 , 20 , 29] . On this basis , we introduced an analogous point mutation , Q223L , in CT2 , to generate a constitutively active protein , which we named CT2CONSTITUTIVELY ACTIVE ( CT2CA ) . To ask if the Q223L mutation abolished GTPase activity , we performed in vitro GTP-binding and GTPase activity assays using fluorescent BODIPY-GTP , where an increase in fluorescence over time corresponds to GTP binding , and a subsequent decrease corresponds to GTP hydrolysis [30] . We first established that CT2 works as an authentic Gα protein , by testing GTP/GDP binding and hydrolysis specificity . CT2 rapidly bound then slowly hydrolyzed fluorescent GTP , with similar kinetics to other vascular plant Gα proteins ( Fig 1A and S1A Fig ) [15 , 30] , and the activity was efficiently competed by non-labeled GTP or GDP but not by ATP or ADP ( S1C Fig ) . As expected , the CT2CA protein had similar GTP-binding , but lacked GTPase activity ( Fig 1A and S1D Fig ) . We further asked if CT2CA interacted with Gβγ in a yeast-3-hybrid ( Y3H ) system . In contrast to CT2 , we found that CT2CA did not interact with the Gβγ dimer , despite being expressed at a similar level as CT2 in the yeast cells ( Fig 1B and S1B Fig ) . In summary , the Q223L point mutation abolished the GTPase activity of CT2 , maintaining it in a constitutively active state that could no longer form a heterotrimeric complex with Gβγ . To test if constitutive activation of CT2 impacted maize development , we introduced the Q223L point mutation into a native CT2 expression construct that also carried an in-frame fusion of mTFP1 at an internal position that maintains full protein function [10] ( Fig 2A ) . After transformation into maize , CT2CA-mTFP1 was correctly localized in a thin line at the cell periphery that co-localized with a plasma membrane ( PM ) counterstain , FM4-64 ( Fig 2B ) , and we confirmed this co-localization following plasmolysis ( Fig 2B ) . We next backcrossed 6 independent transgenic events of CT2CA-mTFP1 into ct2 mutants in a B73 background . Our previous work established that a native CT2-YFP expression construct fully complements ct2 mutant phenotypes , and we found that both CT2CA-mTFP1 and CT2-YFP were expressed at a similar level as the endogenous CT2 [10] ( S2A , S2B and S2D Fig ) . We first asked if CT2CA-mTFP1 was able to complement the vegetative growth defects of ct2 mutants , by measuring plant height and the first leaf length . CT2CA-mTFP1; ct2 plants were significantly taller than ct2 mutants , with longer leaves; however , they were significantly smaller than their normal , ct2 heterozygous siblings with or without the CT2CA-mTFP1 transgene , indicating that CT2CA-mTFP1 only partially rescued the vegetative growth defects of ct2 mutants ( Fig 2C , 2D , 2F and 2G , similar results obtained with 6 independent transgenic events , S2C Fig ) . We also asked if CT2CA-mTFP1 could complement the enlarged meristem phenotypes of ct2 mutants . We again found partial complementation , indicating that CT2CA was only partially functional in meristem regulation ( Fig 2E and 2H ) . Since CT2 is involved in the CLV-WUS pathway by interacting with FEA2 [10] , we tested if CT2CA can still interact with FEA2 in an N . benthamiana transient expression system . The result showed that FEA2-Myc was pulled down by both CT2-YFP and CT2CA-YFP in the co-IP experiment ( S3 Fig ) . Similarly , studies in human and insect cells showed that in some cases G protein subunits and receptors remain associated following receptor activation [31–33] . Further studies will be needed to elucidate the underlying mechanisms . In addition , to ask how CT2CA affected downstream signaling , we measured ZmWUS1 expression in inflorescence transition stage meristems by qRT-PCR . However , we found that ZmWUS1 expression was not significantly changed in ct2 mutants compared to wild type , nor in our constitutively active CT2CA-mTFP1 lines ( S4 Fig ) , similar to other studies involving subtle changes in CLV pathway genes [34 , 35] and reflecting the complex non-linear regulation of the CLV-WUS negative feedback loop . Collectively , our results suggest that ct2ca functioned as a weak allele of ct2 , and that normal GTPase activity and the GDP-GTP exchange cycle is required for full Gα function in maize . We next asked if CT2/Gα function in maize might be compensated by XLGs . We used phylogenetic analysis ( Fig 3A ) to compare the maize XLGs to Arabidopsis , and based on this named them ZmXLG1 ( most similar to AtXLG1 ) and ZmXLG3a and b ( most similar to AtXLG3 ) . We first asked if the three ZmXLGs might function in a heterotrimeric G protein complex , by testing their interaction with a Gβγ dimer in Y3H system . All three were indeed able to interact , similar to CT2 , suggesting that they function in maize heterotrimeric G protein complexes ( Fig 3B ) . To study the functions of ZmXLGs in maize development , we knocked out all three genes using a tandem guide RNA ( gRNA ) CRISPR-Cas9 construct . In one transgenic event , we recovered putative null alleles of all 3 genes , a 1-bp insertion allele for ZmXLG1 , a 4-bp deletion allele for ZmXLG3a , and a 31-bp deletion allele for ZmXLG3b , each within the N terminal half of the protein coding region and before the Gα domain ( Fig 4A ) . Inbreeding these plants produced offspring homozygous for all 3 loci , at the expected ratio . All Zmxlg triple mutant plants showed a striking developmental arrest phenotype , as they were lethal at the seedling stage ( Fig 4B ) , much more severe than in Arabidopsis , where the triple mutants are smaller with reduced fertility , but can still complete the life cycle [27 , 36] . To gain a deeper understanding into the lethal phenotype , we assayed for cell death using trypan blue staining . As shown in S5A Fig , the triple mutants had strong staining , suggesting they were undergoing cell death . We also measured the expression of two immune marker genes , PATHOGENESIS-RELATED PROTEIN 1 ( PR1 ) and PR5 , and found both were significantly up-regulated in the triple mutants , indicating that the lethality may be due to over-activation of immune system ( S5B Fig ) . Rice Gβ mutants also display cell death and lethality [37 , 38] , indicating that the lethal phenotype of G protein mutants is not unique to maize . The reason for these differences between monocot and dicot G protein mutants remains elusive , but may be related to their dual role in immune signaling [13 , 24 , 39 , 40] . Although the Zmxlg triple mutant plants stopped growing soon after germination , we could measure their shoot meristems , and found that they were normal in size and structure ( S6A Fig ) . As the Zmxlg1;3a;b triple mutants were lethal , we next analyzed the developmental phenotype of single or double mutants . Knocking out each single ZmXLG did not alter development; whereas knocking out any two ZmXLGs led to a modest but significant reduction in plant height , but did not affect SAM size ( Fig 4C and 4D and S7 Fig ) , indicating that loss of any two ZmXLGs can be partially compensated by other XLGs or by CT2/Gα . Next , we asked if ZmXLGs function redundantly with the canonical maize Gα , CT2 , by crossing the Zmxlg mutants into a ct2 mutant background . As expected , ct2 mutants were significantly shorter than wild-type siblings [10] , and we found that mutation of any two ZmXLGs dramatically enhanced their dwarf phenotype ( Fig 5A and 5B ) . In addition , mutation of any pair of ZmXLGs significantly increased SAM size in a ct2 mutant background ( Fig 5C and 5D ) , indicating that ZmXLGs are partially redundant with CT2 in SAM regulation . In contrast , although both CT2 and ZmXLGs are expressed in the maize inflorescence , ZmXLG knockouts did not enhance the ct2 inflorescence fasciation phenotype ( S6B , S6C and S8 Figs ) , suggesting that CT2 is the major Gα functioning in inflorescence meristem development . In summary , our results showed that XLGs are partially redundant with CT2 at some stages of development , but that all 3 XLGs redundantly function in early maize development , where they are essential for survival past the germination stage , and cannot be compensated by CT2 . Our previous results indicate that weak alleles of meristem regulatory genes , such as fea2 or fea3 can improve agronomic traits , such as increasing kernel row number ( KRN ) , without the negative yield impacts associated with strong fasciation phenotypes [41 , 42] . The results described above suggest that different Zmxlg mutant combinations reduce maize height , which is an important trait selected during breeding of many cereal crops [43 , 44] . We also found that ct2ca functions as a weak allele of CT2 , and therefore asked if its expression might affect agronomic traits . First , we measured tassel spikelet density , a trait associated with increased meristem size [10 , 42] , of CT2CA-mTFP1-expressing plants in a ct2 homozygous or heterozygous background . ct2 plants expressing CT2CA-mTFP1 had a significantly higher spikelet density compared with normal , ct2 heterozygous siblings with or without the CT2CA-mTFP1 transgene ( Fig 6A and 6B ) . In addition , these plants did not develop stunted , fasciated ears as in ct2 mutants , but made ears of normal length with increased KRN compared with normal , ct2 heterozygous siblings with or without the CT2CA-mTFP1 transgene ( Fig 6C and 6D ) . Since our previous results suggest that there is a positive correlation between the ear inflorescence meristem size and kernel row number [41] , we next checked if this is also true for ct2 plants expressing CT2CA-mTFP1 . Consistently , we found that they had significantly larger ear IMs compared with normal , ct2 heterozygous siblings with or without the CT2CA-mTFP1 transgene ( Fig 6E and 6F ) , but were not fasciated . Leaf angle is another important agronomic trait , because more upright leaves reduce shading and improve photosynthetic efficiency in modern high plant density production systems[45] . ct2 mutants have more erect leaves , however also have negative pleiotropic traits such as extreme dwarfing and very wide leaves [10 , 46 , 47] . Interestingly , we found that plants expressing constitutively active CT2 also had a more erect leaf angle compared with normal , ct2 heterozygous siblings with or without the CT2CA-mTFP1 transgene , without obvious negative pleiotropic phenotypes ( Fig 6G and 6H ) . In summary , we found that ct2 plants expressing a constitutively active CT2/Gα develop phenotypes consistent with a weak allele of ct2 . These finding suggest that the GTPase activity and the GDP-GTP exchange cycle is required for full CT2 function in vivo , but that expression of a constitutively-active version of CT2 can act as a partially functional ( weak ) allele that brings desirable agronomic traits .
Heterotrimeric G protein signaling in mammals and yeast transmits a plethora of developmental and physiological signals from GPCRs to downstream effectors [48 , 49] . Mammals contain many Gα homologs , therefore the full significance of knocking out all Gα signaling has not been addressed . Plants contain a much smaller number , usually a single canonical Gα and ~3 related XLGs [9 , 15 , 22 , 50] . XLGs are evolved from the canonical Gα , and share some redundant functions [25] . In some extreme examples such as moss Physcomitrella patens , the canonical Gα even has been lost during evolution , and its function has been completely replaced by the sole XLG [51] . XLGs have also gained independent functions during evolution , for example , Arabidopsis XLG2 , but not the canonical Gα , interacts with the FLS2 receptor and mediates flg22-induced immune responses [13] . In Arabidopsis , knockouts of all 3 XLGs have no obvious effect on shoot meristem development , and the additional knockout of the canonical Gα leads to only modest effects on development , including a change in leaf shape and slightly larger shoot meristem [25] . These results suggest that the canonical Gα and XLGs work redundantly to regulate shoot development , and heterotrimeric G protein signaling plays a relatively modest role in plant development . In this report , we found that the maize XLGs work both redundantly and independently with the canonical Gα , CT2 . Zmxlg mutations enhanced ct2 null phenotypes in plant height and meristem size during vegetative development , suggesting ZmXLGs function redundantly with CT2 in SAM regulation . However , knocking out all the 3 XLGs in maize leads to a striking early seedling growth arrest and lethality , independent of the presence of CT2 , suggesting ZmXLGs are essential in maize early growth and development . In addition , knocking out ZmXLGs did not enhance ear fasciation , suggesting CT2 is the sole Gα functioning in inflorescence meristem development . Collectively , our results suggest that the maize XLGs and CT2 have overlapping functions at certain stage of development , however , both have evolved specialized functions . While we do not know the signaling pathways of the maize XLGs , it is likely that they interact with receptors involved in plant growth and development , analogous to the interaction between Gα and a CLV receptor [10] . The classic heterotrimeric G protein model established in the mammalian system suggests that Gα is usually in the inactive GDP-bound state , and is activated to switch to the active GTP-bound state by ligand binding to a 7-TM GPCR [9] . However , the plant G proteins , including those from grasses , are spontaneously active in vitro , and it is still under debate if plants have canonical 7-TM GPCRs [15] . Instead , several single TM receptors , such as CLV and innate immune receptors have been found to interact with G proteins [10–13] . Recent studies in Arabidopsis suggest that turning off plant Gα signaling is also an important step for its signal transduction [15 , 52] , indicating that the balance between the active and inactive Gα pool is important to fully exert its function . We found that native expression of CT2CA-mTFP1 in maize partially rescued ct2 mutant phenotypes . Sometimes partial transgene complementation of a mutant is due to improper transgene expression . However , our native CT2-YFP expression construct fully complemented ct2 mutant phenotypes , and CT2CA-mTFP1 was expressed at the same level as CT2-YFP and endogenous CT2 ( S2D Fig ) , so we conclude that the partial complementation is indeed caused by the loss of GTPase activity . In yeast , a constitutively active Gα also similarly only partially complemented the growth defects of Gα null mutants [53] , suggesting that the requirement for GTPase activity is universal . In addition to GTPase activity , GTP binding is also important for the function of Gα . For example , the T475N mutant of Arabidopsis XLG2 , which lacks GTP binding activity , is not able to interact with a downstream effector RELATED TO VERNALIZATION1 ( RTV1 ) [54] . Together , all of these studies suggest that Gα has to bind GTP and to cycle between the active and inactive state to fully exert its function . One explanation for the importance of the cycling is that the Gα controls meristem development through coordinating with the Gβγ dimer pool . Presumably , in both ct2 mutants and CT2CA background , more free Gβγ dimers are released . Arabidopsis Gβ regulates the meristem development via interacting with a CLV-like receptor RPK2 [12] , while the maize Gα , CT2 interacts with another CLV receptor-like protein , FEA2 [10] . It is possible that Gα and Gβ function independently by coordinating with different receptors and downstream effectors at the cell surface , whereas signaling converges at some point . Although their downstream effectors remain largely unknown in plants , Gβ forms a complex with mitogen-activated protein kinases ( MAPKs ) [40] , which may function in the CLV pathway [55] . Therefore , fine-tuning of the active and inactive states of G protein as well as the Gα and Gβγ pools may be important to maintain meristem development , and our study illustrates the complexity of G protein signaling in meristem regulation . Importantly , ct2ca functioned as a weak allele and introduced desirable agronomic phenotypes , similar to many weak alleles that underlie QTLs for crop traits [41 , 42 , 56] . Optimization of traits such as spikelet density , kernel row number , and leaf angle has been key to improvements in maize and other crops , both in improving yield per plant and planting density . Targeting specific regulators such as Gα by using CRISPR to generate weak alleles could enhance multiple yield related phenotypes to meet the food demands of the increasing global population .
Yeast codon-optimized ORFs of CT2 ( GRMZM2G064732 ) , CT2CA , ZmXLG1 ( GRMZM2G127739 ) , ZmXLG3a ( GRMZM2G016858 ) , and ZmXLG3b ( GRMZM2G429113 ) were cloned between the EcoRI and XhoI restriction sites of MCS1 of pGADT7 ( Clontech ) . ZmGB1 ( GRMZM2G045314 ) was cloned between the EcoRI and BamHI restriction sites of MCS1 . ZmRGG2 ( GRMZM6G935329 ) was cloned between the NotI and BglII restriction sites of MCS2 of pBRIDGE ( Clontech ) , respectively . The primer sequences are shown in the supplementary information . The yeast assay was performed in the AH109 yeast strain ( Clontech ) . The double transformants were selected on SC -Trp -Leu ( -LW ) plates . The interaction was tested on the SC -Trp -Leu -His ( -LWH ) medium supplemented with 1 mM 3-Amino-1 , 2 , 4-triazole ( 3-AT ) to suppress histidine synthesis . The HA-tag was detected using the monoclonal anti-HA antibody produced in mouse ( Sigma , clone HA-7 ) . The guide RNAs were designed using the CRISPR-P website ( http://cbi . hzau . edu . cn/crispr/ ) [57] . The multi-gRNA array was synthesized and cloned into pMGC1005 vector by the LR recombination reaction ( Invitrogen ) ( S1 File ) [58] . The construct was introduced into EHA101 and transformed into HiII background using Agrobacterium-mediated transformation by Iowa State University Plant Transformation Facility . The genomic regions spanning the gRNA target sites were amplified by PCR and sequenced . The T0 plants containing lesions in all three XLG genes were backcrossed with ct2 in the B73 background and self-crossed . CT2CA-mTFP1 was constructed by amplification of genomic fragments and fusing with the mTFP1 gene in-frame at an internal position using the MultiSite Gateway Pro system ( Invitrogen ) , as described [10] . All fragments were amplified using KOD Xtreme hot start polymerase ( Millipore Sigma ) and the Q223L point mutation was generated using PCR-based mutagenesis . The ORF of mTFP1 was inserted between the two amino-terminal α helices , αA and αB of CT2 , as described [10] . All the entry clones were assembled in the pTF101 Gateway compatible binary vectors and introduced into the EHA101 Agrobacterium strain . The construct was transformed into HiII background using Agrobacterium-mediated transformation by the Iowa State University Plant Transformation Facility . The T0 plants were backcrossed twice with ct2 mutants in the B73 background . For genotyping , a 1 . 5 kb fragment of the CT2 gene was amplified and digested with AccI , as a single SNP causes a loss of the 5’ AccI site in the ct2-Ref allele . The transgene was amplified using one primer against the mTFP1 sequence and the other primer against the ct2 sequence . Primers are listed in the S1 Table . For the SAM , ear IM , plant height , and leaf angle measurements , the plants were grown in the greenhouse with the light cycle 16/8 h light/dark and the temperature was maintained between ( 26–28°C ) . For the spikelet density and KRN measurement , the plants were grown at the Uplands Farm Agricultural Station at Cold Spring Harbor , New York between June to October . For SAM measurements , maize seedlings were grown in the greenhouse for 15 days and then dissected and fixed in FAA ( 10% formalin , 5% acetic acid , and 45% ethanol ) . The fixed tissues were subsequently dehydrated with 70 , 85 , 95 and 100% ethanol for 30 min each and then immersed in an ethanol-methyl salicylate solution ( 1:1 ) for an additional 60 min . The tissue was then cleared in 100% methyl salicylate for 2 hours . The SAMs were imaged with a Leica DMRB microscope with a Leica MicroPublisher 5 . 9 RTV digital camera system . For IM measurements , ear primordia 2 mm in length were dissected . The pictures were taken using a Hitachi S-3500N scanning electron microscope or a Nikon SMZ1500 dissection microscope equipped with a camera . The SAMs and ear IMs were measured using Image J . The 4-week old shoot apices were harvested for measuring ZmWUS1 ( GRMZM2G047448 ) expression , and 1-wk old seedlings were used to measure CT2 , CT2CA-mTFP1 , and CT2-YFP as well as PR1 ( GRMZM2G465226 ) and PR5 ( GRMZM2G402631 ) expression . qRT-PCR was performed on a CFX96 Real-Time system ( Bio-Rad ) . Total mRNA was extracted using the Direct-zol RNA extraction kit ( Zymo Research ) . The cDNA was synthesized using the iScript Reverse Transcription Supermix ( Bio-Rad ) according to the manufacturer’s manual . The relative expression level of the targeted genes was normalized using ZmUBIQUITIN . Primers are listed in S1 Table . The trypan blue staining was performed using 1-wk old wild-type and Zmxlg triple mutants , as described with slight modifications [59] . Briefly , the whole shoot was immersed in lactophenol containing 2 . 5 mg/mL trypan blue , and heat in a boiling water for 1 min . Then allowing the samples site at room temperature for overnight . The tissue was cleared in chloral hydrate solution ( 25 g of chloral hydrate in 10 ml of H2O ) for 24 hours at room temperature . The ORF of YFP was inserted between the two amino-terminal α helices , αA and αB of CT2 or CT2CA , as described [10] . The entry clones containing 2x35S promoter , CT2 or CT2CA-YFP , and Nos terminator were assembled in the pTF101 Gateway compatible binary vectors . The ORF of FEA2 was fused with the 6xMyc tag sequence and cloned into the pEARLEY301 vector [60] . All the binary vectors were introduced into the GV3101 Agrobacterium and infiltrated into 4-week-old N . benthamiana leaves with a P19 vector to suppress posttranscriptional silencing [61] . The protein extraction and membrane fraction enrichment were performed as described [10] with some modifications . Briefly , the leaves were harvested 3-day post infiltration and ground in liquid nitrogen to a fine powder then suspended in twice the volume of protein extraction buffer containing 150 mM NaCl , 50 mM Tris-HCl pH 7 . 5 , 5% glycerol , and cOmplete , mini , EDTA-free protease inhibitor ( Roche ) and rotated in a cold room for 15 min . Then centrifuge at 4 , 000g for 10 min at 4°C , followed by filtration through Miracloth ( Millipore Sigma ) , resulting a total protein extraction . The extract was then centrifuged at 100 , 000g for 1 h at 4°C to pellet the microsomal membrane fraction . The resulting pellet was re-suspended in 2 ml extraction buffer supplemented with 1% Triton X-100 with a glass homogenizer . Then the lysates were cleared by centrifugation at 100 , 000g for 1 h at 4°C to remove non-solubilized material . For co-immunoprecipitation experiments , solubilized microsomal membrane fractions were incubated with 30 μl magnetic beads coupled to monoclonal mouse anti-GFP antibody ( μMACs , Milteny Biotec , 130-094-3252 ) for 30 min at 4°C . Flow-through columns were equilibrated using 250 μl membrane solubilization buffer before lysates were added . The MicroBead-bound target proteins were magnetically separated , and washed one time with 250 μl membrane solubilization buffer and three times with wash buffer 1 containing 150 mM NaCl , 50 mM Tris pH7 . 5 , 0 . 1% SDS and 0 . 05% IGEPAL-CA-630 followed by one time with wash buffer 2 containing 20 mM Tris , pH 7 . 5 , supplied by the company . Bound target proteins were eluted with 70 μl 1xSDS loading buffer . Following standard SDS-PAGE electrophoresis and blot transfer , FEA2-Myc protein was detected using an anti-Myc antibody generated from mouse ( Millipore Sigma , 05–724 ) and a secondary HRP-coupled anti-mouse antibody ( GE Healthcare Life Sciences , NA931 ) . CT2 or CT2CA-YFP proteins were detected using an HRP-conjugated anti-GFP antibody ( Miltenyi Biotech , 130-091-833 ) . BLAST search against the protein databases of Arabidopsis , maize , rice , and tomato using Arabidopsis GPA1 and maize CT2 was conducted in Phytozome ( www . phytozome . com ) . The sequences were aligned with Clustal X [62] and the phylogenetic tree was constructed using the neighbor-joining model of MEGA7 [63] . One hundred bootstrap iterations were performed . The coding sequence of CT2 was cloned into pPROEX-His vector between restriction sites EcoRI and XhoI . CT2CA was generated using PCR-based mutagenesis ( Primers are shown in the S1 Table ) . Both constructs were transformed into Rosetta DE3 E . Coli cells for protein expression , as described by Urano et al . with modification [15] . The transformed cells were grown to an OD600 of 0 . 6 prior to induction by 0 . 5 mM 1-thio-β-D-galactopyranoside ( IPTG ) in LB medium for 18 hrs at 16°C . Cells were harvested by centrifugation and resuspended in 150 mM NaCl , 50 mM Tris , 10 mM imidazole , 5 mM β-mercaptoethanol ( β-ME ) , 1 mM MgCl2 , 10 μM GDP , and 10% glycerol , adjusted to a final pH of 7 . 5 , and a cOmplete , mini , EDTA-free protease inhibitor tablet ( Roche ) was added . Cells were lysed by passage three times through a cell disruptor ( Avestin ) at greater than 15 , 000 psi , and the lysate was centrifuged at 29 , 000 g for 30 min to produce a clarified lysate . This lysate was loaded onto a cobalt-charged NTA resin column ( GE Life Sciences ) , and washed with 500 mM NaCl , 50 mM Tris , 20 mM imidazole , 5 mM β-ME , 1 mM MgCl2 , 10 μM GDP , and 5% glycerol , pH 7 . 5 . Bound His-tagged protein was eluted with the same buffer including 300 mM imidazole and 10 mM MgCl2 prior to concentration and loading onto a Superdex-200 size-exclusion column ( GE Life Sciences ) equilibrated with 100 mM NaCl , 50 mM Tris , 10 mM MgCl2 , 5 mM β-ME , 10 μM GDP , and 5% glycerol ( final pH 7 . 5 ) . Peak fractions were pooled and dialyzed to an appropriate buffer for later experiments . The BODIPY-GTP assay was performed as described previously with slight modification [64] . Assays were performed at 25°C in a 200 μl reaction volume in the assay buffer ( 20 mM Tris-HCl , pH 8 . 0 and 10 mM MgCl2 ) with 25 μM purified protein and 50 nM BODIPY-GTP . For competition with non-labeled nucleotides , 25 μM of GTP , GDP , ATP or ADP was added to the assay buffer before starting the reaction . The fluorescence ( excitation 485 nm , emission 528 nm ) was recorded every 10 s for up to 40 min using a BioTek Synergy H4 fluorescence microplate reader . For imaging of CT2CA-mTFP1 , roots were counterstained with 1 mg/ml FM4-64 solution in water for 1min and washed with water . Images were taken with a Zeiss LSM 710 microscope , using 458 nm excitation and 488–515 nm emission for detection of mTFP1 and 514 nm excitation and 585–750 nm emission for detection of FM4-64 . For plasmolysis , the tissues were treated with 20% sucrose for 30 min . The significant differences between multiple groups were analyzed using ANOVA followed by the LSD test with Bonferroni correction in the R statistical programming language ( www . R-project . org ) . All experiments were repeated at least twice and similar results were obtained . The result from one repetition is presented . | Maize is one of the most important cereal crops worldwide . Optimizing its yields requires fine tuning of development . Therefore , it is critical to understand the developmental signaling mechanisms to provide basic knowledge to maximize productivity . The heterotrimeric G proteins transmit signals from cell surface receptor and have been shown to regulate many biological processes , including shoot development . Here we study the role of G protein α subunits in maize development by either making the only canonical Gα constitutively active or mutating all other non-canonical Gα subunits ( XLGs ) . We demonstrate that CT2 and XLGs have both redundant and specialized functions in regulating shoot development . Importantly , we show that a constitutively active Gα functioned as a weak allele , which introduced multiple desirable agronomic traits , such as improved kernel row number and reduced leaf angle . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"plant",
"anatomy",
"enzymes",
"brassica",
"enzymology",
"plant",
"physiology",
"g-protein",
"signaling",
"cereal",
"crops",
"plant",
"science",
"model",
"organisms",
"experimental",
"organism",
"systems",
"crops",
"plants",
"research",
"and",
"analysis",
"methods",
"... | 2018 | Role of heterotrimeric Gα proteins in maize development and enhancement of agronomic traits |
The primary mechanism of action of the antibiotic dihydrostreptomycin is binding to and modifying the function of the bacterial ribosome , thus leading to decreased and aberrant translation of proteins; however , the routes by which it enters the bacterial cell are largely unknown . The mechanosensitive channel of large conductance , MscL , is found in the vast majority of bacterial species , where it serves as an emergency release valve rescuing the cell from sudden decreases in external osmolarity . While it is known that MscL expression increases the potency of dihydrostreptomycin , it has remained unclear if this effect is due to a direct interaction . Here , we use a combination of genetic screening , MD simulations , and biochemical and mutational approaches to determine if dihydrostreptomycin directly interacts with MscL . Our data strongly suggest that dihydrostreptomycin binds to a specific site on MscL and modifies its conformation , thus allowing the passage of K+ and glutamate out of , and dihydrostreptomycin into , the cell .
Streptomycin is a member of the aminoglycoside family of antibiotics , with bactericidal properties . The mechanism of action has been extensively studied since its discovery in 1944 . Ultimately , it binds to the S12 protein of the 30S subunit on the ribosome and inhibits translation and can cause misreading of mRNA [1] . However , it is not clear how streptomycin , or its more potent derivative , dihydrostreptomycin ( DHS ) , gets into the cell . With a diameter of about 10 Å and a net charge of 3+ , DHS is not an obvious candidate for easily crossing a biological membrane . Interestingly , upon treatment with the drug , an outward flux of K+ from the cell has been observed prior to any decrease in cell viability [2] , suggesting either the opening of a channel or some sort of loss of cell membrane integrity . MscL is a bacterial mechanosensitive channel of large conductance that responds directly to tension in the membrane [3–6] . It serves as a biological “emergency release valve” that allows a rapid loss of solutes and osmoprotectants , including K+ and glutamate , in response to a sudden decrease in osmolality in the bacterial environment , thus protecting the cell from lysis [7] . Under very special conditions and mutagenesis , small ions have also been shown to be able to pass through the channel into the cytoplasm [7 , 8] . The crystal structure for the MscL homolog from Mycobacterium tuberculosis ( Mt-MscL ) has been solved by the Rees group revealing a homopentameric channel with each subunit having two transmembrane domains [9 , 10] . The first transmembrane domain ( TM1 ) forms the pore and is linked to the second transmembrane domain ( TM2 ) , which is in contact with the lipid bilayer , via a periplasmic loop that is not as conserved as the rest of the protein . The N-terminal domain is a helix that runs parallel to the cytoplasmic membrane; the C-terminal forms a cytoplasmic helical bundle ( Fig 1A ) . Biochemical , structural-functional , and molecular dynamics ( MD ) studies have all been used to study the channel [5 , 6] . In sum , much of the data has suggested that TM1 must undergo a large clockwise ( >100° ) rotation with both transmembranes tilting into the lipid bilayer to open the large 30 Å pore [11] in the channel . The inappropriate opening of such a large pore would be predicted to have negative physiological consequences; indeed , genetic analysis has shown that mutations in key areas of the MscL protein , including the pore , can lead to a MscL channel that gates more easily and , when expressed in vivo , is detrimental to bacterial growth [12 , 13] . In a previous study , we used a high-throughput screen in search of compounds that would target the Escherichia coli MscL ( Ec-MscL ) , increase its activity , and thus inhibit bacterial growth [14] . Surprisingly , one of the positive hits was DHS . Subsequent studies demonstrated that Ec-MscL , as well as MscL orthologues from Bacillus subtilis and Staphylococcus aureus , when expressed in E . coli , increased the potency of DHS relative to cells that were null for the protein . In addition , we found that MscL expression was required for the DHS-dependent K+ flux observed prior to decreased cell viability . However , from the data , it remained unclear if MscL directly or indirectly influenced MscL function , if DHS uses the large MscL pore as a pathway to enter the cell , and if DHS can modify the MscL channel in streptomycin-resistant cells . MD simulations of biomolecules provide a detailed view of structure and dynamics that complement experiments . Increased conformational sampling , enabled by new algorithms and growth in computer power , [15 , 16] now allows a much broader range of events to be observed , providing critical insights , largely inaccessible to experiments . MD simulation of MscL has been used to investigate how the lipid composition affects the structures and dynamics of Mt-MscL , particularly the charge-rich segment RKKEEP located several amino acid distal to the end of TM2 [17] . The gating mechanisms have been extensively studied by MD simulations for both Mt-MscL [18 , 19] and Ec-MscL [20] . All of the MD studies suggest that the interplay between membrane and protein plays a key role in the opening and closing events of the MscL channel . However , because of a lack of specific agonists and antagonists , MD simulations have not before been used to determine how small biomolecules influence the MscL structure or determine if or how they can pass through the MscL channel . Here , we use a combination of genetic screening , biochemical and mutational approaches , computational analyses , MD simulations , and in vivo assays to determine if DHS’ effects on the MscL channel are direct . Our data strongly suggest that DHS directly binds to a specific location in the channel pore and modifies the channel conformation in such a way as to allow the passage of K+ and glutamate out of , and DHS into , the cell .
Under the hypothesis that mutations in MscL that disrupt DHS binding to the channel could lead to increased bacterial growth in the presence of DHS , we designed a 96-well plate screen utilizing a cysteine-scan library previously described [21 , 22] . Briefly , we screened mutations S2C-E107C , omitting strong gain of function ( GOF ) mutations because of their slowed growth phenotype , and loss of function ( LOF ) mutations because of their inability to gate upon normal stimuli . The concentration of DHS used , 6 . 25 μM , was optimized to be just subthreshold for bacteria expressing the wild type Ec-MscL in order to enhance the sensitivity of the assay . At this concentration , growth of cultures expressing the wild type MscL channel was inhibited , on average , by 50% . Positive results were defined as yielding at least a 50% increased growth over cultures expressing wild type MscL in three independent experiments . Only two cysteine mutants were consistently positive by these criteria in all three experiments: L19C and I25C . When highlighted on a model for the structure of the Ec-MscL [23] , which is based on the Mt-MscL crystal structure [10] , these residues lie in the constriction point of the pore of the channel ( Fig 1A ) . Although these residues are not adjacent , one must appreciate that different residues can be exposed because of the twisting of TM1 upon channel opening [24] , and asymmetry has been predicted by several experiments in the opening of the channel [19 , 25 , 26]; thus , these residues may contribute to not just a single but perhaps an evolving binding pocket for a large chemical such as DHS . Previous studies have demonstrated that mutations and post-translational changes in this region can influence channel sensitivity to membrane tension; hydrophilic substitutions and charges within this region make the channel easier to gate [13 , 21] . Hence , it is conceivable that the multicharged ( 3+ ) DHS , when bound within the pore , could also increase MscL sensitivity . Although the screening data , above , are suggestive that these residues participate in DHS binding and mode of action , we sought additional biochemical support . Because cysteine residues can react with sulfhydryl reagents , they can be modified by methanosulfonate ( MTS ) reagents . MTS-PEG5000 potentially gives us the ability to have site-directed PEGylation of any of our cysteine library residues [27] . If the cysteine mutation lies within the DHS binding site but does not fundamentally alter it , then high concentrations of this antibiotic should compete for MTS-PEG5000 binding . Because of its large size , the PEGylated proteins can then be separated from non-PEGylated ones by SDS-PAGE; any potential inhibition of this PEG modification by DHS competition can easily be assessed . Thus , we performed such an experiment on purified L19C and I25C Ec-MscL mutated channels . As seen in Fig 1B , in both instances , DHS at concentrations greater than 500 μM appeared to compete with MTS-PEG5000 . In contrast , a MscL channel with a cysteine mutation in the periplasmic loop region of the protein , L47C , an area that is not expected to interact with DHS , still showed a gel shift due to PEGylation , but this shift was not inhibited by DHS even at high concentrations ( 1 mM ) . ( Fig 1C ) . Finally , wild type MscL , which contains no endogenous cysteines , showed no apparent increase of mass attributed to PEGylation ( Fig 1C ) . This competition was specific to DHS since another aminoglycoside antibiotic , spectinomycin , at concentrations up to 10 mM showed no decrease in the PEGylation band ( S1 Fig ) . Thus , it appears that higher concentrations of DHS , but not spectinomycin , can block the binding of MTS-PEG5000 to the cysteines located at L19 and I25 within the Ec-MscL . If L19 and I25 contribute to DHS binding and mode of action , then one might expect that variations at these sites between orthologues could lead to differences in DHS sensitivity . Both of these residues are highly conserved . In assessing the cases in which differences in residues are seen at these sites , we noted that for the analogous L19 site , the vast majority of orthologues contained the canonical L , with only a small minority containing an M at this location . The Haemophilus influenzae orthologue ( Hi-MscL ) , which has been functionally characterized [28] , was one orthologue within this minority . H . influenzae is , like E . coli , gram-negative , and Hi-MscL is very homologous to the Ec-MscL channel ( Fig 2A ) . Growth for Ec-MscL , Hi-MscL , and mutated channels was measured in the presence and absence of 6 . 25 μM DHS ( Fig 2B ) . Interestingly , while cells expressing Ec-MscL , which contains the L at position 19 , showed a 30% decrease in growth due to DHS , cells expressing wild type Hi-MscL , which has an M at this position , showed only a 16% decrease . If L19 truly contributes to DHS binding and action , then mutating it to an M in Ec-MscL should decrease the DHS sensitivity of cells expressing this mutated channel . Indeed , L19M Ec-MscL expressing cells showed only a 6% decrease in growth in the presence of DHS . Even more significant was the observation that when M19 of Hi-MscL is changed to L , a “super sensitive” channel is formed; cells expressing this mutated channel show a substantial 51% decrease in cell growth . One trivial explanation for these data could be that the orthologues and mutated channels had different sensitivities to membrane tension and thus were more or less likely to be modified by DHS . Hence , these mutated and WT Hi-MscL channels were assayed by both patch clamp and in an in vivo hypo-osmotic down shock assay to assess their function . If the higher sensitivity of Ec-MscL to DHS was due to increased channel sensitivity to membrane tension , then one would expect it to be more sensitive than Hi-MscL; however , the opposite was observed in patch clamp , and no differences were noted between the wild type channels and their respective mutants ( S2A and S2B Fig ) . In addition , all channels were shown to be functional in vivo ( S2C Fig ) , increasing the viability of osmotically fragile cells from osmotic lysis ( survival close to 80% , although at 74% , the L19M Ec-MscL-expressing cells were close to , but statistically less than , the 80% for wild-type Ec-MscL; LOF mutants are usually defined as a 50% decrease in viability compared to wild type [22] . Presumably , these data translate to the concentration at which DHS modifies the channel . To more directly test this , we used MicroScale Thermophoresis ( MST ) [29 , 30] . In initial experiments using this approach , we found the affinity for the wild type Ec-MscL was relatively weak ( Kd = 9 . 81 mM ) . On the assumption that large conformational changes are affecting the signal , and wild-type MscL may be somewhat resistant to such large changes when solubilized , we used the more sensitive mutant K55T , which has a sensitivity identical to that of the Hi-MscL M19L mutant ( in these mechanosensitivity experiments , MscS is used as an internal standard and thresholds measured [31 , 32]; MscL/MscS threshold ratios of 1 . 11 ± 0 . 05 where n = 5 and 1 . 10 ± 0 . 04 , where n = 10 were determined for Ec-MscL K55T and Hi-MscL M19L , respectively ) . Indeed , using MTS , the K55T Ec-MscL gave a better apparent affinity , relative to the wild type channel , of 680 μM . Note that this binding affinity to DHS , which was performed on solubilized protein , is in the approximate range of those observed in our MTS-PEG5000 inhibition assays , which have also been performed on solubilized channels . Consistent with our interpretation that lower concentrations of DHS effect significant changes on the M19L Hi-MscL , this channel gave an even better apparent affinity of 50 μM ( S3 Fig ) . As expected , no affinity could be measured for the unrelated compound isoniazid , which was used as a negative control . Together , these data derived from the Hi- and Ec-MscL orthologues , and complementary mutations , strongly support the notion that L19 contributes to DHS binding and modulation of the MscL protein . To determine the conformational changes in MscL that may occur upon interactions with DHS , we performed MD simulations . The starting structure for all MD simulations is a homology model based on the crystal structure of Mt-MscL resolved at 3 . 5 Å . The details on homology modeling , system setup , molecular docking , and MD simulation protocols are provided in the Materials and Methods . First , 240 nanoseconds MD simulation was carried out for the homology model in a simulation box consisting of 230 POPC lipids , 95 KCl and 32368 water molecules . The overall root-mean-square deviations ( RMSDs ) of α-carbon , about 3 . 6 Å compared to the crystal structure , are quite reasonable ( see Materials and Methods ) . The last snapshot of the MD trajectory was selected as a representative conformation for the following DHS molecular docking study . The first step of a molecular docking study is to identify possible binding sites . Three possible binding sites were discovered by the SiteID module of SYBYL [33] and shown in Fig 3A . GLIDE ( Grid-based Ligand Docking with Energetics ) [34] docking was then performed for DHS ( Fig 3B ) at each binding site . The best docking score , −7 . 46 kcal/mol , was achieved for the magenta pocket formed by ten amino acid residues ( Fig 3C ) . No meaningful docking pose was discovered at the cyan and green binding pockets . As shown in Fig 3D , two hydrogen bonds are formed between DHS and the surrounding residues . One guanidinium functional group of DHS forms a salt bridge with monomer 3: D290 ( D18 ) . Also , monomer 3: I297 ( I25 ) is a magenta binding pocket residue , and monomer 3 L291 ( L19 ) , in magenta ball and stick representation , has a very close contact with DHS and several binding pockets residues , even though itself is not recognized by SiteID as binding pocket residue . These latter findings are consistent with the experimental finding that DHS is a competitor of binding MTSPEG5000 to cysteine at L19C and I25C . We previously speculated that DHS could pass through the MscL channel if the channel partially or fully opened [14] . We therefore tested if DHS could be seen to pass through the channel using MD simulations . First , DHS was manually placed in the center of the Ec-MscL pore with ten different orientations . For each docking orientation , 50 nanosecond MD simulation was first performed with the secondary structure domains ( TM1 , TM2 , S1 and C-terminal helix ) and lipids strongly restrained followed by another 100 nanosecond regular MD simulation without any restraint . In order to observe the passages of DHS through the Ec-MscL channel within a short timeframe for MD simulations , an external electric field ( EEF ) was applied to DHS to accelerate the process . Indeed , a previous study has found that increased membrane potentials in vivo increase the potency of DHS [35] . Note that the experiment was designed so that only the DHS molecule felt this field . Only the best docking orientation , which was recognized according to both the global and ensemble conformational energies sampled by the 100-nanosecond MD simulations , was selected for the subsequent MD simulations with EEF ( Fig 4A ) . With the EEF strength being set to 0 . 2 volt/Å along the Z-axis , we observed the complete procedure of the passing-through event within 12 nanoseconds ( Fig 4B ) . These data have been repeated a total of five times ( twice where the Cl- molecules were set to balance the positive charges ) , all with consistent results with the passage of DHS through the MscL channel . The modified pore is more than accommodating for DHS , which does not have to dehydrate for passage; in fact , DHS is surrounded by tens to hundreds of water molecules when it passes through the channel . A video of the passing-through event is provided as S1 Video . We defined a parameter , ΔZ , the Z-coordinate difference of two geometric centers formed by DHS and five LYS residues located at the end of TM2 ( Fig 4 ) , to describe the passing through event . As demonstrated in Fig 4B , DHS passed through Ec-EscL four times , and the last three took a much shorter time since the channel had already changed conformation . It is also clear that there is a long time period ( about 8 nanoseconds ) that DHS stayed in an intermediate state with a characteristic ΔZ value of 25 Å . We further characterized the passing-through event by calculating the MM-PB energies of all the collected MD snapshots . As shown in Fig 4C , the intermediate state has favorable MM-PB energies consistent with the fact that DHS stays in the middle of the channel for about half of the time of the simulation . The methods and computational details on PB calculations are presented in the Materials and Methods . Moreover , we calculated the displacements of K+ during the passing-through event . The average displacements along Z-axis of 95 K+ , about 6 Å towards the outside of the membrane was observed ( Fig 4D ) . This means that the channel is more open and , although there is no electric field on the ion , it begins to migrate in the direction it would need to flow for efflux from the cell . This finding is consistent with the observation of K+ efflux in in vivo experiments [2 , 14] . The whole procedure can be roughly divided into five stages according to the residence time and the MM-PB energies: ( I ) initial stage ( 0 . 72 nanoseconds ) , ( II ) fast transition stage ( 0 . 4 nanoseconds ) , ( III ) transition stage ( 8 . 8 nanoseconds ) , ( IV ) pre-exiting stage ( 0 . 76 nanoseconds ) , and ( V ) exiting stage ( 0 . 36 nanoseconds ) . The representative conformations of the five stages are shown in Fig 4E . From Stage I to V , DHS moved from the MscL pore to the cytoplasmic side of the membrane progressively . How each secondary structure units ( transmembrane domains TM1 and TM2 , N-terminal S1 and C-terminal helix ) fluctuate along the MD simulation trajectory is discussed further in the Materials and Methods . Interestingly , the RMSDs of the two transmembrane domains and S1 increased slowly from the initial stage to the intermediate state stage and then the final stages; on the contrary , the C-terminal domain has a dramatic RMSDs surge at the final stages ( Stages VI and V ) to accommodate DHS exiting from the channel . The RMSDs of the five representative conformations are summarized in S1 Table . Moreover , we also quantitatively characterized the helical rotation during the MscL channel gating induced by the DHS passing-through event . The axis angles calculated from rotational matrices of least-squares fittings were listed in the S2 Table for the five representative conformations; they are divided by the S1 helices ( S2A Table ) , TM1 helices ( S2B Table ) , TM2 helices ( S2C Table ) , and C-terminal helices ( S2D Table ) . Compared to the closed conformation , the first TM1 helix rotates in a corkscrew manner ( clockwise as seen from the periplasm ) from 10 . 5° in Stage I to 16 . 3° in Stage III and 27 . 5° in Stage V . Large movements also were observed for the C-terminal helical bundle , which largely moved as a unit: the rotational axis angles increased about 10 to 15° for four of the five helices from Stage I to Stage V , although the final exiting stage contributes the most changes . Key residues that directly interact with DHS were identified for representative conformations of the five stages ( Fig 5 , S4 and S5 Figs ) . The conformations of DHS in the five stages are similar , particularly for Stages I-III . DHS adopts a prolonged conformation to facilitate its exit from the channel in Stages IV and V ( S6 Fig ) . We then characterized the MscL channel by calculating the pore radii along the channel coordinate . The “W” shape plot in S7 Fig suggests that there are two bottlenecks ( around -50 and -10 Å ) and the latter is narrower and needs more time for opening . This result is consistent with our MD result ( Fig 4B ) . This plot also suggests that the passage of small molecule like DHS does not lead to a fully open Ec-MscL channel . The MD simulation experiments have made predictions of points of interaction between DHS and the MscL protein . Using both the magenta binding site from SiteID as well as several representative snapshots selected from the MD simulation of DHS passing through Ec-MscL channel , ten residues in MscL were predicted to participate in this interaction: V16 , V17 , D18C , L19 , A20 , A114 , P115 , T116 , K117 , and E118 ( Fig 5 ) . We therefore subjected these cysteine-mutated residues to the MTS-PEG5000 competition assay described above . We chose two residues reflecting the two distinct domains proposed to interact with DHS ( A20C , and E118C ) As shown in Fig 6A , each is positive , as predicted , suggesting they are indeed involved in DHS binding and interactions; the remaining eight residues were also found to be positive and are shown in S8 Fig . In addition , we have tested five residues in the area predicted NOT to interact with DHS , N103 , K105 , K106 , A110 , and D127 . In Fig 6B , we show two of the more distant residues , K105C and D127C , and neither showed inhibition of the PEGylation in the presence of DHS . Shown in S8 Fig are the additional mutated channels; K106C and A110C were negative as predicted , but N103C was positive . This latter finding may be explained by a previously defined conformational change of the protein that makes this site inaccessible as the channel opens; we found that this residue becomes buried within the lipid bilayer upon gating [36] . The evidence that N103 becomes buried within lipids includes phenotypes , including channel sensitivities , obtained after residue modifications designed to embed the residue in the membrane or stabilize it at the membrane interface . In addition , the residue was mutated to tryptophan , and the amount of fluorescence quenching by brominated lipids was measured upon channel gating . All of the data supported a model in which this site becomes buried within the lipids in an early transition state as the channel opens . Assuming this to be the case , it is therefore not surprising that the site is not PEGylated when in the presence of DHS; the DHS is opening the channel and thus burying the 103 site and preventing modification . Given this background , we decided to evaluate the contact numbers for N103 interacting with lipid hydrophobic sites within the MD simulations . The average contact numbers for N103 are 2 . 52 , 4 . 03 , 3 . 38 , 3 . 30 , and 2 . 42 for the five conformational stages defined in Fig 4 . In contrast , the average contact number is only 0 . 01 for the closed conformation . It is noted that the fast transition stage and the transition state stage have the largest and the second largest contact numbers , and the contact numbers decrease gradually once DHS begins passing through the channel . This MD result suggests that N103 can regulate Ec-MscL gating by increasing or decreasing the lipid exposure; these data are entirely consistent with previous experimental data [36] . It therefore seems likely that upon interacting with DHS , the channel undergoes a conformational change that buries this residue within a hydrophobic pocket making it inaccessible for MTS-PEG5000 modification . We previously demonstrated that MscL expression changes the potency of DHS [14] . The finding that increasing the concentration of DHS still leads to decreased viability in sensitive cells suggested MscL is not the only pathway into the cytoplasm , thus explaining why the mscL gene has never been identified in streptomycin suppressor assays . In addition , there is no obvious slowed-growth or decreased-viability phenotype for DHS-resistant strains when in the presence of DHS , demonstrating , not surprisingly , that MscL is not the primary mode of action of the drug . However , we have previously shown that upon treatment with DHS , there is a MscL-dependent flux of both K+ and glutamate in bacteria that were sensitive to streptomycin [14] , strongly suggesting that the DHS-dependent flux of these osmolytes is due to MscL activation . This begs the question: can we measure subtler MscL- and DHS-dependent physiological phenotypes in DHS-resistant cells using flux assays ? Thus far , we have not seen a K+ flux in DHS-resistant E . coli strains . There may be two explanations for this . First , there may be other factors playing a role in the perceived K+ flux; indeed , one interpretation of a previous study was that misfolded and truncated transmembrane proteins may intercalate into the bilayer and compromise membrane integrity [37] . It is possible that this compromising of the membrane is not direct , but in fact the aberrant proteins add tension in the membrane and increase MscL sensitivity to DHS . Under this scenario , MscL-dependent fluxes would only be seen in DHS-sensitive cells . However , a second explanation is that the many high-affinity K+ pumps largely compensate for fluxes in DHS-resistant cells , which are presumably healthier than DHS-sensitive cells upon DHS treatment . Note that these two explanations are not mutually exclusive and both may play a role . To avoid the latter complication of compensatory mechanisms , we assayed for glutamate , which is accumulated largely through synthesis rather than being pumped into the cell by high-affinity pumps; i . e . , fluxed glutamate should not be rapidly taken back into the cell . We used two independent DHS-resistant cell lines from different parental strains that are MscL deficient: MJF367 and PB104 from Frag 1 and AW405 parental strains , respectively . As expected , upon DHS treatment there was no significant loss of glutamate from cells expressing empty plasmid in either of the cell lines ( 98 . 1% of the glutamate remained in the cells for MJF367 and 99 . 2% for PB104’s; t test p-values >0 . 1 and n = 3 for both ) . However , the amount of glutamate remaining in the MJF367 cell line expressing MscL was 78 . 5% , and PB104’s was 83 . 7% . Both of these values , while not as great as the 52% loss seen in streptomycin-sensitive cells [14] , were significantly different from empty plasmid ( p-values <0 . 005 for MJF367 and < . 05 for PB104’s; n = 3 for each ) . Viability experiments were also performed in conjunction with the glutamate experiments , and no significant loss in viability was seen for any strain under any condition . Thus , a modest but statistically significant DHS- and MscL-dependent glutamate flux can be measured in DHS-resistant cells independent of a decreased viability , strongly suggesting that DHS can influence MscL gating independent of DHS toxicity . Together , the data support the direct binding of DHS to a specific site within the MscL channel and its ability to modify MscL such that DHS can pass through the channel into the cytoplasm of the cell . These changes apparently occur in both streptomycin-sensitive as well as streptomycin-resistant cells but may be amplified in the latter due to a previously suggested increase of abnormal membrane proteins adding tension in the membrane and thus increasing the sensitivity of MscL .
Early in the study of the MscL channel , a forward genetics study found that mutations in the channel could lead to cell slowed growth or even death [13] . Briefly , the channel was randomly mutagenized , placed under transcriptional control of the inducible LacUV5 promoter , plated , then replica plated onto plates containing IPTG , which induced expression . Colonies that grew less or failed to grow on the IPTG-containing plates were further studied . In sum , the data demonstrated that mutations at several locations , including what is now known to be the pore constriction site , led to channels that were more sensitive to membrane tension as assayed by patch clamp and , when expressed , lead to the decreased growth or viability observed . On the assumption that small chemical compounds may also be able to make the channel more sensitive , and thus may lead to the development of channel agonists and potential antimicrobial agents , we developed a high-throughput screen [14] . Because MscL senses membrane tension , any compound that intercalates into the membrane could lead to a more sensitive channel; indeed , amphipaths have previously been shown to gate mechanosensitive channels including MscL [38 , 39] . Hence , in the secondary screen we tested the compounds on cells expressing only E . coli MscS , a member of an unrelated second family of bacterial tension-sensing mechanosensitive channels [7] . Surprisingly , four known antibiotics were identified in the screen: DHS , spectinomycin , viomycin , and nifuroxazide . Of the four , DHS appeared to be the most specific and potent and therefore further characterized . We found that MscL expression was required for the DHS-dependent K+ efflux observed prior to a decrease in cell viability . Further , we demonstrated that a DHS-dependent flux of another , larger osmoprotectant fluxed by MscL , glutamate , was also observed on the same time scale and also found to be MscL dependent; here , we demonstrate that some glutamate flux can even be measured in DHS-resistant cells in a MscL-dependent manner . Moreover , MscL channel activity increased in response to DHS as assayed by patch clamp [14] . This latter finding was originally generated using the mutated channel used in the initial screen; not surprisingly , this finding holds true for the wild type E . coli MscL channel using an electrode back-fill approach [31 , 40] , as shown in S9 Fig . However , obvious questions remained: Was DHS directly binding to and modifying the MscL channel ? If so , where is this binding site located ? And is MscL one conduit for passage of the large bulky DHS molecule into the cell cytoplasm ? Here , we present several lines of evidence that DHS binds to the MscL pore , including an in vivo screen of over 90 mutated channels , a MTS-PEG5000 competition assay , analysis of an orthologue with variability within the predicted binding site ( including mutagenesis of the site ) , and MD simulations . The latter showed DHS modifying the channel and passing through it . Finally , the MD simulations made predictions of residues that interact with DHS upon this modification and passage; these predictions were tested by the MTS-PEG5000 competition assay and shown to be true . The most likely binding site using the channel-closed conformation ( predicted by SideID , which is in magenta in Fig 3C ) is relatively stable during DHS passing through the Ec-MscL channel ( S10 Fig ) as well as the binding cavity volumes predicted by the SiteID software . How DHS binding to this site affects the passage of DHS through the channel is an interesting topic and will be pursued in future study . Current models for the open structure of the channel suggest that TM1 corkscrews early in the gating process about 110° in the membrane in a clockwise direction when viewed from the periplasm , then the transmembranes tilt in the membrane and separate like the iris of a camera , opening a large pore approximately 30Å in diameter [5 , 6] . Many of the initial structural changes have been supported by recent crystal structures of the archaeal MscL homolog from Methanosarcina acetivorans trapped in both the closed and closed-expanded intermediate states [41] . The MD simulations of MscL in the presence of DHS do not show full channel opening . Perhaps it is the short time scale of the simulations , but it may also be that additional in vivo environmental factors including ambient membrane tension resulting from the predicted high cell turgor [42] , or the electric field , predicted to be about −140 mV across the bacterial membrane [43 , 44] , play a role in the sensitivity of the channel to DHS modulation . Regardless , many of the changes observed in our simulations are consistent with an opening channel . For example , the corkscrewing ( up to 27 . 5° ) of TM1 is observed in the simulations , as well as a separation of the helices around the pore/vestibule . Potassium , an osmoprotectant that normally can flux through the open channel , even though not under an electrochemical gradient , is observed to move into the channel vestibule; this is also consistent with the DHS-dependent K+ flux observed prior to decreased cell viability [2 , 14] . The observation that the C-terminal helical bundle moves as a unit , remains intact , and the DHS passes asymmetrically through the linkers between TM2 and the C-terminal helical bundle is also expected , because several studies suggest that the bundle is quite stable [45 , 46] and remains intact upon gating [47] . Finally , the cytoplasmic residue N103 has been predicted to insert into , then out of , the membrane in a piston-like fashion upon gating [36]; this is also observed in the MD simulations and is a likely explanation of why this residue gives an apparently false positive in the MTS-PEG5000 competition assay . In sum , although a full opening of the MscL channel is not observed , many of the conformational changes observed are consistent with the initial movements toward a fully open channel . The finding that DHS binds the pore vestibule is , in hindsight , not surprising . The random mutagenesis study discussed above found that many mutations within this region leads to increased channel sensitivity [13] . In addition , it appeared that increasing hydrophilicity or adding charges in the pore/vestibule domain leads to inappropriate channel activity and compromised cell growth [24] . These findings , in conjunction with the crystal structure of the Mt-MscL , led us to propose the “hydrophobic lock” hypothesis , which states that it is the transient exposure of hydrophobic residues within the pore to an aqueous environment that is the major energy barrier for channel opening [48] . Interestingly , DHS interactions appear to be at the interface of two subunits within the pore vestibule . This is reminiscent of the acetylcholine binding site of the nicotinic acetylcholine receptor [49 , 50]; presumably , changes in protein—protein interactions at subunit interface induced by ligand binding can more easily lead to changes in global conformation , and thus channel gating . Indeed , part of the MscL binding pocket is predicted to contain residues from a neighboring TM2 domain . We have previously studied this region of the protein , both the S1 domain [51] , as well as the cytoplasmic portion of TM2 [36 , 52 , 53] , and even the changes in interactions that occur between these domains upon channel gating [54] . These previous studies have found this region to be highly dynamic and undergo large conformational changes upon gating . In sum , the data indicate that DHS binds to and modifies what could be referred to as the gate of the channel , i . e . , the subunit interface at the most constricted portion of the pore , thus leading to a conformational change that allows the efflux of osmoprotectants and the passage of DHS into the cytoplasm .
Mutations for this study were generated using the mega primer technique as previously described [22] . His 6 tags were added to existing mutants on the C-terminus with standard PCR technique . The resulting PCR products were ligated into either pB10d [51] or pET21a ( Novagen ) . Giant spheroplasts were generated from the E . coli strain PB104 ( ΔmscL::Cm ) [31] or MJF612 [56] and used in patch-clamp experiments as described previously [57] . Excised , inside-out patches were examined at RT under symmetrical conditions using a buffer comprised of 200 mM KCl , 90 mM MgCl2 , 10 mM CaCl2 , and 5 mM HEPES pH 6–7 ( Sigma , St . Louis , MO ) . Recordings were performed at −20 mV ( positive pipette ) . Data were acquired at a sampling rate of 20 kHz with a 5 kHz filter using an AxoPatch 200B amplifier in conjunction with Axoscope software ( Axon Instruments , Union City , CA ) . A piezoelectric pressure transducer ( World Precision Instruments , Sarasota , FL ) was used to monitor the pressure throughout the experiments . As previously described [13 , 32 , 57] , MscS was used as an internal standard for determining MscL sensitivity . Measurements were performed using Clampfit10 from Pclamp10 ( Axon Instruments , Union City , CA ) . | Streptomycin is one of the original and best studied antibiotics . Its primary mechanism of action is the interference with protein synthesis by binding to and modifying the function of the bacterial ribosome . However , the antibiotic is quite large , bulky , and is charged , so the mechanisms by which it accesses the inside of the bacterial cell have remained largely unknown . Previously , we have found that the expression of a bacterial mechanosensitive channel , MscL , increases the potency of a variant of this antibiotic , dihydrostreptomycin . Here , we define the dihydrostreptomycin binding site on the MscL channel . We also show how this antibiotic modifies the channel and opens the pore , allowing the diffusion of solutes , such as potassium and glutamate , from the cytoplasm of the cell out to the medium . Finally , we provide evidence that dihydrostreptomycin can pass through MscL and that the channel is thus one pathway by which the drug accesses the inside of the bacterial cell . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"antimicrobials",
"medicine",
"and",
"health",
"sciences",
"neurochemistry",
"crystal",
"structure",
"chemical",
"compounds",
"built",
"structures",
"engineering",
"and",
"technology",
"drugs",
"condensed",
"matter",
"physics",
"microbiology",
"neuroscience",
"organic",
"... | 2016 | Dihydrostreptomycin Directly Binds to, Modulates, and Passes through the MscL Channel Pore |
Dengue viruses ( DENV ) are enveloped single-stranded positive-sense RNA viruses transmitted by Aedes spp . mosquitoes . There are four genetically distinct serotypes designated DENV-1 through DENV-4 , each further subdivided into distinct genotypes . The dengue scientific community has long contended that infection with one serotype confers lifelong protection against subsequent infection with the same serotype , irrespective of virus genotype . However this hypothesis is under increased scrutiny and the role of DENV genotypic variation in protection from repeated infection is less certain . As dengue vaccine trials move increasingly into field-testing , there is an urgent need to develop tools to better define the role of genotypic variation in DENV infection and immunity . To better understand genotypic variation in DENV-3 neutralization and protection , we designed and constructed a panel of isogenic , recombinant DENV-3 infectious clones , each expressing an envelope glycoprotein from a different DENV-3 genotype; Philippines 1982 ( genotype I ) , Thailand 1995 ( genotype II ) , Sri Lanka 1989 and Cuba 2002 ( genotype III ) and Puerto Rico 1977 ( genotype IV ) . We used the panel to explore how natural envelope variation influences DENV-polyclonal serum interactions . When the recombinant viruses were tested in neutralization assays using immune sera from primary DENV infections , neutralization titers varied by as much as ∼19-fold , depending on the expressed envelope glycoprotein . The observed variability in neutralization titers suggests that relatively few residue changes in the E glycoprotein may have significant effects on DENV specific humoral immunity and influence antibody mediated protection or disease enhancement in the setting of both natural infection and vaccination . These genotypic differences are also likely to be important in temporal and spatial microevolution of DENV-3 in the background of heterotypic neutralization . The recombinant and synthetic tools described here are valuable for testing hypotheses on genetic determinants of DENV-3 immunopathogenesis .
Dengue virus ( DENV ) is an enveloped ( + ) RNA virus in the family Flaviviridae , genus Flavivirus transmitted by the bite of Aedes spp . mosquitoes . DENV occurs throughout the tropics and subtropics and infects approximately 50 million individuals annually . There are four distinct serotypes , DENV-1–DENV-4 . While prospective studies have found that most infections are asymptomatic , a proportion of infected persons will develop symptoms that include fever , rash and myalgia [1] , [2] with 2% or less developing the severe disease syndromes of dengue hemorrhagic fever/dengue shock syndrome ( DHF/DSS ) [2] , characterized by hemorrhage , vascular leakage , hypovolemia and , if untreated , shock , end organ failure and death [3] . Approximately 15 , 000–30 , 000 persons die annually from DHF [1] . DHF/DSS has been classically associated with secondary infections that occur in the context of pre-existing heterotypic immunity - leading to hypotheses that DHF/DSS is an immune mediated phenomenon driven by cross-reactive DENV antibodies and/or or DENV specific CD8+ T-cells ( for reviews see: [4] , [5] . Virus genotype also clearly plays an important role in severe disease pathogenesis , as . Multiple studies of DENV molecular epidemiology have found associations between circulating virus genotype and disease severity [6]–[12] . However , the genetic basis of these virulence differences has not been deciphered . One of the fundamental barriers to DENV vaccine development has been concern that a DENV vaccine must be broadly protective against all four serotypes or recipients will risk secondary-like infection and the severe disease associated with naturally acquired secondary infection . Most vaccine trials have assessed protection against all four serotypes using prototype or vaccine related virus isolates [13] and studies need to address the degree to which intra-serotype genotypic differences may affect antibody-mediated immunity to any of the DENV serotypes , including DENV-3 . While genotype specific genetic differences are scattered across the viral genome , the envelope glycoprotein ( E ) is the main target of neutralizing human antibody and is one logical first choice for assessing the genetic basis of differential antibody mediated neutralization of DENV-3 infection . The E glycoprotein exists as a homo-dimer with 3 distinct domains – I , II , and III [14]–[17] , that , on the mature DENV virion , are arranged in a flat herringbone pattern with icosahedral symmetry [14] . Domains I ( EDI ) and II ( EDII ) are linearly discontinuous and fold to form a central eight-stranded ß barrel ( domain I ) with a lateral protrusion ( domain II ) that contains the highly conserved fusion loop required for virion fusion with endosomes . Domain III ( EDIII ) is a continuous peptide that extends from domain I and forms an Ig like fold that is believed to be the ligand for an as yet unidentified cellular receptor . A successful dengue vaccine should induce broadly protective antibodies against all geographic variants of each serotype . The dengue community has long held that primary infection with one serotype confers long-lasting immunity to that serotype , irrespective of the infecting virus genotype . This is based principally on early human challenge trials [18] and multiple observational studies that have shown that , within a particular region , re-infection with the same serotype generally does not occur . Geographic partitioning of DENV genotypes significantly limits our understanding of the role of strain variation in protective immunity , as the vast majority of DENV infected persons in endemic regions never travel to regions where other DENV genotypes are circulating . Several recent findings indicate that genotypic variation may be important in immunity . Recent studies of DENV-3 strain variants using recombinant proteins and whole virus have found that neutralization mAbs raised against one DENV-3 genotype have limited neutralization activity against heterologous genotypes [19]–[22] . After primate vaccination , studies with polyclonal immune sera have also demonstrated variable neutralization of DENV3 strains [23] . In a study of pediatric dengue cases in Thailand , investigators observed significant differences in the ability of sera to neutralize reference and clinical strains of DENV3 [24] . A recent WHO report on dengue neutralization testing highlighted the need for evaluating vaccine induced immune responses using contemporary strains representing the different serotypes and genotypes of dengue [25] . DENV-3 consists of four distinct genotypes: I , II , III and IV , each originally associated with a specific geographic region [26] . Currently genotype I and II are circulating in Asia , genotype III is circulating in the Indian subcontinent , Africa and Latin America , and genotype IV appears to have been displaced but occurred throughout the Caribbean in the 1960s and 70s [7] , [26]–[31] . Here we described the construction of a four-fragment DENV-3 infectious clone platform and a panel of isogenic DENV-3 recombinant viruses that captures DENV-3 E glycoprotein genotypic heterogeneity . While our approach is novel for flaviviruses , human coronavirus ( CoV ) investigators have used a similar system to introduce large , synthesized genomic elements into recombinant viruses to investigate genetic variability in CoV biology and pathogenesis ( see [32]–[35] for examples ) . The CoV systems are a powerful tool for expanding understanding of genetic differences in CoVs and the application to Flaviviruses may prove similarly powerful . We subsequently tested the isogenic recombinant viruses against a panel of immune sera from people exposed to primary or secondary DENV infections . These data demonstrate a role for natural epitope variation in virus neutralization and escape . The molecular clone should also prove to be a valuable tool for studying a variety of other aspects of DENV-3 biology , pathogenesis , immunopathogenesis , epitope mapping and evolution .
The Institutional Review Board of the University of North Carolina at Chapel Hill approved the protocol for recruiting and collecting blood samples from people . Written informed consent was obtained from all donors . Vero E6 cells ( ATCC CRL-1586 ) were maintained in MEM supplemented with 10% FCS ( Gibco ) , non-essential amino acids ( Gibco ) , L-glutamine ( Gibco ) and Anti-Anti antibiotic mix ( Gibco ) at 37°C in 5% CO2 . C6/36 cells ( ATCC CRL-1660 ) were maintained in MEM supplemented with 5% FCS , non-essential amino acids , L-glutamine and Anti-Anti at 28°C in 5% CO2 . The cloning strategy for the DENV-3 clone is illustrated in Figure 1A , and based on strategies employed with CoVs to circumvent sequence instability problems in E . coli [36] , [37] . The clone parent is a 1989 Sri Lankan DENV3 isolate ( genotype III ) designated UNC3001 ( submitted to GenBank ) . To isolate the DENV-3 sub-clones , reverse transcription was performed with AMV reverse transcriptase ( Roche ) and oligodeoxynucleotide primers according to the manufacturer's recommendations using primer BsmbIDen . Following cDNA synthesis , the cDNA was amplified by PCR with Expand Long TAQ polymerase ( Boehringer Mannheim Biochemical ) with cycle settings based on the size of the amplicon . The Dengue genome was amplified from cDNA and cloned as a set of four fragments ( Figure 1 and Text S1 ) . The first fragment , A , was PCR amplified using primer set DEN#1 and DEN2kb− . These primers created a T7 RNA promoter at the 5′ end of the fragment and a BsmBI restriction site at its 3′ end , respectively . The PCR product was gel isolated ( Qiagen QIAquick Gel Extraction Kit ) and then cloned into the pCR-XL TOPO cloning vector ( Invitrogen ) . The second fragment , B , was amplified using primers DEN2kb+ and DENBGL4− . The DEN2kb+ primer introduced a BsmBI site that allowed for the directional ligation of fragments A and B ( Figure 1 and Text S1 ) . The DENBGL4− primer introduced silent changes in the Dengue genome between nucleotides ( nt ) 3150 and 3160 to create a unique BglI site without altering the amino acid sequence . Fragment C was amplified with primers DENBLG3+ and DEN7kb− . This primer set duplicated the BglI site at the 3′end of the B fragment and a naturally occurring BglI site at nt 7031 . The PCR amplicons for both fragments B and C were gel isolated and cloned into the pCR-XL TOPO cloning vector . Fragment D was amplified with primers DEN5kb+ and BsmBIDen . This PCR product , which went from approximately nt 5100 to the 3′ end of the Dengue genome , contained two BglI sites; one at nt 7032 and the other at nt 10186 . The BglI site at nt 10186 was removed using overlapping PCR . Two amplicons – 3′ and 5′ , were generated using primers Dengue15 and Den10198 and primers Den10166 and BsmBIDen , respectively . These two amplicons were joined in an over-lapping extension PCR reaction . The resulting product was digested with SapI and ligated to SapI digested DEN D fragment . This final cDNA fragment , which now had the BglI site at 10186 knocked out , was gel isolated and cloned into the Big Easy v2 . 0 Linear cloning vector ( Lucigen ) . Four to six clones of each fragment were sequence verified . The four DEN cDNAs were isolated from plasmids and directionally ligated to create a full-length cDNA of the dengue viral genome . This full-length cDNA contained only the introduced nucleotide changes , all of which were silent , and could be transcribed with T7 polymerase ( Ambion ) . This RNA produced infectious dengue virus when electroporated into Vero E6 cells . To construct E glycoprotein variant clones ( Figure 1B ) , synthesized envelope genes ( nucleotides 913–2416 of the Dengue genome ) were delivered in puc57 plasmids ( Bio Basic ) . The portion of these envelope genes that needed to be inserted into the A plasmid , was PCR amplified with either a puc57 forward or reverse primer and the Den2kb− primer ( Text S1 ) . These products were digested with BstEII and BsmBI and ligated into the A plasmid which had been digested with the same enzymes . Dengue B plasmids containing the envelope variants were generated by first PCR amplifying the synthetic genes with the Den 2kb+ primer and primer EGENE− ( Text S1 ) and the parent B fragment with primer EGENE+ and DENBGL4− ( Text S1 ) . These products were then digested with BsaI and ligated together . Finally , the ligations were gel purified and cloned into the pCR-XL TOPO cloning vector . To replace the parent clone prM/M gene with a genotype I prM/M gene , RNA from our lab stock genotype I virus UNC3043 , was reverse transcribed with random hexamers and the cDNA was PCR amplified with primers Dengue01+ and Denv900 . The resulting amplicon was digested with BstAPI and PflMI . This product was ligated into the DEN A plasmid corresponding to the Indonesia 1982 genotype I E gene that had been digested using the same enzymes . The resulting plasmid DEN A was sequence verified and used to construct the genotype I recombinant virus . Each plasmid was transformed and propagated in E . coli TOP10 competent cells ( Invitrogen ) and grown on LB plates with selective antibiotics ( A , B , and C containing plasmids selected with kanamycin , D with chloramphenicol ) at 28 . 5°C for 24 hours . Individual colonies were picked , screened and sequenced . The plasmids were subsequently grown to high concentration in selective LB , plasmid purified ( Qiagen Mini-Spin Kit ) and digested as follows according to manufacturers instructions: DEN A with SpeI ( NEB ) followed by calf intestine phosphotase ( NEB ) and BsmBI ( NEB ) yielding a 2 . 0 kb fragment; DEN B with BglI ( NEB ) and BsmbI yielding a 1 . 1 kb fragment; DEN C with BglI yielding a 3 . 9 kb fragment; and DEN D with BglI and BsmbI yielding a 3 . 0 kb fragment . Fragments were gel-isolated ( Qiagen Gel Extraction Kit ) on 0 . 8% agarose gel , mixed in equivalent copy number and ligated with T4 ligase ( NEB ) overnight at 4°C . Full-length transcripts of DENV-3 cDNA constructs were generated in vitro as described by the manufacturer ( Ambion , Austin , Tex; mMessage mMachine ) with the following modifications: For 30-µl reaction mixtures supplemented with 4 . 5 µl of a 30 mM GTP stock , resulting in a 1∶1 ratio of GTP to cap analog and incubated at 37°C for 2 hours . Vero cells were grown to 75% confluence , trypsinized and resuspended in RNAse free PBS at 107 cells/ml . RNA transcripts were mixed with 800 µl of the Vero cell suspension in an electroporation cuvette , and four electrical pulses of 450 V at 50 µF were given with a Bio-Rad Gene Pulser II electroporator . The transfected Vero cells were seeded at 5×106/ml in 75-cm2 flask and incubated at 37°C for 4 days . Two to five ml of supernatant from electroporated Vero cells were passaged on day 4 to 75% confluent uninfected Vero cells in a 75 cm2 flask . Fresh media was added to a final volume of 15 ml . Seven day supernatants were harvested , supplemented to 30% FBS , clarified by centrifugation and frozen at −80°C or passaged serially to amplify a working virus stock . At the time this study was initiated , there were 164 unique , full-length DENV-3 envelope genes available in Genbank , and these sequences were added to 11 Sri Lankan DENV-3 sequences from our laboratory . The 175 envelope amino acid sequences were aligned using ClustalX version 1 . 83 [38] , and one representative sequence was selected for each DENV-3 genotype . The representative sequence was chosen based on amino acid conservation within the genotype cluster , with sequences closest to consensus with no outlier amino acids selected as the representative . Representative sequences chosen were: Genotype I Indonesia 1982 ( GenBank accession# DQ401690 . 1 ) ; Genotype II Thailand 1995 ( GenBank accession# AY676376 ) ; Genotype III Cuba 2002 ( GenBank accession# AY02031 ) ; and Puerto Rico ( PR ) 1977 ( GenBank accession# AY146761 ) . All viruses used in the subsequent experiments were passage three propagated in Vero cells . All passage three clones were sequence verified using previously described methods [39] . To assess viral replication kinetics , each of the DENV-3 clones was inoculated in triplicate onto 95% confluent monolayers of Vero or C6/36 cells in 6 well plates at a multiplicity of infection ( m . o . i ) of 0 . 01 ffu/ml . Cells were incubated at either 37°C for Vero or 27°C for C6/36 cells under maintenance media conditions for the cell line for 60 minutes , after which the innocula were removed and cells washed twice in 3 ml of PBS . Each monolayer was covered in a total volume of 5 ml media . After 60 min , 200 ul of cell supernatant , designated as the Day 0 sample , was taken in duplicate with equal volume media replaced . Samples were supplemented with 30% FCS , clarified by centrifugation and stored at −80°C . Samples were taken in the same manner every 24-hrs for 6 additional days . Virus titers were determined as described below . Sera were collected from adult volunteers with histories of DENV infection [40] and one anonymous donor with dengue infection confirmed by serology ( sample 109 ) . Sera were characterized by flow cytometry at UNC [41] , PRNT60 at the NIH , Bethesda , MD , or PRNT90 at CDC San Juan to confirm past exposure to primary or secondary DENV infections and also to identify the serotype responsible for primary infections . We note that we cannot establish the infecting virus genotype of our experimental sera on neutralization patterns alone . However , only genotypes I , II , and III are currently circulating , and our samples almost certainly capture genotype II ( Thailand ) and III ( Latin America ) based on donor travel history . The FRNT procedure is based on a method previously described by Whitehead [42] . Briefly , twenty-four well plates were seeded with 5×104 Vero cells in MEM supplemented with 5% fetal bovine serum ( FBS ) and grown for 24 hours . Growth media was removed . For virus titration , virus stocks were diluted serially ten-fold from 10−1 to 10−6 and 200 ul of each dilution added to individual wells . After 1 hr incubation on a rocker at 37°C , the wells were overlaid with 1 ml 0 . 8% methylcellulose in OptiMEM ( Gibco ) supplemented with 2% FBS ( Cellgro ) and antibiotic mix ( Gibco Anti-Anti ) . Plates were incubated 5 days at 37°C , 5% CO2 . On day 5 , overlay was removed , cells washed with PBS , fixed in 80% methanol and either stored at −80°C or developed . To develop plates , fixed monolayers were blocked for 10 minutes with 5% instant milk PBS , followed by incubation with anti-flavivirus MAb 4G2 diluted 1∶1000 in blocking buffer for 1 hr at 37°C . Wells were washed with PBS and incubated with horseradish peroxidase ( HRP ) conjugated goat anti-mouse Ab ( Sigma ) diluted 1∶500 in blocking buffer for 1 hr at 37°C . Plates were washed once in PBS and foci developed by the addition of 100 ul of TrueBlue HRP substrate ( KPL ) . Foci were counted on a light box and viral titers calculated by standard methods . For FRNT , MAbs or human sera were serially diluted five-fold from starting dilutions of 1∶5 or 1∶10 . Each dilution was mixed with approximately 30 focus forming units ( ffu ) of virus to a final volume of 200 ul , incubated for 1 hour at 37°C , 5% CO2 and added in triplicate to 24 wells plates and processed as above . Mean focus diameter was calculated from ≥20 foci/clone measured at 5× magnification . Multiple alignments were performed using ClustalX version 1 . 83 [38] and phylogenetic trees of the envelope protein sequences were conducted using Mr . Bayes version 3 . 12 ( Huelsenbeck JP , 2001 ) . Briefly , 175 amino acid envelope sequences were imported into ClustalX and the alignment was performed using default parameters . Structural models of the informative sites were generated using MacPymol ( Delano Scientific ) and the crystal structure of DENV-3 envelope ( PDB 1UZG ) [16] . Mean focus sizes were compared by one-way analysis of variance ( ANOVA ) followed by Dunnett's test for multiple comparisons . Growth curve and FRNT counts were entered into Graphpad Prism ( Version 5 . 00 for OSX , GraphPad Software , San Diego California USA , www . graphpad . com ) . FRNT50 values were calculated by sigmoid dose-response curve fitting with upper and lower limits of 100 and 0 respectively . All error bars show 95% confidence intervals unless otherwise specified . Mean FRNT50 values were compared by one-way ANOVA followed by Tukey HSD multiple comparison test with significance level alpha ( P ) set at <0 . 05 .
The parent DENV-3 clone is a genotype III variant isolated from a Sri Lankan DF patient in 1989 ( Figure 2 ) ( See materials and methods ) . Full-length flaviviruses genomes have been previously described as unstable and toxic in traditional E . coli clone systems [43]–[46] . To disrupt the putative toxic regions and facilitate creation of chimeric DENV-3 clones , the genome was cloned into segmented , sequential fragments . The fragments and junctions in the final platform were chosen to through multiple trials to maximize insert and plasmid stability in E . coli . Clone junctions were based on type IIS restriction enzyme sites ( BsmBI and BglI ) ( Figure 1A ) that allow directional assembly into full-length cDNAs as described in Materials and Methods . After digestion and purification of individual cDNAs , the full-length cDNA was assembled by in-vitro ligation , transcripts were electroporated into cells and recombinant viruses were recovered from first passage Vero cell culture supernatant . Sequence analyses verified indicator mutations within the cDNA clone fragments and no nucleotide mutations were detected in the entire genome of the recombinant virus after three passages in Vero cells ( data not shown ) . To evaluate the role of DENV3 E protein sequence variation on antibody interactions , representative E genes from genotype I , II , III and IV viruses ( Figure 2 , Text S1 ) were selected from 175 published DENV-3 sequences . Each E gene was selected to represent a genotype whose sequence most closely matched a consensus E sequence generated for each genotype . Genotype I is a 1982 Indonesia isolate , genotype II is a 1995 Thailand isolate , genotype III a 2002 Cuba isolate , and genotype IV a 1977 Puerto Rico isolate . A total of 32 informative sites were identified across the representative genotypes ( Materials and Methods ) , forming nine clusters on the surface of the E glycoprotein , relatively evenly distributed through domains I , II and III ( Text S1 ) . To generate clones that would allow testing of variable neutralization , these representative sequences were synthesized by Bio Basic and inserted into the parent clone background , replacing the parent E gene ( Figure 1B ) . Three of the four variant clones were successfully recovered with correct replacement of the E gene alone . One variant , however , Indonesia '82 ( genotype I ) , required the replacement of the parent SL '89 genotype III preM/M gene with a genotype I preM/M gene , supporting earlier studies that co-evolutionary changes in preM/M may be essential for efficient E gene function in select instances [47] . Full-length sequencing of all passage three recombinant virus clones used throughout these experiences found only one nucleotide mutation in one of the five clones , a silent C to T pyrimidine transition mutation at genomic position 7043 in the genotype I virus . Because some DENV clinical isolates do not reliably form plaques on Vero cell monolayers , viral growth on Vero cell monolayers was instead characterized through focus formation ( see Materials and Methods ) . All five clones formed foci on Vero cell monolayers . The parent clone , SL '89 ( III ) and Cuba '02 ( III ) produced moderate sized and relatively uniform foci after 5d growth on a Vero cell monolayer ( Table 1 ) . Clones with Indonesia '82 ( I ) E genes produced marginally smaller foci , while Thailand '95 ( II ) and PR '77 ( IV ) foci were markedly smaller than those formed by the parent clone ( Table 1 ) . The striking difference in plaque phenotype underscores the importance of structural proteins in basic viral biology , and may be due to either E gene differences in the virus envelope or prM-E mismatch in the virus clones , though identifying the particular genetic differences causing the phenotype is beyond this paper's scope . The growth kinetics of the panel of recombinant viruses were characterized in mammalian Vero cells and C6/36 mosquito cells , both of which are a commonly used for DENV propagation and quantification . Both cell lines were infected with the parent and clones at a multiplicity of infection ( MOI ) of 0 . 01 FFU/cell and grown for 216 hours . In Vero cells , the growth curves for the parent virus and the five clones were similar , with all preparations producing focus-forming virus after 24 hours and peak viral titers achieved between 120 hours and 168 hours ( Figure 3A ) . Peak log viral titers ranged from 6 . 68 log FFU/ml for the parent clone to 5 . 10 log FFU/ml for the genotype II clone . Early growth was slower in the genotype I Indonesia recombinant virus , but ultimately reached peak titers equivalent to the other recombinants . Growth kinetics in C6/36 cells were similar to those in Vero cultures except that virus was not detected until 48 hrs post infection ( Figure 3B ) . Peak titers were generally similar , though the parent clone had a single peak log titer of 7 . 70 log FFU/ml that was significantly higher than the other virus samples . The remaining peak titers ranged from 6 . 30 log FFU/ml to 6 . 66 log FFU/ml and did not differ significantly . The genotype I Indonesia clone did show slower kinetics than the other clones , particularly early in infection ( Figure 3B ) . Overall , inter-genotypic E variability had minimal impact on the viruses' growth in tissue culture . To assess the role of DENV E glycoprotein variation on viral neutralization by human polyclonal sera , the isogenic clones were tested against a panel of late convalescent ( >2 years ) human anti-DENV primary and secondary sera collected from individuals in North Carolina who had been infected during foreign travel [40] ( Table 2 and Text S1 ) . The majority of the neutralization tests were repeated in independent experiments , with highly reproducible FRNT50 values ( Text S1 ) . The original infecting virus is not known for any of these sera . Eight primary anti-DENV-3 serum samples were tested against the parent and isogenic recombinant viruses with variable E genes from the different DENV-3 genotypes ( Text S1 ) . The clones did not show differential neutralization patterns against three of the sera; 003 , 005 and 103 ( Figure 4A , B , and E ) . Serum sample 003 was taken from a traveler who acquired a primary DENV-3 infection in Thailand . FRNT50 titers for 003 ranged from 1∶59 for Cuba'02 ( III ) to 1∶203 for Indonesia '82 ( I ) ( Text S1 ) . Serum sample 005 was taken from a traveler who acquired a primary DENV-3 infection in Puerto Rico . Calculated FRNT50 were similar to those observed for 003 , with titers ranging from a low titer of 1∶31 against PR '77 ( IV ) to a high of 1∶118 against the Sri SL '89 ( III ) clone and Indonesia '82 ( I ) ( Text S1 ) . Serum sample 103 was from a traveler infected with DENV-3 in Nicaragua in 1995 . FRNT50s ranged from a low 1∶42 ( PR '77 ( IV ) ) to a high of 1∶117 for Thailand '95 ( II ) ( Text S1 ) . While these FRNT50 values are consistent with most accepted cutoffs for true homotypic neutralization , they are consistently low , with six of the fifteen clone titers in this group less than 1∶60 ( Text S1 ) . More importantly , we found significantly variability neutralization profiles against the five recombinant viruses neutralized with the five remaining homotypic sera tested ( Figures 5C , 5D , 5F , 5G and 5H ) . Though serum sample 011 , from an El Salvador infection , neutralized all five clones , we found a 9-fold difference ( P<0 . 05 ) between the calculated lowest and highest neutralizing titers , with a low neutralizing group consisting of Indonesia '82 ( I ) , 1∶133 , and PR '77 ( IV ) 1∶157 and a second , high neutralizing group included the remaining clones Cuba '02 - 1∶701 , Thailand '95 - 1∶1091 and SL '89 -1∶1172 ( Text S1 ) . Serum sample 033 ( Figure 4D ) , from an infection in India , was similarly potent , with four of the five clone titers greater than 1∶780 , but with the PR '77 ( IV ) recombinant virus again showing a significantly lower neutralization titer at 1∶304 ( Text S1 ) . Against samples 105 ( Figure 4F ) and 118 ( Figure 4G ) , from infections in Thailand and Nicaragua respectively , the neutralization titers differed five ( 105 ) and six ( 118 ) –fold between recombinant viruses expressing Thailand '95 or PR '77 E glycoprotein ( P<0 . 05 ) ( Text S1 ) . The most extreme neutralization differences between the clones were seen using serum 109 ( Figure 4F ) , from a Sri Lanka donor . This serum , with titers of 1∶177 and 1∶280 , efficiently neutralized the Indonesia '82 and Thailand '95 clones respectively , while the genotype III clones were neutralized at much lower dilutions of 1∶40 ( Sri Lanka ) and 1∶15 ( Cuba ) ( Text S1 ) . Thus , we observed significant variation in neutralization across DENV-3 genotypes for five of the eight primary homotypic sera tested . Human anti DENV secondary sera are known to be broadly neutralizing across serotypes , and we would expect it to show relatively high and broad FRNT50 values and resist intra-genotypic variability . To test this assumption , each of the five clones were tested against 009 , serum from a patient who had a secondary DENV infection , in India or Sri Lanka in 2000 . All of the clones were efficiently neutralized at relatively high titers , though the highest ( Indonesia '82 ) and lowest ( Cuba '02 ) did differ significantly ( Text S1 , Figure 4H ) , though this difference was less than three-fold . Heterotypic primary anti-DENV serum may have low-level serotype-cross neutralizing activity , and in one study was shown to be protective for heterotypic infection in some cases [48] . To assess the role of E glycoprotein variation in heterotypic cross-neutralization , the clone panel was tested against representative primary anti-DENV-1 , -2 , and -4 sera ( Table 2 , Text S1 ) . Sample 001 was collected after a primary DENV-2 infection acquired in Sri Lanka in 1996 . 001 had low level but detectable FRNT50s that ranged from 1∶11 to 1∶78 ( Figure 5A , Text S1 ) . Serum 006 FRNT50 titers ranged from 1∶7 to 1∶54 ( Figure 5B , Text S1 ) , and sample 102 , collected after a DENV-4 infection in Honduras , had a similarly scaled FRNT50 range of 1∶12 to 1∶58 ( Figure 5C , Text S1 ) . However , while repeat FRNT against the clone panel with homotypic sera yielded highly reproducible neutralization titers , repeat FRNTs were not reproducible for heterotypic sera ( Text S1 ) , significantly limiting any conclusions that might be drawn from variable heterotypic neutralization .
CJ Lai et al . described the first full-length infectious DENV clone for DENV-4 isolate 814669 ( isolated from a patient in the Dominican Republic in 1981 [49] ) in 1991 [46] . At that time , the authors noted the full-length DENV cDNA was unstable in E . coli . This was overcome by using a two-fragment system that divided the toxic genomic regions . Subsequent DENV-2 New Guinea C [45] and DENV-4 West Pacific '74 clones [44] employed similar fragment based strategies to overcome genomic stability problems , though a single plasmid DENV-2 clone has also seen considerable use ( [50]–[52] for examples ) . Blaney et al . , using the DENV-3 clinical isolate Sleman '78 , described the first , and , until now , only , DENV-3 infectious clone in 2004 [53] . Though based on a full-length cDNA plasmid , successful propagation of the plasmid DNA required inserting a 30 nt linker region containing termination sequences in each of the forward and reverse open reading frames near the E/NS1 junction . To date , the parent Sleman '78 clone has principally been used as a backbone for vaccine candidates [43] , [54] , [55] . Clearly , instability and toxicity have been the principle challenges of developing tractable DENV infectious clones . The smaller DENV cDNA sub-clone platform we employ offers several advantages . The individual fragments are highly stable in E coli and they can be manipulated individually without affecting distant sites on the genome and allow for fragment re-assortment between DENV strains . The type IIS restriction enzymes BglI and BsmbI generates unique 5′ and 3′ overhangs and prevents spurious self-assembly of the sub-clones , a technical problem with all palindromic cutting restriction enzymes [56] . Finally , multiple mutations can be incorporated simultaneously into separate fragments , circumventing iterative mutation and sequencing of the entire molecular clone and allowing for reassortment of fragments . With the exception of the genotype I E gene , the parent molecular clone backbone was receptive to heterotypic E sequences . We suspected that genotype specific prM/M-E interactions between the genotype III parent prM/M and genotype I E accounted for failure to recover viable genotype I E chimera . Genotype I prM/M differs from genotype II , III , and IV in 2 positions . The first is a histidine to lysine mutation at prM/M position 55 . This region is predicted to form a strand between two parallel beta sheets that interacts with the E fusion loop [47] and the polymorphism at position 55 likely explains why the original genotype I clone was not viable . The second difference was a leucine to phenylalanine mutation at position 128 . This polymorphism conserves the hydrophobic character of the residue , hence we think it unlikely that this mutation affected the original genotype I chimera's viability . Replacing the parent prM gene with a genotype I prM established a viable clone and argues that future constructs should include prM and E from the same genotype . However , overall , the clone platform was remarkably stable: full length sequencing of passage three of all five of the clones found only one ( silent ) nucleotide mutation in one - Indonesia '82 - of the five clones . The recombinant viruses grew to equivalent peak titers compared to the parent clone , though Indonesia '82 showed delayed growth kinetics in both cell lines . Different plaque phenotypes emerged with E glycoprotein changes . While chimeric construction may affect interactions between E and the non-structural proteins or directly change RNA-RNA interactions , these effects are likely subtle , given the relatively similar clone growth kinetics in tissue culture , and are unlikely to directly affect chimeric clone neutralization by polyclonal sera . Forty years ago Halstead and others first reported variable neutralization between clinical DENV-3 isolates [57] when they observed that mouse immune sera raised against DENV-3 strain H-87 poorly neutralized low passage wild-type DENV-3 isolates from Thailand . The authors hypothesized that the observed differences in neutralization were due to within serotype antigenic differences . Shortly thereafter , Russell et al . reported similar findings for human immune sera [58] . They found that both human convalescent sera and mouse hyper-immune sera against Tahitian and Caribbean DENV-3 poorly neutralized H-87 and a Thailand 1965 clinical isolate with differences in 50% hemagglutination inhibition ( HI ) titers varying by more that 10-fold . The authors argued that the different titers were evidence of genetic subtypes within DENV-3 , at the time a novel idea , although the genetic basis for this variable phenotype was unclear . The authors also argued that Caribbean strains would be poor vaccine candidates because of their antigenic properties did not elicit broadly neutralizing homotypic antibodies . Despite this early observation of variable neutralization within serotypes , the phenomenon remained largely unexplored , in part because few tools existed to isolate antigenic variation in an otherwise stable genetic background . More recently , Zulueta et al . [22] , found that human sera from acute genotype III DENV-3 infections were essentially non-reactive with recombinant genotype IV EDIII but appropriately reactive with genotype III EDIII . However , this study's findings were significantly limited by the use of pooled acute human sera and binding assays , rather than neutralization assays and individual human polyclonal serum samples . In a related set of experiments , Cuban researchers tested convalescent sera collected from twenty DF and DHF cases from the 2001/2002 Cuban DENV-3 epidemic against a panel of six DENV-3 isolates collected between 2000 and 2002 [19] . The sera PRNT50 titers against clinical isolates from before and after that epidemic differed by nearly 10-fold , with the patients' sera more effectively neutralized virus from after the epidemic than before . However , their observed differences are based on neutralization against wild type viruses representing only genotypes III and IV and only three of the seven viruses used were sequenced . Finally , Thomas et al . [59] , using previously characterized human DENV sera , found that PRNT50 titers were significantly affected by both virus strain and tissue in which the virus was propagated . While these experiments strongly hint at E gene dependent differences in polyclonal antibody neutralization , they do not directly test variability in the neutralization of isogenic DENV-3 viruses encoding clearly defined E gene differences by late convalescent sera . Our results significantly advance both the pioneering early studies of Halstead and Russell as well as the more recent work cited above , all of which collectively argue that antigenic variability in DENV-3 genotypes significantly influences intra-serotypic neutralization responses in in vitro assays . With our panel of sera and E variant clones , we found both dramatically large , up to 19-fold , differences in FRNT50 values and FRNT50 titers as low as 1∶15 for homotypic sera ( Text S1 ) . Our data indicate that variation in E strongly drives these phenotypes , as all other viral proteins were isogenic . Prospective studies of DENV transmission have found that low titer pre-existing neutralizing Ab ( by PRNT ) in endemic areas does not uniformly protect from homotypic infection [24] , and a prospective study of maternal antibody in newborns found that 50% neutralization titers of <1∶50 are often not protective against homologous virus strains , even in endemic settings [60] . Finally , a recent human challenge study in DENV-3 vaccinated subjects found that a PRNT titer 1∶57 in one vaccinated volunteer was only partially protective , and another volunteer developed both fever and viremia with a pre-existing anti-DENV-3 titer of 1∶16 [61] . Current vaccine trials define 50% or 60% neutralization titers of >1∶10 [62] , [63] or 1∶20 as evidence of immunity , potentially lower than the hypothesized protective thresholds suggested by the studies cited above . Some recent vaccine studies by Durbin et al . Guy et al . have begun to test vaccinee sera against representative genotypes [64] , [65] . However , Durbin et al . used early convalescent sera - 42 days post vaccination , which is likely to be more broadly neutralizing is too early post-vaccination to capture the durable , long-term antibody response . Guy et al . similarly evaluated vaccinated vaccine sera against DENV genotypic variants , but used primate rather than human sera and the authors did not specify when the samples were collected post vaccination . The magnitude of the neutralization differences we report may be enough to lead to partial protection or loss of protection in vaccines , depending on the infecting genotype . It is also possible that , in the context of live virus vaccination , broad within serotype protection is conferred even with low titer antibodies , and that genotypic differences will not matter in the context of protection . That said , Genotype IV stands out in our experiments as relatively non-reactive with homotypic human immune sera ( Figure 4B , 4E , 4F ) and raises the question of whether vaccination could potentially create an immunologic “niche” in human hosts that could be exploited by sylvatic or geographically and genetically distant genotypes within a serotype . Our findings serve as a point of departure for studying the important epitopes in the human antibody response to DENV infection , most of which have not yet been defined . Clones that selectively alter the antigenic clusters distributed across E ( Text S1 ) will facilitate initial mapping of the epitopes responsible for differential neutralization . Ideally , identifying the key neutralizing epitopes in the human polyclonal immune response will , in turn , inform rational vaccine and possibly therapeutic monoclonal antibody design - optimizing epitopes to elicit potent neutralizing antibodies . Although only speculative , the DEN3 molecular clone may also prove invaluable for identifying epitopes and antibodies responsible for enhancing dengue infection . | Infectious virus clones are valuable tools for studying how changes in viral genetic codes affect viral biology . Dengue virus is the most important mosquito-borne virus worldwide , yet dengue virus infectious clones have historically been challenging to make and manipulate , making it very difficult to study the variety of genetic changes observed in dengue viruses . Here we describe the construction of a panel of five dengue virus serotype 3 ( DENV-3 ) clones using a novel strategy not previously employed in dengue research . This strategy uses genetic fragments and synthesized genes to introduce genetic changes while minimally affecting the virus . Each of the five recombinant clones was designed to express genetically distinct DENV-3 envelope proteins derived from strains circulating in different regions of the world . We used the recombinant viruses , coupled with DENV-3 sera from geographically defined human cases , to study the impact of E variation on neutralization outcomes . Our data demonstrate that the recombinant viruses varied significantly in their neutralization outcomes , depending on sera . While it has long been presumed that infection , and vaccination , with one serotype confers lifelong protection against all variants of that serotype , our results indicate that this assumption requires a more rigorous assessment by the DENV community . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"clinical",
"immunology",
"immunity",
"virology",
"cloning",
"genetics",
"neglected",
"tropical",
"diseases",
"immunology",
"biology",
"microbiology",
"viral",
"diseases",
"genetics",
"and",
"genomics"
] | 2012 | Development and Characterization of a Reverse Genetic System for Studying Dengue Virus Serotype 3 Strain Variation and Neutralization |
Vertebrate females transfer antibodies via the placenta , colostrum and milk or via the egg yolk to protect their immunologically immature offspring against pathogens . This evolutionarily important transfer of immunity is poorly documented in invertebrates and basic questions remain regarding the nature and extent of parental protection of offspring . In this study , we show that a lipopolysaccharide binding protein/bactericidal permeability increasing protein family member from the invertebrate Biomphalaria glabrata ( BgLBP/BPI1 ) is massively loaded into the eggs of this freshwater snail . Native and recombinant proteins displayed conserved LPS-binding , antibacterial and membrane permeabilizing activities . A broad screening of various pathogens revealed a previously unknown biocidal activity of the protein against pathogenic water molds ( oomycetes ) , which is conserved in human BPI . RNAi-dependent silencing of LBP/BPI in the parent snails resulted in a significant reduction of reproductive success and extensive death of eggs through oomycete infections . This work provides the first functional evidence that a LBP/BPI is involved in the parental immune protection of invertebrate offspring and reveals a novel and conserved biocidal activity for LBP/BPI family members .
The existence of complex immune systems implies that interactions with pathogens represent major selective forces shaping the evolution of animal and plant species [1] . Vertebrate immune systems not only protect the adult organism against infections but also increase reproductive success through parental transfer of innate and adaptive immune factors via the placenta , colostrum and milk or via the egg yolk [2]–[4] . This maternal transfer of immunity is critical for species survival as embryos and neonates are immunologically immature and unable to fight off infections at early life stages . Parental transfer of protection has also been found in invertebrates hosting mutualists and many vertically transmitted arthropod symbionts are able to protect offspring against specific infections [5] , [6] . Despite the impressive advances recently made in characterizing invertebrate immune systems [7] , [8] , data on the nature of the symbiont-mediated or parentally transmitted protection across generations are scarce [9]–[11] . How the estimated 1 . 3 million of invertebrate species [12] protect their offspring against pathogens remains therefore an intriguing question . The freshwater snail Biomphalaria glabrata is particularly well studied as it is the intermediate host of the human blood fluke Schistosoma mansoni , responsible for schistosomiasis affecting millions of people in developing countries [13] . Biomphalaria snails live in various resting water biotopes such as , ponds , marshes , irrigation channels or open sewer drains that are particularly rich in pathogenic organisms . Egg masses are laid on solid substrates under water where they remain for approximately a week before hatching [14] . In a proteomic study on the content of B . glabrata egg masses , 16 defense-related polypeptides were partially identified , among which a lipopolysaccharide binding protein/bactericidal permeability increasing protein ( LBP/BPI ) representing a major protein band [15] . LBP/BPIs are structurally related proteins belonging to the lipid transfer/binding protein ( LT/LBP ) family [16] , which represent important components of the innate immune system against Gram-negative bacterial infections [17] . In mammals , LBPs and BPIs have been extensively studied due to their role in regulating transducing cellular signals from Lipopolysaccharide ( LPS ) [18] , [19] . LBP functions as a carrier of LPS monomers onto CD14 and together with the TLR4-MD2 receptor complex , mediates the activation of monocytes and macrophages , which produce inflammatory mediators [20] . BPI is an antibacterial protein specifically active against Gram-negative bacteria that acts by damaging bacterial membranes [21] . BPI also enhances adaptive immune responses by promoting LPS uptake and presentation to dendritic cells [22] . Although these two proteins present similarities in sequence and activities , they exert different effects on interactions of the host with Gram-negative bacteria [23] . BPI neutralizes the inflammatory properties of LPS decreasing its uptake by LBP whereas LBP is an acute phase protein with LPS-dependent cell stimulatory activity [24] , [25] . These antagonist functions efficiently regulate host response to bacterial invasion and allow the host immune system to return to its normal resting state . The distinction between LBPs and BPIs has not been established in invertebrates . LBP/BPI family members have been reported only in a few invertebrate phyla such as annelids [26] and molluscs [27] , [28] . To date , a single functional study of LBP/BPI has been performed , showing that the oyster Crassostrea gigas expresses a BPI-like protein endowed with the conserved LPS-binding and bacterial permeability increasing activity [27] . As B . glabrata snails apparently heavily invest in the production of LBP/BPI in their eggs [15] , we investigated whether this protein showed the expected anti-bacterial activity and whether it could provide protection against other pathogens .
We first characterized the complete coding sequence of BgLBP/BPI1 ( genbank accession number KC206037 ) from a partial transcript that we had previously identified in an albumen gland cDNA library [29] , as this organ is known in gastropod snails to produce many components of the egg masses , among which the egg perivitelline fluid [30] . Interestingly , BLAST searches against non-redundant protein databases using the BLASTp program revealed that BgLBP/BPI1 corresponded to the “Developmentally regulated albumen gland protein” ( partial sequence; genbank accession number AAB00448 . 1 ) previously identified as being over-expressed in Schistosoma mansoni resistant snails [31] . The sequence of BgLBP/BPI1 displayed the typical features of LBP/BPI family members , including a N-terminal LBP/BPI domain ( pfam PF01273 , Interpro IPR017942 ) containing conserved lysines involved in the interaction with LPS , a central proline-rich domain , and a C-terminal LBP/BPI domain ( Figure S1 ) [17] , [32] . Expression studies showed that the major site of expression for BgLBP/BPI1 was the albumen gland ( Figure 1 , A and B ) . We confirmed , using a specific antibody and western blot analysis followed by mass spectrometry , that the BgLBP/BPI1 gene product is the major protein found in egg masses of B . glabrata ( Figure 1 , C and D ) . We purified the native BgLBP/BPI1 protein from fresh egg masses and estimated its physiologic concentration around 100 µg/ml of egg mass extract , representing 60% of the total protein dry weight . In order to control for trace contamination of the purified native protein by biologically active polypeptides , we produced a recombinant protein in Drosophila S2 cell culture and compared both the native and the recombinant proteins in our assays . Plasmon resonance analysis confirmed that both BgLBP/BPI1 bind LPS and the Lipid A region , which are common to all LPS's , with a range of affinity similar to that of human BPI protein , as shown by their dissociation constants ( Figure 2 ) [33] . In addition , both proteins showed the typical membrane permeabilizing activity leading to bacterial death ( Figure 3 ) that LBP/BPI proteins present toward short-LPS strains of E . coli [34] , demonstrating that the biocidal activity of LBP/BPI proteins against Gram-negative bacteria is conserved in BgLBP/BPI1 [21] , [35] , [36] . Exposure of helminths ( S . mansoni ) , Gram-positive bacteria ( Micrococcus luteus , Bacillus cereus ) , Gram-negative bacteria ( Citrobacter freundii , and Pseudomonas aeruginosa ) and fungi ( Candida albicans and Saccharomyces cerevisiae ) to increasing concentrations of the recombinant BgLBP/BPI1 ( up to the physiological concentration of 100 µg/ml ) had no significant effect on the viability of the microorganisms ( P>0 . 1 , Figure S2 and S3 ) . With the exception of helminths such as S . mansoni , information on natural pathogens of B . glabrata is scarce . Therefore we also wanted to investigate whether oomycetes are sensitive to BgLBP/BPI1 . Oomycetes or water molds are a large group of eukaryotic microbes that can infect plants and animals and can cause devastating diseases in agriculture , aquaculture , and natural ( aquatic ) ecosystems [37] , [38] . The motile zoospores ( infective stage ) and cysts ( germinal stage ) of both Saprolegnia parasitica and Saprolegnia diclina , two well-known pathogens of fresh water fish and their eggs [38] , were exposed to increasing concentrations of BgLBP/BPI1 proteins for 30 min . No effect was observed on Saprolegnia cysts ( not shown ) , but a strong biocidal activity was observed on the infective stage of these pathogens at all protein concentrations ( Figure 4 ) . The viability of S . parasitica or S . diclina zoospores was significantly reduced to 50 . 3% and 68 . 4% or 55 . 8% and 39% with 100 µg/ml of native BgLBP/BPI1 and recombinant BgLBP/BPI1 , respectively ( Figure 4 ) . Interestingly , human BPI also strongly decreased the viability of both oomycete species ( Figure 4 ) , revealing a yet unsuspected activity of this well-studied human immune protein [39] . To better assess the LBP/BPI-dependent anti-oomycete activity , we also tested a plant pathogenic species , Phytophthora parasitica that has the ability to form biofilms [40] like Gram-negative bacteria . Biomphalaria proteins were tested on the three stages of biofilm formation; zoospores , cysts and microcolonies ( Figure S4A ) . Similarly to the Saprolegnia species , only zoospores were affected by the BgLBP/BPI proteins . Both the snail and the human LBP/BPI proteins showed a strong biocidal activity of zoospores from P . parasitica in a dose dependent manner ( Figure 5 and S4B ) . The effect of LBP/BPIs was observed as early as 10 min and resulted in 100% zoospore mortality after one or two hours of exposure to physiological concentrations of 100 µg/ml LBP/BPI proteins ( Figure 5 ) , confirming the strong biocidal activity of LBP/BPIs against oomycete zoospores . In order to assess the role of BgLBP/BPI1 in vivo , we undertook to decrease its expression by using RNA-interference mediated knock-down . Following double strand dsRNA injections , BgLBP/BPI1 protein abundance was analyzed by western blotting and showed a significant decrease in the albumen gland and in the egg masses after 12 and 18 days , respectively ( Figure 6 ) . The number of eggs per clutch collected over 28 days from parents treated with dsRNA of BgLBP/BPI1 was significantly lower than in the control experiment , whereby dsRNA of the luciferase gene was injected . Furthermore , the egg masses of the BgLBP/BPI1 dsRNA-treated snails showed a significant decrease in fecundity ( Table 1 ) . Thereby confirming that the albumen gland , the site of expression of BgLBP/BPI1 , is directly involved in egg mass production [41] . After exposure to zoospores of S . diclina , eggs from parents silenced for BgLBP/BPI1 expression suffered an important decrease in their hatching rate , when compared to snails injected with control dsRNA ( Table 1 ) . The eggs from control-treated parents appeared healthy with a normal development of the embryos ( Figure 7A ) , whereas the egg masses from parents treated with BgLBP/BPI1 dsRNA were covered by oomycete hyphae and the resulting infection impaired dramatically the survival of the snail embryos ( Figure 7B ) .
Many invertebrate species lay fertilized eggs in nutritive egg masses that are highly suitable to the development of microorganisms [42] , [43] . Although parental protection of eggs seems crucial to the survival of species , studies on the immune protection of invertebrate eggs are scarce . For example , an antibacterial activity was shown in eggs from 32 mollusk , 2 polychaete and 1 coral species , out of , respectively 34 , 4 and 11 species tested [9] , [42] . An antibacterial protein , the aplysianin-A was identified from eggs of the gastropod Aplysia kurodai [44] , and an N-acetyl-galactosamine-binding lectin that agglutinates bacteria was identified in eggs from the pulmonate snail Helix pomatia [45] . Together with peptides of aplysianin , peptides of LBP/BPI proteins were identified in a proteomic study on Biomphalaria glabrata egg masses [15] . Here we characterized a B . glabrata LBP/BPI family member that is produced in the albumen gland and abundantly loaded into the egg masses . Consistent with the study on the BPI-like protein from Crassostrea gigas [27] , we showed that the LPS and lipid A-binding activities against Gram-negative bacteria are conserved in BgLBP/BPI1 . BgLBP/BPI1s did not exert any effect against the panel of microorganism tested . However , bacterial permeability activity was observed against E . coli SBS363 , a mutant strain containing short-chain LPS . This is in agreement with previous studies reporting the resistance of Gram-negative bacteria harboring long lipopolysaccharide chain to the activity of hBPI [34] , [46] . In addition to this expected anti-bacterial activity , we discovered a yet unsuspected anti-oomycete activity and demonstrated that BgLBP/BPI1 is a major fitness-related protein affecting both egg production under control conditions and offspring survival in the presence of pathogens . It is possible that the positive effect of BgLBP/BPI1 on the number of eggs produced is related to the glycoprotein nature of the molecule rather than to its antimicrobial activities . Interestingly , the glycoprotein HdAGP , identified from the snail Helisoma duryi albumen gland , was reported as the major nutritive glycoprotein secreted in the perivitelline fluid , and is also sharing sequence similarities with LBP/BPIs [43] . The content in BgLBP/BPI1 may therefore affect egg production as a major nutritive egg mass compound , independently of its antimicrobial action . However , once the egg masses are laid , we demonstrated that the biocidal activity of BgLBP/BPI1 affects offspring survival in the presence of oomycete pathogens . LBP and BPIs are pleiotropic molecules , well characterized for their interactions with LPS from Gram-negative bacteria , but also reported to interact with other organisms such as Gram-positive bacteria and fungi [18] , [47] , [48] . A wide range of lipidic ligands have been reported for human LBP and BPI [18] , [47] , [48] . Our results further evidence the diversity of binding capabilities of LBP/BPIs as both BgLBP/BPI1 and hBPI can interact with an oomycete lipidic ligand that remains to be identified . Oomycetes do share physical characteristics with true fungi , including polarized hyphal extensions but they have a distinct evolutionary history and belong to the kingdom Stramenopila , which also includes brown algae and diatoms [49] . In contrast to fungi , they produce bi-flagellated swimming spores ( zoospores ) and the cell-wall of their cysts is composed of cellulose , β-glucans and hardly any chitin [49] . Oomycetes include some of the most devastating animal and plant pathogens . A few species cause Saprolegniosis in the aquaculture industry [38] . Saprolegnia species are endemic to freshwater habitat worldwide and are partly responsible for declining natural populations of salmonids and amphibians [50] , [51] . Furthermore , the potato and tomato late-blight pathogen , Phytophthora infestans triggered the Irish Famine in the mid-1840s [37] , [50] , [52] . The potent oomycete killing activity of both BgLBP/BPI1 and hBPI was observed with three species belonging to two major oomycete orders , the Perenosporales and Saprolegniales [52] . Our observations demonstrate a conserved and broad-spectrum oomycete killing activity of BgLBP/BPI1 , which may be of interest for both the agricultural and aquacultural sectors [53] . Interestingly , the specificity of this biocidal activity for the zoospore stage suggests that the ligand may be expressed specifically at this developmental stage . To date , despite the economic impact of oomycetes , there is no biochemical information on the membrane compounds of zoospores as studies have focused on identifying surface components of the cell wall of cysts [54] , [55] . Collectively , our results significantly expand our knowledge of the multiple functions of LBP/BPI and highlight their importance in invertebrate biology . We demonstrated that LBP/BPI proteins display a conserved , potent and so far unexpected biocidal activity against zoospores from different oomycete orders . The precise binding and killing activity of the zoospores is unknown , but it is clear that BgLBP/BPI1 represents a major fitness-related protein transferred from parents to their clutches protecting snail eggs from widespread and lethal oomycete infections .
A partial cDNA sequence ( EST GenBank accession number EB709540 ) was used to design specific primers and perform 5′- and 3′-RACE amplification ( 5′3′ RACE kit , 2nd generation - Roche ) according to the manufacturer's instructions . PCR products were cloned into pCR4-TOPO vector ( Invitrogen ) for sequencing . Sequence similarity searches were carried out using NCBI's BLAST-X program [56] against non-redundant databases with default parameters . Global sequence alignments were performed with Clustal W software [57] . The protein domains and signal peptide were predicted with the SMART [58] and SignalP [59] softwares , respectively . Adult Biomphalaria glabrata snails ( albino strain ) were raised in pond water and fed leaf lettuce ad libitum according to previously described procedures [60] . Bacterial and yeast strains used in this study were Micrococcus luteus ( CIP A270 ) , Pseudomonas aeruginosa ( PA14 ) [61] , Bacillus cereus ( ATCC 11778 ) , Citrobacter freundii ( ATCC 8090 ) , Candida albicans ( a pathogenic strain isolated in patient no . 3 by Pr M . Koenig , CHU Strasbourg-Hautepierre ) and Saccharomyces cerevisiae ( Bioreference Laboratory – Institut Pasteur ( Lille , France ) as well as the E . coli SBS363 , a Trp+ galU129 ( truncated LPS ) derivative of E . coli K12 strainD22 ( gift from D . Destoumieux-Garzón , Université Montpellier 2 ) . Bacterial strains were maintained in LB medium at 37°C and yeast in YPD medium at 28°C under standard conditions . S . mansoni miracidia ( swimming infective stage ) were hatched from eggs axenically recovered from 50-days infected hamster livers according to previously described procedures [62] . Oomycete species used in the study were Saprolegnia parasitica , S . diclina and Phytophthora parasitica . Saprolegnia zoospores were obtained as described previously [63] . The average number of zoospores released was approximately 104 zoospores per ml . Phytophthora parasitica was grown in 90 mm-diameter Petri dishes on 20% V8 agar media ( 350 ml V8 juice , 5 g CaCO3 , 3 . 5 g agar ) at 25°C for 6–8 days under continuous light . To induce zoospore release , Phytophthora isolates were placed at 4°C for 30 min . Mycelial cultures were then flooded with 10 ml of warm sterile water and left at 28°C for 30 min [40] . The average number of zoospores obtained was approximately 106 zoospores per ml . Snail organs or tissues , namely albumen gland , hepatopancreas , headfoot , digestive tract and gonads were dissected under a binocular microscope , pooled from 10 individuals and frozen in liquid nitrogen . Snail circulating hemocytes were recovered from hemolymph collected prior to tissue dissections according to previously described procedures [62] . Total RNA and protein were simultaneously isolated using TRIZOL LS Reagent ( Invitrogen ) according to the manufacturer's instructions . Total RNA was quantified using a NanoDrop Spectrophotometer ND-1000 ( Thermo Scientific ) . For cDNA synthesis , 50 ng of RNA from dissected tissues and hemocytes were used for reverse transcription using iScript cDNA Synthesis kit ( Bio-Rad ) and the oligo ( dT ) 20 primer . The relative expression of BgLBP/BPI1 was monitored using Quantitative Real-time PCR on a DNA engine opticon 2 system ( Biorad ) . Primers specific for B . glabrata ribosomal protein S19 ( Genbank accession number CK988928 ) [64] , elongation factor EEF1- α ( Genbank accession number ES482381 . 1 ) and BgLBP/BPI1 , were designed with primer 3 software and used for amplification in triplicate assays . The PCR cycling procedure was as follows: initial denaturation at 95°C for 10 min , followed by 40 cycles of amplification 95°C for 30 s , 60°C for 30 s and 68°C for 30 s for signal collection in each cycle . To assess the specificity of the PCR amplification , a melting curve analysis of the amplicon was performed at the end of each reaction and a single peak was always observed . To examine the distribution of BgLBP/BPI1 protein in snail dissected tissues or egg masses , an anti-BgLBP/BPI1 antiserum was produced in a rabbit using the LAKAHIEKNRLIPDLLSYD and AQDKPGAVLRLNQEALDYGSR peptides and the polyclonal sera were purified using a peptide linked resin column ( Proteogenix ) . Total protein contents of tissues were first determined by the BCA method ( BC assay kit , Uptima ) using albumin as a standard . 15 ug of tissue or egg mass proteins were loaded onto 10% SDS-PAGE gels and either silver stained using standard procedures , or transferred to a PVDF membrane ( 0 . 2 mm pore size ) using a semi-dry blotting system . Western blots were performed using the custom anti-BgLBP/BPI1 antisera ( Proteogenix ) . cDNA corresponding to the open-reading frame of BgLBP/BPI1 was ligated into the pMT/V5/His-A expression vector ( Invitrogen ) . The Drosophila expression system with Schneider 2 ( S2 ) cells ( Invitrogen ) was used to express recombinant C-terminally His-tagged full-length BgLBP/BPI1 ( rBgLBP/BPI1 ) as described previously [65] , [66] . Briefly , S2 cells were transiently transfected by calcium phosphate method with 1 µg of pMT/BgLBP/BPI1/V5/His-A vector and its expression was monitored by SDS-PAGE and western blotting after 3 days of induction with CuSO4 ( 500 mM ) . After confirmation of transient rBgLBP/BPI1 expression , stable cell lines were generated performing co-transfections along with 0 . 1 µg of PJL3 selection vector and 1 µg/ml puromycin . Establishment of stable cell lines and production of rBgLBP/BPI1 were carried out as described previously [66] . Nickel ( II ) -based immobilized metal affinity chromatography ( Qiagen ) in native conditions was performed to purify the recombinant BgLBP/BPI1 protein according to the manufacturer's protocol . The native BgLBP/BPI1 protein ( nBgLBP/BPI1 ) was purified from two-days old B . glabrata egg masses . Egg masses were homogenized in 20 mM acetate buffer , pH 4 . 5 . The crude homogenate was centrifuged at 13000 rpm for 10 min to remove the gelatinous and solid debris . Supernatant containing nBgLBP/BPI1 was loaded onto SP Trisacryl M cation-exchange resin ( BioSepra ) equilibrated in 20 mM acetate buffer , pH 4 . 5 . After washing 3 times with equilibration buffer , nBgLBP/BPI1 was eluted with 1 M NaCl , 20 mM acetate buffer , pH 4 . 5 and quantified by Bradford method . In order to analyze the sequence of the nBgLBP/BPI1 , the purified protein was excised from a 10% SDS-PAGE gel and subjected to a MALDI TOF/TOF-MS analysis ( Proteomic facility , University of Strasbourg , France ) . Protein identification was performed by subjecting the m/z values to Mascott software at an adjusted peptide mass tolerance of 50 . 000 . 000 ppm and/or 0 . 5 Da and at a fragment mass tolerance of 0 . 4 Da . For the subsequent activity assays , the purity of the purified nBgLBP/BPI1 protein was assessed after SDS-PAGE and silver staining . Assessments of both the egg mass volumes used for purification and the final concentration of the purified protein allowed to determine that the natural concentration of nBgLBP/BPI1 is in the range of 100 µg/ml of fresh 2 days old egg masses . Binding of LPS or lipid A to rBgLBP/BPI1 and nBgLBP/BPI1 was assessed with a Biacore 3000 system ( Biacore , GE Healthcare ) . RBgLBP/BPI1 and nBgLBP/BPI1 were immobilized at 7000 response units ( RU ) onto an activated CM5 sensor chip ( Biacore ) according to the manufacturer's instructions . Human BPI ( hBPI - Athens Research , USA ) and BSA proteins were immobilized using the same conditions as positive and negative control proteins , respectively . An activated and blocked flow-cell without immobilized ligand was used as a reference to evaluate nonspecific binding . HBS-EP running buffer ( 10 mM HEPES , 150 mM NaCl , 3 mM EDTA , and 0 . 005% Tween 20 , pH 7 . 4 ) was used for sample dilution and analysis . Purified diphosphoryl lipid A from E . coli F583 Rd mutant and LPS from E . coli O26:B6 ( Sigma ) were sonicated 15 min at 25°C and injected at various concentrations . LPS was diluted at 50 , 100 , 250 , 500 , 1000 and 2000 nanograms and lipid A at 30 , 60 , 100 , 150 , 200 , 300 , 400 , 500 nanograms and passed over the sensor chip at a flow rate of 50 µl/min . Regeneration was achieved with two washes of 20 mM NaOH for 5 min and 150 mM NaOH for 5 min for LPS and lipid A , respectively . Sensor chip was finally equilibrated with HBS-EP buffer for 2 min . All analyses were done at a constant temperature of 25°C . Data analysis was performed after subtraction of the uncoated flow-cell values by using BIAevaluation software version 4 . 1 ( BIAcore ) . The association and dissociation phases of all sensor-grams were fitted globally . Kinetic parameters were then determined using a 1∶1 Langmuir binding model . The effect of proteins on the permeability of bacterial membranes was determined by flow cytometry using E . coli SBS363 and the LIVE/DEAD BacLight Bacterial Viability Kit ( Molecular probes ) . This kit enables assessment of bacterial viability based on membrane integrity by differentiating between bacteria with intact and damaged cytoplasmic membranes [67] . Bacterial culture in mid-logarithmic phase was adjusted to an optical density of A600 = 0 . 003 with poor-broth nutrient medium and treated with 10 , 30 , or 100 µg/ml of rBgLBP/BPI1 and nBgLBP/BPI1 . BSA and hBPI were used at similar concentrations as a negative and positive control , respectively . Samples were incubated 1 , 2 and 6 h at 28°C under vigorous shaking . Bacterial suspensions were stained with LIVE/DEAD BacLight staining reagent mixture ( SYTO 9 and propidium iodide - PI ) as described by the manufacturer . Staining was allowed for 5 min at room temperature in the dark . Flow cytometric measurements were performed on a FACSCanto II flow cytometer ( BD Biosciences ) with a 488 nm argon excitation laser . A total of 60 , 000 events were acquired and analyzed in each sample , using BD FACSDiVa software version 6 . 1 . 3 ( BD Biosciences ) . Results are displayed as a percentage of permeabilized cells with respect to the negative control . The experiments were carried out three times independently . The antibacterial activity was tested on Micrococcus luteus , Bacillus cereus , Citrobacter freundii , and Pseudomonas aeruginosa using the liquid growth inhibition method as previously described [68] . For antifungal activity assays , a similar liquid growth inhibition assay was performed using YPD medium . Microbial growth was controlled by measurement of the optical density at A600 after 6 , 16 and 24 h incubation in proteins at 10 , 50 , 100 ug/ml . Fungal growth was additionally evaluated at 48 h . The percentage of growth ( % growth ) was deduced from the absorbance ( OD ) at 600 nm as previously described [69] . The effect on S . mansoni viability was tested on groups of 10–15 miracidia placed in 24-well plates . Miracidia were exposed to proteins at 10 , 50 , 100 , 200 µg/ml at 28°C . Microscopic observations were performed at 30 min , 1 , 2 , 4 , 6 , 8 and 24 h of incubation . The anti-oomycete assays were adjusted to the characteristics and life-cycles of the oomycete species . Zoospores of S . parasitica and S . diclina were adjusted to 10000 cells/ml and exposed to increasing concentrations ( 5 , 10 , 30 , 100 µg/ml final concentration ) of proteins . Because of the rapid encystment of zoospores into cysts , assessment of mortality was only performed at 30 min ( prior to encystment of live zoospores ) . Zoospores of Phytophthora parasitica were adjusted to 500000 cells/ml and exposed to identical concentrations of proteins ( 5 , 10 , 30 , 100 ug/ml final concentration ) . Microscopic observations of the number of live and dead zoospores were performed after 10 , 30 , 60 , 120 , 240 min of treatment . Zoospore-treated suspensions were stained immediately after incubation with the Live/Dead Cell Assay kit ( Abcam ) as described in the manufacturer's protocol . All conditions were tested in triplicate and assays were performed 3 times independently . BgLBP/BPI1-specific primers containing a T7 promoter sequence were designed to amplify a 420-bp region of BgLBP/BPI1 used as a template for double stand RNA ( dsRNA ) synthesis according to manufacturer's instructions ( MEGAScript T7 kit , Ambion ) . The firefly ( Photinus pyralis ) luciferase gene dsRNA ( pGL3 vector , Promega ) was produced and used as a non-relevant dsRNA control . Each dsRNA ( 15 µg in 10 µl of sterile Chernin's Balanced Salt Solution - CBSS ) was injected into the cardiac sinus of individual snails using a 10 µl Hamilton syringe with a 26 s needle [62] . A second injection was performed 12 days after the first dsRNA injection in order to optimize the knock-down efficiency . Groups of 10 snails were injected either with BgLBP/BPI1 or Luc dsRNA and were maintained under standard conditions . Egg masses produced during 28 days following the first dsRNA injection were scored , collected and observed under a stereoscopic microscope for assessment of the number of eggs in each egg mass . Egg masses laid from day 12 to 21 after the first dsRNA injection were either maintained under control conditions , or exposed to Saprolegnia diclina zoospores at a final concentration of 104cells/ml . Egg masses were microscopically observed during the following 10 days to assess the egg hatching rate . Results are shown as the percentage of eggs hatched . RNA interference experiments were performed three times independently . All data were expressed as mean of three independent experiments plus or minus SE . Differences in relative BgLBP/BPI1 gene expression were tested for statistical significance by one-way ANOVA and the tukey-Kramer test ( Software Prism v . 5 . 0 , GraphPad ) . Data from membrane-permeabilizing and antimicrobial ( antibacterial , anti-oomycete ) assays were analyzed by the chi-square test of independence [70] between treatments ( rBgLBP/BPI1 , nBgLBP/BPI1 , hBPI and BSA ) and the proportion of dead cells , using the computing environment R [71] . Schistosoma mansoni survival curves were analyzed by the Mantel-Cox log-rank test ( Software Prism v . 5 . 0 , GraphPad ) . Results on snail fecundity and on egg viability were statistically analyzed by the likelihood ratio test between nested models [71] . Briefly , three variables were considered in these models; the mean number of egg per snail , the mean number of egg masses per snail and the mean number of eggs per egg mass . To test the effect of the BgLBP/BPI1 silencing on snail fecundity for each of the three variables , linear mixed models were fitted with the function lmr ( LML 4 package of R software ) . In each case the model contains the BgLBP/BPI1 silencing as fixed effect and the time and the replicates as random effects . To normalize the data the mean number of eggs and egg mass per snail were log transformed . To assess the effect of the oomycete infection on the hatching rate of BgLBP/BPI1 silenced eggs , saturated binomial generalized linear mixed models were fitted . This model contained as fixed effects the BgLBP/BPI1 silencing , the oomycete infection , the time and all interactions among these variables; and as random effects , the replicates . To account for over-dispersion , individual level of variability was added . From this model variable selection of fixed effect was based on the AICc ( dredge function of MuMIn package of R software ) then the selected fixed effect was analyzed by the likelihood ratio test . A P value of <0 . 05 was considered statistically significant . Where indicated in figures: *P<0 . 05 , **P<0 . 01 , ***P<0 . 001 . | Vertebrate immune systems not only protect adult organisms against infections but also increase survival of offspring through parental transfer of innate and adaptive immune factors via the placenta , colostrum and milk or via the egg yolk . This maternal transfer of immunity is critical for species survival as embryos and neonates are immunologically immature and unable to fight off infections at early life stages . Parental immune protection is poorly documented in invertebrates and how the estimated 1 . 3 million of invertebrate species protect their eggs against pathogens remains an intriguing question . Here , we show that a fresh-water snail , Biomphalaria glabrata massively loads its eggs with a lipopolysaccharide binding protein/bactericidal permeability increasing protein ( LBP/BPI ) displaying expected antibacterial activities . Remarkably , this snail LBP/BPI also displayed a strong biocidal activity against water molds ( oomycetes ) . This yet unsuspected activity is conserved in human BPI . Gene expression knock-down resulted in the reduction of snail reproductive success and massive death of eggs after water mold infections . This work reveals a novel and conserved biocidal activity for LBP/BPI family members and demonstrates that the snail LBP/BPI represents a major fitness-related protein transferred from parents to their clutches and protecting them from widespread and lethal oomycete infections . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Parental Transfer of the Antimicrobial Protein LBP/BPI Protects Biomphalaria glabrata Eggs against Oomycete Infections |
While there have been studies exploring regulatory variation in one or more tissues , the complexity of tissue-specificity in multiple primary tissues is not yet well understood . We explore in depth the role of cis-regulatory variation in three human tissues: lymphoblastoid cell lines ( LCL ) , skin , and fat . The samples ( 156 LCL , 160 skin , 166 fat ) were derived simultaneously from a subset of well-phenotyped healthy female twins of the MuTHER resource . We discover an abundance of cis-eQTLs in each tissue similar to previous estimates ( 858 or 4 . 7% of genes ) . In addition , we apply factor analysis ( FA ) to remove effects of latent variables , thus more than doubling the number of our discoveries ( 1 , 822 eQTL genes ) . The unique study design ( Matched Co-Twin Analysis—MCTA ) permits immediate replication of eQTLs using co-twins ( 93%–98% ) and validation of the considerable gain in eQTL discovery after FA correction . We highlight the challenges of comparing eQTLs between tissues . After verifying previous significance threshold-based estimates of tissue-specificity , we show their limitations given their dependency on statistical power . We propose that continuous estimates of the proportion of tissue-shared signals and direct comparison of the magnitude of effect on the fold change in expression are essential properties that jointly provide a biologically realistic view of tissue-specificity . Under this framework we demonstrate that 30% of eQTLs are shared among the three tissues studied , while another 29% appear exclusively tissue-specific . However , even among the shared eQTLs , a substantial proportion ( 10%–20% ) have significant differences in the magnitude of fold change between genotypic classes across tissues . Our results underline the need to account for the complexity of eQTL tissue-specificity in an effort to assess consequences of such variants for complex traits .
Gene expression is an essential cellular function whose regulation determines a significant proportion of the phenotypic variance . Using microarrays and recently second generation sequencing ( RNA-seq ) [1] , [2] , major progress has been made in understanding the genetics of human gene expression and identifying loci that drive differential expression across individuals [3] , [4] , populations [5]–[7] and tissues [7]–[11] . This development is especially valuable for the biological analysis of genome-wide association ( GWAS ) signals [12] , which often map to non-genic regions and are thus hard to interpret in the absence of additional information [13] . Transcript abundance is a very proximal endophenotype affected by genetic variation and has already facilitated the identification of candidate susceptibility genes for metabolic disease traits [14] , asthma [15] or Crohn's disease [16] . This has been mostly possible when the tissue of expression was relevant to the interrogated complex trait , as disease phenotypes manifest themselves only in certain tissues . eQTLs discovered in LCLs have primarily helped explain GWAS associations with immunity-related disorders [17] , [18] while associations with obesity-related traits were mostly observed when gene expression was quantified in adipose tissue [9] . Nevertheless , our guess of tissue relevance is yet far from satisfactory [19] , reinforcing thus the incontestable value of measuring expression in multiple cell-types ( including primary tissues reflecting in vivo patterns ) . Transcriptional regulatory networks are expected to dictate tissue-specificity of regulatory effects [20] , but the extent of this is still under debate . Depending on the cell-types compared and the eQTL discovery methods used , current estimates for tissue-specificity of eQTLs range from ∼30% ( liver , adipose tissue ) [21] to 70–80% ( LCL , fibroblasts , T cells ) [7] . In this study we investigated various aspects of tissue-specificity and we emphasize the importance of accounting not only for statistical significance but also for continuous biological properties of regulatory variants , such as fold change in expression . We explored the complexity of the human cis-regulatory variation landscape in three tissues ( LCL , skin and fat ) derived from a subset of female Caucasian twins aged between 40 and 87 years old ( mean 62 years ) from the UK Adult Twin registry [22] . The present study represents the pilot phase of the MuTHER project ( Multiple Tissue Human Expression Resource—http://www . muther . ac . uk/ ) , a major resource initiated to enhance our knowledge about common trait susceptibility by providing genome-wide expression , methylation and eventually transcriptome sequencing information for 855 extensively phenotyped twins ( clinical , anthropometric , life-style information as well as a wide range of biological measurements are available ) .
Gene expression was quantified in LCL , skin and fat using Illumina's whole genome expression array ( HumanHT-12 version 3 ) containing 48 , 803 probes in three technical replicates [E-MTAB-522] . Log2 - transformed expression signals were normalized separately per tissue by quantile normalization across replicates followed by quantile normalization across individuals . 27 , 499 probes mapping uniquely to 18 , 170 Ensembl genes were retained for further analysis . The same individuals had also been genotyped with Illumina's 1M-Duo and 1 . 2M-Duo chips; 865 , 544 SNPs with MAF>1% passed quality check ( QC ) . The overlapping set of successfully genotyped samples with available expression data amounted to 156 individuals for LCL ( 30 MZ pairs , 37 DZ pairs , 22 singletons ) , 160 for skin ( 31 MZ pairs , 37 DZ pairs , 24 singletons ) and 166 for fat ( 31 MZ pairs , 40 DZ pairs , 24 singletons ) . This final dataset was used for eQTL analysis . We tested for SNP-gene expression associations ( eQTLs ) separately in each tissue . We considered only unrelated individuals at a time by separating twins from the same pair and thus performing two independent eQTL analyses per tissue . This study design , hereafter named Matched Co-Twin Analysis ( MCTA ) , permits immediate replication and validation of eQTL discoveries . We used Spearman Rank Correlation ( SRC ) to detect associations and restricted our search to cis effects located within 1Mb on either side of a gene's transcription start site ( TSS ) . Statistical significance was assessed at different thresholds using permutations ( 10 , 000 per gene ) [5] . We detected an abundance of cis eQTLs ( Table S1A ) per tissue at a comparable rate to other studies of similar sample size [5] , [7] . The reported eQTLs appear robust as they replicate well between individuals of the two co-twin groups per tissue . We measured the eQTL overlap in a continuous fashion by taking the significant SNP-gene associations from one co-twin set and estimating the proportion of true associations ( π1 statistic [23] , see Materials and Methods ) on the distribution of corresponding p-values in the reciprocal co-twin validation set . High levels of eQTL replication were observed across co-twins , with a mean π1 of 0 . 93 in skin and 0 . 98 in LCL and fat ( Table 1 ) . We also measured the estimated proportion of true positives among the subset of genes that did not replicate in the co-twin at the same threshold . This too is high ( π1 = 0 . 84 for skin and 0 . 94 for LCL and fat ) , suggesting that exact overlap of genes at a given permutation threshold ( PT ) is an underestimate of eQTL replication due to winner's curse . In other words , we detected eQTLs in the co-twin that clearly replicated the initial findings , but at p-values that marginally missed the initial discovery threshold . To further confirm the robustness of our discoveries , we overlapped the MuTHER LCL results with available eQTL data from two recent independent studies . 40% of the genes for which we detect LCL eQTLs overlap with eQTLs detected in HapMap 3 samples of European ancestry ( CEU ) ( Stranger et al . submitted ) . Likewise , 36% of the cis associations detected by Gibson et . al . in leukocytes derived from 194 southern Moroccan individuals [24] overlap with genes reported in our study . Given the differences in gender distribution , sample preparation or even cell-type tested ( LCL versus leukocytes ) across these studies , the gene overlap observed is reassuring . The observed variation in gene expression is not entirely due to genetic effects . Experimental noise and environmental conditions also affect transcript levels in a global manner . Therefore , it is desirable to remove the effects of such random variables and thus increase the power to detect eQTLs . For this purpose , we employed factor analysis ( FA ) on each tissue separately and corrected for global latent effects on all individuals in each tissue [25] . We fitted various parameters such as number of learned factors and proportion of variance explained , in order to maximize for replication of eQTLs per tissue between twin sets . After performing standard SRC eQTL analysis on the factor-corrected expression data ( SRC-FA ) , we obtained a substantial improvement in eQTL discovery at each of the standard permutation thresholds used ( Table S1B ) . The improvement ( twice as many eQTLs at 10−3 PT ) is consistent in all tissues . The high eQTL replication between twin sets persists after FA , with an additional improvement of true positives detection in skin: π1 = 0 . 95 ( Table 1 ) . As expected , FA correction recovers the majority of the eQTLs discovered with the initial analysis ( 90% of LCL and fat and 80% of skin ) ensuring that proximal genetic effects have not been corrected out . The FA correction enabled the discovery of additional signals ( Table S2 ) likely representing real effects that could not be detected initially due to low power . This is supported by the significant overrepresentation of low association p-values ( π1 = 0 . 99 , Figure 1 ) estimated in the uncorrected data for eQTLs detected only after FA correction . Direct tissue overlap of significant eQTLs supports an extensive level of tissue-specificity for the three tissues , with very similar proportions in both the SRC and SRC-FA analyses ( Figure 2 ) . In the first co-twin set we discovered 858 eQTL genes ( non-redundant union ) at 10−3 PT in all three tissues ( Table 2 ) . Of these , 106 genes ( 12 . 35% ) are shared across all tissues , 139 ( 16 . 2% ) are shared in at least two tissues and 613 genes ( 71 . 44% ) are detected in only one tissue . In skin we detect proportionally fewer tissue-specific effects ( 10 . 02% of skin eQTLs are specific to skin at 10−3 PT ) , an observation likely due to tissue heterogeneity and larger variety of present cell-types . SRC-FA results confirm the estimated ∼30% of eQTLs to be shared in at least two tissues based on threshold eQTL discovery ( Table S3 ) . Tissue-specific effects are largely not due to tissue-specific expression of the underlying transcripts . We detected regulatory variants active only in one tissue for genes that are expressed at high levels in the other two tissues ( Figure S1 ) . The strength of tissue-specificity was investigated further by performing a joint repeated-measures ANOVA analysis with the tissue modelled as a categorical predictor variable ( i . e . tissue type comprised the repeated measure ) . We assessed the relationship to the genotype by inspecting the SNP×tissue interaction p-value term . As expected , we detected a large enrichment of significant SNP×tissue interaction p-values for all associations ( π1 = 0 . 56 ) with tissue-specific effects having higher enrichment ( π1 = 0 . 6 ) than shared ones ( π1 = 0 . 41 ) ( Figure S2 ) . The enrichment in the shared category suggests additional attributes of tissue-specificity beyond statistical significance , as presented in the succeeding fold change analysis . The direction of allelic effects for shared eQTLs ( 10−3 and 10−2 PT ) is consistent across the three given tissues ( Figure S3 ) . As expected , for eQTLs significant in one tissue only the SRC correlation coefficient rho ( reflecting direction and magnitude of effects ) explains a substantially higher fraction of gene expression variation in the tissue of discovery compared to the other two tissues ( identical SNP-gene associations - Figure S4 ) . On the other hand , the amount of expression variance explained by shared eQTLs ( 10−3 PT ) is comparable across tissues . To refine regulatory signals and describe independently acting variants , we mapped eQTLs to recombination hotspot intervals and filtered markers in high LD ( Materials and Methods ) . We found that ∼7% of the genes tested are regulated by more than one independent cis eQTL , with similar estimates obtained from the standard and factor eQTL analysis ( Figure S5 ) . For finer comparison of eQTL effects , we conducted an analysis where sharing was required for both the gene and the genomic interval harboring the eQTL . This analysis yielded similar counts of tissue-shared and specific effects ( Tables S4 , S5 ) , suggesting that the vast majority of shared genes also share regulatory variants across tissues . Furthermore , as shown previously [7] , we observed that eQTLs cluster symmetrically around the TSS , with shared effects being distributed tightly around the TSS and tissue-specific effects spanning a greater range of distances ( Figures S6 , S7 ) . The results described so far are based on thresholds , which are driven by statistical significance . Overlaps at these levels are heavily dependent on power and affected by winner's curse . In addition , eQTLs sharing statistical significance may still have notable effect differences on gene expression levels across tissues , with potentially different biological consequences . Given these caveats , we examined tissue-specificity in a continuous manner by quantifying the proportion of true positives estimated from the enrichment of low p-values ( π1 ) . Specifically , the p-value distribution of significant SNP-probe pairs ( 10−3 PT ) from a reference tissue was investigated in the other two tissues . The p-value distribution in the other tissues indicates a high degree of tissue sharing ( 53 to 80% ) both with the SRC and SRC-FA , varying slightly depending on the reference tissue in the comparison ( Table S6 ) . This suggests that there are effect size differences ( both fold change and amount of variance explained ) among tissues for the same regulatory variants , which is the basis for the previously described higher eQTL tissue-specificity estimates [7] . Overall , 29% of eQTLs ( 1-mean π1 ) are estimated with the continuous approach to be tissue-specific , when comparing the three tissues studied . As described above , tissue overlap of eQTLs should encompass not only sharing of a statistically significant regulatory effect , but also a similar effect size ( fold change in expression ) of that variant across tissues . In this respect , we report the fold change as the difference between the gene expression means of the heterozygous and major homozygous genotypic classes . Within the same tissue , the two co-twin sets are only slightly different in their fold change estimates . These minor differences reflect most probably the winner's curse effect ( 0 . 96 Pearson's correlation of fold change between Twin 1 and Twin 2 in LCL , 0 . 93 in skin and 0 . 93 in fat - Figure 3 , Figures S8 , S9 ) . The difference in estimated effect size is much more apparent however across tissues ( e . g . LCL eQTLs have a 0 . 69 and 0 . 77 fold change correlation with skin and fat eQTLs respectively , skin eQTLs have a 0 . 69 fold change correlation with fat eQTLs ) . This is largely a consequence of eQTL tissue-specificity , but a small effect of winner's curse is also expected ( as observed in the comparison of co-twin sets ) . Furthermore , additional possible hidden tissue-specific effects are implied by the fact that shared eQTLs ( at the same threshold of significance ) don't always share the same effect size across tissues ( LCL fold change correlation of 0 . 78 in skin and 0 . 84 in fat for shared eQTLs i . e . up to 20% difference in fold change magnitude between tissues compared to within-tissue difference ) . This suggests that even statistically tissue-shared eQTLs have additional dimensions of tissue-specificity and their mere discovery in multiple tissues does not guarantee similar magnitude of consequences .
We have performed eQTL analysis in one cell-line ( LCL ) and two primary tissues of clinical importance ( skin – previously uncharacterized and fat ) . For each tissue we report robust eQTLs replicating in independent samples with identical ( MZ ) or on average 50% similar ( DZ ) genetic background using a matched co-twin design ( MCTA ) . To further increase our power to detect eQTLs and uncover smaller genetic effects , we applied factor analysis accounting for global variance components in the data . We refined our signals to detect independently acting cis eQTLs and for most genes we found single associated regulatory variants . When these variants are shared across tissues , they also share the same direction of allelic effects and map to the same recombination hotspot interval . Using threshold-based criteria , tissue overlap of eQTLs supports a large degree of tissue-specificity for the three tissues studied . However , this estimate is dependent on power and we therefore put forth a continuous measure of tissue-specificity that provides a refined view of the decay of statistical significance as well as fold change effect on gene expression . Using this approach we observed a significant overrepresentation of low p-values in all pairwise tissue comparisons , indicating larger proportions of shared statistically significant regulatory effects , some yet to be discovered with bigger sample sizes . However , we also observed significant eQTLs at the same threshold exhibiting differential fold changes in expression between genotypes across tissues . These cases represent tissue-specific effects as well , since differential fold change in expression is likely to have different biological consequences . Overall biological interpretation of regulatory effects - much like in the case of complex traits – is tissue-dependent , highlighting the value of multiple tissue expression datasets . Understanding such complexities and context-dependent effects in the genetic architecture of gene expression and other cellular phenotypes is essential for the interpretation of the biological properties of disease causing variants .
All individuals recruited in this study were Caucasian female twins aged between 40 and 87 years old ( mean age 62 ) . Skin punch biopsies ( N = 196 ) were taken from a relatively photo-protected area adjacent and inferior to the umbilicus . The fat sample was then carefully dissected from the same skin biopsy incision . A peripheral blood sample to generate lymphoblastoid cell lines ( LCL ) was taken contemporaneously . For a full description of the biopsy technique see Text S1 . RNA levels were measured in LCL , skin and fat using Illumina's whole-genome expression array HumanHT-12 version 3 as previously described [5] . Each sample had three technical replicates . Illumina's v3 probes were mapped to unique Ensembl gene IDs by combining and cross-checking two methods . The first approach used Illumina's probe annotation to RefSeq IDs . These were further queried with BioMart ( Ensembl 54 ) for corresponding Ensembl genes . RefSeq IDs mapping to multiple EnsGenes were excluded . The second approach used BLAT to map the 50-mer probe sequences to Ensembl transcripts and to extract genomic locations matching for all 50 bases of the probe sequence . Probes with unique perfect match to the genome and corresponding transcripts matching to the same genes were kept . The union of the two mappings after excluding 196 conflictingly matching probes resulted in 27 , 499 probes corresponding to 18 , 170 autosomal genes available for association analysis . Genotyping has been performed in parallel using Illumina's 1M-Duo and 1 . 2M-Duo custom chips on different subsets of individuals . Before further filtering , there were 106 samples with call rate ( CR ) ≥0 . 90 on the 1 . 2M and 88 samples with CR≥0 . 90 on the 1M chip . Combined intensity files were created for Illuminus [26] by retaining on a per-chromosome basis only SNPs common to both chips . Additionally , any SNPs that moved position between the two chips were removed . Following further quality checks ( Hardy-Weinberg p>10−4 , MAF>1% ) , 865 , 544 SNPs were kept for analysis . The overlapping set of successfully genotyped samples with available expression data amounted to 156 ( LCL ) , 160 ( skin ) and 166 ( fat ) individuals . Log2 - transformed expression signals were normalized separately per tissue as follows: quantile normalization was performed across the 3 replicates of each individual followed by quantile normalization across all individuals . The eQTL analysis was done separately for each tissue . Within each tissue , twins from the same pair were separated by id in two samples analyzed independently . This separation resulted in the following sample size for LCL , skin and fat respectively: Twin 1 ( 74 , 76 , 79 ) and Twin 2 ( 82 , 84 , 87 ) . Associations between SNP genotypes and normalized expression values were conducted using Spearman Rank Correlation ( SRC ) . We considered only SNPs in cis , i . e . within a 1MB window from the TSS . We assess the statistical significance of the nominal associations using permutations as previously described [5] . We call an eQTL significant at 10−3 permutation threshold ( PT ) if the nominal association P-value is greater than the 0 . 001 tail of the minimal P-value distribution resulting from the SNP's associations with 10 , 000 permuted sets of expression values for each gene . We applied a Bayesian factor analysis model [25] to the expression data in each tissue . This approach uses an unsupervised linear model to account for global variance components in the data , and yields a residual expression dataset that can be used in further analysis . We tested a wide range of parameter settings for the model , controlling the amount of variance explained by it . This was achieved by setting the parameters of the prior distributions for gene expression precision ( inverse variance ) and factor weight precision . These random variables are modelled using Gamma distributions , thus we varied their natural exponential family parameters - the prior mean and number of prior observations . We varied the prior mean from 10−6 to 10−2 , and number of prior observations from N*10−3 to N , where N is the number of observations from data , and learned 120 latent factors . In the subsequent analysis , we used for each tissue the residual dataset that gave the best eQTL overlap between the two twin samples . The prior values used for each dataset are given in Table S7 . The eQTL analysis on the corrected expression data was performed identically to the standard analysis: SRC followed by permutation testing . For quantifying eQTL replication and tissue sharing in a continuous way , we used Storey's QVALUE software [23] ( implemented in the R package qvalue 1 . 20 . 0 , default recommended settings ) . The program takes a list of p-values and computes their estimated π0 - the proportion of features that are truly null - based on their distribution ( the assumption used is that p-values of truly alternative cases tend to be close to zero , while p-values of null features will be uniformly distributed among [0 , 1] ) . The quantity π1 = 1−π0 estimates the lower bound of the proportion of truly alternative features , i . e . the proportion of true positives ( TP ) . Replication and sharing between two samples is reported as the proportion of TP ( π1 ) estimated from the p-value distribution of independent eQTLs discovered in sample 1 in the second sample ( exact SNP-probe combinations are tested ) . We refined the eQTL signals in order to characterize likely independent effects per gene . For this purpose , we mapped all common autosomal SNPs to recombination hotspot intervals as defined by McVean et . al [27] . We map significant eQTLs to recombination hotspot intervals and save the most significant SNP per gene . For each gene , SNPs resulting from this mapping are in addition filtered for LD in a pairwise manner ( for each pair with D′>0 . 5 the least significant SNP is ignored ) . This filtering ensures that true shared effects ( interval-gene combinations ) are compared and not just genes . | Regulation of gene expression is a fundamental cellular process determining a large proportion of the phenotypic variance . Previous studies have identified genetic loci influencing gene expression levels ( eQTLs ) , but the complexity of their tissue-specific properties has not yet been well-characterized . In this study , we perform cis-eQTL analysis in a unique matched co-twin design for three human tissues derived simultaneously from the same set of individuals . The study design allows validation of the substantial discoveries we make in each tissue . We explore in depth the tissue-dependent features of regulatory variants and estimate the proportions of shared and specific effects . We use continuous measures of eQTL sharing to circumvent the statistical power limitations of comparing direct overlap of eQTLs in multiple tissues . In this framework , we demonstrate that 30% of eQTLs are shared among tissues , while 29% are exclusively tissue-specific . Furthermore , we show that the fold change in expression between eQTL genotypic classes differs between tissues . Even among shared eQTLs , we report a substantial proportion ( 10%–20% ) of significant tissue differences in magnitude of these effects . The complexities we highlight here are essential for understanding the impact of regulatory variants on complex traits . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genome",
"expression",
"analysis",
"genomics",
"gene",
"expression",
"genetics",
"population",
"genetics",
"biology",
"human",
"genetics",
"genetics",
"and",
"genomics"
] | 2011 | The Architecture of Gene Regulatory Variation across Multiple Human Tissues: The MuTHER Study |
Recombinant interferon-alpha ( IFN-α ) is an approved therapy for chronic hepatitis B ( CHB ) , but the molecular basis of treatment response remains to be determined . The woodchuck model of chronic hepatitis B virus ( HBV ) infection displays many characteristics of human disease and has been extensively used to evaluate antiviral therapeutics . In this study , woodchucks with chronic woodchuck hepatitis virus ( WHV ) infection were treated with recombinant woodchuck IFN-α ( wIFN-α ) or placebo ( n = 12/group ) for 15 weeks . Treatment with wIFN-α strongly reduced viral markers in the serum and liver in a subset of animals , with viral rebound typically being observed following cessation of treatment . To define the intrahepatic cellular and molecular characteristics of the antiviral response to wIFN-α , we characterized the transcriptional profiles of liver biopsies taken from animals ( n = 8–12/group ) at various times during the study . Unexpectedly , this revealed that the antiviral response to treatment did not correlate with intrahepatic induction of the majority of IFN-stimulated genes ( ISGs ) by wIFN-α . Instead , treatment response was associated with the induction of an NK/T cell signature in the liver , as well as an intrahepatic IFN-γ transcriptional response and elevation of liver injury biomarkers . Collectively , these data suggest that NK/T cell cytolytic and non-cytolytic mechanisms mediate the antiviral response to wIFN-α treatment . In summary , by studying recombinant IFN-α in a fully immunocompetent animal model of CHB , we determined that the immunomodulatory effects , but not the direct antiviral activity , of this pleiotropic cytokine are most closely correlated with treatment response . This has important implications for the rational design of new therapeutics for the treatment of CHB .
Approximately 250 million individuals live with chronic hepatitis B ( CHB ) , and over half a million people are estimated to die each year due to CHB-associated liver diseases , such as cirrhosis and hepatocellular carcinoma ( HCC ) [1] . End-points of therapies for CHB are virological response ( durable reduction in serum HBV DNA levels to a degree which varies by therapy ) , serological response ( HBV e antigen ( HBeAg ) loss and seroconversion to anti-HBe in HBeAg-positive patients ) and biochemical response ( normalization of ALT levels ) . However , sustained loss of HBV surface antigen ( HBsAg ) off therapy is currently considered the ideal end-point . Recombinant interferon-α ( IFN-α ) is licensed for the treatment of CHB , but in contrast to potent nucleos ( t ) ides , virologic response is limited to a subset of patients [2] . Conversely , the rate of durable HBsAg loss is higher with IFN-α than with nucleos ( t ) ides , although still only occurs in <10% patients [2] . Despite more than two decades of clinical use , the mechanisms by which IFN-α controls HBV in responders are not well understood [3] . Defining the molecular basis for response remains an important goal , since mechanistic understanding of IFN-α activity could drive rational design of novel immunotherapeutic strategies and may lead to the identification of novel biomarkers of treatment response and/or patient stratification . IFN-α is a pleiotropic cytokine that has both direct antiviral and immunomodulatory properties [4 , 5] . With regard to the former , IFN-α induces the expression of hundreds of interferon-stimulated genes ( ISGs ) , many of which have antiviral effector functions [4] . Although the identification of key restriction factors has been challenging , various studies have indicated that IFN-α induces antiviral effectors of HBV . Most notably , the direct antiviral response to IFN-α has been demonstrated to inhibit the formation or accelerate the decay of replication-competent HBV capsids [6–9] , inhibit virion secretion [10] , reduce transcription from the viral genome ( cccDNA; covalently closed circular DNA ) [11 , 12] , and to induce non-cytolytic degradation of cccDNA [13] . The direct antiviral activity of IFN-α is consistent with the reduction in viral antigen levels by high dose pegylated IFN-α in HBV-infected humanized mice that lack immune cells [14] . The immunomodulatory properties of IFN-α include activation of NK cells and B cells , as well as both direct and indirect activation of CD8+ T cell function [5 , 15] . Despite this potential to activate both innate and adaptive immunity , recent studies have revealed that IFN-α treatment boosts the number and function of NK cells in the periphery , but does not improve peripheral HBV-specific CD8+ T cells responses [16–19] . Antiviral and mechanistic studies of IFN-α treatment of HBV infection have been performed in vitro , in transgenic and immunodeficient mouse models , and in peripheral blood from CHB patients , but there is very little data regarding the intrahepatic response to IFN-α treatment in an immunocompetent host . A baseline ( i . e . pre-treatment ) intrahepatic transcriptional signature of response to treatment with pegylated IFN-α and adefovir ( response defined as HBeAg loss , HBV DNA <2 , 000 IU/mL and ALT normalization ) has recently been described [20] . However , due to the difficulty in obtaining multiple liver biopsy specimens from chronically infected HBV patients , longitudinal evaluation of the intrahepatic response to IFN-α treatment is only possible with an animal model . Since ethical and cost considerations limit the use of chimpanzees for biomedical research and there is no small animal model of natural HBV infection , we selected the woodchuck model for this purpose . The Eastern woodchuck ( Marmota monax ) is naturally infected with WHV , a hepadnavirus which is genetically closely related to human HBV [21] . WHV infection displays a disease course similar to that in HBV-infected persons [21] . Although the woodchuck model has been used in a number of studies to characterize antiviral response to IFN-α treatment [22 , 23] these studies relied on adenovirus delivery of woodchuck IFN-α or utilized a recombinant human hybrid ( B/D ) IFN-α . Furthermore , these studies did not define the molecular basis of antiviral response . We recently described the sequencing , assembly and annotation of the woodchuck transcriptome , together with the generation of custom woodchuck microarrays . Using this new platform , we established the translational value of the woodchuck model and characterized the immune determinants of WHV clearance during self-limiting infection [24 , 25] . Since these studies yielded important insights into immune responses in the liver during hepadnavirus infection , in the current study we used a similar approach to characterize the intrahepatic transcriptional signature associated with antiviral response to recombinant woodchuck IFN-α treatment .
The amino acid sequence and in vitro antiviral activity of woodchuck IFN-α5 ( wIFN-α ) have previously been described [26 , 27] . wIFN-α was expressed , purified and biological activity confirmed as described in the Methods . The tolerability and pharmacodynamic activity of wIFN-α were then evaluated in a single dose study in WHV-negative woodchucks . Subcutaneous administration of a single dose of 2 , 20 or 200 μg wIFN-α per animal ( n = 3/group ) , induced dose-dependent increases in ISG and cytokine mRNA expression in the blood relative to the placebo group ( S1 Fig ) . Pharmacokinetic ( PK ) analysis of serum wIFN-α levels was not performed due to the lack of a sufficiently sensitive quantitative method ( see Methods ) . There was a trend towards changes in several hematological and clinical chemistry parameters at the higher doses , although these were likely due to the drawing of large blood volumes over a short time period . The antiviral efficacy of wIFN-α was then evaluated in a repeat-dose study in adult woodchucks chronically infected with WHV . To model vertical transmission in humans , chronic infection in these animals was established by neonatal WHV infection . The study design is described in Fig 1 . To match the frequency of non-pegylated IFN-α dosing in CHB patients , animals ( n = 12/group ) were dosed subcutaneously three times per week ( TIW ) on Monday , Wednesday and Friday with either placebo ( vehicle control ) or wIFN-α for a total of 15 weeks . Based on activity and safety considerations from the single dose study in WHV-negative woodchucks , the 20 μg dose was selected as the starting dose for the efficacy study . Initially wIFN-α was given for 7 weeks at a low dose of 20 μg/animal TIW . However , since an interim analysis indicated that this dose did not induce a significant decline in serum WHsAg or WHV DNA ( Figs 2 and 3A ) , at the start of week 7 the wIFN-α dose was increased to 100 μg/animal TIW . Thus , in the wIFN-α treatment group , animals received a low dose of wIFN-α for 7 weeks ( 21 doses total ) , followed by a high dose of 100 μg/animal for another 8 weeks ( 24 doses total ) . Note that one animal in this group ( M1004 ) was excluded from the analyses described below since it was likely naturally clearing WHV as the study initiated ( Table 1 ) . In contrast to low dose ( 20 μg ) wIFN-α , high dose ( 100 μg ) wIFN-α treatment induced a rapid decline in serum WHsAg and WHV DNA ( Figs 2 and S2 ) , which was statistically significant relative to the placebo group ( Fig 3A ) . The maximum reduction of serum WHsAg and WHV DNA was at week 16 in most animals , with a mean maximal reduction of 2 . 0 log10 for WHsAg and 3 . 0 log10 for viral load . Notably , wIFN-α treatment induced the complete loss of detectable ( <20 ng/mL ) WHsAg in one animal ( F1022 ) , although WHV DNA was still detectable ( >1 , 000 genome equivalents ( ge ) /mL ) at all time-points ( S2 Fig ) . After completion of treatment there was WHsAg and WHV DNA rebound in most woodchucks , albeit not always to pre-treatment levels ( Figs 2 and S2 ) . There was a high degree of variability in the antiviral response of individual woodchucks in regard to the kinetics and magnitude of serum WHsAg and WHV DNA decline , as well to the time interval between cessation of treatment and return of these viral parameters to pre-treatment levels ( S2 Fig ) . For correlative analyses with treatment response ( see below ) , response groups were defined as the following: R , responder ≥1 log10 reduction in WHsAg at week 15 ( end-of-treatment ) and week 23 ( end-of study ) ( n = 3 animals ) ; PR , partial responder ≥1 log10 reduction in WHsAg at week 15 but not week 23 ( n = 2 animals ) ; NR , non-responder <1 log10 reduction in WHsAg at week 15 and week 23 ( n = 2 animals ) ( Table 1 ) . Notably , baseline ( pre-treatment ) levels of serum WHsAg and WHV DNA were comparable in these different treatment response groups ( Table 1 ) . The four animals in the wIFN-treatment group that did not survive until end-of study ( see below ) , together with animal M1004 which was likely naturally clearing infection , were excluded from treatment response analyses ( Table 1 ) . High dose wIFN-α treatment significantly reduced intrahepatic cccDNA , WHV DNA replicative intermediate ( RI ) and WHV RNA levels ( Figs 3B and S3 ) . Reductions in these intrahepatic parameters typically correlated with reductions in serum WHsAg and viral load ( Table 1 ) . Only two woodchucks ( M1004 and F1022 ) with sustained WHsAg reduction developed consistently detectable anti-WHs antibodies ( S1 Table ) , one of which ( M1004 ) was likely naturally clearing WHV as the study initiated ( Table 1 ) . The overall seroconversion rate was therefore 0/9 ( placebo group ) and 1/7 ( wIFN-α group ) for animals that survived until end-of-study ( excluding M1004 ) . wIFN-α treatment was well-tolerated , and there were no signs of overt toxicity based on gross observations , body weights , hematology or clinical chemistry . Although several animals died during treatment , the causes of death ( e . g . HCC-related conditions , biopsy complications ) were likely not treatment related ( Table 1 ) . There was a trend towards elevated serum ALT and AST levels during high dose treatment , but on a group level these overall differences were not statistically significant ( Fig 3C ) . This is reflected in a poor temporal association between peak antiviral response and elevation of ALT , AST and SDH in some animals ( Fig 4 ) . Similarly , even though there was considerable fluctuation in liver histology scores in both placebo and wIFN-α groups ( S1 Table ) , antiviral response was correlated temporally with an increase in liver inflammation in some ( although not all ) wIFN-treated animals ( S4 Fig ) . Conversely , baseline liver enzyme levels and pre-treatment histology scores were comparable in the different treatment response groups ( Figs 4 and S4 ) . wIFN-α treatment induced dose-dependent increases in blood ISG mRNA expression . There was significant induction at both low and high dose levels , with a larger increase observed for the higher dose ( Fig 5A ) . In contrast , only high dose treatment significantly induced the expression of various T helper cell type 1 ( TH1 ) -type cytokines ( Fig 5B ) . Given that only high dose treatment was associated with a significant antiviral response , this suggests cellular immunity ( and associated cytokines ) may play a role in and/or be a useful biomarker of treatment response . Although comparative analysis is limited by small animal numbers in each response group , a role for cellular immunity in antiviral response is also suggested by the significant difference in IFN-γ expression in animals with an on-treatment response ( R and PR ) relative to those with no treatment response ( NR ) ( S5 Fig ) . As outlined in Fig 1 , intrahepatic transcriptional profiles of placebo-treated and wIFN-treated animals were determined by RNA-Seq at various times during the study . RNA-Seq was performed rather than using the microarray platform from previous studies [24 , 25] because this method has superior concordance with qRT-PCR data [28] and also enabled generation of a more complete ( version 2 ) woodchuck transcriptome assembly ( S2 Table ) . Principal Component Analysis ( PCA ) demonstrated that wIFN-α treatment substantially altered gene expression within the liver of chronic carrier animals ( S6 Fig ) . In contrast to the significant difference in antiviral response , there were only relatively modest differences ( restricted to PC#2 ) between intrahepatic transcriptional changes induced by low dose ( 20 μg ) and high dose ( 100 μg ) wIFN-α treatment . A gene module approach [29] confirmed that there was substantial modulation of intrahepatic gene expression by wIFN-α overall , with only moderate differences between low and high dose treatment ( Fig 6 ) . The modular signature for wIFN-α treatment revealed an increase ( >10% of the transcripts in each module significantly up-regulated ) in the number of differentially expressed genes in the IFN response ( Module , M3 . 1 ) , cytotoxic cell ( NK cell/CD8+ T cell ) ( M2 . 1 ) , plasma cell ( M1 . 1 ) , B cell ( M1 . 3 ) , myeloid cell lineage ( M1 . 5 and M2 . 6 ) and inflammation ( M3 . 2 ) modules ( Fig 6 ) . Consistent with an increase in liver inflammation in many wIFN-treated animals ( S4 Fig ) , the transcriptional data suggest that wIFN-α induced migration of immune cells into the liver and/or proliferation of intrahepatic immune cells . In contrast to the differential antiviral response ( Fig 3A and 3B ) and dose-dependent ISG induction in the periphery ( Fig 5A ) , module analysis revealed a striking increase ( >80% of the transcripts significantly up-regulated ) in the intrahepatic IFN response module ( M3 . 1 ) at all on-treatment time-points , regardless of wIFN-α dose ( Fig 6 ) . Consistent with the modular analysis , low dose and high dose wIFN-α treatment were both associated with strong induction of a large number of intrahepatic ISGs , including many antiviral effector genes ( Fig 7A , cluster 3 ) . Furthermore , there was no apparent difference between the intrahepatic expression of these ISGs in animals with a treatment response ( R and PR ) and those with no treatment response ( NR ) . Comparable induction of select ISGs in the liver by low and high dose wIFN-α treatment ( regardless of treatment response ) was confirmed by qRT-PCR ( Fig 7B , S4 Table ) . Taken together , these data indicate that the antiviral response to wIFN-α does not correlate with the intrahepatic expression of the majority of ISGs , suggesting they do not play a key role in the antiviral response to treatment ( see Discussion ) . Furthermore , pre-treatment ( week -3 ) ISG levels were comparable in the different response groups ( Fig 7A ) , indicating that baseline ISG expression was not an important determinant of treatment response . In the context of defining the molecular basis of IFN-α treatment response , the APOBEC proteins are ISGs of particular interest since various family members have been reported to be restriction factors for HBV [13] . It is therefore notable that the intrahepatic expression profile of APOBEC3H ( A3H ) was unlike the majority of antiviral ISGs , in that it was selectively induced by high dose wIFN-α treatment ( Table 2 ) . However , the degree of A3H induction was modest ( maximum 3 . 6-fold ) relative to many other ISGs , consistent with low A3H induction by IFN-α in purified primary human hepatocytes [13] . Furthermore , intrahepatic induction of A3H was only statistically significant at end-of-treatment ( week 15 ) , suggesting that it is not likely to be a main mediator of the wIFN-α antiviral response . In contrast to A3H , A3D and A3F were not significantly modulated ( FDR<0 . 05 , FC>2 ) by wIFN-α treatment . Other APOBEC3 family members ( including A3A ) were not available in the woodchuck transcriptome assembly . Since there was a strong association between wIFN-α dose and antiviral response ( Fig 3A and 3B ) , we reasoned that determining which genes were selectively induced by high dose wIFN-α would enable the identification of genes and/or pathways closely associated with treatment response . This approach identified genes that were selectively modulated during high dose wIFN-α treatment ( S7 Fig , high dose n = 468 ) , as well as genes induced only by low dose treatment ( low dose n = 29 ) or by both low and high dose wIFN-α ( low & high dose n = 775 ) . The full gene list from each set is displayed in S8 Table . Consistent with the previous analyses , module analysis ( M3 . 1 ) and Ingenuity Pathway Analysis ( IPA ) confirmed significant induction of an IFN-α response at both low dose and high dose wIFN-α treatment ( S8 Fig ) . In contrast , module analysis revealed that cytotoxic cell ( NK cell/CD8+ T cell ) responses were selectively induced by high dose wIFN-α treatment , and hence were temporally associated with treatment response ( Fig 8A ) . Significant enrichment of NK and T cell signatures with high dose wIFN-α treatment was confirmed by IPA ( Fig 8B ) . To complement the approach focused on identifying genes selectively induced by high dose wIFN-α , Weighted gene coexpression network analysis ( WGCNA ) was used to identify modules of co-regulated treatment-induced genes that correlated most closely with antiviral response ( S5 Table , Modules 1 and 2 ) . These modules were also significantly enriched for NK and T cell associated genes ( S9 Fig ) , consistent with the trend for induction of an NK/T cell signature in animals that had an antiviral response to treatment ( M2 . 1 , S10 Fig ) . Notably , these diverse analytical approaches identified common intrahepatic transcriptional signatures associated with treatment response , suggesting that NK/T cells play an important role in the antiviral response to wIFN-α treatment . On the individual gene level , induction of T cell associated genes ( CD3D , CD8A ) suggests that there is migration of T cells into the liver and/or proliferation of intrahepatic T cells during high dose wIFN-α treatment ( Table 2 ) . Expression of the T cell TH1-type transcription factor T-bet ( TBX21 ) was also significantly induced during high dose treatment ( Table 2 ) . Strikingly , qRT-PCR analysis revealed that T-bet expression was strongly induced by high dose treatment in animals with treatment response but not in animals without an antiviral response ( Fig 8C and S6 Table ) . This is notable since it may indicate improved functionality ( antigen-specific proliferation and IFN-γ production ) of intrahepatic HBV-specific CD8+ T cells , particularly since high dose wIFN-α also induced IL-12 expression ( Fig 5B ) [30] . However , it is important to note that this transcriptional analysis cannot determine whether T-bet is expressed by virus-specific or virus non-specific CD8+ T cells , or potentially other cell types [31] . Induction of NKG2D ( KLRK1; activating receptor ) expression , but not NKG2A ( KLRC1; inhibitory receptor ) , CD16 ( FCGR3A ) or CD56 ( NCAM1 ) ( Tables 2 and 3 ) , is consistent with activation , but not migration or proliferation of intrahepatic NK cells . As discussed previously , the peak antiviral response to treatment and elevation of liver injury biomarkers were temporally correlated in some animals ( Fig 4 ) , indicating that wIFN-α induced killing of WHV-infected hepatocytes . This biochemical evidence of liver damage is consistent with intrahepatic induction of the receptor-mediated cell death genes TRAIL ( TNFSF10 ) , Fas ( FAS ) and Fas ligand ( FASLG ) and the cytotoxic effector gene perforin ( PRF1 ) during high dose treatment ( Table 3 ) . These genes , as well as a death receptor signaling pathway ( S8 Fig ) , were also significantly induced ( on a group level ) by low dose wIFN-α , consistent with liver enzyme elevations in some animals during this treatment period ( Fig 4 ) . Notably , although there was substantial induction of TRAIL expression ( >17-fold ) by high dose wIFN-α in two animals with a treatment response , one responder animal had only modest intrahepatic TRAIL induction ( animal F1013; maximal 6-fold induction ) , and an animal with no treatment response had the greatest TRAIL induction ( animal F1014; 118-fold ) ( S6 Table ) . This overall poor correlation of intrahepatic TRAIL with treatment response suggests that additional antiviral mechanisms may be required to control infection . CD8+ T cells and NK cells have the potential to inhibit HBV infection by non-cytolytic mechanisms mediated by IFN-γ and TNF-α , as well as by killing infected cells via cytotoxic effector molecules . It is therefore notable that the two genes induced to the greatest degree in the liver by high dose wIFN-α treatment , PLA2G2A and CXCL9 , are IFN-γ responsive genes [24] ( Table 2 ) . Furthermore , PLA2G2A , CXCL9 and other IFN-γ inducible genes ( as well as IFN-γ itself ) are members of a subset of intrahepatic ISGs that correlated with wIFN-α dose ( Figs 7A , Cluster 2 , and S11 ) . Strikingly , a large number of IFN-γ-regulated genes ( e . g . MHC class I and II ( HLA ) genes , CXCL9 ) were also induced in the liver of chimpanzees during clearance of acute HBV infection [32] . In addition , although the on-treatment profile was not determined , MHC class I and II genes as well as CXCL9 were also up-regulated prior to treatment in the liver of CHB patients that subsequently responded to pegylated IFN-α and adefovir treatment compared to non-responder patients [20] . Consistent with the association between blood IFN-γ expression and antiviral response ( S5 Fig ) , high dose wIFN-α significantly induced intrahepatic CXCL9 expression in animals with a treatment response , but not in those without an antiviral response ( Fig 8C and S6 Table ) . These data indicate that IFN-γ-mediated , non-cytolytic mechanisms may play a role in the antiviral response to wIFN-α treatment . This is supported by the observation that the initial reduction in WHsAg and WHV DNA by high dose treatment in two responder animals ( F1013 and F1022 , weeks 7–15 and 7–11 , respectively ) occurred in the absence of substantial liver enzyme elevations ( Fig 4 ) . In both animals , there were subsequently modest increases in liver enzyme levels together with a further decrease in viral levels , suggesting that the antiviral response induced by high dose wIFN-α treatment is mediated by both cytolytic and non-cytolytic NK/T cell responses . In addition to positive effects on antiviral immunity , wIFN-α also induced various counter-regulatory mechanisms that may have limited the antiviral response to treatment . Notably , intrahepatic mRNA levels of the inhibitory T cell receptor PD-1 ( PDCD1 ) and its ligand PD-L1 ( CD274 ) were significantly increased during wIFN-α treatment ( Tables 2 and 3 ) . Intrahepatic expression of indoleamine 2 , 3-dioxygenase 1 ( IDO1 ) , which limits the availability of the essential amino acid tryptophan and produces immunosuppressive kynurenine to locally suppress T cells [33] , was also significantly increased by wIFN-α treatment ( Table 3 ) . Furthermore , high dose wIFN-α modestly elevated intrahepatic FOXP3 mRNA levels ( Table 2 ) , which suggests treatment-associated migration and/or proliferation of T regulatory cells ( Tregs ) that may negatively regulate CD8+ T cell and NK cell function . In contrast , expression of IL-10 ( IL10 ) and TGF-β ( TGFB1 ) , immunosuppressive cytokines produced by Tregs and various other cells , was not significantly modulated by treatment ( Table 3 ) .
Recombinant IFN-α has been used to treat CHB for over 20 years , but the molecular basis of treatment response remains poorly understood [3] . Previous transcriptome analyses have shown there are important parallels between the immune response to WHV in woodchucks and HBV in man [24] , and that self-limiting hepadnavirus infection in woodchucks and chimpanzees share key immunological features [25] . Together these studies suggest that the woodchuck is a relevant model to study the mechanisms that govern antiviral response to IFN-α . Consequently , we characterized the intrahepatic transcriptional profile of WHV chronic carrier woodchucks during treatment with recombinant woodchuck IFN-α . Treatment with wIFN-α produced variable antiviral effects , inducing multi-log reduction in serum WHV DNA and WHsAg in a subset of animals , and sustained WHsAg loss and seroconversion to anti-WHsAb in one animal , while not exerting antiviral effects in other animals . Importantly , the variability and degree of antiviral response in these animals are comparable to those observed with CHB patients treated with pegylated IFN-α [2 , 34] . Furthermore , viral rebound in the WHV-infected woodchucks was typically observed following cessation of wIFN-α treatment , consistent with the low rate of durable HBsAg loss in patients treated with IFN-α [2] . Together , these data reveal important parallels between the IFN-α treatment response of chronic hepadnavirus infection in woodchucks and man , establishing the translational value of the woodchuck model for characterizing the immune correlates of IFN-α treatment response . Since various studies have demonstrated that IFN-α can directly inhibit HBV [6 , 11 , 13] , a striking finding of this study was that the antiviral response to wIFN-α did not correlate with the intrahepatic induction of the majority of antiviral ISGs . Since WHV is sensitive to the direct antiviral effects of wIFN-α in vitro [23 , 35] , our data suggest that IFN-induced antiviral effectors of WHV do not play a key role in the antiviral response to treatment in vivo . However , there are several important caveats to consider . Firstly , since there are a large number of antiviral ISGs , and not all were available in the woodchuck transcriptome ( e . g . APOBEC3A ) , intrahepatic expression of antiviral ISGs that were not evaluated in this study may correlate with treatment response . In addition , since intrahepatic transcriptional analysis was restricted to 6 hours post-dose , it is possible that the expression of certain ISGs with slower induction kinetics may be associated with the antiviral response to IFN-α . Secondly , although low dose wIFN-α induced intrahepatic ISG expression but not a significant antiviral response , it is conceivable that prolonged ISG expression ( 7 weeks ) by low dose treatment played an important role in the antiviral response subsequently induced by higher dose wIFN-α . Finally , transcriptional analysis of whole biopsy tissue cannot define cell-specific ISG expression , which may be important in treatment response [36] . This is also an important caveat if hepatocytes and non-parenchymal cells ( e . g . Kupffer cells ) display markedly different sensitivity to ISG induction , since it may preclude accurate correlation of treatment response with induction of antiviral ISGs in infected cells using whole biopsy tissue . The significant difference in ISG induction by low and high dose wIFN-α in the blood but not the liver of woodchucks chronically infected with WHV is noteworthy considering a recent study demonstrating that HBV can inhibit IFN-α signaling in human hepatocytes [37] . This suggests that WHV may limit ( although not abrogate ) wIFN-α signaling in woodchuck hepatocytes . Alternatively , induction of USP18 , SOCS1 and SOCS3 ( Table 3 ) and/or other inhibitors of IFN-α/β receptor signaling may limit the intrahepatic ISG response to wIFN-α treatment . Since liver biopsies were not taken after wIFN-α treatment of WHV-negative animals , and there is currently no sensitive , quantitative wIFN-α ELISA ( see Methods ) , additional studies will be required to determine whether there are significant differences in PK-PD responses to wIFN-α treatment in WHV-negative and WHV-infected animals . In contrast to intrahepatic ISG expression , the expression of other gene sets showed a correlation with antiviral response . Both NK/T cell and IFN-γ transcriptional signatures in the liver were increased in animals with antiviral response to wIFN-α treatment . The peak antiviral response was also associated with liver enzyme elevations in some ( although not all ) animals . Collectively these data suggest that the antiviral response induced by wIFN-α treatment was mediated by both cytolytic and non-cytolytic NK/T cell responses . The correlation of liver injury biomarkers with antiviral response is notable because host-induced ALT flares are associated with IFN-α treatment response in CHB patients [38] . The association of intrahepatic NK cell and IFN-γ transcriptional signatures with antiviral response to treatment is also striking because NK cells in CHB patients have a markedly impaired capacity to produce IFN-γ [39 , 40] . This dysfunctional phenotype can be reversed ( at least in NK cells in the periphery ) by treatment with IFN-α [17] , which suggests that NK cell IFN-γ production may represent a common mechanism of IFN-α antiviral response to chronic hepadnavirus infection in woodchucks and man . Clearance of acute HBV infection in chimpanzees is also characterized by an intrahepatic IFN-γ transcriptional signature [32] , suggesting that there are important parallels between the immunological mechanisms of natural clearance of HBV and those induced by IFN-α treatment . Recent studies have revealed that IFN-α treatment does not improve peripheral HBV-specific CD8+ T cell responses [16–18] . In view of the aforementioned NK cell activation by IFN-α , this failure to augment virus-specific CD8+ T cell responses may be explained , at least in part , by the observation that NK cells can directly kill HBV-specific CD8+ T cells via TRAIL and other mechanisms [41] . The induction of an intrahepatic NK signature as well as TRAIL expression suggests that WHV-specific CD8+ T cell responses may be inhibited by similar mechanisms during wIFN-α treatment . Conversely , IFN-induced protection of antiviral CD8+ T cells might limit NK regulation of T cell immunity in this setting [42 , 43] , consistent with the induction of an intrahepatic T cell transcriptional signature coupled with significant elevation of T-bet ( TBX21 ) mRNA during wIFN-α treatment . In addition to potentially inducing NK cell killing of virus-specific CD8+ T cells , wIFN-α treatment induced various counter-regulatory mechanisms , including intrahepatic PD-1 ( PDCD1 ) and PD-L1 ( CD274 ) expression , which may also have limited antiviral CD8+ T cell function in the liver . However , it is important to the note that a limitation of the woodchuck model is that it is challenging to confirm that changes in gene expression are associated with corresponding changes in protein levels and/or cellular function . This is particularly important for characterization of CD8+ T cell specificity in the context of wIFN-α treatment , since studies in HBV transgenic mice as well as CHB patients indicate that antigen-nonspecific inflammatory cells ( including nonvirus-specific CD8+ T cells ) can accumulate to high levels in the liver under inflammatory conditions [44 , 45] . Unfortunately , blood volume and biopsy material limitations precluded functional analysis of WHV-specific CD8+ T cells in the current study . In addition , the lack of woodchuck-specific immunological reagents prevented immunophenotyping of WHV-specific CD8+ T cells by flow cytometry . Attempts to develop high-quality monoclonal antibodies against woodchuck CD56 and CD8a to enable detection of NK and CD8+ T cells , respectively , by immunohistochemistry were also not successful . Therefore , additional studies in immunocompetent models of natural infection and/or CHB patient biopsies will be required in order to define the relative contribution of intrahepatic NK and virus-specific CD8+ T cells to IFN-α treatment response . In summary , by studying recombinant IFN-α in an immunocompetent animal model of CHB , this study provided new insights into the immune mechanisms that mediate the antiviral response to treatment . In addition , various immune pathways were identified that may act to limit treatment response . These findings have important implications for the design of new therapeutics for CHB , and also provide rationale for evaluating combinations of immunotherapeutic agents currently in development .
The sequence of woodchuck IFN-α5 ( wIFN-α ) has previously been described [26] . Recombinant wIFN-α was expressed by transient transfection of human embryonic kidney ( HEK ) 293F cells using the FreeStyleTM 293 expression system according to the manufacturer’s instructions ( Invitrogen , Inc . , Carlsbad , CA ) . Culture supernatant was filtered and then purified by two chromatographic steps . Firstly , after adjusting to pH 6 . 0 with 50 mM KH2PO4 , pH 5 . 0 , the sample was loaded on a 5 mL SP HP Hi Trap ( GE Healthcare , Little Chalfont , Buckinghamshire , UK ) that had been pre-equilibrated with 50 mM KH2PO4 , pH 6 . 0 . The wIFN-α was then eluted with a 17 column-volume salt gradient from 0–500 mM NaCl . Fractions were analyzed via SDS-PAGE and wIFN-containing fractions were pooled . Secondly , size exclusion chromatography on Superdex 75 ( GE Healthcare , Little Chalfont , Buckinghamshire , UK ) was performed in 20 mM His/HCl , 140 mM NaCl pH 6 . 0 . The eluted wIFN-α was filtrated with a 0 . 22 μM syringe filter and stored at -80°C . The wIFN-α concentration was determined by measuring optical density ( OD ) at 280 nm . Purity and monomer content were confirmed by SDS-PAGE and SE-HPLC , respectively , and the integrity of the wIFN-α amino acid backbone was verified by Nano Electrospray QTOF mass spectrometry . The protein was kept in a storage buffer ( 20 mM His/HCl , 140 mM NaCl pH 6 . 0 ) prior to dosing . The endotoxin level of the wIFN-α preparation was <0 . 454 EU/mL . The in vitro biological activity of wIFN-α was confirmed by dose-dependent induction of mRNA levels of the interferon-stimulated genes ( ISGs ) Mx1 and OAS1 in woodchuck PBMCs ( n = 2 animals ) treated with 0 . 1 , 1 and 10 μg/mL wIFN-α . The animal protocol and all procedures involving woodchucks were approved by the Georgetown University IACUC ( Protocol Number: 11–006 ) and adhered to the national guidelines of the Animal Welfare Act , the Guide for the Care and Use of Laboratory Animals , and the American Veterinary Medical Association . All woodchucks used in this study were obtained from Northeastern Wildlife . Prior to the study , male woodchucks were confirmed negative for WHV surface antigen ( WHsAg ) and for antibodies against WHsAg ( anti-WHsAb ) and WHV core antigen ( anti-WHc ) . Animals were assigned to four groups ( n = 3/group ) using stratification based on body weight , clinical biochemistry and hematology . Animals received a single subcutaneous dose of 2 , 20 or 200 μg wIFN-α , or a placebo control ( all n = 3/group ) . Various measurements ( body weight , body temperature , clinical serum chemistries , and CBCs ) were obtained to monitor drug safety . All woodchucks used in this study were obtained from Northeastern Wildlife . These woodchucks were born in captivity and were infected at 3 days of age with the cWHV7P2a inoculum containing WHV strain WHV7-11 . cWHV7P2a has the same biological and virological characteristics as the cWHV7P2 inoculum as both were derived from cWHV7P1 [46] . Chronically infected animals were all anti-WHs negative , with detectable serum WHV DNA , WHsAg and anti-WHc at approximately 1 year post-infection . Absence of liver tumors in woodchucks with low GGT was confirmed by ultrasonography . Chronic WHV carrier woodchucks were assigned and stratified by gender , body weight , and by pretreatment serum markers ( WHsAg and WHV DNA concentrations , serum GGT and SDH activities ) into treatment and placebo groups ( n = 12/group ) . The study design and sampling scheme are summarized in Fig 1 . The PK of wIFN-α was not measured due to the lack of a suitable analytical method . Although a wIFN-α ELISA has previously been described [27] , it was discovered during method development that one of the antibodies likely recognized the 6xHis tag of the antigen used for immunization , which was not present in our preparation of wIFN-α . Despite extensive screening of available anti-human , anti-macaque , anti-mouse and anti-pig IFN-α antibodies ( PBL , Piscataway , NJ ) , as well as additional anti-woodchuck IFN-α antibodies ( Digna Biotech , Pamplona , Spain ) , none were identified that robustly detected wIFN-α in an ELISA format . Serum WHV DNA was quantified by two different methods depending on concentration: dot blot hybridization or real time PCR assay on a 7500 Real Time PCR System instrument ( Applied Biosystems , Foster City , CA ) as described previously [47] . Serum WHsAg and anti-WHsAb were measured by WHV-specific enzyme immunoassays as described [48] . Liver WHV RNA was measured quantitatively by Northern blot hybridization as previously described [49] . Liver WHV DNA replicative intermediates ( RI ) and WHV cccDNA were quantitatively determined by Southern blot as previously described [50] . The revised woodchuck transcriptome assembly ( version 2 ) consists of a previous assembly ( version 1 ) , generated with Roche-454 sequencing data [24] , that was merged with newly assembled contiguous transcripts ( contigs ) from Illumina sequencing data of the 24 animals from the current study ( n = 12 placebo , n = 12 wIFN-α treated ) . The main improvement of version 2 over version 1 is that the sequencing depths of the Illumina data is significantly higher than that of 454 and therefore resulted in a higher dynamic range and increased number of genes as compared to assembly version 1 ( S2 Table ) . The assembly method of transcriptome version 2 consisted of three stages: 1 ) initial contig assembly , 2 ) contig annotation and 3 ) contig refinement . First , Illumina RNA-Seq paired-end reads from liver samples were assembled using Trinity [51] ( release 2011-08-20 ) . The obtained contigs were further refined and merged by applying the sequence assembly algorithm PHRAP [52] . As a result , the number of contigs was reduced by about 25% and the contig lengths were increased . Second , all contigs were subjected to an in-house developed gene annotation pipeline which performs sequence homology searches within reference transcript databases from other species . First , woodchuck contigs were mapped to transcripts from RefSeq reference database containing human , mouse , and rat transcripts using BLAST [53] , with a 1 . e-5 E-value cutoff . Matches with the highest BLAST scores were further pair-wise aligned by applying the Needleman-Wunsch algorithm [54] in order to obtain more accurate alignments and to calculate the sequence identities ( i . e . number of identical nucleotides in percentage of alignment length ) between RefSeq transcripts and woodchuck contigs . If the identity difference between the two best hits exceeded 25% , then the top gene was used for contig annotation . Only contigs that could be mapped to known mouse , rat or human genes were used for further data processing . Because the assembly often contained more than one contig per gene , a final sequence refinement was then performed to remove redundancies . Contigs annotated with identical genes were subjected to the CAP3 assembler [55] , and as a result , the number of contigs was further reduced and the sequence lengths of numerous contigs were increased . Sequencing libraries were created using Illumina’s TruSeq RNA sample preparation kit ( San Diego , CA ) according to manufacturer’s protocol . Total RNA was purified using oligo ( dT ) magnetic beads , fragmented , and reverse-transcribed using SuperScript II ( Invitrogen , Inc . , Carlsbad , CA ) to synthesize first strand cDNA . After second strand synthesis , Illumina specific adapters containing unique barcodes were ligated to the ends of the double-stranded cDNA . Fragments containing adapters on both ends were then enriched and amplified with PCR , quantified with qPCR , and run on the Agilent Bioanalyzer DNA-1000 chip to estimate fragment size . Samples were then multiplexed and sequenced on the Illumina 2500 . The data was demultiplexed using CASAVA and run through FastQC ( http://www . bioinformatics . babraham . ac . uk/projects/fastqc/ ) to assess sequencing data quality . Paired-end 50 nucleotide read data from mRNA-Seq were mapped against the revised woodchuck transcriptome with Bowtie2 [56] and prioritized for concordant paired alignments with unique hits . The resulting SAM/BAM files were processed with SAMtools [57] to yield count data that was normalized and processed by DESeq [58] for differential expression analysis and subsequent pattern recognition and pathway analysis . Multiple testing correction was performed using the method of Benjamini and Hochberg [59] . Principal component analysis was performed with Partek Genomics version 6 . 6beta ( Partek , St . Louis , MO ) . Heatmaps of the expression data were generated by unsupervised hierarchical clustering of least square means expression values , after z-score normalization across samples . The enrichment of differential genes relative to the gene modules described previously [29] was calculated with R version 2 . 13 . 2 ( http://www . r-project . org ) using the humanized gene symbols for the woodchuck genes . Gene Set Enrichment Analysis ( GSEA ) was performed as previously described [60] , with ranks determined by the multiplicative product of the fold-change and–log ( FDR ) values for each gene . Weighted gene coexpression network analysis ( WGCNA ) [61] was performed within the R statistics environment . Pathway analysis was performed using Ingenuity Pathway Analysis ( Ingenuity Systems , Redwood City , CA ) . Total RNA was isolated using the RNeasy Mini Kit ( Qiagen Inc . , Redwood City , CA ) with on-column DNase digestion using the RNase-Free DNase Set ( Qiagen Inc . , Redwood City , CA ) . Following reverse transcription into cDNA with the Transcriptor First Strand cDNA Synthesis Kit ( Roche Applied Sciences , Indianapolis , IN ) , samples were analyzed by real time PCR on a 7500 Real Time PCR System instrument ( Applied Biosystems , Inc . , Foster City , CA ) using EagleTaq Universal Master Mix ( Roche Applied Sciences , Indianapolis , IN ) . Target gene expression was normalized to 18S rRNA expression . The primers and probes used in this study are displayed in S7 Table , or have previously been described [24 , 62–65] . Note that , although there was insufficient mRNA from biopsy samples to perform extensive qRT-PCR validation of gene expression , in contrast to microarray , RNA-Seq has high concordance with qRT-PCR data [28] . | Approximately 250 million people are chronically infected with HBV , and over 500 , 000 people die every year because of associated liver diseases . IFN-α has been used to treat patients with chronic HBV infection for over 20 years , but it is not well understood why some patients respond to treatment and others do not . In large part , this is because it is not practicable to obtain liver samples to characterize the intrahepatic response to IFN-α in patients with different treatment outcomes . In this study we used the woodchuck model of chronic HBV infection to study how IFN-α changes gene expression patterns in the liver during treatment . Surprisingly , we found that the treatment response did not correlate with the expression of antiviral effector genes that have previously been shown to mediate the direct antiviral effects of IFN-α in vitro . Instead , we found that the response to IFN-α treatment was associated with the presence of select immune cells ( natural killer cells and T cells ) in the liver . Our work also indicates that these immune cells inhibit the virus by killing infected cells , as well as in ways that do not require killing of liver cells . Altogether , our study suggests that new therapies that stimulate these immune cells in the liver may hold promise for the treatment of chronic HBV infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Intrahepatic Transcriptional Signature Associated with Response to Interferon-α Treatment in the Woodchuck Model of Chronic Hepatitis B |
Environmental enteropathy ( EE ) is a poorly understood condition that refers to chronic alterations in intestinal permeability , absorption , and inflammation , which mainly affects young children in resource-limited settings . Recently , EE has been linked to suboptimal oral vaccine responses in children , although immunological mechanisms are poorly defined . The objective of this study was to determine host factors associated with immune responses to an oral cholera vaccine ( OCV ) . We measured antibody and memory T cell immune responses to cholera antigens , micronutrient markers in blood , and EE markers in blood and stool from 40 Bangladeshi children aged 3–14 years who received two doses of OCV given 14 days apart . EE markers included stool myeloperoxidase ( MPO ) and alpha anti-trypsin ( AAT ) , and plasma endotoxin core antibody ( EndoCab ) , intestinal fatty acid binding protein ( i-FABP ) , and soluble CD14 ( sCD14 ) . We used multiple linear regression analysis with LASSO regularization to identify host factors , including EE markers , micronutrient ( nutritional ) status , age , and HAZ score , predictive for each response of interest . We found stool MPO to be positively associated with IgG antibody responses to the B subunit of cholera toxin ( P = 0 . 03 ) and IgA responses to LPS ( P = 0 . 02 ) ; plasma sCD14 to be positively associated with LPS IgG responses ( P = 0 . 07 ) ; plasma i-FABP to be positively associated with LPS IgG responses ( P = 0 . 01 ) and with memory T cell responses specific to cholera toxin ( P = 0 . 01 ) ; stool AAT to be negatively associated with IL-10 ( regulatory ) T cell responses specific to cholera toxin ( P = 0 . 02 ) , and plasma EndoCab to be negatively associated with cholera toxin-specific memory T cell responses ( P = 0 . 02 ) . In summary , in a cohort of children 3–14 years old , we demonstrated that the majority of biomarkers of environmental enteropathy were positively associated with immune responses after vaccination with an OCV .
The immunogenicity and efficacy of oral vaccines are lower in developing countries compared to those in developed countries . This hypo-responsiveness is predominantly seen in lower age group children [1 , 2 , 3 , 4] . Environmental enteropathy ( EE ) is an acquired syndrome , characterized by villous blunting , crypt hyperplasia , and increased intraepithelial lymphocytes and pro-inflammatory cytokine responses [5 , 6] . EE is known to be common in settings with poor water , sanitation and hygiene infrastructure such as in low- and lower-middle income countries . It is hypothesized that repeated enteric infections are the underlying cause of this subclinical condition , which also result in reduced efficacy of vaccines . However , the mechanisms underlying this entity are not well understood [7 , 8 , 9] Cholera is a dehydrating diarrheal disease endemic in more than 50 countries across the world . It is caused by infection with Vibrio cholerae , present in contaminated water and food [10] . The disease burden is highest in children under 5 years of age [11] . However , currently available and WHO recommended oral cholera vaccines ( OCV ) show less pronounced immune responses and protection in this age group of children compared with older children and adults [1 , 12 , 13] . Little is known regarding how host factors , such as malnutrition , micronutrient deficiency , and enteropathy , affect vaccine-induced immune responses in children . The objective of this study was to determine how host factors impact immune responses in children receiving oral cholera vaccine . To do so , we examined markers of enteropathy and micronutrients in the plasma and stool of 40 Bangladeshi children who received two doses of an OCV , Dukoral , and correlated those with immune responses to the vaccine .
As previously reported [12 , 19] , we carried out this study in the urban slums of Mirpur , an area of Dhaka , Bangladesh . After obtaining written informed consent of the parents/ guardians as well as assent of the children wherever applicable , we enrolled children ( age 3–14 years ) that were healthy , did not have diarrhea in the preceding 14 days , and who were given two doses of a licensed killed whole cell oral cholera vaccine ( OCV ) Dukoral ( WC-rBS ) , containing recombinant cholera toxin B subunit , 14 days apart . Individuals were recruited from February 22 , 2011 to March 07 , 2011 . We obtained anthropometric measurements before vaccination and measured height for age ( HAZ ) Z-scores using WHO Multicenter Growth Reference Study Child Growth Standards; subjects were excluded if they had a Z score of less than -2 . Subjects were also excluded if they had stool microscopy positive for parasites or had any febrile illness or antibiotic treatment in the past week . We obtained blood samples before vaccination ( D0 ) and 7 days after the second dose of vaccination ( D21 , Fig 1 ) . We also collected stool samples before vaccination ( D0 ) . This study was approved by the Ethical and Research Review Committees of the icddr , b and the Institutional Review Board of Massachusetts General Hospital . We determined vibrioicidal titers as previously described [15] . We quantified LPS- ( prepared in house ) [16] and cholera toxin B subunit- ( CTB , gifts of A . M Svennerholm , Göteborg University ) specific IgA , IgG and IgM antibody responses in plasma using a previously described kinetic enzyme-linked immunosorbent assay ( ELISA ) [14 , 17 , 18 , 19] . T cell responses by FASCIA ( Flow cytometric Assay of Specific Cell-mediated Immune responses in Activated whole blood ) We performed FASCIA for evaluation of antigen stimulated lymphoblast subpopulations in blood , as previously described [12 , 19] . For in vitro antigenic stimulation , cholera holotoxin containing the G33D variant homopentameric B subunit ( mCT , gifts of Randall K . Holmes , University of Colorado ) [21] was used . After 6 days of culture , we separated the supernatant from the stimulated cells by centrifugation and added a protease inhibitor cocktail , followed by storing supernatants at -80°C for subsequent cytokine analysis by Luminex . To characterize the cell populations by surface expression of markers , we incubated them with various antibodies , including anti-CD3-phycoerythrin -Texas Red ( Invitrogen , CA ) , anti-CD4-Amcyan , anti-CD45RA-V450 , anti-integrin 7-PE , anti-CXCR5-AF488 , anti-CCR7-PE-Cy7 , and anti-CCR9-AF647 fluorochrome conjugated monoclonal antibodies ( BD Bioscience , San Jose , CA ) . We used ammonium chloride ( Sigma ) solution to lyse red blood cells and the remaining lymphoblasts were resuspended in stabilizing fixative ( BD Bioscience , San Jose , CA ) . We then acquired cell populations using a FACSAria III flowcytometer ( BD Bioscience , San Jose , CA ) and analyzed the data using the FACSDiva ( BD Bioscience , CA ) and the FlowJo software ( TreeStar Inc . , Oregon ) . Cellular proliferative responses are presented as the ratio ( denoted as stimulation index , SI ) of lymphoblast count with antigenic stimulation to the count without any stimulation . SI value greater than “1” indicates V . cholerae antigen-specific stimulation in compared to without a V . cholerae antigen or no stimulation . Stored FASCIA culture supernatants were analyzed for different cytokines as per manufacturers’ instructions using the Milliplex human cytokine/ chemokine kit ( Millipore Corp . , MA ) and the Bio-Plex 200 system ( Bio- Rad , Pennsylvania ) . We selected cytokines for analysis based on their importance in relation to infection , cell differentiation and relation to gut enteropathy [12 , 22] . We performed ELISA to measure soluble CD14 ( sCD14; R&D Systems , Minneapolis , MN , USA ) , endotoxin core IgG antibodies ( EndoCAb; Hycult Biotech , Uden , Netherlands ) , and intestinal fatty acid binding protein ( i-FABP; R&D Systems , Minneapolis , MN , USA ) in plasma specimens [23] . All assays followed the instructions specified by the manufacturer . The samples were diluted at a ratio of 1:1000 , 1:200 and 1:10 for sCD14 , EndoCAb and i-FABP , respectively . We used flat-bottom MaxiSorp plates ( Nunc ) ( Thermo # 442404 ) for the i-FABP assay . We quantified levels of MPO ( Alpco , Salem , NH , USA; and Immundiagnostik , Bensheim , Germany ) in fecal extracts as described in the package insert , using a dilution factor of 1:200 . We assessed the concentration of AAT ( ImmuChromGmBH , NC , USA ) in fecal extracts , according to the package insert , using a dilution of 1:400 [7] . We measured retinol binding protein ( RBP ) and 25-OH vitamin D ( VitD ) by ELISA ( R&D Systems and Roche , respectively ) from plasma on day 0 . We analyzed antibody responses as the fold change of titer from day 0 to 21 . Memory T cell and cytokine responses were analyzed as the absolute change from day 0 to 21 . For our comparison of EE markers between younger and older children , we conducted statistical comparisons using the Mann Whitney U test . To look at associations between immune responses and host factors , we conducted multiple linear regression analysis with Least Absolute Shrinkage and Selection Operator ( LASSO ) regularization to identify predictive host factors truly informative for each response of interest , and the final model for each response was determined based on the optimal tuning parameter using 10 fold cross-validation criteria . The LASSO method was used in this study since there was no high co-linearity among potential host factors . From the soft-threshold property of the LASSO in a linear model [24] , the estimated regression coefficient is biased toward zero . To mitigate these bias problems , we reported a more unbiased estimation of the regression parameters from unpenalized multivariate linear regression using the selected factors in the LASSO . The age and gender covariates were added in all unpenalized multivariate linear regression models as default factors . All the LASSO analyses were performed using the "glmnet" package in R ( www . r-project . org ) . The unpenalized multivariate linear regression was fitted using the function "lm" in R ( www . r-project . org ) .
We measured enteropathy markers and micronutrients in 40 children given 2 doses of a cholera toxin B subunit-containing oral cholera vaccine 14 days apart ( Fig 1 ) . There were 20 children ages 3–5 ( “young children” ) and 20 children ages 7–14 ( “older children” ) . All the participants completed day 21 follow up . The characteristics of the cohort , by age grouping , are shown in Table 1 . We did not find any age-specific differences in retinol binding protein or vitamin D levels . We also found comparable levels of enteropathy markers between young children and older children , with the exception of sCD14 , which was higher in older children ( Fig 2 ) . We did not find any gender-specific differences in levels of EE markers . We assessed the variance inflation factor and did not find high levels of co-linearity between host factors ( VIF <4for all variables , range 1 . 18–1 . 78 ) . We had previously reported V . cholerae antigen-specific immunological responses for this cohort of oral cholera vaccinees [12 , 19] . We conducted multivariate linear regression using selected factors from LASSO regularization , modeled with age and gender , to determine associations between immune responses and host factors , including micronutrient and EE markers . We showed the results of this analysis in Table 2 , where all host factor ( s ) associated with each immunologic outcome with P < 0 . 10 are included . For each host factor , we determined an estimated effect as the changes in immunologic response with each one unit increase in the host factors . We found the LPS IgG response to be positively associated with plasma vitamin D ( P = 0 . 02 ) , i-FABP ( P = 0 . 01 ) , and sCD14 levels ( P = 0 . 07 ) . We found LPS IgA and CTB IgG responses to be positively associated with stool MPO ( P = 0 . 02 and 0 . 03 , respectively ) . We found that CT-specific effector memory T cell responses were positively associated with plasma i-FABP ( P = 0 . 01 ) and negatively associated with plasma EndoCab ( P = 0 . 02 ) , and that the CCR9 gut-homing effector memory T cell subset had similar associations as the parent population . We found a negative association between IL-10 produced by T cells stimulated with CT , and stool AAT ( P = 0 . 02 ) .
Live oral vaccines including oral cholera vaccines fail to show similar levels of immunogenicity in developing countries as in developed countries [2 , 3 , 4] . Many factors such as maternal antibodies ( placental or breast milk ) , infections at the time of vaccination , or gut enteropathy [2 , 6 , 8 , 25 , 26] , are hypothesized to be the underlying causes of these decreased immune responses . However , the degree to which each factor influences immune responses to vaccines are yet to be elucidated . Among the factors , EE is prevalent in children of low-income countries , and we hypothesized that EE may be a cause of the lower immunogenicity of the oral cholera vaccine Dukoral in Bangladesh . However , our data did not support this hypothesis , but in contrast , suggested that environmental enteropathy ( as evidenced by high enteropathy biomarkers ) was positively associated with increased immunogenicity to vaccine antigens , particularly for antibody responses . We examined several proposed biomarkers of enteropathy [5 , 7 , 27 , 28] . Intestinal Fatty Acid-Binding Protein 2 ( i-FABP2 ) is mostly expressed in intestinal enterocytes [29] , and studies have shown that i-FABP2 is an indicator of enterocyte damage and is associated with intestinal inflammation [30] . i-FABP2 is also involved in cell repair and proliferation [31 , 32] that may play crucial roles in immune responses to vaccines . A study of HIV enteropathy showed a concordant increase in i-FABP2 and duodenal helper T cells ( CD4+ ) [28] . Here , we found that cholera toxin ( CT ) specific effector memory and gut homing memory T cell responses after vaccination were positively associated with i-FABP2 . Similarly , MPO is known to be a specific marker of intestinal inflammation , and its concentration in stool has been associated with deficits in linear growth [33] [7] . We found that LPS IgA and CTB-IgG antibody responses to vaccination were positively associated with the concentration of MPO in stool . Soluble CD14 ( sCD14 ) is a glycoprotein that mediates the interaction of lipopolysaccharide ( LPS , endotoxin ) with antigen presenting cells such as macrophages to produce pro-inflammatory signalling in the presence of gram-negative bacteria [34 , 35 , 36] . sCD14 is expressed mainly by macrophages , and to a lesser extent by neutrophils [37] . In this study , we found a statistically non-significant association ( P = 0 . 07 ) between Vibrio cholerae LPS-specific IgG antibody responses to vaccination and sCD14 . Studies of oral vaccine responses in infancy have shown sCD14 to be both positively and negatively associated with oral polio vaccine antibody responses , which depended on the age of the child at which sCD14 was measured ( 6 weeks vs . 18 weeks of age ) [38] . We hypothesize that inflammation at different ages may have differential effects on the outcome of the immune responses . AAT is a protease inhibitor that crosses into the intestinal lumen as a result of increased gut permeability , and has been used as a marker of protein losing enteropathies [39] . High fecal AAT levels have also been associated with mucosal ulceration in acquired immunodeficiency syndrome ( AIDS ) patients [40] . Notable elevations of AAT are also seen in patients with shigellosis [41] . In this study , we found a negative association of CT specific , anti-inflammatory IL-10 responses with the level of AAT , consistent with the above findings that EE markers are associated with pro-inflammatory responses against vaccines . Campbell et al reported in a gut enteropathy study that the level of a regulatory cytokine ( TGF-β ) decreases with an increase in inflammatory cytokines [20] . While we saw a positive association between vitamin D levels and LPS antibody responses , we did not see any association between age , gender , blood group , or malnutrition ( as assessed by HAZ score ) and vaccine responses . While this exploratory study was not powered to detect such differences , our results suggest that enteropathy markers may be better predictors of vaccine responses than the markers of nutrition that were studied . Lower immunity to enteric vaccines in developing countries in children is now well established through a number of studies , including ones examining oral cholera vaccine responses [2 , 3 , 4] . Among the factors hypothesized [2] to play a role in decreased immunity to OCV is gut enteropathy , which is commonly seen in resource-limited settings and believed to play a crucial role in underperformance of enteric vaccines . However , contrary to this hypothesis , we found a positive correlation of markers of gut enteropathy and immune responses to a cholera vaccine ( Dukoral ) in a cohort of Bangladeshi children 3–14 years of age . We hypothesize that the increased inflammation seen in enteropathy may facilitate a greater number of antigen presenting cell encounters with vaccine antigens , and that this is accompanied by increased B and T cell responses . However , more detailed studies with mucosal specimens and other enteric vaccines are needed to further our understanding of this observation . The majority of published studies of the effect of enteropathy on vaccine responses have focused on infants [7 , 38] . This is one of the first studies to examine enteropathy markers in older children , and we postulate that enteropathy in older children may not have the same deleterious effects as those seen in infants . Notably , due to ethical and logistical limitations in infants , most of the enteropathy markers used in our study have only been histologically validated in adults . The study has several limitations . Notably , we did not account for the intestinal microbiota , which has recently been shown to impact vaccine responses in infancy [42] . There may also be other unmeasured confounders that could impact immune responses . Additionally , we did not look at mucosal responses such as secretory IgA , nor at memory B cell responses . Lastly , we excluded severely malnourished children , and thus our findings may not be applicable to this population . Despite these limitations , we demonstrate here a positive association of enteropathy markers with immune responses to an oral vaccine . Further studies are warranted to delineate the mechanism through which this occurs . | Cholera is a life-threatening diarrheal disease that affects millions of people worldwide . Currently available oral cholera vaccines are less effective in young children , and some have hypothesized that this is related to environmental enteropathy , a problem in the gut characterized by alterations in intestinal permeability , absorption , and inflammation , which mainly affects young children in resource-limited settings . We measured cholera-specific immune responses in 40 Bangladeshi children aged 3–14 who received an oral cholera vaccine . We then identified host factors , such as enteropathy biomarkers , sex , age , and micronutrient status , associated with each immune response . Unexpectedly , we found enteropathy biomarkers to be positively associated with immune responses to vaccine , underlining the complexity of the interaction between enteropathy and oral vaccine immunogenicity . | [
"Abstract",
"Introduction",
"Methods",
"and",
"Materials",
"Results",
"Discussion"
] | [
"enteropathies",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"body",
"fluids",
"immunology",
"tropical",
"diseases",
"vaccines",
"preventive",
"medicine",
"bacterial",
"diseases",
"gastroenterology",
"and",
"hepatology",
"neglected",
"tropical",
"dise... | 2016 | Biomarkers of Environmental Enteropathy are Positively Associated with Immune Responses to an Oral Cholera Vaccine in Bangladeshi Children |
Repression of somatic gene expression in germline progenitors is one of the critical mechanisms involved in establishing the germ/soma dichotomy . In Drosophila , the maternal Nanos ( Nos ) and Polar granule component ( Pgc ) proteins are required for repression of somatic gene expression in the primordial germ cells , or pole cells . Pgc suppresses RNA polymerase II-dependent global transcription in pole cells , but it remains unclear how Nos represses somatic gene expression . Here , we show that Nos represses somatic gene expression by inhibiting translation of maternal importin-α2 ( impα2 ) mRNA . Mis-expression of Impα2 caused aberrant nuclear import of a transcriptional activator , Ftz-F1 , which in turn activated a somatic gene , fushi tarazu ( ftz ) , in pole cells when Pgc-dependent transcriptional repression was impaired . Because ftz expression was not fully activated in pole cells in the absence of either Nos or Pgc , we propose that Nos-dependent repression of nuclear import of transcriptional activator ( s ) and Pgc-dependent suppression of global transcription act as a ‘double-lock’ mechanism to inhibit somatic gene expression in germline progenitors .
How germ cell fate is established and maintained is a century-old question in developmental , cellular , and reproductive biology . Metazoan species have two distinct modes of germline specification [1] . In some species , germline progenitors are characterized by inheritance of a specialized ooplasm , or the germ plasm , which contains maternal factors necessary and sufficient for germline development [2–7] . In other species , germline progenitors are specified by inductive signals from surrounding tissues [8 , 9] . Irrespective of the mode of germline specification , transcriptional repression of somatic genes is common in germline progenitors [10–16] , implying that this phenomenon is critical for separation of the germline from the soma . In Drosophila , the germ plasm is localized in the posterior pole of cleavage embryos ( stage 1–2 ) , and is partitioned into germline progenitors called pole cells ( stage 3–4 ) . In pole cells of blastoderm embryos ( stage 4–5 ) , the genes required for somatic differentiation are transcriptionally repressed by two maternal proteins in the germ plasm , Polar granule component ( Pgc ) and Nanos ( Nos ) [10 , 15 , 17] . Pgc is a Drosophila-specific peptide that suppresses RNA polymerase II-dependent transcription in pole cells by inhibiting positive transcriptional elongation factor b ( P-TEFb ) function [17] . By contrast , Nos is an evolutionarily conserved protein that plays an essential role in germline development in various animals [18] . For example , in Drosophila , pole cells lacking Nos ( nos pole cells ) can adopt a somatic , rather than a germline , fate [19] . Furthermore , depletion of Nos is reported to show ectopic expression of somatic genes , such as fushi tarazu ( ftz ) , even-skipped ( eve ) , and the sex-determination gene Sex lethal ( Sxl ) , in pole cells [15] . Thus , maternal Nos is required in pole cells for repression of somatic genes and establishment of the germ/soma dichotomy . However , the mechanism by which Nos represses somatic gene expression remains unknown . Nos acts as a translational repressor of mRNAs that harbor a discrete sequence motif called Nanos Response Element ( NRE ) in the 3´ UTR . NRE contains an evolutionarily conserved Pumilio ( Pum ) -binding sequence , UGU trinucleotide [20–22] . In abdominal patterning , Pum represses translation of maternal hunchback ( hb ) mRNA by binding to NREs in its 3´ UTR and recruiting Nos to the RNA/protein complex [23 , 24] . Deletion of the NREs from hb mRNA causes its ectopic translation in the posterior half of embryos , which in turn suppresses abdomen formation [25 , 26] . Furthermore , deletion of NREs causes hb translation in pole cells [25 , 26] , suggesting that NRE-dependent translational repression occurs in pole cells . Indeed , Nos represses translation of head involution defective ( hid ) mRNA in pole cells in an NRE-like-sequence-dependent manner [27] . In addition , Nos and Pum repress Cyclin B translation in pole cells by binding to a discrete sequence containing two UGU trinucleotides ( Cyclin B NRE ) [26] . These findings led us to speculate that Nos , along with Pum , represses somatic gene expression in pole cells by suppressing translation of mRNAs containing NRE or UGU in their 3´ UTRs . Here , we report that , in pole cells , Nos , along with Pum , represses translation of importin-α2 ( impα2 ) /Pendulin/oho31/CG4799 mRNA , which contains an NRE-like sequence in its 3´ UTR [28] . The impα2 mRNA encodes a Drosophila Importin-α homologue that plays a critical role in nuclear import of karyophilic proteins [28–31] . Nos inhibits expression of a somatic gene , ftz , in pole cells by repressing Impα2-dependent nuclear import of the transcriptional activator , Ftz-F1 . Based on our observations , we propose that Nos-dependent inhibition of nuclear import of transcriptional activators and Pgc-dependent global transcriptional silencing act as a ‘double-lock’ mechanism to repress somatic gene expression in pole cells .
Maternally supplied impα2 mRNA is distributed throughout cleavage embryos . When embryos develop to the blastoderm stage , impα2 mRNA is degraded in the somatic region , but not in pole cells , resulting in enrichment of impα2 mRNA in pole cells [28] ( Fig 1A ) . However , we found that expression of Impα2 protein was at background levels in pole cells [28] ( Fig 1D and 1G ) . Because impα2 mRNA contains a sequence very similar to the NRE ( hereafter , NRE-like sequence ) in its 3´ UTR [25 , 28] ( Fig 2A ) , we assumed that impα2 mRNA is a target of Nos/Pum-dependent translational repression in pole cells . To investigate this possibility , we first monitored the expression of the Impα2 protein in pole cells of embryos lacking maternal Nos or Pum ( nos or pum embryos , respectively ) . In these pole cells , expression of Impα2 protein was higher than in those of control ( nos/+ ) embryos ( Fig 1D–1F and 1I and S1 Fig ) . Because neither nos nor pum mutation affected the impα2 mRNA level in pole cells ( Fig 1B and 1C ) , these observations show that Nos and Pum repress protein expression from the impα2 mRNA in pole cells . We next investigated whether this repression is mediated by the NRE-like sequence in the impα2 3´ UTR . To this end , impα2 mRNA , with or without the NRE-like sequence ( impα2 WT and impα2 ΔNRE , respectively ) ( Fig 2A ) , was maternally supplied to embryos , and their protein expression was examined in pole cells at the blastoderm stage . Because a triple Myc tag sequence was inserted at the C-terminal end of the coding sequence , protein expression from these mRNAs could be monitored using an anti-Myc antibody . When impα2 WT mRNA was supplied to normal ( y w ) embryos , the tagged protein was expressed at low levels in the soma , but was barely detectable in pole cells ( Fig 2B , 2F and 2G ) . By contrast , the tagged protein from impα2 ΔNRE mRNA was detected in normal pole cells ( Fig 2C , 2F and 2G ) . Similar protein expression was observed in pole cells lacking Nos ( nos pole cells ) , when impα2 WT mRNA was supplied ( Fig 2E , 2F and 2G ) , as well as when impα2 ΔNRE mRNA was supplied ( Fig 2D , 2F and 2G ) . Because the frequency of tagged protein expression from impα2 ΔNRE mRNA did not increase in cells lacking Nos ( Fig 2F and 2G ) , these results indicate that the NRE-like sequence mediates Nos-dependent repression of Impα2 protein expression in pole cells . The NRE-like sequence of impα2 mRNA contains two UGU trinucleotides ( Fig 2A ) . The UGU trinucleotide is a core sequence of an RNA motif ( Nos-Pum SEQRS motif: 5´-HWWDUGUR ) that was highly enriched in a SEQRS ( in vitro selection , high-throughput sequencing of RNA , and sequence specificity landscapes ) analysis of the Nos–Pum–RNA ternary complex ( Fig 7 in the article [22] ) . Hence , we asked whether Pum and Nos form a ternary complex with impα2 mRNA in an NRE-like sequence–dependent manner . To address this question , we performed electrophoretic mobility shift assay ( EMSA ) using the Pum RNA-binding domain and the Nos protein containing Zn finger motifs and C-terminal region , which are reported to form a Nos–Pum–target RNA ternary complex in vitro [22] . We found that Nos and Pum together , but neither alone , formed a complex with impα2 RNA containing an NRE-like sequence ( WT ) ( Fig 2H and S2 Fig ) , whereas alteration of the NRE-like sequence ( mut ) ( S2 Fig ) abolished this interaction ( Fig 2H ) . These results demonstrate that Nos and Pum are able to interact with the impα2 3´ UTR in an NRE-like sequence–dependent manner . The observations described above led us to conclude that Nos , along with Pum , directly represses impα2 translation in pole cells . Impα2 is a Drosophila homologue of Importin-α that mediates nuclear import of karyophilic proteins with classical nuclear localization signal ( NLS ) [28–31] . We predicted that ectopic production of Impα2 in nos pole cells would cause aberrant nuclear import of NLS-containing karyophilic proteins . To explore this possibility , we focused on a transcriptional activator , Ftz-F1 , which contains a classical NLS and is expressed throughout early embryos , including pole cells [32–34] . In normal embryos , Ftz-F1 was enriched in the cytoplasm of pole cells , although it was in the nuclei of somatic cells ( Fig 3A , 3B , 3E , 3F , 3J and 3K ) . In the absence of maternal Nos , the percentage of embryos with Ftz-F1 signal accumulating in pole-cell nuclei was higher than in normal embryos ( Fig 3C , 3G , 3H and 3J ) . Furthermore , the nuclear/cytoplasmic ratio of Ftz-F1 signal intensities in nos pole cells was higher than in normal pole cells ( Fig 3G , 3H and 3K ) . To determine whether this aberrant concentration of Ftz-F1 was caused by mis-expression of Impα2 , we expressed Impα2 ectopically in pole cells of normal embryos ( Fig 1H and 1I ) . To this end , impα2 mRNA in which the 3´ UTR was replaced with the nos 3´ UTR , was maternally supplied under the control of the nos promoter; the mRNA was localized to the germ plasm and pole cells under the control of the nos 3´ UTR [35 , 36] . The percentage of these embryos ( impα2-nos3´UTR embryos ) with Ftz-F1 in pole-cell nuclei and the nuclear/cytoplasmic ratio of Ftz-F1 intensities in their pole cells were higher than those of normal pole cells ( Fig 3D , 3I , 3J and 3K ) . These observations suggest that mis-expression of Impα2 in pole cells caused by depletion of maternal Nos results in aberrant nuclear import of Ftz-F1 . Depletion of maternal Nos results in ectopic expression of the somatic genes ftz , eve and Sxl in pole cells [15] . Because Ftz-F1 is required for proper expression of ftz in the soma [37–41] , we asked whether mis-expression of Impα2 causes ectopic expression of ftz in pole cells . In normal embryos , ftz mRNA was expressed in seven stripes of somatic cells [42] , but never expressed in pole cells [percentage of embryos expressing ftz in pole cells ( pe ) = 0%; number of embryos examined ( n ) = 283] ( Fig 4A and 4G ) . By contrast , in impα2-nos3´UTR embryos , ftz mRNA was rarely detectable in pole cells ( pe = 8 . 9% , n = 45 ) ( Fig 4B , 4C and 4G ) . We assumed that this low frequency of ftz expression was due to Pgc-mediated silencing of global mRNA transcription . To test this idea , we expressed Impα2 in pole cells of embryos lacking maternal Pgc ( pgc impα2-nos3´UTR embryos ) , and found that the frequency of ftz expression was drastically increased ( pe = 51 . 4% , n = 74 ) ( Fig 4F and 4G ) , compared to those of impα2-nos3´UTR embryos ( P < 0 . 01 ) and the embryos lacking Pgc ( pgc embryos ) ( pe = 34 . 9% , n = 109 , P < 0 . 05 ) ( Fig 4D , 4E and 4G ) . A similar situation was observed in embryos lacking both Pgc and Nos activities ( pgc nos embryos ) ( Fig 5F and 5G ) . The percentage of embryos expressing ftz in pole cells was 82 . 8% ( n = 209 ) ( Fig 5H ) , an increase relative to 35 . 8% in pgc embryos ( n = 203 , P < 0 . 01 ) ( Fig 5B , 5C and 5H ) , and 7 . 2% in nos embryos ( n = 69 , P < 0 . 01 ) ( Fig 5D , 5E and 5H ) . Furthermore , ectopic ftz expression in pgc nos pole cells was suppressed by injecting double-stranded RNA ( dsRNA ) against impα2 ( Table 1 ) . Therefore , we conclude that ectopic expression of ftz in pole cells is cooperatively repressed by Nos-dependent suppression of Impα2 production and Pgc . In addition to ftz expression , eve was expressed ectopically in pole cells of pgc impα2-nos3´UTR embryos ( S3 Fig ) . Ectopic eve mRNA and its protein expression were significantly higher in pgc impα2-nos3´UTR pole cells than pgc or impα2-nos3´UTR pole cells ( S3 Fig ) . We next examined expression of the sex-determination gene Sxl in early pole cells , because Sxl is also repressed by nos in both male and female pole cells [15] . In males , Sxl mRNA expression was rarely detectable in pole cells of nos , impα2-nos3´UTR , pgc , and pgc impα2-nos3´UTR embryos ( P > 0 . 1 , vs . y w ) ( S4 Fig ) . By contrast , in females , the percentage of embryos expressing Sxl mRNA in pole cells was significantly higher in pgc impα2-nos3´UTR embryos than in impα2-nos3´UTR , and pgc embryos ( S4 Fig ) . These results indicate that eve and Sxl , like ftz , are cooperatively repressed in pole cells by Impα2 depletion and Pgc-dependent transcriptional silencing . Because there is no evidence for the involvement of Ftz-F1 in eve and Sxl expression , it is likely that Impα2 mediates nuclear import of other transcriptional activator ( s ) for eve and/or Sxl in pole cells . Nos is required in pole cells for mitotic quiescence , repression of apoptosis , and proper migration to embryonic gonads [19 , 43–45] . Hence , we asked whether mis-expression of Impα2 causes defects in these processes . First , using an antibody against a phosphorylated form of histone H3 ( PH3 ) , a marker of mitosis [46] , we investigated whether pole cells enter mitosis in stage 7–9 embryos . Premature mitosis was detected in pole cells of nos embryos , as described previously [43] , but never in pole cells of impα2-nos3´UTR or pgc impα2-nos3´UTR embryos ( Fig 6A ) . Second , using an antibody against cleaved Caspase-3 , a marker of apoptosis , we investigated whether pole cells enter apoptosis in stage 10–16 embryos . Pole cells never expressed the apoptotic marker in impα2-nos3´UTR embryos , whereas in pgc impα2-nos3´UTR embryos , 20 . 4% of pole cells expressed the apoptotic marker ( Fig 6B ) . The latter was statistically indistinguishable from pgc pole cells ( Fig 6B ) , which has been reported to enter apoptosis [47] . These data indicate that mis-expression of Impα2 does not affect apoptosis of pole cells even in the absence of pgc function . Last , we investigated whether mis-expression of Impα2 affects pole cell migration . The ability of pole cells to migrate properly into the embryonic gonads was never impaired in impα2-nos3´UTR embryos ( Fig 6C ) , and the percentage of pole cells entering the gonads in pgc impα2-nos3´UTR embryos was statistically indistinguishable from that of pgc pole cells ( Fig 6C ) , which has been reported to exhibit migration defect [47] . These observations indicate that mis-expression of Impα2 does not induce premature mitosis , apoptosis , or mis-migration of pole cells . This can be partly explained by the facts that Cyclin B and hid mRNAs are the targets for Nos-dependent translational repression regulating mitosis and apoptosis in pole cells , respectively [27 , 43] . During the course of the experiments described above , we happened to observe that impα2-nos3´UTR interacts genetically with the pgc mutation to cause dysgenic gametogenesis ( Fig 7 ) . Because almost all of the ovaries in females derived from pgc mothers mated with y w males were agametic , as reported previously [17] , we examined the effect of impα2-nos3´UTR in pgc/+ background ( Fig 7A ) . The percentage of dysgenic ovaries in pgc/+ impα2-nos3´UTR females derived from pgc/+ impα2-nos3´UTR mothers mated with y w males was significantly higher than those in pgc/+ and impα2-nos3´UTR females ( Fig 7A ) . In the dysgenic ovaries , almost all of the egg chambers fail to complete the vitellogenic stage , and consequently only a few mature oocytes were present ( S5 Fig ) . Furthermore , the percentages of dysgenic and agametic testes in pgc impα2-nos3´UTR males derived from pgc impα2-nos3´UTR mothers mated with y w males were higher than those in pgc and impα2-nos3´UTR males ( Fig 7B ) . In these testes , the abundance of Vasa-positive germline cells was reduced ( dysgenic ) or absent ( agametic ) ( S5 Fig ) . Because dysgenic and agametic gonads were barely detectable in females and males derived from reciprocal crosses ( Fig 7 ) , our data suggest that mis-expression of Impα2 from maternal transcript , concomitant with maternal pgc depletion in pole cells , causes defects in gametogenesis . However , we cannot test whether concomitant depletion of maternal Nos and Pgc causes a similar phenotype because nos pole cells degenerate before adulthood , even when apoptosis in these cells is genetically repressed [19] . Expression of Importin-α subtypes is spatio-temporally regulated in the soma during development in multiple animal species , including Drosophila , and they control nuclear transport of unique karyophilic proteins to activate different sets of somatic genes [30 , 48–54] . Drosophila genome contains three Importin-α family genes: impα1 , 2 , and 3 [28 , 49 , 55] . impα1/Kap-α1/CG8548 mRNA is not detectable in pole cells during early embryogenesis [56 , 57] , and its protein product is ubiquitously expressed at a very low level throughout embryogenesis [48] . By contrast , maternal impα3/Kap-α3/CG9423 mRNA is detectable in germ plasm during pole cell formation [58 , 59] , and production of Impα3 protein is upregulated during the blastoderm stage [55 , 58] ( S6 Fig ) . Because Impα3 production was independent of maternal nos activity ( S6 Fig ) , it is likely that Nos-dependent repression of Impα2 production is solely responsible for suppression of somatic gene expression in pole cells . By contrast , pole cells become transcriptionally active during gastrulation [60–64] , when Impα2 is undetectable in these pole cells [28] . Thus , the onset of zygotic transcription in pole cells may require Impα3-dependent nuclear import of transcription factors , in addition to the disappearance of Pgc and the alteration in chromatin-based regulation [10 , 17] . After gastrulation , maternal impα2 mRNA is rapidly degraded in pole cells , and neither impα2 mRNA nor protein is detectable in the germline before adulthood [28] . This suggests that maternal impα2 is dispensable for germline development , and that maternal impα2 mRNA partitioned into early pole cells must be silenced by Nos and Pum in order to suppress mis-expression of somatic genes . We found that depletion of maternal Nos activities caused mis-expression of ftz in pole cells . Although ftz expression was barely observed in pole cells lacking only maternal Nos , it was partially derepressed in pole cells in the absence of Pgc alone ( Figs 4G and 5H ) , probably because a trace amount of Ftz-F1 enters pole cell nuclei even in the absence of the impα2 translation . Therefore , we propose that a subset of somatic genes , including ftz and eve , are repressed in pole cells by two distinct mechanisms: Nos-dependent repression of nuclear import of transcriptional activators and Pgc-dependent silencing of mRNA transcription . Pgc inhibits P-TEFb-dependent phosphorylation of Ser2 residues in the heptad repeat of the C-terminal domain ( CTD ) of RNA polymerase II , a modification that is critical for transcriptional elongation [17]; thus , mRNA transcription in pole cells is globally suppressed by Pgc . By contrast , Nos inhibits transcription of particular genes by repressing Impα2-dependent nuclear import of the corresponding transcriptional activators . Nos is evolutionarily conserved and expressed in the germline progenitors of various animal species [18] . In C . elegans , nos-1 and -2 are essential for rapid turnover of maternal lin-15B mRNA , which encodes a transcription factor that would otherwise cause inappropriate transcriptional activation in primordial germ cells [65] . In the germline progenitors of Xenopus embryos , Nos-1 , along with Pum , destabilizes maternal VegT mRNA and represses its translation to inhibit somatic ( endodermal ) gene expression , which is activated by VegT protein [16] . Furthermore , in the germline progenitors ( small micromeres ) of sea urchin embryos , Nos silences maternal mRNA encoding a deadenylase , CNOT6 , to stabilize other maternal mRNAs inherited into small micromeres [66] . Here , we demonstrate that Nos inhibits translation of maternal impα2 mRNA in pole cells in order to suppress nuclear import of a transcriptional activator for somatic gene expression . Based on these observations , we propose that Nos silences maternal transcripts that are inherited into germline progenitors but deter the proper germline development . In addition to Nos-dependent silencing of maternal transcripts , transient suppression of RNA polymerase II elongation is observed during germline development of a wide range of animals , including Drosophila , C . elegans , Xenopus , and an ascidian , Halocynthia roretzi [17 , 67–69] . Therefore , we propose that the ‘double-lock’ mechanism achieved by Nos and global suppression of RNA polymerase II activity plays an evolutionarily widespread role in germline development .
y w was used as a normal strain . nosBN/nosBN [35 , 70] or nosBN/nosBN Df ( 3L ) H99 [19] were designated as nos/nos . nosBN/TM3 or nosBN/TM2 were designated as nos/+ . In ( 3R ) Msc/T ( 1;3 ) FC8 [23 , 71] , pgcΔ1/pgcΔ1 , pgcΔ1/Df ( 2R ) X58-7 , and pgcΔ1/CyO [17] are referred to as pum/pum , pgc/pgc , pgc/Df , and pgc/+ , respectively . nosBN and In ( 3R ) Msc/T ( 1;3 ) FC8 flies were gifts from R . Lehmann . nos-gal4VP16 ( nos-gal4 ) ( a gift from R . Lehmann ) [64] was used as a germline-specific driver . y1 M{vas-int . Dm}ZH-2A w*; M{3×P3-RFP . attP}ZH-58A ( Bloomington Drosophila Stock Center , Stock No . 24484 ) was used as y vas-φ-zh2A w; ZH-attP-58A [72] . Developmental stages of Drosophila embryos were determined according to Campos-Ortega and Hartenstein [77] . In this study , stage-4 embryos that had finished the 13th somatic nuclear division and retained round nuclei before cellularization were referred to as "late stage-4 embryos" . Antibody staining of embryos was performed as described previously [43] . For anti-Impα2 staining , embryos were fixed in 2 ml of 1:1 mixture of heptane and fixative I [3 . 7% formalin in PBS ( 130 mM NaCl , 7 mM Na2HPO4 , 3 mM NaH2PO4 ) ] for 10 min with vigorous shaking . Two different antibodies were used , anti-Impα2 23aa and anti-Impα2 2/3 ( gifts from B . M . Mechler ) , which were raised against the 23-amino acid residues of the C-terminal region and two-thirds of Impα2 protein , respectively [28 , 48] . For the experiments shown in Fig 1 , rabbit anti-Impα2 23aa antibody ( 1:50 dilution ) and Alexa Fluor 488-conjugated anti-rabbit IgG antibody ( 1:200 dilution , Molecular Probes ) were used . For the experiments shown in S1 Fig , rabbit anti-Impα2 2/3 antibody ( 1:40 dilution ) and biotinylated anti-rabbit IgG antibody ( 1:200 dilution , Vector Lab . ) were used . The signal was amplified using Vectastain ABC-AP kit ( Vector Lab . ) , and then detected with 5-bromo-4-chloro-indolyl phosphate ( BCIP ) /nitroblue tetrazolium ( NBT ) ( Boehringer Mannheim ) . Embryos were dehydrated in graded alcohol and mounted in Eukitt ( O . Kindler ) . We observed no significant difference in the results obtained using these two antibodies , except that anti-Impα2 23aa antibody often caused a non-specific signal on the embryo surface . For double-staining with anti-Ftz-F1 antibody and propidium iodide ( Fig 3 ) , embryos were fixed in 2 ml of 1:1 mixture of heptane and fixative II ( 4% paraformaldehyde in PBS ) for 5 min with vigorous shaking . Rabbit anti-Ftz-F1 antibody ( 1:500 dilution , a gift from H . Ueda ) and Alexa Fluor 488-conjugated anti-rabbit IgG antibody ( 1:500 dilution , Molecular Probes ) were used . The embryos were treated with RNase , and then stained with propidium iodide ( Sigma ) , as described previously [43] . For double-staining with anti-Ftz-F1 antibody and DAPI , the embryos were treated with DAPI ( 1 μg/ml , Sigma ) for 10 min , after anti-Ftz-F1 staining . For anti-Eve staining , embryos were fixed in 2 ml of 1:1 mixture of heptane and fixative II for 5 min with vigorous shaking . Guinea pig anti-Eve antibody 634 [1:200 dilution , Asian Distribution Center for Segmentation Antibodies at National Institute of Genetics ( NIG ) , Japan] [78] and Cy3-conjugated anti-guinea pig IgG antibody ( 1:500 dilution , Jackson ImmunoResearch ) were used . For the experiments shown in Fig 2 , embryos were fixed in 2 ml of 1:1 mixture of heptane and fixative I for 20 min . Mouse anti-Myc antibody 9E10 [1:100 dilution , Developmental Studies Hybridoma Bank ( DSHB ) at the University of Iowa] and HRP ( horse-radish peroxidase ) -conjugated anti-mouse IgG antibody ( 1:500 dilution , Bio-Rad ) were used . The signal was enhanced using the TSA-Biotin System and Streptavidin-FITC ( PerkinElmer Life Sciences , Inc . ) . For staining with antibodies against Vasa , PH3 , cleaved Caspase-3 , and Impα3 , embryos were fixed in 2 ml of 1:1 mixture of heptane and fixative II for 20 min . The following antibodies were used: chick anti-Vasa antibody ( 1:500 dilution , lab stock ) , rabbit anti-PH3 antibody ( 1:200 dilution , Upstate Biotechnology ) , rabbit anti–Caspase-3 antibody ab13847 ( lot no . 593692 , 1:1000 dilution , Abcam ) , and mouse anti–dKap-α3 antibody 5E3 ( 1:500 dilution , a gift from C . S . Parker ) . Signal was detected using Cy3-conjugated anti–chick IgY antibody ( 1:500 dilution , Jackson ImmunoResearch ) , Alexa Fluor 488–conjugated anti–rabbit IgG antibody A-11034 ( 1:500 dilution , Molecular Probes ) , or Alexa Fluor 488–conjugated anti–mouse IgG antibody A-11029 ( 1:500 dilution , Molecular Probes ) , as appropriate . Antibody staining of ovaries and testes was performed as previously described for the ovary [79] . Chick anti-Vasa antibody ( 1:500 dilution , lab stock ) and Alexa Fluor 488–conjugated anti–chick IgY antibody A-11039 ( 1:500 dilution , Molecular Probes ) were used . All embryos , ovaries , and testes stained with fluorochrome-conjugated secondaries were mounted in Vectashield ( Vector Laboratories ) or ProLong Diamond ( Molecular Probes ) . Z-stack confocal images were taken from each embryo using a Zeiss LSM 5 Pascal ( Zeiss ) , Zeiss LSM 510 Meta ( Zeiss ) , Leica TCS-NT ( Leica ) , or Leica TCS-SP8 ( Leica ) confocal microscope . Optical slices were analyzed using Zeiss LSM 5 Image Browser ( Zeiss ) , ImageJ , or Fiji software . In Figs 2F , 3J , 6A and S1 , the numbers of signal-positive pole cells located from the top to median plane of embryos were counted in confocal serial images . Digoxigenin ( DIG ) -labeled RNA probes were synthesized with SP6 , T7 , or T3 RNA polymerase in the presence of DIG-labeled uridine triphosphate ( UTP ) ( Boehringer-Mannheim ) , using full-length impα2 cDNA clone K9 , full-length 1817-bp ftz cDNA ( a gift from H . Ueda ) , a 985-bp eve cDNA fragment ( ntd 231–1215 of GenBank accession no . BT029151 ) , or an 848-bp Sxl cDNA fragment ( ntd 572–1419 of GenBank accession no . NM167112 ) as the template . Whole-mount in situ hybridization of embryos was performed essentially according to the methods reported by Tautz and Pfeifle [80] , with several modifications [81] . For staining with impα2 probe , fixed embryos were treated at 23°C for 3 min with PBT ( 130 mM NaCl , 7 mM Na2HPO4 , 3mM NaH2PO4 , 0 . 1% Tween-20 ) containing 50 μg/ml Proteinase K , and the reaction was immediately stopped by treating twice for 30 sec each with PBT containing 2 mg/ml glycine . Hybridization was performed for 16 hr at 60°C in hybridization solution ( 50% formamide , 5 x SSC , 0 . 1% Tween 20 , 0 . 05 mg/ml heparin , 0 . 1 mg/ml yeast tRNA ) containing 0 . 6 μg/ml impα2 RNA probe . Post-hybridization washing was performed six times ( 30 min each ) at 60°C in a solution containing 50% formamide , 5× SSC , and 0 . 1% Tween-20 . Embryos were incubated for 30 min with Fab fragments of anti-DIG antibody conjugated with HRP ( 600 U/l , Boehringer-Mannheim ) , then the signal was enhanced using the TSA-Biotin System ( PerkinElmer Life Sciences , Inc . ) and Streptavidin-Cy3 conjugate ( 1:2000 dilution , Jackson ImmunoResearch ) . For staining with ftz , eve or Sxl probe , the fixed embryos were treated at room temperature for 15 min with PBT containing 7 μg/ml Proteinase K , and then the reaction was stopped as described above . Hybridization was performed for 16 hr at 56°C in hybridization solution containing 0 . 5 μg/ml of ftz , eve or Sxl RNA probe . The embryos were washed five times ( 30 min each ) at 56°C in hybridization solution , and then rinsed in PBT containing 75% , 50% , and 25% hybridization solution for 5 min each , and in PBT five times for 5 min each . The embryos were incubated with HRP-conjugated anti-DIG antibody ( 300 U/l ) for 16 hr at 4°C , and the signal was enhanced using TSA-Plus Fluorescein System ( PerkinElmer Life Sciences , Inc . ) . Embryos were mounted in Vectashield ( Vector Laboratories ) or ProLong Diamond ( Molecular Probes ) . Z-stack confocal images were taken from each embryo using a Zeiss LSM 5 Pascal ( Zeiss ) , Leica TCS-NT ( Leica ) , or Leica TCS-SP8 ( Leica ) confocal microscope . Optical slices were analyzed using Zeiss LSM 5 Image Browser ( Zeiss ) , or Fiji software . In Figs 4G , 5H , S3 and S4 , the numbers of signal-positive pole cells located from the top to median plane of embryos were counted in confocal serial images . Embryos from late stage 4 to stage 6 were stained with anti-Impα2 23aa antibody . Serial optical sections ( 1 . 3 μm thick , 4–5 sections per pole cell ) were obtained using a confocal microscope ( LSM 510 Meta , Zeiss ) . Embryos from late stage 4 to stage 5 were stained with anti-Impα3 antibody , and serial optical sections ( 1 . 0 μm thick , 6–8 sections per pole cell ) were obtained using a TCS-SP8 confocal microscope ( Leica ) . Fluorescence intensities from the area occupied by individual pole cells ( judged by the outline of the cell in the DIC image ) were determined in sections through the median plane of pole cells . Fluorescence intensities were measured in all pole cells located within 15 μm of the top section of confocal serial images . Average fluorescence intensities ( intensity/pixel ) were calculated . Embryos from late stage 4 to stage 5 were double-stained with anti-Ftz-F1 antibody and propidium iodide or DAPI as described above . Under a confocal microscope ( Zeiss LSM 510 Meta , Zeiss ) , serial optical sections ( 1 . 3 μm thick , 4–5 sections per pole cell ) were obtained . We examined all pole cells located within 18 . 2 μm of the top section of confocal serial images . To quantify Ftz-F1 distribution in the nucleus of a single pole cell , fluorescence intensities from the area occupied by the nucleus were determined for each section using ImageJ , and then summed . The nuclear area was judged as the propidium iodide- or DAPI-positive area . To quantify Ftz-F1 distribution in the cytosol of pole cells , we measured fluorescence intensity from the whole area of a pole cell ( judged by the outline of the cell in the DIC image ) , and then fluorescence intensity of the cytosolic area was calculated by subtracting the nuclear intensity from the whole-cell intensity . Average fluorescence intensities ( intensity/pixel ) were calculated for both nuclear and cytoplasmic areas , and the ratio of nuclear to cytoplasmic intensity was calculated . Template DNA was amplified from impa2 cDNA clone K9 by PCR using forward primer 5´-GCGCGAATTAACCCTCACTAAAGGGCTCCCGAACAGATCGTCG-3´ ( ntd 1483–1500 of GenBank accession no . BT003258 ) and reverse primer 5´- GCGCGAATTAACCCTCACTAAAGGGAATCATTCAAACAATTCATTATTTATTGACAACTTTG-3´ ( complementary to ntd 2447–2483 of GenBank accession no . BT003258 ) , both of which contain the promoter sequences for T3 RNA polymerase ( shown in bold ) at their 5´-ends . dsRNA was transcribed in vitro from the amplified DNA with T3 RNA polymerase ( MEGAscript T3 kit , Ambion ) . dsRNA ( 0 . 1 nl of a 1 . 7 μg/μl solution ) was injected into the posterior pole of pgc nos embryos at early stage 2 . Because knockout of maternal impα2 mRNA results in developmental arrest at early cleavage stage [82] , we performed partial knockdown of impα2 mRNA by precisely regulating the injection volume using a thin glass needle ( hole diameter = 3 μm ) . Injected embryos were fixed in a 1:1 mixture of heptane and fixative II for 20 min , and the vitelline membrane was removed in PBS using a tungsten needle . Fixed embryos were processed for in situ hybridization with an antisense ftz RNA probe , as described above . The pole cells located within 30 μm of median section of confocal serial images were counted . Recombinant Nos and Pum proteins were expressed in KRX E . coli cells ( Promega ) as described previously [22] using the Nos expression plasmid pFN18K NosZC ( aa 289–401 ) ( a gift from A . C . Goldstrohm ) and the Pum expression plasmid pFN18K Pum RNA-binding domain ( aa 1091–1426 ) ( a gift from A . C . Goldstrohm ) . For Nos expression , cells were cultured in 2×YT medium with 25 μg/ml kanamycin and 2 mM MgSO4 at 37°C to an OD600 of 0 . 7–0 . 9 , and then protein expression was induced with 0 . 1% ( w/v ) rhamnose for 3 hr . For Pum expression , cells were cultured at 37°C in the same medium to an OD600 of 0 . 6 , and then at 16°C to an OD600 of 0 . 7–0 . 9 . Protein expression was induced with 0 . 1% rhamnose for 14–16 hr at 16°C . Nos and Pum proteins were purified essentially as described by Weidmann et al . [22] , with the following modifications . Nos and Pum proteins with Halo tag were purified by incubating with Magne HaloTag beads ( Promega ) overnight at 4°C . Beads were washed three times with Wash Buffer ( 50 mM Tris-HCl pH 8 . 0 , 2 mM MgCl2 , 1 M NaCl , 1 mM DTT , 0 . 5% [v/v] NP-40 ) , and three times with Elution Buffer ( 50 mM Tris-HCl pH 7 . 6 , 150 mM NaCl , 1 mM DTT , 20% [v/v] glycerol ) . Then , the beads were resuspended in Elution Buffer containing AcTEV protease ( Invitrogen ) and incubated for 24 hr at 4°C to cleave Nos or Pum protein from the Magne HaloTag beads . The beads were then removed using a MagneSphere magnetic separation stand ( Promega ) . Synthetic Cy5-labeled impα2 RNA fragment ( IDT , Tokyo ) , shown in S2 Fig , were used in EMSA . RNA-binding reactions were performed in RNA-binding buffer ( 50 mM Tris-HCl pH 7 . 6 , 150 mM NaCl , 2mM DTT , 2 μg/ml BSA , 0 . 01% [v/v] NP-40 , 20% [v/v] glycerol ) . Target RNA ( 100 nM ) , purified Pum ( 1 . 2 μM ) , and Nos ( 1 . 2 μM ) were incubated in RNA binding buffer for 3 hr at 4°C . Native polyacrylamide TBE mini-PROTEAN gel ( 5% , Bio-Rad ) was pre-run for 2 . 5 hr at 50 V , and then 10 μl of each sample was loaded and the gel was run at 50 V for 2 hr 10 min at 4°C . A Typhoon FLA 7000 laser scanner ( GE Healthcare ) was used to image EMSA . | Identification of the molecular mechanism underlying germline segregation from the soma is a fundamental goal of reproductive , cellular , and developmental biology . In many animal species , repression of somatic gene expression in germline progenitors is critical for the germ/soma segregation . In Drosophila , germ plasm , a specialized ooplasm partitioned into germline progenitors , contains maternal factors sufficient to repress somatic differentiation . Here , we show that a subset of somatic genes is derepressed when two maternal factors , Nanos ( Nos ) and Polar granule component ( Pgc ) are concomitantly suppressed . While Pgc is known to suppress RNA polymerase II ( Pol II ) activity , how Nos achieves this effect remains obscure . We find that Nos represses production of Importin-α2 that is essential for nuclear import of transcriptional activators for somatic gene expression in germline progenitors . Thus , we propose that Nos-dependent inhibition of nuclear import of transcriptional activators and Pgc-dependent suppression of Pol II activity acts as a ‘double-lock’ mechanism to ensure tight inhibition of somatic gene expression in germline progenitors . Since Nos is evolutionarily conserved , and a transient suppression of Pol II is a trait of germline progenitors of diverse animal species , the ‘double-lock’ mechanism may play a widespread role in germ/soma segregation . | [
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Materials",
"and",
"methods"
] | [
"nuclear",
"import",
"3'",
"utr",
"messenger",
"rna",
"cell",
"processes",
"dna",
"transcription",
"developmental",
"biology",
"protein",
"expression",
"untranslated",
"regions",
"sequence",
"motif",
"analysis",
"molecular",
"biology",
"techniques",
"embryos",
"research... | 2019 | Maternal Nanos inhibits Importin-α2/Pendulin-dependent nuclear import to prevent somatic gene expression in the Drosophila germline |
Rare variation in protein coding sequence is poorly captured by GWAS arrays and has been hypothesized to contribute to disease heritability . Using the Illumina HumanExome SNP array , we successfully genotyped 191 , 032 common and rare non-synonymous , splice site , or nonsense variants in a multiethnic sample of 2 , 984 breast cancer cases , 4 , 376 prostate cancer cases , and 7 , 545 controls . In breast cancer , the strongest associations included either SNPs in or gene burden scores for genes LDLRAD1 , SLC19A1 , FGFBP3 , CASP5 , MMAB , SLC16A6 , and INS-IGF2 . In prostate cancer , one of the most associated SNPs was in the gene GPRC6A ( rs2274911 , Pro91Ser , OR = 0 . 88 , P = 1 . 3×10−5 ) near to a known risk locus for prostate cancer; other suggestive associations were noted in genes such as F13A1 , ANXA4 , MANSC1 , and GP6 . For both breast and prostate cancer , several of the most significant associations involving SNPs or gene burden scores ( sum of minor alleles ) were noted in genes previously reported to be associated with a cancer-related phenotype . However , only one of the associations ( rs145889899 in LDLRAD1 , p = 2 . 5×10−7 only seen in African Americans ) for overall breast or prostate cancer risk was statistically significant after correcting for multiple comparisons . In addition to breast and prostate cancer , other cancer-related traits were examined ( body mass index , PSA level , and alcohol drinking ) with a number of known and potentially novel associations described . In general , these findings do not support there being many protein coding variants of moderate to high risk for breast and prostate cancer with odds ratios over a range that is probably required for protein coding variation to play a truly outstanding role in risk heritability . Very large sample sizes will be required to better define the role of rare and less penetrant coding variation in prostate and breast cancer disease genetics .
For most common diseases and traits the genetic basis underlying susceptibility has yet to be completely revealed . While genome-wide association studies ( GWAS ) have been remarkably successful in identifying common genetic variants associated with risk , the effect sizes of the risk alleles have been modest ( relative risk , RR of 1 . 1–1 . 4 ) and in most cases , even in sum , they can explain only a fraction of familial risk or disease heritability . GWAS have relied almost exclusively on Illumina and Affymetrix SNP arrays , with SNP content selected primarily from HapMap to capture a large fraction of common variation in coding and non-coding regions in populations of European ancestry . The vast majority of alleles with frequencies <5% , and in particularly those with frequencies ≤1% , have not been tested . This low allele frequency spectrum of genetic variation represents a very large fraction of all variation in the human genome . Thus , to date , a large fraction of genetic variation has yet to be explored with respect to disease etiology . It is possible that the majority of less common ( 1–5% ) and rare variants ( <1% ) will have weak effects , like the GWAS-identified common variants , and if this is the case then very large studies will be required for their discovery . An alternative hypothesis is that less common and rare variants convey larger relative risks than common variants , and indeed this assumption is required in order that rare variants contribute meaningfully to the understanding of inherited susceptibility . Such enhancement of effect sizes for rarer alleles may be especially relevant to rare coding variants given their dominant role in the etiology of “Mendelian” disorders ( e . g . the OMIM database [1] ) . Support for the hypothesis that rare coding variation also profoundly affects risk of certain “complex” diseases is growing and there are now a number of such examples including rare missense variants in CHEK2 , ATM , NBS1 , RAD50 , BRIP1 , and PALB2 in breast cancer [2] , rare coding mutations in RAD51D and BRIP1 in ovarian cancer [3] , [4] , as well as rare coding variants in genes implicated in hyperglyceridemina [5] and colorectal cancer adenomas [6] . More recently , whole-genome and candidate gene sequencing studies have revealed rare coding variants in ALDH16A1 for gout [7] and a number of genes ( NOD2 , IL23R , CARD9 , IL18RAP , CUL2 , C1orf106 , PTPN22 and MEC19 ) involved in inflammatory bowel disease [8] . Studies in prostate cancer have reported rare gene coding mutations in BRCA2 ( found in 2% of cases <55 years ) to be associated with greater risk of prostate cancer ( RR>4 . 5 ) and more aggressive disease [9] , [10] . For many of these examples , in addition to single SNP association testing , burden of rare variation analyses have been applied to increase the number of observations in the comparison groups ( and thus the statistical power ) , and to provide statistical support for the involvement of the gene which is not achieved when examining large number of SNPs in any given gene . To date , a lack of technology to survey the genome and accurately enumerate and test the variants in large numbers of samples has limited the exploration of less common and rare alleles . In the past year the Illumina Infinium HumanExome array ( or “exome chip” ) has been developed in collaboration with investigators who combined whole-exome sequencing conducted in >12 , 000 individuals of primarily European ancestry as well as in small numbers of other racial/ethnic minorities including African Americans , Hispanics , and Asians; the content on the array includes >200 , 000 putative functional exonic variants and is aimed to provide comprehensive testing on all non-synonymous variants above 0 . 1% frequency in Europeans . In the present study , we have utilized this array to test the hypothesis that there are less common and rare functional variants in the coding regions of genes that convey risk for breast and prostate cancer of greater magnitude than the common variants revealed through GWAS . We tested both single markers as well as gene summaries of the burden of rare alleles in multiethnic studies of invasive incident breast cancer and prostate cancer in the Multiethnic Cohort study ( MEC: 3 , 141 breast cancer cases , 4 , 675 incident prostate cancer cases and 8 , 021 controls ) . In addition we conducted exploratory analyses of rare variants in relationship with several breast and prostate cancer-related traits ascertained at baseline in the entire MEC sample ( n = 15 , 837 ) .
In the ethnic-pooled breast cancer analyses ( 2 , 984 cases and 7 , 545 controls ) , the most significant predicted protein-altering variant was a rare SP variant rs145889899 at the splice donor site in the second intron of the gene LDLRAD1 ( OR = 3 . 74 , p = 2 . 5×10−7 ) , which was almost exclusively seen in African Americans , this variant was statistically significant at our exome-wide level ( nominal p<3 . 9×10−7 , see Methods ) . Of the top 10 ranked associations , the remaining 9 involved NS variants ( p-values ≥1 . 3×10−6 , Table 2 , Table S1 ) . None of the other associations met the Bonferroni adjustment for multiple comparison testing . All of the 10 most associated variants , were quite rare and present mainly or exclusively in one or two ethnic groups . The genes containing the most significant SNPs for breast cancer ranged widely in apparent function ( see Table 2 ) with GWAS associations reported with SNPs in CFB ( complement factor B ) for age-related macular degeneration [11] , BAZ2A for platelet counts [12] and ACADS for metabolic traits [13] . Table S1 gives information for the 100 most significant associations for breast cancer , both overall and by ethnic group when including all SNPs passing quality control ( not just the non-synonymous , splice site and nonsense variants described here ) . For ER- breast cancer ( n = 441 cases ) many associations ( 358 ) with very rare SNPs were nominally significant using the score test but the p-values failed to stand up to further investigation using exact logistic regression ( the exact p-values ranged from 3×10−5 to 0 . 21 ) . The many small p-values apparently reflected overly liberal behavior of the score test when alleles are rare and when there are many more cases than controls . In order to reduce discussion of a large number of likely false positive tests we consider in the subtype analyses only SNPs with at least 10 minor alleles seen over all cases and controls . With this restriction we found a total of ten globally significant SNPs ( using the score test ) . However , p-values from exact logistic regression for these SNPs were again far less striking ( ranging from 3×10−5 to 1 . 5×10−3 ) . When restricted to estrogen receptor-positive ( ER+ ) cases ( n = 1 , 688 ) ( and screening out SNPs with less than 10 minor alleles seen ) the most significant coding SNP was a rare NS variant in UMODL1 ( exm1573155 , Ala542Thr , OR = 7 . 28 , p = 9 . 8×10−7 ) ( Table 2 , Table S2 ) . This SNP had a frequency of just over 0 . 2% in African Americans controls and 0 in the other groups . No associations are reported for this gene in the GWAS catalog . Neither this SNP nor any others were significant after correction for multiple testing . In ethnic-specific analyses of overall breast cancer only one additional SNP ( in FANCI ) met our criteria ( p<3 . 9×10−7 ) of global significance . This NS variant ( rs62020347 , Pro55Leu ) was common in European Americans , African Americans , and Latinos ( 3–8% frequency ) but was only associated with risk among European Americans ( MAF 8% , OR = 0 . 47 , p = 1 . 8×10−7 ) and was weakly associated with risk overall ( p = 0 . 02 ) ( Table S1 ) . Table 3 summarizes the most significant findings from the gene burden ( sum of coding variants ) analysis based on all common and rare ( ≤1% ) functional SNPs in each gene . Further details are given in Table S5 . For overall breast cancer no gene burden sum passed the Bonferroni criteria ( 3×10−6 ) for global significance for testing approximately 17 , 200 genes ( see Methods ) . The strongest associations were seen for MMAB ( p = 5 . 0×10−5 ) , SLC16A6 ( p = 5 . 0×10−5 ) and INS-IGF2 ( p = 1 . 2×10−4 ) . The MMAB gene is close to non-exonic SNPs that have been associated with HDL cholesterol [14] and one of those GWAS SNPs ( the intronic variant rs7134594 ) was among our top 100 single SNP associations with breast cancer ( Table S1 ) . INS-IGF2 contains an intronic SNP that has been associated with type 1 diabetes [15] . Restricting the gene burden analysis to only SNPs with overall frequency ≤1% gave non-significant associations as well ( p>8×10−6 ) and none of the top five genes in these analyses have globally significant GWAS associations reported . For ER+ breast cancer , the burden of rare SNPs in gene FGFBP3 was nominally globally associated ( p = 6×10−7 ) although follow-up using exact logistic regression gave a larger p-value ( 1 . 0×10−4 ) . This gene included five rare SNPs and no reports of any GWAS associations for SNPs near this gene are found in the GWAS catalog . When examining ER- breast cancer , the burden of variants in MMAB remained one of the strongest associations ( p = 2 . 0×10−5 ) . The burden of coding SNPs ( all of which were rare ) in EGR2 was the leading association in the ER- analysis with a p-value from the score test of 1 . 2×10−11 . A variant upstream of EGR2 has been associated in a GWAS of Ewings sarcoma [16] . Rare variant burdens also met our criteria for global significance for CNR1 ( p = 1 . 7×10−10 ) , FKSG83 ( p = 1 . 5×10−8 ) , GATM ( p = 4 . 8×10−7 ) , and ACSBG1 ( p = 5 . 3×10−7 ) . Again as for the single SNP results for ER- disease , these p-values were found to be overly liberal compared to an exact test ( the smallest exact logistic regression p-value was 2 . 8×10−5 for ACSBG1 ) For overall prostate cancer ( 4 , 376 cases and 7 , 545 controls ) none of the single SNP associations with prostate cancer met the Bonferroni adjustment for multiple comparison testing ( nominal p<3 . 9×10−7 ) . The top two associations found for prostate cancer were for rare NS variants in F13A1 ( rs140712764 , Val170Ile , OR = 28 . 0 , p = 9 . 1×10−7 ) and ANXA4 ( rs146778617 , Val315Phe , OR = 4 . 52 , p = 6 . 0×10−6 ) , Table 4 , see also Table S2 . Gene F13A1 is a coagulant factor gene not obviously related to prostate cancer etiology . ANXA4 encodes a protein that has been discussed as a possible marker for gastric cancer [17] . Of note , the third most significant association was for a common NS variant in GPRC6A ( rs2274911 , Pro91Ser , OR = 0 . 88 , P = 1 . 3×10−5 ) . This gene is nearby to RFX6 , which harbors an intronic variant ( rs339331 ) that has been reported in a GWAS of prostate cancer in Japanese men [18] . The SNP rs2274911 is common in all populations ( MAFs of 24–43% ) ( Table 4 ) and the protective effect of the minor allele was generally consistent in each group ( OR = 0 . 78 to 0 . 95 , over the five groups ) . This NS variant is correlated with the known intronic variant ( rs339331 , which is included on the Illumina HumanExome array ) in all populations ( r2 between 0 . 74 and 0 . 98 ) ; in conditional analyses neither of these two SNPs remained significant after the other was forced into the model ( P>0 . 2 ) ; thus these two variants are probably capturing the same signal , with the NS SNP in GPRC6A a potentially plausible susceptibility variant . The top 10 ranked associations ( Table 4 ) were all NS variants and 4 were common with a MAF>10% in all ethnic groups . When restricted to advanced cases ( n = 499 ) , similarly as for ER- breast cancer , many associations with very rare SNPs were nominally significant using the score test ( 69 total for SNPs with less than 10 minor alleles observed ) but the p-values failed to stand up to further investigation using exact logistic regression ( with p-values all <3×10−5 ) . In order to reduce discussion of a large number of likely false positive tests we considered in subtype ( advanced/nonadvanced ) analyses only SNPs with at least 10 minor alleles seen over all cases and controls used in the analysis . Of the remaining SNPs we found that four NS SNPs with at least 10 minor alleles present were nominally significant using the score test criteria ( Table 4 , Table S4 ) . These included NS variants in KLHL30 ( exm280349 , Arg108His , OR = 13 . 9 , p = 1 . 7×10−9 ) , PPP1R15A ( rs45533432 , Arg65Gly: OR = 4 . 67 , p = 1 . 2×10−8 ) , MUC12 ( rs143984295 , Ala101Thr , OR = 14 . 4 , p = 1 . 5×10−8 ) and RP1 ( rs114797722 , Ala1326Pro , OR = 13 . 4 p = 2×10−8 ) . These SNPs were all quite rare in the four largest populations ( 0 . 1%–1% ) . P-values from exact logistic regression for these SNPs were again less significant with p-values between 1 . 4×10−6 and 4 . 6×10−4 ) . For non-advanced disease ( n = 3 , 666 cases ) , the strongest associations were with the same SNPs as overall prostate cancer ( rs140712764 in F13A1 , rs146778617 in ANXA4 , rs2274911 in GPRC6A ) and also with rs61746620 in ZKSCAN2 ( Ala574Val , OR = 13 . 4 , p = 1 . 3×10−5 ) , although none of these were significant at our Bonferroni criteria . None of the gene burden analyses were significant for overall prostate cancer after correcting for multiple comparisons ( p<3×10−6 ) either when including common coding variants or when restricting the results to SNPs with frequency ≤1% ( Table 5 , Table S5 ) . When the analysis was restricted to advanced prostate cancer , four gene burdens ( for SAMD1 , FOXF2 , NOL4 and CPA3 ) were significant using the score test but not by exact logistic regression ( p = 2 . 5×10−3 , 3 . 3×10−3 , 5 . 0×10−3 and 3 . 4×10−6 respectively ) . No notable findings were observed when only localized prostate cancer was assessed . We also examined additional cancer-related traits: body mass index ( BMI ) , alcohol intake , as well as circulating PSA levels ( Table S8 ) . A number of NS variants have already been strongly associated with many of these traits , such as rs671 ( Glu504Lys ) in ALDH2 with alcohol intake [23] , rs17632542 [Ile179Thr] in KLK3 and circulating PSA levels [24] , [25] and rs198977 [Arg250Trp] in KLK2 and the ratio of free to total PSA [26] . For each trait , the 10 most associated variants on the array ( including non-functional SNPs , i . e . GWAS SNPs ) are provided in Table S9 . We also observed a number of suggestive associations at p<3 . 9×10−7 with rare coding variants in some genes that are biologically plausible for each trait . Three variants were strongly associated with blood PSA levels ( chr19: Hg19 position: 4552446 , Thr326Met , SEMA6B , 0 . 1% MAF in African Americans and monomorphic in all of the other populations , beta = 3 . 8 , p = 3 . 8×10−9; rs17632542 , Ile136Thr , beta = −0 . 4588 , p = 1 . 0×10−8 MAF . 06 in European Americans; rs148595483 , Asn322Lys , CCDC78 , 0–0 . 1% MAF across populations , beta = −2 . 9 , p = 2 . 4×10−8 ) . We also found a number of significant associations with very rare NS variants that were observed in 2–7 individuals and BMI ( rs146199292 , Asn31Lys , OSBPL11 , beta = 19 . 9 p = 1 . 2×10−10; rs149954327 , Leu458Val , STON1-GTF2A1L , beta = 15 . 2 p = 1 . 5×10−9; rs146922831 , Lys608Asn , LRGUK , beta = 9 . 2 , p = 3 . 0×10−8 ) . The variants were very rare in African Americans with frequencies <0 . 09% and monomorphic in all of the other populations except for rs146199292 in Latinos ( 0 . 02% ) . Variations in these genes have been reported in association with conditions related with BMI , including cardiovascular risk factors , type 2 diabetes and polycystic ovarian syndrome [27] , [28] , [29] . The carriers of these rare alleles were clustered at the extreme high end of the BMI distribution . All these potentially novel associations will need further follow-up . This paper presents an initial investigation of the role of coding variation in the genetics of breast and prostate cancer . Our initial analysis fails to find strong evidence for the hypothesis that relatively rare coding variation is highly determinative of breast or prostate cancer risk either overall or by subtype . Our sample sizes in each racial/ethnic group were each relatively small ( roughly 1 , 000 cases and 2 , 000 controls in the largest groups ) however these sample sizes are large enough to detect risk alleles with moderate to large effects ( odds ratios of 3–13 ) appearing in quite low frequency ( 0 . 1–1% ) and to examine whether such coding variation underlie ( by so-called synthetic association [30] ) many GWAS associations . While caution is advised in interpreting our results , especially for other than European racial/ethnic groups ( since the array utilized was predominantly based upon sequence information for Europeans and is not expected to cover other groups equally well ) , it appears that future studies to understand the relationship between rare coding variation and breast and prostate cancer risk will likely require the very large sample sizes needed to target much less penetrant alleles . Our analyses consisted of both single variant analysis and simple gene burden analyses . The gene burden analyses consisted of summing the minor alleles of coding variants including either all coding variants regardless of their frequency , or only those variants with MAF <1% in our overall sample . While this gene-burden test assumes implicitly that all coding variants have the same direction of effect , this is reasonable given that the power of detecting rare protective alleles in a case-control study such as this one ( where controls can be regarded as representative of the population ) is much less than the power to detect rare risk alleles . The rare variant sum therefore is not very sensitive to the presence of rare protective alleles in a gene . One association for breast cancer , a single SNP in LDLRAD1 , appeared to pass our established level of global significance ( p<3 . 9×10−7 ) when all cases were examined . No associations ( either single SNPs or gene burdens ) were globally significant for overall prostate cancer . Subset analyses , by ER status for breast cancer or advanced/non-advanced for prostate cancer generally failed to show believable associations . While the score test gave many “globally significant” associations these apparently reflected excess type I error of this test when both the number of cases is small compared to the number of controls and when the SNPs were rare . This breakdown in reliability is similar to that seen for the uncorrected Pearson chi-square test ( a special case of the score test when no covariates are present ) , which is well-known to have poor control of type I error when the expected number of cases is very small for a cell . Following-up such associations with exact logistic regression implemented in SAS ( Cary , NC ) provided larger p-values not globally significant using our criteria . Nevertheless a number of suggestive findings were observed that are worthy of further attempts at replication: The splice site variant rs145889899 in LDLRAD1 ( our top finding for overall breast cancer ) is found in low frequency ( <1% ) in African American controls ( higher of course in cases since this is nominally a risk variant ) , and only seen among cases in the other groups . No associations with any disease or phenotype have to date been reported for this gene . Among the other genes highlighted in Table 2 or Table 3 , associations have been reported for SNPs in SLC19A1 and CASP5 for renal cancer [31] , [32]; BAZ2A has been reported to be up-regulated in CLL patients [33] . Also notable is a strong link between SNPs in EGR2 ( ER- association ) and risk of Ewing's sarcoma [16] . For prostate cancer ( all cases ) the third strongest association result was for a common NS coding variant ( rs2274911 ) in GPRC6A that is in very high LD with the known intronic GWAS variant rs339331 . In our data the NS variant was slightly more associated ( Table S3 ) with prostate cancer risk ( p = 1 . 3×10−5 ) than was rs339331 ( p = 2 . 1×10−5 ) . The coding SNP is arguably a more likely causal variant than the intronic SNP since expression of GRPC6A is substantially increased in prostate cancer cell lines , and mice deficient in GRPC6A show retarded prostate cancer progression [34] . In addition , GRPC6A deficiency in mice also attenuates the rapid signaling responses to testosterone , an androgen that is critical for initiation and progression of prostate cancer [35] . Other suggestive findings for prostate cancer include SNPs in a variety of genes such as F13A1 expression of which has been associated with bone metastasis in prostate cancer [36] , ANXA4 which is up-regulated in gastric and other cancers [17] , NSD1 where cryptic translocations may be involved in AML occurrence [37] and MUC12 , expression of which has been reported to be a prognostic marker in colon cancer [38] . The burden of rare SNPs in FGFBP6 ( one of the stronger association seen for breast cancer ) was also among the top associations for overall prostate cancer ( Table 5 , p = 1 . 5×10−4 ) . We evaluated also associations in regions surrounding known ( GWAS ) risk alleles as a partial fine-mapping exercise; we specifically focused upon ( 1 ) coding alleles reported to be in high LD ( in Europeans using 1000 Genomes data ) with the index marker , and ( 2 ) other ( generally less common ) coding alleles within 500 kb of the GWAS alleles , that might show associations that could underlie ( by synthetic association [30] ) GWAS associations . A number of GWAS risk alleles are in reasonable LD ( r2>0 . 3 ) with coding SNPs on the array and several of the latter show nominal associations ( p<0 . 05 ) with breast cancer risk including SNPs in STXBP4 , ZNF45 , and ZNF404 which are all worth evaluating as candidate loci potentially explaining the index GWAS associations . For prostate cancer , a similar observation is made most notably for GPRC6A but also for MLPH ( GWAS index = rs7584330 , chromosome 2 , p = 0 . 003 ) , PDLIM5 ( rs12500426 , chromosome 4 , p = 0 . 019 ) , RNMTL1 ( rs684232 , chromosome 17 , p = 0 . 024 ) , KLK3 ( rs2735839 , chromosome 19 , p = 0 . 0046 ) , and RTEL1 ( rs6062509 , chromosome 20 , p = 0 . 001 ) . Previous reports [24] , [39] have highlighted the NS SNP rs17632542 in KLK3 as highly associated with PSA level and a highly significant risk variant in fine-mapping of the locus near rs2735839 [39]; while no report for prostate cancer exists for coding SNPs in RTEL1 , another NS SNP , rs3208008 , in RTEL1 has been found to be associated with glioma risk [40] . Other coding SNPs that could include causal variants producing synthetic associations ( associations of rare with common SNPs of high penetrance ) include SNPs in genes INS-IGF2 , ZFYVE26 , C16orf46 , UNC13A , NRIP1 and CCDC91 for breast cancer and SNPs in SNED1 and PASK for prostate cancer . These do not have high r2 with the GWAS variants as they are mostly rare ( and are >100 kb away from the index signal ) but their nominally strong associations ( p-values<1×10−3 ) might possibly be indicative of signals extending for many thousands of base pairs , although it will take much larger studies to verify or refute this . We found little evidence that the NS , SP , or nonsense variants captured by the HumanExome SNP array that fall within known or suspected high risk genes for breast or prostate cancer are meaningfully associated with either cancer . The Illumina array does not directly interrogate the rare , high-risk mutations , such as frameshift mutations in BRCA1 or BRCA2 ( e . g . c . 68_69delAG ) [41] , as very few indels are included on this array ( just 136 were examined here ) . The inability to address frameshift mutations either within known risk genes or more widely is a limitation of this report . Other limitations include the focus on Europeans in the development of the array ( as seems to be particularly reflected in the relatively small fraction of SNPs found to be polymorphic in Japanese Americans ) , and the loss of some targeted SNPs in the manufacturing process and in our QC procedures . In addition , this technology ( unlike exome sequencing ) cannot address the role of either private variation or of variants too rare to have been reliably identified during the discovery phase of the development of the array . Genotyping cases and controls from our prospective cohort allowed us an opportunity to examine other cancer-related phenotypes and traits for which data and specimens had been collected prior to breast or prostate cancer diagnosis . While two of these endpoints ( BMI , alcohol ) were based on self-report , we were able to strongly replicate a number of known associations such as rs671 in ALDH2 with alcohol intake which is proof of principle that the exome array has the potential to reveal biologically relevant coding variants . Apparently novel findings for PSA , BMI , and alcohol consumption will need to be replicated in large-scale exome association analyses; hopefully making the results from these preliminary analyses in a multiethnic population broadly available will contribute to novel discoveries and further understanding the genetic basis of these traits . In order for rare variants to play an important role in explaining missing heritability [42] even in composite they must have effects that are larger in magnitude than those observed for common SNPs . Roughly speaking , for a given allele the contribution to additive heritability ( under a liability model for example [43] ) is proportional to 2b2p ( 1-p ) where b is the log odds ratio ( OR ) and p is the frequency for that allele . Under simplifying assumptions ( such as limited selection and constant population sizes ) population genetics theory [44] indicates that there should be approximately as many variants “moderately rare” with frequency in the range 0 . 1 to 1% as there are the common variants in the range 5 to 50% that have been the targets of GWAS studies to date . However , in order that variants in the frequency range from 0 . 1 to 1% have the same composite effects on risk as do those in the frequency range from 5 to 50% then the magnitude of effect sizes must be considerably larger than for the common variants; if ORs for common variants lie in the range from 1 . 1 to 1 . 3 then ORs in the range from 2 to 6 are needed for the rare and common alleles to have similarly sized roles in disease susceptibility ( assuming that the same fraction of all rare alleles are risk variants as for common alleles ) . Moreover , under the hypothesis that the coding regions of the genome ( ∼1% of the total genome ) by themselves play an profound role in disease susceptibility these ORs would likely need to be skewed even higher – i . e . if rarer variation in 1% of the genome was to play as much a role as does common variation over the entire genome then the existence of ORs above 10 or even greater for such variation may arguably be a necessary consequence . Realistically our study only begins the assessment of whether a range of effects for “moderately rare” coding variants is possible: the detectable ORs in this study range from approximately 3 to 13 for alleles with frequency 1 to 0 . 1% , respectively . While these are large ORs the above argument indicates that such effect sizes are not unreasonable if rarer protein coding variation plays a similar role in the heritability of risk as does common variation genome-wide . Our failure to find such ORs for the rarer alleles may be providing evidence against coding variation having a predominant role in breast and prostate cancer heritability and risk ( outside of high risk families ) . In summary , the analyses and methods described here do not support NS variants on the current exome chip as conveying moderate to high risk for breast and prostate cancer . While some suggestive findings are noted it is likely that very large sample sizes of the order that can be only developed through collaborative efforts such as those now engaged in the NCI GAME-ON post-GWAS meta-analysis of common variants , will be required in order to further the understanding of the role of rare NS and other coding variation in disease genetics . Exome sequencing of high-risk families will continue to be important to reveal biologically relevant coding variants for these cancers , both for insertion/deletion variants that were not covered by the current array , and to capture rarer variation ( including private variants ) that cannot be captured except by sequencing .
This work has been performed according to relevant national and international guidelines . Written consent was obtained at the time of DNA sample collection . The Institutional Review Boards at the University of Southern California and University of Hawaii approved of the study protocol . The MEC consists of more than 215 , 000 men and women in California and Hawaii aged 45–75 at recruitment , and comprises mainly five self-reported racial/ethnic populations: African Americans , Japanese , Latinos , Native Hawaiians , and European Americans [45] . Between 1993 and 1996 , adults enrolled in the study by completing a 26-page mailed questionnaire asking detailed information about demographic factors , personal behaviors , and prior medical conditions . Potential participants were identified through driver's license files from Departments of Motor Vehicles , voter registration lists , and Health Care Financing Administration data files . Incident breast and prostate cancer , as well as stage and hormone receptor status was identified by linkage of the cohort to the Surveillance , Epidemiology , and End Results cancer registries covering Hawaii and California . Between 1995 and 2006 , blood specimens were collected prospectively from ∼67 , 000 participants for genetic and biomarker analyses . Currently , the breast cancer case-control study nested in the MEC includes 3 , 141 women diagnosed with invasive breast cancer and 3 , 721 frequency-matched controls without breast cancer , matched by race/ethnicity and age ( in 5-year age categories ) . The case-control study of prostate cancer includes 4 , 675 men diagnosed with incident prostate cancer and 4 , 300 male controls without prostate cancer . The Institutional Review Boards at the University of Southern California and University of Hawaii approved of the study protocol . Genotyping of the Illumina Human Exome BeadChip ( n = 247 , 895 SNPs ) was conducted at the USC Genomics Core Laboratory . DNA extraction of buffy coat fractions was conducted using the Qiagen protocol . Cases and controls were randomly placed across ethnic-specific plates for each cancer type . All samples had DNA concentrations >10 ng/ul . Initial genotype definitions were based on auto-clustering 6 , 404 samples across all populations which had call rate >0 . 99 ( African American 1883 , Japanese American 1823 , Latino 1008 , European American 1690 ) using the GenomeStudio software ( V2011 . 1 ) . Following genotype calling on all samples ( >16 , 000 ) , manual inspection was conducted of the following SNPs: 1 ) SNPs with call rate <0 . 98 ( n = 3 , 317 ) , 2 ) monomorphic SNPs with call rate <1 ( n = ∼15 , 000 ) , 3 ) SNPs with minor allele frequency between 0 and 0 . 001 and call rate <1 ( n = ∼31 , 500 ) , 4 ) SNPs with >1 replicate error based on sample duplicates ( ∼1 , 000 , discussed below ) , 5 ) SNPs with apparent differences in minor alleles frequencies >15% across ethnic-specific 96 sample plates ( n = 798 ) , or other evidence of batch/plate effects on allele frequency ( n = 18 , 188 ) , 6 ) all mitochondrial SNPs and all SNPs on the X and Y chromosomes ( n = 5 , 574 ) , and 7 ) autosomal SNPs out of Hardy-Weinberg Equilibrium in more than one ethnic group with p value<0 . 001 and at least one ethnic group with p value<0 . 00001 ( n = 827 ) . During the inspections we in total inspected cluster plots for approximately 70 , 000 SNPs ( counting overlapping SNPs in the categories above ) and genotypes were manually edited for 27 , 506 SNPs . Of the 15 , 837 samples described above genotyping was successful with call rates ≥98% for 15 , 573 samples; of these we removed 17 samples for which reported sex conflicted with assessment of X chromosome heterozygosity , and 651 samples based on relatedness . Relatedness was determined using the IBD calculation in plink [46] , and we removed one of each estimated MZ twin , sibling , parent-offspring , half sibling , or first cousin pairs . In the analysis , we also removed SNPs with <98% call rates ( n = 2 , 531 ) . To assess genotyping reproducibility we included 338 replicate samples which passed genotyping QC; among these samples the concordance rate of heterozygote calls , number concordant/ ( number concordant+number discordant ) , was 99 . 6% or greater for all replicate samples ( average 99 . 99% ) . The final analysis dataset included 245 , 339 SNPs genotyped on 2 , 984 breast cancer cases and 3 , 568 controls , and 4 , 376 prostate cancer cases and 3 , 977 controls . We relied on documentation files obtained from the University of Michigan posted on ftp://share . sph . umich . edu/exomeChip/IlluminaDesigns/ for the assessment of SNP type ( i . e . NS , SP ) , and the amino acid affected . The array also includes SNPs that do not code for protein changes including synonymous SNPs , and other intergenic SNPs including ancestry informative markers , and GWAS identified risk SNPs for a number of diseases and outcomes . All SNPs were analyzed and their results shown in Tables S1 , S2 , S3 , S4 , S5 , S6 , S7 , S8 , S9 . However our primary analysis focused on the 191 , 032 putative functional variants in the following categories ( NS , SP and stop gain or loss ) that passed quality control procedures discussed above . We estimated principal components in the entire sample using EIGENSTRAT [47] based on 2 , 887 autosomal ancestry informative markers on the array . We adjusted for the top 10 principal components in all analyses . Recognizing that many variants are only polymorphic in a few racial/ethnic groups , we give power analysis for a study with 1 , 000 cases and 2 , 000 controls ( roughly the number of cases and controls in each of the four largest populations ) by odds ratio ( 1–200 ) and allele frequencies ranging from 0 . 0001 to 0 . 1 ( Figure S2 ) . The Bonferroni criteria for significance in this study is calculated to be 0 . 05 divided by the largest number of polymorphic SNPs in any population ( African Americans , ∼125 , 000 ) or roughly 3 . 9×10−7 . For the gene burden analysis the Bonferroni criteria is 0 . 05 divided by the number of genes considered or roughly 3×10−6 . We had 80% power to detect odds ratios of 3 . 3 or above for SNPs with a frequency of 0 . 01 and odds ratios in the range 13 or above for SNPs of frequency 0 . 001 in single SNP analyses . Power for the gene burden analysis depends upon the number of polymorphic SNPs in a given gene . Using a Poisson approximation ( i . e . with variance assumed to be equal to the mean ) a gene with 10 variants each of frequency 0 . 001 gives power of 80% to detect a per minor allele OR of 3 . 1 . For genes with many more variants ( 100 ) of the same frequency detectable ORs per minor allele are 1 . 6 or greater . For common variants present in all ethnic groups we had much greater power to detect associations , for example we had 80% power to detect a 20% allele with an OR of 1 . 24 in the global analyses; for the region-specific analyses we have 80% power to detect a 20% allele with an OR of 1 . 17 in a region with 100 variants and 1 . 14 in a region with 10 variants . | For breast and prostate cancer , GWAS have revealed many risk variants ( >70 for each cancer as of this report ) . All together the common variants in these regions explain only a minority of familial risk of these cancers . Using the Illumina HumanExome SNP array , we explored the hypothesis of rare coding variation contributing to breast and prostate cancer risk in a sample of African American , Latino , Japanese , Native Hawaiian , and European American breast and prostate cancer cases and controls from the Multiethnic Cohort study . While only one association exceeded significance thresholds after correcting for multiple comparisons , a number of suggestive associations involving genes previously reported to be associated with a cancer-related phenotype were noted . Our results do not generally support a major role of protein-coding variants with odds ratios over a range that is probably required for protein coding variation to play a truly outstanding role in risk heritability . If very rare and/or less penetrant coding variants underlie disease heritability of these cancers , then very large sample sizes ( i . e . consortia ) will be required for their discovery . | [
"Abstract",
"Introduction",
"Results",
"Materials",
"and",
"Methods"
] | [
"genome-wide",
"association",
"studies",
"cancer",
"genetics",
"genetic",
"polymorphism",
"genetics",
"population",
"genetics",
"biology",
"genetics",
"and",
"genomics"
] | 2013 | Genome-Wide Testing of Putative Functional Exonic Variants in Relationship with Breast and Prostate Cancer Risk in a Multiethnic Population |
Chagas disease , caused by the flagellate parasite Trypanosoma cruzi affects 8–10 million people in Latin America . The mechanisms that underlie the development of complications of chronic Chagas disease , characterized primarily by pathology of the heart and digestive system , are not currently understood . To identify possible host genetic factors that may influence the clinical course of Chagas disease , Human Leucocyte Antigen ( HLA ) regional gene polymorphism was analyzed in patients presenting with differing clinical symptoms . Two hundred and twenty nine chronic Chagas disease patients in Santa Cruz , Bolivia , were examined by serological tests , electrocardiogram ( ECG ) , and Barium enema colon X-ray . 31 . 4% of the examinees showed ECG alterations , 15 . 7% megacolon and 58 . 1% showed neither of them . A further 62 seropositive megacolon patients who had undergone colonectomy due to acute abdomen were recruited . We analyzed their HLA genetic polymorphisms ( HLA-A , HLA-B , MICA , MICB , DRB1 and TNF-alpha promoter region ) mainly through Sequence based and LABType SSO typing test using LUMINEX Technology . The frequencies of HLA-DRB1*01 and HLA-B*14:02 were significantly lower in patients suffering from megacolon as well as in those with ECG alteration and/or megacolon compared with a group of patients with indeterminate symptoms . The DRB1*0102 , B*1402 and MICA*011 alleles were in strong Linkage Disequilibrium ( LD ) , and the HLA-DRB1*01-B*14-MICA*011haplotype was associated with resistance against chronic Chagas disease . This is the first report of HLA haplotype association with resistance to chronic Chagas disease .
Following an extensive control program consisting of vector control , serological screening in blood banks and identification and treatment of congenital transmission , the estimated number of people infected with Trypanosoma cruzi , the causal agent of Chagas disease in Latin America , has fallen from approximately 20 million in 1981 to around 8–10 million in 2009 [1] , [2] . Most of the seropositive patients are chronically infected and more than 10 , 000 deaths are estimated to occur annually from the disease [1] , [3] , [4] . Cardiac myositis and autonomous neuroplexus degeneration of the digestive tract are major histopathological alterations that can arise during Chagas disease , and may lead to cardiac failure , digestive abnormalities , megacolon or megaesophagus . Based on these pathologies , there are often considered to be three major clinical forms of Chagas disease; cardiac , digestive and indeterminate [5] , [6] . This variation in pathological manifestation has been reported to relate to differences in host immune response , such as the ability to control parasitaemia , the strength of inflammatory reactions , and the induction of autoimmune like responses [7]–[11] . Indeterminate phase T-cells have been reported to correspond with modulatory responses such as increased IL-10 production by CD4+CD28− T cells and the expression of CTLA-4 , a ligand that is involved in modulation of T-cell responses by CD8+ T-cells [12] , [13] . Additionally , the unregulated production of IFN-gamma by CD8+ T-cells in cardiac Chagas disease patients has been reported , which might result in the destruction of heart tissue due to its cytotoxic effect [12] . The highly polymorphic HLA Class I and II molecules determine the efficiency of T . cruzi epitope presentation to T lymphocytes that could affect the clinical course of Chagas disease [14] , [15] . Several HLA alleles and haplotypes have been reported to be associated with Cardiac Chagas disease in Chile , Venezuela , Brazil and Guatemala [16]–[18] . The HLA region contains not only classical HLA genes but a wide variety of immunologically relevant genes , such as nonclassical class I genes ( MICA , MICB ) [19] , [20] , and class III genes ( TNF-alpha , beta ) , that may be involved in pathogenesis [21] , [22] . In the present study , we investigated HLA class I ( A , B , MICA , MICB ) , Class II ( DRB1 ) and Class III ( TNF-alpha ) gene polymorphism in seropositive chronic Chagas patients in Bolivia , characterized by electrocardiogram ( ECG ) , barium enema colon X-ray examinations and/or surgical operation .
The study subjects were described previously [23] . Two hundred and ninety one patients with chronic Chagas disease ( 136 men and 155 women , mean age 45 years ) were recruited from the Centro Nacional de Enfermedades Tropicales ( CENETROP ) ( 91 men and 119 women ) , Hospital Primero de Mayo ( 12 men and 7 women ) and from post-operative patients at the Hospital Universitario Japonés ( HUJ ) ( 33 men and 29 women ) in Santa Cruz , Bolivia . Upon medical examination of patients other than the HUJ patients , if serological tests ( Indirect Haemaglutination assay ( IHA ) and Indirect Immunofluorescence test ( IIF ) [3] ) were positive , they were asked to participate in the study and signed informed consent was obtained . Informed consent was also obtained for the HUJ post-operative patients , using the same protocol . ECG abnormalities were diagnosed based on the Minnesota Code Criteria and were confirmed independently by two cardiologists . Colon X-ray with barium enema examination was performed for the detection of megacolon . To exclude the possibility of including individuals who were asymptomatic upon examination , but who may not have had adequate time post-infection for symptoms to become apparent , those under 30 years of age were excluded . Finally , 229 seropositive Chagas outpatients in Santa Cruz , Bolivia , were examined by ECG and/or barium enema colon X-ray as shown in Table 1 . The 62 post-operational patients from HUJ were confirmed to be suffering from Chagas megacolon during the admission period . The experimental protocol was approved by the Institutional Ethical Review Committee of the Institute of Tropical Medicine , Nagasaki University ( No . 0210170018 ) and the Centro Nacional de Enfermedades Tropicales ( CENETROP ) . Genomic DNA was extracted from 10 mL of whole blood containing 10 mM EDTA using a DNA extraction kit ( QIAGEN GmbH , FRG ) and was stored at −20°C until use . MICA , MICB , TNF-alpha promoter region and MICA-TM typing was performed as previously described [19]–[21] . Sequences were obtained from a 3730 DNA Analyzer ( Applied Biosystems , USA ) and submitted to the Assign- software ATF ( Conexio Genomics ATF , Australia ) for allele identification . The TNFP alleles were determined according to Higuchi T et al ( 1998 ) and Ubalee R et al ( 2001 ) [21] , [22] . MICA-TM alleles were typed by sequencing on a 3730 DNA analyzer ( Applied Biosstems , USA ) and analyzed with the GeneMapper Software Version 3 . 7 ( Applied Biosystems , USA ) . The fluorescent primers , 5′ F ( GCC CAG TGT ATA ACA AGT CCC-6-FAM ) and 3′ R ( CCT TAC CAT CTC CAG AAA CTG C ) were used . Typing was carried out according to the manufacturer's specification for LABType SSO Typing , testing for each locus using LUMINEX Technology ( ONE LAMBDA , INC , USA ) and the retrieved output was analyzed by HLA Fusion software ( ONE LAMBDA , INC , USA ) for allele identification . Statistical analysis was performed at the two and four digits levels . Allele frequencies less than 5% were removed from the analysis . The statistical significance and odds ratio ( OR ) of allele frequencies between each group was determined by Chi square and Fisher's exact tests using the StatsDirec software ( StatsDirect Ltd , UK ) . P-values were considered significant when <0 . 05 following Bonferoni correction ( Pc ) . Hardy-Weinberg Equilibrium , linkage disequilibrium ( LD ) and Haplotype analyses were calculated with PyPopWin32 . 0 . 7 . 0 software [24] .
In the two digits analysis ( Table S1 , S3 , S5 ) , we observed a significant decreased frequency of alleles DRB1*01 and B*14 in the megacolon patients compared with indeterminate patients ( Table 2 ) . The frequency of HLA-B*14 was also significantly lower in the patients with ECG alteration compared to the indeterminate patients . In the four digits analysis ( Table S2 , S4 , S6 ) , DRB1*01:01 , *01:02 and B*14:02 had the same tendency towards lower allele frequencies in the megacolon patients compared with the indeterminate patients . The frequency of the HLA-B*14:02 allele was significantly lower in the ECG alteration and/or megacolon patients compared with those of indeterminate symptoms ( Table 2 ) . The frequency of MICA*011 was also significantly lower in the complication positive groups ( Table 2 , S7 ) . None of the TNF-alpha promoter region and MICB ( Table S11 and S8 ) alleles tested here were significantly associated with the clinical groups compared with the indeterminate group . Linkage disequilibrium ( LD ) was calculated for all combinations of alleles ( Table S1 , Table S2 , Table S3 , Table S4 , Table S5 , Table S6 , Table S7 , Table S8 , Table S9 ) , and significantly strong LDs was observed between the alleles , HLA-DRB1*01 , A*33:01 , B*14:02 and MICA*011 as shown in Table 3 . Except for HLA-A*33:01 , all the linkage group alleles showed the same tendency of being associated with protection against megacolon ( Table 2 ) . HLA-A*01:01 showed strong LD with HLA-B*08:01 , DRB1*0301 and MICB*008 . Although HLA-A*01:01 showed a non-significant tendency of association with non-megacolon patients , none of the LD group alleles showed any similar tendency ( Table 2 ) . No LD was observed between any HLA alleles and TNF-alpha promoter haplotypes .
We have previously shown that there was no association between T . cruzi lineage or sub lineage and the clinical manifestation of Chagas disease in samples from Bolivia [23] . Here , we have analysed the same samples in order to determine if there were any associations between clinical disease and host genetic variation . We analysed polymorphic genes located in the MHC region , consisting of three sub regions; class I , II and III . Human MHC , HLA class I and II molecules play a crucial role in determining individual acquired immune responsiveness through the presentation of pathogen-derived peptides to CD8+ and CD4+ T-cells . In the class III sub region , there are a variety of genes related to immunity such as complement , TNF-alpha , Lymphotoxin , etc . [24] . Our study revealed that the HLA-DRB1*01-B*14-MICA*011 haplotype was significantly associated with protection from Chagasic megacolon . As the linkage disequilibrium between the HLA-B*14 and DRB1*01 was strong ( Table 3 ) , it was difficult to determine the primary associated locus within the haplotype . HLA-class I is the antigen-presenting molecule on the cell-surface of infected host cells which stimulates microbe-derived antigen specific CD8+ T-cells . Therefore , HLA-B*14 itself could be directly related to protective T-cell immunity as was suggested by the Chagas disease mouse model [25] . If HLA-B*14 is more efficient at stimulating protective T-cells through the binding of antigenic peptides , other HLA-class I molecules that share the same antigen binding motif should show the same protective association . HLA-B*14 consists of three 4-digit alleles , HLAB*14:01 , 14:02 and 14:06 that share the same antigen binding motifs but the latter two alleles are so rare that it was impossible to analyse their effect . As the HLA-B*14 alleles belong to the B27 supertype group that share the same anchoring residues in the peptide binding groove , we analyzed the member alleles B*38:01 , 39:04 , 39:05 . 39:06 , 39:14 , 48:01 , 48:03 for their total effect on protection against complication , and found no association ( Tables S6 and S10 ) [26] . This finding did not , therefore , support the hypothesis that the association between HLA-B*14 and protection against clinical Chagas disease is driven by the ability of the gene product to more effectively stimulate protective T-cells than other alleles of this gene . The peripheral human and mice CD8+T-cells reactive to Trypanosome antigens have been identified [27] . Interestingly , HLA-A*02:01 restricted epitopes from cruzipain and FL-160 were frequently recognized by PBMC of patients with Cardiopathy [28] . Moreover , an HLA-A2 tetramer experiment showed that the number of IFN-gamma producing amastigote-specific CD8+ T-cells inversely correlated with the severity of the disease [29] . It will be interesting to see if the same phenomenon occurred in the HLA-B*14 patients . Immuno-regulatory mechanisms have been reported to be associated with clinical forms including Treg cells [30] , NKT cells and NK cells [31] . Whereas the HLA involvement in the induction of Treg cells is not yet clear , HLA-class I can interact with NK cells to suppress their activity through various inhibitory receptors such as KIR . HLA-B*14 belongs to the Bw6 family , the members of which preferentially stimulate specific members of the KIR family [32] , [33] , which could regulate NK cell activity during inflammation . HLA-non classical class I , MICA*011 , which was closely linked to HLA-B*14 and DRB1*01 might also be functional as MICA is known to stimulate gamma-delta T-cells in the gut mucosa; a phenomenon that could relate to megacolon [34] . HLA-class II can present antigen to CD4+ T-cells so HLA-DRB1*01 may also be directly involved in the pathogenesis as well as HLA-B*14 . Many autoimmune diseases are reported to be associated with specific HLA-class II alleles [35] . Auto-reactive processes that involve the activation of cytokine producing T-cells may occur during infection . As was previously suggested , autoimmune mechanisms in the pathogenesis of chronic Chagas heart and colon may be regulated by the HLA-class II . We analyzed 4-digit HLA-DRB1 alleles for association with Chagas disease clinical manifestations . As shown in Table 2 , the HLA-DRB1*01 group included three alleles , DRB1*01:01 , 01:02 , 01:03 and all of them showed the same protective tendency when compared between the megacolon and indeterminate symptom groups . As two of them , 01:01 and 01:02 shared the same peptide-binding motif [36] , we considered that the HLA-DR molecule itself was functionally related to resistance to megacolon . It was previously reported that the DRB1*01 allele was associated with susceptibility to Chagas cardiomyopathy in Venezuela [37] . This opposite association to the megacolon resistance observed in the present work requires further clarification . Despite the strong LD shown within the HLA-DRB1*01-B*14 haplotype , there was no strong linkage between TNF-alpha promoter alleles that may influence the levels of its production by immune cells [21] . However , between HLA-DRB1 and HLA-B loci , 1270 kb of class III sub region containing more than 60 genes such as complements , heat shock proteins , 21-hydroxylase , are present that might be relevant to pathogenesis . The HLA-B*14:02-DRB1*01:02 haplotype was reported to be associated with V281L polymorphism in 21-hydroxylase in African-American and Caucasian populations [38] . The same kind of abnormality was also reported in the HLA-A*01:01-B*08-DRB1*03 haplotype that is associated with several diseases such as allergy and viral infectious diseases [39] . It was not significantly associated with Chagas . However , HLA-A*01:01 also showed a decreased frequency in the megacolon patients . Whole sequencing of the class III region of the associated haplotypes would be the next target for clarification of any genetic resistance . We have no information relating to the immunological characteristics of the individuals who poscessed those haplotypes that might be associated with lymphocyte activation during a chronic infection . We did , however , analyze the relationship between individuals' specific antibody titers and their HLA alleles ( data not shown ) but there was no clear correlation . About 7 . 1% of the seropositive indeterminate individuals were estimated to carry this haplotype; therefore further to identify its characteristic immunological function is feasible . To our knowledge , this is the first report of resistant HLA haplotype association with chronic Chagas diagnosed by the active examination of silent colon lesion . | Chronic Chagas disease consists of four different forms categorized on the basis of their clinical manifestations , namely; cardiac , digestive , cardiodigestive and indeterminate . In Latin America , there are 8–10 million seropositive persons who are at risk of , or have already developed serious clinical complications and who have limited access to effective treatment . The cardiac and digestive forms are characterized by tissue damage caused by persistent infection of Trypanosoma cruzi and are thought to be modulated by host immunity . In our large scale screening for chronic Chagas disease in Santa Cruz , Bolivia , hearts and colons of 229 seropositive patients were examined . We found 31 . 4% of patients had abnormal electrocardiograms ( ECGs ) , 15 . 7% presented with megacolon , 5 . 2% had a combination of abnormal ECG and megacolon , and 58 . 1% were of indeterminate status . Previously , we attempted to ascertain whether parasite genetic polymorphism might account for the differences in clinical manefestations , by analyzing parasite DNA taken from the same study group ( with the addition of a further 62 megacolon post-operational patients ) . We found no relationships between parasite lineages and clinical disease form . The present study reveals that host HLA polymorphisms associate with clinical manifestations of Chagas . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"clinical",
"immunology",
"neglected",
"tropical",
"diseases",
"immunology",
"parasitic",
"diseases"
] | 2012 | Protective Human Leucocyte Antigen Haplotype, HLA-DRB1*01-B*14, against Chronic Chagas Disease in Bolivia |
Diarrhea is an important cause of morbidity and mortality in all regions of the world and among all ages , yet little is known about the fraction of diarrhea episodes and deaths due to each pathogen . We conducted a systematic literature review to identify all papers reporting the proportion of diarrhea episodes with positive laboratory tests for at least one pathogen in inpatient , outpatient and community settings that met our inclusion and exclusion criteria . We identified a total of 25 , 701 papers with possible etiology data and after final screening included 22 papers that met all inclusion and exclusion criteria . Enterotoxigenic Escherichia coli and V . cholerae O1/O139 were the leading causes of hospitalizations . In outpatient settings , Salmonella spp . , Shigella spp . , and E . histolytica were the most frequently isolated pathogens . This is the first systematic review which has considered the relative importance of multiple diarrhea pathogens . The few studies identified suggest that there is a great need for additional prospective studies around the world in these age groups to better understand the burden of disease and the variation by region .
Diarrhea is an important cause of morbidity and mortality in all regions of the world and among all ages [1] , [2] . For children 5 years of age and older , adolescents , and adults mild to moderate diarrhea can lead to absenteeism from school or work and may require treatment by a health care provider . More severe diarrhea can lead to hospitalization; serious sequelae such as Guillain Barre' syndrome and hemolytic uremic syndrome; and in some cases death [3] , [4] . Though most diarrhea episodes are self limiting and dehydration can usually be controlled with oral rehydration therapy , it would be ideal to be able to prevent diarrhea , especially the more severe episodes which have a higher likelihood of progressing to complications or death . Some prevention strategies such as improved water and sanitation and basic hygiene practices are generalizable and thus do not require knowledge of diarrhea etiology , but others such as vaccines would benefit greatly from a comprehensive understanding of the overall burden of pathogen-specific diarrheal disease . Recent advances have led to the development of an effective rotavirus vaccine which is now recommended for young children as part of the routine immunization schedule [5] . A vaccine for cholera that could be useful in some settings in all ages has been available for several years , and is now recommended by the WHO for persons living in endemic areas [6] . The number of pathogens that are responsible for diarrheal disease goes far beyond rotavirus and Vibrio cholerae; however , the fraction of diarrhea episodes and deaths due to each pathogen is unclear , and thus uncertainty may inhibit prioritization of funding for research and disease control programs . There have been numerous studies conducted in countries around the world to determine the presence of one or more pathogens in diarrheal stools . While isolated studies provide important pieces of information , it is difficult to draw conclusions with regard to the importance of various pathogens without looking at a complete spectrum of agents simultaneously . We conducted a systematic literature review of diarrhea etiology studies to better understand the likely distribution of pathogen-specific diarrhea episodes and deaths in older children , adolescents and adults . To our knowledge this is the first systematic review designed to compile the data from multiple pathogens which might be applied to annual incidence and mortality rates in these age groups .
We first calculated the un-weighted mean , median and inter-quartile range for each pathogen for each type of patient population separately . There was only 1 community study [10] so for all analyses we combined this study with the studies conducted in outpatient settings . We then categorized inpatient and outpatient studies based on the number of pathogens reported by the authors in the methods and results sections of each published study: single pathogen studies , those reporting 2–4 pathogens , and those reporting at least 5 pathogens . For each of these categories we calculated the weighted mean for each pathogen . Among inpatient studies we removed the 1 study conducted in a high income setting to enable separate calculations for high vs . low and middle income countries separately . We stratified studies by inpatient or outpatient status and by the number of pathogens identified in each study to present the best possible summary data to approximate the most likely pathogen distributions for diarrhea mortality and all episodes , respectively .
We identified 25 , 701 papers with possible etiology data ( Figure 1 ) . After screening 5 , 986 abstracts and 932 papers , we found 45 that met our inclusion and exclusion criteria . Twenty-two papers met all inclusion and exclusion criteria and described the study populations with regard to inpatient , outpatient or community study populations [10] , [11] , [12] , [13] , [14] , [15] , [16] , [17] , [18] , [19] , [20] , [21] , [22] , [23] , [24] , [25] , [26] , [27] , [28] , [29] , [30] , [31] , [32] ( Table 2 ) . Twenty three additional papers met initial screening criteria but were subsequently excluded from the analysis presented here because they lacked information with regard to the patient population ( i . e . inpatient vs . outpatient ) or did not differentiate the results by population type ( Table 3 ) [20] , [33] , [34] , [35] , [36] , [37] , [38] , [39] , [40] , [41] , [42] , [43] , [44] , [45] , [46] , [47] , [48] , [49] , [50] , [51] , [52] , [53] . In table 4 we present the unweighted mean and median proportion of stools which tested positive for each pathogen in both in- and out-patient settings . In this analysis V . cholerae O1/O139 and ETEC were the leading causes of hospitalization . In inpatient populations Aeromonas spp . Yersinia spp . Cryptosporidium spp . , V . parahaemolyticus , P . shigelloides , and C . difficile were each found in <2% of patients . In out-patient settings , Salmonella spp . , Shigella spp . , and E . hystolitica were isolated the most frequently . In outpatient populations , EHEC , Campylobacter spp . , Aeromonas spp . and Yersinia spp . were found in <2% of patients . Very few studies tabulated data such that the co-occurance of more than one pathogen in a diarrheal stool could be ascertained and few tested a broad enough spectrum of pathogens to be able to quantify the proportion of episodes from which no currently recognized pathogen could be identified . In Table 5 we present the analysis of inpatient studies stratified by the number of pathogens sought among those studies conducted in low and middle income countries . We separately present the results for the single analysis which included more than 4 pathogens conducted in a high income setting [19] . There were very few single pathogen studies thus it is difficult to identify a trend as one progresses from single to comprehensive studies with at least 5 pathogens . In the studies conducted in low and middle income countries which identified at least 5 pathogens , 28 . 1% of hospitalized patients had tested positive for ETEC and 20 . 7% tested positive for V . cholera O1/O139 . For high income/low mortality countries , one study found that 14% of hospitalized patients tested positive for Campylobacter spp . followed by 11 . 5% of samples testing positive for Salmonella spp . [19] . For outpatient studies we only identified studies of single pathogens and those which looked for more than 4 pathogens ( Table 6 ) . The difference in proportion of stools testing positive for a particular pathogen is most noticeable for Shigella spp . where 34 . 3% of episodes were positive for Shigella in studies that sought only that pathogen , vs . only 9 . 4% positive among studies which looked for 5 of more pathogens .
This is the first systematic review which has considered the relative importance of multiple diarrhea pathogens for all regions of the world among children 5 years and older , adolescents , and adults using studies published in the peer reviewed literature . We stratified our results by inpatient vs . outpatient settings because it is likely that the distribution of pathogens differs by diarrhea severity . We found ETEC and V . cholerae O1/O139 to be the most frequently isolated pathogens among patients hospitalized for diarrhea; together they were observed in more than 49% of samples from patients in low and middle income countries . Because these studies were conducted in cholera endemic areas this is not surprising; the importance of cholera will depend on whether the study was done in an endemic or epidemic area thus these results are not possible to generalize to all countries . Rotavirus , which is known to be a leading cause of death among young children , was not found to be as important among older persons providing additional evidence suggesting immunity with increasing age . In outpatient settings , Salmonella spp . , Shigella spp . , and E . histolytica were the most frequently isolated pathogens . Because little is known about the care-seeking behavior for community-acquired diarrhea among children 5 years of age and older and adults , additional data are needed in this age group to determine the distribution of pathogens in the community . Because blood in the stool is common for illnesses due to Shigella spp . , Campylobacter spp . , and E . histolytica and may occur with Salmonella spp . it is possible that the isolation of these pathogens would be higher than in a true community-based setting due to an increase in care-seeking behavior for illnesses with the presence of blood in the stool . We only identified one community-based study; thus , separate estimates for outpatient and community studies were not possible . The overall scarcity of the data used to produce these estimates is a major limitation . This is particularly concerning when generalizing across regions and when making assumptions about variations which are likely among low , middle , and high income countries based on variation in geography and risk factors . Given the few studies meeting our criteria for inclusion in the review , it is not possible to account for the additional differences in study populations by region or over time which might have also influenced the spectrum of pathogens due to changes in pathogens chosen for isolation , pathogens circulating in a community , and baseline characteristics of the study population . An additional limitation of this review is the time span of the included studies and thus heterogeneity of laboratory methods for some key pathogens . In the last 30 years , diagnostic methods have evolved for many pathogens , such as diarrheagenic E . coli and E . histolytica . New laboratory methods , including PCR , and antigen detection assays have increased sensitivity and decreased risk of misclassification substantially . Because some reports included in this review used older laboratory methods there is a risk that data from these may under- or over-represent the prevalence of selected pathogens . However , because of the overall paucity of data we chose to include these studies however caution should be taken when interpreting the results for these selected pathogens for which laboratory methods have improved dramatically over the past 30 years . In this review we stratified studies by those that sought a single pathogen and those that considered multiple pathogens . Because single pathogen studies often pick study sites based on a known prevalence of a particular pathogen it can be expected that the observed rates would be higher than in studies where multiple pathogens are being isolated . This was especially true for Shigella spp . where we found the weighted mean dropped from 35 . 3% in the single pathogen studies to 9 . 4% in the multiple pathogen studies . In addition , outpatient studies did not look specifically for some pathogens such as V . cholerae O1/O139 thus limiting the inference about non severe episodes . We recognize that we did not capture the true burden of every possible pathogen that might cause diarrhea because many pathogens occur in outbreaks and these may not have been included in these ongoing disease surveillance studies . For example , we only found one study that included detection of norovirus [15] meeting our study inclusion criteria of at least 12 mo of surveillance . Norovirus is known to be seasonal and a frequent cause of epidemics , so may be underestimated in our review . Similarly because we did not include outbreak data , pathogens that are more typically observed in outbreaks may have been missed if they were not known to be endemic in the study area . Because we identified very few studies that tested for 5 pathogens or more and most were from South Asia , we were not able to assess regional differences in pathogen importance . For pathogens that are not known to be prevalent globally such as V . cholerae O1/O139 this is especially problematic . Ideally unique pathogen distributions would be developed for each region and for large countries , such as Brazil , China , and India . National level community-based surveillance and inpatient reports would enable countries to better understand the local burden of disease by pathogen and better design prevention programs . In this analysis we have treated all data as isolated proportions yet we recognize that this is not the case for many diarrhea episodes . Many patients have multiple pathogens and likewise for some patients , no pathogen is found . Because we had very few studies seeking multiple pathogens and even fewer reporting mixed infections , we were not able to conduct a more complex analysis to control for the role of multiple infections . We also recognize that the identification of a pathogen in the stool does not necessary mean that it is the cause of the illness . Many patients are asymptomatic carriers and thus the prevalence of some pathogens might be found at the similar proportions in healthy individuals . These pathogens have a lower pathogenicity than those that are never or rarely identified in the stools of asymptomatic individuals . Only one study in our final data set provided data for asymptomatic controls , thus a full analysis to control for asymptomatic carriage was not possible . This study is the first to systematically review the literature on the etiology of diarrhea in children ≥5 years of age , adolescents and adults and provides an important overview of the distribution of pathogens responsible for both infection and possible death . The few studies identified suggest that great caution must be taken when interpreting these limited data . Many limitations have been identified suggesting the need for additional prospective studies around the world in these age groups . Understanding the burden of pathogen specific diarrheal disease and the variation by region is important for planning effective control programs for the overall reduction of diarrhea disease among persons of all ages . | Diarrhea is an important cause of illness and death around the world and among people of all ages , but unfortunately we often do not know what specific bacterium or virus causes the illness . We conducted a review of the scientific literature with the goal of finding published studies that identified bacteria and viruses among patients with diarrhea in the community and in hospital settings . We initially found nearly 26 , 000 papers on this topic but narrowed the list to 22 studies that met all of our specific criteria for inclusion in our review . Among patients hospitalized for diarrhea , E coli and Vibrio cholerae were found in more than 49% of people living in middle income and poor countries . Among patients who sought care from their doctor on an outpatient basis , Salmonella spp . , Shigella spp . , and E . histolytica were most often found . In our review we focused on the differences in the distribution of pathogens between patients in inpatient vs . outpatient settings because these estimates may best approximate what we would expect to see if the distribution were applied to global estimates of diarrhea deaths vs . uncomplicated illnesses . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"public",
"health",
"and",
"epidemiology/global",
"health",
"public",
"health",
"and",
"epidemiology",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"infectious",
"diseases/gastrointestinal",
"infections"
] | 2010 | Etiology of Diarrhea in Older Children, Adolescents and Adults: A Systematic Review |
Collective cell migration in cohesive units is vital for tissue morphogenesis , wound repair , and immune response . While the fundamental driving forces for collective cell motion stem from contractile and protrusive activities of individual cells , it remains unknown how their balance is optimized to maintain tissue cohesiveness and the fluidity for motion . Here we present a cell-based computational model for collective cell migration during wound healing that incorporates mechanochemical coupling of cell motion and adhesion kinetics with stochastic transformation of active motility forces . We show that a balance of protrusive motility and actomyosin contractility is optimized for accelerating the rate of wound repair , which is robust to variations in cell and substrate mechanical properties . This balance underlies rapid collective cell motion during wound healing , resulting from a tradeoff between tension mediated collective cell guidance and active stress relaxation in the tissue .
Collective cell migration is central to tissue morphogenesis , wound repair and cancer metastasis [1] . During tissue repair after wounding [2] , or during closure of epithelial gaps [3 , 4] , collective cell migration enables the regeneration of a functional tissue . Gap closure is usually mediated by two distinct mechanisms for collective cell movement [5–7] . First , cells both proximal and distal to the gap can crawl by Arp2/3 driven forward lamellipodial protrusions [6–8] . Secondly , cells around the gap can collectively assemble a supracellular actomyosin cable , known as a purse-string , which closes tissue voids via active contractile forces [6 , 9] . It remains poorly understood how these two modes of collective cell movement , driven by the assembly of distinct actin network architectures , are regulated in diverse biophysical conditions . Many experimental studies have provided key insights into the physical forces driving collective cell migration [7–13] . Recent in vitro wound healing experiments have shown that closure of large wounds is initiated by cell crawling , followed by the assembly of purse string that dominates closure at smaller wound sizes [12 , 13] . Purse-string acts like a cable under contractile tension , pulling in the wound edge at a speed proportional to its local curvature [14] . By contrast , crawling driven closure occurs at a constant speed , regardless of wound morphology [7] . However , it remains unknown how the mechanochemical properties of individual cells and their interactions with the extracellular matrix regulate crawling and purse-string based collective cell motion . While experiments are limited in the extent to which mechanical effects are separated from biochemical processes , theoretical and computational models can decouple these variables precisely . Extensive theoretical work has been done to model collective cell migration during tissue morphogenesis and repair [15–21] . However , existing models do not explain how individual cells adapt their migratory machineries and interactions with neighboring cells to move collectively like a viscous fluid while maintaining tissue cohesion . Continuum models of tissues [22] as viscoelastic fluids [13 , 16] or solids [14 , 15 , 17 , 23] have been successful in describing collective flow and traction force patterns observed experimentally . However , such macroscopic models cannot capture cellular scale dynamics , and therefore unsuited for connecting individual cell properties to collective cell dynamics . On the other hand , cell-based computational models , including the Cellular Potts Model [24 , 25] , Vertex Model [26 , 27] , phase-field [28] or particle-based models [20 , 29 , 30] explicitly account for dynamic mechanical properties of individual cells and their physical interactions . However , these models have not yet been developed to integrate the mechanics of cell motion with cell-substrate adhesions and intracellular cytoskeletal dynamics . It remains poorly understood how migrating cells sense changes in their physical environment and translate those cues into biomechanical activities in order to facilitate collective motion . This is particularly important for epithelial wound healing , where wound edge cells actively remodel their cytoskeletal machineries and the resulting modes of motility in response to changes in wound size , shapes and substrate properties [12 , 14 , 31] . To overcome these limitations , we propose an integrative modeling framework that incorporates the mechano-chemical coupling of cell motion and adhesion with stochastic transformation between protrusive and contractile cell behaviors . In contrast to previous cell-based models of wound healing [18 , 31 , 32] , our approach explicitly accounts for the spatiotemporal regulation of protrusive and contractile activities , cell-matrix interactions , adhesion turnover , and cell polarity . Using this model , we ask: How do migrating cells sense changes in their physical environment ? How do cells regulate their modes of motilities to optimize the speed of collective motion ? What roles do tissue mechanical properties play in stress propagation and relaxation during wound repair ? In particular , we find that an optimum mixture of protrusive and contractile cell activities at the wound edge accelerates the rate of wound healing under diverse conditions . The optimum mixed mode of migration is robust to changes in substrate rigidity , wound shape , intercellular adhesions and cortical tension . A unique insight offered by our study is that a mixture of protrusive and contractile activities promotes faster wound repair by optimizing the tradeoff between collective cell guidance and local stress relaxation . Finally , we propose a fundamental mechanism by which tissues can locally fluidize to drive rapid collective cell motion while maintaining their overall mechanical integrity .
Our model consists of several computational components that simulate: ( 1 ) mechanical interactions between cells , ( 2 ) biochemical dynamics ( protrusions , adhesions ) , and ( 3 ) transitions between distinct cell motility modes . Mechanical interactions between cells are simulated using the vertex model for epithelial mechanics [18 , 21 , 26 , 27 , 33–35] , where the geometry of each cell is defined by a two-dimensional polygon , with mechanical energy given by: E i = K ( A i - A 0 ) 2 + Γ P i 2 + γ P i . ( 1 ) The first term in ( 1 ) represents the energy cost for cell compressibility , where Ai is the area of cell i , A0 is the preferred cell area , and K is the elastic constant . The second term , Γ P i 2 , is the energy due to contractile forces in the actomyosin cortex . The last term in ( 1 ) represents the interfacial tension between cells , which is the difference between cortical tension and the cell-cell adhesion energy per unit length . The elastic substrate is modeled as a triangular mesh of harmonic springs ( Methods ) . Focal adhesion complexes are modeled as stiff springs that anchor the cell vertices to the substrate mesh , with attachment and detachment rates given by kon and koff , respectively ( see Methods ) . The net mechanical force acting on the cell vertex α is given by Fα = −∂Etot/∂xα , where E tot = ∑ i = 1 n E i + E adh is the total mechanical energy of the cells and the cell-substrate adhesions . In addition to mechanical forces ( Fig 1A ) , cells within the bulk tissue actively move with a self-propulsion velocity v 0 p ^ i ( Fig 1B ) , where p ^ i defines the polarity vector for cell motion , and v0 is the self-propulsion speed . Cells at the wound leading edge initiate motion by crawling towards the wound center [12 , 13] , with a force fp ( Fig 1A ) . At each time step , crawling cell fronts can transition to a purse-string at a constant rate kp . This leads to an increased line tension on the wound edge due to actomyosin contractility ( Fig 1A ) ( see Methods ) . Assuming over-damped dynamics , cell vertex α at the wound edge moves as: μ d x α d t = F α + f p α , ( 2 ) where μ is the friction coefficient . Cell vertices in the bulk of the tissue move according to following equation of motion μ d x α d t = F α + 1 n α ∑ i ∈ α μ v 0 p ^ i , ( 3 ) where the last term is the averaged self-propulsion force over nα neighboring cells sharing the vertex α ( Fig 1B ) . We estimate the model parameters from available experimental data ( Methods , Table 1 ) . To elucidate the mechanisms of collective cell motion during wound repair , we simulated healing of a circular wound for a mixed modality of closure: kp = 4 hr−1 . Initially , cells close the wound by crawling ( Fig 1C ) , but over time they switch to the purse-string mode , resulting in rapid contraction of cell edges lining the wound periphery ( Fig 1C , S1 Video ) . To quantify the spatiotemporal patterns of collective cell motion , we calculated spatially averaged radial and azimuthal velocities as a function of the radial distance from the wound center at each time point ( Fig 1D and 1E ) . Initially , both radial and azimuthal velocities are highest around the wound edge and decay with distance inside the monolayer . As crawling cells pull on the substrate , the resultant traction forces point radially outwards and away from the wound ( Fig 1C and 1F ) . Halfway through the closure process , the purse-string fully assembles ( Fig 1G-inset ) and the traction forces switch to pointing radially inwards ( Fig 1F ) , in quantitative agreement with experimental data [31] . Consistent with experiments , tangential traction stresses are comparable in magnitude with the radial components of the traction stress ( S2B Fig ) . Our model reproduces the experimental observation that focal adhesions are oriented towards the wound center for crawling cells [31 , 36] ( S3A and S3C Fig ) . By contrast , purse-string adhesions have a higher probability of orienting tangentially at the leading edge than crawling cells ( S3B , S3D and S3E Fig ) . As closure proceeds , the band of high radial velocities around the wound narrows ( Fig 1D ) , while the azimuthal velocity narrows and decreases around the wound ( Fig 1E ) . This results in more coordinated inward motion of the cells . Increasing kp from 0 ( crawling only ) to 1000 hr−1 , monotonically increases the proportion of wound perimeter covered by the purse-string over time ( Fig 1G-inset ) . For non-zero values of kp , wound area shrinks in a biphasic manner: an initial slow exponential decay , followed by fast exponential decay , consistent with experimental data [36] . In contrast to the mixed mode of closure ( Fig 1C and 1G ) , the traction forces for crawling mediated closure are always directed radially outwards ( S2C Fig ) , because crawling cells pull on the substrate . While further inside the monolayer the traction forces point radially inwards as the rear end of crawling cells retract via cortical contraction . In purse-string mediated closure , the wound shape remains circular throughout ( S2 Video ) , in contrast to the ruffling morphology observed for crawling cell fronts ( S2 Video ) . Traction forces point into the gap , and increases in magnitude as the wound size gets smaller ( S2D Fig ) . For a fixed set of parameters , we find that a balance of purse-string and crawling mediated closure results in faster wound healing ( Fig 1G ) . To determine how the relative proportion of purse-string and lamellipodia is optimized for rapid collective motion , we turned to examine how the purse-string assembly rate ( kp ) regulates wound closure time for varying physical properties of the cells , the underlying substrate , wound size and shape . Since the speed of cell crawling and the magnitude of traction forces are sensitive to substrate rigidity [37 , 38] , we first investigated the role of substrate stiffness on wound closure time . To this end , we varied the substrate Young’s modulus , Es , and the purse-string assembly rate , kp , for fixed physical properties of the tissue and the wound . We find that wound closure time increases with Es for higher values of kp , but remained insensitive for crawling mediated closure ( Fig 2A ) . Strikingly , there exists an optimum value of kp ( corresponding to mixed modality ) for any value of Es , which results in minimum closure time ( Fig 2A ) . For fixed kp , the strain energy transmitted to substrate decreases monotonically with increasing stiffness for Es > 0 . 5kPa ( Fig 2B ) ( see Methods for calculation details ) . For all values of Es and kp , faster wound closure coincides with higher strain energy transmitted to the substrate , signifying a positive correlation between energy cost and the speed of wound healing . Our results agree with experimental findings that wound closure time is not sensitive to changes in substrate stiffness for moderate to high rigidities [31 , 36] . On very soft substrates ( < 500 Pa ) , our model predictions are inconsistent with experiments by Anon et al [7] , who showed that crawling-based migration fails to close wounds on very soft gels ( ∼ 100 kPa ) , as lamellipodia do not form . This may be captured by implementing additional biochemical feedback mechanisms between protrusive activity and substrate stiffness , beyond the scope of our model . As Es is increased , purse-string driven motion slows down . To quantify the dependence of closure time on stiffness , we calculated the Pearson’s correlation coefficient between wound closure time and substrate stiffness for different modes of wound closure ( Fig 2C ) . We find that purse-string based motility slows down with increasing stiffness , with a positive correlation coefficient significantly different from zero ( p-value < 0 . 05 ) . In contrast , crawling based motility and have the least significant correlation coefficient ( p-value > 0 . 05 ) . The sensitivity of purse-string driven motility to substrate rigidity ( Fig 2A and 2C ) can be explained by a mechanical force balance argument ( Fig 2D and 2E ) . Purse-string driven contractile forces drag the border cells into the gap , in competition with cortical tension retracting the rear cell edges . This results in a large net resistive force from the deforming elastic substrate ( Fig 2D ) . By contrast , crawling cells pull the substrate backwards at the wound edge and contractile forces pull the substrate forward at the cell rear ( Fig 2E ) . This dipole-like traction pattern results in a net assistive force from the substrate , pointing towards direction of cell crawling . During mixed mode of migration , a combination of net assistive and resistive forces should therefore lead to the least sensitivity to substrate stiffness . To test this hypothesis , we computed the net radial traction force , Fr , on the substrate under the first row of cells at the wound edge . We then calculate the time-averaged ratio between the radial force and the radial velocity , vr , of the wound edge , to obtain an effective friction coefficient: μeff = 〈Fr/vr〉 ( Fig 2F ) . We find that μeff monotonically increases in magnitude with increasing substrate stiffness ( for all modes of migration ) , consistent with previous theoretical predictions [39] . For all values of substrate stiffness , purse-string motion leads to the highest positive μeff , suggesting high resistance and sensitivity to substrate rigidity . Crawling driven motility leads to negative μeff , indicative of assistive motion . By contrast , the mixed mode of migration leads to the lowest magnitude of μeff , i . e . least drag from the substrate . Rigidity sensing by different modes of collective migration is expected to be strongly coupled to focal adhesion kinetics . While we have assumed constant rates of binding and unbinding of cell-substrate adhesions , experiments have demonstrated that integrin-ligand pairs form catch bonds [40] , such that koff decreases under low forces and increases under larger forces . To test if the mechanosensitivity of cell-substrate adhesion bonds impact our results , we implemented a catch bond model for adhesions , assuming a single bound state and two unbinding pathways [41] ( see Methods ) . As a result , the crawling mode of closure is now more sensitive to changes in substrate stiffness , with closure time increasing with stiffness , before decreasing at higher stiffnesses due to increased adhesion lifetime ( S4 Fig ) . Purse-string driven closure shows an increase in sensitivity compared to the default case , while the mixed mode of closure is least sensitive to changes in substrate stiffness . However , the mixed mode of migration is always the fastest , irrespective of force sensitivity of the adhesions . Aside from mechanosensitivity of different modes of wound closure , the driving force for closure is expected to be strongly dependent on the relative proportion of purse-string and crawling cells . Since the actomyosin purse-string is a cable under tension , the driving force for closure is proportional to the wound curvature . As a result , purse-string driven closure is expected to be sensitive to the wound geometry [12 , 14] . By contrast , crawling driven closure has been found to reduce wound area at a constant speed [7] . Therefore , we sought to investigate how the coaction of purse-string and crawling based motilities modulate collective motion for varying wound morphologies . For circular wounds of varying radii we recapitulate the experimental result that closure time increases with wound radius ( Fig 3A ) [7] . However , the optimum purse-string assembly rate ( kp ) for fastest closure decreases with wound radius , such that closure time is highly sensitive to kp for larger wounds . This is because purse-string driven forces are higher near the end of closure , and that purse-string force is low in the beginning of closure of a large wound . For larger wound radii , an optimum mixture of purse-string and protrusive cell crawling leads to fastest closure . We find that the average strain energy on the substrate increases monotonically with wound radius for kp ( Fig 3B ) , but is more sensitive to wound size for purely crawling mediated migration ( kp = 0 ) . Next we simulated elliptical shaped wounds of fixed area but varying aspect ratios . We find that regardless of the migratory mode , closure time decreases with increasing aspect ratio ( Fig 3C ) . In addition , there exists an optimum value of kp for a given aspect ratio that leads to minimal closure time . Thus , a mixed mode of closure is always the fastest , but isn’t much faster than crawling mediated closure for high aspect ratio wounds . This is because crawling cells advance at a constant speed perpendicular to the wound edge . Therefore only the short axis distance must be crossed for the wound to close ( S4 Video ) ( S5A Fig ) . For purse-string driven closure , the high curvature ends of elliptical wounds move rapidly inwards , leading to faster closure than circular shapes ( S5C Fig ) . At all values of aspect ratio , strain energy is inversely proportional to closure time ( Fig 3D ) . Since purse-string behaves as a contractile cable , then for wounds with concave morphologies ( positive curvatures ) , cells should be pulled away from the wound by the purse-string tension . To investigate this we simulated concave wound shapes as in ref . [14] . For varying degrees of concavity ( with fixed area ) , we observed that a mixed mode of closure leads to fastest wound closure ( Fig 3E ) . To quantify the relationship between wound healing speed and curvature , we measured the local velocity and curvature at the wound perimeter . We find that the purse-string velocity is proportional to the curvature , crawling velocity is curvature-independent , while a mixture of crawling and purse-string leads to faster collective motion , with velocity decreasing with curvature ( Fig 3F , S5 Video ) . These findings quantitatively agree with experimental data [14] . Previous studies suggest the possibility that purse-string and lamellipodia-based migration during wound healing can be geometrically coupled [14 , 29 , 31] , such that the formation of protrusive borders may be directly coupled to the assembly of purse-string cables on neighboring wound edges with opposite curvatures . Such a mechanism is not captured by a purely stochastic transition between protrusive and contractile activities . To this end , we implemented a model of curvature sensing motility of the wound leading edge , similar to Ref . [29] , where the switching between crawling and purse-string mechanisms is regulated by the local curvature of the wound ( S7 Fig ) . Based on this model , if the curvature of a cell’s leading edge is larger than a threshold curvature , it contracts via purse-string . Otherwise , the cell moves via protrusive crawling ( Methods , S7 Fig ) . We applied this model to wounds with non-uniform curvatures as in Fig 3E . Consequently , the convex regions move forward by crawling , whereas contractile purse-string cables assemble in the concave regions . We find that for all three concave shapes in Fig 3E , the curvature sensing mechanism closes the wound at least as fast as in the mixed case with stochastic switching of motility modes ( Fig 3E–3G ) . We note that the curvature-sensing mechanism may not be applicable to the closure of undamaged epithelial gaps where purse-string cables do not form [7] . Our cell-based model predicts many differences in collective cell motility driven by contractile and protrusive activities ( Figs 1–3 ) . In particular , purse-string tension rounds the wound edge and leads to solid-like , radial deformation of the tissue ( Fig 3F-inset ) . By contrast , crawling cells ruffle the wound leading edge ( Fig 3G , S3 and S5 Videos ) , suggestive of lack of guided motion . To quantify differences in tissue deformation and their relationship to collective motion , we measured the angle ( θ ) between cell center velocity and the unit vector pointing towards the wound center ( Fig 4A ) . In purse-string driven closure ( kp = 1000 hr−1 ) , the angle distribution shows a single peak at θ = 0 , corresponding to radially inward deformation ( Fig 4B ) . By contrast , crawling cells ( kp = 0 ) have a wider distribution of angles , with secondary peaks at θ = ±π ( Fig 4B ) , representing outward motion from cell neighbor exchanges ( Fig 4A ) . To quantify the distributions , we define collective cell guidance , G , as the probability that a cell moves towards the wound center: G = ∫ - π / 2 π / 2 P ( θ ) d θ , which monotonically increases with increasing kp ( Fig 4C ) . Since tissue deformation properties depend on cortical tension , cell contractility , and cell-cell adhesions [34 , 42–44] , we investigate how cellular mechanical properties regulate collective guidance ( G ) . We can rewrite the mechanical energy of cells ( Eq ( 1 ) ) as: E i = K ( A i - A 0 ) 2 + Γ ( P i - P 0 ) 2 , ( 4 ) where P0 = −γ/2Γ is the preferred cell perimeter . The non-dimensional shape parameter p 0 = P 0 / A 0 controls cell shape anisotropy and the emergent rigidity of confluent tissues [45] . Increasing p0 reduces cortical tension relative to cell-cell adhesions , which softens the tissue . It has been shown that confluent tissues behave like a jammed solid for p0 < 3 . 81 , whereas it exhibits fluid-like behaviour for p0 > 3 . 81 [45] . Activity in the form of cell motility , division , or death can fluidize tissues further by lowering the critical p0 for rigidity transition [46–48] . In our model , activity arises from self propulsion ( v0 ) ( S1B and S8 Figs ) , and cell crawling whose relative strength is regulated by kp . We find that increasing p0 decreases G , regardless of kp ( Fig 4C ) . The decrease in G with increasing p0 arises from an increased rate of cellular neighbor exchanges ( T1 transitions ) that locally fluidizes the tissue ( Fig 4D ) . Surprisingly , for a fixed p0 , T1 rates in the wounded tissue is highest for intermediate values of kp , resulting in minimum closure time ( Fig 4E ) . With higher p0 , cells have a higher preferred perimeter , such that both contractile and protrusive motilities experience lower mechanical resistance from tension in the border cells ( S6 Video ) ( S9 Fig ) . This enables a faster reduction in wound area as compared to rigid tissues with lower p0 ( Fig 4E ) . These findings elucidate the mechanical basis for rapid collective migration via a mixture of protrusive and contractile cell activities . Purse-string driven tension maximizes collective cell guidance and leads to the lowest frequency of tissue rearrangements , such that cell movements are impeded by mechanical resistance from the surrounding tissue . By contrast , purely crawling motion exhibits the lowest collective guidance due to randomized protrusions of individual cells at the wound leading edge . We find that an optimum mixture of crawling and purse-string leads to intermediate collective guidance , while maximizing the frequency of local tissue rearrangements ( Fig 4D ) . This mechanism of active fluidization enables tissues to locally relax their mechanical stress , promoting rapid wound healing . When intercalations are disabled in the model , tissue mechanical energy increases due to increase in cell elongation around the wound ( S10 Fig ) . This results in cell jamming and slowing down of wound closure . Therefore , cell intercalations , promoted by a mixture of contractile and protrusive forces , lead to efficient wound closure by minimizing both tissue mechanical energy and wound closure time . Recent experiments , however , suggest that cells may not necessarily try to minimize energy or closure time during wound healing [36] . But rather , they tend to coordinate the assembly of diverse actin architectures to conserve the amount of mechanical work done per unit time .
Our cell-based computational model quantitatively captures a wide range of experimental trends including the patterns of collective cell motion and traction stress organization for crawling and purse-string mediated wound closure ( Fig 1 ) . We reproduced the experimentally observed size-dependence of wound closure times , the curvature dependence of purse-string velocity , and independence of cell crawl speeds to variations in wound morphology . We predict that increasing aspect ratio of the wound speeds up closure as crawling cells can rapidly cross the short axis of the wound , whereas purse-string cables can generate rapid movements on regions of high curvature ( Fig 3 ) . Robust to variations in substrate and tissue mechanical properties , we find that an optimum proportion of protrusive and contractile motilities accelerates wound closure . While purse-string driven motion slows down on stiffer gels due to an increased resistance from drag on the substrate , crawling driven migration is largely independent of substrate stiffness ( Fig 2 ) . We find that a mixed mode of collective migration is more efficient regardless of substrate stiffness . Robust to parameter variations , an increase in closure speed is associated with an increase in the strain energy transmitted to the underlying substrate ( S11 Fig ) . As a result , migrating cells actively dissipate more mechanical energy to their environment in order to speed up collective motion . A source of active stress dissipation comes from cellular neighbor exchanges that locally fluidize the tissue , resulting in faster wound closure ( Fig 4 ) . These T1 transitions have previously been observed in vivo , during wound closure in Drosophila embryo epidermis [49] . T1 transitions are also observed in our in vitro laser-ablation experiments on MDCK monolayers , where the number of cells at the wound edge decreases over time via wound edge intercalations ( S12 Fig ) . In our model , the mechanism of active fluidization via intercalation is promoted by a mixture of protrusive and contractile activities of wound edge cells , and reduced contractility or increased cell-cell adhesion in the bulk of the tissue . The ability to actively remodel an elastic tissue , coupled with tension-driven collective cell guidance , constitute the two key mechanisms for rapid directed motion in adherent environments . While the stress relaxation mechanism in our model comes only from cell neighbor exchanges , other dissipative mechanisms can also be triggered by mechanical forces including cell shape fluctuations [50] , cell division [51] or cell death [4] . In these cases , our prediction will remain very similar , with the rate of cell movement into free space augmented by the sum of relaxation rates of various dissipation modes [46] . A future challenge is to identify the molecular pathways that activate distinct stress relaxation modes during tissue development and regeneration .
We model the substrate as a triangular mesh of springs with a spring constant ks . The Young’s modulus of the substrate is given by E s = 2 k s / 3 h s , where hs is the substrate thickness , and the Poisson’s ratio for a triangular mesh is ν = 1/3 . Since focal adhesions and cellular traction forces typically localize at the cell periphery [44] , we implement adhesions at the cell boundaries . We model the focal adhesion complexes as stiff springs with stiffness kf , which connect the cell vertices with the substrate mesh . Bound focal adhesions can detach stochastically with a rate koff , whereas unbound cell vertices can attach to the nearest node of the substrate mesh with a rate kon . The resultant force on the cell vertex is , f adh α = - ∂ E adh α ∂ x α , ( 5 ) E adh α = σ α k f 2 ( | x α - r α | - | x 0 α - r 0 α | ) 2 , ( 6 ) where E adh α , is the adhesion energy , σα is the state variable for cell-substrate attachment ( 0: detached; 1: attached ) , rα is the position of the substrate mesh connected to xα , and x 0 α and r 0 α are the initial positions of the cell and the substrate vertices at the time of adhesion formation . Each cell carries a unit polarity vector , p ^ i , which represents the front/rear polarization of a motile cell [52] . The polarity vector is an internal state variable of cell that specifies the preferred orientation of cell motion , not their actual direction of motion . Cells in the bulk of the tissue , i . e . not on the wound edge , move due to self-propulsion [47] . The polarity of a bulk cell i is defined by a unit vector with angle θi that undergoes rotational diffusion: ∂ t θ i = η i ( t ) , ⟨ η i ( t ) η j ( t ′ ) ⟩ = 2 D r δ i j δ ( t - t ′ ) , ( 7 ) where Dr is the rotational diffusion constant , and ηi ( t ) is a Gaussian white noise with mean 0 and variance 2Dr . The self-propulsion of cell i results in a force on the vertex α as: 1 n α ∑ α ∈ i μ v 0 p ^ i , where v0 is the self-propulsion speed , and the sum is over all neighboring cells to vertex α ( S1B Fig ) . Here , we have neglected alignment interactions between cell polarity vectors in the bulk of the tissue , which can drive coherent swirling motion of cell collectives [53] . Without such polarity alignment rules , cell velocity vectors remain correlated over ∼ 5 cell diameters due to mechanical interactions ( S13 Fig ) , somewhat less than the correlation lengths measured in experiments in the absence of a wound [54] . To model lamellipodia based crawling , we allow cell vertices at the wound edge to protrude in the direction of polarity before attaching to the substrate ( S1 Fig ) . This pushes the cell front outwards , while cortical tension pulls the rear of the cell forwards . The polarity vector of cells along the wound points into the gap , and is determined by the mid-point of the wound edges . The direction of protrusion is given by the unit vector v ^ α i of wound cell i , which makes half the angle between the two lines joining the centroid of cell i to the vertices on the wound that neighbour other cells , i . e . are on the boundary of internal and external edges ( S1C Fig ) . This ensures contact inhibition of locomotion [30] , preventing collision of two neighbouring cells . For a cell i neighbouring the wound , the crawling force on vertex α on the wound edge is given by: f p α = f p ( 1 - σ α ) v ^ i α , where fp is the protrusion force magnitude . For simplicity we have assumed that fp is independent cell-substrate adhesions . However , protrusive activity remains strongly correlated to focal adhesion kinetics , since the frequency of the protrusions is controlled by the rate of focal adhesion binding and unbinding . As a consequence of this feedback , increasing the duty ratio of adhesions leads to slower crawl speeds and increased closure time ( S4D Fig ) . Here we describe the model where the switching between crawling and purse-string modes is dependent on the local geometry of the wound leading edge . At each time step in the simulation , cells at the wound edge makes a decision to switch its motility phenotype based on the local curvature of the wound edge . We calculate the curvature of a wound edge cell as the inverse of the radius of a circle inscribed to that cell edge . Curvature is defined as positive if the wound is convex ( e . g . a circle ) , and negative otherwise . If the curvature is above a threshold value , then the cell switches to a purse-string mode . If the curvature is below the threshold value , then the cell moves by crawling . As a result , cells typically start by crawling and switch to the purse-string mode as the wound shrinks in size , consistent with experimental findings [31] . To determine the optimum value of the threshold curvature , we varied the threshold curvature for switching to a purse-string mode , and computed the resultant wound closure time for a given initial wound shape . The optimum threshold curvature is given by the curvature value that minimizes wound closure time , as shown in S7B Fig . We implemented a catch-bond model for cell-substrate adhesions , where the detachment rate of the adhesion bonds , koff is a function of the bond tension , f , as given below: k off ( f ) = k 0 e - f / f 0 + k 1 e - f / f 1 . ( 8 ) The functional form for the detachment rate is taken from a catch bond model for integrin-ligand bonds that assumes a single bound state and two unbinding pathways [41] . The parameters k0 , k1 , f0 , and f1 have previously been estimated for single integrin ligand bonds [55] . Based on that estimate , we calibrate these parameters for the coarse-grained adhesion bonds in our simulations that represent several ligand-integrin pairs . We used parameter values of k0 = 25 hr−1 , k1 = 0 . 006 hr−1 , f0 = 3 . 125 μN , and f1 = 0 . 6944 μN , which results in the default unbinding rate at zero force , and showed high sensitivity to substrate stiffness . The vertex model is implemented using Surface Evolver [56] . We generate a wound by removing any cells that lie totally or partially within the wounded area . Edges surrounding the wound are then moved to the target wound shape . We then relax the energy of the remaining cells without adhesions so that all vertices on the wound lie on the target wound perimeter and system is at an energy minimum . To initiate gap closure , cells around the wound are set to crawling mode . We then execute the following steps ( S1 Fig ) until wound closure: Table 1 lists the parameters used in our simulations . The number of cells was chosen to be large enough to avoid finite size effects and displacement on the outer row of the cells . To confirm this , we ran wound healing simulations using different numbers of cells . As the number of cells increases from 50 , closure time increases and then quickly plateaus after cell count reaches 100 ( S14A Fig ) . We use a default value of 150 cells , but increase the cell number ( in the range 150-250 ) while running simulations for wounds with larger sizes ( Fig 3A ) . Substrate node density was chosen to be small enough so that a cell vertex is always close to a node in the substrate spring mesh , allowing focal adhesions to form with a relatively short length . As shown in S14B Fig , we find little dependence of closure time on node density , and use 0 . 6 μm−2 as the default value . The preferred area of the cell , A0 , is chosen to be approximately the same as the average area of MDCK cells in wound healing assays [13 , 31] . The preferred perimeter P0 is chosen so that the cell shape index , p 0 = P 0 / A 0 is close to the value for a regular hexagon , enabling us to study the effects of cell shape anisotropy on wound healing speed . The substrate stiffness was chosen as a typical value for gels used in in vivo wound healing assays [31]; the Poisson’s ratio of 1/3 for the substrate is a consequence of using a triangular mesh of linear springs . The Young’s modulus of the substrate defines the force scale in the simulations . The wound radius was chosen to be in the range 5-30 μm , similar to those in experimental studies [12 , 13 , 31] . Purse-string tension was estimated by taking the product of the force generated by a single myosin motor , 3 pN [57] , with the typical number of myosin motors in a contractile ring of length 15 μm and thickness 1 μm , 105 [58] , which gives a tension of 300 nN . Next , we fit parameters for cell area and perimeter elasticities , K and Γ , adhesion binding and unbinding rates , koff and kon . Together , these parameters determine the overall tissue motility and the magnitude of traction force generation . Thus we fit them simulataneously to the experimental data for typical closure speed and traction force magnitudes generated during closure [12 , 13 , 31] . In addition , we examine the spatiotemporal pattern of traction forces generated during closure . For example , traction stresses are normally localized around the wound but are not evenly distributed around the perimeter . Low adhesion time leads to smooth closure but little traction force while higher adhesion binding times lead to an even distribution of traction around the wound but the closure dynamics are less smooth . We estimate the protrusion force , fp , by comparing to single cell crawling speeds of 15 μm hr−1 [7] . To this end , we simulated a single crawling cell with a fixed polarity vector , and calibrated fp to the value that resulted in a crawl speed of 15 μm hr−1 . Internal motility speed was set to a similar value as cell crawling speeds . Dependence of wound closure time for variations in fp and γps are shown in S2F Fig . Whereas , the dependence of closure time on internal motility , v0 is shown in S8A and S8C Fig . The range of purse-string assembly rates were chosen so that the minimum value , kp = 0 yields pure crawling , the maximum , kp = 1000 hr−1 , yields 100% purse-string coverage , and intermediate values produce a combination of purse-string and crawling . We record displacements of the substrate mesh , u = r − r0 , at each timestep during the simulation . These vectors are then interpolated to a square grid , from which strain is evaluated using the finite difference discretization of: ϵ k l = 1 2 ( ∂ k u l + ∂ l u k ) , where k and l are in-plane spatial coordinates . The resultant stress is: σ k l = E s ν ( 1 + ν ) ( 1 - 2 ν ) δ k l ϵ m m + E s ( 1 + ν ) ϵ k l . ( 9 ) The traction stress is calculated using Tk = hs∂l σkl . The computed traction force vectors in the square grid are in excellent agreement with forces directly inferred from spring displacements in the triangular mesh ( S15 Fig ) . The strain energy density is given by U = 1 2 ϵ k l σ k l . For each simulation we calculate the mean strain energy as total strain energy transmitted to the substrate averaged over simulation time , T: ⟨ SE ⟩ = 1 T ∫ 0 T d t ∫ A d A h s U ( x , y , t ) . ( 10 ) Madin-Darby Canine Kidney ( MDCK . 2 ) cells ( ATCC , Manassas , VA ) were cultured in Eagle’s Minimum Essential Medium ( ATCC ) containing 10% fetal bovine serum ( GIBCO Life Technologies ) and 1% penicillin/streptomycin at 37°C and 5% CO2 in a humidified incubator . MDCK . 2 cells are stably transfected with a plasmid construct encoding for FTRActinEGFP ( a gift from Sergey Plotnikov , University of Toronto ) . Polyacrylamide gels are polymerized onto a glass coverslip at a ratio of 12%:0 . 086% polyacrylamide:bis-acrylamide to create a gel with an elastic modulus of 12 . 2 kPa [59] . After polymerization is complete , the polyacrylamide gels are reacted with 2mg/mL Sulfo-SANPAH ( Thermo Fisher Scientific ) and incubated with 1mg/mL Type 1 rat tail collagen ( Corning , high concentration ) for 2 hours in the dark [60] . Excess collagen is removed by rinsing with 1X Phosphate-buffered saline . Confluent cell monolayers were grown on a polyacrylamide gel substrate with an elastic modulus of 12 . 2 kPa . Wounds were formed by laser ablation of a single cell using a 435 nm wavelength laser ( Andor Technology , Belfast , Northern Ireland ) . Cell death causes monolayer retraction for ∼20 min after which the wounds close . | Many developmental processes involve collective cell motion , driven by migratory behaviours of individual cells and their interactions with the extracellular environment . An outstanding question is how cells regulate their internal driving forces to maintain tissue cohesiveness while promoting the requisite fluidity for collective motion . Progress has been limited by the lack of an integrative framework that couples cellular physical behavior with stochastic biochemical dynamics underlying cell motion and adhesion . Here we develop a cell-based computational model for collective cell migration during epithelial wound repair that integrates tissue mechanics with active cell motility , cell-substrate adhesions , and actomyosin dynamics . Using this model we show that an optimum balance of protrusive cell crawling and actomyosin contractility drives rapid directed motion of cohesive cell groups , robust to variations in cell and substrate physical properties . We further show that disparate modes of individual cell migration can cooperate to accelerate collective cell migration by fluidizing confluent tissues . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"velocity",
"stiffness",
"mechanical",
"properties",
"cell",
"motility",
"medicine",
"and",
"health",
"sciences",
"classical",
"mechanics",
"focal",
"adhesions",
"geometry",
"biological",
"locomotion",
"physiological",
"processes",
"developmental",
"biology",
"mathematics",... | 2018 | Cooperation of dual modes of cell motility promotes epithelial stress relaxation to accelerate wound healing |
Type 2 diabetes is characterized by insulin resistance of target organs , which is due to impaired insulin signal transduction . The skeleton of signaling mediators that provide for normal insulin action has been established . However , the detailed kinetics , and their mechanistic generation , remain incompletely understood . We measured time-courses in primary human adipocytes for the short-term phosphorylation dynamics of the insulin receptor ( IR ) and the IR substrate-1 in response to a step increase in insulin concentration . Both proteins exhibited a rapid transient overshoot in tyrosine phosphorylation , reaching maximum within 1 min , followed by an intermediate steady-state level after approximately 10 min . We used model-based hypothesis testing to evaluate three mechanistic explanations for this behavior: ( A ) phosphorylation and dephosphorylation of IR at the plasma membrane only; ( B ) the additional possibility for IR endocytosis; ( C ) the alternative additional possibility of feedback signals to IR from downstream intermediates . We concluded that ( A ) is not a satisfactory explanation; that ( B ) may serve as an explanation only if both internalization , dephosphorylation , and subsequent recycling are permitted; and that ( C ) is acceptable . These mechanistic insights cannot be obtained by mere inspection of the datasets , and they are rejections and thus stronger and more final conclusions than ordinary model predictions .
Insulin is the primary hormone in control of whole body energy metabolism in human beings . The hormone is secreted to the blood circulation by the β-cells , located in the islands of Langerhans in the pancreas . The adipose tissue and the adipocytes are important targets for insulin control of energy metabolism . Failure of the adipocyte and other target cells to properly respond to insulin , insulin resistance , is often associated with obesity and is a distinguishing feature of type 2 diabetes . Insulin controls cellular metabolism by binding to the insulin receptor ( IR ) at the surface of the cell ( reviewed in [1] ) . In response to insulin-binding the intracellular β-subunits of the transmembrane receptor , which carry protein kinase activity , autophosphorylate on specific tyrosine residues . Thus autophosphorylated , the IR is active against a set of intracellular signal mediator proteins , in particular the insulin receptor substrate-1 ( IRS1 ) , which becomes phosphorylated on tyrosine residues . Phosphotyrosines in IRS1 are recognized by proteins , containing a SH2-domain , which by binding to phospho-tyrosine become activated to transduce the insulin signal further downstream . The signaling eventually affects cellular metabolism , for example through an increase of glucose uptake or inhibition of lipolysis . Many of the downstream intermediary steps in the insulin signaling network of the target cells remain unidentified . However , also apparently well-characterized early aspects of insulin signal transduction remain incompletely understood and may thus also reveal novel features of importance for insulin action , both in normal and in disease states . At the plasma membrane of adipocytes , the IR has been shown to be localized in plasma membrane microdomains , invaginations of the membrane , referred to as caveolae [2] . It is important that in human fat cells , but for instance not in rat adipocytes , the IRS1 is co-localized with the IR in caveolae [3] . In conjunction with insulin-binding the IR is internalized by endocytosis [4] , [5] , but the function of IR endocytosis has not been demonstrated . It may be to turn off signaling , e . g . , by dephosphorylation of the receptor , by downregulating the number of IRs at the cell surface , or by clearing insulin from the circulation . Conversely , endocytosis may be a part of the signal transduction per se , e . g . , by gaining access to downstream signaling intermediates or by providing for compartmentalization of the signaling . It has not yet been possible to determine experimentally which of these alternatives that are of highest importance , at the various time-scales involved in the signaling [6]–[15] . Conversely , the insulin-controlled internalization of IR has been shown to depend on IR autophosphorylation [13] , [16] , [17] , but to be independent of downstream activation of IRS or phosphatidylinositol-3 kinase [17] . To gain further insight into which mechanisms that are most active during the early events of insulin signaling , we have measured the transient phosphorylation of IR and IRS1 during the first ten minutes after a step increase in extracellular insulin concentration . The mechanistic explanation to such transient data is typically not evident from a mere inspection of the time courses . Nevertheless , such data contains valuable information on the active mechanisms in a complex system , and measurements of rapid transient responses is one of the most widely used methods for characterization of technical systems [18] . In such studies , the information in the data is typically extracted from the data using a model based hypothesis testing approach . Such an approach is different from the kind of large-scale gray-box modeling approaches that typically are used in systems biology studies . Two such related models are [19] , [20] , and large-scale gray-box models are in general characterized by the fact that many more interactions are included than can be tested from the existing data . Conversely , in the hypothesis testing tradition followed here , we do not include all known mechanisms in the models . This typically corresponds to setting parameters to zero in a comprehensive model , and a key question is whether the included mechanisms are sufficient , necessary , or not sufficient , to explain the data . This gives information on which mechanisms that may , must , and cannot be significantly active during the specific time-scale . Apart from the overall methodology , the work also makes use of several non-trivial theoretical results and methods that can be re-used in other analyses of signaling systems .
We examined the extent of phosphorylation of IR and IRS1 on tyrosine residues in human adipocytes . In three separate experiments , data were collected at 10 time points during 15 min , following a step increase from 0 to 0 . 1 µM in insulin concentration ( Figure 1 ) . The experimental set-up is limited to measurements of relative changes , i . e . , all signals come with an unknown scaling factor . We measured phosphorylated and total IR and IRS1 by SDS-PAGE and immunoblotting . To achieve a robust measurement signal , the extent of phosphorylation of both IR and IRS1 were divided by total amount of IR and IRS1 , respectively . The resulting signals are therefore proportional to the relative degree of phosphorylation of IR and IRS1 . The rapid initial transient response was higher than the quasi-steady state level attained after about 5 min for both IR and IRS1 ( Figure 1 ) . This transient behavior is referred to as the overshoot in the data . The overshoot is clearly present both in each individual time course , and in their mean values . We now use a model based hypothesis testing approach , to translate these experimental observations to mechanistic insights . Three hypotheses are considered as possible mechanistic explanations to the observed overshoot . The first of these hypotheses , hypothesis A , assumes that the overshoot is generated by an interplay between the autophosphorylation and protein phosphatase activity at the plasma membrane only . It is interesting to consider the possibility whether such mechanisms might be the only ones significantly active in the IR signaling subsystem , since we are only considering the first few minutes of the response . The analysis shows that this possibility can be rejected based on the information in the collected data . While hypothesis B is like A with the additional possibility of endocytosis , hypothesis C is also like A but with the additional possibility of feedbacks from downstream signaling intermediates . There exist , indeed , in the literature suggestions of feedback from downstream signaling intermediates to , e . g . , the phosphotyrosine protein phosphatase activity [22]–[25] . We here suggest an archetypical version of such a feedback , to show that hypothesis C also provides an acceptable explanation of the data . The suggested model structure ( ℳf; Figure 3 ) includes activation of IRS1 and its subsequent activation of X , which refers to some non-identified downstream signaling intermediate . The notation X is chosen in order to illustrate the fact that it is impossible to conclude without further experiments which specific feedback that is most likely to generate the observed behavior in the experimental data , and only that any feedback of the given character would be sufficient . The feedback to PTP illustrates the archetypical feedback , which also could be illustrated by a direct feedback to the IR , by for instance its serine phosphorylation [26] . The agreement between this model structure and the data is just as convincing as that for the minimal model ℳi , b . Since it is sufficient that a single model structure from a given class produces a satisfactory explanation , in order for the whole class to be acceptable , we have now shown that hypothesis C is an alternative explanation to hypothesis B .
This paper has two parts . The first part reports a rapid overshoot in IR and IRS1 phosphorylation upon insulin stimulation of human fat cells . These observations , although interesting in themselves , do not provide any mechanistic insights by themselves , and mere inspection and reasoning around the data is not sufficient to evaluate which mechanisms that may and may not explain the given data in a satisfactory manner . The second part of the paper analyzes three biologically realistic and plausible mechanistic explanations: ( A ) direct phosphorylation and dephosphorylation of IR at the plasma membrane only; ( B ) the additional possibility of IR endocytosis; ( C ) the alternative additional possibility of feedback to IR from downstream intermediates . Our analysis has shown that A is not a satisfactory explanation , that B provides such an explanation if both internalization and subsequent recycling are included , and that hypothesis C provides such an explanation . The mechanistic insights obtained here are the result of model based hypothesis testing and there are some important properties of such studies that should be pointed out . In a hypothesis testing framework , the most interesting result is when a model may be rejected . A rejection is also the kind of conclusion that is hardest to achieve . Ideally , all parameter values in all model structures belonging to the class of model structures corresponding to the tested explanation should be evaluated before a rejection has been shown . Conversely , evidence of the sufficiency of a mechanistic explanation is shown already by the existence of a single model structure at a single parameter point which gives a satisfactory agreement . Further , a model rejection is a strong statement since it will not be altered when new data are collected ( unless , of course , the new data would point to errors in the previous data ) . The conclusions drawn here are thus not typical model predictions to be tested in validation experiments , but evaluations of possible mechanistic explanations for a given data set . The significance of this modeling approach becomes evident when comparing with a previous modeling work by Sedaghat et al . [20] . That model structure is an example of a large-scale mechanistically detailed model for insulin signaling , and it includes both internalization of the insulin receptor and feedback effects from downstream metabolic intermediates to IRS1 . Interestingly , the feedback signals do generate an overshoot in IRS1 phosphorylation . However , the Sedaghat model does not predict an overshoot in the IR phosphorylation , and must generally be revised to serve as a ( single ) explanation to our experimental data [27] . More importantly , however , the single model structure in [20] was evaluated at a single parameter point , and [20] is therefore a qualitatively different type of study than ours . A main drawback of such purely forward-simulation based studies is that most parameter values are unknown , especially in vivo . Analysis at a single parameter point is of course problematic if the chosen parameter values are unrealistic ( which is the case for instance for the internalization constant in [20] ) . However , also if all parameter values are realistic , one does not know which model predictions and parameter values ( i . e . active mechanisms ) are necessary consequences of the given data and model structure , and which model predictions are merely outcomes of more or less arbitrarily chosen parameter values . An example of a stronger model prediction is for instance that for ℳm , PTP herein , which says that all parameter values that give an acceptable agreement with our experimental data must also give a steady state concentration of ( IR⋅ins ) ⋅PTP larger than 25% . Finally , it should be noted that not even an ordinary model rejection , reporting a lack of agreement with the data , may be done without global searches among all realistic parameter values . There are also other related works . Interestingly , a transient overshoot in the phosphorylation of internalized IR has been reported [14] . However , that work did not provide any mechanistic explanations . A simulated model of signaling by the epidermal growth factor ( EGF ) receptor has been found to exhibit a transient phosphorylation overshoot when endocytosis of the receptor is included in the model [19] . However , the EGF receptor has a different mechanism of activation than IR , and there does not exist a thorough hypothesis testing approach that evaluates which mechanisms that may , and may not , produce such an overshoot . In a more recent time-course modeling of IR phosphorylation and endocytosis in Fao hepatoma cells [28] , no transient phosphorylation overshoot was included , neither in the experimental data , nor in the model . Further , the authors used the Akaike Information Criterion ( AIC ) hypothesis testing approach to distinguish between all possible model structures . The AIC simply chooses a model as the best one , by weighting model agreement against number of parameters . This means that AIC does not provide any statistical measure on whether any of the evaluated models show an acceptable agreement with the data , or what the statistical significance of the conclusions are . That means that the AIC test alone would not have been sufficient to find the main conclusions and rejections provided in this article . Our statistical testings are based on a number of assumptions . For instance , the noise in the system is approximated by white and Gaussian signals appearing exclusively in the measurements . This means that intrinsic system noise has been neglected , as have the indications that experimental noise from immunoblotting might be log-normal . To compensate for this limited complexity of the noise model , the variance of the noise has been exaggerated , and for many analyses only the most prominent features of the data ( primarily the overshoot ) have been used for the rejections . Other limitations in our assumptions are due to our usage of ordinary differential equations ( ODEs ) . This means that stochastic effects from individual particles , or individual cells , and subtle spatial phenomena ( everything besides the internalisation itself ) all are disregarded . These approximations have been judged acceptable since the available data do not allow for a more detailed inspection of the processes . It will be an important step forward in our understanding of these processes when we can measure data containing spatially resolved single cell data , and when we can more realistically describe processes in micro-environments such as caveolae , where IR and IRS1 are situated . So far , we can only speculate what the corresponding conclusions might be in such studies . For instance , the number of IR proteins per fat cell has been estimated to >2×105 [29] , and this should , according to generic studies such as [30] , mean that molecular stochastic effects are insignificant , at least if the assumption of fast diffusion within the cell is valid . However , when it comes to incorporating the caveolae micro-environment properties , the fundamental kinetics will probably change ( see e . g . [31] ) , and we have to-date no good guidelines for how such generalisations change the properties of a system . In any case , despite these limitations , statistical assessments of the degree of uncertainty underlying model rejections do provide more detailed and objective statements than those based on simulations and/or subjective judgments alone . Most importantly , we have been able to draw mechanistic insights from a given set of time-series data; these mechanistic insights could not have been drawn using only classical biochemical reasoning .
Samples of subcutaneous abdominal fat were obtained from female patients at the University Hospital of Linköping . Patients with diabetes were excluded . Pieces of adipose tissue were excised , during elective abdominal surgery and general anesthesia , at the beginning of the operation . The study was approved by the Local Ethics Committee and participants gave their informed approval . Rabbit anti-insulin receptor β-chain polyclonal and mouse anti-phosphotyrosine ( PY20 ) monoclonal antibodies were from Transduction Laboratories ( Lexington , KY , USA ) . Rabbit anti-IRS1 polyclonal antibodies were from Santa Cruz Biotech . ( Santa Cruz , CA , USA ) . Insulin and other chemicals were from Sigma-Aldrich ( St . Louis , MO , USA ) or as indicated in the text . Adipocytes were isolated by collagenase ( type 1 , Worthington , NJ , USA ) digestion as described [32] . At a final concentration of 100 µl packed cell volume per ml , cells were incubated in Krebs-Ringer solution ( 0 . 12 M NaCl , 4 . 7 mM KCl , 2 . 5 mM CaCl2 , 1 . 2 mM MgSO4 , 1 . 2 mM KH22PO4 ) containing 20 mM Hepes , pH 7 . 40 , 1% ( w/v ) fatty acid-free bovine serum albumin , 100 nM phenylisopropyladenosine , 0 . 5 U/ml adenosine deaminase with 2 mM glucose , at 37C on a shaking water bath . For analysis after 20–24 h incubation , cells were incubated at 37C , 10% CO2 in the same solution mixed with an equal volume of DMEM containing 7% ( w/v ) albumin , 200 nM phenylisopropyl adenosine , 20 mM Hepes , 50 UI/ml penicillin , 50 µg/ml streptomycin , pH 7 . 40 . Before analysis cells were washed and transferred to the Krebs-Ringer solution . Cells were then incubated at 37C with 100 nM insulin for the indicated time period . Cell incubations were terminated by separating cells from medium by centrifugation through dinonylphtalate . The cells were immediately dissolved in SDS and β-mercaptoethanol with protease and protein phosphatase inhibitors , frozen within 10 sec , and thawed in boiling water to minimize postincubation signaling modifications in the cells and protein modifications during immunoprecipitation [32] . Equal amounts of cells as determined by lipocrit , that is total cell volume , were subjected to SDS-PAGE and immunoblotting . After SDS-PAGE and electrotransfer membranes were incubated with indicated antibodies that were detected using ECL+ ( Amersham Biosciences ) with horseradish peroxidase-conjugated anti-IgG as secondary antibody , and evaluated by chemiluminescence imaging ( Las 1000 , Image-Gauge , Fuji , Tokyo , Japan ) . By two-dimensional electrofocusing ( pH 3–10 ) - SDS-PAGE analysis and immunoblotting against phosphotyrosine and IRS1 , >95% of the tyrosine phosphorylated 180-kD band was determined to represent IRS1 [33] . In this paper we evaluate three mechanistic hypotheses for the explanation of experimentally observed phosphorylation dynamics . Each of these three hypotheses are too general to correspond to a single mathematical model that can make specific predictions , which can be compared with data . For this reason we consider classes of model structures in our analysis . In practise , these classes are approximated by a large number of specific models . In this paper we restrict ourselves to the consideration of models described by ODEs . The general form of an ODE is given by ( 1A ) ( 1B ) where x∈ℝn is the n-dimensional column vector containing the state variables ( concentrations denoted by a square bracket ) , f is a well-behaved ( e . g . , continuous and differentiable ) function , p∈ℝr contains the parameters , and y contains the measurement signals whose relation to the state variables and the parameters is given by the function h . Spatial transport in the form of endocytosis and recycling is described by the introduction of compartment specific state variables , where the subscript i denotes state variables that have been internalized . All the models are uniquely given by the figures according to standard interpretation of such figures; examples and more details are included in the Text S1 . Models are sought to be rejected in several different ways . The first way is rejection through analysis of a corresponding transfer function form . Transfer functions are commonly used for linear models [34] , while the models considered here are nonlinear . Nevertheless , for the specific input studied ( a step function ) , we can find equivalent linear models giving exactly the same responses , i . e . , without approximations . This holds for all models accept ℳf , and to see it on a more general level , consider the systemwhere A ( ⋅ ) is an ℝn×n-valued function , x ( 0 ) ∈ℝn is the state vector , e . g . , concentrations of relevant substances , and u ( t ) is the input to the system . If u ( t ) changes from 0 to u0 at t = 0 , the state vector x ( t ) will follow the same trajectory as for the system ( 2 ) where δ ( t ) is the Dirac function . In other words , we can study the impulse response of a linear system instead . Taking the Laplace transform of Equation 2 yields ( 3 ) For instance , for ℳm , a we have the following state-space description ( derived in the Text S1 ) ( 4A ) ( 4B ) and the following transfer function description ( 5 ) with X ( s ) being the Laplace transform of . Note the pole in s = 0 , which means marginal stability . Biologically , this is due to the mass conservation , saying that [IR]+[IRp] is constant . For the same reason , all our considered model structures will contain a pole in s = 0 . Now , X ( s ) in Equation 5 can mathematically be interpreted as the step response of the transfer function ( 6 ) This allows us to transfer standard results from linear systems theory to our specific application . We have derived two general results that allow for rejection by direct inspection of the transfer functions , i . e . , without considering specific parameter values; these are presented in the following two subsections . Lemma 1 Consider a stable , linear time-invariant system with transfer function G ( s ) having real poles and no zeros . Then , the impulse response of G ( s ) is positive for all t>0 , i . e . , the system cannot display an overshoot . Proof . Since G ( s ) has real poles , we can write G ( s ) as a cascade of first-order transfer functions Gi ( s ) , i . e . , each with an impulse responsewhere H ( t ) is the Heaviside function , i . e . , H ( t ) = 1 if t≥0 , 0 otherwise . The lemma can now be proved by induction . Assume that has a positive impulse response yk ( t ) . Then the impulse response yk+1 ( t ) for satisfiesfor all t>0 . Now , since the step response of G ( s ) can be obtained by integration of the impulse response , it follows that if G ( s ) has only real poles and no zeros , its step response is monotonously increasing , which means that no overshoot may occur . For more details on conditions for positive impulse responses , see [35] . Consider the system Assume that it is stable and has real poles . As described above the impulse response of this system is equivalent to the step response of the model ℳm , c . The transfer function of the above system can be computed as Note that the system has a pole at s = 0 . Its impulse response equals the step response of Since this transfer function has no zeros and real poles , its step response does not display any overshoot , according to Lemma 1 . Therefore the final possibility would be that the overshoot in the models like ℳm , c would be generated by non-real poles . However , this generates damped oscillations , and this is not seen in the data . Nevertheless , to be sure that no erroneous conclusions are drawn because of this interpretation of the data , also the first models in Figure 2 have been rejected by a χ2 test . For those models where a transfer function analysis is not sufficient for rejection , specific parameter values are needed: these are determined by parameter optimization . The resulting model is thereafter subjected to statistical tests , primarily χ2 tests . Models can also be rejected if they are biochemically unrealistic in some other way , even though they show an acceptable agreement with the data . All models that can not be rejected in any of these ways are considered as acceptable explanations of the given data set . Further details on the parameter optimization and on the statistical testing are available in the Text S1 . Finally , note that even though all models except for ℳf may be analyzed using a transfer function study , this analysis gives a non-conclusive result for many more models than that , e . g . , because the models may produce an overshoot , but it is unclear what its shape may be; for all those models we applied the more general optimization and statistical testing approach . | Insulin is a central player in maintaining energy balance in our bodies and in type 2 diabetes , where the effect of insulin on its target tissues is diminished . Insulin acts on cells by binding to specific insulin receptors ( IRs ) at the cell surface . This triggers a series of events , including attachment of phosphate to IR , activation of downstream proteins that eventually mediate the signal to specific targets in the cell , and internalization of IR to the inner cytosolic part of the cell . The importance , time relations , and interactions between these events are not fully understood . We have collected experimental time-series and developed a novel analysis method based on mathematical modeling to gain insights into these initial aspects of how insulin controls cells . The main conclusion is that either IR internalization and the subsequent recycling back to the cell surface or feedbacks from downstream proteins ( or both ) must be significantly active during the first few minutes of insulin action . These conclusions could not have been reached from the experimental data through conventional biological reasoning , and this work thus illustrates the power of modeling to improve our understanding of biological systems . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"diabetes",
"and",
"endocrinology/type",
"2",
"diabetes",
"computational",
"biology/systems",
"biology",
"cell",
"biology/cell",
"signaling"
] | 2008 | Model-Based Hypothesis Testing of Key Mechanisms in Initial Phase of Insulin Signaling |
Helicobacter pylori ( Hp ) injects the CagA effector protein into host epithelial cells and induces growth factor-like signaling , perturbs cell-cell junctions , and alters host cell polarity . This enables Hp to grow as microcolonies adhered to the host cell surface even in conditions that do not support growth of free-swimming bacteria . We hypothesized that CagA alters host cell physiology to allow Hp to obtain specific nutrients from or across the epithelial barrier . Using a polarized epithelium model system , we find that isogenic ΔcagA mutants are defective in cell surface microcolony formation , but exogenous addition of iron to the apical medium partially rescues this defect , suggesting that one of CagA's effects on host cells is to facilitate iron acquisition from the host . Hp adhered to the apical epithelial surface increase basolateral uptake of transferrin and induce its transcytosis in a CagA-dependent manner . Both CagA and VacA contribute to the perturbation of transferrin recycling , since VacA is involved in apical mislocalization of the transferrin receptor to sites of bacterial attachment . To determine if the transferrin recycling pathway is involved in Hp colonization of the cell surface , we silenced transferrin receptor expression during infection . This resulted in a reduced ability of Hp to colonize the polarized epithelium . To test whether CagA is important in promoting iron acquisition in vivo , we compared colonization of Hp in iron-replete vs . iron-deficient Mongolian gerbils . While wild type Hp and ΔcagA mutants colonized iron-replete gerbils at similar levels , ΔcagA mutants are markedly impaired in colonizing iron-deficient gerbils . Our study indicates that CagA and VacA act in concert to usurp the polarized process of host cell iron uptake , allowing Hp to use the cell surface as a replicative niche .
Helicobacter pylori ( Hp ) is a mucosal colonizer that infects the stomachs of more than half of the world's population [1] . Chronic Hp infection is a major cause of gastric and duodenal ulcer disease , a risk factor for gastric cancer [2] , and recently has also been associated with iron deficiency anemia [3] , [4] . During colonization of the stomach , a significant number of Hp ( ∼20% ) adhere to the host cell surface via various adhesins [5]–[7] . We have previously reported that Hp can colonize and replicate directly while adhered to the epithelial surface , and can grow in this niche even in conditions where growth of the free-swimming bacteria is not supported [8] . The contact-dependent Hp virulence factor CagA , which is injected directly into host cells via the bacterium's type IV secretion system , plays an important role in enabling Hp colonization of the epithelium [8] . This occurs via a local perturbation of epithelial polarity , and can occur without gross disruption of epithelial integrity [8] . Since an important role of the epithelial barrier is to sequester and compartmentalize molecules that may be useful for colonizing microbes , we speculated that Hp has evolved specialized mechanisms to perturb cell polarity to acquire essential factors directly from the polarized epithelium . However , the nature of the factors transferred from the host cells to the bacteria and the molecular mechanisms involved remain unclear . Successful colonization of mucosal surfaces by bacteria implies an ability to extract essential micronutrients from their immediate environment , either from epithelial secretions near the cell surface , from the polarized host cells themselves , and/or from the interstitial side , across the epithelial cell layer . Iron is a micronutrient critical for the survival and growth of many mucosal colonizers and its availability controls expression of bacterial virulence factors in Hp and several other pathogens [9]–[15] . In the host however , free iron exists in extremely limited quantities , since it is sequestered from the mucosal surface through various mechanisms , including the epithelial barrier blocking access to the interstitium , binding of interstitial iron by transferrin , sequestration of intracellular iron by ferritin , and chelation of mucosal iron by lactoferrin [13] , [14] . While Hp is known to possess several iron uptake systems , the sources of iron that Hp utilizes during colonization of the gastric mucosa remain unclear [16] . Unlike other mucosal colonizers that possess siderophore-mediated mechanisms for uptake of iron [11] , Hp has not been shown to synthesize siderophores [16] . While the acidity of the gastric lumen releases iron from ingested food [17] , Hp is not found in the gastric lumen but rather colonizes the neutral environment of the epithelial cell surface and the overlying mucus layer [18] . In this microenvironment , iron is complexed with lactoferrin or with other glycoproteins found in the mucus [13] , [19] , [20] . Hp is unable to compete with partially saturated lactoferrin for iron acquisition [21] , and its ability to obtain iron complexed with mucus glycoproteins is unknown . In the interstitium , iron is tightly bound to transferrin and Hp cannot compete with partially saturated transferrin for iron [21] . However , Hp is able to utilize iron from fully saturated transferrin [21] , which is the major form endocytosed by epithelial cells [22] . In this study , we utilize a model polarized epithelium to show that Hp colonizing the apical surface are able to acquire iron from host epithelial cells . We also show that the Hp virulence factors CagA and VacA work in concert to shift the basolateral transferrin/transferrin receptor recycling process apically , directing transferrin and its receptor to sites of microcolony formation on the cell surface . Silencing expression of the transferrin receptor interferes with the colonization of the epithelium by Hp . Finally , we show , using a Mongolian gerbil model of Hp infection , that host iron depletion results in a decreased ability of CagA-deficient Hp to colonize the gastric niche .
We previously reported that wild type ( WT ) Hp are able to colonize the apical cell surface of polarized MDCK monolayers when the apical medium bathing the cells contains only DMEM , a medium that cannot support Hp growth in vitro [8] . However , CagA-deficient Hp ( ΔcagA ) are defective in colonizing this niche . This suggested that we could use this system to identify host factors obtained by WT Hp but unavailable to ΔcagA by enriching the apical medium with specific nutrients and testing which would rescue the growth defect of ΔcagA . We found that addition of iron in the form of ferric chloride to the apical chamber partially rescued the growth defect of ΔcagA in a saturable and dose dependent manner ( Figure 1 ) . In contrast , exogenous iron in the apical chamber of monolayers infected with WT did not lead to increased growth of the bacteria , indicating that rescue of the mutant is not due to a general enhancement of growth by a mechanism unrelated to CagA ( Figures 1A and 1B ) . Imaging of bacteria adhered to the polarized monolayer revealed that iron added to the apical chamber enabled the growth of ΔcagA microcolonies adhered to the epithelium ( Figure 1C ) . In addition , iron added to DMEM was not sufficient to sustain Hp in vitro ( Figure S1A ) , suggesting that Hp colonizing the cell surface obtain not just iron , but also other essential factors through interaction with host cells . The rescue of ΔcagA growing on the cell surface by iron suggests that CagA affects host epithelial cell function to allow Hp access to micronutrients that are found in the epithelium or across its barrier . Host epithelial cells acquire iron largely via transferrin receptor-mediated endocytosis [22] . In the serum where transferrin is normally found , the population of transferrin molecules is 20% – 40% saturated with iron [23] , [24] . We thus wondered whether Hp growing on the cell surface gain access to interstitial transferrin and utilize this form of iron for survival . However , in vitro reports have shown that Hp cannot grow on partially saturated transferrin [21] . We found that Hp growth on the cell surface was inhibited by the presence of partially saturated transferrin in the apical medium , indicating that Hp is unable to compete with partially saturated transferrin for iron even in the presence of cells ( Figure 2A ) . Since Hp is susceptible to iron chelation by transferrin , and because transferrin is the major chelator of iron in extracellular fluids of the interstitial space , these results suggest that CagA's mechanism of action is not simply to injure the epithelium and provide paracellular diffusion of transferrin . We wondered if , instead , Hp utilizes the epithelial barrier as a shield from noxious macromolecules in the interstitial space , and whether the bacteria can actively obtain nutrients from the epithelium . To test this , we added transferrin to the basolateral co-culture medium at a concentration sufficient to chelate serum iron and inhibit Hp growth in broth ( Figure S1B ) , and determined if the basal transferrin would inhibit growth of the apical bacteria . As shown in Figure 2B , transferrin at a concentration that quickly kills Hp in the apical chamber had no inhibitory effect across the polarized epithelium , indicating that Hp interaction with the apical cell surface allows the bacteria to utilize the epithelium as a filter for iron . Epithelial cells selectively acquire transferrin saturated with iron ( holotransferrin ) via endocytosis , as its affinity for the transferrin receptor is 2000X greater than iron-free transferrin [22] . Since Hp growing on the cell surface are protected from the toxic effects of partially saturated transferrin by the epithelial barrier , we wondered whether the bacteria can obtain iron from holotransferrin , the form of transferrin taken up into cells . A recent report suggested that , in a chemically defined in vitro media system , Hp can obtain iron from holotransferrin , but not partially saturated transferrin [21] . We confirmed these results and tested whether holotransferrin affected the growth of Hp microcolonies on the cell surface by adding it to the apical medium of the Transwell culture system . Not only was holotransferrin not toxic to WT microcolonies , addition of holotransferrin to the apical chamber in fact led to partial rescue of ΔcagA ( Figures 2C and 2D ) . These experiments indicate that Hp growing on the cell surface are able to utilize iron from holotransferrin , even though they cannot compete with partially saturated transferrin for iron , and suggest the hypothesis that acquisition of holotransferrin from within host cells is one mechanism by which Hp acquire iron during mucosal colonization . Since CagA has multiple effects on epithelial physiology and appears to aid Hp in iron acquisition from host cells , we asked if Hp colonization of the apical cell surface affects host cell transferrin recycling . To study this , we utilized MDCK cells stably expressing human transferrin receptor [25] , as this allowed us to visualize and quantify transferrin binding to its receptor and its uptake through the use of human transferrin conjugated to a fluorophore , which is readily available ( Figure S2A ) . These cells stably expressing human transferrin receptor formed polarized monolayers , and WT Hp colonized the apical cell surface while ΔcagA exhibited a 100X defect in colonization ability , as seen with untransfected MDCK cells ( Figure S2B ) . Polarized MDCK cells stably expressing human transferrin receptor were either left uninfected or infected with WT or ΔcagA for two days before fluorescent transferrin assays were performed . In this assay , internalization of transferrin was synchronized by first adding transferrin to the basal chamber on ice . This allowed basolateral binding of transferrin to its receptor , but inhibited its uptake . Unbound transferrin was subsequently washed away , and the cells were warmed to 37°C to allow endocytosis to proceed . As expected , endocytosis was inhibited on ice and transferrin bound to the basolateral membranes of the polarized cells ( Figure 3A , top panels ) [22] , [25] . There was no significant difference in the amount of transferrin bound in WT vs . ΔcagA-infected monolayers ( Figure 3A , top panels ) . Also , by immunoblot , the expression level of the transferrin receptor under the different conditions was similar ( Figure 3B ) . However , when endocytosis was allowed to proceed at 37°C for 30 minutes , significantly higher amounts of transferrin were observed inside the WT-infected monolayers as compared to uninfected or ΔcagA-infected monolayers ( Figures 3A and 3C ) . We found similar results in Caco-2 cells ( human colon carcinoma cells ) ( Figure S3A ) , indicating that these results are recapitulated in multiple epithelial lines . To determine if this difference is due to CagA vs . the density of colonizing bacteria on the cell surface , we repeated this experiment while adding exogenous iron to the apical chamber to rescue the growth defect of ΔcagA . In these conditions , the numbers of WT and ΔcagA growing on the cell surface were similar , yet the same difference in internalized transferrin was observed ( Figures S3B and S3C ) . Finally , complementation of ΔcagA with the cagA gene ( CagA* ) led to restoration of the phenotype of increased transferrin internalization on infection with the bacteria ( Figure 3C ) . These experiments suggest that CagA delivery into host cells increases the amount of internalized transferrin . Once injected into the host cell , CagA is tyrosine phosphorylated by host Src- and Abl-family tyrosine kinases at several repeated sites in the C-terminal end containing EPIYA motifs [26]–[28] . CagA then acts as an adaptor protein that stimulates signaling downstream of growth factor receptor tyrosine kinases [29]–[32] . Since growth factor signaling increases transferrin uptake [33] , we studied whether CagA phosphorylation is necessary for its effects on transferrin internalization . We found that the ability of CagA to increase internalized transferrin depends on the presence of the EPIYA motifs , as infection of polarized monolayers with a mutant lacking these phosphorylation domains resembled ΔcagA infection ( Figure 3C ) . These results indicate that CagA injected by Hp microcolonies on the apical cell surface increases transferrin internalization through receptor tyrosine kinase-like signaling , and suggest that this leads to an increased availability of iron for the colonizing bacteria . However , under normal conditions , transferrin should not be released to the apical side of an epithelium , since its recycling is confined to the basolateral membrane where the receptor is exclusively found [22] . This suggests that infecting bacteria perturb not just uptake , but also localization of the transferrin/transferrin receptor complex . Hp is known to affect host cell polarity and intracellular trafficking [34]–[38] , and our previous study showed that perturbation of host cell polarity is involved in enhancing colonization of the polarized epithelium [8] . We therefore wondered if Hp colonization might lead to mis-sorting of the transferrin receptor , and hence transferrin and possibly iron , to sites of bacterial microcolony growth on the apical surface . To address this , we fixed uninfected or infected polarized MDCK monolayers in conditions that do not permeabilize the membrane and then applied antibodies against the transferrin receptor only to the apical side . In this manner , we can detect whether small amounts of transferrin receptor are found at the apical membrane without detecting internalized or basolateral transferrin receptor . We observed distinct puncta of immunolabeled transferrin receptor localized at the apical membrane near the Hp microcolonies ( Figure 4A ) . This mislocalization of the transferrin receptor does not occur immediately after bacterial attachment to the apical surface , since monolayers fixed after 5 minutes of infection did not show puncta of transferrin receptor on the apical side ( Figure S4A ) . This implies that Hp can affect host cell polarity locally to mislocalize basolateral proteins to sites of microcolony growth . To confirm these findings , we used a different technique that allows selective biotinylation of surface proteins of the polarized epithelium ( Figure 5A ) [39] , [40] . We selectively biotinylated the basolateral surface on ice , and then allowed internalization and recycling of the biotinylated basolateral proteins for 30 minutes at 37°C . We then stained the apical surface without permeabilization with fluorophore-conjugated streptavidin to detect mislocalized basolateral proteins . As with the previous results , we found that Hp microcolonies are associated with membrane patches containing basolateral proteins that were mis-sorted to the apical membrane ( Figures 5B and 5C , bottom panels , and Figure 5D ) . To confirm that endocytosis is required , we repeated the staining on infected monolayers that were fixed immediately after surface biotinylation on ice . These monolayers did not contain apical biotin staining ( Figures 5B and 5C , top panels , and Figure 5D ) . To determine whether this phenomenon is generalizable to other polarized cell models , we repeated the staining of transferrin receptor in Caco-2 cells . We obtained similar results as with MDCK cells , indicating that Hp can induce mis-sorting of the transferrin receptor in multiple epithelial lines ( Figure S5 ) . Finally , to determine whether mislocalization of basolateral proteins is restricted to a subset of proteins or affects all basolateral proteins , we used antibodies to other basolateral markers . E-cadherin is a cell-cell adhesion molecule that is normally absent from the apical membrane . Using apical staining of non-permeabilized cell monolayers with E-cadherin antibodies , we had previously shown that E-cadherin is absent from most of the apical membrane except for sites of cell extrusion [41] . When we applied these antibodies to Hp-infected monolayers , we did not find E-cadherin associated with Hp microcolonies ( Figure S4B ) , indicating that not all basolateral proteins are mislocalized during Hp colonization . Examination of monolayers infected with ΔcagA indicated that CagA-deficient bacteria are less efficient but still able to recruit transferrin receptor to the sites of bacterial microcolonies ( Figure 4 ) . Quantification of the amount of transferrin receptor observed apically at the sites of ΔcagA microcolonies showed that the median amount ( 514 arbitrary units/bacterium ) was less than that observed with WT ( 774 arbitrary units/bacterium ) , but this was not statistically significant ( Figure 4B ) . The ability of ΔcagA to still cause mislocalization of host cell transferrin receptor apically to sites of bacterial microcolonies suggested that other bacterial factor ( s ) may be involved in this phenomenon . Another major virulence factor of Hp , vacuolating cytotoxin ( VacA ) , has been reported to interfere with the endocytic pathway of host cells [38] . We therefore decided to test the role of VacA in the mislocalization of transferrin receptor . We constructed a VacA-deficient Hp mutant ( ΔvacA ) , and found that in monolayers infected with ΔvacA , transferrin receptor was absent from the sites of bacterial microcolonies on the apical cell surface ( Figure 4 ) . This result was confirmed in infected Caco-2 cell polarized monolayers ( Figure S5 ) . We also complemented ΔvacA with the vacA gene , and found that this reconstitution ( VacA* ) restored the ability of the bacteria to mislocalize transferrin receptor apically to sites of bacterial microcolonies ( Figure 4B ) . Selective biotinylation of basolateral epithelial proteins and quantification of the amount of biotin subsequently associated with apical microcolonies indicated that both ΔcagA and ΔvacA had significantly less biotin associated with microcolonies than WT ( Figure S6 ) . This suggests that both CagA and VacA are involved in disruption of polarity , and that the transferrin receptor is not the only molecule that is mislocalized apically . We also found that a mutant deficient in both CagA and VacA ( ΔcagAΔvacA ) was still able to mislocalize proteins to the sites of bacterial microcolonies ( Figure S6 ) , implying that CagA and VacA are not the only bacterial factors involved in this process , although they do appear to play major roles . Together , these results indicate that Hp colonization of the apical cell surface leads to mislocalization of a subset of basolateral proteins to the apical cell membrane at sites of bacterial microcolony growth , with the Hp virulence factors CagA and VacA playing major roles in this process . One of the proteins mislocalized to these sites is the transferrin receptor , and its mislocalization is primarily dependent on VacA . Given the role of VacA in transferrin receptor mislocalization , we next tested whether VacA affects the ability of Hp to colonize the apical cell surface of a polarized epithelium . We observed that ΔvacA have approximately a 10X decrease in bacterial counts at day 5 post-infection , as compared to WT ( Figure 6A ) . In the presence of rich media in the apical chamber , ΔvacA grow as well as WT , indicating that this phenotype is not due to an in vitro growth defect of the mutant ( Figure S7 ) . To determine whether the decrease in bacterial counts correlates with a defect of colonization of the apical cell surface , we examined microcolony formation on polarized cells infected with ΔvacA by confocal immunofluorescence microscopy . Initial adherence of ΔvacA to the apical cell surface was no different from WT , with an average of 9 bacteria adhered/100 cells in each case ( p = 0 . 5 ) . However , ΔvacA formed significantly smaller microcolonies on the cell surface at day 5 post-infection ( Figures 6C and 6D ) . Complementation of ΔvacA ( VacA* ) led to restoration of the ability of the bacteria to effectively colonize the polarized epithelium , with the formation of large microcolonies ( Figure 6D ) . We also tested whether the ΔvacA mutant is defective in iron acquisition from the host , by supplementing the apical media with iron . Exogenous addition of iron to the apical chamber led to rescue of the colonization defect shown by ΔvacA ( Figure 6B ) . To examine if CagA and VacA act in similar or different pathways in aiding Hp in cell surface colonization , we tested the mutant deficient in both CagA and VacA ( ΔcagAΔvacA ) . This double mutant resembled the single CagA-deficient mutant , and exogenous addition of iron apically partially rescued its ability to colonize the polarized epithelium ( Figures 6E and 6F ) . The phenotype observed is not due to the mutant having an in vitro growth defect , as growth of this double mutant was similar to WT in the presence of rich media added to the apical chamber ( Figure S7 ) . These results indicate a role for VacA in Hp colonization of the polarized epithelium , and suggest that CagA and VacA work in concert to aid Hp acquisition of iron from , and colonization of , host polarized epithelial cells . The data presented above indicate that the Hp virulence factors CagA and VacA both affect normal recycling and trafficking of the transferrin/transferrin receptor complex , which is the major iron uptake mechanism of epithelial cells [22] . Furthermore , mutants in both virulence factors are defective in colonizing the polarized cell surface and these defects can be partially rescued by exogenous addition of iron . This suggests that Hp perturbation of the transferrin/transferrin receptor recycling pathway might be used by the bacteria to obtain iron from host cells . To directly test if the transferrin receptor pathway is important for Hp colonization of the polarized epithelium , we silenced expression of the transferrin receptor during infection and asked whether this affects microcolony growth . We designed siRNAs directed against the canine transferrin receptor and selected two that produce very effective knockdown of transferrin receptor expression ( Figure 7A ) . Cells transfected with a mixture of these two siRNAs were seeded on Transwell filters and allowed to polarize before infection with WT . A first observation was that Hp was able to recruit the minimal amount of transferrin receptor still expressed by the host cells to the sites of bacterial microcolonies ( Figure 7B ) . More importantly , Hp formed significantly smaller microcolonies on the apical cell surface of monolayers where transferrin receptor expression had been knocked down , as compared to monolayers transfected with a control siRNA ( eGFP ) ( Figures 7C and 7D ) . We made use of the fact that the siRNAs directed against the canine transferrin receptor are highly specific and do not cross-react with human transferrin receptor ( Figure S8A ) to test that the phenotype observed was not due to off-target effects of the siRNAs . In MDCK cells stably expressing human transferrin receptor , knockdown of endogenous canine transferrin receptor expression left expression of human transferrin receptor intact ( Figure S8A ) . Hp allowed to colonize the apical surface of these cells formed microcolonies similar in size to those formed by Hp colonizing cells transfected with a control siRNA ( Figure S8B ) . This indicates that the decreased ability of Hp to colonize the apical cell surface after knockdown of transferrin receptor expression in MDCK cells is specifically due to decreased expression of transferrin receptor in those cells . Collectively , these results show that host cell transferrin receptor is functionally important in enabling Hp colonization of the apical surface of a polarized epithelium . They also suggest that CagA and VacA-mediated perturbation of transferrin/transferrin receptor recycling allows Hp to acquire iron from the host cells . Our data suggest a model in which Hp colonization of the apical cell surface leads to mis-sorting of a subset of the transferrin/transferrin receptor complex and transcytosis of the complex from the basolateral to the apical surface at the sites of bacterial microcolony growth . We determined whether transferrin is transcytosed apically by adding biotinylated transferrin to the basolateral media and then assaying for the presence of biotinylated transferrin in the apical chamber after a 24-hour incubation period . We detected a 1 . 5-fold increase in the amount of biotinylated transferrin in the apical chamber of WT-infected monolayers as compared to uninfected monolayers ( Figure 8 and Figure S9 ) . This increase was dependent on CagA , since monolayers infected with ΔcagA had similar amounts of biotinylated transferrin in the apical chamber as uninfected monolayers ( Figure 8 ) . To determine that this increase was due to transcytosis , and to control for possible paracellular leakage of macromolecules , we also added biotinylated albumin to the basolateral chamber . In contrast to transferrin , the small amount of biotinylated albumin detected in the apical chamber was the same irrespective of whether the monolayers were infected with WT or ΔcagA , or left uninfected ( Figure 8 and Figure S9 ) . Infection with the complemented ΔcagA ( CagA* ) resulted in restoration of the phenotype of increased transferrin transcytosis ( Figure 8 ) . These findings indicate that Hp colonization of the apical cell surface results in transcytosis of transferrin to the apical side of the cell . Hp colonizes multiple niches in the stomach ( i . e . free-swimming in the mucus layer vs . the cell surface ) , and since CagA-negative strains are common in nature , it is likely that in vivo , Hp utilizes multiple modes of iron acquisition . However , our findings suggest that CagA may be more important for the bacteria when colonizing hosts that are iron-depleted or during conditions of poor dietary iron content . To address this question , we utilized the Mongolian gerbil model of Hp infection to determine if the iron status of the host would affect WT or ΔcagA colonization . Mongolian gerbils were maintained either on a regular diet containing 250 ppm iron , or an iron-deficient diet containing <1 ppm iron , for 3 weeks prior to infection , and then through the duration of the infection ( see Materials and Methods for details ) . Dietary restriction of iron has been shown to result in decreased host iron levels in mice [42]–[44] , and we verified that the animals placed on the iron-deficient diet had reduced iron levels , as measured by inductively coupled plasma-mass spectrometry of liver samples ( Figure S10A ) . For these infections , we used the Hp strain 7 . 13 , which has been previously shown to reproducibly colonize the Mongolian gerbil stomach , is able to deliver CagA into host cells , and whose isogenic ΔcagA mutant exhibits a defect in colonization of the polarized epithelium in vitro , similar to the Hp strain G27-MA [8] , [45] . In conditions of growth in nutrient-rich broth , 7 . 13 WT and ΔcagA grow with similar kinetics ( Figure S10B ) . We infected Mongolian gerbils with either 7 . 13 WT or ΔcagA , and examined bacterial loads at 6 or 8 weeks after infection . 7 . 13 WT colonized iron-replete and iron-deficient animals to similar levels ( Figure 9 ) . The isogenic ΔcagA mutant also colonized iron-replete animals to a similar level as WT . However , 7 . 13 ΔcagA showed a significant decrease in colonization levels in the stomachs of iron-deficient animals , both in comparison to WT in iron-deficient animals and in comparison to ΔcagA in iron-replete animals ( Figure 9 ) . We also found that bacteria could not be recovered from 9/16 animals in the ΔcagA-infected iron-deficient group of animals , unlike the WT-infected iron-deficient group , in which Hp could be recovered from all 14 animals infected , at the end of the experiment ( Figure 9 ) . These results mirror our in vitro findings that CagA plays an important role in enabling effective acquisition of iron from the host during Hp colonization of its gastric epithelial niche .
Bacterial virulence factors are defined as molecules associated with disease . The major virulence factors of Hp , CagA and VacA , are epidemiologically linked to disease and possess multiple biological properties that can be deleterious to host cells [46]–[49] . However , understanding how these molecules function during an infection requires asking not just how virulence factors disrupt the host cell , but also how such effects benefit the bacteria . Our previous study had established that Hp can grow as microcolonies attached to the apical cell surface of a polarized epithelium , even in conditions where the free-swimming bacteria are rapidly killed [8] . CagA helps the bacteria form microcolonies and exploit this niche by perturbing cell polarity [8] . Here , we extend the concept that bacteria perturb host cell polarity to use the apical cell surface as a replicative niche . We show that bacterial virulence factors alter polarized host intracellular trafficking , suggesting a novel mechanism by which the bacteria are able to acquire essential micronutrients from host cells and colonize the apical cell surface . We found that exogenous iron added to the media bathing bacterial microcolonies on the apical cell surface partially rescues the Hp ΔcagA mutant growth defect . This suggests that one of the factors Hp extract from epithelial cells is iron , and that one of CagA's benefits to the bacteria involves facilitating iron acquisition from the epithelium . Furthermore , Hp is able to colonize the apical membrane of the epithelium even in the presence of excess iron chelators bathing the interstitial side of the epithelium , suggesting that it can acquire iron directly from the host cells without gross disruption of epithelial integrity . We note that attachment of Hp to non-polarized host cells had previously been reported to result in upregulation of expression of several annotated iron uptake proteins [50] . Because uptake of iron into host epithelial cells is a polarized process , occurring largely basolaterally through the transferrin/transferrin receptor recycling pathway [22] , we probed whether this pathway is manipulated by Hp during colonization of the cell surface . In the presence of an intact epithelial barrier , microbes on the apical surface do not have access to interstitial iron-bound transferrin or its recycling . In addition , partially saturated transferrin , which is the form found in the interstitium , is toxic to Hp [21] . However , Hp is able to utilize iron from holotransferrin [21] , which is the form that eukaryotic cells preferentially bind and uptake due to its higher affinity for the transferrin receptor , as compared to partially saturated or iron-free transferrin [22] , [51] . This suggested that Hp may be able to utilize the epithelium both as a barrier against the toxic effects of partially saturated transferrin , and as a source of holotransferrin . We observed that the apical cell surface colonization defect of ΔcagA mutants can be partially rescued by addition of holotransferrin to the apical chamber , suggesting that intracellular holotransferrin could be one possible iron source for colonizing bacteria . For Hp microcolonies to utilize holotransferrin as an iron source without destroying the epithelium , polarized uptake and recycling of transferrin would have to be perturbed . Both CagA and VacA have biological properties that could be involved in this process . For example , CagA is known to be able to induce receptor tyrosine kinase ( RTK ) -like signaling [29]–[31] , and growth factor RTK activation has been shown to increase uptake of transferrin in other models [33] . CagA has also been shown to affect cell polarity and the assembly of the epithelial junctions [35]–[37] , [45] , both of which could influence sorting of basolateral molecules [52]–[54] . VacA , another major virulence factor of Hp , has previously been reported to affect endosomal trafficking [38] . We therefore hypothesized that CagA and VacA could have effects on the host cell transferrin/transferrin receptor recycling pathway . We found that CagA injected into host cells by Hp growing as microcolonies on the apical cell surface increased internalized transferrin . This required signaling via the EPIYA motifs in the C-terminus of the protein , which are important in the activation of RTK-like signaling [29] , [30] , [32] . We also observed that a subset of the transferrin receptor population is mislocalized from the basolateral surface specifically to sites of bacterial microcolonies on the apical cell surface . In our model system , this mislocalization is largely dependent on the action of VacA , as a ΔvacA Hp mutant failed to recruit transferrin receptor apically . We were able to show a direct involvement of the transferrin/transferrin receptor recycling process in the colonization of the polarized epithelium by Hp , since siRNA knockdown of transferrin receptor expression resulted in decreased growth of Hp microcolonies on the cell surface . Our model suggests that Hp colonizing the apical cell surface induce mis-sorting of a subset of the transferrin/transferrin receptor complex apically . In accord with this , we observed significantly increased transcytosis of transferrin from the basal to the apical compartment and its release into the apical media when the epithelium is infected by WT Hp . Several important questions remain regarding this model . First , it is clear that iron acquisition from the epithelium is only one of several mechanisms of iron acquisition by Hp . Hp can live as a free-swimming population in vitro without the need for contact with epithelial cells . In vivo , Hp may obtain iron through several mechanisms and from various sources . For example , Hp residing as free-swimming bacteria in the mucus layer may obtain iron conjugated in mucus glycoproteins , or perhaps from the release of dietary iron by the acidic stomach lumen , or from inflammatory exudates , although these potential sources need study . Of note is that in addition to its role in causing peptic ulcers and its association with gastric cancer , Hp infection has recently been associated with iron deficiency anemia unrelated to blood loss [3] . Indeed , the latest Maastricht Consensus Report recommends diagnosis and treatment of Hp in cases of unexplained iron deficiency anemia [55] . There are now several reports in the literature showing that Hp eradication improved or cured previously unexplained cases of iron deficiency anemia [4] , [56] , [57] . Second , it will also be important in future studies to directly visualize the transfer of iron from the host to the bacteria and the precise molecular mechanisms involved . The observation that specific basolateral molecules , such as the transferrin receptor , are highly concentrated at sites of bacterial microcolonies in comparison to the rest of the apical membrane suggests that these sites are a very specialized and enriched microenvironment . However , methods that allow the visualization and quantification of iron in a spatially defined manner at sub-micrometer resolution are still at early stages of development [58] , [59] . Technical limitations such as these , as well as difficulties associated with maintaining a polarized monolayer in which the transferrin receptor or other important host molecules have been downregulated over prolonged periods of time , will first have to be overcome before the exact mechanisms by which Hp extracts iron from the host epithelium can be fully understood . Several of the known effects of CagA and VacA on host cell physiology may have important roles in altering trafficking of transferrin and other molecules for the benefit of colonizing bacteria . For example , we showed here that signaling through the EPIYA motifs of CagA increase host internalization of transferrin . CagA also has multiple effects mimicking growth factor activity which could influence the way in which gastric epithelial cells uptake and process nutrients that could be used by colonizing Hp [29]–[31] , [60] . In addition , we had previously reported that CagA's perturbation of host cell polarity via its actions on Par1b are important in enabling Hp colonization of the apical cell surface [8] . Of note , one of Par1b's functions is in organizing microtubules , and it has been speculated to play a crucial role in regulation of intracellular vesicle transport [61] . In addition to its effects on endosomal trafficking , VacA has been shown to act as a pore on membranes , and may independently increase transcellular and paracellular flow of iron and other essential factors [48] , [62] , [63] . We did not observe transferrin receptor mislocalization to sites of bacterial attachment in preliminary experiments with purified VacA added exogenously to ΔvacA-infected polarized cell monolayers ( data not shown ) . The delivery of VacA can be local through contact by adhered bacteria [64] , as well as at a distance through diffusion of soluble VacA [48] . Since localized VacA delivery during cellular infection may not be equivalent to generalized intoxication , it will be interesting to define how these two forms of delivery differ in their effects on polarized epithelia . It is also possible that VacA is necessary , but not sufficient , for this process . The potential cooperation between CagA and VacA to affect polar transport of micronutrients to colonizing microcolonies is exciting , since CagA and certain VacA alleles have been linked genetically [65] , and possible interplay between their functions have been reported [66]–[70] . Different subtypes of CagA and VacA may also have varied activities in aiding Hp colonization of the epithelium . For example , the “East Asian” form of CagA has been reported to bind more strongly to Par1b than the “Western” form of CagA [71] . How do such differences translate to the ability of different strains of the bacteria to colonize the epithelium ? Our findings suggest that CagA and VacA may act in concert in a novel way of inducing the host epithelium to relinquish essential micronutrients that does not necessitate destroying the epithelium or causing gross leakage of interstitial macromolecules into the lumen . Further studies will reveal the exact pathways that are usurped by CagA and VacA in altering host transferrin trafficking , visualize how iron is delivered to Hp on the cell surface , and perhaps uncover additional iron acquisition methods by Hp . Our use of a simplified experimental system allowed us to focus on the microenvironment of the apical cell surface as a replicative niche , and to uncover one possible mechanism by which Hp may obtain iron from host epithelial cells . We subsequently found that ΔcagA mutant Hp have a decreased ability to colonize the stomachs of Mongolian gerbils that are iron-deficient whereas WT bacteria do not , indicating that CagA aids Hp in acquiring iron from the host during in vivo bacterial colonization . Our findings also raise the question of how host micronutrient levels impact the pathogenesis of colonizing Hp . For example , might decreased levels of micronutrients in the host lead to increased pathogenicity of the bacteria ? We note here that expression of both cagA and vacA have been reported to be upregulated in conditions of iron starvation in vitro [9] , [72] , and that the ferric uptake regulator ( Fur ) protein indirectly regulates vacA expression [72] , [73] . Previous epidemiological data had suggested an association between iron deficiency and gastric cancer , although this data predates our understanding of the role of Hp infection [74] . It will now be interesting to examine epidemiologically the possible roles of CagA and VacA in the context of iron deficiency in human populations . We suggest that manipulation of host epithelial polarity is akin to utilizing the epithelium as a filter to acquire essential micronutrients while maintaining a barrier that protects the microbes from innate immune defenses present in the interstitial space . Perturbation of host epithelial polarity by Hp appears to be a specific and subtle process , not due to a generalized loss of epithelial polarity , since biotinylated proteins from the basolateral membrane form patches specifically under apical bacterial microcolonies , and not all basolateral proteins are found in these membrane patches . Although we focused on transferrin receptor mislocalization in this study , the technique of basal biotinylation of host membrane proteins in infected monolayers suggests that there are multiple proteins that are mislocalized to the apical sites of bacterial growth . The identity of these proteins may shed other important insights into the way microbes colonize the epithelial surface . CagA and VacA play major roles in this mislocalization , but other bacterial factors do appear to be involved as well . The ability to manage host epithelial polarity is emerging as an important facet of bacterial-host interactions . For example , Listeria monocytogenes has been shown to take advantage of cell polarity changes for invasion [41] , [75] , and mislocalization of basolateral proteins to the apical cell surface as a result of bacterial infection has also been described for other pathogens such as enteropathogenic Escherichia coli and Pseudomonas aeruginosa [76] , [77] . What are the mechanisms by which specific basolateral molecules are sequestered to the sites of bacterial microcolonies ? During P . aeruginosa infection of polarized epithelia , phosphatidylinositol 3 , 4 , 5-trisphosphate ( PIP3 ) , a basolateral membrane lipid important in cellular signaling and polarity , is mislocalized to the sites of bacterial attachment on the apical cell surface [77] , [78] . PIP3 is an intriguing candidate in the context of our results with the transferrin receptor , as it has recently been shown to also localize to recycling endosomes , and to be important for the proper sorting of recycling cargo , such as the transferrin receptor [79] . Furthermore , both CagA and VacA have been implicated in the activation of the phosphatidylinositol 3-kinase pathway , which catalyzes production of PIP3 [80] , [81] . It will be interesting to test if PIP3 is recruited to sites of Hp microcolonies on the apical cell surface , and whether PIP3 function is involved in Hp's colonization of the polarized epithelium . In summary , our results show that iron is one important factor that Hp is able to obtain from host cells during colonization of the apical cell surface , and illustrate a way in which CagA and VacA work in concert to aid the bacteria in establishing a replicative niche . We hypothesize that growth on the epithelial surface involves more than just iron acquisition from the host . For example , it has previously been shown that Hp can acquire cholesterol directly from host cells via contact [82] . We speculate that Hp has evolved sophisticated mechanisms to manipulate host cell physiology for its own benefit , and that these features have side effects that result in disease in a subset of the infected population . Instead of destroying the epithelium as may occur in some types of acute bacterial infection , mucosal colonizers like Hp use more subtle mechanisms of local epithelial perturbation . We propose that mucosal colonizers may be utilizing the polarized epithelium as a “filter” both to protect themselves from potentially toxic host defense molecules , and to selectively extract micronutrients that are present inside the host in a form that is usable by the bacteria . Future exploration of the nature of host proteins associated with apical bacterial microcolonies , and the possible role of other bacterial factors in perturbing polarity , will give us a better understanding of how Hp and other mucosal colonizers affect the epithelial surface for their benefit .
All animal experiments were performed in accordance to NIH guidelines , the Animal Welfare Act , and US federal law . Such experiments were carried out with the approval of the Institutional Animal Care and Use Committee of Vanderbilt University , which has been accredited by the Association of Assessment and Accreditation of Laboratory Animal Care International ( AAALAC ) . All animals were housed in an AAALAC-accredited research animal facility fully staffed with trained technical , husbandry , and veterinary personnel . Madin-Darby Canine Kidney II ( MDCK ) cells ( kindly provided by W . James Nelson , Stanford University , Stanford , CA ) [83] , and MDCK cells stably expressing human transferrin receptor ( kindly provided by Suhaila White and Suzanne Simon , The Salk Institute , La Jolla , CA ) [25] , were maintained in DMEM ( Gibco ) containing 5% fetal bovine serum ( FBS ) ( Gibco ) , at 37°C in a 5% CO2 atmosphere . Caco-2 cells ( ATCC ) were maintained in DMEM containing 10% FBS , at 37°C in a 5% CO2 atmosphere . Polarized MDCK and Caco-2 monolayers were cultured by seeding cells at confluent density onto 12 mm , 0 . 4 µm-pore polycarbonate tissue culture inserts ( Transwell filters; Corning Costar ) . Polarized MDCK monolayers were maintained as previously described [8] . Caco-2 cells on Transwell filters were allowed to fully polarize for 3 weeks before use in assays . Apical medium was changed to DMEM one day after seeding , and basal medium ( DMEM + 10% FBS ) was changed daily during this time . Hp strain G27-MA and its isogenic ΔcagA mutant have been previously described [8] , [35] . Complementation of G27-MA ΔcagA has been previously described [8] . An isogenic ΔvacA mutant of strain G27-MA was constructed by deletion of the vacA open reading frame ( ORF ) beginning from the start codon to 14 base pairs before the end of the ORF , and replacement with the aphA gene ( conferring kanamycin resistance ) , by a PCR based method without recombinant cloning [84] , [85] . Complementation of G27-MA ΔvacA was accomplished by natural transformation with a construct containing the vacA ORF with the cat gene ( conferring chloramphenicol resistance ) immediately downstream , flanked by the upstream and downstream regions of vacA to allow for homologous recombination . Verification of the ΔvacA mutant and of the complemented strain ( VacA* ) was performed by immunoblotting bacterial whole cell lysates with a polyclonal rabbit anti-VacA antibody ( Austral Biologicals ) ( Figure S11A ) . Immunoblots of bacterial whole cell lysates with a rabbit anti-CagA-N-terminus antibody [8] or a rabbit anti-VacA antibody did not show significant differences in CagA or VacA expression resulting from deletion of vacA or cagA respectively ( Figure S11B ) . A G27-MA ΔcagAΔvacA double mutant was obtained by natural transformation of the single chloramphenicol-resistant ΔcagA mutant with the ΔvacA deletion construct , and selection on Columbia blood agar plates containing 25 µg/ml kanamycin and 25 µg/ml chloramphenicol . G27-MA carrying a mutated CagA that cannot be phosphorylated ( EPISA ) was constructed by transformation with a previously described allele [35] , with the aphA gene inserted immediately downstream of the mutant CagA sequence for selection of transformants . Hp strain 7 . 13 and its isogenic ΔcagA mutant have been previously described [8] , [45] . Routine culture of Hp on Columbia blood agar plates and co-culture of Hp with MDCK cells were as previously described [35] , [86] . Unless otherwise indicated , Hp from co-cultures were used for infections . Co-culture media for Hp with MDCK cells consists of DMEM + 5% FBS +10% Brucella broth +10 µg/ml vancomycin . Transferrin was saturated with iron essentially as previously described [87] . 9 mg of human transferrin ( Sigma ) was added to 1 . 5 ml of 0 . 25 M Tris-Cl , pH 8 , containing 10 µM of NaHCO3 . 30 µl of a mixture of 100 mM disodium nitrilotriacetate ( Sigma ) and 12 . 5 mM FeCl3 ( Sigma ) was then added to the solution . After incubation at 37°C for 1 hour , the sample was passed through a HiTrap desalting ( Sephadex G-25 Superfine ) column ( GE Healthcare ) , previously equilibrated with a solution of 0 . 02 M Tris-Cl , pH 7 . 4 , containing 0 . 15 M NaCl . The ratio of absorbance at 465 nm to 280 nm was measured to provide an estimate of the amount of iron bound by transferrin [88] . To remove unbound iron from the preparations , the samples were subsequently treated with Chelex resin ( BioRad ) [21] , according to the manufacturer's recommended batch method protocol . For Hp growth assays in media without cells , Hp grown overnight on Columbia blood agar plates were resuspended in DMEM , and aliquots inoculated into the appropriate media in 6-well plates . Where added , ferric chloride ( FeCl3 , Sigma ) was added at a final concentration of 100 µM , and human transferrin ( Sigma ) was added at a final concentration of 75 µg/ml . Data are shown as means ± SD . We also tested holotransferrin ( prepared as described above ) added to DMEM at a final concentration of 75 µg/ml , which does not allow growth of Hp without the presence of cells ( Figure S1C ) . Infection of polarized MDCK monolayers with Hp was carried as previously described [8] . In brief , Hp ( ∼108 bacteria/ml ) were added to the apical chamber , allowed to adhere for 5 minutes , and cell monolayers washed 5 times with fresh DMEM to remove non-adherent bacteria . Appropriate media was added back to the apical chamber , and the cells incubated at 37°C in a 5% CO2 atmosphere . Basal media was changed daily . After sampling for colony forming unit ( CFU ) counts from the apical chamber each day , cell monolayers were washed 3 times with fresh DMEM , before appropriate media added back and the cells returned to the incubator . Data are shown as means ± SD . Hp express both CagA and VacA during colonization of the polarized epithelium as evaluated by immunoblotting ( Figure S12A ) . We had showed previously that Hp is able to deliver CagA into MDCK cells [8] , and verified here that Hp growing on the apical cell surface is also able to deliver VacA into the host cells by immunofluorescence staining with a mouse monoclonal anti-VacA antibody ( Santa Cruz Biotechnology ) ( Figure S12B ) . DMEM is iron-poor , containing 0 . 248 µM ferric nitrate ( Invitrogen media formulation ) , in contrast to Brucella broth media often used in Hp culture , which has been reported to contain 13 . 9 µM of iron [89] . Where used , FeCl3 ( Sigma ) was added to the media in the apical chamber at a final concentration of 100 µM ( unless otherwise stated ) . Holotransferrin ( prepared as described above ) or transferrin ( Sigma ) was added to the media in the apical chamber at 75 µg/ml when used . For experiments with transferrin added to the co-culture media in the basal chamber , transferrin was added at a final concentration of 75 µg/ml . Samples were processed for confocal immunofluorescence as previously described [41] . Mouse anti-ZO-1 and mouse anti-transferrin receptor antibodies ( Zymed ) were used at 1∶300 dilution . Mouse monoclonal antibody rr1 , which recognizes an extracellular epitope of E-cadherin [41] , [90] , [91] , was used at 1∶100 dilution . Chicken anti-Hp antibodies [35] were used at 1∶200 dilution . Mouse anti-VacA ( Santa Cruz Biotechnology ) was used at 1∶100 dilution . Anti-IgG Alexa-fluor conjugated antibodies of appropriate fluorescence and species reactivity ( Molecular Probes ) were used for secondary detection . For transferrin and transferrin receptor experiments , an anti-chicken IgG Dylight 405 conjugated antibody ( Rockland Immunochemicals ) was used for secondary detection of the chicken anti-Hp antibodies . We did not use the 488 nm channel in these experiments , and visualized transferrin or transferrin receptor in the 594 nm channel to avoid overlap of the fluorescence spectra and prevent signal bleed-through from one channel into the next . Alexa-fluor 647-conjugated phalloidin ( Molecular Probes ) was pseudocolored yellow in these experiments . For all other experiments , either Alexa-fluor 594 or 647-conjugated phalloidin were used for visualization of the actin cytoskeleton and pseudocolored red or blue respectively . Nuclei were visualized with DAPI ( Invitrogen ) . Samples were imaged with a BioRad MRC-1024 confocal microscope , or with a Zeiss LSM 700 confocal microscope , and z-stacks reconstructed into 3D using Volocity software ( Improvision ) . Quantification of microcolony sizes was carried out as previously described [8] . Statistical differences between the data sets were determined by non-parametric Mann-Whitney test . MDCK cells stably expressing human transferrin receptor were seeded on Transwell filters and allowed to polarize before use in assays . Polarized cells were left uninfected or infected with Hp from the apical side for two days . For assays with polarized Caco-2 cells , Hp were infected from the apical side for 18 hours . Monolayers were washed 3 times apically and 5 times basally with fresh , pre-warmed DMEM . DMEM was added back to the apical chamber , and DMEM + 25 µg/ml human transferrin conjugated to Alexa Fluor 594 ( Invitrogen ) added to the basal chamber . The samples were then incubated on ice for 30 minutes . After this time , monolayers were washed 5 times with fresh DMEM basally . Samples were then either immediately fixed and processed for confocal immunofluorescence , or co-culture media added back to the basal chamber and the samples incubated for 30 minutes at 37°C in a 5% CO2 atmosphere before fixation and processing . For quantification of transferrin signal , we randomly sampled 300 µm X 300 µm optical fields by confocal microscopy . Stacks containing the full thickness of the monolayers were acquired at 0 . 5 µm z-steps . The stacks were reconstructed in 3D and the fluorescence sum of the transferrin signal present in the monolayers was measured . The background fluorescence was calculated from voxels imaged below the monolayers in each sample . Statistical differences between the data sets were determined by non-parametric Mann-Whitney test . Lysates were prepared for Western blots as previously described [8] . For samples from polarized monolayers on Transwell filters , cells from 3 filters were pooled for each lysate . Samples were separated by SDS-PAGE , and transferred to nitrocellulose membranes for immunoblotting . Mouse anti-transferrin receptor ( Zymed ) was used at 1∶5000 . Mouse anti-GAPDH ( EMD Chemicals Inc . ) was used at 1∶10000 . Rabbit anti-CagA-N-terminus [8] was used at 1∶10000 . Rabbit anti-VacA ( Austral Biologicals ) was used at 1∶2500 . Goat anti-mouse or anti-rabbit IgG Alexa-fluor 660-conjugated antibodies ( Molecular Probes ) were used for secondary detection . To visualize total protein , SDS-PAGE gels were stained with Coomassie Blue ( Sigma ) . A LI-COR Odyssey Scanner was used for signal detection ( LI-COR Biosciences ) . Polarized monolayer samples were fixed and non-permeabilized , apical staining carried out as previously described [41] . For quantification of transferrin receptor signal associated with bacterial microcolonies , 100 µm X 100 µm optical fields were randomly sampled by confocal microscopy . The 3D reconstructions of the confocal stacks were used to collect the fluorescence voxel volume of each microcolony stained with anti-Hp antibodies , and the fluorescence sum of the transferrin receptor signal associated with the microcolonies . For background measurements , regions on the apical cell surface of ∼20 µm3 with no bacteria were also measured for the fluorescence sum of the transferrin receptor signal present . Statistical differences between the data sets were determined by non-parametric Mann-Whitney test . Polarized monolayers were rinsed 3 times with Ringer's buffer ( 154 mM NaCl , 7 . 2 mM KCl , 1 . 8 mM CaCl2 , 10 mM HEPES , pH 7 . 4 ) . Ringer's buffer containing 200 µg/ml sulfo-NHS-SS-biotin ( Pierce ) was added to either the basal or apical chambers for selective basal vs . apical membrane protein biotinylation [40] . Samples were incubated on ice for 30 minutes . The cells were then washed 5 times with Tris-saline ( 120 mM NaCl , 10 mM Tris-HCl , pH 7 . 4 ) , and 3 times with DMEM . Samples were then either immediately fixed and processed for confocal immunofluorescence , or incubated for 30 minutes at 37°C in a 5% CO2 atmosphere before fixation and processing . Non-permeabilized , apical staining with Alexa Fluor 488-conjugated streptavidin ( Molecular Probes ) was carried out as described previously [41] . Quantification of biotin signal associated with bacterial microcolonies was carried out as described for the transferrin receptor . Statistical differences between the data sets were determined by non-parametric Mann-Whitney test . siRNAs directed against canine transferrin receptor were designed by Ambion , Applied Biosystems Inc . , using their Silencer Select siRNA design algorithm . Two canine transferrin receptor-targeted siRNAs were used in the experiments here – 5′ GCAGAAAAGUUGUUUGAAA , and 5′ CCUAUGAUCUUGAAUUGAA . A Silencer Select Validated siRNA directed against human transferrin receptor ( 5′ GGUCAUCAGGAUUGCCUAA ) and a control enhanced green fluorescent protein ( eGFP ) Silencer siRNA ( #AM4626 ) were also obtained from Ambion . siRNAs were transfected into MDCK cells using a reverse transfection protocol with Lipofectamine RNAiMAX Transfection Reagent ( Invitrogen ) . For each well to be transfected in a 6-well plate , 50 pmoles of RNAi duplex and 7 . 5 µl of Lipofectamine RNAiMAX Transfection Reagent were gently mixed in 500 µl of OPTI-MEM I Reduced Serum Medium . The mixture was incubated at room temperature for 20 minutes , and 5×105 cells suspended in 2 ml of DMEM + 5% FBS added to the mixture . The cells were then incubated at 37°C in a 5% CO2 atmosphere for 1–3 days . Samples were collected for Western blotting as previously described [8] . For Hp infection of polarized cells transfected with siRNA , cells reverse-transfected with siRNA as above were trypsinized after a 24 hour incubation at 37°C in a 5% CO2 atmosphere , and seeded at high confluent density onto Transwell filters . 30 hours after seeding , infection of the polarized epithelial cells was then carried out as described earlier for the Transwell Hp growth assays , with DMEM present in both the apical and basal chambers . Quantification of microcolony sizes was carried out as previously described [8] . Statistical differences between the data sets were determined by non-parametric Mann-Whitney test . MDCK cells stably expressing human transferrin receptor were seeded on Transwell filters and allowed to polarize before use in assays . Co-culture media was added basally , and DMEM apically . Polarized cells were left uninfected or infected with Hp from the apical side for 2 days . After this time , the media in the Transwell basal chamber was replaced with co-culture media containing 50 µg/ml biotinylated transferrin ( Invitrogen ) and 20 µg/ml biotinylated albumin ( Sigma ) . The basal chamber contains 1 . 5 ml of media , while the apical chamber contains 0 . 5 ml of media . After a 24 hour incubation at 37°C in a 5% CO2 atmosphere , the media from the apical chamber was collected and diluted 1∶1 in 2X SDS-sample buffer . Samples were separated on a SDS-PAGE gel , and transferred to nitrocellulose membranes for immunoblotting . Detection of biotinylated transferrin and biotinylated albumin was carried out by blotting with streptavidin conjugated to Alexa Fluor 660 ( Molecular Probes ) , and scanning on a LI-COR Odyssey Scanner ( LI-COR Biosciences ) . The detection limit and linear range of measurements of the biotinylated transferrin and biotinylated albumin were determined from standard curves generated by use of a dilution series of the co-culture media containing 50 µg/ml biotinylated transferrin and 20 µg/ml biotinylated albumin , diluted 1∶1 in SDS-sample buffer . Mongolian gerbils ( Harlan Laboratories ) were placed on a regular , iron-replete diet ( Modified TestDiet AIN-93M with 250 ppm iron ) , or an iron-deficient diet ( Modified TestDiet AIN-93M with no iron ) ( TestDiet , Purina Mills , LLC ) , for 3 weeks prior to infection . Animals were then inoculated with either Hp strain 7 . 13 WT or a 7 . 13 ΔcagA mutant , and sacrificed 6–8 weeks post-inoculation [92] , [93] . The iron-replete and iron-deficient diets were maintained as appropriate for each group of animals throughout the course of the experiment . Colonization was determined by quantitative culture [92] , [93] , and liver samples were also collected at the time of sacrifice for analysis of total iron content . Iron analysis was performed using inductively coupled plasma-mass spectrometry , carried out by Applied Speciation and Consulting , LLC . Statistical differences between the data sets were determined by non-parametric Mann-Whitney test . | Helicobacter pylori ( Hp ) is a bacterium that chronically infects the stomach of humans and can lead to serious illness . To survive in the stomach , the bacteria intimately interact with the epithelial lining . Some inject the virulence protein CagA into the host cells , and we previously showed that CagA helps Hp survive and grow directly on the epithelial cell surface . Iron is one of the limiting factors that infectious bacteria must acquire from their host . Using a model polarized epithelium system , we discovered that CagA is able to alter the internalization , intracellular transport , and polarity of the transferrin/transferrin receptor iron uptake system . This allows the bacteria to shuttle iron across the epithelium and suggests a novel mechanism of iron acquisition from host cells , enabling Hp growth on the cell surface . Another major virulence factor of Hp , VacA , is also involved in this process . To test the role of CagA in iron acquisition in vivo , we infected iron-deficient Mongolian gerbils and found that CagA-deficient bacteria had a decreased ability to colonize the stomach . Our study illustrates how microbes that chronically infect our mucosal surfaces can manipulate the epithelium to acquire micronutrients from host cells and grow on the cell surface . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"microbial",
"pathogens",
"biology",
"microbiology",
"host-pathogen",
"interaction",
"bacterial",
"pathogens"
] | 2011 | Helicobacter pylori Perturbs Iron Trafficking in the Epithelium to Grow on the Cell Surface |
Associating spatial locations with rewards is fundamental to survival in natural environments and requires the integrity of the hippocampus and ventral striatum . In joint multineuron recordings from these areas , hippocampal–striatal ensembles reactivated together during sleep . This process was especially strong in pairs in which the hippocampal cell processed spatial information and ventral striatal firing correlated to reward . Replay was dominated by cell pairs in which the hippocampal “place” cell fired preferentially before the striatal reward-related neuron . Our results suggest a plausible mechanism for consolidating place-reward associations and are consistent with a central tenet of consolidation theory , showing that the hippocampus leads reactivation in a projection area .
Successful foraging requires that animals maintain a representation of a multitude of reward properties including the location at which a reward can be found . Forming a place–reward association is thought to depend critically on the communication between the hippocampal formation and the ventral striatum ( VS ) . Cells in the hippocampus proper ( HC ) [1] , [2] and adjacent subiculum [3] show location-specific firing ( i . e . , “place fields” ) , and these structures are crucial for spatial and contextual learning [2] , [4]–[6] . Neurons in the VS fire in relation to rewards , as they are expected or actually delivered , as well as to cues predictive of reward [7]–[9] . Receiving information from a range of structures such as the HC , amygdala , prefrontal cortex , and midline thalamic nuclei [10]–[13] , the VS is thought to utilize information of reward-predicting cues and contexts to guide goal-directed behavior [8] , [14] , [15] . This process is under strong control of the mesolimbic dopaminergic system , and its disruption has been associated with neuropsychiatric conditions such as drug addiction and obsessive-compulsive disorder [16]–[18] . Although the hippocampal formation projects directly to the VS , and this connection has been implicated in contextual conditioning [19] , it is unknown how neural representations of contextual and motivational information are integrated and stored to enable the learning of place-reward associations . In several brain areas , neuronal patterns evoked during behavior are reactivated during subsequent sleep [20]–[24] . Through modification of synaptic connections , this reactivation has been theorized to constitute an important step in memory consolidation [25]–[28] . Because hippocampal CA1 pyramidal cells exhibiting place fields during active behavior have been demonstrated to reactivate during sleep , it may be reasonably assumed that this replay pertains to spatial and contextual information [20] , [21] , [29] , [30] . In contrast , reactivation in the VS is dominated by reward-related information [31] . Joint reactivation of HC and VS may enable the formation of a memory trace comprising both contextual and motivational components . In this study , we recorded activity from neuronal ensembles in the rat HC and VS simultaneously during wake and sleep episodes to examine whether the HC and VS reactivate coherently and to reveal the temporal dynamics of this process . First , during active behavior , much of the dynamics of hippocampal processing is governed by the theta rhythm , which has been hypothesized to function as a “read-in” or encoding mode for information acquisition and provides a means to temporally align spike sequences by way of theta phase precession [26] , [32]–[34] . Therefore , we studied whether neural activity modulation by this rhythm in the awake state is correlated to reactivation during sleep . A second foremost question in this field , not yet addressed in previous multi-area recording studies [22] , [24] , [35] , is whether cross-structural replay depends on the type of behavioral information coded by cell assemblies . To address this question , we investigated whether reactivation is preferentially associated with the expression of place fields and reward-related neural responses . Third , we planned to utilize joint HC-VS recordings to test a central tenet of theories of memory consolidation [25]–[28] . These theories posit that , after a learning experience , long-term episodic and declarative memories become gradually independent of hippocampal storage because this structure would repeatedly retrieve stored associative information over time and thereby orchestrate consolidation of memory traces in the neocortex and other target sites . A key point in these hypotheses is that replay is initiated and orchestrated by the HC , which prompted us to examine whether hippocampal activity leads the VS during reactivation .
Four rats were implanted with a tetrode drive allowing joint HC-VS recordings of spike trains of multiple neurons and local field potentials ( LFPs ) in each area . Daily recording sessions were composed of an episode of reward searching behavior flanked by two episodes of rest , which rats spent on a “nest” next to the track . The task was to run along a triangular track repeatedly and in one direction . On each lap , one of three reward wells was baited with a drop of one of three corresponding reward types; i . e . , sucrose solution , vanilla desert , or chocolate mousse . An example of joint HC and VS ensemble recordings during track running and sleep is shown in Figure 1 . First , we assessed reactivation of neuronal patterns using an explained variance ( EV ) method based on the spike correlations of cell pairs across all simultaneously recorded neurons [23] , [36] . The EV reflects the extent to which the variance in the distribution of spike correlations during postbehavioral rest is statistically accounted for by the correlation pattern found during track running , factoring out the correlations present in prebehavioral rest . Joint HC-VS reactivation was examined during rest periods in which the rat was immobile , using only spike correlations between pairs composed of one HC and one VS neuron . We found coherent , cross-regional reactivation between ensembles of the HC and VS as expressed by an EV of 9 . 7±3 . 0% , which was significantly higher than the control measure , the reverse explained variance ( REV: 1 . 4±0 . 5% , p<0 . 01 , n = 21 sessions; Figure 2A and 2B; Figures S1 and S2; Table S1; Text S1 ) . In the analyses conducted in this research , putative interneurons were excluded from the neuronal population , but it should be noted that including these interneurons yielded similar reactivation values ( EV: 9 . 4±2 . 7% , REV: 1 . 9±0 . 7%; p<0 . 01 ) . Analysis of the temporal dynamics of reactivation in 20-min blocks of concatenated rest revealed a gradually decaying reactivation which was significant for at least 1 h of postbehavioral rest ( Figure 2C ) . Within periods of rest , reactivation was prominent especially during quiet wakefulness–slow-wave sleep episodes ( QW-SWS; n = 13 sessions ) , but it was not significant for rapid eye movement ( REM ) sleep ( Figure 2D ) . The lack of pattern recurrence during REM sleep was not attributable to its relatively short duration , undersampling of spikes , or its late occurrence after sleep onset compared to QW-SWS ( Figure S3; Table S2; Text S1 ) . Reactivation in the HC [36] and VS [31] occurs markedly during sharp wave-ripple complexes , i . e . , short-lasting , high-frequency oscillations in the hippocampal LFP with associated bursts of large-scale neuronal firing that characterize QW-SWS [2] , [37] . The same trend was observed for joint reactivation; however , the difference in reactivation values for time windows of 200 ms following ripple onset ( “Ripples” ) and during 200 ms following the onset of interripple intervals ( “Intervals” ) did not reach statistical significance , which most likely relates to the high variability across sessions ( cf . [31] ) ( Ripples: EV: 5 . 9±2 . 6% , REV: 1 . 0±0 . 3%; Intervals: EV: 1 . 9±0 . 8% , REV: 0 . 5±0 . 2%; EV and between Ripples and Intervals [EV−REV]: n . s . ; n = 14 sessions ) . The existence of joint HC-VS reactivation raises the question of which physiological and behavioral factors are associated with the strength of this process . We examined three not mutually exclusive factors pertaining to ( 1 ) the modulation of the neural activity patterns by theta oscillations , ( 2 ) the correlation of neuronal firing patterns with behavioral parameters , and ( 3 ) the order in which neurons in different areas were activated . First , we computed the degree to which cells in each pair fired together , and then all of these correlation values per episode were pooled across sessions and animals . We next formed subgroups of cell pairs by partitioning the complete set of correlation values on the basis of the factor under scrutiny . Reactivation values were computed for these subgroups , and statistical significance was assessed by applying a bootstrapping procedure with resampling of pooled correlation values [22] , [23] . The two-stage model of memory trace formation posits that theta oscillations are crucial for encoding information in the HC in the awake , active state [26] . The hippocampal theta rhythm may also have a role in governing the temporal organization of activity in target structures to ensure efficient communication [38] , [39] . Thus , our first hypothesis holds that HC-VS reactivation will only be strong when information is cross-structurally aligned during encoding by a common temporal framing , the theta oscillation , creating windows of near-synchronous firing . During track running , robust theta oscillations were observed in the HC and VS . In both areas , cells were observed whose firing patterns were modulated by the local theta rhythm ( n = 121 out of 263 , 46 . 0% , in HC , and n = 20 out of 243 , 8 . 2% , in VS ( Figure 3A; FigureS4A ) . Ventral striatal units that were modulated by the local theta rhythm generally showed firing rate modulation also by hippocampal theta oscillations ( n = 20 , 8 . 2% ) . When the peak of the theta oscillation recorded near the hippocampal fissure was taken as synchronizing time point , CA1 cells fired at an average angle of 199 . 9±6 . 1° ( range: 43 . 9°–356 . 4° ) and VS cells fired at a slightly later phase ( 217 . 1±26 . 0° , range: 4 . 5°–336 . 3°; n . s . ) . Cell pairs were divided on the basis of modulation by the hippocampal theta rhythm , resulting in four subgroups: Both Cells ( n = 140 ) , HC only ( n = 1 , 273 ) , VS only ( n = 81 ) , and None Modulated ( n = 1 , 422 ) . Reactivation was observed for all but the VS only groups; its strength was significantly stronger in the Both Cells group compared to the None Modulated and HC only groups ( Figure 3A; Table S3 ) . Our second hypothesis departed from the assumption that spike patterns are not reactivated equally , but are reprocessed especially when they convey behaviorally relevant information . For the HC-VS system , we predicted that cells expressing spatial ( HC ) and reward-related information ( VS ) should be preferentially reactivated . Location-specific firing was found for 102 out of 263 ( 38 . 8% ) hippocampal cells . Place fields were distributed uniformly across the track; there was no indication that place fields occurred more frequently near reward sites or corners of the track . In contrast , a subset of VS neurons fired in close temporal relationship with reward site visits ( 41 out of 243 cells , 16 . 9%; Text S1 ) . Reward-related responses were generally increments in firing rate and could be generated at one , two , or all three reward sites . Furthermore , they were often sensitive to either the presence or absence of reward . In line with previous studies on the VS [9] , [31] , [40] , we will apply the term reward-related to all VS units showing significant responses time-locked to reward site visits . Depending on the expression of place fields and reward-related correlates , cell pairs were grouped in four categories: Double Correlates ( n = 192 ) , Place Field only ( n = 941 ) , Reward-related Correlate only ( n = 287 ) , or No Correlates ( n = 1496 ) . The Double Correlates group showed very strong reactivation ( EV: 22 . 9% , REV: 0 . 1% ) , whereas reactivation in the other three subgroups in this partition was not significant ( Figure 3B; Table S3 ) . Accordingly , the strength of coactivation of a place cell and a reward-related VS cell , expressed in the Pearson correlation coefficient , was positively correlated to the degree of spatial overlap of the firing fields on the track during task performance and postbehavioral sleep , but not during prebehavioral rest ( prebehavioral rest: n . s . ; track running: R2 = 0 . 25 , p<1×10−12 , postbehavioral rest: R2 = 0 . 03 , p<0 . 02; n = 192 ) . A long-standing assumption in memory consolidation theory holds that the HC initiates and orchestrates reactivation in its projection areas [25]–[28] . This general process may be realized in several ways ( see Text S1 ) . In the HC-VS system , evidence suggests a particular variant of replay in which hippocampal ripples initiate reactivation locally and subsequently trigger a wave of enhanced excitability in the VS [23] . This variant implies that reactivation should be strong when a particular firing order is maintained: during replay , a hippocampal cell should fire predominantly in advance of a VS cell . During behavior , VS firing may also precede HC firing , but this order should not be associated with strong reactivation . The HC→VS order would also be consistent with the unidirectionality of the projection from HC to VS [13] . Despite the finding that sleep reactivation occurs in a “forward” direction , meaning that the order of firing during sleep is similar to the order during the preceding behavior [29] , [30] , [41] , this critical assumption has yet to be confirmed or refuted . Hence , our third hypothesis holds that reactivation is strong when the information flow is organized according to a leading role of the HC . The firing order of each cell pair was assessed by computing cross-correlograms [42] , [43] and determining which order of firing was most prevalent using a “temporal bias” measure [29] . Three subgroups were distinguished , i . e . , HC→VS pairs ( n = 608 ) , VS→HC pairs ( n = 796 ) , and No Clear Order , which included pairs that did not show a preferred firing order ( n = 1 , 512 ) . The HC→VS group reactivated strongly ( EV: 15 . 2% , REV: 0 . 0% ) . Reactivation was also observed for the other groups , although the observed strengths were significantly lower than for the HC→VS group . ( Figure 3C; Table S3 ) . Variations in reactivation measured across all of the subgroups partitioned according to each of the three factors could not be attributed to differences in varying numbers of cell pairs , differences in correlation strengths , or differences in spike counts ( Text S1 ) . Altogether , these results suggest that all three factors analyzed—modulation of both cells by the hippocampal theta rhythm , maintenance of the HC→VS firing order , and expression of a combination of a place field and a reward correlate—are associated with strong reactivation . However , since a reactivating cell pair may display multiple characteristics at the same time ( e . g . , behavioral correlates and a particular firing order; see Figure S4 ) , we used a multilinear regression model to test whether the contribution of each cell pair to the session EV value was dependent on firing order , theta modulation , behavioral correlates , or any combination of these characteristics . First , the relative contribution of each cell pair to the session reactivation was estimated by excluding a pair from the simultaneously recorded population and recomputing the reactivation values . The difference between the session EV minus the EV after pair exclusion represents the estimated contribution of that pair to the session EV . Multilinear regression showed that both the expression of a double correlate and the HC→VS firing order were significant factors in explaining the contribution to the session EV ( p<0 . 02 and p<0 . 002 , respectively; theta modulation was not significant , p = 0 . 6 ) . The combination of firing order and expression of a double correlate predicted the pair's contribution better than either one alone ( p<0 . 0002 ) . This analysis confirms the importance of the HC→VS firing order and expression of combined place and reward information during track running for subsequent reactivation and identified theta modulation as a less significant indicator . We tested whether reactivation in the subgroup reactivating most strongly , i . e . , the Double Correlates , was sparsely distributed as we previously showed for VS ensembles [31] . First , we assessed the contribution of each cell pair to the reactivation as explained above . To find an indication of how many cell pairs can be excluded to abolish reactivation , the pairs were sorted in descending order in terms of their contribution to the reactivation value [EV−REV] . Then the pairs were excluded one by one in a cumulative fashion starting with the highest contributor from the population , and reactivation values were computed each time a next pair was excluded . If the 17 ( 17/192 , 8 . 9% ) most contributing cell pairs were left out of the population , the [EV−REV] dropped below 5 . 0% . A total of 34 ( 17 . 7% ) pairs could be removed before the [EV−REV] level decreased to 0 . 0% . This analysis indicates that , consistent with VS ensembles , reactivating cell pairs were also sparsely distributed in the HC-VS population . We next explored whether HC-VS cell pairs fire in the same order during reactivation as during behavior and whether replay is accelerated relative to active behavior . For each pair of neurons that showed a place field and a reward-related correlate , we constructed three cross-correlograms , one for each task-episode ( n = 192 , Figure 4A ) . We compared the time offsets during active behavior and rest for pairs that showed significant peaks in the cross-correlograms of track running and in at least one of the rest episodes ( n = 53 , 27 . 6% ) . The time offsets of the peaks during track running were positively correlated to those in postbehavioral rest ( R2 = 0 . 09 , p<0 . 05; n = 47 ) , but not to those of prebehavioral rest ( Figure 4B; n = 26 ) . Thus , the recurrent firing patterns reflected the preceding experience . In this analysis , spatial overlap between the firing fields of a cell pair turned out not to be a prerequisite for concurrent firing during subsequent sleep , as 29 . 8% of the cell pairs that showed peaks in the cross-correlograms for task performance and postbehavioral rest exhibited nonoverlapping firing fields on the track . The peak offset in the cross-correlograms of track running ranged from −4 . 5 to 3 . 8 s and was significantly correlated to the spatial distance between the firing fields ( R2 = 0 . 27 , p<0 . 001 ) . To determine whether the order of firing on the track was preserved or reversed in the subsequent rest episode the offset sign ( + or − ) of the cross-correlogram peak relative to zero was considered . Peaks during track running and postbehavioral rest were consistently found with the same offset sign ( 43/47 = 89% , sign test , p<0 . 0001 ) , which demonstrates that replay took place in a forward direction . In combination with the strong reactivation of cell pairs that exhibited the HC→VS firing order during track running observed in the subgroup-based reactivation analysis ( Figure 3C ) , the preservation of firing order suggests that replay should be dominated by activity patterns in which HC firing largely precedes VS firing , both during track running and postbehavioral rest . Indeed , in the large majority of cell pairs that showed forward reactivation , the hippocampal cell fired preferentially before the striatal cell during both periods ( 36/43 = 83 . 7% ) , indicating that most of the reactivating cell pairs express a HC→VS order during behavior and sleep ( sign test , p<0 . 0001 ) . Thus , not only is the firing order preserved from the behavior to ensuing sleep , but apparently the HC also takes the lead in replay and the VS follows . An additional analysis on all cell pairs with significant cross-correlogram peaks yielded similar results and confirmed that the preferential firing order during reactivation was not attributable to a lack of VS→HC correlations during track running ( see Text S1 ) . Like cross-structural reactivation , reactivation within hippocampal and ventral striatal ensembles also took place in a forward direction ( see Text S1 ) ( cf . [29] for HC ) . Replay may occur at a different time scale than applicable during behavior [29] , [30] , [41] . We examined whether joint HC-VS firing patterns were replayed on an accelerated time scale . Peak times in the postbehavioral rest cross-correlogram occurred significantly closer to zero than during track running ( track: −525 . 5±201 . 9 ms , postbehavioral rest: −53 . 2±28 . 5 ms; p<0 . 01; n = 47 ) , showing an approximately10-fold compression ( Figure 4 ) . Replayed patterns appeared compressed and not merely truncated , because the shape of the cross-correlogram peaks with offsets of up to several seconds during behavior were re-expressed during sleep , including their offset from zero ( Figure 4A ) .
Altogether , our results demonstrate coherent reactivation between the HC and a subcortical structure , and identify two major factors governing cross-structural reactivation in the HC-VS system , suggesting a plausible mechanism for consolidation of associative place-reward information . The first factor that significantly correlated to strong HC-VS reactivation bears on the dependence of reactivation on the coding of behaviorally relevant information . Cell pairs that exhibited a double correlate—one place field plus one reward-related correlate—showed the strongest reactivation among all four subgroups . In addition , the contribution to reactivation by individual pairs depended specifically on such a coexpression of behavioral correlates . The near-synchronous reiteration of spatial and motivational information during sleep may serve to integrate these types of information and support the learning of place-reward associations . Such associations are essential to predict and localize desired food and liquids within a known environment and are therefore fundamental to foraging behavior and learned behaviors such as conditioned place preference [5] , [19] , [44] . Like intra-area ventral striatal reactivation [31] , cross-structural replay is dependent on a relatively small subset of cell pairs , indicating that it is a sparsely distributed phenomenon . If replay indeed supports memory consolidation , the formation of associations of a specific place-reward combination is likely to depend on a small minority of cells in the HC-VS circuitry . The second significant factor in joint HC-VS replay is the preferred firing order of HC and VS cells , consistent across track running and subsequent sleep . The HC→VS firing order during track running was associated with a significantly elevated reactivation as compared to other temporal relationships , and the cell pair contributions to reactivation depended on this specific firing order . This organization of firing order obeys the direction of the anatomical projection [13] , [45] and presents necessary , although not sufficient , evidence for a central tenet of consolidation theory , proposing the HC to initiate reactivation in its target structures , as predicted by Marr [25] and subsequent theorists [26]–[28] . Our data provide several indications supporting that the observed cross-structural reactivation is the consequence of a coordinated process between the HC and the VS rather than of two separately , or coincidentally , reactivating ensembles . First , the temporal relationship between a pair of task-related hippocampal and ventral striatal cells was relatively consistent across task phases , resulting in significant peaks in the cross-correlograms of a substantial number of pairs both during track running and postbehavioral rest . If joint reactivation was just coincidental , the temporal firing relationship between cells in different structures is expected to be random , contrary to what was observed . Second , the timing of the peaks during track running and postbehavioral rest was correlated in the cross-correlogram analysis ( Figure 4B ) , and furthermore , the time scale of sequential activation of firing patterns during postbehavioral rest was compressed compared to track running . Thus , temporal firing relations were consistent across different overall brain states ( awake active versus SWS ) and on accelerated time scales . Observing such results is very unlikely if the two structures would be reactivating without a systematic temporal relationship . Third , during behavior we identified pairs that fired in the order HC→VS and also in the order VS→HC . During reactivation in the postbehavioral rest , however , we found an overrepresentation of HC→VS pairs . If two ensembles reactivated independently , one would expect the ratio of HC→VS and VS→HC pairs to be similar during behavior and reactivation . An important finding is that the joint reactivation is compressed by a factor of ten compared to the behavioral time scale of neuronal activation . Thus , at least several seconds of “real-time” joint place-reward information during behavior are brought together in a time frame of hundreds of milliseconds during sleep . This further supports the plausibility of a mechanism for the associative storage of place and reward information by way of synaptic weight changes in the HC-VS system . If a cell from the hippocampal formation , coding place , fires consistently and briefly in advance of a VS cell signaling reward ( Figure 4 ) , spike timing–dependent plasticity may be induced in their connection [46] , [47] . Cross-correlogram analysis revealed that joint reactivation is not restricted to neuronal pairs that exhibit overlapping firing fields; peaks that were separated by up to about 4 . 5 s during behavior were found to recur during postbehavioral rest . In a scenario in which a series of place fields is followed by a reward-related correlate , this indicates that value information during SWS is not only paired with locations nearby , but also with more remote stages of the path leading to the reward site . Formation of reward-predicting representations should , by definition , obey the temporal order of predictor-reward events , a requirement that is met by the preferential HC→VS firing order during replay . In principle then , the characteristics of hippocampal-striatal replay are suitable for mediating the “backwards” association between reinforcements and cues and contexts situated progressively earlier in time . This temporally backwards referral is a key feature of conditioning theory and models of reinforcement learning [48]–[50] . Although the causal role of ensemble reactivation in memory consolidation remains to be proven , the temporally ordered cross-structural replay of spatial and motivational information during sleep illuminates a plausible offline mechanism by which information processed in different parts of the brain can be integrated to enable the composition and strengthening of memory traces comprising various attributes of a single learning experience .
All experimental procedures were in accordance with Dutch national guidelines on animal experimentation . Four male Wistar rats ( 375–425 g; Harlan ) were individually housed under a 12/12-h alternating light-dark cycle with light onset at 8:00 am . All experiments were conducted in the animal's inactive period . During training and recording periods , rats had access to water during a 2-h period following the experimental session , whereas food was available ad libitum . Rats were chronically implanted with a microdrive [51] containing five individually movable tetrodes directed to the dorsal hippocampal CA1 area ( 4 . 0 mm posterior and 2 . 5 mm lateral to bregma ) and seven to the VS ( 1 . 8 mm anterior and 1 . 4 mm lateral to bregma ) [52] . Reference electrodes were placed in the corpus callosum dorsal to the HC , and near the hippocampal fissure . A skull screw inserted in the caudal part of the parietal skull bone served as ground . Unit activity , local field potentials , and position data were acquired on a 64-channel Cheetah recording system ( Neuralynx ) . Spike sorting was performed offline using custom cluster-cutting software as described in Text S1 . Recordings of hippocampal CA1 neurons were made from 103 locations between 2 . 6 mm and 4 . 8 mm posterior and between 1 . 2 mm to 2 . 8 mm lateral to bregma compared to an atlas of the rat brain [52] . Ventral striatal tetrodes were situated between approximately 2 . 2 and 1 . 2 mm anterior to bregma and between 1 . 6 and 3 . 0 mm laterally . From a total of 140 recording sites , 58% was estimated to be situated in the core region and 42% in the shell region of the VS . Although most sessions were likely to contain recordings from both the core and shell region , six sessions were identified to contain core-only recordings . No gross differences were observed in the number , firing rate , or appearance of behavioral correlates that were estimated to be recorded from the core and the shell region . Moreover , cross-regional reactivation was observed for the core-only sessions , with EV and REV values similar to those observed for other sessions . Therefore , core and shell recordings were pooled . Pre- and postbehavioral rest episodes included all periods of at least 20 s in which the rat was in the flower pot and remained motionless; i . e . , episodes of movement were excluded from analysis . Within these periods of rest , episodes of SWS were characterized by the presence of large irregular activity and the occurrence of sharp wave-ripple complexes in the LFP of the CA1 pyramidal layer [2] , [37] . Ripples were detected each time the squared amplitude of the filtered LFP trace ( 100–300 Hz ) crossed a threshold of 3 . 5 standard deviations ( SD ) for at least 25 ms . Because incidentally short periods of quiet wakefulness may have been included in SWS episodes , as these two phases share principal LFP characteristics , this state is referred to as quiet wakefulness–slow-wave sleep ( QW-SWS ) . REM sleep periods were indicated by an elevated ratio ( >0 . 4 ) of spectral density in the theta band ( 6–10 Hz ) to the overall power of the LFP trace recorded near the hippocampal fissure . Their borders were refined upon visual inspection of the trace . Modulation of a cell's firing pattern to the theta oscillation was determined by first filtering LFP traces recorded from the hippocampal fissure and the VS using a Chebyshev type-1 bandpass filter between 6 and 10 Hz . Binned spikes ( 10°/bin ) were then plotted relative to the theta peaks of two successive theta periods . The spike distribution was considered nonuniform when the Rayleigh score was <1×10−5 . The phase angle of the spikes was determined by computing the Hilbert transform of the filtered theta signal . Firing of a unit was considered as being modulated by the theta rhythm when shuffling of the spikes abolished the nonuniformity of the spike firing distribution as assessed with the Rayleigh score . To characterize spatially selective firing fields , instantaneous firing rates were computed for bins of 50 ms . The spatial position of the rat's head was determined by creating a one-dimensional representation of the track and using a resolution of 2 . 3 cm . Mutual information was computed between the binned spike trains and the position , and corrected for finite sampling size [56] , [57] . A cell was considered to express a place field if its firing rate during track running was at least 0 . 3 Hz and if it carried at least 0 . 25 bits/spike of spatial information . Peri-event time histograms were constructed for the rewarded and nonrewarded condition for each reward site and were synchronized on crossings of offline installed “virtual photobeams” positioned at the points where the rat was just reaching the reward sites . Reward-related responses were assessed within a period of 1 s before and 1 s after arrival at a reward site , using a bin resolution of 250 ms . Spike counts in the eight bins comprising the reward period were each compared to three separate control bins taken from the corner passage opposite to the well under scrutiny within the same lap . A bin of the reward period was only considered significantly different when the Wilcoxon matched-pairs signed rank test indicated significance from each of the three control bins ( p<0 . 01 ) . A reward-related response comprised one or more bins that were significantly different from control bins . Control bins did not show marked deviations from overall baseline firing as checked in peri-event time histograms and plots of the spatial distribution of firing rates . Differences between the responses at different reward sites were assessed with a Kruskal-Wallis test ( p<0 . 05 ) followed by a MWU ( p<0 . 05 ) , whereas rewarded versus nonrewarded conditions were compared using MWU ( p<0 . 05 ) . Cross-correlograms were constructed according to Perkel et al . [58] and Eggermont [43] . Spikes were binned into 10- or 50-ms intervals , and the cross-correlation was examined across at least three time windows; viz . [−500 , 500] ms , [−2 , 000 , 2 , 000] ms and [−5 , 000 , 5 , 000] ms . The firing order of pairs of hippocampal and striatal cells was assessed with the temporal bias method [29] using cross-correlograms with the spikes of the striatal cell serving as reference . The ordinate expressed spike counts per second , which was integrated across intervals of 200 ms before ( I ) and after ( II ) zero . The difference between II minus I divided by the sum of the spike counts determined the temporal bias score . If this score was negative , the HC was determined to fire preferentially before the striatal cell . If this score was positive , the preferred firing order was in the opposite direction . The classification No Clear Order was assigned when the scores ( I and II ) were approximately equal or when the cross-correlogram did not have a clear single maximum . To estimate the significance of peaks in the cross-correlograms , the mean expected number of joint spike counts μ and the levels of μ±3 SD ( corresponding to p = 0 . 0013 ) were computed to provide indications for nonrandom excursions of spike counts above or below the expected range [43] . Each cross-correlogram was then subjected five times to a spike-shuffling subtraction procedure [42] , [58] . Peaks were accepted as significant only when they exceeded the +3 SD threshold above the mean in each of the five repetitions . | Thinking back to an exciting event often includes the scene in which the event took place . Associations between specific places and emotional events are consolidated in memory , but how this is achieved is currently unknown . Two brain areas involved in learning such associations are the hippocampus and the ventral striatum , which represent spatial and emotional information , respectively . A highly valuable object in an environment will prompt humans and animals to take action , such as approaching the object . Here , we demonstrate that a combination of spatial and emotional aspects of a learning experience is replayed in the hippocampus and the ventral striatum during sleep , which is likely to contribute to the consolidation and strengthening of memory traces . This reactivation is coordinated such that the spatial information in the hippocampus is activated shortly before the emotional information in the ventral striatum . This finding is consistent with a central prediction from Memory Consolidation Theory , namely that the hippocampus initiates and orchestrates replay in connected brain areas . In addition , sleep replay occurs at a time scale about ten times faster than during the actual experience , which makes it a mechanism suitable for strengthening synaptic connections associating place with reward . Our results shed new light on the distributed way the brain processes , links , and retrieves different aspects of memories . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"neuroscience/behavioral",
"neuroscience",
"neuroscience/cognitive",
"neuroscience",
"physiology/cognitive",
"neuroscience",
"neuroscience/neuronal",
"signaling",
"mechanisms",
"neuroscience/animal",
"cognition"
] | 2009 | Hippocampus Leads Ventral Striatum in Replay of Place-Reward Information |
Enterovirus 71 ( EV71 ) is causing life-threatening outbreaks in tropical Asia . In Taiwan and other tropical Asian countries , although nationwide EV71 epidemics occur cyclically , age-specific incidence rates of EV71 infections that are critical to estimate disease burden and design vaccine trials are not clear . A nationwide EV71 epidemic occurred in 2008–09 in Taiwan , which provided a unique opportunity to estimate age-specific incidence rates of EV71 infections . We prospectively recruited 749 healthy neonates and conducted follow-ups from June 2006 to December 2009 . Sera were obtained from participants at 0 , 6 , 12 , 24 , and 36 months of age for measuring EV71 neutralizing antibody titers . If the participants developed suspected enterovirus illnesses , throat swabs were collected for virus isolation . We detected 28 EV71 infections including 20 symptomatic and 8 asymptomatic infections . Age-specific incidence rates of EV71 infection increased from 1 . 71 per 100 person-years at 0–6 months of age to 4 . 09 , 5 . 74 , and 4 . 97 per 100 person-years at 7–12 , 13–24 , and 25–36 months of age , respectively . Cumulative incidence rate was 15 . 15 per 100 persons by 36 months of age , respectively . Risk of EV71 infections in Taiwan increased after 6 months of age during EV71 epidemics . The cumulative incidence rate was 15% by 36 months of age , and 29% of EV71 infections were asymptomatic in young children .
Enterovirus 71 ( EV71 ) was first isolated in California , USA , in 1969 . Since then , EV71 has been identified globally . The clinical spectrum of EV71 infection ranges from asymptomatic infection , to mild hand-foot-mouth disease ( HFMD ) , and severe cases with central nervous system ( CNS ) , and cardiopulmonary involvement [1] , [2] . Recent studies have further demonstrated that CNS-complicated EV71 infections could cause long-term cognitive and motor deficits [3] , [4] . Globally , two patterns of EV71 outbreaks have been reported: small-scale outbreaks with few CNS-complicated cases and deaths , and large-scale outbreaks with frequent CNS-complicated cases and deaths [1] . The latter pattern occurred in Bulgaria , with 44 deaths in 1975 [5]; in Hungary , with 45 deaths in 1978 [6]; in Malaysia , with 29 deaths in 1997 [7]; in Taiwan , with 78 deaths in 1998 [8]; in Singapore , with 5 deaths in 2000 [9]; and recently in China , with more than 100 deaths in 2007 , 2008 , and 2009 [10]–[12] . Since the 1998 epidemic , EV71 has continued to cause nationwide epidemics again in 2000–2001 , 2004–2005 , and 2008–2009 in Taiwan [12]–[22] . No antiviral against EV71 is currently available , so development of EV71 vaccines has become a national priority in Taiwan and China , and several organizations in Asia are planning clinical trials of EV71 vaccines [12] . To design clinical trials of EV71 vaccines , age-specific incidence rates of EV71 infections are required to identify target populations , estimate disease burdens , select endpoints of clinical efficacy , and estimate sample size . Taiwan has had a national surveillance system for severe enterovirus infections since 1998 . Age-specific incidence rates of EV71-related severe infections during the 1998 epidemic have been estimated to be 27 . 3 , 37 . 1 , 30 . 0 , and 23 . 1 per 100 , 000 for children aged <6 , 6–11 , 12–23 , and 24–35 months , respectively [23] , which are too low to be a suitable clinical endpoint . Alternatively , EV71-related mild infections such as herpangina and HFMD could be suitable clinical endpoints; but their age-specific incidence rates are not available in Taiwan . We initiated a longitudinal cohort study in 2006 to estimate age-specific incidence rates of EV71 infection in young children in northern Taiwan .
Institutional review board approval was obtained from Chang Gung Memorial Hospital ( CGMH ) following the Helsinki Declaration; and written informed consent was obtained from all mothers of participating infants . Pregnant women having prenatal exams in CGMH were invited to participate in the study starting in June 2006 . Sera were obtained from participating pregnant women and their children for measuring EV71 neutralizing antibody titers in the following schedule: pregnant women immediately before delivery; neonates at birth ( cord blood ) ; and infants at 6 , 12 , 24 , and 36 months of age . About 20–30 participants were recruited every month . Parents were educated to contact study staff when their children developed suspected enterovirus illnesses ( herpangina or HFMD ) . During enterovirus seasons , study staff actively contacted parents about suspected enterovirus illnesses by e-mail , telephone or text message . If the participating children developed suspected enterovirus illnesses , throat swabs were collected from these participating children for virus isolation . In previous studies , throat swabs were more sensitive than rectal swabs or feces for isolating EV71 viruses in acute EV71 cases [12] . Occasionally , pediatricians also collected throat swabs from the participating children who developed non-specific febrile illness for virus isolation during enterovirus seasons . We chose CGMH as a study site because it has large obstetric and pediatric populations and serves residents from both rural and urban areas in northern Taiwan [14] , [22]–[24] . In additional to the infant cohort study , clinical samples ( throat swabs , rectal swabs , or stool samples ) are routinely collected for virus isolation from hospitalized pediatric patients with suspected enterovirus infections ( herpangina , HFMD , or non-specific febrile illness ) in CGMH . Monthly distribution of EV71 isolation in pediatric inpatients the study hospital is based on the routine statistics and is anonymized . This information was used to decide EV71 seasons for data analysis . Data related to serum EV71 antibody titers in pregnant women , neonates and 6-month-old infants have been published previously [24] and this report focused on incidence rate of EV71 infections in the first three years of life . In this study , evidence of herpangina included oral ulcerations on anterior tonsillar pillars , soft palate , buccal mucosa , or uvula . Evidence of HFMD included oral ulcers on the tongue and buccal mucosa , and a vesicular rash on the hands , feet , knees , or buttocks . Nonspecific febrile illness was defined as a rectal temperature greater than 38°C without other symptoms . Laboratory evidence of EV71 infection was defined as the isolation of EV71 from a throat swab; or a ≥4-fold rise or seroconversion in EV71 neutralizing antibody titers in paired sera samples . Asymptomatic EV71 infection was defined as a seroconversion in EV71 neutralizing antibody titers in paired sera samples without display of clinical symptoms during the period of collection of paired sera . Samples were inoculated into human embryonic fibroblast , LLC-MK2 , HEp-2 , and rhabdomyosarcoma cell cultures . When enteroviral cytopathic effect involved more than 50% of the cell monolayer , cells were scraped and subjected to indirect fluorescent antibody staining with enterovirus monoclonal antibodies [25] . Laboratory methods for measuring EV71 serum neutralizing antibody titers followed standard protocols [26]–[27] . Twofold serially diluted sera ( 1∶8–1∶512 ) and a virus working solution containing 100 TCID50 of EV71 strain E59-TW-2002 ( B4 genotype ) were mixed on 96-well microplates and incubated with rhabdomyosarcoma cells . A cytopathic effect was observed in a monitor linked with an inverted microscope after an incubation period of 4 to 5 days . The neutralization titers were read as the highest dilution that could result in a 50% reduction in the cytopathic effect . Each test sample was run simultaneously with cell control , serum control , and virus back titration . The starting dilution was 1∶8; the cutoff level of seropositivity was set at 8 . For deciding serostatus ( positive or negative ) , sera were tested only at 1∶8 . Since no EV71 infection was detected in our study cohort in 2006 and 2007 [22] , observation time ( person-months ) for participants who returned for follow up during 2008 and 2009 were summed and converted to person-years to calculate age-specific incidence rates and 95% confidence interval ( 95% CI ) based on the Poisson distribution [27] . Cumulative incidence rates were calculated for participants who returned for follow up; and 95% CIs of cumulative incidence rates were calculated based on the binominal exact distribution . Neutralization antibody titers were log-transformed to calculate the GMT and 95% CI . The statistical association between two nominal or ordinal variables was tested by the χ2 test , Fisher's exact test , or the Mantel-Haenszel χ2 test for trend , as appropriate . All statistical analyses were performed using Microsoft Excel ( Microsoft , Redmond , WA , USA ) or SAS ( SAS Institutes , Cary , NC , USA ) .
In CGMH , clinical samples are routinely collected for virus isolation from hospitalized pediatric patients with suspected enterovirus infections . From 2007 to 2009 , 4691 pediatric inpatients were tested for enterovirus cultures and the enterovirus isolation rates in these three years were 14% ( 207/1475 ) , 21% ( 420/1969 ) , and 15% ( 181/1247 ) , respectively . EV71 was isolated from 123 inpatients , including 5 cases in 2007 , 104 cases in 2008 , and 14 cases in 2009 . Only 5 sporadic EV71 inpatients were detected in 2007 , all in the middle of the year . In January 2008 , EV71 inpatients began being detected again , with the numbers growing rapidly before peaking in June of that year and disappearing after July 2009 ( Fig . 1 ) . Overall , the pattern was consistent with that observed in the prospective cohort study . Between June 2006 to June 2009 , 749 neonates were recruited into the cohort study . In 2008–09 , 124 herpangina and 9 HFMD cases were detected and 128 of them provided clinical samples ( throat swabs or serum ) for laboratory diagnosis . In addition , 20 children developing non-specific illnesses including fever , upper respiratory illness , and viral exanthema also provided clinical samples for laboratory diagnosis . Among 118 symptomatic children providing throat swabs for virus culture , 40 ( 34% ) had enterovirus isolated and 5 were EV71 . In total , 20 symptomatic EV71 infections were detected in the children's cohort study , including 5 herpangina cases , 6 HFMD cases , and 9 cases with the non-specific illnesses ( Table 1 ) . None of the EV71 cases identified in the cohort developed complications . Among the 20 symptomatic EV71 infections , all developed seroconversion and 9 of them provided throat swabs for virus culture in acute phase . In these 9 cases , EV71 and cytomegalovirus were isolated from 5 and 2 cases , respectively . Overall , the virus isolation rate was significantly lower than the seroconversion rates for detecting EV71 infections ( 4 . 2% vs . 16 . 7% , P<0 . 01 , Fisher's exact test ) . In addition , 8 asymptomatic EV71 infections were detected by seroconversion . Between January 2008 and December 2009 , 307 , 391 , 294 , and 66 children returned for follow up and serum collections at 6 , 12 , 24 , and 36 months of age , respectively . Overall , the follow-up rates at 6 , 12 , 24 and 36 months of age were 73% , 67% , 55% and 51% , respectively . The major reason for loss of follow-up was refusal of providing blood samples . As shown in Table 2 , age-specific incidence rates of EV71 infections in young children increased gradually from 1 . 71 per 100 person-years at 0–6 months of age to 4 . 09 , 5 . 74 , and 4 . 97 per 100 person-years at 7–12 , 13–24 , and 25–36 months of age , respectively . Table 3 shows the cumulative incidence rates , which increased from 0 . 65% ( 95% CI , 0 . 08% , 2 . 33% ) at 6 months of age to 6 . 46% ( 95% CI , 3 . 94% , 9 . 91% ) and 15 . 15 ( 95% CI , 7 . 51% , 26 . 1% ) at 24 and 36 months of age , respectively . Two children who acquired EV71 infections before 6 months of age were both asymptomatic and one of them had neutralizing antibody titer in cord blood ( antibody titer 16 ) . Among 7 children who acquired EV71 infections at 7–12 months of age , 2 were asymptomatic and 5 were symptomatic . The 2 asymptomatic cases both had neutralizing antibody titer in cord bloods and their maternal antibodies declined to undetectable level at 6 months of age . In contrast , only 2 of the 5 symptomatic cases had detectable EV71 neutralizing antibody in cord blood and their maternal antibodies also declined to undetectable level at 6 months of age . The 2 symptomatic cases with detectable maternal antibody in cord blood were infected at 7 and 9 months of age and the 3 symptomatic cases without detectable maternal antibody in cord blood were infected at 7 , 8 and 9 months of age .
EV71 continues to cause disease with the potential for life-threatening infections in Asian children . Development of EV71 vaccines has become a national priority in Taiwan and China . To design clinical trials of EV71 vaccines , age-specific incidence rates of EV71 infections are required to identify target populations and select study endpoints . In this study , risk of EV71 infections greatly increased after 6 months of age during EV71 epidemics , which is consistent with national surveillance data of severe EV71 infections and decline of maternal antibody by 6 months of age [23]–[24] . Furthermore , the cumulative incidence rate in young children was about 15% by 36 months of age after follow-up for 2 years , which is consistent with the results of seroprevalence studies conducted after the 1998 epidemic [23] , [28] . The follow-up rate by 36 months of age was about 51% in our study but the major reason for loss of follow-up was refusal of providing blood samples . Meanwhile , 17 ( 61% ) of the 28 EV71-infected children continue to return for follow-up . Overall , follow-up rates in EV71-infected and uninfected children differed slightly . Therefore , the substantial loss of follow-up in our study may not cause significant bias in calculation of incidence . However , our cohort study was conducted in northern Taiwan and more severe EV71 cases were reported in southern Taiwan than in northern Taiwan during the 2008–2009 epidemic . It should be cautious about extrapolating finding of this study to whole Taiwan . In addition , EV71 epidemics occurred every 3–4 years in the past 10 years in Taiwan , which may greatly affect the estimation of age-specific incidence rates . In addition , in our prospective cohort 29% of EV71 infections in young children were asymptomatic , 32% had non-specific illness , and 39% developed herpangina/HFMD . In a retrospective study conducted in Taiwan in 1999 , Chang et al . found that 29% of 484 EV71-seropositive children <6 years of age developed herpangina/HFMD and 71% were asymptomatic [23] . In a prospective hospital-based case-finding study conducted in a medical center in 2001–02 , 6% of 183 EV71 infections in children <18 years of age were asymptomatic , 13% developed non-specific illnesses , and 71% developed herpangina/HFMD [14] . It is hard to compare our study with the other two studies due to differences in study designs , age groups , laboratory diagnosis , and genotypes of circulating EV71 . Overall , the retrospective study could not differentiate EV71 infections with non-specific illnesses from asymptomatic EV71 infections and the prospective hospital-based case-finding study would underestimate the proportion of asymptomatic infections . Therefore , our prospective infant cohort study is more likely to reflect occurrences of EV71 infections in the community . Due to small sample size , our prospective study could not detect any complicated EV71 infections . In the prospective hospital-based case-finding study , 21% of 183 EV71 infections in children <18 years of age developed neurological complications such as meningitis and encephalitis [14] . Based on national severe enterovirus surveillance with virus isolation and two cross-sectional serosurveys , Lu et al . estimated that 130617 Taiwanese children aged <3 years were infected with EV71 infections in 1998 and that 273 ( 0 . 21% ) of these infected children developed neurological complications [28] . Overall , the prospective hospital-based case-finding study would overestimate the proportion of EV71 infections with neurological complications and the national surveillance data would underestimate the proportion of EV71 infections with neurological complications . Interestingly , we found that the 2 children who acquired EV71 infections before 6 months of age were both asymptomatic and their EV71 neutralizing antibody titers at birth , 6 and 12 months of age were 16 , 256 and 32 , and <8 , 128 and no serum collected at 12 months of age , respectively . Chang et al . [23] also found that children <6 months of age were more likely to develop asymptomatic EV71 infections than children >6 months of age . These observations may indicate that maternal antibody may provide protection against symptomatic EV71 infections in young infants even though EV71-specific maternal antibody may have declined to undetectable levels in these young infants . Current standard methods for laboratory diagnosis of EV71 infections include virus isolation and serum neutralizing antibody testing . In the hospital-based case finding study , serum neutralizing antibody testing ( ≥4-fold rise ) was more sensitive than virus isolation for detecting EV71 infections [12] , which was consistent with the finding of our infant cohort study . However , the serum neutralizing antibody testing needs to collect paired sera , which are not always available . Several studies have found that serum IgM tests based on single clinical specimen were more sensitive than virus isolation for detecting EV71 infections but had a major drawback of high false positives , especially for patients infected with Coxsackievirus A16 [29] , [30] . In addition , molecular tests based on single clinical specimen were also more sensitive than virus isolation for detecting EV71 infections but had a major limitation of high cost [31]–[33] . Our cohort study collected serial sera to detect EV71 infections and would also be more sensitive than molecular tests based on a single clinical specimen . Overall , enterovirus surveillance systems based only on virus isolation for detecting EV71 infections would significantly underestimate disease burdens of EV71 infections [7]–[9] . However , after introducing vaccines , vaccinated children would possess EV71 antibodies which would confound serological diagnosis of EV71 infection so it would be necessary to detect and serotype enteroviruses using virus isolation and molecular methods . Overall , harmonized and integrated laboratory diagnosis methods are urgently needed to make national enterovirus surveillance data comparable [12] . Like poliomyelitis viruses , vaccination would be the most cost-effective intervention to prevent EV71-related diseases in endemic countries . In Taiwan , the target population of EV71 vaccines has been identified to be infants <6 months of age [23] , [24] . Our prospective infant cohort study has found that EV71-related HFMD/herpangina is a suitable endpoint of vaccine efficacy trials in Taiwan . In addition , the cumulative incidence rate of EV71 infections during an EV71 epidemic was about 15% by 36 months of age , and 29% of EV71 infections were asymptomatic in young children , that are critical for designing vaccine clinical trials . Several organizations are planning clinical trials of EV71 vaccines in Asia; and EV71 vaccines could be available in the near future [12] . To successfully introduce EV71 vaccines in epidemic areas , each country needs to well characterize its EV71 epidemiology [34] , [35] . Our prospective infant cohort study would be helpful to other countries for understanding their EV71 epidemiology . | Enterovirus 71 ( EV71 ) was first isolated in California , USA , in 1969 . Since then , EV71 has been identified globally . Recently , EV71 caused several life-threatening outbreaks in young children in tropical Asia . Development of EV71 vaccines becomes national priority in several Asia countries including Taiwan . To design clinical trials of EV71 vaccines , age-specific incidence rates of EV71 infections are required to identify target populations , estimate disease burdens , select endpoints of clinical efficacy , and estimate sample size . In Taiwan , nationwide EV71 epidemics occurred every 3–4 years but age-specific incidences of EV71 infection are not available . In 2006 , we initiated a prospective cohort study in northern Taiwan to recruit neonates and follow up them . In 2008–09 , a nationwide EV71 epidemic occurred and we found that age-specific incidence rates of EV71 infection increased from 1 . 71 per 100 person-years at 0–6 months of age to 4 . 09 , 5 . 74 , and 4 . 97 per 100 person-years at 7–12 , 13–24 , and 25–36 months of age , respectively . The cumulative incidence rate was 15% by 36 months of age , and 29% of EV71 infections were asymptomatic in young children . These findings would be helpful to development of EV71 vaccines in Taiwan and other Asian tropical countries . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"public",
"health",
"and",
"epidemiology",
"epidemiology",
"pediatrics",
"and",
"child",
"health",
"pediatrics"
] | 2012 | Incidence Rates of Enterovirus 71 Infections in Young Children during a Nationwide Epidemic in Taiwan, 2008–09 |
Alveolar echinococcosis ( AE ) is caused by the metacestode stage of Echinococcus multilocularis . The inflammatory response to this infection is influenced by the interaction of the parasite with the host . We aimed to analyze human liver lesions infected with Echinococcus multilocularis and the changes of the cellular infiltrates during albendazole ( ABZ ) treatment . We analyzed liver tissue samples from 8 untreated patients , 5 patients treated with two daily doses of 400 mg ABZ for up to two months and 7 patients treated for more than two months with the same ABZ therapy . A broad panel of monoclonal antibodies was used to characterize the lesion by immunohistochemistry . A change in the cellular infiltrate was observed between the different chemotherapy times . During the initial phases of treatment an increase in CD15+ granulocytes and CD68+ histocytes as well as in small particles of Echinococcus multilocularis ( spems ) was observed in the tissue surrounding the metacestode . Furthermore , we observed an increase in CD4+ T cells , CD20+ B cells and CD38+ plasma cells during a longer duration of treatment . ABZ treatment of AE leads to morphological changes characterized by an initial , predominantly acute , inflammatory response which is gradually replaced by a response of the adaptive immune system .
Human alveolar echinococcosis ( AE ) , caused by the tapeworm Echinococcus multilocularis , is considered one of the most pathogenic zoonosis in humans with endemic areas in the Northern hemisphere and also in Western China [1 , 2] . The adult worms live in the intestine of canids , such as the red fox ( Vulpes vulpes ) . Eggs are released into the environment through the feces of canids . In humans who accidentally ingest the eggs , the multi-vesicular metacestode shows a tumor-like growth pattern predominantly in the liver . However , the disease may spread to other organs [3] . If possible , first-line treatment is radical surgery , accompanied by treatment with benzimidazole derivatives . Lifelong treatment is necessary if the patient has non-resectable lesions . The metacestode stage of E . multilocularis consists of a cellular germinal layer surrounded by an acellular laminated layer . The laminated layer synthesized by the germinal layer is the histological hallmark of the lesion [4] . Since the laminated layer is rich in polysaccharide protein complexes , these fragments have a high affinity to PAS staining and are well recognized on histological examination . The central core of the lesion is necrotic and may contain particles of protoscoleces and fragments of the laminated layer [5]; this zone is surrounded by a cellular infiltrate [6] . The monoclonal antibody Em2G11 is specific for the Em2 antigen of the E . multilocularis metacestode and exclusively stains the laminated layer as well as the cyst content in tissue sections . Additionally , the antibody marks acellular Em2-antigen-positive small particles of E . multilocularis ( spems ) inside and outside the main lesion [7] . These particles are probably shed due to the growth of the metacestode and/or the inflammatory response [8] and may play a modulatory role in the immunological process during the infection [9] . Infection with E . multilocularis in humans is characterized by modulation of the immune response , which allows the parasite to escape the immune response of the host [10] , [11] . This phenomenon is reflected by changes in the cytokine profile and the T-helper cell response . During the course of inflammation , the acute inflammatory Th1 response is gradually converted into a Th2 response in mice , reflecting the chronic phase of AE [12 , 13] . The severity of the disease may depend on the genetic background of the host and on the acquired disturbances of the Th1-related immunity [12 , 14 , 15] . The laminated layer of the metacestode , particularly its carbohydrate components , plays a major role in the evasion of cellular and humoral immunomechanisms and , furthermore , in tolerance induction and immunomodulation [16] . The Em2 antigen is a T cell-independent antigen and the response against Em2 antigen has been shown to lack antibody maturation [9] . Moreover , in contrast to Em492 antigen , the Em2 antigen does not lead to anti-CD3 apoptosis . Em492 stimulates peritoneal macrophages to produce high levels of nitric oxide leading to an inhibition of murine splenocyte proliferation [11] , therefore acting as an immunosuppressant [17] . Th2-type and anti-inflammatory cytokines , IL-10 and TGF-β , as well as nitric oxide , are involved in the maintenance of tolerance and partial inhibition of cytotoxic mechanisms [12 , 18] . The complex immune response during infection is characterized by an abnormal production of various interleukins , such as interleukin-5 [19 , 20] , IL-27 [21] , high levels of IL-10 and TGF-beta [18 , 22] . Simultaneously , insufficient production of IL-12 [23] , IL-31 , IL-33 [21] , IFN-gamma and cytotoxic T cells leads to an enhanced tolerance towards the parasite . It has been shown that the Fibrinogen-like-protein 2 ( FGL2 ) , a CD4+CD25+ regulatory T cell effector molecule secreted by T regulatory cells , plays a crucial role in the immune response by suppressing Th1 and Th17 responses [24] . In line with this observation , mice infected with E . multilocularis eggs showed up-regulation of FGL2 in the liver [25] . Long-term treatment with ABZ has improved the 10-year survival rate in comparison with untreated historical controls from 6–25% to 80–83% [26][27] . ABZ binds to beta-tubulin and inhibits absorptive functions in E . multilocularis [28] . ABZ also binds to beta-tubulin in the human host , which is very similar with more than 90% identical amino acids between the parasite and humans [29] . Treatment results in an inhibition of metacestode proliferation , and leads to destruction of protoscoleces; it inhibits formation of the germinal layer and therefore of the metacestode [27] . ABZ treatment is regarded as parasitostatic [30] , [31]; in some cases benzimidazoles show a parasitocidal action [32] in vitro . Here , we hypothesized that treatment with ABZ may have an influence on the cellular infiltrate of E . multilocularis in infected human liver tissue . Therefore , we conducted a morphological and immunohistological analysis of 20 human liver tissue lesions ( 12 treated/8 untreated with ABZ ) using a broad panel of antibodies to characterize the lesions .
The study has been approved; see: Zentrale Ethikkommission bei der Bundesärztekammer . Mitteilungen: Die ( Weiter- ) Verwendung von menschlichen Körpermaterialien für Zwecke medizinischer Forschung . Dtsch Arztebl . 2003: 1632 . Human liver tissue samples from 20 patients were retrieved from the paraffin archives of the Institute of Pathology , University of Ulm . The samples were anonymized according to German law for correct usage of archival tissue for clinical research [33] . Of the 20 liver specimens with histologically confirmed E . multilocularis infection , 8 samples are from untreated patients ( patients #1–8 ) , 5 samples from patients treated with 2 x 400 mg ABZ ( Eskazole ) per day for up to two months ( patients #9–13 ) and 7 patients treated for more than two months with 2 x 400 mg ABZ ( Eskazole ) per day ( patients # 14–20 ) . 17 cases were resection samples and 5 cases were cutting needle biopsies with more than 90% of representative tissue of the lesion . Cutting needle biopsies were performed as a diagnostic step regarding a liver lesion of unknown significance . The clinical characteristics of the patients are shown in Table 1 . The cohort was divided into three groups: group 1 = no therapy , group 2 = treatment of up to two months with a range between 4 and 37 days’ treatment , group 3: treatment of more than two months . These groups were formed on the basis of samples of patients available for the analysis . Hematoxylin and eosin ( HE ) and Periodic Schiff staining ( PAS ) staining , as well as immunohistochemistry , were performed according to standard protocols [7] . The resection specimens and biopsies were fixed in 4% buffered formaldehyde for at least 36 hours . Serial sections of about 3 μm from paraffin blocks with representative tissue were performed with a microtome . Paraffin was dissolved with xylol and ethanol . For antigen retrieval , different pretreatment methods were used according to the companies’ recommendation . As primary antibodies , the following antibodies were used: monoclonal antibody CD3 ( F7 . 2 . 38 , 1:100 dilution , DAKO , Glostrup , DK ) , CD4 ( 4B12 , 1:200 dilution , DAKO ) , CD8 ( C8/144B , 1:200 , DAKO ) , CD15 ( MMA , 1:300 , BD; Erembodegem , BEL ) , CD20 ( L26 , 1:500 , DAKO ) , CD38 ( SPC32 , 1:100 , Menarini , Florence , IT ) ) , CD68 ( PG-M1 , 1:100 , DAKO ) , eosinophil major basic protein ( EMBP ) antibody ( BMK-13 , 1:25 , Zytomed Systems; Berlin , GER ) , Em2G11 ( 1:100; kind gift of Peter Deplazes , Institute of Parasitology , University of Zürich , Switzerland ) and FGL2 ( 1:4000 , Abnova; Taipeh , TW ) . The primary antibody was diluted in Antibody Dilution solution ( DAKO ) and each slide was incubated with 50 μl in a humid chamber at room temperature for 30 min . The DAKO REAL Detection System , Alkaline Phosphatase/Red ( DAKO , Carpintera , CA , USA ) was used as the detection system according to the manufacturer’s protocols . As negative controls , staining was performed without the primary antibody . The evaluation of the immunohistological stainings was carried out in a blinded fashion by three observers ( TFEB; FJR; JN ) at a multihead microscope . Five different high-power observation fields of one section centering on the inflammation zone between the normal liver parenchyma and the necrosis were analyzed and the stained cells counted ( 400x magnifications ) . The average for each section was calculated . Regarding the possible sample error , we stained representative sections of two different blocks of tissue of three cases with the whole antibody panel . In sections of two cutting needle biopsies , only four positions were evaluated . To measure the quantity of spems , we determined the percentage of the whole necrotic area with a typical pattern of spem staining using a 25x magnification . To register the lymphatic aggregates , we counted all lymphatic aggregates larger than 1 mm within an area of 1 . 2 cm in diameter . The average and the standard deviation for each evaluation , was calculated . Furthermore , we performed a two-sided t-test type 3 with unequal variance with Excel ( Microsoft Office 2007 ) and IBM SPSS ( Statistic Version 21 , IBM Corp . ) . The result was regarded as significant for p-values p< 0 . 05 .
Using HE staining , we first defined the microscopical parameters of the lesion . All lesions , with and without treatment with ABZ , had a central necrosis of varying diameters in common; next followed an inner circle close to the necrotic zone , characterized by epithelioid cells and granulocytes , and an outer circle with lymphocytes followed by hepatic tissue . Between the outer and the inner zone , a fibrotic layer of varying diameter was found . Of note , fragments of protoscoleces were found only once in 20 samples ( case # 10 ) . On analysis of the different parameters of the inflammatory infiltrate , histological differences and similarities were noted . All lesions had the following immunohistological characteristic in common: In a CD68 staining , the macrophages and epithelioid cells were highlighted in the inner zone . CD15+ granulocytes were intermingled with the CD68+ cells in the inner circle but not in the outer circle . Some positive EMBP eosinophilic granulocytes were resident in the inner zone . The outer circle was characterized by a mixture of CD8+ and CD4+ T cells; CD8+ cells were generally more frequent than CD4+ cells . CD20+ B cells were mixed with the T cells . CD38+ plasma cells were interspersed predominantly in the outer zone . Regarding the time course of ABZ treatment , we noted differences in the composition of the cellular infiltrate , which varied in relation to the duration of treatment . In the lesions of patients treated up to two months , CD68+ and CD15+ cells were more prominent in the inner zone compared to non-treated lesions and lesions treated for over two months . With respect to T cells , CD4+ cells increased significantly and the number of CD8+ cells remained largely stable during the course of treatment . CD20+ B cells and CD38+ plasma cells increased significantly ( Figs 1 and 2 , S1 Table ) with plasma cells outnumbering B cells . Along with the increased number of T and B cells , we noted an increase in lymph follicles larger than 1 mm consisting of CD4+/CD8+ T cells and CD20+ B cells . Small particles of E . multilocularis ( spems ) increased during the course of treatment . Spem staining was most prominent during the initial treatment phase up to two months ( Fig 1 ) . Spems were detected in the necrotic area , in sinusoids , vessels and lymph follicles around the lesion . In contrast , FGL2+ cells showed a tendency to decrease under therapy , being lowest in patients having a short duration of treatment ( Fig 1 ) . These results show that the inflammatory infiltrate changes during early and late treatment with an increase in macrophages and granulocytes during the first six weeks of treatment followed by a shift to a specific cellular response with an increase in CD4+ T cells during early response; by contrast , the number of CD8+ T cells remains stable during treatment . Furthermore , CD20+ B cells , plasma cells and lymph follicles generally increase during late treatment ( Fig 3 ) . Immunohistological analysis of sections of two different paraffin blocks from one resection specimen from the same patient showed almost identical immunohistological results . This was repeated for two samples from three patients and therefore confirmed the method . The cohorts were also analyzed with a cut-off level of one month . The results showed the same trends as described for the two-months threshold ( S1 Fig ) .
To date , only very few histological/immunohistological studies characterizing the inflammatory infiltrates around the human lesions of E . multilocularis before and after treatment have been performed [34 , 35] . We detected a morphological spectrum of lesions caused by an infection with E . multilocularis which is characterized by irregular sized metacestodes [36] and the lamellar layer , which is the hallmark of these lesions [7] . We detected protoscolex remnants in only one sample , which is in agreement with the literature ( less than 10% ) [3 , 37] . Based on our morphological and immunohistological data , the lesion is characterized by an inner layer with a cellular composition typical of a non-specific response consisting of macrophages and granulocytes and an outer layer consisting of T and B cells . This confirms data from mice infected with E . multilocularis , which showed an intense granulomatous infiltration in the periparasitic area of the lesions [37] . Peripheral blood mononuclear cells and polymorph nuclear granulocytes are activated after stimulation in vitro with E . multilocularis vesicles and synthesize interleukin-8 ( IL-8 ) [38] and monocyte chemo attractant protein-1 ( MCP-1 ) [39] . IL-8 leads to neutrophil migration and activation [40] and MCP-1 attracts and activates macrophages [41] and is an attractant for CD4+ and CD8+ T cells . These findings are reflected by our data showing that T cells are present in the lesions in situ and are increased during ABZ therapy . In mice infected with E . multilocularis , the number of CD4+ and CD8+ cells is reduced , probably due to the diminished ability of antigen-presenting cells to present conventional antigens [42] . Furthermore , there is an elevation in CD4+ T cells in abortive or died-out lesions and active metacestodes are indicated by higher levels of CD8+ T cells [34] . ABZ acts as an intracellular tubulin inhibitor [28] and prevents metacestode formation . In mice , treatment with ABZ leads to loss of integrity in the germinal layer and a reduction in tumor mass [43] . Liance et al . [37] showed that rodents inoculated with E . multilocularis material from treated human patients have a decreased larval development in contrast to inoculation with samples from untreated patients . At a high concentration , ABZ leads to a collapse of the alveolar architecture of the parasite , partially dissolving the laminated layer followed by an invasion of the lesion with host inflammatory cells , such as histocytes , lymphocytes , neutrophils and eosinophils [44] . Reduction of the width of the laminated layer upon therapy [45] was confirmed in our study and degradation of the laminated layer may contribute to the observed increase of spems in and around the lesion , such as sinusoids , vessels and lymph follicles , which may influence the immune reaction [7] . In support of our immunohistological finding ABZ treatment has been shown to affect differentiated cells of E . multilocularis including the tegument , which is responsible for the production of the laminated layer [46]; therefore , it might be hypothesized that by this mechanism ABZ treatment leads to an increased immunohistological detection of spems . Taken together , we found an overall increase in the number of immune cells during the course of treatment with ABZ . This effect was enhanced in the first weeks of treatment with ABZ . Our findings support the view that the non-specific immune reaction is activated at the beginning of treatment with an increase in macrophages and granulocytes , which then reduce during later treatment; our data suggest that this response is shifted towards the specific immune response , dominated by B and plasma cells which , however , do not eliminate the infection . Therefore , ABZ treatment supports the activation of the host immune system by reducing the immunosuppressive functions of the parasite . Our data suggest that , by reducing the metabolism of the metacestode during ABZ treatment and dissolution of the laminated layer , more parasite antigens are exposed and detected by the immune system and that this may lead to a more specific immune response . Supportive of this finding is that protoscoleces not protected by the laminated layer are killed by macrophages [47] . Furthermore , there are several parasite excretory/secretory products with suppressive effects on the immune system of the host [48]; by damaging the tegument , the function of these products may be reduced and may , in turn , lead to an increase in the immune response of the host . FGL2 , secreted by macrophages and T regulatory cells , leads by various mechanisms to an suppressed immune status of the host and to a progression of the metastatic growth [49] . It has been shown in mice that FGL2 suppresses the Th1 and Th17 immune response and supports the Th2 response [24] . Our finding that the FGL2 effector molecule is reduced during ABZ treatment corresponds to these observations . This indicates that treatment with ABZ may lead to a change in the immune response towards a Th1-shifted immune response by down-regulation of FGL2 . To summarize , our histological study confirms and extends findings of in vitro and in vivo studies in mice and humans infected with E . multilocularis and may help to explain the mechanism of action of ABZ during the course of treatment of patients with an initial acute inflammatory response that is gradually replaced by the adaptive immune system . The finding that spems are increased during early treatment may point to a role of spems as mediators of this inflammatory response . | Alveolar echinococcosis ( AE ) is a life-threatening disease in humans caused by the larval stages of E . multilocularis . It has been shown that the infection in humans is associated with a modulated immune response . Depending on multiple factors , such as the stage of disease , total or partial surgical resection and albendazole ( ABZ ) therapy are treatments of choice . ABZ is known as a parasitostatic drug that has to be administered for years to suppress metacestode development . Here we compared human liver lesions before and after short and long term treatment with ABZ by immunohistochemistry using a broad panel of antibodies . We found a change in the cellular infiltrate , characterized by a shift to an infiltrate rich in T cells , B cells and plasma cells during long-term treatment with ABZ , including a pronounced detection of small particles of E . multilocularis ( spems ) . We argue that ABZ treatment is likely to change the cellular infiltrate , leading to an enhancement of the host immune response during treatment . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"immunology",
"parasitic",
"diseases",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"signs",
"and",
"symptoms",
"white",
"blood",
"... | 2017 | Albendazole increases the inflammatory response and the amount of Em2-positive small particles of Echinococcus multilocularis (spems) in human hepatic alveolar echinococcosis lesions |
The trematode flatworms of the genus Schistosoma , the causative agents of schistosomiasis , are among the most prevalent parasites in humans , affecting more than 200 million people worldwide . In this study , we focused on two well-characterized strains of S . mansoni , to explore signatures of selection . Both strains are highly inbred and exhibit differences in life history traits , in particular in their compatibility with the intermediate host Biomphalaria glabrata . We performed high throughput sequencing of DNA from pools of individuals of each strain using Illumina technology and identified single nucleotide polymorphisms ( SNP ) and copy number variations ( CNV ) . In total , 708 , 898 SNPs were identified and roughly 2 , 000 CNVs . The SNPs revealed low nucleotide diversity ( π = 2×10−4 ) within each strain and a high differentiation level ( Fst = 0 . 73 ) between them . Based on a recently developed in-silico approach , we further detected 12 and 19 private ( i . e . specific non-overlapping ) selective sweeps among the 121 and 151 sweeps found in total for each strain . Functional annotation of transcripts lying in the private selective sweeps revealed specific selection for functions related to parasitic interaction ( e . g . cell-cell adhesion or redox reactions ) . Despite high differentiation between strains , we identified evolutionary convergence of genes related to proteolysis , known as a key virulence factor and a potential target of drug and vaccine development . Our data show that pool-sequencing can be used for the detection of selective sweeps in parasite populations and enables one to identify biological functions under selection .
In addition to their obvious importance as threats to physical and economical well-being , parasites constitute an interesting group of organisms in which to investigate adaptation and selection . Parasites closely interact with their hosts and entirely depend on them for reproduction and survival . Thus , any change in a host population , which for example decreases parasite ability to penetrate host tissue , will reciprocally select for a change in the parasite such as mechanisms favouring evasion of host resistance to infection . Such an evolutionary arms race [1] , [2] has been well studied in Schistosoma mansoni during the interaction with its intermediate host snail [3]–[6] . S . mansoni is a parasitic platyhelminth infecting humans in Africa , the Arabian Peninsula , and South America . It is responsible for the most severe parasitic disease after malaria in terms of morbidity [7]–[10] killing 200 , 000 people ( WHO Technical Report Series 912: prevention and control of schistosomiasis and soil transmitted helminthiasis ( WHO , Geneva , 2002 ) ) . S . mansoni's life cycle is characterized by the passage through two obligatory hosts . Parasite eggs are emitted with the faeces of the definitive human or rodent host , but can also accumulate in the liver and cause the symptoms of the disease . When the eggs in host faeces come into contact with water , free-swimming larvae ( miracidia ) hatch and actively seek their specific intermediate host snails . After active penetration through the tegument , the parasite develops via a primary ( mother ) sporocyst , then daughter sporocysts releasing the cercariae that infect the vertebrate host . Then , sexual differentiation takes place within this definitive host and the mating of male and female worms leads to new egg production . In natural populations , snail/schistosome combinations present different levels of compatibility ( i . e . the ability for the parasite to penetrate and develop in the host ) [3] , [4] . Previous comparative approaches between a Brazilian ( BRE ) and Guadeloupean ( GH2 ) S . mansoni strains showed that while the first is compatible with a sympatric Biomphalaria glabrata strain from Brazil , the latter is much less compatible [7] . Compatibility levels of these strains ( and several others not presented in this study ) are stable after several years of maintenance under laboratory conditions ( Supplementary figure S1 ) . This particular feature has made it possible to elucidate partially the molecular basis of the compatibility polymorphism at the global proteomic [3] , [4] , [7]–[11] and epigenetic scales [12] . In addition , these strains present significant differences in several life history traits , such as chronobiology in cercarial emission [11] and number of mother sporocysts [7] . They are therefore ideal models to investigate signatures of selection at the whole genome scale in S . mansoni and to elucidate the genetic basis of phenotypic variation for this parasite . Three major classes of polymorphisms are responsible for variations in the genotype: ( i ) single nucleotide polymorphisms ( SNPs ) , a modification of the nucleotide information at a single position , ( ii ) insertions and deletions ( indels ) and ( iii ) structural polymorphisms such as copy number variation ( CNV ) , resulting from tandem duplications of genome segments . These variations can produce by chance favourable , neutral or deleterious phenotypic variations leading to greater , equal or lower fitness , respectively . With the recent advent of Next Generation Sequencing ( NGS ) , hundreds of complete eukaryotic genomes are now available together with huge amounts of data on gene expression and genomic polymorphisms . Standard methods from population genetics to detect selection [3] , [4] , [7] , [13] , [14] can theoretically be applied to large-scale genomic data , but improvements have been needed to take into account the SNP ascertainment process [15] . These improved methods have been successfully applied to answer different biological questions dealing with the adaptive process , such as biological invasions [11] , [16] , gene selection in human populations [7] , [17] , and domestication [18] , [19] . An obstacle for small organisms is the amount of DNA required for library generation and sequencing . In this context , pool-sequencing provides an alternative . The method consists of DNA extraction from a large number of individuals from a population ( pool ) followed by massive sequencing . SNP frequencies and distributions are then extracted from the sequencing data . While in principle straightforward , current methods for the detection of selective sweeps from pooled sequence SNPs had to be adapted . We used pool-HMM [20]–[22] in the current work . In this statistical method , the inference of selective sweeps ( i . e . the elimination of standing variation in regions linked to a recently fixed beneficial mutation ) is based on the allele frequency spectrum ( AFS ) assessed in a sliding window along each chromosome . Our approach allowed sequencing a sufficient number of individuals at moderate costs and to apply population genetics approaches without affecting population genetics estimators [23] . CNV detection is often performed by microarray-based methods comparing two genomes [1] but recent developments such as CNV-seq [24] have extended the method to NGS data . It takes advantage of the variation in the number of short reads aligned in a sliding window along each chromosome to assess CNV at the whole genome scale . As differences in CNV within the same species have been recognized to be involved in adaptive evolution [25] , [26] , we used NGS data to investigate the proportion of this type of variation in our comparative approach . In the case of S . mansoni , we benefit from the availability of a genome assembly [27] , facilitating the use of NGS-based strategy for the detection of regions under selection . In this work , we have characterized SNPs from whole genome pool-sequencing data and described their distribution before applying the new population genomics method of Boitard et al . [21] , [22] for detecting selective sweeps as signatures of past selection in these two populations . We also describe differences in copy number variations ( CNV ) between the two populations as another signature of selection . We finally investigated functional aspects of all genomic regions corresponding to either private selective sweeps or structural variation and proposed evolutionary and ecological interpretations to these genetic differences found between and within BRE and GH2 strains at the whole genome scale .
We adhered to national ethical standards established in the writ of February 1st , 2013 ( NOR : AGRG1238753A ) setting the conditions for approval , planning and operation of establishments , breeders and suppliers of animals used for scientific purposes and controls . The Ministère de l'Agriculture et de la Pêche and Ministère de l'Education Nationale de la Recherche et de la Technologie provided permit A 66040 to our laboratory for experiments on animals and certificate for animal experimentation ( authorization 007083 , decree 87–848 ) for the experimenters . Housing , breeding and animal care followed national ethical standards . DNA of twenty individual worms ( 10 males and 10 females ) of each strain was individually extracted and genotyped using 15 microsatellite markers . Methods for DNA extraction and microsatellite amplifications were previously published [28] . Two Schistosoma mansoni strains , one collected in Brazil ( BRE ) and the other in Guadeloupe ( GH2 ) were used in this study . Each strain was maintained in its sympatric intermediate host ( the mollusk Biomphalaria glabrata ) and in the mouse ( Mus musculus ) or the hamster ( Mesocricetus auratus ) as a definitive vertebrate host . Genomic DNA was isolated from a pool of a hundred adult individuals . Parasite tissues were digested with 300 mg/L Protease K ( Merck , Darmstadt , Germany ) in 20 mM TRIS pH 8; 1 mM EDTA; 100 mM NaCl; 0 . 5% SDS at 55°C overnight . DNA was extracted by two successive rounds of phenol/chloroform followed by chloroform extraction . Precipitation of DNA was done by adding an equal volume of isopropanol/sodium acetate at room temperature [29] and DNA was collected by centrifugation at 12000 rpm for 30 min . After washing with 1 mL of 70% ethanol and air drying , DNA was dissolved in 200 µL of ultrapure water . Quality control and quantification were performed using a spectrophotometer ( BioPhotometer , Eppendorf AG , Hamburg , Germany ) . Twenty µg of DNA was sent to the Montpellier ( France ) GenomiX facility for Illumina sequencing . Both pooled DNA samples thus represented a random sample from BRE and GH2 populations . If not otherwise stated , reagents were purchased from Sigma-Aldrich ( St . Louis , USA ) . If not otherwise stated , software was used with default parameters . The Fastx-toolkit version 0 . 0 . 13 ( http://hannonlab . cshl . edu/fastx_toolkit/index . html ) was used for quality control and initial cleaning of the sequencing reads . Adaptators were first removed and the three last bases were trimmed from reads because they showed global poor quality scores ( Phred quality score <24 ) in most reads . We then filtered reads by their quality score and retained only reads for which at least 90% of the bases had a minimum Phred quality score of 24 ( corresponding to less than 1 incorrect base call in 100 and more than 99% of base call accuracy ) . Paired-end data were then considered as two subsets of single-end data . Reads were aligned onto the reference genome version 5 . 2 [27] with the Bowtie software version 0 . 12 . 7 [30] . Because 47% of the S . mansoni genome consists of repeated sequences [31] we did not allow for alignment of reads that matched more than one time to avoid false positives in subsequent SNP calling . We allowed 2 mismatches and used the “best” and “strata” options . SNP calling was done with Freebayes software version 0 . 9 . 5 ( https://github . com/ekg/freebayes ) with parameters –min-coverage –min-alternate-fraction –min-alternate-total –pooled –ploidy 20 . We tested a minimum coverage of 10 or 20 reads per SNP position ( –min-coverage 10 or 20 ) with a minimum of 2 reads supporting the variant allele ( –min-alternate-total 2 ) and with a minimum frequency of variant of 0 . 02 ( –min-alternate-fraction 0 . 02 ) . By retaining only SNP that met these conditions , we prevented mistakes in SNP discovery due to errors during base calling and we allowed the discovery of rare variants ( up to a frequency of 0 . 02 ) . Among all SNPs , we identified rare and frequent SNPs by using a 20% threshold as proposed in other studies [17] , [32] . In silico verification of SNP calls was performed by choosing at random 20 SNPs per chromosome and per strain for each minimum coverage value 10 and 20 , corresponding to a total of 640 SNPs . We visually checked the number of reads , the number of variant allele and the alternative base identity with the Next Generation Sequence Assembly Visualization software Tablet [33] . All further analyses were done on the dataset obtained with the option –min-coverage 20 . To confirm SNP calling , we PCR amplified ( primers in Table S8 ) and re-sequenced 14 DNA fragments covering a total of 22 SNPs . We used a specific melting temperature for each primer pair ( Table S8 ) . Sequencing was performed for each PCR product in forward and reverse directions by the GATC biotech AG Company ( Konstanz , Germany ) using Sanger technology . Sequences were aligned on the reference sequence with Sequencher software version 4 . 5 ( Gene Codes Corporation , Ann Arbor , USA ) and SNPs were visually confirmed . We used CNV-seq to detect copy number variation [24] using a 10−6 p-value threshold . Briefly , this algorithm uses variation in sequence coverage in a sliding window between genomic reads mapped to a reference genome to detect variations in sequence copy numbers . Plots were drawn using the associated cnv package in R . We manually checked protein-coding genes contained in the 20 largest genomic regions and the 20 regions with the highest difference in copy number between the BRE and GH2 strain by using the coordinates of the regions in the Schistosoma mansoni local Gbrowse instance of the genome ( http://genome . univ-perp . fr ) . We used standard tools available on a local Galaxy instance [34] to extract information from the pool-sequencing data . Samtools version 0 . 1 . 18 [33] was used to produce mpileup files with read coverage information and call quality from the alignment BAM files . Nucleotide diversity was evaluated using the unbiased Tajima's Pi and Watterson Theta estimators proposed by Futschik and Schlotterer [35] and implemented in Popoolation [36] . Briefly , along each chromosome , each estimator was computed in 50 kb sliding windows ( with an overlap of 25 kb between consecutive windows ) with the following additional options –min-count 2 –min-coverage 4 –max-coverage 400 –min-qual 20 . Considering smaller ( 10 kb ) or larger ( 100 kb ) window size did not affect the results . Fixation index ( Fst ) was assessed on SNPs by using Popoolation2 [37] . We considered all SNPs for which at least 6 reads supported the minor allele for both population simultaneously ( –min-count 6 ) , and with a coverage ranging between 20 and 200 reads ( –min-coverage 20 ; –max-coverage 200 ) . Mean fixation index between BRE and GH2 was calculated as the average of all Fst values obtained for individual SNPs . In the current version of the genome , the Z and the W chromosomes are put into a single linkage group [27] despite their physical separation into two chromosomes . SNPs could therefore reflect differences between Z and W and not between target and reference genome . For this reason we excluded them from further analyses of selection . To identify footprints of selection on the 7 autosomes of Schistosoma mansoni from BRE and GH2 pool sequences , we relied on the approach recently proposed by Boitard et al . [21] implemented in the pool_hmm program [22] . The following option were used to run the analyses:-C 1000 -k 1e-10 –pred -t θw , where θw corresponds to the average unbiased Watterson theta estimator of nucleotide diversity for the population of interest ( see above ) . This method relies on the study of the allele frequency spectrum ( AFS ) within sliding windows along the genomic sequences . The AFS is expected to be distorted in regions subjected to selection . The model allows estimating the probability of each SNP to belong to one of the three possible ( hidden ) states ( i ) neutral , ( ii ) intermediate and ( iii ) selected . In practice , one of the critical parameter of the model ( defined with the –k option ) is the transition probability q assumed between states . The larger the q , the less evidence is required for transition to selection and the more sweep candidates will be detected . In our analyses , we used q = 10−10 but also tested less stringent values ( q = 10−9 ) . As shown in Table S9 and as expected , this leads only to a slightly higher number of footprints . We identified overlapping and private selective sweeps between the two strains by comparing their genomic position for each chromosome . As some genomic region swept in one strain could overlap several shorter regions swept in the other strain ( and vice versa ) , we counted overlapping regions as the exact number of genome fragments really overlapping ( i . e . counting the number of shorter instead of counting only the larger region covering them ) . Private selective sweeps of each strain correspond to all genomic regions that strictly did not overlap any sweeps of the other strain . Identification of known transcripts in private selective sweeps for both strains was performed using the genome coordinates of the regions and the Schistosoma mansoni local Gbrowse instance of the genome . Confidence index was calculated for each specific selective sweep as the maximum of −log ( 1-qi ) over the window , where qi is the posterior probability of hidden state “Selection” given after simulations . To characterize the molecular functions of the transcripts contained in selective sweeps , we first performed functional annotations of a recent S . mansoni transcriptome ( unpublished results ) by using the Blast2GO software [38] . After the blast step , we mapped gene ontology ( GO ) terms ( BRE: NodeScore = 10 , alpha = 0 . 4; GH2: NodeScore = 15 , alpha = 0 . 2 ) . We then scanned the proteins with Interproscan [39] and performed GO-Enzyme code mapping to improve annotations before running the annotation step . We finally merged results from these three annotation methods before making functional analyses . We checked if our lists of transcripts matched with proteins that were identified in an earlier study comparing these two strains at the proteomic level [3] using tblastn on the 5 . 2 version of the S . mansoni genome . Results were verified by visual inspection on a local GBrowse instance of the genome and transcriptome . We then performed functional analyses with Blast2GO tools for the list of transcripts from selective sweeps but not for the list of protein-coding genes from CNV data because this latter list was not exhaustive ( see the “Structural variants” previous section ) . Enrichment analysis was performed for each strain by using the Fisher's exact test with the P-value filter mode set at the default value of 0 . 05 ( Bonferroni correction is applied ) . Combined graphs were then drawn for each strain from two kinds of data: i ) the total list of transcripts found in selective sweeps and ii ) the reduced list of transcripts obtained after enrichment analysis . In both cases , Score alpha and Seq Filter value were set at the default values of 0 . 6 and 5 respectively . Graphs displaying process or function were built on the node score criterion set at the value of 10 or 15 depending on the graph complexity obtained . We also compared molecular pathways in which selection was found by building KEGG maps for each strain within the Blast2GO application . SNPEff was used to scan synonymous and non-synonymous SNPs in exons of the whole genome of our parasite model . We downloaded the GFF3 file for the latest assembly ( v5 . 2 , nov . 2011 ) from the Sanger centre ( ftp://ftp . sanger . ac . uk/pub/pathogens/Schistosoma/mansoni/genome/GFF/Smansoni_gff_21032012 . tar . gz ) and modified it using custom scripts so that gff data and fasta data were separated . We then built a SnpEff database using snpEff . jar with “build -gff3” . A total of 13 , 385 genes and 14 , 395 transcripts were detected . The snpEff . config file was modified accordingly and snpEff run with the following parameters: -c snpEff . config -i vcf -o txt -upDownStreamLen 5000 -no None –stats . If not otherwise mentioned , statistical analyses were done on R version 2 . 15 . 1 [40] . To test if the SNP density was correlated to the chromosome or the strain , we constructed linear models and tested for the significance of parameters ( p>0 . 05 ) by analyses of variance ( ANOVAs ) . Sequencing reads are available at the NCBI sequence read archive under study accession number SRP016500 ( alias PRJNA177787 ) . Freebayes output ( SNPs ) is available in vcf and mpileup format at http://2ei . univ-perp . fr/ ? page_id=2007 and http://methdb . univ-perp . fr/downloads/ .
Since pool-sequencing is not yet an established technique for detection of selection we reasoned that it would be suitable to use strains with presumably low genetic diversity . We had earlier characterized the proteome of two laboratory strains of S . mansoni of different geographic origins [41] and we had characterized their life history traits such as host compatibility in detail [42] . A comparison of the epigenomes of both strains had identified differences in chromatin structure in several loci , among them a metalloprotease of the neutral endopeptidase ( NEP ) family potentially involved in immuno-modulation of B . glabrata [12] . A preliminary assessment of genetic variation in both strains was performed by sequencing 15 microsatellites previously described [28] . All microsatellites markers of the Brazilian ( BRE ) strain were fixed and homozygous . In the Guadeloupean ( GH2 ) strain only one marker ( SMDO11 ) showed two distinct genotypes . Among the 14 fixed markers in the two strains , 6 are shared by BRE and GH2 . We concluded that both strains were sufficiently genetically homogenous to perform massive sequencing of a population and phenotypically sufficiently well characterized to potentially link sequence variations to phenotypic traits . Sequencing of BRE and GH2 strains produced roughly 500 , 000 , 000 raw clusters each ( Table S1 ) . About 50% passed quality checking ( Table S1 ) and were subsequently aligned to the unique sequences of the reference genome ( strain NMRI ) . Unsurprisingly , we successfully aligned only 46 . 95% and 59 . 29% of high-quality reads for BRE and GH2 strains , respectively . This is consistent with the large proportion ( 47% ) of repetitive sequences in the S . mansoni genome [31] . Sequencing data are available under study accession number SRP016500 ( alias PRJNA177787 ) at the NCBI sequence read archive . We used the pool-HMM method [20]–[22] to detect selective sweeps from pool-sequencing data . The question of selective sweeps on sex chromosomes was not addressed in this study . We detected a total of 121 and 151 selective sweeps across the 7 autosomes for BRE and GH2 , respectively ( Figure 2 ) . We counted a total of 146 overlapping regions , and identified 12 and 19 private selective sweeps for BRE and GH2 , respectively ( Figure 2 ) which were differently distributed along the 7 chromosomes ( Table 3 ) . Most of these private selective sweeps spanned 100 kb to 1 Mb which corresponded to larger regions than those identified with this method for the X chromosome of Drosophila melanogaster [21] . Along autosomes , we identified 10 , 982 and 14 , 811 SNPs in exons for BRE and GH2 respectively . Proportions of non-synonymous ( NonSyn- ) and synonymous ( Syn- ) SNPs within and outside swept regions were similarly distributed for both strains ( Supplementary figure S6 ) . Interestingly , 49% of them were Non-Syn-SNPs and found within selective sweeps . We also found more than three times more Non-SynSNPs than Syn-SNPs in private selective sweeps .
In this study we have sequenced the genomes of two Schistosoma mansoni populations , BRE and GH2 , originating from Brazil and Guadeloupe , respectively . Next generation pool-sequencing was applied to this metazoan parasite to analyse genomic variations ( SNP and CNV ) and to scan for selective sweeps . By mapping genome reads of these two strains to the S . mansoni NMRI strain reference genome [27] , [52] , we discovered hundreds of thousands of SNPs and used these data to make the first SNP map for S . mansoni at the whole genome scale . We made these data available for further analyses . Although both strains have been maintained in the lab for more than thirty years ( corresponding to approximately one hundred full life-cycles ) , life history traits such as compatibility levels with the intermediate host as well as chronobiology were maintained overtime ( Supplementary figure S1 ) [7] , [11] . Earlier investigations had concluded that populations maintained in the laboratory rapidly decreased in diversity to become monomorphic based on the analyses of nine neutral microsatellite markers [28] . It was also previously demonstrated that the diversity of S . mansoni decreased dramatically after the first life cycles of laboratory maintenance , based on the analyses of 15 microsatellite markers [28] . Our global comparative approach of BRE and GH2 strains was thus fully justified by their high level of inbreeding and their well-described compatibility polymorphism at the phenotypic and molecular scales [3] , [7] , [12] , [41] , [51] , [53] . The use of pool-sequencing allowed for evaluating the genome-wide diversity ( SNP distribution and density , SNP frequencies and copy number variations ) for each population . Intra-population diversity was low ( confirming the results of the earlier microsatellite study ) based on the whole-genome nucleotide diversity indexes Tajima's Pi and Watterson Theta and compared to other studies considering nucleotide diversity at this scale [54] , [55] . Moreover , several lines of evidence indicate a high divergence between populations: ( i ) we found a high number ( and thus a high density ) of single-base polymorphisms for both strains when compared to the NMRI reference genome ( 1 . 35 and 1 . 96 mutated site per kb ) , which was clearly higher than the values ( i . e . between 0 . 312 and 0 . 857 per kb ) found after the re-sequencing of lab strains from Entamoeba histolytica , another human parasite [56] ; ( ii ) a large majority of SNPs ( 90 . 8% ) we identified were private , either to BRE ( 35 . 8% ) or GH2 ( 55 . 0% ) strains , indicating specific polymorphic loci; ( iii ) mean fixation index was clearly very high , certainly as a result of the only few shared and variable SNPs ( Table 1 ) and ( iv ) large differences in copy number variations were identified between both strains . All these genome-wide diversity results are fully consistent with the geographic separation of the two strains in the wild since their introduction in the New World from Africa , hampering any genetic exchange , and separate cultivation for decades in the laboratory ( Supplementary figure S1 ) . Interestingly , rare SNPs represent the most important class of private SNPs , thus demonstrating that long maintenance in the lab conditions , even if it would largely decrease diversity , could not fully erase segregating polymorphisms . Using SNP frequencies within each population , we applied a recently published method of population genomics to identify selective sweeps [21] , [22] . The sweep detection method of Pool-hmm [21] , [22] uses both the density of segregating sites and the allele frequency pattern among segregating sites to distinguish sweep regions from neutral regions . In populations with low genetic diversity the length of regions with low genetic diversity in the genome is expected to be higher and therefore mapping resolution lower . Low genetic diversity of our populations will affect the power of the method ( false negative sweep detection rate ) and the resolution of the sweeps but not its robustness ( false positive sweep detection rate ) . The high number of selective sweeps for both BRE and GH2 strains indicated that a high selective pressure has operated at the whole genome scale . Attributing these signatures of selection to either long evolutionary history or adaptation to laboratory conditions will now require comparison with strains sampled from the field . We therefore focused on private ( i . e . specific non-overlapping ) selective sweeps that may not correspond to adaptation to the common laboratory conditions but to long-term evolutionary processes instead . It also helped to increase resolution because the private selective sweeps all constituted smaller regions than other sweeps . We may have here under-estimated the number of regions involved in adaptation and missed some relevant regions that would be identified in further work on field strains . The robustness of the method we used ensures a true and confident discovery of private sweeps , which was confirmed by the high proportion of non-synonymous SNPs that was found within selective sweeps ( Supplementary figure S6 ) . As differences in copy number variation ( CNV ) within a same species have been recognized to be involved in adaptive evolution [25] , [26] , [57] , we also investigated such variations as clues of selection between BRE and GH2 . We used the recently published CNV-seq method [24] that takes advantage of the high throughput sequencing data and is suitable for pairwise comparisons . The high number ( 2 , 003 in total ) of CNV found across the whole-genome of S . mansoni reinforced our hypothesis of high selective pressure and even suggested that these selective pressures were different between both strains and contributed to shape the specific genomic landscape observed for each strain . Functional annotations and analyses gave more biological significance to these specific signatures of selection . Exploring private selective sweeps , we identified 592 ( BRE ) and 791 ( GH2 ) transcripts that have been potentially subjected to selection , but some of them could also have been identified because of their genomic proximity with truly selected genes . We therefore regrouped genes in regions under selection and/or with CNV by function using GO terms . Genes that did not form functional groups ( outliers ) were considered as having little or no significance . Overall , biological processes specific to each strain emerged from functional analysis , among which there is cell-cell adhesion for BRE and reduction-oxidation reactions , potentially involved in ROS production or ROS scavenging , for GH2 . Earlier studies postulated that an evolutionary arms race between snail host and parasite operates preferentially on immune effectors for the Brazilian strain and on immune recognition for the Guadeloupean strain [3] , [42] . This is consistent with negative selection process acting on these pathways in the two strains . Two pathways involved in N-glycan biosynthesis were also found under selection in BRE . This finding was consistent with previously observed differences between BRE and GH2 in their glycosylation level for SmPoMucs [51] . As SmPoMucs were shown to be involved in compatibility polymorphism , and because the level of glycosylation is directly related to compatibility , we argue that differential selection in the N-glycan biosynthesis pathway may be responsible for this compatibility polymorphism between the two strains . In summary , we find correspondence between observed life-history traits of the different strains and genomic regions under selection . However , we see this rather on the biological function and metabolic pathway level and not on the level of individual genes . Despite strong genomic divergence between BRE and GH2 strains , functional analyses revealed interesting evolutionary convergence . From non-overlapping lists of transcripts ( because only private sweeps were analysed ) , we highlighted two common biological processes and three common molecular pathways as targets of selection . More particularly , we independently identified numerous proteases in genomic swept regions or in regions showing copy number variations in both strains , indicating an evolutionary convergence to this function . Proteases are indeed key virulence factors of parasites and particularly trematodes , as they are involved in a number of biological processes such as host tissue invasion/migration , nutrition from host substrates ( e . g . haemoglobin degradation ) , immune evasion and more generally host-parasite interactions [58] , [59] . Phylogenomic analysis had shown that protease-encoding genes , including cathepsins for which we found two genes under selection in the BRE strain , were expanded in the Schistosoma lineage [60] . Cathepsins are secreted proteases and due to their importance in host-parasite interactions they are considered to be promising targets for the development of novel chemotherapeutic drugs and vaccines against schistosomiasis . Simoes et al . [61] previously described SNPs in the cysteine protease Cathepsin B vaccine target gene that are involved in significant conformation changes leading to an alteration in antibody binding to the protein . Variability among Schistosoma species at the biochemical level has been described earlier for cathepsin B-like activity [62] . Our results suggest that different selective pressures operated on this gene , and we propose that using this as a target for vaccine development could rapidly become inefficient in natural populations . Cathepsin D is an aspartyl protease originating from successive gene duplication events in the parasite lineage after its diversification from other metazoans and was proposed to be involved in adaptation to the parasitic lifestyle [49] . Here , we also showed that this gene could be under differential selection between strains of the S . mansoni species . Further analyses of these two particular genes have to be conducted to investigate the functional basis of this evolutionary signature . From CNV data , we identified an M13 family peptidase , a type II membrane metallo-endopeptidases acting on extracellular substrates [48] , belonging to a family of proteases with other member involved in one of the epigenetic differences found between these two strains in a previous study [12] . Characterized members of the family such as neprilysin act on polypeptides smaller than 40 amino-acids . Considering evolutionary relationships among members of the family , it has been proposed that proteases of this family fulfil a broad range of physiological roles [63] . In nematodes , a neprilysin-like protease is involved in locomotion and pharyngeal activity [64] . Combination of our results indicated that the differences in BRE and GH2 life history traits [3] , [11] could result from the selection and maintenance of different proteases , leading for example to different compatibility levels with the intermediate host . Altogether , these elements support the hypothesis that the proteolysis function has evolved through different ways within different populations of the S . mansoni species . Evolutionary convergence was generally described between species [65] or communities [66] to explain adaptation of different species to a particular life style . Proteolysis could thus be a promising function to further focus on because of its key role in parasitic lifestyle , either to better understand parasite evolution and adaptation to its hosts or to develop new treatment or preventive strategies . Since the function of most genes in the S . mansoni genome has not yet been confirmed by functional studies , further investigations are nevertheless needed to objectively demonstrate the above conclusions . It is important to note that the 47% of repetitive elements in the S . mansoni genome were not investigated in this work because we excluded all reads that matched more than one time to the genome to allow SNP and CNV identification . We may have thus under-estimated the number of regions under selection . The recent de novo repeat assembly [31] should facilitate such a study on the repetitive part of the genome . The identification of SNPs in segmental duplications could also open new perspectives in S . mansoni evolutionary studies . As gene duplication is a mechanism of genomic adaptation to a changing environment [67] , it could be largely involved in parasite adaptation as it was suggested earlier with peptidase families [49] . In conclusion , our work provides one of the first examples of a comparative genomic approach based on population sequencing . A genome-wide comparison of single-nucleotide and structural polymorphisms combined with population and functional analyses allowed us here to identify signatures of selection in two S . mansoni populations . Even if all the life history traits and proteome complexity have not found their genetic bases through this study , we were able to identify selection acting on specific functions involved in parasitic lifestyle . Notably , our integrative approach highlighted that selection can act on a same function but through different pools of genes , which clearly suggests evolutionary convergence within schistosomes . However , high-throughput genotyping of a larger number of populations ( with the pool-sequencing method ) is expected to shed more light into evolutionary history and bases of adaptation of Schistosoma mansoni . In particular further comparative analyses of field isolates of Old World and New World strains could help to better understand the evolutionary history and diversification of the parasite since its “out-of-Africa” origin [68] . Such future studies would also help us to identify more polymorphisms associated with the intermediate host compatibility and to clarify how the rapid adaptation of the parasite to Biomphalaria glabrata has occurred during the colonization of the New World . | Adaptation of parasites to their environment is governed by the principle of selection . Favourable mutations are fixed in populations while deleterious mutations are progressively eliminated . Here , we aimed to find signatures of selection in two strains of Schistosoma mansoni , the causative agent of intestinal schistosomiasis . The strains differ in specific characters , in particular in their capacity to infect intermediate host snails . The reason for this is unknown and understanding it could help control the spreading of the disease . Finding footprints of adaptation to different snail hosts would lead to the discovery of genes that are particularly important for the interaction . Since a single parasite does not contain sufficient DNA to be sequenced , we pooled several individuals , sequenced them as a whole analysed them . In the regions under selection we found genes that are indeed linked to the parasitic lifestyle . We also discovered that natural selection led to diversification of genes that are related to proteolysis , the process by which the parasite destroys host tissue . The related proteins are considered good targets for drug development and vaccination . Our results suggest that in natural populations many variants of these genes exist and that they evolve rapidly , which might hamper therapeutic approaches . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2013 | Private Selective Sweeps Identified from Next-Generation Pool-Sequencing Reveal Convergent Pathways under Selection in Two Inbred Schistosoma mansoni Strains |
Rabies virus ( RABV ) is a highly neurotropic pathogen that typically leads to mortality of infected animals and humans . The precise etiology of rabies neuropathogenesis is unknown , though it is hypothesized to be due either to neuronal death or dysfunction . Analysis of human brains post-mortem reveals surprisingly little tissue damage and neuropathology considering the dramatic clinical symptomology , supporting the neuronal dysfunction model . However , whether or not neurons survive infection and clearance and , provided they do , whether they are functionally restored to their pre-infection phenotype has not been determined in vivo for RABV , or any neurotropic virus . This is due , in part , to the absence of a permanent “mark” on once-infected cells that allow their identification long after viral clearance . Our approach to study the survival and integrity of RABV-infected neurons was to infect Cre reporter mice with recombinant RABV expressing Cre-recombinase ( RABV-Cre ) to switch neurons constitutively expressing tdTomato ( red ) to expression of a Cre-inducible EGFP ( green ) , permanently marking neurons that had been infected in vivo . We used fluorescence microscopy and quantitative real-time PCR to measure the survival of neurons after viral clearance; we found that the vast majority of RABV-infected neurons survive both infection and immunological clearance . We were able to isolate these previously infected neurons by flow cytometry and assay their gene expression profiles compared to uninfected cells . We observed transcriptional changes in these “cured” neurons , predictive of decreased neurite growth and dysregulated microtubule dynamics . This suggests that viral clearance , though allowing for survival of neurons , may not restore them to their pre-infection functionality . Our data provide a proof-of-principle foundation to re-evaluate the etiology of human central nervous system diseases of unknown etiology: viruses may trigger permanent neuronal damage that can persist or progress in the absence of sustained viral antigen .
Rabies is a fatal neurological disease of animals and humans for which there is no treatment once symptoms develop . The disease is caused by infection of the central nervous system ( CNS ) with the single-stranded RNA virus , Rabies virus ( RABV ) . Infection results in dramatic neurological symptoms—aggression , hyperactivity , muscle weakness , paralysis , coma—invariably leading to fatality . The precise etiology of rabies pathogenesis is unknown and hypothesized to be either neuronal death or dysfunction . However , whether infected neurons can survive infection and the resultant immune response is unknown . Moreover , if these neurons survive , whether they are functionally restored to their pre-infection competence has not been determined in vivo for RABV , or for any neurotropic virus . Analysis of human brains post-mortem reveals surprisingly little tissue damage and neuropathology , considering the dramatic clinical symptomology [1] , [2] . As seen for other viral infections both RABV replication and resultant anti-viral immune responses are believed to be non-cytolytic; the latter mediated by cytokines , including type I interferons , and neutralizing antibodies [3]–[7] . Acute infection induces global upregulation of proinflammatory and innate immunity genes , including IL-6 , TNF-α , type I interferons , complement cascade genes , and toll-like receptors within the brain [8]–[10] . Though there is some evidence that infection induces morphologic changes in infected neurons [11] , [12] , there is a distinct lack of overt histopathological changes indicative of apoptosis or necrosis [1] , [2] . Attempts to recapitulate this ex vivo have been difficult; some viral strains induce neuronal apoptosis in tissue culture , while others do not [12]–[18] . This demonstrates the importance of studying neuronal cell fate in an animal model . An alternative hypothesis is that neuronal dysfunction , rather than cell death , is responsible for the clinical features and fatal outcome in rabies . Neurological abnormalities are obvious , but studies in experimental rabies have revealed other phenomena , including disappearance of rapid eye movement ( REM ) sleep and initiation of facial twitching , called myoclonus , prior to development of classic symptoms . It was also found that brain electrical activity terminated about 30 minutes before cardiac arrest , indicating that cerebral death precedes organ failure [19] , [20] . This correlates with functional deficiencies observed during acute experimental RABV infection , including altered expression of proteins involved in synapse communication and ion homeostasis [21] , as well as neuronal depolarization and decreased neurotransmitter binding [22]–[26] . A challenge in studying the longevity of infected neurons is identifying and isolating them after resolution of the acute viral infection . In the absence of a permanent “mark” on once-infected cells , it is impossible to decisively answer the question as to whether infected cells survive and regain their pre-infection functionality . Instead , more general metrics of CNS health have been used , including histopathological assessments , absence of cell death ( via assays with known limitations in vivo , such as TUNEL staining ) , and in vitro studies using neuroblastoma cells or primary neurons that may not faithfully recapitulate the biology of an infected neuron in vivo . Furthermore , control of viral replication by host immune responses in immunocompetent animal models may limit the infection to only a few cells , and apoptotic loss of these few cells may not be readily detected by most methods . Here we present a novel approach to study neuronal cell fate after RABV infection . We infected Cre reporter mice with a sub-lethal dose of recombinant RABV expressing Cre-recombinase ( RABV-Cre ) to switch neurons constitutively expressing tdTomato ( red ) to expression of a Cre-inducible EGFP ( green ) , permanently marking neurons that had been infected in vivo . This model allowed us to monitor neuronal survival after infection and to isolate neurons that resolved infection to characterize gene expression profiles relative to uninfected neurons . Our results support the notion that the majority of neurons survive infection , but remain impaired; this may account for the CNS disease caused by this neuropathogen .
In order to identify and isolate cells ( specifically , neurons ) after resolution of a viral infection , we adopted a double-fluorescent Cre reporter mouse model [27] . These Cre reporter mice constitutively and ubiquitously express membrane-targeted tandem dimer Tomato ( tdTomato ) ; upon exposure to Cre recombinase , the tdTomato gene is deleted , and membrane-targeted EGFP is induced ( Figure 1D ) . We generated a recombinant rabies virus ( RABV-Cre ) expressing Cre-recombinase ( Figure 1A ) [28] , [29] , which , in combination with the Cre reporter mouse , provided a model to permanently change the color of an infected cell from red to green , even after the virus was cleared . Cre was modified by the addition of a 5′ nuclear localization signal to promote high efficiency , in vivo recombination between loxP sites , and cloned into the an empty recombinant RABV backbone previously published as “BNSP” [30] . This recombinant virus is based on the SAD B19 RABV vaccine strain , which is moderately pathogenic after intracranial inoculation [30] , but known to infect neurons efficiently [31] . Of note , to study long-term effects of RABV infection on neurons , it was imperative to choose a viral strain that efficiently infected the brain but that resulted in full recovery of the host: our previous work indicated that doses of this strain could be delivered that resulted in extensive neuronal infection , a robust antiviral response , and no mortality [30] . RABV-Cre was recovered by standard methods [29] and the fitness of the recombinant RABV , as well as the expression of the inserted Cre gene , was evaluated ex vivo . BSR fibroblasts were infected with RABV or RABV-Cre and analyzed by western blot for protein expression ( Figure 1B ) as well as one-step growth curves for viral replication ( Figure 1C ) . These assays demonstrated that insertion of Cre between the RABV nucleoprotein ( N ) and phosphoprotein ( P ) genes had no effect on the rate of viral replication , or relative viral gene expression ( Figure 1B and C ) . To analyze the functionality of the virus-encoded Cre , primary mouse fibroblasts harvested from Cre reporter mice were infected with RABV or RABV-Cre . Functional expression of Cre was indicated by EGFP expression in Cre reporter mouse fibroblasts after infection with RABV-Cre but not after infection with wildtype RABV ( Figure 1D and E ) . Of note , these cells are double-labeled with EGFP and tdTomato; this delayed loss of tdTomato is likely due to a long protein half-life . Published data shows loss of tdTomato upwards of 9 days post-excision [27] , though this may vary from tissue to tissue . Neuronal survival following viral infection next prompted us to investigate the functional integrity of these “cured cells” . We used fluorescence activated cell sorting ( FACS ) to isolate the EGFP+ population ( “infected” ) and the tdTomato+ population ( “uninfected” ) from individual mice 3 months post-infection ( Supplemental Figure S1 ) , a time when viral transcription was undetectable ( Figure 7 ) . Of note , the endogenous fluorescence of these populations precluded the need for intracellular staining that compromises RNA integrity . Affymetrix Gene Microarray was used to measure transcriptional profiles of the infected and uninfected cells from two independent cell sorting events ( n = 2 for each group ) . This analysis identified 1248 genes differentially expressed between the sorted groups ( infected vs . uninfected , ≥1 . 5 fold change , p<0 . 05 ) , 361 genes of which were down-regulated , 887 genes that were up-regulated . Gene expression levels ranged from 3 . 09-fold up-regulation to 2 . 72-fold down-regulation , and only 127 genes differed by more than 2-fold ( Supplemental Table S1 ) . Though the overall change in gene expression was modest , these transcriptional changes may be implicated in neuronal dysfunction . We saw no decrease in neuron-specific genes classically down-regulated in neurological disorders; in fact , there was up-regulation of neuronal receptors ( glutamate receptor , GABA receptor ) , ion channels ( potassium , sodium , and hydrogen ) , neurotransmitter transporter ( Cacna1b ) , and synapse-specific genes ( synaptotagmin I , synaptophysin ) ( Supplemental Table S2 ) , strongly supporting our contention that these analyzed cells are of neuronal origin . However , it is not a given that the change in expression of “neuron-specific” genes is responsible for the dysfunction of the neuron . More likely it is the virus' interaction with ubiquitous cellular genes that induces functional changes within the neuron . For example , genes encoding proteins that play critical roles in ion homeostasis , exocytosis , or mitochondrial function have been shown to have a great effect on neuronal function – however these genes are commonly expressed in a large variety of cells . Ingenuity Pathway Analysis ( IPA ) was used to identify biological functions most significantly impacted; cell-to-cell signaling was the most significant molecular and cellular function impacted , whereas behavior and nervous system development/function were the two most significant physiological systems impacted ( Figure 8 ) . Analyzing the regulation of groups of related genes , IPA predicted decreased function ( those with an absolute z-score value≥1 . 96 ) in the following areas: neurite growth/outgrowth , organization of cytoskeleton , organization of cytoplasm , and microtubule dynamics ( genes involved in these functions are listed in Table 1 ) .
Neurons are particularly vulnerable to the consequences of viral infection and/or the anti-viral immune response in the brain . Evidence suggests that neuropathology induced directly or indirectly by infection may lead to symptoms of disease . For example , caspase-dependent apoptosis has been implicated in pathogenesis of vesicular stomatitis virus ( VSV ) and WNV , and cytomegaloviruses induce lysis of the host cell [37]–[39] . For RABV , the basis of neuropathogenesis is unknown , and predicted to be due either to neuronal death or dysfunction . A challenge in studying the longevity of infected neurons is identifying and isolating them after resolution of the acute viral infection . To study rabies , approaches used in the past have included histopathology , measurements of cell death ( via TUNEL assay ) , and in vitro studies that may not represent the biology of an infected neuron in vivo . These assays have known limitations and may account for the inconsistent results in this research area , making it hard to draw a conclusion as to neuronal cell fate after viral infection . Here we examined the survival and functionality of neurons infected with RABV using a novel and innovative approach: the Cre reporter mouse model . Of note , our results are based on attenuated RABV which only models what may happen in a natural RABV infection . We appreciate the fact that pathogenic rabies virus is invariably lethal and few survive to the later timepoints evaluated in this study . However , by using a less virulent strain of virus , we were able to sufficiently infect brain neurons without killing the mice , and study the impact of the infection on these cells long after clearance of substantial inflammation and bystander effects , which may complicate data analysis . More pathogenic RABV variants could be used , though the window of time between infection and death is brief and catching these cells for analysis would be challenging . However , we speculate that in the rare event of an animal ( or person ) surviving a RABV infection , the changes within the previous infected neurons are similar to what have been observed for the attenuated RABV . Though we do not have direct evidence that infection with a pathogenic RABV will result in similar changes in gene expression as observed here for the attenuated RABV , there is indirect support for this hypothesis . This is based on the findings that the major difference between a pathogenic and attenuated RABV is the faster spread within the CNS of the pathogenic strain [40] which , in combination with the lack of immunogenicity of the pathogenic strain , normally results in death of the infected host . However , the lack of any pathology within the RABV-infected brain , which is especially true for an infection with pathogenic RABV strain , [15] , [41]–[45] strongly argues for functional changes within the infected neurons , including pathogenic strains . Our results show that experimental RABV does not induce cell loss as a result of direct or indirect cytopathic effects of the virus , nor a cytotoxic immune response . RABV G has been implicated in apoptotic signaling in vitro; for many years it was believed that the quantity of RABV G expressed correlated with both apoptosis and pathogenicity [15] , [46] . Recently this paradigm was challenged , and abundance alone was shown not to be the only determinant of apoptosis [47] . Préhaud et al . demonstrated that pathogenic and attenuated strains recruit different intracellular proteins that mediate either cell survival or cell death , respectively [18] . The in vivo relevance of their work remains to be determined; in vitro infection of pure cultures of neurons , at a high multiplicity of infection , in the absence of an intact immune response may not represent events in vivo . Our results indirectly support the claims that RABV immunity is rapid and mediated by non-cytolytic immune responses , likely antibody and cytokine secretion [4] . The role of type I interferons , specifically , has been of great research interest . RABV-infected neurons can produce type I interferons in vitro [48] , and interferon-α receptor knock out mice are unable to control the virus and ultimately succumb to infection [49] . What impact these cytokines have on neuronal integrity is unclear . Though CD8+ T cells have been found in brain sections of human rabies victims [50] , their deletion appears to have no significant effect on the survival of challenged mice [4] . Our data supports the hypothesis that RABV pathogenesis is not due to loss of neurons , but rather to neuronal dysfunction . Evolutionarily this makes sense , as preservation of the neuronal network by inhibition of apoptosis and limitation of inflammation is advantageous for RABV to complete its lifecycle . The Cre reporter model allowed us to isolate “cured” neurons from mice 3 months post-infection to study their transcriptional profiles compared to uninfected cells from the same mice . Comparing one cell population to another in the same mouse acted as an internal control , avoiding differences based on age , treatment , and normal biological variance . We show evidence that neurons exhibit permanent transcriptional differences from uninfected cells long after resolution of the viral infection . We used a bioinformatics tool , Ingenuity Pathway Analysis ( IPA ) , to identify biological functions most significantly impacted by this dysregulated gene set . Our results showed a modest change in gene transcription ( rarely over 3-fold difference between infected and uninfected ) with most affected genes involved in cell-to-cell signaling and nervous system development/function . The most interesting aspect of the transcriptional data was the prediction of functional consequences potentially induced by the pattern of gene changes: decreased neurite growth , decreased organization of cytoskeleton , decreased organization of cytoplasm , and decreased microtubule dynamics . For example , we observed a down-regulation of many key microtubule/microtubule-associated genes by microarray . Our results , which were acquired through an unbiased assay , are in agreement with others; Li et al . found similar down-regulation of microtubule-related genes in acute pathogenic RABV infection , and this directly correlated to degeneration of neuronal processes [51] . Furthermore , decreased neurite growth and dendritic/axonal swelling has been detected by microscopic examination of infected neurons in culture [11] , [12] , [18] , [51] , though it may be strain dependent . Together , these hypothesized gene expression changes may reflect an overall failure to reorganize the cytoskeleton for growth or repair of cellular processes . Future examination of live EGFP labeled cells purified from the brains of these Cre reporter mice may fully elucidate the functionality , or lack thereof , of once-infected neurons compared to never-infected neurons . Furthermore , alternative approaches , such as laser capture microdissection , may be useful in future efforts to monitor transcriptional changes in specific sub-populations of neurons . We anticipate that an understanding of the deficiencies of RABV infected cells will provide useful insights into the development of novel treatment options for the acute disease based on restoring lost cellular function . The incubation period , or period between infection and development of symptoms , is extremely variable and can range from 1 week to over 1 year in duration . Post-exposure prophylaxis is highly effective during this period; however , once symptoms develop , survival is extremely rare . Our data provides hope that living neurons may be a possible target for intervention . If research can elucidate the precise effect of viral infection of brain neurons , it may be possible to develop novel treatments effective during these late stages of disease . This data also provides reason to be concerned with the use of recombinant rabies vectors for treatment of neurological disorders or infections . This model may also be used as a tool to study the regenerative capacity of neurons . We looked at neuronal gene expression from one time point only , but analyzing the expression of select genes at several points over a period of time would provide information regarding the ability of these cells to regenerate . Furthermore , viral genome was still present at the 3 month timepoint; is functionality regained after complete genome clearance ? Latent herpesvirus induces neuronal deficits , though this is likely due to transcription of latency-associated genes [52] . We believe that the RABV genome is persisting due to the stabilizing effects of the viral nucleocapsid as described for other negative-stranded RNA viruses , including VSV [53] , and RABV does not express latency-associated genes . We do not anticipate that the persistent genome will play a role in neuronal dysfunction , though this would need to be confirmed experimentally . Lastly , this model may be used to study the link between viral infection and chronic neurodegenerative diseases . This link has been suggestive for other abortive or latent viral infections in which the host survives ( e . g . herpesviruses , measles , Polio , Epstein-Barr ) . Multiple sclerosis , for example , is suspected by some of having a viral etiology . A virus-induced disturbance may be , in a subset of patients , the starting point for autoimmune demyelination . There is compelling evidence for involvement of Human herpesvirus 6 ( HHV-6 ) , as viral antigen has been found in sclerotic lesions , and virus-specific IgM antibodies increase in relapsing-remitting MS patients [54] , though proving a causative link between viral infection and chronic dysfunction has been difficult . The Cre mouse model proposed here provides a powerful tool to study other neurotropic viruses and show causation between virus and chronic disease . Our data clearly demonstrates that neurotropic viral infection can cause persistent neurological deficits long after active infection has been cleared , and that once-infected neurons may not be restored to their pre-infection phenotype . The implications of these findings are substantial , as surviving neurons , which have a remarkable capacity to recover following damage [55] , [56] , may be targeted for novel treatment options in rabies and other neurological infections by restoring compromised cellular functions .
All animals were handled in strict accordance with good animal practice as defined by the relevant international ( Association for Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) ( Accreditation Status TJU: Full ) ) and national ( TJU Animal Welfare Assurance Number: A3085-01 ) , and all animal work was approved by the Institutional Animal Care and Use Committee ( IACUC ) at Thomas Jefferson University TJU . Animal use protocols are written and approved in accordance with Public Health Service Policy on Humane Care and Use of Laboratory Animals , The Guide for the Care and Use of Laboratory Animals . TJU IACUC protocol number 414 I ( Pathogenesis of rabies virus and vaccine vectors in mice ) was utilized in this study . Cre reporter mice , strain B6 . 129 ( Cg ) -Gt ( ROSA ) 26Sortm4 ( ACTB-tdTomato , -EGFP ) Luo/J , were purchased from the Jackson Laboratory , USA . Heterozygotes were bred and offspring were genotyped as recommended by Jackson Laboratories . Mice were anesthetized with 4% isoflurane by inhalation and inoculated intranasally with 1×105 ffu virus diluted to 20 µL with phosphate-buffered saline ( PBS ) ( 10 µL per nostril ) . Mice were sacrificed at the indicated time points post-infection and brain tissue harvested for RNA and immunohistochemistry . cBNSP is an infectious clone based on the RABV vaccine strain SAD B19 , which contains two single restriction sites ( BsiWI and NheI ) for inserting foreign genes [30] . Enterobacteria phage P1 Cre Recombinase ( Genbank accession number X03453 ) was engineered to include a 5′ nuclear localization signal ( ATG GCA CCC AAG AAG AAG AGG AAG ) to promote high efficiency in vivo recombination between loxP sites , and is hereafter referred to simply as Cre ( Cre cDNA was kindly provided by Dr . Jianke Zhang , Thomas Jefferson University , Philadelphia ) . The Cre gene was amplified using forward primer 5′-cac CGT ACG acc atg gca ccc aag aag aag-3′ ( BsiWI site in caps , ATG start codon for the nuclear localization signal is underlined ) and reverse primer 5′-cga GCT AGC cta atc gcc atc ttc cag cag g-3′ ( NheI site in caps ) , and cloned into cBNSP using the unique BsiWI and NheI sites between RABV nucleoprotein ( N ) and phosphoprotein ( P ) . The resulting cDNA was termed RABV-Cre . RABV-Cre was recovered as previously described [30] . Multi- and single-step growth curves were conducted on BSR cells as previously described [29] . Cre functionality was evaluated in vitro by infecting primary fibroblasts cultured from Cre reporter mice . To isolate primary fibroblasts , leg muscle was dissected from euthanized adult Cre reporter mice and dissociated using a protocol adapted from Blau et al . [57] . Briefly , muscle tissue was minced into 3- to 4-mm pieces in PBS , incubated in an enzyme mixture of collagenase D ( 0 . 75 units/mL , Roche Applied Science , 11088858001 ) , Dispase II ( 1 . 2 units/mL , Roche Applied Science , 10295825001 ) , and 2 . 5 mM CaCl2 for 45 min in a 37°C water bath , triturating with a 5 mL pipette 3 times throughout the incubation . Ten milliliters of PBS was added to the cell suspension and passed through a 70 µm filter , spun at 300×g for 5 min , resuspended in DMEM supplemented with 10% FBS and penicillin-streptomycin ( DMEM10 ) , and plated in a standard T75 tissue culture flask . To enrich for fibroblasts , cells were allowed to attach to the flask for 1 h at 37°C , at which time the unattached cells were aspirated and the adherent cells were passaged in DMEM10 for 5 to 7 d . Cre reporter primary fibroblasts plated at 80–90% confluency in 6-well plates were infected with 2 . 4×106 ffu recombinant virus at 37°C . After 96 h , the cells were assayed for color change from TdTomato to EGFP by fluorescent imaging and flow cytometry . For imaging , plated cells were fixed with 4% PFA ( pH 7 ) for 20 min at 4°C and viewed under a fluorescence microscope . For flow cytometry , trypsinized cells were fixed in suspension with 4% PFA ( pH7 ) for 20 min at 4°C , washed once in PBS supplemented with 2% BSA , and analyzed on a BD FACSCalibur ( 50 , 000 events collected ) . NA cells plated at 80–90% confluency in 12-well plates were infected with 3×106 ffu recombinant virus at 37°C in serum-free media . After 1 h , inoculum was replaced with RPMI supplemented with 5% FBS and penicillin-streptomycin and incubation was continued at 34°C . After 48 h , the cells were washed in PBS , lysed on ice in RIPA buffer ( 25 mM Tris pH 7 . 5 , 150 mM NaCl , 1% NP-40 , 0 . 4% sodium deoxycholate , 1 mM EDTA ) containing 1× protease inhibitor cocktail ( Sigma ) , and centrifuged at 12 , 000×g for 10 min . The protein concentration in the supernatant was determined using a BCA kit ( Pierce , 23227 ) and the supernatants were denatured with urea buffer ( 125 mM Tris–HCl pH 6 . 8 , 8 M urea , 4% sodium dodecyl sulfate , 5% beta-mercaptoethanol , 0 . 02% bromophenol blue ) at 56°C . Five micrograms of protein were resolved on an SDS-10% polyacrylamide gel and transferred to a nitrocellulose membrane in Towbin buffer ( 192 mM glycine , 25 mm Tris , 20% methanol ) . The membrane was then blocked in TBST ( 100 mM Tris-HCl pH7 . 9 , 150 mM NaCl , 0 . 01% Tween20 ) containing 5% dried milk at 4°C for several hours . After blocking , the membrane was incubated overnight with rabbit anti-Cre polyclonal antibody ( Novagen , 69050 ) at a dilution of 1∶10 , 000 , anti-RABV serum from reconvalescent RABV-infected mice diluted 1∶6 , 000 , and a mouse monoclonal antibody against Actin ( Sigma , A5441 ) , all diluted 1∶250 , 000 in TBST containing 5% BSA . After washing , the blot was incubated for 1 h in anti-rabbit-HRP conjugate and anti-mouse-HRP conjugate , both diluted 1∶50 , 000 in blocking buffer . Bands were developed with SuperSignal West Pico Chemiluminescent substrate ( Pierce , 34080 ) . Immediately after dissection , the mouse brains were bisected laterally using a sterile scalpel . One half was immediately immersed in RNAlater ( Qiagen , 1 mL/100 mg tissue ) for the purpose of RNA isolation . The second half was placed in 4% PFA ( pH 7 . 0 ) for immunohistochemical analysis . Immediately after dissection , brains were fixed 24 h in 4% paraformaldehyde ( pH 7 . 0 ) and cryoprotected by sequential saturations in 10% , 20% , and 30% sucrose/PBS ( each for 24 h ) . Samples were embedded , frozen , and cut by the Kimmel Cancer Center's Pathology Core Facility . Brains were embedded in Tissue-Tek O . C . T . compound ( Sakura ) , frozen and cut at −20°C on a Microm HM550 cryostat ( Thermo Scientific ) into 10 µm sections , and mounted onto charged slides ( Thermo Scientific Superfrost Plus ) . Slides were stored at −20°C and either directly imaged or stained for cell- and virus-specific antigens and then imaged . In preparation for staining , sections were permeabilized in 0 . 2% TritonX-100/PBS for 1 h at room temperature ( RT ) , washed in 0 . 05% TritonX-100/PBS ( wash buffer ) , and blocked in wash buffer supplemented with 5% BSA . For neuronal staining , slides were stained with a 1∶100 dilution of mouse anti-NeuN ( MAB377; Millipore ) , washed 3× , stained with 1∶300 dilution of Pacific Blue goat anti-mouse ( P-10993; Invitrogen ) . For astrocyte staining , slides were stained with a 1∶250 dilution of rabbit anti-GFAP ( NB300-141; Novus ) , washed 3× , stained with 1∶300 dilution of Pacific Blue goat anti-rabbit ( P-10994; Invitrogen ) . For RABV P staining , slides were directly stained with a 1∶300 dilution of AlexaFluor647 mouse anti-RABV P antibody . This antibody was generated by conjugating AF647 ( A20173; Invitrogen ) to purified RABV P-specific IgG produced from hybridoma cells kindly provided by Dr . Danielle Blondel , Gif sur Yvette , France [58] . Specifically , antibody-containing supernatant was purified using Nunc ProPur Midi G Kit , dialyzed in PBS , and conjugated . Images were acquired using a Leica DM5000B fluorescence microscope equipped with the DFC340FX camera or a Zeiss LSM 510 META Confocal microscope . Brain tissues immersed in RNAlater were transferred to RLT Buffer ( Qiagen ) supplemented with beta-mercaptoethanol ( 10 µL BME/1 mL RLT ) at a ratio of 100 µL RLT-BME per 10 mg tissue . Tissue was homogenized with Hard Tissue Omni Tip probes ( Omni International ) . RNA was isolated using the RNAeasy Mini Kit ( Qiagen ) according to the manufacturer's protocol . A 15 min on-column DNaseI digest ( Qiagen ) was included for all samples during the purification . RNA concentration and purity were determined using the NanoDrop 2000c ( Thermo Scientific ) . RNA was reverse transcribed into cDNA using the Omniscript Reverse Transcription Kit ( Qiagen ) according to the manufacturer's protocol . Each 20 µL reaction contained 2 µg of purified RNA , 10 units RNaseOut ribonuclease inhibitor ( Invitrogen ) , and 0 . 5 µM primer . Reaction mixtures were incubated at 37°C for 1 h , followed by 5 min at 95°C to inactivate the enzyme . All primers/probes used throughout were designed using Sigma-Aldrich's OligoArchitect and purchased from Sigma-Genosys . The following reverse transcription primers were used , 5′ to 3′: RV-N genome ( CAT GGA ACT GAC AAG AGA ) , messenger RNA ( for subsequent QPCR of RV-N sense message; TTT TTT TTT TTT TTT TTT TTV; V = G , C , or A ) , EGFP ( CGG ATC TTG AAG TTC ACC ) , RPL13A housekeeping gene ( CTT TTC TGC CTG TTT CCG TA; part of MHK-1 primer set from RealTimePrimers . com ) . Quantitative analysis of all genes was conducted on the MX3005P QPCR Machine ( Agilent Technologies ) . All genes ( except RPL13A , described below ) were quantified using the QuantiFast Probe PCR Kit containing ROX internal reference dye ( Qiagen , 204256 ) according to the manufacturer's protocol . Each of these 20 µL reactions contained 2 µL of the reverse transcription reaction , 0 . 4 µM each primer , and 0 . 2 µM TaqMan probe . QPCR cycling began with one hot start cycle of 95°C for 15 min , followed by 45 amplification cycles of 95°C for 15 sec , and 60°C for 1 min ( data acquired at end of each step ) . RPL13A was quantified using Brilliant II SYBR Green QPCR Master Mix ( Agilent; 600828 ) according to the manufacturer's protocol . Each of these 20 µL reactions contained 2 µL of the reverse transcription reaction and 0 . 1 µM each primer . QPCR cycling began with one cycle of 95°C for 15 min , followed by 45 cycles of 95°C for 30 sec , 58°C for 1 min ( data acquired at end of this step ) , and 72°C for 30 sec . Sequences of the QPCR forward primer , reverse primer , and TaqMan probe ( respectively ) are as follows , 5′ to 3′: RV-N anti-sense genome and RV-N sense message ( CAT GGA ACT GAC AAG AGA , TGC TCA ACC TAT ACA GAC , [6-FAM]ATG CGT CCT TAG TCG GTC TTC TC[TAMRA] ) , EGFP ( AGC TGG AGT ACA ACT ACA , CGG ATC TTG AAG TTC ACC , [6-FAM]ATG CCG TTC TTC TGC TTG TCG[TAMRA] ) , RPL13A housekeeping gene ( ATG ACA AGA AAA AGC GGA TG , CTT TTC TGC CTG TTT CCG TA , [no probe]; primers from MHK-1 primer set from RealTimePrimers . com ) . Primer pairs for all genes were validated by measuring product linearity ( R2>0 . 99 ) and amplification efficiency ( E = 98–102% ) . SYBR Green reactions were further validated for specificity by running dissociation curves after quantification ( using default program on MX3005P ) and by running PCR products on 2 . 5% agarose gel . All samples were run in triplicate alongside negative controls ( water and No RT ) . For absolute quantification of RV-N anti-sense genome and RV-N sense message , an eight point standard curve was generated from 10-fold serial dilutions of cDNA of known copy number ( ranging 108 to 101 transcripts ) . Copy numbers were normalized to RPL13A housekeeping gene . For relative quantification of EGFP , the DDCt method [59] was used to measure fold-change of EGFP relative to RPL13A housekeeping gene . Whole brains with olfactory bulbs attached were dissected from adult ( >8 wks ) Cre reporter mice and dissociated using a well-described protocol adapted from Huettner and Baughman [60] , [61] . Brain tissue was minced into 3- to 4-mm pieces in Earle's Balanced Salt Solution ( EBSS , HyClone , SH3002902 ) . Papain enzyme ( Worthington Biochemical , LS003126 ) was preactivated in EBSS/0 . 5 mM EDTA/1 mM L-cysteine by incubating 10 min at 37°C . Minced tissue was incubated in 20 units/mL preactivated papain and 125 units/mL DNase I ( Worthington Biochemical , LS002006 ) for 85 min in a 37°C water bath , triturating gently with a 10 mL pipette twice throughout the incubation . The cell suspension was passed through a 70 µm filter , spun at 300×g for 5 min , and resuspended in 3 mL of dilute protease inhibitor solution ( EBSS , 1 mg/mL ovomucoid inhibitor [Worthington Biochemical , LS003085] , 1 mg/mL BSA , 125 units/mL DNase I ) . A discontinuous density gradient was made by layering the cell suspension on top of 5 mL concentrated protease inhibitor ( EBSS , 10 mg/mL ovomucoid inhibitor , 10 mg/mL BSA ) and centrifuging at 70×g for 8 min to separate cells ( pellet ) from debris ( supernatant ) . Cells were washed once in PBS supplemented with 2% BSA , and FACS-purified based on EGFP and TdTomato fluorescence utilizing a Coulter MoFlo sorter . EGFP-positive and EGFP-negative populations were saved for RNA extraction . RNA was isolated using the RNAeasy Mini Kit ( Qiagen ) according to the manufacturer's protocol . RNA was quantified on a NanoDrop 2000c ( Thermo Scientific ) , followed by RNA quality assessment on an Agilent 2100 Bioanalyzer ( Agilent , Palo Alto , CA , USA ) . Amplification and labeling was performed using the Ovation Pico WTA-system V2 RNA amplification system ( NuGen Technologies , Inc . ) . Briefly , 50 ng of total RNA was reverse transcribed using a chimeric cDNA/mRNA primer , and a second complementary cDNA strand was synthesized . Purified cDNA was then amplified with ribo-SPIA enzyme and SPIA DNA/RNA primers ( NuGEN Technologies , Inc . ) . Amplified DNA was purified with Qiagen MinElute reaction cleanup kit . The concentration of Purified ST-cDNA was measured using the Nanodrop . 2 . 5 µg ST-cDNAs were fragmented and chemically labeled with biotin to generate biotinylated ST-cDNA using FL-Ovation cDNA biotin module V2 ( NuGen Technologies , Inc . ) . Affymetrix gene chips , mouse gene 1 . 0 ST array ( Affymetrix , Santa Clara , CA ) , were hybridized with fragmented and biotin-labeled target ( 2 . 5 µg ) in 110 µl of hybridization cocktail . Target denaturation was performed at 99°C for 2 min and then 45°C for 5 min , followed by hybridization for 18 h . Arrays were then washed and stained using Genechip Fluidic Station 450 , and hybridization signals were amplified using antibody amplification with goat IgG ( Sigma-Aldrich ) and anti-streptavidin biotinylated antibody ( Vector Laboratories , Burlingame , CA , USA ) . Chips were scanned on an Affymetrix Gene Chip Scanner 3000 , using Command Console Software . Background correction and normalization were done using Iterative Plier 16 with GeneSpring V11 . 5 software ( Agilent , Palo Alto , CA , USA ) . 1 . 5-fold differentially expressed gene list was generated . The differentially expressed gene list was loaded into Ingenuity Pathway Analysis ( IPA ) 5 . 0 software ( http://www . ingenuity . com ) to perform biological network and functional analyses . The microarray data can be accessed at the GEO - repository access number GSE38975 ( http://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? acc=GSE38975 ) | Rabies is an ancient and fatal neurological disease of animals and humans , caused by infection of the central nervous system ( CNS ) with Rabies virus ( RABV ) . It is estimated that nearly 55 , 000 human RABV fatalities occur each year , though this number is likely much higher due to unreported exposures or failure of diagnosis . No treatment has been identified to cure disease after onset of symptoms . Neurovirologists still do not know the cause of rabies' dramatic symptoms and fatality , though it is thought to be due to neuronal loss or dysfunction . Here , we use a novel approach to permanently and genetically tag infected cells so that they can be identified after the infection has been cleared . This allowed us to define neuronal survival time following infection , and to assess neuronal function through gene expression analysis . We found that RABV infection does not lead to loss of neurons , but rather induces a permanent change in gene expression that may be related to the ability of RABV to cause permanent CNS disease . Our study provides evidence that viral infection of the brain can initiate long-term changes that may have consequences for nervous system health , even after the virus has been cleared from the CNS . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"viral",
"transmission",
"and",
"infection",
"microbiology",
"rabies",
"neglected",
"tropical",
"diseases",
"infectious",
"diseases",
"viral",
"clearance",
"biology",
"cell",
"biology",
"virology",
"neurons",
"cellular",
"types",
"molecular",
"cell",
"biolog... | 2012 | Immune Clearance of Attenuated Rabies Virus Results in Neuronal Survival with Altered Gene Expression |
A major goal in post-genome biology is the complete mapping of the gene regulatory networks for every organism . Identification of regulatory elements is a prerequisite for realizing this ambitious goal . A common problem is finding regulatory patterns in promoters of a group of co-expressed genes , but contemporary methods are challenged by the size and diversity of regulatory regions in higher metazoans . Two key issues are the small amount of information contained in a pattern compared to the large promoter regions and the repetitive characteristics of genomic DNA , which both lead to “pattern drowning” . We present a new computational method for identifying transcription factor binding sites in promoters using a discriminatory approach with a large negative set encompassing a significant sample of the promoters from the relevant genome . The sequences are described by a probabilistic model and the most discriminatory motifs are identified by maximizing the probability of the sets given the motif model and prior probabilities of motif occurrences in both sets . Due to the large number of promoters in the negative set , an enhanced suffix array is used to improve speed and performance . Using our method , we demonstrate higher accuracy than the best of contemporary methods , high robustness when extending the length of the input sequences and a strong correlation between our objective function and the correct solution . Using a large background set of real promoters instead of a simplified model leads to higher discriminatory power and markedly reduces the need for repeat masking; a common pre-processing step for other pattern finders .
The rapid emergence of experimental techniques that can probe for functional elements at whole-genome scales[1] necessitates computational methods to analyze data in these settings . In particular , methods that locate promoters or measure gene expression on genome-wide scales ( e . g . [2] , [3] ) must be complemented by algorithms that can find the active regulatory elements within the larger promoters . Ab initio computational search for transcription factor binding sites ( TFBS ) in DNA sequences is often termed “motif discovery” . “Motif” here refers to a general pattern describing what DNA sequences the transcription factor binds[4] . Motif discovery is one of the classical problems in computational sequence analysis and can be briefly stated as: Given a set of sequences containing one or several short overrepresented sites , locate these and produce a model describing them . There are two main avenues used to attack this problem: i ) enumerative algorithms based on word counting , such as [5] , [6] , and ii ) pattern-based approaches often using position specific weight matrices ( WMs ) , which scores sites based on position specific weights [4] . Since the binding preferences of transcription factors ( TFs ) are not easily captured by a single word or consensus string , pattern-based approaches can give solutions closer to the biological reality and it has been argued that the matrix score is related to the binding energy [7] , [8] . However , such approaches correspond to the problem of finding local , optimal multiple alignments , which is NP-complete [9] . Therefore , almost all pattern-based motif finders use statistical optimization methods such as Gibbs sampling or expectation maximization [10] , [11] . A typical instance of motif discovery starts with a set of upstream promoter regions of co-expressed genes suspected to be co-regulated and by extension more likely to be under control by the same regulatory machinery . This set is called the “positive set” and most methods proceed from here by locating motifs that are in some way statistically overrepresented in this set . The most successful applications of motif discovery have been in organisms whose regulatory information is densely aggregated around transcription start sites , such as Saccharomyces cerevisiae ( baker's yeast ) . In mammalian genomes , regulatory information is spread out over wider regions , which makes “pattern drowning” a significant issue; in other words , the information in the regulatory sites is too small to stand out in the large genomic region of interest . In this context , the accuracy of contemporary pattern finders is not sufficient for many biologically important problems [12] . Most methods operate with some notion of a background model describing “generic DNA” against which the over-representation is measured . The model is often a multinomial or a Markov model . The choice of model is important for obtaining good results [13] , [14] . However , most such models have difficulty in capturing the complexity of the highly heterogeneous mammalian genome sequence , which has a multitude of different promoter architectures[15] , numerous interspersed repeats , low complexity sequences , CpG islands , etc . [16] . Instead of simplifying the underlying DNA sequence by a general model , we take this to its extreme conclusion and use a very large set of promoters as the actual background instead of building a model describing the sequences in the promoters . For simplicity , we use the term “negative set” to describe the background set; this is strictly speaking not true as sites could occur in this set at a much lower frequency , since real promoters are sampled randomly . By contrasting the sets , it is possible to see what common features make the sequences in the positive set unique . Discriminatory motif searching is not a new idea; several methods have been developed that take advantage of a negative set [17]–[24] . However , many of these use word-based models [19]–[21] , which might not capture the diversity of binding sites . Others again use PWMs , but have binary hit models that do not distinguish between hits as long as they are over a threshold [22] . A discriminatory approach similar to ours has been combined with the use of expression data [18] , but depending on the regions that are being investigated this might often not be available or even possible . We adopt an approach similar to DEME [23] to identify the most discriminative set of motifs by modeling the sequence labels ( positive or negative ) rather than using the conventional generative approach[10] , [11] . However , there are some important differences to DEME . Firstly , DEME uses a global string-based search followed by a local gradient refinement , which may miss patterns that are not well-represented by a consensus string , whereas we use a global optimization technique ( simulated annealing ) for optimizing the model , which does not have this limitation , although it may have others ( see below ) . Secondly , our method ( Motif Annealer - MoAn ) uses and optimizes a threshold , and uses an enhanced suffix array ( ESA ) to speed up pattern searches . Thirdly , in MoAn the length of the motif is also optimized . DEME is also particularly targeted towards proteins while our approach is intended for use with DNA . Specifically , we use conditional maximum likelihood to estimate the WMs and their thresholds such that the probability of the positive and negative sets is maximized ( see Methods ) . Thus , the resulting matrices cannot be derived from the frequency matrix for the sites found – it is rather the matrices that lead to the best discrimination . The probability of a sequence is calculated as a product of the probabilities given by the matrices matching above a threshold and a simple null model for non-matching regions . From this and prior probabilities for matches in the positive and negative sets , the probability of the set label ( positive or negative ) is calculated . In this probability the background model cancels . The total likelihood is a product of the class probabilities for all sequences ( positive and negative ) . This conditional likelihood leads to a non-trivial optimization problem which is handled using simulated annealing ( see Methods ) , where we iteratively change the WMs and their thresholds , retaining changes that lead to higher discriminatory power using the Metropolis-Hastings algorithm [25] , [26] . Given sufficient iterations , the method guarantees convergence on the optimally discriminatory motifs . To cope with the vast size of the sets we utilize a highly efficient data structure , the ESA , for searching DNA for pattern instances[27] . With reasonable cutoffs , this reduces the computation by an order of magnitude[28] .
The relationship between our objective function and the correct solution was assessed by plotting the MoAn scores against the sensitivity obtained in all five runs on each of the 84 sets ( not just the best from each run ) ( Fig . 2 ) . There is a clear correlation ( Pearson CC: 0 . 90 ) between these two measures . There is a similar correlation with other measures , such as the nCC ( Fig . S1 ) . This finding is important , because it indicates that the raw score is an indication of quality independent of the motif analyzed . It also shows that choosing the best scoring run of several will often give the best result . Aside from the problem with decreasing sensitivity as the length of the input sequences increase , repetitive sequences represent a severe problem for motif discovery , as these will often seem to be over-represented , and therefore it is common to mask these repeats . However , masking is always arbitrary , and some repeats are functional [31] , [32] , so indiscriminate repeat masking is not optimal . When using a large negative set , repeat masking is unnecessary since repeats , if commonly occurring , will feature in the negative set and therefore be avoided as potential hits in the positive . At the same time , we can avoid the reverse problem – if a type of repeat actually is over-represented in the positive set , it can still be found . To demonstrate the insensitivity to repeats on a practical level , we planted repetitive sequences in each of the positive sets with a slightly higher frequency than the real motifs and ran our predictor on these sets both with the normal background and with a background similarly spiked with repeats . Specifically , we planted 1 to 10 consecutive instances of CACTA with a probability of 60% in each sequence . Fig . 3 shows , as expected , that the results do not deviate much from the repeat-less run when repeats are planted in both the positive and negative sequences , while the method picks up the repeats instead when there are no repeats in the negative set . We also performed this test using decoy motifs instead of repeats with similar results ( Text S4 , Fig . S2 ) . Evaluation of methods on real data is difficult and often a poor indication of general performance due to lack of insight into the correct solution [12]; on the other hand , it is necessary to show that the method can be applied to real problems . MoAn and four other methods were run on a collection of real data sets consisting of the binding sites of four human and mouse factors from the PAZAR database[33] and their associated genomic sequence . The sets were split by organism into 7 sets and the regions adjacent on the genome were merged resulting in sets ranging in size from 14 to 118 . The merging means that the base sequences can have a varying number of sites and may be of different lengths . The sets were then subsequently enlarged by adding an equal number of randomly selected promoters to increase the difficulty ( Text S6 and Table S5 ) and also padded with their cognate upstream and downstream regions of varying lengths ( 200–1200 , as in the synthetic evaluation ) to estimate the impact of noise . Fig . 4 shows the performance over the real sets . MoAn's performance is clearly superior , but not as spectacular as in the more controlled environment with synthetic sequences . We speculate that the reason for this is that the background and foreground of the synthetic sets are essentially sampled from the same pool ( RefSeq promoters ) , while we have made no effort to customize the background for the PAZAR sets . If the genomic environment of the factors differ from normal promoter sequences this could lead to a reduced performance . There are also fewer sets ( 7 versus 12 ) in this evaluation leading to a higher variability . We report additional trials using ChIP-chip data in supplementary material ( Text S7 , Fig . S3 and Tables S6 , S7 ) . MoAn has also been used successfully to discriminate between binding regions of human ESR1 and its paralog ESR2; the results were comparable with matrix-scanning approaches with pre-defined motifs[34] . An additional aspect of the motif finding problem is that TFs often work by forming complex interactions [35] . Examples include mutually exclusive and cooperative binding . Clusters of TFBSs are commonly termed cis-regulatory modules , and are often responsible for tissue-specific expression . We try to capture these interactions by incorporating co-occurrence of sites from different motifs into our model , with the goal of further increasing predictive power . To test whether our objective function is capable of capturing interactions between factors we constructed a set where co-occurrence of sites from different motifs occurs . We randomly chose 5 pairs of new motifs ( Table S2 ) and planted their corresponding sites in a positive set of 100 promoters with a 40% chance of co-occurrence and 10% of single occurrence . We then spiked the background set with sites from each of the motifs ( 10% chance each for all sequences ) to mimic a situation where it is the interactions of the two sites rather than single sites that are responsible for the regulation . MoAn was then run in co-occurrence mode and compared to two single-occurrence runs in a series . In the serial runs we masked out the predictions from the first iteration before running the second iteration . In Fig . 5 the ASP and nCC is plotted . In our experiment three of the pairs turned out to be composed of motifs with relatively low information , leading to poor performance . However , the two remaining ones show that modeling of co-occurrence can significantly improve performance . This extended model is unfortunately computationally taxing and requires more than twice the number of iterations compared to the single prediction .
In this work we have shown the value of using a large negative set instead of a pre-defined background model in motif discovery . Using raw sequences more accurately portrays the background than any general model and therefore higher discriminatory power is achieved . This method is also much less sensitive to “pattern drowning” in larger sequences , which is a bottleneck in computational analysis of mammalian regulatory regions . However , while our method takes a significant step towards routine motif discovery on large sequences , the problem cannot be considered fully solved . In particular , MoAn accuracy may be further improved by incorporating information on evolutionary constraints ( phylogenetic footprinting ) [36] or DNA accessibility[24] , [37] . In our opinion DEME is the best runner up of the methods . It often predicts the correct motif and has a high sensitivity , but often at the cost of a large number of false positives as it predicts also in those sequences not containing a site . MoAn seems to be better at balancing the sensitivity and specificity . On the other hand DEME is also given an artificial advantage by having the correct motif length as input and it is uncertain how advantageous this is . Weeder performed surprisingly poorly given its stellar performance in a recent evaluation[12] . This might be due to motif selection which we did according to the most redundant motif , but was in [12] done in a more complicated manner not part of the current Weeder package . This procedure led to no predictions on several of the harder sets which might give Weeder a statistical advantage ( as discussed in [12] ) . A concern that might be raised is that optimizing a cutoff might lead to a conservative estimate of binding sites at the expense of weaker sites . However , assessing this is hard since experiments have their own thresholds in the post-analysis and any evaluation of MoAn's threshold will be dependant upon those . Investigations where we artificially forced the cutoff to remain low , lead to a reduction in performance ( data not shown ) . We address this potential problem indirectly by providing a matrix that can be used to search sequences at a lower threshold . Future improvements of MoAn will focus on the optimization algorithm , which currently is not robust enough to always produce reliable results . In our current implementation we avoid this problem by running the algorithm many times to see that the solution is stable .
nTP Number of nts part of a site correctly predicted . nTN Number of background nts correctly predicted . nFP Number of background nts predicted to be part of a motif . nFN Number of nts part of a site predicted as background . sTP Number of real sites that share over 50% of its nts with a predicted site . sFP Number of predicted sites that share less than 50% of its nts with a real site . sFN Number of real sites that share less than 50% of its nts with a predicted site . Note that we are more conservative with respect to the site prediction than [12] in that we demand at least half of the nucleotides overlapped to get a single sTP . Derived from the basic statistics: A sequence is assumed to be described by a mixture model consisting of a background distribution and a set of WMs describing the binding affinities of the TFs . The WMs contain log-odds scores of the type: ( 1 ) where is the position in the WM , is a letter in the DNA alphabet and is the probability of having letter at position in the motif described by . The score of a matrix aligned at a position in a sequence is therefore: ( 2 ) where is the DNA letter at position in sequence . The aim is to discriminate between two sets of sequences , where label denotes the positive set and the negative . The prior probability of binding site occurrence in a sequence contained in set is called . We assume that there is a marked difference in the site occurrence between the two sets and want to construct a score that captures how well a set of WMs describe this difference . Using two WMs as an example , and , there are four possible ways for a sequence to be generated . With prior probability it contains no sites and is only generated by the background model . Or , with prior probability , it contains a single site ( one of the two ) corresponding to one WM positioned at nucleotide number ( is equal to 1 or 2 corresponding to the two different matrices ) . This is written , where is the score of the matrix aligned to the nucleotides at position ( eq . 2 ) and 2 is the base of the log scores contained in the WM . Note that the log scores in a WM are divided by the background model , so the background ( ) cancels out in sites where the motif occurs . The final case , with prior probability , is the co-occurrence of two sites in a sequence , which is . However , this is only correct when the sites are not overlapping since otherwise the overlapping nucleotides would be included in the product twice . Therefore we disallow overlaps . For efficiency reasons , we do not calculate the score in its entirety . We assume that it is the strong sites that contribute the most to the equation and introduce a cutoff for each WM on the minimum score of a site . This enables an efficient search in the ESA . This is not without biological merit since WM scores and binding energies for known TFs are correlated , and at some point the binding energies of a TF and a poor binding sequence must be too small to matter [4] . It is also a standard method to use when scanning with known matrices [38] . So we only consider sites that score above a threshold , which is called for matrix . Then the probability of a sequence from the set being generated by the WMs is ( 3 ) where is the expectation over of over all predicted sites: ( 4 ) with being the step function ( 1 above 0 and zero otherwise ) . The co-occurrence expectation is defined in a similar way with overlaps disallowed . The effective weight of no sites ( 5 ) accounts for extra weight given to no sites due to alignments not meeting the threshold . With this definition , is the probability or generative model of the sequence conditioned on the WM and threshold , . To find the WMs that best explain the difference in occurrence between the sets we use a discriminative objective function based on the probability of the labels given the sequences and WMs , formally: ( 6 ) This is the logistic likelihood function for binary classification , see e . g . [39] . The discriminative model can thus be viewed as logistic regression with an adaptive set of basis functions . For multiple sequences assumed to be independent , the joint probability is the product of the single sequence probabilities over all sequences in both the positive and negative set: ( 7 ) We refer to this function as the ( log likelihood ) score , . Based on the sequence density we can use Bayes theorem to calculate the probability of the label given the WMs , the thresholds , and the sequence : ( 8 ) We observe that the prior probability of is proportional to the number of sequences in the set divided by the total number of sequences . A very high threshold will give no matches , and the probability will then be a constant given by the priors and the size of the two sets . Matches that score above the threshold in the negative set will lower the score and matches above the threshold in the positive set will increase the score , so the game is to obtain as many high-scoring matches in the positive set as possible without introducing too many matches in the negative set . The prior is conservative in our runs in that we are strict about promoting hits in the positive set , but only moderately strict about disallowing negative hits . For a single matrix the prior on is 0 . 01; : 0 . 99; : 0 . 80; : 0 . 20 . For two matrices: : 0; : 0 . 1; : 0 . 9; : 0 . 80; : 0 . 15; and : 0 . 05 . These priors can be set by the user if prior knowledge is available about the set ( i . e . a high confidence negative set or an uncertain positive set ) . In the evaluation we deliberately chose a probability of having a site ( 0 . 5 ) in a sequence very different from the model prior ( 0 . 99 ) to avoid giving our own method a big advantage . It shows that the method is not very sensitive to the choice of prior . The objective function outlined above is optimized using simulated annealing [40] . Informally , it proceeds by iteratively proposing a candidate solution and then accepting or rejecting it depending on how good it is compared to the current solution . It sometimes accepts changes for the worse and therefore possesses the power to escape local maxima . The hope is that it will converge on a solution that is close to optimal . Formally , this translates to a walk over the search space where in the current state , the next state is either the same or the candidate solution depending on their relative scores and a temperature parameter . The temperature parameter is lowered for each iteration using as default an exponential cooling scheme ( for details see Text S5 ) , thus incrementally constraining the neighborhood of accepted changes . Candidate solutions are proposed by applying one of several steps outlined in the list below . In the case of multiple matrices , only one is changed at a time . We perform all steps on a integer “count” matrix which is then translated into a log-odds WM prior to searching the ESA , but notice that the “count” matrix does not represent actual letter frequencies in the selected sites . The steps are: Note that for the extend and decrease step there is a minimum and maximum number of columns for a motif . The default for these are 5 and 15 respectively . The matrix is initialized with random counts and the cutoff is also selected uniformly according to the last step in the list above . Termination of the optimization is only based on the number of iterations which is by default set to a rather conservative value of 30 million iterations . Time requirements for a single run is variable depending on the set size , but was for our runs comparable to NestedMICA ( single threaded ) and considerably faster than Weeder's “large” run and DEME . Source code as well as data sets is freely available at the author's web site: http://moan . binf . ku . dk | In the years following the sequencing of the human genome focus have shifted towards trying to understand how this blueprint results in the diversity of cells that we observe . Part of the answer lies in the regulation of transcription and how the proteins responsible for this recognize where they should attach to the DNA . This is a well studied problem , but most methods developed for this have a hard time dealing with the heterogeneity of the mammalian genomes . Here we present a method that greatly improves the efficiency of this search by contrasting the DNA with a large number of background DNA sequences . This enables us to handle repetitive segments of the genome that may be functional , but are usually considered intractable by most methods . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"computational",
"biology/sequence",
"motif",
"analysis",
"computational",
"biology/transcriptional",
"regulation"
] | 2009 | Discovery of Regulatory Elements is Improved by a Discriminatory Approach |
Catenation links between sister chromatids are formed progressively during DNA replication and are involved in the establishment of sister chromatid cohesion . Topo IV is a bacterial type II topoisomerase involved in the removal of catenation links both behind replication forks and after replication during the final separation of sister chromosomes . We have investigated the global DNA-binding and catalytic activity of Topo IV in E . coli using genomic and molecular biology approaches . ChIP-seq revealed that Topo IV interaction with the E . coli chromosome is controlled by DNA replication . During replication , Topo IV has access to most of the genome but only selects a few hundred specific sites for its activity . Local chromatin and gene expression context influence site selection . Moreover strong DNA-binding and catalytic activities are found at the chromosome dimer resolution site , dif , located opposite the origin of replication . We reveal a physical and functional interaction between Topo IV and the XerCD recombinases acting at the dif site . This interaction is modulated by MatP , a protein involved in the organization of the Ter macrodomain . These results show that Topo IV , XerCD/dif and MatP are part of a network dedicated to the final step of chromosome management during the cell cycle .
DNA replication of a circular bacterial chromosome involves strong DNA topology constraints that are modulated by the activity of DNA topoisomerases [1] . Our current understanding of these topological modifications comes from extensive studies on replicating plasmids [2 , 3] These studies suggest that positive supercoils are formed ahead of the replication fork , while precatenanes are formed on newly replicated sister strands . At the end of a replication round , unresolved precatenanes accumulate in the region of replication termination and are converted to catenanes between the replicated sister chromosomes . Neither precatenanes or catenanes have been directly observed on chromosomes but their presence is generally accepted and failure to resolve them leads to chromosome segregation defects and cell death [4] . Topo IV is a type II topoisomerase formed by two dimers of the ParC and ParE subunits and is the main decatenase in Esherichia . coli [5] . in vitro , its activity is 100 fold stronger on catenated circles than that of DNA gyrase [6] . Topo IV activity is dependent on the topology of the DNA substrate; Topo IV activity is strongest on positively supercoiled DNA and has a marked preference for L-braids , which it relaxes completely and processively . Topo IV can also unlink R-braids but only when they supercoil to form L-plectonemes [7–9] . In vivo , DNA gyrase appears to have multiple targets on the E . coli chromosome [10–12] , whereas Topo IV cleavage sites seem to occur less frequently [11] . Interestingly , Topoisomerase IV activity is not essential for replication itself [13] but is critical for chromosome segregation [14] . The pattern of sister chromatid separation has been shown to vary upon Topo IV alteration , leading to the view that precatenanes mediate sister chromatid cohesion by accumulating for several hundred kilobases behind the replication forks keeping the newly replicated DNA together [13 , 15] . The regulation of Topo IV and perhaps the accessibility of the protein to chromosome dimers was proposed to be an important factor controlling chromosome segregation [15 , 16] . Topo IV activity can be modulated by a number of proteins including MukB and SeqA . MukB , is an SMC-related protein in E . coli and is reported to bind to the C-terminus of Topo IV [17] to enhance Topo IV unlinking activities [18 , 19] . MukB also appears to be important in favoring the formation of Topo IV foci ( clusters ) near the origin of replication [20] . SeqA , a protein involved in the control of replication initiation , and Topo IV also interact [21] . These interactions may play a role in sister chromatid segregation at the late segregating SNAP regions near the origin of replication of the chromosome [16] . Beside its role in the resolution of precatenanes , Topo IV is mostly required in the post-replicative ( G2 ) phase of the cell cycle for the resolution of catenation links . Indeed , Espeli et al . showed that Topo IV activity is mostly observed during the G2 phase , suggesting that a number of catenation links persist after replication [22] . Recent cell biology experiments revealed that in G2 , the terminal region ( ter ) opposite oriC segregates following a specific pattern [23–25] . Sister ter regions remain associated from the moment of their replication to the onset of cell division . This sister-chromosome association is mediated by the Ter macrodomain organizing protein , MatP [26] . At the onset of cell division , the FtsK DNA-translocase processes this region , releasing the MatP-mediated association . This process ends at the dif site , when the dimeric forms of the sister chromosomes are resolved by the XerC and XerD recombinases . A functional interaction between the MatP/FtsK/XerCD-dif system and Topo IV has long been suspected . FtsK interacts with Topo IV , enhancing its decatenation activity in vitro [27 , 28] and the dif region has been reported as a preferential site of Topo IV cleavage [29] . This functional interaction has been poorly documented to date and is therefore remains elusive . In this study we have used genomic and molecular biology methods to characterize Topo IV regulation during the Escherichia coli cell cycle on a genome-wide scale . The present work revealed that Topo IV requires DNA replication to load on the chromosome . In addition , we have identified two binding patterns: i ) regions where Topo IV binds DNA but is not engaged in a cleavage reaction; ii ) numerous sites where Topo IV cleavage is frequent . We show that Topo IV-mediated removal of precatenanes is influenced by both local chromatin structure and gene expression . We also demonstrate that at the dif site , Topo IV cleavage and binding are enhanced by the presence of the XerCD recombinase and the MatP chromosome-structuring factor . The enhancement of Topo IV activity at dif promotes decatenation of fully replicated chromosomes and through interaction with other DNA management processes , this decatenation ensures accurate separation of the sister chromosomes .
To identify Topo IV binding , we performed ChIP-seq experiments in ParE and ParC Flag tagged strains . The C-terminus fusions of ParE and ParC replaced the wild-type ( WT ) alleles without any observable phenotypes ( S1 Fig ) . We performed three independent experiments , two ParE-flag IPs and one ParC-flag IP , with reproducible patterns identified in all three experiments . A Pearson correlation of 0 . 8 , 0 . 9 and 0 . 7 was observed for ParC-ParE1 , ParE1-ParE2 and ParC-ParE2 respectively . A map of enriched regions observed in each experiment is represented on Fig 1A ( red circles ) . Four of the highly-enriched sites are illustrated at a higher magnification in Fig 1A—right panels . Interestingly one of these sites corresponds to the dif site ( position 1 . 58Mb ) , which has previously been identified as a strong Topoisomerase IV cleavage site in the presence of norfloxacin [29] . We also observed strong enrichment over rRNA operons , tRNA and IS sequences . To address the significance of the enrichment at rRNA , tRNA and IS , we monitored these sites in ChIP-seq experiments performed in the same conditions with a MatP-flag strain and mock IP performed with strain that did not contain any flag tagged protein . Both MatP and Mock IP presented significant signals on rRNA , tRNA and IS loci ( S2 Fig ) . This observation suggested that Topo IV enrichment at rRNA , tRNAs and IS was an artifact of the ChIP-Seq technique . By contrast no enrichment was observed at the dif site in the MatP and mock-IP experiments ( S2 Fig ) , we therefore considered dif to be a genuine Topo IV binding site and compared every enriched region ( >2 fold ) with the dif IP . We filtered the raw data for regions presenting the highest Pearson correlation with the dif signal ( >0 . 7 ) . This procedure discarded many highly enriched regions ( Fig 1A orange circles ) . We identified 19 sites throughout the chromosome where Topo IV IP/input signal suggested a specific binding for at least two of the experiments ( Fig 1A , outer circle histogram , S1 Table ) . Most Topo IV binding sites span a 200 bp region . These sites frequently overlapped intergenic regions , with their mid-points located inside the intergenic region , and did not correlate with any identifiable consensus sequence . In addition to dif , which exhibited a 10-fold enrichment , three other sites were strongly enriched . These sites corresponded to positions 1 . 25Mb ( 9 . 4x ) , 1 . 85Mb ( 31x ) and 2 . 56Mb ( 19x ) on the chromosome ( Fig 1A , right panels ) . Beside these specific sites , Topo IV IP showed non-specific enrichment in the oriC proximal half of the chromosome . This bias was not a consequence of locus copy number , as the enrichment remained after copy number normalization ( Fig 1B ) . We used MatP-Flag IP [30] and a control IP in a strain that does not contain a Flag tagged gene to differentiate non-specific Topo IV binding from experimental noise ( S3A Fig ) . In addition , Topo IV enrichment was also observed in GC rich regions of the chromosomes ( S3B Fig ) . Importantly , the ori/ter bias was not a result of the GC% bias along the chromosome since it was still explicit after GC% normalization ( S3C Fig ) . More precisely , the Topo IV binding pattern closely followed gene dosage for a ~3Mb region centered on oriC ( S3D and S3E Fig and S1 Text ) . In the complementary ter-proximal region , gene dosage ( input reads ) was higher than the ChIP-seq profile , suggesting that the nonspecific Topo IV binding was lower or lasts for a shorter time in the cell cycle ( since these data are population-averaged ) . The Terminus region that is depleted in Topo IV binding ( 1 . 6Mb ) surpassed , by far , the size of the Ter macrodomain ( 800kb ) . The influence of Topo IV on sister chromatid interactions [15] prompted the question of how Topo IV would follow replication forks and bind to the newly replicated sister chromatids throughout the cell cycle . We performed ChIP-seq experiments in E . coli dnaC2 strains under conditions suitable for cell cycle synchronization of the entire population . Synchronization was achieved through a double temperature shift , as described previously [15] . Using these conditions , in each cell , S phase is initiated on one chromosome , lasts for 40–45 min and is followed by a G2 phase ( 20 min ) ( S4 Fig ) . We analyzed ParE binding before the initiation of replication , in S phase 20 min ( S20 ) and 40 min ( S40 ) after the initiation of replication and in G2 phase . The synchronization of replication in the population was monitored by marker frequency analysis of the Input DNA ( Fig 1C ) . The profile observed for bacteria that did not replicate at non-permissive temperature was strictly flat , but the S20 replication profile presented two sharp changes of the marker frequency slope around positions 500kb and 2700kb . This suggested that each replication fork had crossed approximately 1000 to 1300 kb in 20 min . The S40 replication profile demonstrated that most cells had finished replication , with the unreplicated region being limited to 300 kb around dif in no more than 20% of the bacteria . In G2 phase the marker frequency was flat . We used flow cytometry to demonstrate that at G2 , the amount of DNA in each bacterium was double compared to that of the G1 bacteria , indicating that cytokinesis has not yet occurred ( S4 Fig ) . We analyzed Topo IV binding at specific binding sites ( Fig 1D ) . Binding at these sites was strongly impaired in the absence of replication . Binding at every site started in the S20 sample and was maximal in the S40 or G2 samples , without showing any marked decrease , even in the oriC-proximal region . These observations suggest that Topo IV binds to specific sites during S phase . However , since enrichment was observed for non-replicated loci and was maintained for a long time after replication , it was not compatible with a model of Topo IV migration with the replication forks . Synchronization experiments with a higher temporal resolution are required to clarify this observation . To measure Topo IV cleavage at the binding sites , we took advantage of the fact that norfloxacin covalently links Topoisomerase II to the gate segment of DNA and prevent its relegation [31] . We first monitored Topo IV activity on the Topo IV enriched regions ( 1 . 2 , 1 . 8 , 2 . 5 , 3 . 2 Mb and dif ) by incubating bacteria with norfloxacin for 10 minutes before genomic extraction and performing Southern blot analysis to detect the cleaved DNA products [10 , 29] . This revealed cleavage fragments induced by both DNA Gyrase and Topo IV poisoning in the WT strain , but only Topo IV cleavage in a nalR strain where DNA Gyrase is resistant to norfloxacin . Among the 5 tested sites , only two displayed clear Topo IV cleavage at the expected position ( Fig 2A ) . As expected , the dif site exhibited strong cleavage . Moreover cleavage was also observed at position 2 . 56 Mb . However the 1 . 2 , 1 . 8 and 3 . 2 Mb sites did not show any Topo IV mediated cleavage in the presence of norfloxacin . The above result prompted us to investigate Topo IV cleavage at the genome-wide scale . We performed IPs in the presence of norfloxacin as a crosslinking agent instead of formaldehyde . Following this step , all downstream steps of the protocol were identical to that of the ChIP-Seq assay . We referred to this method as NorflIP . The NorflIP profile differed from the ChIP-seq profile ( Fig 2B ) . Regions immunoprecipitated with Topo IV-norfloxacin cross-links were frequently observed ( Fig 2C orange circle ) . Similarly to the ChIP-seq experiments , the NorflIP profile revealed strong enrichment over the rRNA operons and IS sequences but not at the tRNA genes ( S5A Fig ) . We used a Southern blot cleavage assay to demonstrate that these signal did not correspond to Topo IV cleavages ( S5B Fig ) . The NorflIP peaks correspond to a ~170 bp forward and reverse enrichment signal separated by a 130 bp segment , which is not enriched . This pattern is the consequence of the covalent binding of Topo IV to the 5’ bases at the cleavage site . After Proteinase K treatment the cleaving tyrosine residue bound to the 5’ extremity resulted in poor ligation efficiency and infrequent sequencing of the cleaved extremities . ( S6A and S6B Fig ) This observation confirmed that we were observing genuine Topoisomerase cleavage sites . We used this pattern to define an automatic peak calling procedure ( S6C Fig ) that identified between 134 and 458 peaks in the three NorflIP experiments , two experiments performed with ParC-Flag and one with ParE-Flag ( Fig 2C purple circles and Fig 2D ) . We observed a total of 571 possible sites in the three experiments with about half of the sites common to at least two experiments and approximately 88 sites common to all three experiments ( S1 Table ) . We analyzed sequencing reads for the three experiments around the dif , 0 . 2 Mb and 1 . 92Mb positions . It revealed abrupt depletions of forward and reverse reads in a 100bp center region suggesting that it corresponds to the site of cleavage . We extrapolated this result for every peak to estimate the cleavage positioning of Topo IV ( ~150bp downstream of the center of the forward peak , S6D Fig ) We manually validated 172 sites that were common to ParC-1 and ParE-1 experiments ( S1 Table ) for further analysis . The Topo IV cleavage at the dif site was the most enriched of the chromosome ( ~ 30 fold ) , fourteen sites were enriched from 5 to 10 fold and other positions were enriched from 2 to 5 fold ( Fig 2E ) . Most NorflIP sites did not correspond to significant peaks in the ChIP-seq experiment ( Fig 2E ) . We also did not observe any cleavage for the majority of the strong binding sites observed by ChIP-seq . This is illustrated for the binding site at 1 . 85 Mb ( Fig 2E ) . We verified several Topo IV cleavage sites by Southern blot , a significant cleaved DNA fragment was observed at the expected size for each of them ( Fig 2F ) . Southern blotting experiments following DNA cleavage in the presence of norfloxacin on synchronized cultures revealed that , like its binding , Topo IV cleavage is coordinated with DNA replication . In good agreement with ChIP-seq experiments , increased cleavage was observed as soon as 20 minutes after initiation of replication for the dif and 2 . 56 Mb sites ( Fig 2G ) . The general genomic distribution of Topo IV cleavage sites was not homogeneous; a few regions had a large number of sites clustered together , while the 1 . 2Mb– 2 . 5 Mb region contained a low density of sites ( Fig 2H ) . We further analyzed the distribution of cleavage sites in the terminus and the oriC regions . In the terminus region , the average distance of consecutive cleavage sites was long ( around 30 kb in the 1 . 5–2 . 5 Mb region ) compared to 8 kb in the 0 . 8–1 . 5 Mb or the 2 . 5–3 . 1 Mb regions ( S7A Fig ) . The oriC region displays a mixed distribution ( S7B Fig ) , a high density of sites near oriC flanked by two depleted regions , including the SNAP2 region [16] . At the gene scale , the mid-point of Topo IV cleavage signal can be localized inside genes ( 82% ) or intergenic regions ( 16% ) but it presents a bias toward the 5’ or 3’ gene extremities ( S7C Fig ) . Since the cleavage signal spans approximately 200bp , nearly 50% of the sites overlapped , at least partly , with intergenic regions that account for only 11% of the genome . Finally , we did not identify any robust consensus between sets of Topo IV cleavage sites . The only sequence traits that we identified are a bias for GC dinucleotides near the center of the sites ( S7D Fig ) and an increased spacing of GATC motifs around cleavage sites ( S7E Fig ) . The bias in the distribution of cleavage sites ( Fig 2H ) was very similar to the Topo IV binding bias revealed by ChIP-seq ( Fig 1C ) . NorflIP and ChIP-seq data were compared on Fig 3A . Despite the lack of corresponding ChIP-seq enrichment at the position of most highly enriched NorflIP sites , a number of consistencies were observed between these two data sets . Overall the NorflIP and ChIP-seq datasets had a Pearson correlation of 0 . 3 and the averaged data ( 1 kb bin ) revealed a Pearson correlation of 0 . 5 . First a small amount of local enrichment in the ChIP-seq experiments was frequently observed in the regions containing many cleavage sites ( Fig 3A and 3C ) . This led us to consider that trapped Topo IV engaged in the cleavage reaction could contribute to a small amount of local enrichment in the ChIP-seq experiments . Second , both Topo IV cleavages and binding sites were rare in highly expressed regions ( Fig 3A ) , only one of the 172 manually validated Topo IV cleavage site overlapped a highly expressed region . However cleavages sites were more frequently , than expected for a random distribution , observed in their vicinity ( Fig 3C and S8 Fig ) . Thirty percent ( 50/172 ) of the Topo IV sites are less than 2 kb away from the next highly expressed transcription unit ( Fig 3 ) . We explored correlations between the localization of Topo IV cleavages and binding sites of various NAPs thanks to the Nust database and tools [32] . A significant correlation was only observed for Fis binding sites ( Fig 3B ) . Sixty eight genes present both Fis binding [33] and Topo IV cleavage ( P value 2x10-03 ) . Thirty-three of the 172 manually validated cleavage sites overlapped at least partially with a Fis binding site , 80 of them are located less than 400 bp away from a Fis binding site . At the genome scale this correlation is difficult to observe ( Fig 3A ) , but close examination clearly revealed overlapping Topo IV cleavages and Fis binding sites ( Fig 3C ) . Fis binding sites are more numerous than Topo IV cleavage sites , therefore a large number of them do not present enrichment for Topo IV ( Fig 3C ) . By contrast , Topo IV peaks are excluded from H-NS rich regions ( Fig 3A , 3B and 3C ) . Only one of the 172 manually validated Topo IV cleavage site overlapped with an H-NS binding site . As observed for highly expressed regions TopoIV cleavage sites were frequently observed at the border of H-NS rich regions ( Fig 3C ) . Moreover H-NS rich regions contain less Topo IV than the rest of the chromosome ( Fig 3A–3D and S9A Fig ) . H-NS rich regions correspond to an AT rich segment of the chromosome ( Fig 3C and 3D ) . Indeed background level of Topo IV binding and cleavage were significantly reduced in AT rich regions ( S9B Fig ) . In rare occasions binding of H-NS has been observed in regions with a regular AT content ( Fig 3C ) , notably Topo IV binding and cleavage were also reduced in these regions . This observation suggested that H-NS itself rather than AT content limits the accessibility of Topo IV to DNA . This observation was confirmed by the identification of Topo IV cleavage in regions with an AT content ranging from 20 to 80% ( S9C and S9D Fig ) . We performed Southern blot analysis of Topo IV cleavage on representative sites to test whether gene expression and chromatin factors influenced Topo IV site selection . First , we observed that the exact deletion of cleavage sites at position 1 . 92 Mb and 2 . 56 Mb did not abolish Topo IV cleavage activity ( Fig 3D and 3E ) . Second , since these loci also contain a Fis binding site overlapping Topo IV cleavage signal , we deleted the fis gene . However , deletion of the fis gene did not modify Topo IV cleavage ( Fig 3D and 3E ) . Finally we performed cleavage assays in the presence of rifampicin to inhibit transcription . To limit the pleiotropic effects of rifampicin addition we performed the experiment with a 20 min pulse of rifampicin . Rifampicin treatment abolished Topo IV cleavage ( Fig 3E ) . These results suggest that gene expression rather than chromatin factors influences Topo IV targeting . Our analysis confirms that the dif region is a hot spot for Topo IV activity [29] . Indeed , ChIPseq and NorflIP show that Topo IV binds to and cleaves frequently in the immediate proximity of dif . We measured DNA cleavage by Topo IV in the presence of norfloxacin in various mutants affecting the structure of dif or genes implicated in chromosome dimer resolution . Southern blot was used to measure Topo IV cleavage ( Fig 4A ) . We observed that exact deletion of dif totally abolished Topo IV cleavage . Interestingly , the deletion of the XerC-binding sequence ( XerC box ) of dif was also sufficient to abolish cleavage , while the deletion of the XerD box only had a weak effect . Deletion of the xerC and xerD genes abolished Topo IV cleavage at dif . However , cleavage was restored when the catalytically inactive mutants XerC K172A or XerC K172Q were substituted for XerC ( Fig 4B ) . This suggests that the role of XerCD/dif in the control of Topo IV activity is structural and independent of XerCD catalysis . Deletion of dif or xerC did not significantly alter cleavage at any of the other tested Topo IV cleavage sites ( Fig 4C ) . This suggests that influence of XerC on Topo IV is specific to dif . To evaluate the role of XerCD-mediated Topo IV cleavage at dif , we attempted to construct parEts xerC , parEts xerD and parCts xerC double mutants . We could not obtain parCts xerC mutants by P1 transduction at any tested temperature . We obtained parEts xerC and parEts xerD mutants at 30°C . The parEts xerC double mutant presented a growth defect phenotype at 30°C and did not grow at temperature above 35°C ( Fig 4D ) . The parEts xerD mutant presented a slight growth defect at 37°C compared to parEts or xerD mutants . None of the parEts mutant grew above 42°C . Next , we used quinolone sensitivity as a reporter of Topo IV activity . To this aim , we introduced mutants of the FtsK/Xer system into a gyrAnalR ( nalR ) strain; Topo IV is the primary target of quinolones in such strains . The absence of XerC , XerD , the C-terminal activating domain of FtsK or dif exacerbated the sensitivity of the nalR strain to ciprofloxacin ( Fig 4D ) . We therefore concluded that the impairment of Topo IV was more detrimental to the cell when the FtsK/Xer system was inactivated . Among partners of the FtsK/Xer system the absence of XerC was significantly the most detrimental , suggesting a specific role for XerC in this process . The above results suggest an interaction between Topo IV and the XerCD/dif complex . We therefore attempted to detect this interaction directly in vitro ( Fig 4E and 4F ) . We performed EMSA with two fluorescently labeled linear probes , one containing dif and the other containing a control DNA not targeted by Topo IV in our genomic assays . Topo IV alone bound poorly to both probes ( Kd > 100nM ) . Binding was strongly enhanced when XerC or both XerC and XerD were added to the reaction mix . In contrast , Topo IV binding to dif was slightly inhibited in the presence of XerD alone . These results were consistent with the observation that deletion of the XerC box but not of the XerD box inhibited Topo IV cleavage at dif and pointed to a specific role for XerC in Topo IV targeting . The control fragment showed that these effects are specific to dif . Topo IV-XerC/dif complexes were stable and resisted a challenge by increasing amount of XerD ( S10A Fig ) . The positive influence of XerCD on TopoIV binding was also observed on a negatively supercoiled plasmid containing dif . In the presence of XerCD ( 50nM ) , a delay in the plasmid migration was observed with 40nM of TopoIV . By contrast , 200 nM was required in the absence of XerCD ( S10B Fig ) . The Southern blot cleavage assay showed that overexpression of the ParC C-terminal domain ( pET28parC-CTD ) strongly reduced cleavage at dif but enhanced cleavage at the Topo IV site located at 2 . 56Mb . This suggested that , as observed for MukB [17] , Topo IV might interact with XerC through its C-terminal domain ( Fig 4G ) . We assayed the effects of the reported Topo IV modulators and proteins involved in chromosome segregation the activity of Topo IV at dif . MukB has previously been shown to influence the activity of Topo IV [17 , 18] . We measured Topo IV cleavage in a mukB mutant at dif and at position 2 . 56 Mb , cleavage was reduced at dif but no significant effect was observed at position 2 . 56Mb ( Fig 5A ) . We did not detect any effect of a seqA deletion on Topo IV cleavage at either position ( Fig 5B ) . We next assayed the effect of MatP , which is required for compaction and intracellular positioning of the ter region as well as for the its progressive segregation pattern ending at dif [25 , 26] . The Topo IV cleavage at dif was significantly impaired in the matP mutant ( Fig 5C ) . The Topo IV cleavage site at position 1 . 9Mb is included in the Ter macrodomain , but cleavage at this site was almost unchanged in the absence of MatP ( Fig 5C ) . Introduction of a matP deletion into the nalR strain yielded an increase in ciprofloxacin sensitivity ( Fig 5D ) . We also constructed a parEts matP double mutant . Growth of this strain was significantly altered compared to the parEts parental strain at an intermediate temperature ( Fig 5E ) . Such a synergistic effect was not found when combining the matP deletion with a gyrBts mutation . Taken together , these results led us to consider that MatP itself or the folding of the Ter macrodomain might be important for Topo IV targeting at dif . Since the FtsK/Xer/dif system is dedicated to post-replicative events that are specific to a circular chromosome , it was tempting to postulate that the activity of Topo IV at dif is also dedicated to post-replicative decatenation events and is strictly required for circular chromosomes . To address this question , we used E . coli strains harboring linear chromosomes [34] . In this strain , expression of TelN from the N15 phage promotes linearization of the chromosome at the tos site inserted a 6kb away from dif . Indeed , chromosome linearization suppresses the phenotypes associated with dif deletion [34] . We analyzed cleavage at the dif site by Topo IV in the context of a linearized chromosome . Cleavage was completely abolished; showing that Topo IV activity at dif is not required on linear chromosomes . This effect was specific to the dif site , since cleavage at the 1 . 9Mb site remained unchanged after chromosome linearization ( Fig 5F ) . We next assayed if the phenotypes associated with matP deletion , i . e . , formation of elongated cells with non-partitioned nucleoids [26] , depend on chromosome circularity . Strikingly , most of the phenotypes observed in the matP mutant were suppressed by linearization of the chromosome ( Fig 5G ) . Interestingly , the frequency of cleavage at dif sites inserted far ( 300 kb ) from the normal position of dif or in a plasmid were significantly reduced compared to the WT situation ( S11 Fig ) confirming that Topo IV cleavage at dif is specific to circular chromosomes .
Whole genome analysis of Topo IV binding by ChIP-seq revealed approximately 10 Topo IV binding sites across the E . coli genome . Among them , only 5 sites were strongly enriched in every experiment and these were mapped to positions 1 . 25 , 1 . 58 ( dif ) , 1 . 85 , 2 . 56 and 3 . 24 Mb . We did not identify any consensus sequence that could explain specific binding to these sites . Band shift experiments at the dif site and the 1 . 25 Mb site revealed that Topo IV binding is not sequence-dependent . This led us to favor models involving exogenous local determinants for Topo IV binding as it is the case for the dif site in the presence of XerC . Because XerC is only known to bind to dif , we could speculate that other chromatin factors might be involved in Topo IV targeting . Topo IV and Fis binding sites [33] overlap more frequently than expected ( Nust P value 10e-03 [32] . Topo IV and Fis binding sites overlap at the positions 1 . 25 and 2 . 56 Mb; it is therefore possible that Fis plays a role in defining some Topo IV binding sites . However our EMSA , cleavage and ChIP experiments did not show any cooperative binding of Topo IV with Fis . In spite of its co-localization with Topo IV , Fis does not contribute in defining Topo IV binding or cleavage sites . Nevertheless , the role of the chromatin in Topo IV localization was also illustrated by the strong negative correlation observed for the Topo IV and H-NS bound regions . H-NS rich regions were significantly less enriched for nonspecific Topo IV binding than the rest of the chromosome . We postulated that loci where Topo IV is catalytically-active could be identified by DNA cleavage mediated by the quinolone drug norfloxacin . We designed a new ChIP-seq strategy that consisted of capturing DNA-norfloxacin-Topo IV complexes . We called it NorflIP . Three independent experiments show that Topo IV was trapped to a large number of loci ( 300 to 600 ) with most of these loci observed in two out of three experiments . A hundred of these loci were identified in all three experiments . Dif presented a strong signal in the NorflIP as in the ChIP-seq but this is not the case for most of the other ChIP-seq peaks . NorflIP peaks presented a characteristic pattern suggesting that they are genuine DNA-norfloxacin-Topo IV complexes . Considering that norfloxacin does not alter Topo IV specificity , our results suggest that for Topo IV the genome is divided into five categories: i ) Loci where Topo IV binds strongly but remains inactive for most of the cell cycle; ii ) Loci where Topo IV is highly active but does not reside for very long time; iii ) Loci where we observed both binding and activity ( dif and 2 . 56 Mb ) ; iv ) regions where Topo IV interacts non-specifically with the DNA and where topological activity is not stimulated; v ) regions where non-specific interactions are restricted ( the Ter domain , chromatin rich regions ( tsEPODs [35] , H-NS rich regions ) . Detection of norfloxacin-mediated genomic cleavage by pulse field electrophoresis has previously revealed that when Topo IV is the only target of norfloxacin the average fragment size is 300–400 kb while it drops to 20 kb when Gyrase is the target [11] . This suggests that , for each cell , no more than 10 to 20 Topo IV cleavages are formed in 10 min of norfloxacin treatment . To fit this observation with our data , only a small fraction ( 10–20 out of 600 ) of the detected Topo IV cleavage sites would actually be used in each cell . This might explain why Topo IV cleavage sites were hardly distinguishable from background in the ChIP-seq assay ( Fig 3 ) . This is in good agreement with the estimation that the catalytic cycle only provokes a short pause ( 1 . 8 sec ) in Topo IV dynamics [36] . The mechanism responsible for the choice of specific Topo IV cleavage sites is yet to be determined . As indicated by our findings that deletion of the cleavage site resulted in the formation of a new site or sites in the vicinity , cleavage is not directly sequence-related . We observed several biases that might be involved in determination of cleavage sites ( GC di-nucleotide skew , GATC spacing , positioning near gene ends or intergenic regions , proximity with highly expressed genes and Fis binding regions ) . Interestingly inhibition of transcription with rifampicin inhibits Topo IV cleavage ( Fig 3 ) . This raises the possibility that transcription , that can be stochastic , may influence stochastic determination of Topo IV activity sites . The influence of transcription could be direct , if RNA polymerase pushes Topo IV to a suitable place , or indirect if the diffusion of topological constraints results in their accumulation near barriers imposed by gene expression [37 , 38] . This accumulation could then , in turn , signal for the recruitment of Topo IV . Synchronization experiments revealed that , like Topo IV binding at specific sites , Topo IV cleavage activity is enhanced by chromosome replication . Enrichment was the highest in late S phase or G2 phase; it seems to persist after the passage of the replication fork at a defined locus . Enrichment in asynchronous cultures was significantly reduced compared to S40 or G2 synchronized cultures suggesting that Topo IV is not bound to the chromosome for the entire cell cycle . Unfortunately our experiments did not have the time resolution to determine at what point of the cell cycle Topo IV leaves the chromosome and if it would leave the chromosome during a regular cell cycle . The role of DNA replication of Topo IV dynamics has recently been observed by a very different approach [36] . The authors propose that Topo IV accumulates in the oriC proximal part of the chromosome in a MukB and DNA replication dependent process . These observations are in good agreement with our data and suggest that Topo IV is loaded on DNA at the time of replication , accumulate towards the origin of replication and remains bound to the DNA until a yet unidentified event triggers its release . Formation of positive supercoils and precatenanes ahead and behind of the replication forks respectively , could be the reason for Topo IV recruitment . One could hypothesize that MukB is used as a DNA topology sensor that is responsible for redistribution of Topo IV . However we only detected a modest effect of mukB deletion on Topo IV cleavage at dif ( Fig 5 ) . Putative events responsible for Topo IV release could be , among others , complete decatenation of the chromosome , SNAPs release , or stripping by other proteins such as FtsK . Non-specific Topo IV binding presents a very peculiar pattern; it is significantly higher in the oriC proximal 3Mb than in the 1 . 6Mb surrounding dif . This pattern is not simply explained by the influence of replication ( S3 Fig ) . Interestingly , ChIP-seq and ChIP-on-Chip experiments have already revealed a similar bias for DNA gyrase [12] and SeqA [39] . The CbpA protein has been shown to present an inverse binding bias [40] , with enrichment in the terminal region and a reduction in the oriC proximal domain . The HU regulon has also presented a similar bias [41] . The terminus domain defined by these biases always comprises the Ter macrodomain but it extends frequently beyond the extreme matS sites . The role of MatP in the definition of these biases has not yet been tested . The group of G . Mushelishvili proposed a topological model to interpret the DNA gyrase and HU regulon biases , suggesting that HU coordinates the global genomic supercoiling by regulating the spatial distribution of RNA polymerase in the nucleoid [41] . Topo IV could benefit from such a supercoiling gradient to load on the chromosome . Interestingly , the strongest Topo IV binding and cleavage sites are localized inside the Terminus depleted domain . One possibility could be that these sites minimize Topo IV binding to adjacent nonspecific sequences . Alternatively one can propose that a regional reduction of non-specific binding creates a selective advantage for optimal loading on to specific sites . Dif was the strongest Topo IV cleavage site detected by NorflIP , it was also detected in the ChIP-seq assays . We have used Southern blot to analyze the determinants involved in this activity . The binding of XerC on the xerC box of dif and the region downstream of the xerC box are essential . In vitro , XerC also strongly favors binding of Topo IV at dif . Interestingly XerD and the xerD box did not improve Topo IV binding or cleavage . We propose that XerC works as a scaffold for Topo IV , simultaneously stimulating its binding and its activity . Topo IV activity at dif is also dependent on the circularity of the chromosome , suggesting that when topological constraints can be evacuated through chromosome ends , Topoisomerase IV does not catalyze strand passage at dif . This suggests that topological complexity is directly responsible for Topo IV activity . Topo IV cleavage activity at dif is not influenced by SeqA or FtsK , which are two known Topo IV partners . Interestingly , mukB and matP deletion mutants slightly reduced this activity . The synergistic effect observed when a matP deletion is combined with a parEts mutation suggests that MatP indeed influences Topo IV activity . The phenotypes of the matP mutant are rescued by the linearization of the chromosome . A similar rescue has been observed for the dif mutant [34] . Therefore it is likely that a significant part of the problems that cells encounter in the absence of matP corresponds to failure in chromosome topology management , either decatenation or chromosome dimer resolution [25] . In conclusion , we propose that genomic regulation of Topo IV consists of: ( 1 ) Topo IV loading during replication , ( 2 ) Topo IV binding to specific sites that may serve as reservoirs , ( 3 ) Topo IV activation to remove precatenanes or positive supercoils in a dozen of stochastically chosen loci ( 4 ) XerC and MatP ensuring the loading of Topo IV at the dif site for faithful decatenation of fully replicated chromosomes .
ParE-flag and ParC-flag C-terminus fusions were constructed by lambda red recombination [42] . Cultures were grown in LB or Minimal medium A supplemented with succinate ( 0 . 2% ) and casamino acids ( 0 . 2% ) . Cells were fixed with fresh Formaldehyde ( final concentration 1% ) at an OD600nm 0 . 2–0 . 4 . Sonication was performed with a Bioruptor Pro ( Diagenode ) . Immunoprecipitations were performed as previously described 26 . Libraries were prepared according to Illumina's instructions accompanying the DNA Sample Kit ( FC-104-5001 ) . Briefly , DNA was end-repaired using a combination of T4 DNA polymerase , E . coli DNA Pol I large fragment ( Klenow polymerase ) and T4 polynucleotide kinase . The blunt , phosphorylated ends were treated with Klenow fragment ( 3’ to 5’ exo minus ) and dATP to yield a protruding 3- 'A' base for ligation of Illumina's adapters which have a single 'T' base overhang at the 3’ end . After adapter ligation DNA was PCR amplified with Illumina primers for 15 cycles and library fragments of ~250 bp ( insert plus adaptor and PCR primer sequences ) were band isolated from an agarose gel . The purified DNA was captured on an Illumina flow cell for cluster generation . Libraries were sequenced on the Genome Analyzer following the manufacturer's protocols . Norfloxacin ( final concentration 2μM ) was added to the cultures at OD600nm 0 . 2 LB for 10 min before harvesting . Sonication and immunoprecipitation were performed as described for the ChIP-seq assay . Sequencing results were processed by the IMAGIF facility . Base calls were performed using CASAVA version 1 . 8 . 2 . ChIP-seq and NorflIP reads were aligned to the E . coli NC_000913 genome using BWA 0 . 6 . 2 . A custom made pipeline for the analysis of sequencing data was developed with Matlab ( available on request ) . Briefly , the number of reads for the input and IP data was smoothed over a 200bp window . Forward and reverse signals were added , reads were normalized to the total number of reads in each experiment , strong non-specific signals observed in unrelated experiments were removed , data were exported to the UCSC genome browser for visualization and comparisons . The strongest peaks observed with NorflIP experiments ( dif and 1 . 9 Mb ) present a characteristic shape ( S6 Fig ) that allows the automatic detection of lower amplitude peaks but preserves the characteristic shape . We measured Pearson correlation coefficient with the dif and the 1 . 9 Mb site for 600bp sliding windows over the entire genome . Peaks with a Pearson correlation above 0 . 72 were considered as putative Topo IV cleavage sites . Sequencing data are available on the GEO Repository ( http://www . ncbi . nlm . nih . gov/geo/ ) with the accession number GSE75641 . Data were plotted with the Circos tool [43] and UCSC Archaeal Genome Browser [44] . Cleavage of DNA by Topo IV in the presence of Norfloxacin was monitored by Southern blot as previously described [10] . DNA was extracted from E . coli culture grown in minimal medium supplemented with glucose 0 . 2% and casaminoacids 0 . 2% . Norfloxacin ( final concentration 10μM ) was added to the cultures at OD 0 . 2 for 10 min before harvesting . DNA was transferred by neutral blotting on nitrocellulose membranes . For synchronization experiments a flash freeze step in liquid nitrogen is included before harvesting . Quantification was performed with Image J software . Experiments were conducted using Cy3-coupled probes harboring the dif site and a Cy5-coupled dye as control . Reactions were carried out in EMSA reaction buffer ( 1mM spermidine , 30mM potassium glutamate , 10mM DTT , 6mM magnesium chloride , 10% glycerol , pH 7 . 4 ) . Reactions were incubated for 15 min at RT , loaded on 4% native PAGE gel at 25 volts and then run at 125 volts for 2 hours . Gels were then visualized using a Typhoon FLA 5000 scanner ( GE healthcare Life Science ) . EMSA of plasmids were performed with unlabeled supercoiled plasmid in the same reaction buffer . Electrophoresis was performed in a 0 . 8% agarose gel in 0 . 5x TAE buffer at 4°C for 80 min at 150V . DNA labeling was performed with SYBR green . | DNA topoisomerases are ubiquitous enzymes that solve the topological problems associated with replication , transcription and recombination . Type II Topoisomerases play a major role in the management of newly replicated DNA . They contribute to the condensation and segregation of chromosomes to the future daughter cells and are essential for the optimal transmission of genetic information . In most bacteria , including the model organism Escherichia coli , these tasks are performed by two enzymes , DNA gyrase and DNA Topoisomerase IV ( Topo IV ) . The distribution of the roles between these enzymes during the cell cycle is not yet completely understood . In the present study we use genomic and molecular biology methods to decipher the regulation of Topo IV during the cell cycle . Here we present data that strongly suggest the interaction of Topo IV with the chromosome is controlled by DNA replication and chromatin factors responsible for its loading to specific regions of the chromosome . In addition , our observations reveal , that by sharing several key factors , the DNA management processes ensuring accuracy of the late steps of chromosome segregation are all interconnected . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"bacteriology",
"chemical",
"characterization",
"molecular",
"probe",
"techniques",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"microbiology",
"dna",
"replication",
"bacterial",
"genetics",
"dna",
"molecular",
"biology",
"techniques",
"microbial",
... | 2016 | Mapping Topoisomerase IV Binding and Activity Sites on the E. coli Genome |
As antimicrobial signalling molecules , type III or lambda interferons ( IFNλs ) are critical for defence against infection by diverse pathogens , including bacteria , fungi and viruses . Counter-intuitively , expression of one member of the family , IFNλ4 , is associated with decreased clearance of hepatitis C virus ( HCV ) in the human population; by contrast , a natural frameshift mutation that abrogates IFNλ4 production improves HCV clearance . To further understand how genetic variation between and within species affects IFNλ4 function , we screened a panel of all known extant coding variants of human IFNλ4 for their antiviral potential and identify three that substantially affect activity: P70S , L79F and K154E . The most notable variant was K154E , which was found in African Congo rainforest ‘Pygmy’ hunter-gatherers . K154E greatly enhanced in vitro activity in a range of antiviral ( HCV , Zika virus , influenza virus and encephalomyocarditis virus ) and gene expression assays . Remarkably , E154 is the ancestral residue in mammalian IFNλ4s and is extremely well conserved , yet K154 has been fixed throughout evolution of the hominid genus Homo , including Neanderthals . Compared to chimpanzee IFNλ4 , the human orthologue had reduced activity due to amino acid K154 . Comparison of published gene expression data from humans and chimpanzees showed that this difference in activity between K154 and E154 in IFNλ4 correlates with differences in antiviral gene expression in vivo during HCV infection . Mechanistically , our data show that the human-specific K154 negatively affects IFNλ4 activity through a novel means by reducing its secretion and potency . We thus demonstrate that attenuated activity of IFNλ4 is conserved among humans and postulate that differences in IFNλ4 activity between species contribute to distinct host-specific responses to—and outcomes of—infection , such as HCV infection . The driver of reduced IFNλ4 antiviral activity in humans remains unknown but likely arose between 6 million and 360 , 000 years ago in Africa .
Vertebrates have evolved the capacity to coordinate their antiviral defences through the action of proteins called interferons ( IFNs ) [1] , which are small secreted signalling proteins produced by cells after sensing viral infection . IFNs bind to cell surface receptors , commencing autocrine and paracrine signalling via the ‘JAK-STAT’ pathway . Through this mechanism , IFNs induce expression of hundreds of ‘interferon-stimulated genes’ ( ISGs ) that establish a cell-intrinsic ‘antiviral state’ and regulate cellular immunity and inflammation [2 , 3] . Thus , IFNs are pleiotropic in activity and modulate aspects of protective immunity and pathogenesis [4] . Three groups of IFNs have been identified ( types I–III ) , with the type III family ( termed IFNλs ) being the most recently discovered [5 , 6] . Emerging evidence highlights the critical and non-redundant role that IFNλs play in protecting against diverse pathogens , including viruses , such as norovirus [7] , influenza virus [8] and flaviviruses [9]; bacteria [10]; and fungi [11] . While IFNλs induce nearly identical genes to type I IFNs , differences in signalling kinetics and cell-type specificity contribute to their specialisation [12 , 13] . Hence , as a consequence of selective expression of the IFNλ receptor 1 ( IFNλR1 ) co-receptor on epithelial cells [13] , type III IFNs play a significant role in defence of ‘barrier tissues’ , such as the gut , respiratory tract and liver [reviewed in 14]; the second co-receptor for IFNλ is IL10-R2 , which is expressed more broadly . Although important for host defence , some IFNs are highly polymorphic [15] . In humans , a number of genetic variants in the type III IFN locus ( containing IFNλs 1–4 ) have been identified and are associated with clinical phenotypes relating to viral infection [16–18] . Although many of these variants are in linkage disequilibrium , the major functional variant is thought to lie in the IFNL4 gene [19] . This causative variant is a single substitution/insertion mutation converting the ‘ΔG’ allele to a ‘TT’ allele ( rs368234815 ) , thereby yielding a frameshift which leads to loss of active human IFNλ4 ( HsIFNλ4 ) [18] . Genome-wide association studies have convincingly demonstrated a seemingly counter-intuitive correlation between the IFNL4 ΔG allele and reduced clearance of hepatitis C virus ( HCV ) infection , i . e individuals who produce HsIFNλ4 clear HCV infection with reduced frequency in the presence or absence of antiviral IFN therapy [17 , 18] . Although IFNλ4 is highly conserved among mammals , the ‘pseudogenising’ TT allele of HsIFNλ4 has evolved under positive selection in some human populations suggesting that expression of the wild-type protein likely conferred a fitness cost during recent human evolution [20] . Expression of IFNλ4 is tightly controlled and reduced in human as well as Gorilla cells following viral infection compared to IFNλ3 [21] . The mechanism underlying the contribution of HsIFNλ4 to viral persistence in HCV infection is not well understood but is associated with enhanced ISG induction . Moreover , a common natural variant of HsIFNλ4 ( P70S ) [18] , which has reduced signalling capacity , is also linked with improved HCV clearance [22] . Thus , there is a spectrum of HsIFNλ4 activity in humans as a consequence of natural variation that has a significant influence on chronic HCV infection , with wt HsIFNλ4 representing the protein with apparently the greatest antiviral activity . Whether other human IFNλ4 variants exist in addition to P70S , which affect antiviral activity , has not been explored fully . In this study , we have examined human genetic data to identify other possible naturally occurring IFNλ4 variants and performed comparative analysis with mammalian orthologues in species closely related to humans . We provide evidence that the antiviral potential for the most common form of IFNλ4 in humans has attenuated activity due to a single amino acid substitution . In addition , we propose that acquisition of the attenuating substitution arose very early during human evolution but that some populations do encode a more active variant . Mechanistically , our data show that the reduced antiviral potential of human IFNλ4 results from a likely dual defect in secretion and potency .
Firstly , we undertook genetic and functional comparisons of natural human IFNλ4 coding variants present in the human population . We identified 15 non-synonymous HsIFNλ4 variants in the 1000 Genomes Project Database [23] ( Fig 1A and S1 Data ) , including three previously described variants ( C17Y , P60R and P70S; >1% global frequency , classified as ‘common’ ) [18] . The remaining 12 variants were classified as rare ( <1% global frequency ) . The African population harboured the largest number of , as well as the most unique , variants . Interestingly , three rare variants ( A8S , S56R and L79F ) were shared exclusively between African and American populations , which may have arisen due to relatively recent movements of people perhaps through the transatlantic slave trade . Variants were located in regions of functional significance in the HsIFNλ4 protein ( Fig 1B and S1A–S1C Fig ) , such as the predicted signal peptide ( amino acids 1–24 ) , surrounding the single glycosylation site ( N61 ) [both of which are required for secretion of active protein] , and helix F that is predicted to interact with the IFNλR1 receptor ( variants 151–158 ) [24] . Interestingly , the variants in helix F were clustered in the N-terminal portion of the predicted helix . Based on the above predictions , we hypothesised that some of these variants may have phenotypic effects on HsIFNλ4 function . Of note , no variants were found in helix D , which is predicted to contribute to interaction with the IL-10R2 receptor , nor on the IL-10R2-interacting face of the protein . The functional impact of variation on HsIFNλ4 has only been assessed for the common P70S variant and so we sought to screen all other variants in activity assays . To determine whether variants affected HsIFNλ4 antiviral activity , they were introduced independently into an expression plasmid that produced HsIFNλ4 with a C-terminal ‘FLAG’ tag . Transient transfection of the expression plasmids into human ‘producer’ cells ( HEK-293T cells ) allowed harvesting of active HsIFNλ4 in the cell supernatant ( referred to herein as conditioned media [CM] ) thereby enabling analysis of the effects of variants on HsIFNλ4 production , glycosylation , secretion and potency; a similar approach has been successfully adopted previously to determine the relative activities of secreted HsIFNλ3 and HsIFNλ4 as well as a HsIFNλ4 variant that is not glycosylated [24] . We chose to screen the function of the panel of HsIFNλ4 variants on the interferon-competent hepatocyte cell line HepaRG cells [25] . Firstly , we investigated the antiviral activity of variants by titrating them against encephalomyocarditis virus ( EMCV ) , a highly IFN-sensitive and cytopathic virus used to measure IFN-mediated protection [26] ( Fig 1C and S2A Fig ) . We also measured their capacity to induce two major ISGs , MX1 and ISG15 , by RT-qPCR ( Fig 1D and 1E and S2B and S2C Fig ) and validated the ISG15 mRNA data by determining production of unconjugated ‘mono’ ISG15 and high-molecular weight ISG15-conjugates ( ‘ISGylation’; S2D Fig ) . In addition , we constructed a series of negative controls ( plasmids expressing EGFP and the frameshift TT variant of HsIFNλ4 ) , a positive control ( HsIFNλ3op ) for comparative analysis to examine HsIFNλ4 activity , and three HsIFNλ4 variants , which do not occur naturally but were included as they could alter post-translational modification ( N61A which ablates glycosylation ) or potential receptor interactions ( F159A and L162A located in helix F ) , respectively [27] . Negative controls ( EGFP or the frameshift TT variant ) gave very low induction of ISG15 and MX1 and no detectable antiviral activity in the EMCV assay whereas the positive control ( HsIFNλ3op ) was highly active in both assays ( S2A–S2C Fig ) . The non-natural variants N61A and F159A almost abolished activity compared to wt HsIFNλ4 and HsIFNλ3op while L162A gave slightly less activity in the ISG induction assay but activity was reduced to a greater extent in the EMCV assay . In a previous report , ablating glycosylation at N61 substantially reduced activity of secreted HsIFNλ4 in an ISG induction assay [24] . Thus , our assay systems recapitulated findings from previous studies with similar assays and provided a range of activities to assess the impact of the natural HsIFNλ4 variants . Our analyses on the natural HsIFNλ4 variants revealed that only three variants ( P70S , L79F and K154E ) consistently and substantially modulated antiviral activity and signalling compared to wt HsIFNλ4 ( Fig 1C–1E ) . The impact of these variants was particularly pronounced in the EMCV assay that measures the dilution giving 50% activity over a large range of dilutions ( Fig 1C ) . Our results confirmed previous observations on the lower activity of the P70S variant [22] and demonstrated that the rare L79F variant had a similar phenotype . By contrast , the K154E variant substantially enhanced antiviral activity and ISG induction . These effects on activity for P70S , L79F and K154E did not arise from differences in the levels of HsIFNλ4 intracellular production or changes to glycosylation ( S3A and S3B Fig ) . However , variants S56R and R60P ( R60P is a common variant in Africa ) did lead to marked reductions in the glycosylated form of HsIFNλ4 as demonstrated by the mean ratio of glycosylated:non-glycosylated protein ( S3 Fig ) but did not greatly alter their antiviral activity in contrast with our findings with the N61A non-natural variant , which abolished both glycosylation and antiviral activity of conditioned media ( S2A–S2C Fig and S3A and S3B Fig ) . From this screen , we concluded that three non-synonymous variants in HsIFNλ4 ( P70S , L79F and K154E ) , identified as either common or rare alleles in the human population , affect the antiviral activity of the protein . Examining the global distribution of genetic variation can help understand its origins , evolution and functional consequences . P70S is a common variant that is found worldwide ( in every population in the 1000 Genomes Database ) . By contrast , L79F and K154E are rare and , based on evidence in the 1000 Genomes Project Database , geographically restricted to West Africa/Americas , and central Africa , respectively ( Fig 1A ) . We were able to obtain DNA from lymphoblastoid cell lines developed from the 2 individuals encoding the L79F variant; our analysis revealed that the West African subject carried the SNP for this variant but not the individual from the Americas see Materials and Methods ) . ) . Thus , we could verify the occurrence of the L79F variant in Africa but not in the Americas . From further interrogation of the 1000 Genomes Database [28] , the HsIFNλ4 K154E variant was present in two individuals from different African rainforest ‘Pygmy’ hunter-gatherer populations ( Baka and Bakola ) in Cameroon ( S4A Fig ) . The Bakola individual was homozygous for the ΔG allele , indicating that the K154E variant would be encoded on one of the functional ΔG HsIFNλ4 alleles . The Baka subject was heterozygous at rs368234815 ( ΔG/TT ) and thus only one allele ( ΔG ) would produce full-length HsIFNλ4 , presumably K154E . The other allele ( TT ) is a pseudogene and would not lead to expression of functional HsIFNλ4 . Each of the Baka and Bakola individuals also had additional non-synonymous HsIFNλ4 variants ( V158I and R151P , Baka and Bakola individuals respectively ) ; these variants were included in our functional screen of HsIFNλ4 variants but did not significantly alter activity ( Fig 1C–1E and S2A–S2C Fig ) . From analysis of the Genome Aggregation Database ( gnomAD ) [29] , the SNP variant resulting in the K154E substitution was found in a further 29 out of a population of 8 , 655 African individuals . Thus , this SNP is rare in the African population ( 0 . 003% ) compared to the combined Baka/Bakola groups ( 20% ) . K154E was not found in other East or Southern African hunter-gatherer populations ( such as Hadza and Sandawe ) nor in the African San , who have the oldest genetic lineages among humans [30] ( S4B Fig ) ; it was also not identified in Neanderthal and Denisovan lineages ( denoted as ‘archaic’ in S4B Fig ) . However , E154 is encoded in the IFNλ4 orthologue for the chimpanzee , Pan troglodytes ( Pt ) , our closest mammalian species . Notably , the human TT allele encodes a potential K154 codon [18] suggesting that the E154K substitution arose in humans prior to IFNL4 pseudogenisation . Together with the fact that nearly all humans encode K154 , these data suggest that the less active E154K substitution emerged early during human evolution after the divergence of our last common ancestor with chimpanzees . Since a lysine residue encoded at positon 154 is unique to humans compared to other mammalian species ( Fig 2A ) [31] , we compared wt HsIFNλ4 and its K154E variant to wt PtIFNλ4 and an equivalent ‘humanised’ PtIFNλ4 E154K mutant in both the EMCV and ISG induction assays as well as a CRISPR-Cas9 cell line in which the EGFP coding region had been introduced into the endogenous ISG15 gene upstream of and in-frame with the ISG15 open reading frame ( ORF ) ( S5 Fig ) . This cell line offered advantages over other approaches since it facilitated measurement of ISG induction of an endogenous gene by assessing EGFP fluorescence across a range of dilutions of secreted IFNλs ( S5B Fig ) . Although intracellular expression levels of each IFNλ4 variant were similar , ( Fig 2B ) , wt PtIFNλ4 was significantly more active than HsIFNλ4 in each assay and had approximately equivalent activity to the HsIFNλ4 K154E variant in signalling as well as antiviral assays ( Fig 2C–2E ) . Converting PtIFNλ4 to encode the E154K variant significantly decreased activity to levels that were similar to those for wt HsIFNλ4 ( encoding lysine at position 154 ) . Extending the analysis to include rhesus macaque IFNλ4 ( Macaca mulatta , MmIFNλ4 ) gave the same pattern whereby wt MmIFNλ4 with E154 had greater activity than its K154 variant . However , wt MmIFNλ4 was less active than either the human or chimpanzee IFNλ4 with E154 indicating that other genetic differences likely modified MmIFNλ4 activity in our assays . Consistent with the hypothesis that additional genetic differences affect susceptibility to E154K , introducing a lysine into the equivalent position of HsIFNλ3 had a much lesser effect on its activity compared to IFNλ4 ( Fig 2C–2E ) . Overall , we observed a similar ~100-fold enhancement of activity for E154 over K154 for each of the IFNλ4 orthologues in anti-EMCV activity and EGFP IFN reporter induction . Thus , we conclude that wt HsIFNλ4 has attenuated activity principally because of a single amino acid change at position 154 . To broaden analysis of the impact of a lysine residue compared to a glutamic acid at position 154 in HsIFNλ4 , antiviral assays were conducted with other human viruses that are less sensitive to exogenous IFN compared to EMCV , and on different cell lines . Specifically , we used HCV infection in Huh7 cells as well as infectious assays with influenza A virus ( IAV ) and Zika virus ( ZIKV ) in A549 cells against single high dilutions of each IFN . As controls , we also included the less active P70S and L79F HsIFNλ4 variants alongside HsIFNλ3op in these assays . Using the HCVcc infectious system in Huh7 cells , HsIFNλ4 K154E significantly decreased both viral RNA abundance compared to wt protein and exhibited a trend towards a lower number of infected viral antigen ( NS5A ) -positive cells ( Fig 3A , upper and lower panels respectively ) . Furthermore , we performed assays examining HCV entry ( HCV pseudoparticle system [HCVpp] ) , viral RNA translation and RNA replication ( both assessed with the HCV sub-genomic replicon system ) . There was no significant difference in the efficiency of HCVpp infection between wt HsIFNλ4 and any of the three variants tested in the MLV-based pseudoparticle assay ( Fig 3B , upper panel ) . However , we did observe a greater inhibition when the non-HCV E1E2-containing PPs were used , potentially reflecting the higher efficiency or different mode of entry of HCVpp entry compared to non-glycoprotein-containing retroviral PPs that could saturate an inhibitory response ( Fig 3B , lower panel ) . wt HsIFNλ4 reduced HCV RNA replication compared to EGFP and introducing the K154E mutation into wt HsIFNλ4 gave a further significant reduction in replication . ( Fig 3C , upper panel ) . However , primary translation of input viral RNA was not affected by HsIFNλ addition ( Fig 3C , lower panel ) . To examine further the inhibitory effect of wt HsIFNλ4 and the K154E variant on viral RNA replication , Huh7 cells that constitutively expressed a HCV sub-genomic replicon [Tri-JFH1; 32] were treated with both forms of the protein over several passages ( S6A Fig ) . Our results revealed a consistent decrease in HCV RNA levels over 8 passages spanning 25 days with the K154E variant exerting a greater inhibition on RNA replication compared to wt HsIFNλ4; consistent with this conclusion , fewer sub-genomic replicon-bearing cells survived treatment with HsIFNλ4 K154E than the wt protein ( S6B and S6C Fig ) . Thus , HsIFNλ4 reduces HCV RNA replication and the K154E variant exerts greater potency against this stage in the virus life cycle . HsIFNλ4 K154E also reduced titers of IAV and ZIKV to a greater extent than wt protein in A549 cells ( ~10-fold; Fig 3D and 3E ) . Although this was only statistically significant in the context of IAV , a similar trend was evident with ZIKV for the K154E variant compared to wt HsIFNλ4 . We found that the P70S and L79F variants consistently reduced the ability of wt HsIFNλ4 to protect against infection in most assays . Taken together , our data further confirmed the greater antiviral activity associated with converting a lysine residue at position 154 in HsIFNλ4 to a glutamic acid residue . The enhanced antiviral activity of HsIFNλ4 E154 against multiple viruses in different cell lines suggested that this variant may differentially affect global transcription of antiviral ISGs . To test this hypothesis and examine the impact of HsIFNλ4 on global transcription , A549 cells were treated with wt and variant forms of HsIFNλ4 that had different antiviral activities and transcriptional changes were analysed by RNA-Seq at 24 hrs post stimulation ( Fig 4 ) . A549 cells were used because they recapitulate the functional differences in HsIFNλs as observed in other cell types and are widely used as a cell line model for epithelial antiviral immunity . The data revealed that K154E induced the broadest profile of significantly differentially-regulated genes ( n = 273 ) compared with either the wt protein ( n = 178 ) or the P70S variant ( n = 115; Fig 4A–4C and S2 Data ) . The pattern of genes induced by the positive control HsIFNλ3op and HsIFNλ4 K154E were very similar ( Fig 4B and 4C ) . From IPA pathway analysis , all HsIFNλs induced the same transcriptional programmes with differences in the overall significance of these pathways , most notably enhancement of the antigen presentation and protein ubiquitination pathways with the K154E variant ( Fig 4D ) . Many of the differentially-expressed genes shared by HsIFNλ4 wt , K154E and P70S included known restriction factors with antiviral activity ( e . g . IFI27 , MX1 , ISG15; Fig 4E ) although the magnitude of induction was consistently greatest for HsIFNλ4 K154E ( Fig 4F ) . There were also several ISGs that only achieved significant induction by K154E and HsIFNλ3op ( e . g . IDO1 , IRF1 and ISG20; Fig 4E and 4F ) . We predict that the apparent selectivity by IFNλ4 K154E results from the greater potency of this variant compared to wt . HsIFNλ4 and the P70S variant allowing genes to reach the significance threshold ( Fig 4F ) . Enhanced production of antiviral genes in cells treated with HsIFNλ4 E154 would explain differences in antiviral activity against EMCV , HCV , IAV and ZIKV . Direct in vivo validation of our transcriptomic findings alone on the enhanced activity of the K154E variant would require liver biopsy samples from either HCV-infected Pygmies or chimpanzees combined with equivalent samples from infected humans encoding wt HsIFNλ4 . This was not possible since such tissue samples are not available from the Pygmy population infected with HCV and biopsies from acutely infected individuals are exceptionally rare . Moreover , chimpanzees are no longer used for experimental studies for ethical reasons . Therefore , we compared lists of reported differentially-expressed genes during acute HCV infection in humans and chimpanzees from the available literature . In the case of humans , there is only one report that analyses the transcriptional response in acute infection [33] . For chimpanzees , gene expression analysis is available from four independent studies [34–37] which include longitudinal data from serial biopsies . Therefore , all of the data was collated and we focused our comparisons on periods when human and chimpanzee biopsies were taken across the same time period after initial HCV infection ( between 8 and 20 weeks post infection ) . Comparative gene expression analysis revealed distinct host responses in humans and chimpanzees as well as overlapping differentially-regulated genes ( Fig 5A and S3 Data ) . In chimpanzees , the transcriptional profile contained significantly expressed genes that were type I/III IFN-regulated ISGs known to restrict HCV infection ( RSAD2 , IFI27 and IFIT1 ) [2] , as well as genes involved in antigen presentation and adaptive immunity ( HLA-DMA and PSMA6 ) . These genes were not significantly differentially expressed in humans , whose response was mainly directed towards up-regulation of pro-inflammatory genes ( for example , CXCL10 , CCL18 and CCL5 ) and metabolism genes ( AKR1B10 and HKDC1 ) ( Fig 5A and S3 Data ) . This was consistent with previous characterisation of the human acute response to HCV infection that failed to detect a major type I/III IFN signature but predominantly found a type II or IFN-gamma-mediated response [33] . From the available longitudinal data , the ‘chimpanzee-biased’ differentially-expressed genes were induced early in infection and remained significantly up-regulated during the acute phase following an early peak after infection ( S7A and S7B Fig ) . Differences were also reflected in pathway analysis in terms of the most significant pathways and their overall levels of significance ( S3 Data ) . For example , the ‘chemokine-mediated signalling pathway’ was upregulated in humans but not chimpanzees whereas the T cell receptor signalling pathway which was modulated in chimpanzees was not significantly altered in humans . Inspection of the raw data from humans indicated that many apparently ‘chimp-biased genes’ were expressed but did not reach significance in the original study . These genes were typically induced at a lower level in the human group when compared to averaged values for chimp studies across the similar time period ( Fig 5B ) . Furthermore , there was a greater induction of antiviral ISGs in chimpanzees during chronic infection in comparison to humans although to a less pronounced effect ( S7C Fig ) . From examining the in vivo biopsy data , we identified a group of 29 chimpanzee-biased genes in liver biopsies that were seemingly up-regulated during acute infection to a greater extent compared to humans . Comparing this set of genes to those from the RNA-Seq transcriptomic data obtained in vitro ( Fig 4 ) showed that the majority ( 17/29 genes ) of the chimpanzee-biased genes were induced by HsIFNλ4 stimulation , with approximately half ( 8 genes ) of those being significantly up-regulated to a great extent with K154E compared to wt , including MX1 , IFITM1 , IFIT1 , IFIT3 , TRIM22 and IFI44L ( Fig 5C ) . Thus , there are similarities between our in vitro analysis and published in vivo studies that would correlate with differences in IFNλ4 activity between humans and chimpanzees . Having established the greater antiviral potential for the E154 IFNλ4 variant and its apparent evolutionary relevance , we set out to determine the possible basis for its enhanced activity . No crystal structure for HsIFNλ4 is available but a homology model based on comparison with the IFNλ3 structure has been reported [24] . We expanded this predicted model based on both of the IFNλ1 and IFNλ3 crystal structures to explore the possible impact of K154E , P70S and L79F on IFNλ4 function ( Fig 6A and S8 Fig; [38 , 39] ) . As has been previously described , the sequences in helix F , which binds to IFNλR1 , are relatively well conserved [18 , 24] . The position equivalent to amino acid 154 in IFNλ4 is a glutamic acid in both IFNλ1 and IFNλ3 ( amino acid position 176 in IFNλ1 and 171 in IFNλ3 ) and its side chain faces inward towards the opposing IL10R2-binding helices C and D ( Fig 6A ) . The free carboxyl group of glutamic acid forms non-covalent intramolecular interactions with two non-linear segments on IFNλ1 and 3 ( IFNλ1 residue K64 , and in IFNλ3 K67 and T108 ) . In IFNλ4 , these E154-interacting positions are not conserved compared to IFNλ1/3 although homologous positions do exist with biochemically similar residues ( IFNλ4 R60 , and R98 that lies just upstream of the residue homologous to IFNλ3 T108 ) . To test whether the biochemical properties of glutamic acid at position 154 contribute to IFNλ4 activity , a panel of variants was constructed with biochemically distinct amino acids ( R154 , L154 , A154 , D154 and Q154 ) . Firstly , intracellular expression of each variant at position 154 was approximately equivalent ( Fig 6B ) . In signalling assays , the order of activity was E>Q/D>A>L>K>R ( S9A Fig ) . We found a similar pattern in the EMCV antiviral assays except that L154 had the least activity ( Fig 6C ) . We interpret these findings to conclude that E154 is biochemically the most favoured residue at this position with regards to antiviral potential , and that substitution of E154 to lysine results in the lowest potency for IFNλ4 activity . Interestingly , both Q154 and D154 had ‘intermediate’ activity compared to E154 and K154 , suggesting that side chain length and negative charge are important to maximise the activity of IFNλ4 . In a final series of experiments aimed at giving further insight into the mechanism of action of IFNλ4 K154E , we compared the relative activities and abundance of different IFNλ4 variants in cell lysates ( i . e . intracellular protein ) and supernatants ( i . e . extracellular protein ) . As wt HsIFNλ4 is poorly secreted into the supernatant from transfected cells in the absence of enrichment [24 , 40] , IFNλ4 in CM was immunoprecipitated using an anti-FLAG antibody . In antiviral assays , the activity of human , chimpanzee and macaque E154 variants from cell supernatants , IP fractions and lysates was greater than the corresponding K154 variants in agreement with our earlier results ( Fig 6D and Fig 2C–2E ) . Moreover , the D154 and R154 variants yielded patterns for cell lysates , cell supernatants and immunoprecipitated IFNλ4 protein such that D154 had intermediate activity between E154 and K154 while R154 had approximately equivalent activity to K154 ( S9B Fig ) . Thus , each variant displayed a similar pattern of activity irrespective of the source of IFNλ4 . From Western blot analysis , the E154 and K154 variants for each individual species were detected at similar levels in cell lysates ( S9C Fig ) . The HsIFNλ4 D154 and R154 variants were expressed to slightly higher and lower levels respectively compared to E154 and K154 from humans . Paradoxically , we did not find the same pattern in IFNλ4 abundance for immunoprecipitated protein derived from cell supernatants . Thus , we were able to detect greater amounts of the E154 variants for human , chimpanzee and macaque IFNλ4 compared to their K154 variants ( Fig 6E and S9D Fig ) . It was not possible to reliably detect macaque K154 or human R154 variants . HsIFNλ4 D154 had levels intermediate between the E154 and K154 variants . By Western blot and subsequent densitometry analysis , the relative abundance of IFNλ4 E154 and K154 variants in cell lysates for any species differed by 1 . 3 fold yet the approximate fold increase in antiviral activities were significantly greater and on average 16-fold . For the secreted IFNλ4 variants , we found that not only was there a higher abundance of E154 to K154 protein ( 9-fold ) , but activity was 41-fold higher for E154 than K154 variants , which results in a significant 3 to 4-fold rise in antiviral activity not explained by protein abundance ( S9E Fig ) . With the exception of macaque K154 , the FLAG antibody detected a putative breakdown product of about 11kDa in each of the samples , which we presume arose from cleavage by an unknown intracellular protease as it was also detected in cell lysates . The amount of this lower molecular weight product followed the same pattern as the full-length protein in that there was more with E154 than K154 thus cleavage does not explain differences in antiviral activity . Towards the end of the study , the split NanoLuc reporter system became available [41] , which comprised an 11 amino acid HiBiT tag that could replace the FLAG tag at the C-terminal end of the HsIFNλ4 variants and reconstitute luciferase activity ( Fig 6F ) . The advantage of this approach was that the relative secretion of each variant could be determined by comparing enzyme activity from cell lysates and culture media in a highly-quantitative manner . Moreover , the antiviral activity of HiBiT-tagged HsIFNλ4 variants could be compared to their respective FLAG-tagged versions . To demonstrate the capacity of the system to quantify secretion , we generated HiBiT-tagged versions of wt HsIFNλ3 and HsIFNλ4 with and without N-terminal signal sequences , which would abrogate secretion . Removing the signal sequences from either HsIFNλ protein reduced secretion by fivefold ( Fig 6F and 6G ) . The non-natural N61A variant that is not glycosylated was secreted ~2-fold less efficiently than wt HsIFNλ4 . In agreement with our data using FLAG-tagged HsIFNλ4 , the K154E variant of HsIFNλ4 was secreted about 3 times more efficiently than the wt protein ( Fig 6F and 6G ) . In a subsequent screen of all HsIFNλ4 natural variants with the HiBiT tag system , we also validated the data shown in Fig 1C and S2 Fig ( S10A–S10C Fig ) . Moreover , HiBiT-tagged HsIFNλ4 K154E was >10-fold more active than the wt form ( S11C Fig ) . Thus , both FLAG- and HiBiT-tagged forms of K154E gave higher antiviral activity and were secreted more efficiently than wt HsIFNλ4 . Lastly , to examine secretion in more detail , cells that had been transfected with HiBiT-tagged HsIFNλ3 and HsIFNλ4 variants were treated with Brefeldin A ( BFA ) and Monensin to block ER and Golgi transport respectively ( S11A–S11C Fig ) . Treatment with these inhibitors disrupted secretion of HsIFNλ3 by 12- ( BFA ) and 25-fold ( Monensin ) . For the HsIFNλ4 K154E variant , each of the inhibitors blocked secretion by about 10-12-fold . In the case of wt HsIFNλ4 , BFA inhibited secretion to a greater extent ( ~20-fold ) than for either HsIFNλ3 or HsIFNλ4 K154E whereas Monensin decreased secretion to a similar extent for all three proteins . The explanation for the slightly greater inhibition of HsIFNλ4 by BFA requires further investigation . Overall , our data conclusively demonstrate that glutamic acid at position 154 promotes greater antiviral potential by enhancing both IFNλ4 secretion from cells and its intrinsic potency .
Our analysis suggests that the Homo IFNλ4 orthologue acquired the E154K substitution , yielding a less active protein , after the genetic divergence of the hominid Homo and Pan ancestral lineages ( estimated to be at most 6 million years ago in Africa [42] ) but before human/Neanderthal divergence ( ~370 , 000 years ago , [43] ) . Subsequently , the IFNL4 gene acquired two further variants , the P70S and TT alleles that are now common in the human population [18] . Acquisition of each of these alleles either further reduced ( P70S ) or abolished ( TT ) IFNλ4 activity . Other rare variants have arisen in humans with little impact on HsIFNλ4 antiviral potential based on our in vitro assays , except for variants L79F and K154E , which lower and increase activity respectively . To us , the most intriguing of these variants is K154E , which was found in a high proportion of rainforest ‘Pygmy’ hunter-gatherers from west central Africa [28] but was rarely present in the African population . Since this variant was not present in the genetic data for San and Archaic Neanderthal and Denisovan human lineages , we speculate that these populations likely reacquired K154E following divergence of chimpanzees and humans . However , with the ever-increasing availability of genetic data from ancient and extant human populations , it may be possible to identify other populations carrying the E154 variant; in particular more details on the demographics of the 29 African individuals in the gnomAD database [29] may address whether there are additional specific groups who carry this more active variant and how it arose . The factors responsible for divergent functional evolution of the IFNL4 gene within and between species are not known . It has been demonstrated that loss of IFNL4 has evolved under positive selection in some human populations thus we speculate that differences in exposure to certain pathogenic microbes has driven evolution of the E154 variant . On the one hand , type III IFN signalling enhances disease and impedes bacterial clearance in mouse models of bacterial pneumonia [44] . This suggests that IFNλ4 with a lower activity could be beneficial during non-viral infections although a link between IFNL4 genotype and bacterial infection in humans has not yet been made . Conversely , we postulate that the presence of more active IFNλ4 exemplified by E154 in Pygmies and chimpanzees may be linked to increased exposure to zoonotic viral infections in the Congo rainforest , such as pathogenic Filovirus infections [45] . For decades , experimental studies in chimpanzees have provided unique insight into HCV infection [46] but they do not present with identical clinical outcomes as human subjects . For example , chimpanzees have been reported to clear HCV infection more efficiently than humans [47] , rarely develop hepatic diseases similar to humans [48] , and are refractory to IFNα therapy [49] . Moreover , HCV evolves more slowly in infected chimpanzees , possibly due to a stronger immune pressure that reduces replication compared to humans [50] . In humans , IFNL4 genetic variants are associated with , and thought to regulate , each of these characteristics [18 , 51 , 52] . Although myriad factors could explain these phenotypic differences , including differences in antagonism of the immune response by HCV or changes in IFNL transcription for example , we propose that the greater antiviral activity of PtIFNλ4 compared to HsIFNλ4 contributes to the distinct responses to HCV infection in the two species . Acute HCV infection in human cells in vitro and chimpanzees in vivo selectively stimulates type III over type I IFNs , which are effective at signalling in hepatocytes [53 , 54] . Notably , there is no apparent type I/III IFN gene expression signature in liver biopsies from humans with acute HCV infection [33] . Differences in IFN signalling during HCV infection have been postulated to explain the ability to control HCV infection in cell culture or following IFN-based therapy in humans [55 , 56] . Our comparative meta-analysis of the available literature revealed an apparent enhanced expression of ISGs with anti-HCV activity as well as genes involved in antigen presentation and T cell mediated immunity in chimpanzees compared to humans . Our analysis cannot be considered conclusive given the disparate nature of the retrospective studies used in our comparisons and therefore , there are a number of caveats ( e . g . the expression levels of the IFNL genes; the relative expression of receptors in the two species ) . Nonetheless , we speculate that enhanced expression of ISGs in chimpanzee liver due to higher IFNλ4 activity could lead to greater control of viral infection by both inducing antiviral genes and by coordinating a more effective adaptive T cell response , which is critical for clearance and pathogenesis during HCV infection [57] . Based on the above speculation , we would predict that the response to HCV infection in chimpanzees may be similar in Pygmies with the K154E variant . A recent study in Pygmies from Cameroon , including the Baka and Bakola groups , showed low seroprevalence of 0 . 6% and no evidence of chronic HCV infection [58] . Interestingly , infection in non-Pygmy groups in Cameroon has a seroprevalence of ~17% [59] . One explanation for this difference could be higher IFNλ4 activity in populations with the K154E variant , which may enhance HCV clearance . In our study of three primate orthologues , glutamic acid at position 154 in IFNλ4 provided greater antiviral activity and enhanced its ability to induce antiviral gene expression . A functional comparison of human and chimpanzee IFNλ4 orthologues has been explored previously but no significant differences in signalling activity were observed [31] . There are substantial differences in the methodologies used in our study and that of Paquin et al . , which could explain our ability to detect divergent activity , for example the size of the tag attached to IFNλ4 and dose of protein used in assays . Our observed functional differences between E154 and K154 did not correlate with levels of intracellular accumulation or glycosylation . However , we did find that the more active E154 variants for human , chimpanzee and macaque IFNλ4 were detected at higher levels in the immunoprecipitated fractions from cell supernatant ( CM ) compared to the K154 variants; the D154 and R154 variants also were detected at a lower level than E154 . Interestingly , endogenous , wt HsIFNλ4 with K154 is not secreted to detectable levels compared to wt HsIFNλ3 [24 , 40 , 60] . Detection of low levels of secreted wt HsIFNλ4 by Western blot analysis requires exogenous expression of the protein and precipitation of material in the cell supernatant . Poor release of HsIFNλ4 was not due to differences in the signal peptide of HsIFNλ4 or HsIFNλ3 but secretion could be ablated if the single N-linked glycosylation site was mutated [24] . Our data indicate that position 154 regulates release of intracellular IFNλ4 . Moreover , IFNλ4s with E154 are more potent than those with K154 when correcting for the difference in amounts of protein . This increase in potency for E154 was detected in both IP protein and lysates as well as HsIFNλ4 with different C-terminal tags . Moreover , we observed that the difference between E and K is greater in the cell released fraction than the cell lysate . The reason for this discrepancy could be explained by a number of factors that are outside of the scope of this study . Based on our modelling of the IFNλ4 structure and further mutational analysis , glutamic acid is apparently the optimal residue at position 154 . At the biochemical level , glutamic acid has the capacity to form electrostatic bonds with charged residues in the IFNλ4 protein and moreover it possesses a side chain which could contribute greater flexibility for such interactions . Notably , replacing glutamic acid with either aspartic acid or glutamine gave higher IFNλ4 activity than either non-polar or positively-charged residues . These potential E154-mediated interactions occur in the region of the protein devoid of cysteine-bonds likely making the interaction between helix F ( IFNλR1-binding ) and the loop connecting helices C and D ( IL-10-R2-binding ) particularly flexible . The putative greater structural stability facilitated by E154 may inherently increase the structural integrity of IFNλ4 making this variant more competent for secretion and more potent in signalling through the IFNλR1-IL10R2 surface receptor complex . Increased binding to IFN receptor complexes has been shown to enhance signalling by type I IFNs [61 , 62] . Further biophysical studies using highly-purified recombinant protein measuring affinity and avidity of HsIFNλ4 wt and K154E for each receptor molecule [as in 27 , 39] combined with studies on the mechanism of IFNλ4 release will help address these hypotheses . To conclude , our study further supports a significant and non-redundant role for IFNλ4 in controlling the host response to viral infections yet one whose activity has been repeatedly attenuated during human evolution , commencing with E154K . Taken together , this provides the foundation for more detailed investigation into the mechanism of action of IFNλ4 and its overall contribution to host immunity in regulating pathogen infection .
All available human IFNL4 genetic variation along with associated frequency and ethnicity data for the human population were collected from the 1000 Genomes Database available at the time of study ( June 2016 ) [23] ( http://browser . 1000genomes . org/index . html ) . The reference sequence for the human genome contains the frameshift ‘TT’ allele and so potential effects of variants on the HsIFNλ4 predicted amino acid sequence were identified manually following correction for the frameshift mutation ( TT to ΔG ) . The effect of all single nucleotide polymorphisms ( SNPs ) on the open reading frame ( ORF ) was thus assessed and re-annotated as synonymous or non-synonymous changes resulting in the selection of coding variants reported here . Inspection of whole genome sequence data from African hunter-gatherers was carried out using previously published datasets [28] . We remapped the raw reads of six San individuals ( four Juǀʼhoan and two ‡Khomani San ) in the Simons Genomic Diversity Project [30] to the human reference genome ( hg19 ) and conducted variant calling using the haplotype caller module in GATK ( v3 ) . Two Juǀʼhoan individuals were heterozygous at rs368234815 ( TT/ΔG genotype , S2 Data ) . The genotypes of rs368234815 in Neanderthal and Denisovan were extracted from VCF files that were downloaded from http://cdna . eva . mpg . de/denisova/VCF/hg19_1000g/ and http://cdna . eva . mpg . de/neandertal/altai/AltaiNeandertal/VCF/ . Neanderthal and Denisovan genetic data contained only ΔG alleles ( S2 Data ) . Independent validation of the SNP variant that gives rise to the K154E substitution in HsIFNλ4 was obtained from gnomAD ( http://grch37 . ensembl . org/Homo_sapiens/Variation/Population ? db=core;r=19:39737353-39738353;v=rs377155886;vdb=variation;vf=58909380 and https://www . ncbi . nlm . nih . gov/projects/SNP/snp_ref . cgi ? rs=rs377155886 ) . Amino acid sequences for mammalian IFNλ genes were obtained from NCBI following protein BLAST of the wt HsIFNλ4 polypeptide sequence . Multiple alignments of IFNλ amino acid sequences were performed by MUSCLE using MEGA7 . Accession numbers of specific IFNλs used in the experimental section of this study were as follows: HsIFNλ1: Q8IU54; HsIFNλ3 , Q8IZI9 . 2; and for IFNλ4: Homo sapiens AFQ38559 . 1; Pan troglodytes AFY99109 . 1; Macaca mullata XP_014979310 . 1; Pongo abelii ( orangutan ) XP_009230852 . 1 , Bos taurus ( cow ) XP_005219183 . 1 , Felis catus ( cat ) XP_011288250 . 1 . To validate the L79F variant , genomic DNA isolated from Epstein-Barr virus-immortalised B-cell lymphoblastoid cell lines from individuals ( HG03095 and NA19658 ) identified through the 1000 Genomes Project and International HAPMAP project as probands with the L79F substitution ( rs564293856 G>A SNP ) were obtained from the Coriell Institute for Medical Research . PCR was used to generate an amplicon of ~300 base pairs corresponding to the region that includes rs564293856; amplicons were column-purified ( Qiagen PCR-Clean Up kit , Qiagen ) . Internal forward and reverse primers within the amplicon were employed to determine the sequence of the region across SNP rs564293856 by Sanger sequencing ( S13A and S13B Fig ) . The resulting chromatograms were inspected manually and the sequences at rs564293856 were called when identified using primers from both directions . HG03095 was confirmed as heterozygous for rs564293856 as predicted although we failed to detect the variant in NA19658; HG03095 and NA19658 were individuals who originated from Africa and America respectively . The homology model of the HsIFNλ4 structure used in Fig 6 and S8 Fig was generated using the RaptorX online server ( http://raptorx . uchicago . edu ) . The resultant HsIFNλ4 structural model was then structurally aligned with both HsIFNλ1 ( PDB 3OG6 ) [38] and HsIFNλ3 ( PDB 5T5W ) [39] . Visualization , structural alignments , and figures were generated in Pymol ( The PyMOL Molecular Graphics System , Version 1 . 8 ) . DNA sequences encoding the ORFs of HsIFNλ4 , PtIFNλ4 and MmIFNλ4 ( based on accession numbers above ) were synthesized commercially with a carboxy-terminal DYKDDDDK/FLAG tag using GeneStrings or Gene Synthesis technology ( GeneArt ) . As a positive control for functional assays , the HsIFNλ3 ORF was codon optimised ( human ) to ensure robust expression and antiviral activity , and is termed ‘HsIFNλ3op’; the exception to this were the HiBiT-tagged versions of HsIFNλ3 which were not codon optimised . All IFNλ4 coding region sequences were retained as the original nucleotide sequence without optimisation . Synthesized DNA was cloned into the pCI mammalian expression vectors ( Promega ) using standard molecular biology techniques . At each cloning step , the complete ORF was sequenced to ensure no spurious mutations had occurred during plasmid generation and manipulation . Single amino acid changes were incorporated using standard site-directed mutagenesis protocols ( QuickChange site-directed mutagenesis kit [Agilent] , or using overlapping oligonucleotides and Phusion PCR ) . A549 ( human lung adenocarcinoma ) , U2OS ( human osteosarcoma ) , MDCK ( Madin-Darby canine kidney ) and HepaRG ( non-differentiated human hepatic progenitor ) cells were obtained from Chris Boutell; Huh7 ( human hepatoma ) cells were obtained from Charles Rice , Rockefeller University , USA; HEK293T ( human embryonic kidney ) cells were obtained from Sam Wilson; Vero ( African Green Monkey kidney ) cells were obtained from Arvind Patel . Cells were grown in DMEM growth media supplemented with 10% FBS and 1% penicillin-streptomycin except for HepaRG and genome-edited derivatives ( generated during this study ) ; these cells were cultured in William’s E medium supplemented with 10% of FBS , 1% penicillin-streptomycin , hydrocortisone hemisuccinate ( 50 μM ) and human insulin ( 4 μg/mL ) . Huh7 cells harbouring the HCV JFH-1 sub-genomic replicon [Tri-JFH1; 32] were cultured in the presence of 500 μg/ml G418 . All cells were grown at 37°C with 5% CO2 . Cell lines were routinely tested for mycoplasma and no contamination was detected . To inhibit secretion of HsIFNλs , cells were treated with BFA and Monensin at a final concentration of 5 μg/ml . BFA was purchased in DMSO-dissolved form ( 10 mg/ml ) and Monensin sodium salt was dissolved in methanol to the same concentration as BFA; both inhibitors were purchased from Sigma-Aldrich ( UK ) . Plasmid DNA generated from bacterial cultures ( GeneJET plasmid midiprep kit , ThermoScientific ) was introduced into cells by lipid-based transfection using Lipofectamine 2000 or Lipofectamine 3000 ( ThermoFisher ) following manufacturer’s instructions . To produce IFN-containing conditioned media ( CM ) or measure protein production , HEK293T ‘producer’ cells were grown to near-confluency in 12 ( ~4 x 105 cells per well ) or 6-well ( ~1 . 2 x 106 cells per well ) plates and transfected with plasmids ( 2 μg ) in OptiMEM ( 1–2 ml ) overnight . At approximately 16 hours ( hrs ) post transfection ( hpt ) , OptiMEM was removed and replaced with complete growth media ( 1–2 ml ) . CM containing the extracellular IFNλs was harvested at 48 hpt and stored at -20°C before use . Although antiviral activity was observed at 16 hpt , we chose 48 hpt to harvest CM to ensure robust production and secretion of each IFNλ . Intracellular IFNλs also were harvested from transfected cells at 48 hpt . CM was removed and replaced with fresh DMEM 10% FCS ( 2 ml ) and then frozen at -70°C . To prepare cell lysates with IFNλ activity , plates were thawed and the cell monolayer was scraped into the media and clarified by centrifugation ( 5 minutes [mins] x 300 g ) before use . CM or lysates were diluted in the respective growth medium for each cell line before functional testing as described in the text . Two-fold serial dilutions of CM were used in titration of anti-EMCV activity and ability to induce EGFP in an IFN-reporter cell line . Single CM dilutions of 1:4 ( HepaRG and A549 ) or 1:3 ( Huh7 ) were chosen based on initial experiments for gene expression and non-EMCV antiviral activity measurements to allow measurement of both high and low activity variants . Immunoprecipitation of extracellular FLAG-tagged IFNλ4 present in the supernatant of transfected cells was carried out using an anti-FLAG M2 antibody-bound gel as described by the manufacturer’s guidelines ( Sigma Aldrich ) . Immunoprecipitated IFNs were used in activity assays and for Western blot analysis . Briefly , resin with anti-FLAG antibody ( 40 μl ) and supernatants ( 1 ml ) were thawed on ice . Beads were washed repeatedly in ice cold buffer before being incubated with IFNs in CM for 2 hrs at 4°C while rocking . Bead-bound IFN was pelleted by centrifugation , washed and eluted with FLAG peptide ( 100 μl ) . Positive and negative controls were ‘BAP-FLAG’ and buffer only , respectively . Centrifugation conditions were 8 , 200 x g for 30 sec at 4°C . One quarter ( 25 μl ) of total immunoprecipitated protein was loaded onto gels for Western blot analysis . The FLAG tag at the C-terminal ends of IFNλs was replaced with the 11 amino acid HiBiT tag without a linker ( amino acid sequence: N-VSGWRLFKKIS-C [41] ) using PCR and subsequent cloning into the pC1 vector . To block secretion from cells , the first 26 amino acids of HsIFNλ3 and 23 amino acids of HsIFNλ4 , corresponding to their respective N-terminal signal peptide sequences , were removed using a similar approach . The sequences of all plasmid constructs were verified . Luciferase activity was measured following transfection of plasmids using the Nano-Glo HiBiT Lytic Detection System and its Extracellular Detection System counterpart following manufacturers protocols ( Promega , UK ) . Briefly , for measuring intracellular enzyme activity , supernatant was removed from transfected cells and lytic buffer was added directly to the cell monolayer before incubation for 10 minutes; all lysed material was transferred to a 1 . 5 ml Eppendorf tube and luciferase activity was measured in a luminometer . For determining extracellular activity , 100 μl of the CM was removed from the culture medium and mixed 1:1 with both the LgBiT and substrate for 10 minutes prior to measuring luciferase activity; the luciferase value for secreted HiBiT-tagged HsIFNλ was then calculated based on the total volume of culture medium . Ratios to determine the extent of secretion of each IFNλ were generated based on total intracellular and extracellular luciferase values . Total cellular RNA was isolated by column-based guanidine thiocyanate extraction using RNeasy Plus Mini kit ( genomic DNA removal ‘plus’ kit , Qiagen ) according to the supplier’s protocol . cDNA was synthesised by reverse transcribing RNA ( 1 μg ) using random primers and the AccuScript High Fidelity Reverse Transcriptase kit ( Agilent Technologies ) ; the recommended protocol was followed . Relative expression of mRNA was quantified by qPCR ( 7500 Real-Time PCR System , Applied Biosystems ) of amplified cDNA . Probes for ISG15 ( Hs01921425 ) , Mx1 ( Hs00895608 ) and the control GAPDH ( 402869 ) were used with TaqMan Fast Universal PCR Master Mix ( Applied Biosystems ) . The results were normalised to GAPDH and presented in 2−ΔΔCt values relative to controls as described in the text . HCV genomic RNA was quantified by RT-qPCR as described previously [32] . IFN-competent cells ( A549 ) were stimulated with IFN CM ( 1:4 dilution ) in 6-well plates ( ~1 . 2 x 106 cells ) for 24 hrs and global gene expression was assessed by RNA-Seq , using three biological replicates per condition . Sample RNA concentration was measured with a Qubit Fluorometer ( Life Technologies ) and RNA integrity ( RIN ) was determined using an Agilent 4200 TapeStation . All samples had a RIN value of 9 or above . 1 . 5 μg of total RNA from each sample was prepared for sequencing using an Illumina TruSeq Stranded mRNA HT kit according to the manufacturer's instructions . Briefly , polyadenylated RNA molecules were captured , followed by fragmentation . RNA fragments were reverse transcribed and converted to dsDNA , end-repaired , A-tailed , ligated to indexed adaptors and amplified by PCR . Libraries were pooled in equimolar concentrations and sequenced in an Illumina NextSeq 500 sequencer using a high output cartridge , generating approximately 25 million reads per sample , with a read length of 75 bp . 96 . 3% of the reads had a Q score of 30 or above . Data was de-multiplexed and fastq files were generated on a bio-linux server using bcl2fastq version v2 . 16 . RNA‐Seq analysis was performed using the Tuxedo protocol [63] . Briefly , reads from 3 replicates per condition were aligned and junctions mapped against the human reference transcriptome hg38 using Tophat2 with the default settings except library type . Transcriptome assembly was performed using Cufflinks supplying annotations from the reference genome hg38 and the differential gene expression was calculated using Cuffdiff . Differential gene expression was considered significant when the observed fold change was ≥2 . 0 and FDR/q‐value was <0 . 05 between comparisons . Pathway analysis was carried out using Ingenuity Pathway Analysis [IPA] ( Ingenuity Systems , Redwood City , CA , USA ) . Cell growth media was removed and monolayers were rinsed once with approximately 0 . 5 ml PBS before lysis using RIPA buffer ( ThermoFisher ) containing protease inhibitor cocktail ( 1x Halt Protease inhibitor cocktail , ThermoFisher , or cOmplete , Mini , EDTA-free Protease Inhibitor Cocktail , Sigma Aldrich ) for 10 mins at 4°C before being frozen at -20°C overnight . Lysates were collected into a 1 . 5 ml sample tube and clarified by centrifugation ( 12 , 000 x g for 15 mins ) . Samples ( 10 μl ) from the soluble fraction were heated to 90°C for 10 mins with 100 mM dithiothreitol ( DTT ) -containing reducing lane marker at 90°C for 10 mins . Samples were run on home-made 12% SDS-PAGE gels alongside molecular weight markers ( Pierce Lane marker , Thermofisher ) before wet-transfer to nitrocellulose membrane . Membranes were blocked using a solution of 50% PBS and 50% FBS for 1 hr at room temperature and then incubated overnight at 4°C with primary antibodies in 50% PBS , 50% FBS and 0 . 1% TWEEN 20 . Secondary antibodies were incubated in 50% PBS , 50% FBS and 0 . 1% TWEEN 20 for 1 hr at room temperature . Membranes were washed four times ( 5 mins each ) following each antibody incubation with PBS containing 0 . 1% TWEEN 20 . After the 4th wash following incubation with the secondary antibody , the membrane was washed once more in PBS ( 5 mins ) and kept in ddH20 until imaging . Primary antibodies to the FLAG tag ( 1:1000 ) ( rabbit , lot . 064M4757V , LiCor ) and α-tubulin ( 1:10000 ) ( mouse , lot . GR252006-1 , LiCor ) were used along with infra-red secondary antibodies ( LI-COR ) to anti-rabbit ( donkey [1:10 , 000] , 926–68073 ) and anti-mouse ( donkey [1:10 , 000] , C50422-05 ) to allow protein visualisation . Pre-stained , Pageruler Plus marker was used to determine molecular weights ( ThermoFisher ) . Membranes were visualised using the LI-COR system on an Odyssey CLX and the relative expression level of proteins determined using LI-COR software ( Image Studio ) . An IFN reporter HepaRG cell line was generated to measure IFN activity by introducing the EGFP ORF fused to the ISG15 ORF separated by ribosome skipping sites by CRISPR-Cas9 genome editing . We chose to introduce EGFP in-frame to the N-terminus of the ISG15 ORF since it is a robustly-induced ISG . We also introduced the blasticidin resistance gene ( BSD ) for selection purposes . BSD , EGFP and ISG15 were separated using ribosome skipping 2A sequences ( P2A and T2A ) . Transgene DNA was flanked by homology arms with reference to the predicted target site . Homology donor plasmids for CRISPR-Cas9 knock-in were generated through a series of overlapping PCR amplifications using Phusion DNA polymerase followed by sub-cloning into pJET plasmid . Plasmids for CRISPR-Cas9 genome editing ( wt SpCas9 ) were generated using established protocols [64] in order to create plasmids that would direct genome editing at the 5’ terminus of the HsISG15 ORF ( exon 2 ) . pSpCas9 ( BB ) -2A-Puro ( PX459 ) V2 . 0 was a gift from Feng Zhang ( Addgene plasmid # 62988 ) . All sequences are available by request . HepaRG cells grown in 6 well dishes were co-transfected with CRISPR-Cas9 editing plasmids targeting the beginning of the ISG15 ORF in exon 2 ( exon 1 contains only the ATG of the ORF ) , and homology donor plasmids described above ( 1 μg each ) using Lipofectamine 2000 and the protocol described above . Transfected cells were selected using puromycin ( Life Technologies ) ( 1 μg/ml ) and blasticidin ( Invivogen ) ( 10 μg/ml ) until non-transfected cells were no longer viable . Selected cells were cloned by single cell dilution , expanded and tested for EGFP induction following IFN stimulation . Positioning of the introduced transgene was assessed by PCR amplification on isolated genomic DNA from individual clones ( S5C Fig ) . Primers were designed to include one primer internal to the transgene and another external to the transgene and found in the target loci ( sequences available on request ) . For use as an effective IFN reporter cell line , cells had to demonstrate robust induction of EGFP expression following stimulation with IFN and evidence of specific introduction of the transgene . This study uses clone ‘G8’ of HepaRG . EGFP-BSD-ISG15 cells . We have not tested whether there is a single transgene integration site or multiple ones nor confirmed that the EGFP produced following stimulation by IFNs results from the expression of the specifically-introduced transgene rather than off-target integration , which is theoretically possible . We do not predict this would affect the cells’ ability to act as a reporter cell line . For use in IFN reporter assays , stimulated cells ( in 96 well plates stimulated for 24 hrs; ~5 x 104 cells per well ) were washed , trypsinised and fixed in formalin ( 1% in PBS ) at room temperature for 10 mins in the dark before being transferred to a round-bottomed plate and stored at 4°C in the dark until measurement of EGFP fluorescence . Non-stimulated cells were used as negative controls and the change in % EGFP-positive cells was assessed by flow cytometry using a Guava easyCyte HT ( Merck Millipore ) . For fluorescence microscopy , EGFP induction was measured by indirect immunofluorescence of stimulated cells that were fixed and permeabilised on coverslips prior to antibody binding . An EGFP primary antibody ( 1:1000 , rabbit ab290 Abcam ) was used followed by a fluorescent anti-secondary antibody ( 1:500 , Goat anti Rabbit Alexa-Fluor , Thermo Fisher , 568nm ) . Samples were counter-stained using DAPI and visualised with a confocal laser-scanning microscope ( Zeiss LSM 710 ) under identical conditions . Antiviral activity of IFNλs was determined using encephalomyocarditis virus ( EMCV ) , influenza A virus ( IAV; A/WSN/1933 ( H1N1 ) ) , Zika virus ( ZIKV; Brazilian strain PE243 ) [65] and HCV ( HCVcc chimeric clone Jc1 ) [66] . EMCV was grown on Vero cells followed by titration on U2OS cells by plaque assay . IAV stocks were generated on MDCK cells and titrated by plaque assay on MDCK cells with protease ( TPCK-treated trypsin , Sigma Aldrich ) . ZIKV was titrated on Vero cells by plaque assay . For all plaque assays , cells were grown in 12 or 6-well plates to ~90% confluency before inoculation with serial 10-fold dilutions of virus stocks in serum-free Optimem . Inoculum remained on the cells for 2 hrs before removal and the monolayers were rinsed with PBS ( 1x ) and semi-solid Avicell overlay ( Sigma Aldrich ) was added . For EMCV and IAV , 1 . 2% Avicell was used , diluted in 1x DMEM 10% FCS , 1% penicillin-streptomycin . For IAV titration , TPCK-treated trypsin was added ( 1 μg/ml ) . For ZIKV plaque assay , 2x MEM was used instead of 1x DMEM . HCVcc Jc1 was generated as described previously by electroporation of in vitro transcribed viral RNA into Huh7 cells and harvested at 72 hrs post electroporation . After filtration of the supernatant , HCVcc Jc1 stocks were titrated by TCID50 on Huh7 cells and stored at 4°C before use . HCVcc Jc1 TCID50 assays were performed using anti-NS5A antibody [67] . Infected cells at 72 hrs post infection were fixed and permeabilised with ice-cold methanol . Cells were rinsed in PBS , blocked with 3% FCS in PBS at room temperature and incubated overnight with mouse monoclonal anti-NS5A antibody ( 9E10 ) at 4°C . After removal of the antisera , cells were rinsed 3 times with PBS containing 0 . 1% TWEEN 20 , and then incubated in the dark at room temperature for 1 hr with secondary antibody [Alexa-fluor 488nm anti-mouse ( donkey ) ] . Cells were finally washed with PBS containing 0 . 1% TWEEN 20 and NS5A-expressing cells were visualized with a fluorescent microscope . Cells stimulated with IFNλs were infected with viruses at the following multiplicities of infection ( MOI ) : EMCV ( MOI = 0 . 3; added directly to the media ) ; IAV ( MOI = 0 . 01 ) ; ZIKV ( MOI = 0 . 01 ) ; HCVcc ( MOI = 0 . 05 ) . For IAV , ZIKV and HCVcc , the inoculum was incubated with cells for at 2 ( IAV/ZIKV ) or 3 hrs ( HCVcc ) in 0 . 5–1 . 0 ml serum-free Opti-MEM/DMEM at 37°C before removal . Cells were rinsed with PBS and then incubated with fresh growth media for the allotted time ( 24 hrs for EMCV , 48 hrs for IAV and 72 hrs for ZIKV and HCVcc ) . At the times stated for individual experiments , infected-cell supernatants were harvested and infectivity was titrated by plaque assay . IAV , ZIKV and HCVcc antiviral assays were all carried out in 12 well plates except for measurement of HCVcc infectivity by indirect immunofluorescence , which was measured in a 96 well plate . In the case of EMCV , a cytopathic effect ( CPE ) protection assay was employed to assess infectivity ( 26 ) . Here , HepaRG cells were plated in a 96-well plates ( ~5 x 104 cells per well ) and , when confluent , were incubated with two-fold serial dilutions of CM or lysate for 24 hrs before the addition of EMCV . At 24 hrs post infection with EMCV , media was removed , cell monolayers were rinsed in PBS and stained using crystal violet ( 1% in 20% ethanol in H20 ) for 10 mins . Crystal violet stain was then removed and stained plates were washed in water . The dilution of ~50% inhibition of EMCV-induced CPE was marked visually and the difference determined relative to wt HsIFNλ4 . Luciferase-expressing MLV pseudoparticles containing the E1 and E2 glycoproteins from JFH1 HCV strain were generated as described [68] along with their corresponding JFH1 E1-E2 deficient controls ( particles generated only with MLV core ) and used to challenge IFNλ-stimulated Huh7 cells . Huh7 cells grown in 96-well plates overnight ( seeded at 4 x 103 cells per well ) were stimulated with IFNλs for 24 hrs and transduced with HCVpp . 72 hrs later , cell lysates were harvested and luciferase activity was measured ( Luciferase assay system , Promega ) on a plate reading luminometer . For HCV RNA replication assays , RNA was transcribed in vitro from a sub-genomic replicon ( HCV-SGR ) expressing GLuc ( wild-type and non-replicating GND ) [69] . In vitro transcribed RNA ( 200 ng ) was transfected using PEI ( 1:1 ) into monolayers of Huh7 cells in 96-well plates overnight ( seeded at 4 x 103 cells per well ) that had been stimulated with IFNλs ( 24 hrs ) . At the specified time points , total supernatants ( containing the secreted GLuc ) from treated Huh7 cells were collected and replaced with fresh growth media . 20μl ( ~10% of total volume ) was used to measure luciferase activity and mixed with GLuc substrate ( 1x ) ( 50 μl ) and luminescence ( as relative light units , RLUs ) was determined using a luminometer ( Promega GloMax ) . Pierce Gaussia Luciferase Flash Assay Kit ( ThermoFisher ) was used and the manufacturer’s instructions were followed . Previously published datasets of intrahepatic differentially-expressed genes from liver biopsies were used to compare human and chimpanzee transcriptomic responses to early HCV infection . At first , we used reported lists of differentially-expressed genes between humans and chimpanzees but further validated observations with raw data from human studies . Studies focusing on acute HCV infection ( 0 to 26 weeks ) in humans and chimpanzees were acquired through manual literature search using Pubmed and gene lists were compiled . For chimpanzees , data was acquired from 4 studies [34–37] and one report was employed for human data [33] . The study by Dill et al . comprised single biopsy samples from each of six individuals , while in toto the chimpanzee studies combined data from ten animals with multiple , serial biopsies . All studies were carried out using similar Affymetrix microarray platforms except Nanda et al . who used IMAGE clone deposited arrays . Although similar microarrays measured different numbers of genes we focused on ‘core’ shared genes from chimpanzee studies . Humans were infected with HCV genotype ( gt ) 1 ( n = 2 ) , gt3 ( n = 3 ) and gt4 ( n = 1 ) while chimpanzees were experimentally infected with HCV gt1a ( n = 6 ) , gt1b ( n = 3 ) and gt2a ( n = 1 ) . The human dataset included individuals with IL28B rs12979860 genotypes T/T , C/T and C/C but no association between IL28B genotype and gene expression was noted ( 33 ) . Gene names and fold-changes were manually converted to a single format ( fold change rather than log2 fold change for example ) to allow comparative analysis . Human biopsies were taken between two and five months after presumed infection following known needle-stick exposure , and serial chimpanzee biopsies were taken at different time points from between one week and one year after HCV infection . For comparative purposes , differentially-expressed genes in chimpanzees were included if they were detected during a time period overlapping with the human data . We identified a ‘core’ set of chimpanzee differentially-expressed genes ( independently characterized in at least two studies ) and compared them to the single human transcriptome study data at equivalent time points ( between 8 and 20 weeks post-infection ) . This approach generated a set of core chimpanzee genes ( genes found differentially-expressed in at least 2 studies , >2 fold change compared to controls and during the time frame compared to humans ) for comparison with the human data . This is reflected in the ten-fold higher numbers of differentially-regulated genes found in the one human study compared to the ‘core’ ( reduced ) set assembled from four chimpanzee studies . To validate these findings , we used three studies of chronic infection for which data were available [56 , 70 , 71] , two from humans employing RNA-Seq and one from chimpanzees using microarray measurement . Gene lists were extracted and a core human list was produced and compared to that from chimpanzees . For shared genes , the fold change values were compared for humans and chimpanzees . The ratio of chimpanzee induction to human induction was calculated . These gene sets were compared to determine their degree of species-specificity or species-similarity using Venn diagram analysis ( http://bioinfogp . cnb . csic . es/tools/venny/ ) . The gene lists for humans and core genes for chimpanzees are shown in the S1 Data . For the chimpanzee-biased genes , mean expression values were determined at each time point from individual animals . For non-transcriptomic analysis ( transcriptomic analysis is outlined above ) , Graphpad Prism was used for statistical testing , which included Students’ T test and ANOVA and post-hoc tests ( Dunnett’s test ) where appropriate as described in figure legends . **** , p = <0 . 0001; *** , p = <0 . 001; ** , p = <0 . 01; * , p = <0 . 05 , are used throughout to denote statistical significance . Unless explicitly stated , in our analysis , the term biological replicate refers to measurements made from different wells during an experiment on the same day while independent experiment refers to measurements made from assays carried out on different days . Ethical approval was not required for analysis of human or animal samples as no such material was used in this study . All human and animal genomic and transcriptomic datasets were obtained from either publicly-available data or published data . We note that all data collected in the original studies was subject to ethical approval , details of which are available in the original study manuscripts . | Natural genetic variation and its influence on the outcome of viral infection is a topical area given the wealth of genetic data now available . However , understanding how clinical phenotype is affected by genetic variation at the molecular level is often lacking yet critical for any insight into immunity and disease . It is known that variants in the antiviral ‘interferon lambda 4’ ( IFNL4 ) gene significantly influence outcome of hepatitis C virus ( HCV ) infection in humans . Counter-intuitively , those producing IFNL4 have greater risk of establishing chronic HCV infection , compared to individuals with an inactive variant , although the underlying mechanisms remain poorly understood . From a comprehensive screen of all natural human variants , we show that the most common form of IFNλ4 is less able to protect human cells from pathogenic virus infection than the equivalent protein from our closest living relative the chimpanzee . This is as a result of a single amino acid substitution that impedes its release from cells and reduces antiviral gene expression . Our observed differences in activity correlated with divergent host responses in HCV-infected livers from humans and chimpanzees . We suggest that human IFNL4 evolution places humans at a disadvantage when infected with pathogens such as HCV . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"luciferase",
"pathology",
"and",
"laboratory",
"medicine",
"hepacivirus",
"pathogens",
"enzymes",
"microbiology",
"vertebrates",
"enzymology",
"orthomyxoviruses",
"animals",
"mammals",
"primates",
"viruses",
"physiological",
"proces... | 2018 | A polymorphic residue that attenuates the antiviral potential of interferon lambda 4 in hominid lineages |
Distinct Trypanosoma cruzi genotypes have been considered relevant for patient management and therapeutic response of Chagas disease . However , typing strategies for genotype-specific serodiagnosis of Chagas disease are still unavailable and requires standardization for practical application . In this study , an innovative TcI/TcVI/TcII Chagas Flow ATE-IgG2a technique was developed with applicability for universal and genotype-specific diagnosis of T . cruzi infection . For this purpose , the reactivity of serum samples ( percentage of positive fluorescent parasites-PPFP ) obtained from mice chronically infected with TcI/Colombiana , TcVI/CL or TcII/Y strain as well as non-infected controls were determined using amastigote-AMA , trypomastigote-TRYPO and epimastigote-EPI in parallel batches of TcI , TcVI and TcII target antigens . Data demonstrated that “α-TcII-TRYPO/1:500 , cut-off/PPFP = 20%” presented an excellent performance for universal diagnosis of T . cruzi infection ( AUC = 1 . 0 , Se and Sp = 100% ) . The combined set of attributes “α-TcI-TRYPO/1:4 , 000 , cut-off/PPFP = 50%” , “α-TcII-AMA/1:1 , 000 , cut-off/PPFP = 40%” and “α-TcVI-EPI/1:1 , 000 , cut-off/PPFP = 45%” showed good performance to segregate infections with TcI/Colombiana , TcVI/CL or TcII/Y strain . Overall , hosts infected with TcI/Colombiana and TcII/Y strains displayed opposite patterns of reactivity with “α-TcI TRYPO” and “α-TcII AMA” . Hosts infected with TcVI/CL strain showed a typical interweaved distribution pattern . The method presented a good performance for genotype-specific diagnosis , with global accuracy of 69% when the population/prototype scenario include TcI , TcVI and TcII infections and 94% when comprise only TcI and TcII infections . This study also proposes a receiver operating reactivity panel , providing a feasible tool to classify serum samples from hosts infected with distinct T . cruzi genotypes , supporting the potential of this method for universal and genotype-specific diagnosis of T . cruzi infection .
Trypanosoma cruzi , the etiological agent of Chagas disease [1] infects 6–7 million people worldwide , mainly in Latin America causing serious consequences for public health and national economies [2] . Geographical variations in the prevalence of clinical forms and morbidity of Chagas disease in different countries have been recorded [3] . Although the factors underlying the clinical heterogeneity of Chagas disease are still not completely understood , it has been suggested that different clinical outcome may be associated with the genetic diversity of T . cruzi isolates observed in the Americas [4] . Moreover , differences in therapeutic response of distinct T . cruzi genotypes have been also reported previously in mice infection [5–8] . Typing strategies for genotype-specific diagnosis of Chagas disease to identify the six T . cruzi discrete typing units ( DTU ) , named TcI , TcII , TcIII , TcIV , TcV and TcVI [9] have already been developed , including biochemical and molecular methods [4] . However , none of these methods allows a full resolution when used individually and a combinatory three-marker sequential typing strategy is usually required to confirm the T . cruzi genotype [10–12] . Straightforward , genotyping methods to identify the T . cruzi DTUs are currently available , but research is still required to optimize sensitivity and simplify methods so that they can be easily applied in clinical laboratories . In fact , molecular methods require a measurable parasite load to directly identify T . cruzi DTUs in samples . Because of this , the approaches used for T . cruzi genotyping requires parasite isolation by hemoculture/xenoculture followed by in vitro growth that may lead to clonal selection [13–16] . A feasible solution to overcome these problems is the design and development of genotype-specific serology to provide a current/historical profile of T . cruzi DTUs infecting an individual patient [17–20] . Moreover , genotypic-specific serodiagnosis has the potential to predict therapeutic response and provide insights upon re-activation episodes . Recently , a flow cytometry-based assay , named Chagas-Flow ATE ( Amastigote , Trypomastigote and Epimastigote ) , has been developed for simultaneous measurement of anti-amastigote , anti-trypomastigote and anti-epimastigote antibodies displaying high performance for the diagnosis and post-therapeutic monitoring of Chagas disease [21] . Aiming at optimizing the Chagas-Flow ATE for universal and genotypic-specific diagnosis of T . cruzi infection , the present study proposed the development of modified Chagas-Flow ATE , using parallel batches of distinct T . cruzi genotypes as target antigens . Standard T . cruzi strains , representative of three major genotypes ( TcI , TcII and TcVI ) were used to setup the Chagas-Flow ATE-IgG2a methodology . High-dimensional data handling were applied to select the sets attributes ( “target-antigen/serum dilution/cut-off” ) applicable for universal and genotypic-specific diagnosis of T . cruzi-infection . A receiver operating reactivity panel was proposed as a feasible tool to identify hosts infected with distinct T . cruzi genotypes . The results demonstrated the high-quality performance of TcI/TcVI/TcII Chagas-Flow ATE-IgG2a for universal and genotype-specific diagnosis of T . cruzi infection .
All animals included in this study were maintained at the Animal Science Center of the Universidade Federal de Ouro Preto , Ouro Preto , MG , Brazil , in strict accordance with the Brazilian College of Animal Experimentation Guidelines for ethical conduct in use of animals in research . Efforts were performed to reduce animal suffering . The study protocols were approved by the Ethics Committee on Animal Experimentation of the Federal University of Ouro Preto ( Protocol approval numbers #2013/48 from December , 6th , 2013 for the experimental infection and collected blood by ocular plexus puncture in mice ) . Standard T . cruzi strains , representative of three major genotypes [9] , involved in the domestic cycle of Chagas disease in Brazil , were used to setup the TcI/TcVI/TcII Chagas-Flow ATE-IgG2a methodology for the serodiagnosis of T . cruzi infection . The Colombiana , acronyms “COL” ( TcI ) [22] , CL ( TcVI ) [23] and Y ( TcII ) T . cruzi strains were used in this study [24] . All isolates were obtained from the T . cruzi cryobank at Grupo de Genômica Funcional e Proteômica de Leishmania spp e Trypanosoma cruzi , Centro de Pesquisas René Rachou ( CPqRR-FIOCRUZ/ MG ) . The T . cruzi strains were maintained by consecutive in vivo passages in Swiss female mice . Blood samples obtained from infected mice were used for experimental infection as well as for preparation of target antigens ( amastigote-AMA , trypomastigote-TRYPO and epimastigote-EPI ) used on each TcI/TcVI/TcII Chagas-Flow ATE-IgG2a platform . Female Swiss mice ( n = 118 , 28–30 days old ) , obtained from the Animal Science Centre at the Universidade Federal de Ouro Preto ( UFOP ) , MG , Brazil , were maintained in temperature-controlled room with access to water and food ad libitum . Animals were subdivided into four groups referred as T . cruzi-infected TcI/Colombiana/COL strain ( n = 36 ) , TcVI/CL strain ( n = 36 ) or TcII/Y strain ( n = 36 ) as well as non-infected mice ( NI , n = 10 ) . The infection was confirmed in all T . cruzi-infected mice , by positivity at fresh blood examination performed at day 7 , 10 or 15 post-infection . The serum samples used for the TcI/TcVI/TcII Chagas-Flow ATE-IgG2a serology were prepared from whole blood samples collected by ocular plexus puncture . Samples were collected from non-infected controls and T . cruzi-infected mice ( day 90 and day180 post-infection ) were inactivated at 56°C for 30 min and stored at -20°C until use . Considering the animal mortality during the experimental timeline , the final number of animals/group were ( TcI/Colombiana strain , n = 29 , TcVI/CL strain , n = 29 , TcII/Y strain , n = 35 and NI , n = 10 ) . The amastigote/trypomastigote/epimastigote forms of TcI , TcVI and TcII T . cruzi genotypes were obtained as described previously by Alessio et al . ( 2014 ) [21] . Enriched trypomastigotes ( TRYPO ) and amastigote ( AMA ) preparations were obtained from desynchronized in vitro tissue cultures ( L929 cell-line ) harvested at day 4–6 and 8–15 post-inoculation , respectively . Epimastigote forms were obtained at log-phase growth of axenic culture in liver infusion tryptose medium [25] . Live amastigotes and trypomastigotes as well as fixed epimastigotes forms were stained with fluorescein isothiocyanate ( FITC ) as described by Alessio et al . ( 2014 ) [21] . Briefly , AMA/TRYPO mix and EPI suspensions ( 1x107 parasites/mL ) were stained with FITC ( 100μg/mL for TcI/Colombiana strain and 200μg/mL for TcVI/CL strain and TcII/Y strain ) for 30 min at 37°C . After staining , AMA/TRYPO mix were kept at 37°C for 60min and EPI preparation stored at 4°C for 24h prior to use . The three FITC-labeled parasite preparations were mixed accordingly to obtain an equivalent proportion of AMA ( 33% ) , TRYPO ( 33% ) and EPI ( 33% ) in the final ATE-parasite Mix Platforms , monitored by flow cytometry checking performed prior use . The FITC-labeling approach led to a differential staining phenomenon previously described by Alessio et al . , ( 2014 ) [21] that allowed the segregation of AMA , TRYPO and EPI organisms in distinct clusters , based on the FITC ( Fluorescence 1- FL1 ) vs Forward Scatter ( FSC ) dot plot distribution ( Fig 1 ) . An outline of the TcI/TcVI/TcII Chagas-Flow ATE-IgG2a applied to the serodiagnosis of T . cruzi infection is provided in the Fig 1 . The method comprises two steps referred as: i ) Experimental Procedure ( Fig 1A ) and ii ) Analysis of genotype-specific anti-T . cruzi IgG2a reactivity ( Fig 1B ) . Data mining for universal and genotype-specific diagnosis of T . cruzi infection was first performed by non-parametric Kruskal—Wallis test followed by Dunns' multiple comparison post-test to compare the overall reactivity profile of TcI/TcVI/TcII Chagas-Flow ATE-IgG2a . Significant differences were considered at p ≤ 0 . 05 . The performance indices ( global accuracy defined by the area under the curve-AUC , sensitivity-Se and specificity-Sp ) for the pair of attributes ( “target antigen/serum dilution” ) selected for universal diagnosis purposes were determined by the receiver operating characteristic ( ROC ) curve , scatter plot distribution and Two-Graph ROC curve ( TG-ROC ) analysis . Histogram plot distributions and nonlinear regression analysis was used for comparative analysis of pair of attributes ( “target antigen/serum dilution” ) selected for genotypic-specific diagnosis purposes . The global median was calculated for each pair of attributes ( “target antigen/serum dilution” ) to define putative cut-off edges to segregate the reactivity amongst T . cruzi-infected hosts . Scatter plot distribution was used for performance analysis of sets of selected attributes ( “target-antigen/serum dilution/cut-off” ) applicable for genotypic-specific diagnosis of T . cruzi-infection . The GraphPad Prism software , Version 5 . 0 ( San Diego , CA , USA ) was used for statistical analysis and graphic arts . Decision trees were built for the set of selected attributes ( “target-antigen/serum dilution/cut-off” ) to create algorithms ( root and branch attributes ) to classify T . cruzi in distinct population/prototype scenarios ( TcI-infection/Colombiana vs TcVI/CL vs TcII/Y ) and ( TcI-infection/Colombiana vs TcII/Y ) . The WEKA software ( Waikato Environment for Knowledge Analysis , version 3 . 6 . 11 , University of Waikato , New Zealand ) was used for decision tree construction . Step-wise discriminant analysis was applied to determine the global accuracy and the leave-one-out-cross-validation-LOOCV values . The R-project for statistical computing software , Version 3 . 0 . 1 was used for discriminant analysis . The algorithm C4 . 5 was used to build the decision tree using an implementation named J48 . This method analyzed all characteristics to select a minimum set of markers that could efficiently separate study groups .
The overall profiles of TcI/TcVI/TcII Chagas-Flow ATE-IgG2a reactivity observed for T . cruzi infected mice ( TcI/Colombiana strain , TcVI/CL strain and TcII/Y strain ) and non-infected controls are presented in the Fig 2 . The reactivity of individual samples were assessed for distinct target-antigen ( amastigote-AMA , trypomastigote-TRYPO and epimastigote-EPI ) from T . cruzi genotype I—Colombiana strain ( Fig 2—left panels ) , genotype VI—CL strain ( Fig 2—middle panels ) and genotype II—Y strain ( Fig 2—right panels ) along the titration curves ( serum dilutions ranging from 1:500 to 1:64 , 000 ) . Comparative analysis allowed the selection of pair of attributes ( “target antigen/serum dilution” ) with the most promising perspective to be used for universal and genotypic-specific diagnosis of T . cruzi infection . The pair of attributes “anti-TcII TRYPO reactivity at 1:500” presented the highest significant difference between non-infected mice and all T . cruzi-infected hosts ( TcI/Colombiana , TcVI/CL and TcII/Y strains ) , and therefore was further evaluated for universal diagnosis purpose ( Fig 2—light gray continuous rectangle ) . The pairs of attributes with putative applicability to genotype-specific diagnosis of T . cruzi infection comprise: ( “anti-TcII AMA reactivity at 1:1 , 000”; “anti-TcI TRYPO reactivity at 1:4 , 000” and “anti-TcVI EPI reactivity at 1:1 , 000” ) . The pair of attributes “anti-TcII AMA reactivity at 1:1 , 000” presented the highest ability to distinguish the lower reactivity of hosts infected with TcI/Colombiana strain from the higher reactivity observed for hosts infected with TcVI/CL or TcII/Y strains ( Fig 2—right dark gray dotted frame ) . The pair of attributes “anti-TcI TRYPO reactivity at 1:4 , 000” presented the highest ability to discriminate lower reactivity of hosts infected with TcII/Y strain from the intermediate reactivity observed for hosts infected with TcVI/CL strain and the higher reactivity observed for hosts infected with TcI/Colombiana strain ( Fig 2—left dark gray dotted frame ) . The pair of attributes “anti-TcVI EPI reactivity at 1:1 , 000” presented the most relevant potential to distinguish the lower reactivity of hosts infected with TcII/Y strain from the higher reactivity observed for hosts infected with TcI/Colombiana or TcVI/CL T . cruzi strains ( Fig 2—middle dark gray dotted frame ) . Together , these pairs of attributes were selected for further performance assessment applicable to the genotype-specific diagnosis of T . cruzi infection . The performance of the pre-selected pair of attributes “anti-TcII TRYPO reactivity at 1:500” applied to the universal diagnosis of T . cruzi infection is present in the Fig 3 . Comparative analysis demonstrated that the median value of “anti-TcII TRYPO reactivity at 1:500” differ significantly between non-infected mice and all T . cruzi-infected hosts ( TcI/Colombiana , TcVI/CL and TcII/Y strains ) ( Fig 3A ) . ROC curve analysis indicated the PPFP value of 20% as the cut-off edge with excellent performance indices ( area under the curve-AUC = 1 . 0 along with Sensitivity-Se and Specificity-Sp of 100% ) ( Fig 3B ) . Scatter plot distribution of individual values illustrates the ability of this set of attributes to completely segregate the serum samples of the NI and T . cruzi-infected hosts ( Fig 3C ) . Additional analysis by TG-ROC confirmed the selected PPFP value of 20% as the best cut-off for universal diagnosis of T . cruzi infection using the selected set of attributes ( Fig 3D ) . The overall reactivity profile of the pre-selected pairs of attributes ( “anti-TcII AMA reactivity at 1:1 , 000”; “anti-TcI TRYPO reactivity at 1:4 , 000” and “anti-TcVI EPI reactivity at 1:1 , 000” ) are shown in the Fig 4 . Data mining was carried out by histogram graph and trendlines drawn by non-linear regression analysis . The results showed that the “anti-TcII AMA reactivity at 1:1 , 000” of serum samples from hosts infected with TcI/Colombiana strain displayed a nearly unimodal distribution in the region of PPFP values = 10% , contrasting with the bimodal distribution of serum samples from hosts infected with TcVI/CL and TcII/Y strains that shows a shift towards higher PPFP values ( Fig 4A ) . The analysis of “anti-TcI TRYPO reactivity at 1:4 , 000” revealed a clear polarization of serum samples from hosts infected with TcI/Colombiana strain with a unimodal distribution in the region of PPFP values around 100% , contrasting with a unimodal distribution in the region of PPFP values around 0% observed for serum samples from hosts infected with TcII/Y strain . Again , a typical bimodal distribution was noticed for serum samples from hosts infected with TcVI/CL strain ( Fig 4B ) . The histogram distribution of “anti-TcVI EPI reactivity at 1:1 , 000” revealed a clear Gaussian unimodal distribution of serum samples from hosts infected with TcI/Colombiana and TcVI/CL strains within the region of PPFP values around 50% . On the other hand , the unimodal distribution observed for serum samples from hosts infected with TcII/Y strain showed a clear shift towards PPFP values < 50% ( Fig 4C ) . Comparative analysis of median reactivity pattern of the selected pairs of attributes confirmed the trend observed by histogram and non-linear regression analysis , pointing out the ability of “anti-TcII AMA reactivity at 1:1 , 000” to segregate hosts infected with TcII/Y strain ( and TcVI/CL strain ) apart from those infected with TcI/Colombiana strain ( Fig 4D ) . On the other hand , the “anti-TcI TRYPO reactivity at 1:4 , 000” was able to segregate hosts infected with TcI/Colombiana strain ( and TcVI/CL strain ) apart from those infected with TcII/Y strain ( Fig 4E ) . Moreover , the “anti-TcVI EPI reactivity at 1:1 , 000” was capable to discriminate the hosts infected with TcVI/CL strain ( and TcI/Colombiana strain ) apart from those infected with TcII/Y strain ( Fig 4F ) . Aiming at making the TcI/TcVI/TcII Chagas-Flow ATE-IgG2a applicable to the genotypic-specific diagnosis of T . cruzi infection , the overlaid trendlines for the overall reactivity ( S1A Fig ) along with the global median PPFP value of each pair of pre-selected attributes ( S1B Fig ) were employed to establish potential cut-off edges to categorize individual samples as they present negative ( <cut-off ) or positive ( >cut-off ) reactivity . Using this approach , specific cut-off edges were defined for each pre-selected pairs of attributes ( “anti-TcII AMA reactivity at 1:1 , 000”; “anti-TcI TRYPO reactivity at 1:4 , 000” and “anti-TcVI EPI reactivity at 1:1 , 000” ) , comprising PPFP = 40% , PPFP = 50% and PPFP = 45% , respectively ( S1B Fig ) . Diagrams were used to compile the reactivity patterns and calculate the proportion of negative and positive results for each selected set of attributes ( “target-antigen/serum dilution/cut-off” ) . Data analysis showed that the set of attributes “anti-TcII AMA reactivity at 1:1 , 000 , cut-off = 40%” were able to show positive results in 74% of hosts infected with TcII/Y strain ( and 55% of TcVI/CL strain ) apart from 14% of those infected with TcI/Colombiana strain ( S1C Fig ) . Moreover , the set of attributes “anti-TcI TRYPO reactivity at 1:4 , 000 , cut-off = 50%” showed positive results in 83% of hosts infected with TcI/Colombiana strain ( and 59% of TcVI/CL strain ) contrasting with 14% of those infected with TcII/Y strain ( S1C Fig ) . Furthermore , the set of attributes “anti-TcVI EPI reactivity at 1:1 , 000 , cut-off = 45%” showed positive results in 72% of hosts infected with TcI/Colombiana strain ( and 55% of TcVI/CL strain ) distinct from 27% of those infected with TcII/Y strain ( S1C Fig ) . Scatter plot distribution further illustrated the pre-selected sets of attributes segregated the reactivity of hosts infected with distinct T . cruzi genotypes , emphasizing the performance of “anti-TcII AMA reactivity at 1:1 , 000 , cut-off = 40%” to discriminate the majority of the hosts infected with TcI/Colombiana strain and the ability of “anti-TcI TRYPO reactivity at 1:4 , 000 , cut-off = 50%” and “anti-TcVI EPI reactivity at 1:1 , 000 , cut-off = 45%” to discriminate the majority of the hosts infected with TcII/Y strain . In general , considerable proportion of hosts infected with the hybrid TcVI/CL strain presented positive results using the pre-selected set of attributes ( S1D Fig ) . The performance of combined TcI/TcVI/TcII Chagas-Flow ATE-IgG2a was evaluated in two population/prototypes ( TcI/Colombiana vs TcVI/CL vs TcII/Y strains and TcI/Colombiana vs TcII/Y strains ) selected to exemplify the distribution of human T . cruzi infection in distinct geographical regions around the world . Data analysis was carried out using the sets of pre-selected attributes ( “anti-TcII AMA reactivity at 1:1 , 000 , cutoff = 40%”; “anti-TcI TRYPO reactivity at 1:4 , 000 , cut-off = 50%” and “anti-TcVI EPI reactivity at 1:1 , 000 , cut-off = 45%” ) , as presented in the Fig 5 . Three-dimensional plots were built to obtain a panoramic snapshot provided by the combined reactivity of the three sets of pre-selected attributes ( Fig 5A and 5D ) . Data analysis was carried out in two population/prototypes ( TcI/Colombiana vs TcVI/CL vs TcII/Y strains , Fig 5A ) and ( TcI/Colombiana vs TcII/Y strains , Fig 5D ) . The results obtained for the first population/prototype ( TcI/Colombiana vs TcVI/CL vs TcII/Y strains ) demonstrated clearly that sera samples from hosts infected with TcI/Colombiana strain was confined in a region of high “anti-TcI TRYPO reactivity at 1:4 , 000” ( left lateral axis ) and low “anti-TcII AMA reactivity at 1:1 , 000” ( vertical axis ) . In contrast , sera from hosts infected with TcII/Y strain presented a shift towards lower “anti-TcI TRYPO reactivity at 1:4 , 000” ( left lateral axis ) and higher “anti-TcII AMA reactivity at 1:1 , 000” ( vertical axis ) . A slight translocation of samples from hosts infected with TcII/Y strain towards lower “anti-TcVI EPI reactivity at 1:1 , 000” ( right lateral axis ) was also observed . A notable evidence was that the sera samples from hosts infected with TcVI/CL strain displayed a typical interweaved distribution pattern ( Fig 5A ) . The dichotomic reactivity pattern of the three sets of pre-selected attributes was more evident when data analysis was performed in the second population/prototype which included only hosts infected with TcI/Colombiana vs TcII/Y strains ( Fig 5D ) . Decision tree analyses were built for the two population/prototypes ( Fig 5B and 5E ) . The algorithm proposed for the first population/prototype indicated the “anti-TcI TRYPO reactivity at 1:4 , 000 , cut-off = 50%” as the root attribute , followed by “anti-TcII AMA reactivity at 1:1 , 000 , cut-off = 40%” as the first branch and “anti-TcVI EPI reactivity at 1:1 , 000 , cut-off = 45%” as the second branch to classify sera samples from hosts infected with TcI/Colombiana vs TcVI/CL vs TcII/Y strains with a moderate global accuracy ( 68 . 8% , LOOCV = 58 . 0% ) ( Fig 5B ) . Data obtained for the second population/prototype indicated that the same decision tree algorithm presented high global accuracy ( 93 . 8% , LOOCV = 87 . 5% ) ( Fig 5E ) . Bar charts were constructed to illustrate the categorical classification proposed by the decision trees , demonstrating the number of animals that ranked within each branch amongst the T . cruzi-infected hosts for the first population/prototype ( TcI/Colombiana vs TcVI/CL vs TcII/Y strains , Fig 5C ) and the second population/prototype ( TcI/Colombiana vs TcII/Y strains , Fig 5F ) . Data demonstrated that the algorithm applied to the first population/prototype was not able to clusterize the serum samples from hosts infected with the TcVI/CL strain that display a spread ranking within branches ( Fig 5C ) . On the other hand , the algorithm applied to the second population prototype yielded lower classification error with only four samples misplaced within branches ( one sample from TcI/Colombiana strain and three from TcII/Y strain ) ( Fig 5F ) . Discriminant analysis of combined TcI/TcVI/TcII Chagas-Flow ATE-IgG2a for genotype-specific diagnosis of T . cruzi infection groups performed for the two population/prototypes are provided in the S2 Fig . Data analysis demonstrate that for the first population/prototype , the combined TcI/TcVI/TcII Chagas-Flow ATE-IgG2a categorize 82 . 8% of serum samples from hosts infected with TcI/Colombiana strain apart from 82 . 9% of those infected with TcII/Y strain . However , only 38% of serum samples from hosts infected with TcVI/CL strain were clusterized in a particular branch ( S2A Fig ) . If we consider the scenario represented by the second population/prototype , the data showed that 96 . 6% of serum samples from samples TcI/Colombiana strain were correctly classified apart from 91 . 4% of those infected with TcII/Y strain ( S2B Fig ) . Reactivity boards were constructed using the pre-selected set attributes , including “anti-TcII TRYPO reactivity at 1:500 , cut-off = 20%” for universal diagnosis purpose and “anti-TcI TRYPO reactivity at 1:4 , 000 , cut-off = 50%” , “anti-TcII AMA reactivity at 1:1 , 000 , cut-off = 40%” and “anti-TcVI EPI reactivity at 1:1 , 000 , cut-off = 45%” for genotype-specific diagnosis ( Fig 6 ) . The reactivity at selected serum dilutions ( Fig 6—dashed frames ) were used to further create a receiver operating reactivity panel , applicable for universal diagnosis ( NI vs T . cruzi-infected hosts , Fig 6B ) and for genotype-specific diagnosis applied to the first population/prototype ( TcI/Colombiana vs TcVI/CL vs TcII/Y strains , Fig 6B ) and the second population/prototype ( TcI/Colombiana vs TcII/Y strains , Fig 6C ) . When using the TcI/TcVI/TcII Chagas-Flow ATE-IgG2a applied for the universal diagnosis purpose , the attribute α-TII 500 ( >20% ) presenting a positive score ( + ) define the presence of T . cruzi infection , while a negative score ( - ) ruled out the presence of T . cruzi infection . The reactivity panel for TcI/TcVI/TcII Chagas-Flow ATE-IgG2a applied for the genotype-specific diagnosis regardless the population/prototype scenario indicated scores sequences of α-TI ( trypomastigote TcI ) 4 , 000 ( >50% ) /α-AII ( amastigote TcII ) 1 , 000 ( >40% ) /α-EVI ( epimastigote TcVI ) 1 , 000 ( >45% ) to define the T . cruzi infection with distinct genotypes defined as: ( +/-/not applicable ( na ) or -/-/+ ) for TcI/Colombiana and ( -/+/na or -/-/- ) for TcII/Y strain infection . The extensions of the score ( +/+ ) do not allow the proper identification the T . cruzi infection genotype , since in can belongs to hosts infected with TcI/Colombiana ( +/+/+ ) , TcVI/CL ( +/+/na ) or even TcII/Y strain ( +/+/- ) ( Fig 6B and 6C ) . Together , the proposed receiver operating reactivity panels for TcI/TcVI/TcII Chagas-Flow ATE-IgG2a provided a feasible tool to classify the serum samples as they belong to the true respective groups , supporting the potential of this method for universal and genotype-specific diagnosis of T . cruzi infection .
The broad genetic variability of T . cruzi has been related to biological characteristics ( infectivity , parasitemia , tissue tropism , mortality during the acute phase of infection [13 , 26–30] and susceptibility/resistance to drugs [5 , 8 , 31–34] in murine model and infectivity , replication and differentiation in vector o ) [35] , epidemiological characteristics [4 , 36 , 37] and clinical manifestations [38–40] of Chagas disease . Therefore , the knowledge of parasite genetics may offer insights about the biology of the parasite , patient’s treatment outcome , clinical aspects of human disease , as well as how to establish epidemiological surveillance and control of Chagas disease [41 , 42] . So , the genetic diversity of T . cruzi infection may also influence the sensitivity of the techniques used for Chagas disease diagnosis [43–45] . The currently available methods for genotype-specific diagnosis of T . cruzi infection , most based on molecular biology approaches present distinct levels of complexity and in general display high specificity but moderate sensitivity [4 , 10–12 , 46–52] . Moreover , a combination of several genetic markers is necessary to detect and distinguish the T . cruzi genotypes [10–12] . Furthermore , the majority of these methods can not directly be performed in biological and clinical samples , requiring previous parasite isolation by hemoculture/xenoculture followed by in vitro growth and maintenance that may lead to clonal selection [14 , 53–55] . Besides , it is known that the parasitemia in patients and reservoirs of T . cruzi are variable and that the success of parasite isolation is dependent on the host parasitemia . The full extent of lineage distribution in nature using genetic markers it is not known due to the low levels of circulating parasitemia and possible lineage-specific tissue sequestration [34 , 56–58] . In addition , it has been proposed that parasite isolates from blood may not necessarily represent the full set of strains current in the individual , hence some strains of T . cruzi can be confined to tissues [11 , 13 , 16 , 15] . In general , PCR ( polymerase chain reaction ) based genotyping has limitations that hamper the analysis of large numbers of samples . Therefore , the development of methods for diagnosis and serotyping of Chagas disease are urgently required . Attempting to address this matter , Mendes et al . ( 2013 ) [18] have described a set of B-cell epitopes able to discriminate TcI and TcII infections , demonstrating the potential of these targets for Chagas disease serotyping . Later on , the putative TcI epitope reported by Mendes et al . ( 2013 ) [18] was found to be conserved across all T . cruzi lineages by studies developed by Bhattacharyya et al . ( 2014 ) [19] . Moreover , samples from animals infected with TcVI presented cross-reactivity with a range of T . cruzi-derived peptides , suggesting the need of improved antigen search and the development of a robust panel of strain-specific epitopes to achieve a method applicable in large epidemiological studies [18] . Aiming at developing innovative serological approaches for universal and improved genotypic-specific diagnosis of T . cruzi experimental infection , our goal was optimize the Chagas-Flow ATE methodology , proposed originally by Alessio et al . ( 2014 ) [21] . The present approach is based on parallel batches of distinct T . cruzi genotypes as target antigens , employing parasites strains of three more important genotypes associated with human infection and Chagas disease ( TcI , TcVI and TcII ) [4 , 38 , 59] although others genotypes exist such as TcIII [60–62] , TcIV recently described as associated to oral transmission [63 , 64] and TcV associated to the classical clinical forms of Chagas disease ( cardiac and digestive ) [39] . Previous studies have demonstrated that patients from distinct geographic areas , infected with different genotypes of T . cruzi seem to display differential serological pattern upon serodiagnosis of Chagas disease , when employing distinct methodological approaches and different T . cruzi target antigens [4 , 43 , 44 , 65] . Verani et al . ( 2009 ) [44] have demonstrated that the performance of two serological tests , using serum samples from distinct geographical regions ( Bolivia and Peru ) displayed distinct sensitivity , ranging from 26 . 6%-87 . 5% , corroborating the hypothesis that intrinsic features of regional parasite strains may influence the serological tests . Studies using six recombinant antigens of T . cruzi tested in samples from Argentina , Brazil , Chile , Colombia , El Salvador , Guatemala , Honduras and Venezuela also reported discrepancy in the serological reactivity ranging from 79% to 100% [43] . In this study , we have intended to evaluate the performance of combined TcI/TcVI/TcII Chagas-Flow ATE-IgG2a for universal diagnosis of T . cruzi infection , simulating two population prototypes that may represent the geographic distribution of T . cruzi infection in the Latin America . The first population prototype , represent regions were TcI , TcVI and TcII genotypes are co-endemic . In the second population setting , we intended to evaluate the performance of combined TcI/TcVI/TcII Chagas-Flow ATE-IgG2a to discriminate the infections with TcI/Colombiana vs TcII/Y strains . Our findings demonstrated that regardless the population prototype , the TcI/TcVI/TcII Chagas-Flow ATE-IgG2a presented an outstanding performance for universal diagnosis of T . cruzi infection using the set of attributes “anti-TcII TRYPO reactivity at 1:500 , cut-off = 20%” ( Fig 3 ) . In fact , although the sensitivity of TcI/TcVI/TcII Chagas-Flow ATE-IgG2a varies according to the target antigen employed , the TcII TRYPO antigen was able to detect seroreactivity in all mice infected with distinct T . cruzi genotypes ( Figs 2 and 3 ) . Corroborating with our study , Bhattacharyya et al . ( 2014 ) [19] also observed that all sera from patients with chronic Chagas disease recognized the T . cruzi TcII lysate antigen preparation . Previous studies have also demonstrated the influence of parasite genotype on the pattern of antibody response in experimental models [66] . It has been described that the profile of lytic antibodies varies when distinct T . cruzi strains are used as targets , suggesting that genotypic-specific antigenic features may be involved in the induction of lytic antibodies [67–69] . Moreover , Di Noia et al . ( 2002 ) [70] have reported that two T . cruzi antigens , named small surface antigen of trypomastigotes ( TSSAI and TSSAII ) presented the ability of genotypic-specific recognition of T . cruzi infection . These studies were expanded by Bhattacharyya et al . ( 2010 , 2014 and 2015 ) [17 , 19 , 20] that reported that TSSA pepII/V/VI isoforms were able to distinguish samples of hosts infected with distinct T . cruzi genotypes . Based on these findings , the authors proposed that TSSA isoforms are feasible serological markers to identify a T . cruzi lineage in human and experimental infection . However , the TSSA pepI did not yield significant reactivity , suggesting that novel targets for TcI is still required . The innovative TcI/TcVI/TcII Chagas-Flow ATE-IgG2a methodology presented a high-quality performance to segregate infections with TcI/Colombiana , TcVI/CL or TcII/Y strain . The performance of combined TcI/TcVI/TcII Chagas-Flow ATE-IgG2a for genotype-specific diagnosis of T . cruzi infection differs depending on the population prototypes used to represent distinct geographic regions of T . cruzi infection in the Latin America . In the first prototype ( TcI/TcVI/TcII ) , our data demonstrated that the proposed method showed a moderate global accuracy ( 68 . 8% , LOOCV = 58 . 0% ) to discriminate the infections with TcI/Colombiana vs TcVI/CL vs TcII/Y strains . On the other hand , the combined TcI/TcVI/TcII Chagas-Flow ATE-IgG2 was capable to discriminate the infections with TcI/Colombiana vs TcII/Y strains in the second population prototype ( TcI/TcII ) with high global accuracy ( 93 . 8% , LOOCV = 87 . 5% ) ( Fig 5 ) . Overall , hosts infected with TcI/Colombiana and TcII/Y strains displayed opposite patterns of reactivity with “anti-TcI TRYPO” and “anti-TcII AMA” and hosts infected with TcVI/CL strain showed a typical interweaved distribution pattern ( Figs 4 and S1 ) . This phenomenon may reflect the phylogenetic origin of DTUs [4] where TcI and TcII are ancestral DTUs presenting polar characteristics [5 , 27 , 28 , 30 , 71–73] , whereas the TcVI has a hybrid origin , showing intermediate characteristics of both polar genotypes [73 , 74] . In conclusion , based on the receiver operating characteristic , the TcI/TcVI/TcII Chagas-Flow ATE-IgG2a seems to be a feasible tool to classify the serum samples as they belong to the true respective groups infected with distinct T . cruzi genotypes ( Fig 6 ) , suggesting its applicability for both , universal and genotype-specific diagnosis of T . cruzi infection in clinical laboratories . The proposed methodology includes essential advantages such as high sensitivity and specificity , ease to perform , using a wide range of antigenic preparation into a single flow cytometric platform [21 , 75–79] . Future derivation of TcI/TcVI/TcII Chagas-Flow ATE-IgG2a as the development of a suitable ELISA or multiplex beads assay would contribute to practical applications in routine clinical laboratories , since the original version of this fluorescence-based methodology is more reliable for applications in reference laboratories . Additional tests are under investigation to establish accuracy of TcI/TcVI/TcII Chagas-Flow ATE-IgG2a to identify mixed infections with distinct T . cruzi genotypes . An extension of this study may be applicable to other genetic groups not included in this work ( TcIII , TcIV and TcV ) . Further studies including serum samples from patients with genotypic-specific diagnosis Chagas disease performed by molecular methods are currently under investigation as a proof-of-concept to propose a prototype for clinical purposes , epidemiological studies and post-therapeutic monitoring applications . | Chagas disease remains a significant public health issue infecting 6–7 million people worldwide . The factors influencing the clinical heterogeneity of Chagas disease have not been elucidated , although it has been suggested that different clinical outcome may be associated with the genetic diversity of T . cruzi isolates . Moreover , differences in therapeutic response of distinct T . cruzi genotypes have been also reported . Typing strategies for genotype-specific diagnosis of Chagas disease to identify the T . cruzi discrete typing units ( DTU ) have already been developed , including biochemical and molecular methods , however the techniques have limitations . The majority of these methods can not directly be performed in biological and clinical samples . In addition , it has been proposed that parasite isolates from blood may not necessarily represent the full set of strains current in the individual as some strains can be confined to tissues . The improvement of genotype-specific serology to identify the T . cruzi DTU ( s ) present in a given host may provide a useful tool for clinical studies . In the present investigation , we developed an innovative TcI/TcVI/TcII Chagas Flow ATE-IgG2a technique with applicability for universal and genotype-specific diagnosis of T . cruzi infection that may contribute to add future insights for genotype-specific diagnosis of Chagas disease . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"biogeography",
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"engineering",
"and",
"technology",
"population",
"genetics",
"tropical",
"diseases",
"parasitic",
"diseases",
"parasitic"... | 2017 | Performance of TcI/TcVI/TcII Chagas-Flow ATE-IgG2a for universal and genotype-specific serodiagnosis of Trypanosoma cruzi infection |
Endemic Burkitt's lymphoma ( eBL ) arises from the germinal center ( GC ) . It is a common tumor of young children in tropical Africa and its occurrence is closely linked geographically with the incidence of P . falciparum malaria . This association was noted more than 50 years ago . Since then we have learned that eBL contains the oncogenic herpes virus Epstein-Barr virus ( EBV ) and a defining translocation that activates the c-myc oncogene . However the link to malaria has never been explained . Here we provide evidence for a mechanism arising in the GC to explain this association . Accumulated evidence suggests that eBL arises in the GC when deregulated expression of AID ( Activation-induced cytidine deaminase ) causes a c-myc translocation in a cell latently infected with Epstein-Barr virus ( EBV ) . Here we show that P . falciparum targets GC B cells via multiple pathways to increase the risk of eBL . 1 . It causes deregulated expression of AID , thereby increasing the risk of a c-myc translocation . 2 . It increases the number of B cells transiting the GC . 3 . It dramatically increases the frequency of these cells that are infected with EBV and therefore protected from c-myc induced apoptosis . We propose that these activities combine synergistically to dramatically increase the incidence of eBL in individuals infected with malaria .
Endemic Burkitt's lymphoma ( eBL ) is an extremely common tumor of young children in tropical Africa [1] . Genetic , phenotypic and transcriptional analysis suggests that it originates from germinal center ( GC ) cells [2] , [3] although it actually grows in extrafollicular locations . It is defined by a well described chromosomal translocation between the c-myc oncogene and one of the immunoglobulin loci that results in constitutive activation of the oncogene leading to uncontrolled growth of the cell [4] , [5] , [6] . Recent studies indicate that this translocation may be mediated as a consequence of deregulated expression of the enzyme AID ( Activation-induced cytidine deaminase ) [7] , [8] , [9] . AID is highly expressed in GC B cells and is normally responsible for the processes of somatic hypermutation and class switch recombination of immunoglobulin genes as they undergo affinity maturation in the GC [10] . This restricted expression of AID further supports the notion that eBL originates in the GC . eBL is also closely associated with two infectious agents , P . falciparum malaria and Epstein-Barr virus ( EBV ) [1] , [11] , [12] . The distribution of the tumor in Africa closely matches that of hyper- and holoendemic malaria [12] while EBV was originally discovered in eBL tumor biopsies . Subsequently , we have learned a great deal about the molecular mechanism behind eBL pathogenesis and the transforming ability of EBV . EBV is a B lymphotropic herpes virus that can drive the activation and proliferation of newly infected B cells by expressing a series of latent proteins and noncoding RNAs that collectively are referred to as the growth transcription program[13] , [14] . In vivo , however , EBV establishes a lifelong , quiescent , persistent infection in resting memory B cells [15] , [16] . The virus makes the transition in vivo from a newly infected activated B cell blast to a resting memory B cell via passage through the GCs of the tonsillar lymphoid tissue [[17] , [18] . In doing so it recapitulates the mechanism by which normal B cells become memory B cells ( for a detailed description of the mechanism see [14] ) . Normally deregulation of c-myc expression such as is found in eBL would lead to apoptotic death of the cell; however , evidence suggests that exposure to the EBV growth program prior to entry into the GC [14] , [19] and viral genes expressed in the GC [20] , [21] are sufficient to convey a level of resistance to this apoptosis . Thus , the cells in the GC most likely to tolerate the c-myc translocation are the ones already latently infected with EBV . This also places EBV at the site of eBL origin , the GC . Interestingly GC cells carrying EBV express only a limited subset of the latent proteins ( default transcription program ) and even these become silenced as the infected cells enter the memory compartment [16] , [17] , [22] . Here the virus only expresses small non-coding RNAs including ∼40 miRNAs . The exception is that they also express the viral DNA tethering protein EBNA1 when the cells occasionally divide as part of normal memory B cell homeostasis [23] . Viral gene expression in eBL resembles the infected dividing memory B cells , not the GC cell: i . e . viral gene expression is limited to the viral DNA tethering protein ( EBNA1 ) and the non-coding RNAs . This led us to propose that eBL is a tumor of a GC cell that has left the lymph node to become a resting memory B cell but is unable to do so because it continues to proliferate , driven by the deregulated c-myc oncogene . While understanding the function of c-myc and EBV in eBL has progressed , the role of P . falciparum malaria has remained poorly understood . P . falciparum malaria is immunosuppressive [24] and there is considerable evidence that this leads to much higher viral burdens of EBV [25] , [26] . However it is well documented that increased EB viral loads associated with immunosuppression [27] predispose to EBV positive immunoblastic lymphoma not Burkitt's lymphoma [28] . We hypothesize that malaria plays multiple roles in eBL pathogenesis . First , we propose that malaria has the capacity to induce deregulated expression of AID thereby increasing the likelihood of the translocation event . Second , we suggest that the higher viral burdens lead to an increased frequency of newly infected B cell blasts in the tonsils . This results in more EBV infected B cells transiting the GC and consequently a higher frequency of cells in the GC able to tolerate a c-myc translocation . Taken together malaria infection would both increase the likelihood of c-myc translocations in the GC and the probability that it will occur in a cell that can tolerate it , namely an EBV infected cell . In this paper we have sought to test this hypothesis .
To provide direct support for our hypothesis we have sought in vitro evidence that P . falciparum can stimulate AID expression . Tonsil B cells were incubated with malaria extract prepared by lysing red blood cells infected with P . falciparum . As controls we used the known Toll like receptor 9 ( TLR9 ) agonist CpG and costimulation with IL-4 and CD40 ligand . Figure 1A shows a time course of AID mRNA induction with various combinations of stimulants . It is apparent that optimal induction requires a combination of T cell help ( CD40 ligand and IL-4 ) and the TLR9 agonist CpG , with peak activation occurring after 5 days of culture . In our hands , CpG alone had minimal or no effect on AID induction although it was extremely potent in driving B cell proliferation ( not shown ) . These results are consistent with previous reports [29] . Figure 1B shows the same experiment where extracts from lysed red blood cells infected with P . falciparum were also tested . Similarly to CpG , the parasite extract had minimal or no effect when added alone but showed strong stimulation of AID expression in combination with IL-4 and CD40 ligand . This suggests that the parasite would only stimulate AID expression when T cell help is available , i . e . in the GC . As with CpG activity peaked by day 5 . At this time , the parasite extract was three times as effective as CpG in inducing AID mRNA . The parasite extract differed from CpG in that when added alone it was not able to stimulate B cell proliferation ( not shown ) . Optimal stimulation of AID expression by CpG requires costimulation through the BCR . This is demonstrated in Figure 2 , where inclusion of sIg cross-linking increased stimulation two and a half fold compared to CpG alone . Comparison to parasite extract in the same experiment demonstrated that sIg cross-linking plus CpG were about as effective as parasite extract alone and that sIg cross-linking did not significantly enhance the effect of parasite extract . Extracts from uninfected RBCs showed no activity either alone or in combination with other stimuli ( Figure 2 ) . We conclude that the parasite is able to generate a signal that is as effective as the TLR9 and BCR signals combined . To confirm that the increase in AID mRNA stimulated by the parasite was reflected in increased AID protein expression , we repeated the stimulation experiments and examined the resulting cells for AID by flow cytometry . As may be seen in Figure 3A–C , parasite extract induced comparable levels of AID protein expression to that achieved with CpG when combined with sIg cross-linking . It is noteworthy that in some cases the relative level of protein expression measured by FACS did not match the levels seen for the mRNA ( see for example Unstimulated versus CD40L +IL4 ) . This likely reflects that the cells are newly stimulated in culture and there is a lag between the synthesis of AID mRNA and the production of the protein . We conclude , therefore , that P . falciparum is a potent antigen independent stimulator of AID expression . When combined with T cell help signals it is at least as effective as the combination of CpG and BCR cross-linking . It has been reported that the P . falciparum metabolic product of hemoglobin digestion , hemozoin [30] , is a ligand for TLR9 , but this has only been tested with dendritic cells [31] , [32] . It has not been shown for B cells . To test if hemozoin could be taken up by B cells we have incubated hemozoin and CpG with two B cell lines , BL2 ( EBV negative BL line ) and IM171 ( spontaneous EBV positive lymphoblastoid cell line ) . As may be seen in Figure 4A , both were readily taken up by the B cells ( Hemozoin crystals were visualized using reflection microscopy and CpG is tagged with Alexa-488 ) . To test if hemozoin could thereby act to stimulate AID expression we have compared the stimulatory activity of hemozoin to CpG and sIg . The result is shown in Figure 4B . Hemozoin alone had no effect , however in combination with sIg cross-linking it was two and a half fold more effective than sIg alone , comparable with CpG plus sIg . No appreciable increase was observed when parasite DNA was added ( 3 ug/ml ) either alone or together with hemozoin . This suggests that the hemozoin preparation alone was sufficient for the activity . These results demonstrate that hemozoin is one component of the P . falciparum extract that is capable of stimulating AID expression however there must be another component that is mimicked by sIg cross-linking to obtain the optimal stimulation obtained with whole parasite extracts . To test the hypothesis that malaria is associated with higher levels of AID expression in vivo , we have examined and compared purified GC cells from tonsils obtained from malaria infected and uninfected patients who were matched for age , sex and socioeconomic class . The results are summarized in Figure 5A–B . mRNA was isolated from similar numbers of GC B cells . For all samples AID and c-myc mRNA levels were normalized to β-actin and the level in the GC population is expressed relative to that in a standard naïve B cell population ( calibrator ) isolated from a single Boston tonsil . The level of AID mRNA in the GC cells from the control population are comparable with what has been reported previously [33] . In comparison , the levels of AID mRNA in the malaria tonsils were significantly higher , about 5 fold on average . However , although all the values for the malaria tonsil were above the range of the controls , the spread was large such that some samples had levels 8–13 fold higher than controls . The results for c-myc transcripts were less striking . Most of the malaria samples showed little or no significant difference from controls with the exception that two of the nine samples tested clearly had a markedly elevated level of c-myc mRNA . The level of c-myc mRNA did not correlate with the levels of AID mRNA . These results confirm our prediction that the levels of AID expression are higher in the tonsil GC B cells of individuals infected with malaria . To test the hypothesis that malaria is associated with higher levels of EBV infected cells in the GC , the GC cells from the same sets of tonsils described above were also analyzed for the presence of EBV . We recovered similar numbers of cells from both sets of tonsils and the fraction of B cells was also similar ( Figure 6A ) . The fraction of GC cells in the control tonsils was ∼32% ( Figure 6B ) , consistent with what we have seen in previous studies . However , the frequency in the malaria tonsils was unusually high ∼50% . When we analyzed the frequency of EBV infected cells in the two sets of tonsils , an even more dramatic difference was observed ( Figure 6C and Table 1– note the log-scale in the Figure ) , with the level being ∼50 fold higher on average in the malaria tonsils compared to controls ( log mean: 2 , 275/107 versus 51/107; median 2 , 580/107 versus 52/107 ) . Taking the increased numbers of GC cells into account , this means that the malaria tonsils have approximately 70 fold more EBV infected cells in their GCs . These results confirm our prediction that the levels of EBV infected GC B cells should be higher in the tonsils of individuals infected with malaria . As may be seen in Figure 6D , although both EBV infection and AID expression were elevated in all of the malaria tonsils there was no correlation between the levels of EBV infection and the level of AID in either malaria or control samples . We conclude that individuals infected with malaria have an increased level of EBV infected cells and AID mRNA expression in their tonsil GC cells , but these levels are not correlated . We have shown above that the level of c-myc transcripts is significantly elevated in GC cells from the tonsils of a subset of individuals with malaria . However , there has been controversy as to whether c-myc is actually expressed in the GC [34] , [35] , [36] , [37] . Therefore , to confirm that we were detecting c-myc expression in GC cells we performed several experiments as shown in Figure 7 . GC cells ( CD10+ tonsil B cells ) were positive for c-myc expression when stained with a c-myc specific antibody and analyzed by flow cytometry ( Figure 7A ) . The presence of c-myc protein in our GC cell preparations was confirmed by Western blot ( Figure 7B ) , where the signal was specifically blocked by the myc specific peptide used to raise the antibody . The specificity and correct location of the protein in the nucleus was confirmed by analysis with the ImageStream ( Figure 7C ) . We conclude , therefore , that our studies support the current opinion that c-myc is expressed in GC cells .
It is now more than 50 years since the association between P . falciparum malaria and eBL was first proposed [11] , [12] . Since that time , confirmation of a direct link and a mechanism to explain it has been lacking . Here we have presented studies on the effect of malaria in the context of the GC and provided evidence for a multifactorial effect of malaria that can account for the increase risk of eBL . These include the activation of AID expression , possible heightened c-myc transcription , increasing the numbers of B cells transiting the GC and increasing the fraction of these cells that are EBV infected . The common component linking these effects is the GC . The GC is the structure where immunoglobulin genes undergo somatic hypermutation and class switch recombination [38] , [39] , mediated by AID [10] , as the cells undergo affinity maturation . It is this enzyme that is responsible for causing the c-myc translocation characteristic of eBL [7] , [8] , [9] . However , the GC is also the site where newly infected EBV blasts undergo the transition to become resting latently infected memory B cells [14] , [16] , [22] . Therefore , in the presence of malaria the number of cells able to tolerate a c-myc translocation ( EBV infected ) in the GC are increased and the probability of a c-myc translocation is also increased ( AID activation ) . Thus , it is the increased probability of a fortuitous collision of AID and EBV in GC cells , both exacerbated by malaria that leads to eBL . A further observation confirming that these events occur within the GC was the finding that the P . falciparum extract alone had little or no ability to induce AID . Maximal induction required co-stimulation with CD40 ligand and IL-4 . This means that the malarial parasite can only work to induce AID expression if T cell help is also provided . Since T cell help is specifically provided in the GC [40] , [41] , this firmly places the role of P . falciparum in the induction of AID in the GC and would seem to rule out fortuitous activation elsewhere . This would explain why eBL is uniquely a tumor of GC cells . We observed no difference in the percentage of total B cells per tonsil between the malaria-endemic and the non-malaria regions , however , GC cells are ∼2 fold higher in the malaria background . This suggests that malaria ( or malaria background ) does not disrupt the size of the B cell pool but increases the likelihood that more B cells , and therefore EBV-infected cells , either enter the GC or that the cells stay longer in the GC . Either way , this increases the chances that some B cells , including EBV infected cells , will develop deleterious mutations and translocations . In individuals with malaria we detected a wide range in the number of GC cells containing EBV and in the degree of AID expression . It is tempting to speculate that those with the rare fortuitous coincidence of extremely high levels for both may be the likely candidates for tumor development . We have provided compelling in vitro evidence to support the claim that P . falciparum induces AID expression , the first such evidence . Importantly , the parasite induces AID in an antigen independent , i . e . deregulated fashion . Such expression is known to be a predisposing factor for the c-myc translocation . EBV infection of B cells in vitro , as well as EBV associated proteins , have also been shown to induce AID expression [42] , [43] , [44] . It is conceivable therefore that EBV and malaria could even cooperate in the induction of AID . We have shown that P . falciparum is capable of eliciting AID expression at least as effectively as the combination of CpG and BCR cross-linking , suggesting that the parasite can provide both signals . We have identified hemozoin , the crystalline by product of P . falciparum digestion of hemoglobin [30] and a reported TLR9 ligand [31] , [32] , as one of the parasite components responsible . Consistent with its signaling through TLR9 , hemozoin is only effective at inducing AID in the presence of BCR cross-linking . Thus , it is likely that there is a second , as yet unidentified , parasite derived ligand that provides the surrogate BCR signal . A likely candidate for this is the P . falciparum specific protein PfEMP1 encoded by the var gene family [45] , which has been shown previously to bind to and activate B cells through the BCR [46] . This would also explain the specific link of P . falciparum with eBL since only this species of malaria expresses PfEMP1 . The prediction is therefore , that a combination of hemozoin and PfEMP1 are providing the requisite TLR9 and sIg signals . Since TLR9 is a receptor for polynucleotides it was somewhat surprising that hemozoin was able to stimulate AID expression in the absence of parasite DNA . It is controversial as to whether hemozoin does [32] or does not [31] need or be associated with DNA in order to signal through TLR9 . It is possible that hemozoin may be signaling via a TLR9 independent mechanism in our system since it has been reported that it can signal through other pathways including the inflammasome [47] , [48] . However , there was a high death rate in our in vitro assays therefore it is possible that endogenous unmethylated DNA from the dead cells was being trafficked into the endosomal compartment by hemozoin to stimulate TLR9 and consequently lead to AID induction . Demonstration of an increased viral burden of EBV in individuals with P . falciparum malaria is not a new observation [25] , [26] , however in this paper we have shown a direct mechanistic consequence of this elevation that provides a risk factor for eBL namely an increased number of latently infected cells in the GC . This effect is not modest , with the mean frequency of EBV infected GC B cells from the malaria tonsils being about 50 fold more than the frequency of EBV infected GC B cells from the controls . Combined with the increase in the number of total GC cells this results on average in there being 70 times more EBV infected cells in the GCs of individuals with malaria compared to controls . We have shown previously that healthy individuals have at any time on average approximately 3 EBV infected cells per GC [49] . Our results here indicate that this would increase to around 150–200 per GC for an individual with malaria . This is a significant increase in risk since the consequence is an extremely high number of ( EBV infected ) cells each of which are primed to tolerate the c-myc translocation that emanates from overexpressed AID . The increased frequency of latently infected cells seen in the malaria samples ( ∼50 fold ) is very similar to what we have reported previously for immunosuppressed patients and patients with SLE [27] , [50] . Indeed , it has been reported previously that malaria is immunosuppressive for T cell responses [51] , including those directed against EBV [24] . It is tempting therefore to speculate that children with malaria are immunosuppressed and that this explains the risk for eBL . However , we have pointed out previously that immunosuppression is a risk factor for post-transplant lymphoproliferative disorder ( PTLD ) -like disease , i . e . immunoblastic lymphoma not eBL [14] , [22] . Thus , immunosuppression alone is not sufficient to explain eBL . The results presented here also provide further support for the GC model of EBV persistence . This model , which is now generally accepted , holds that EBV establishes a persistent infection by driving newly infected blasts through the GC to become latently infected resting memory B cells [14] , [16] , [22] . A direct prediction of this model is that a higher viral burden should produce a higher number of EBV infected cells transiting the GC and that prediction has been fulfilled in this study . Furthermore , the studies presented here provide a powerful functional confirmation of the model in that they provide an explanation for the link between malaria and eBL . Thus , the model predicts that it is the elevated rate of passage of virus infected cells through the GC , together with deregulated AID expression , that explains the origin of eBL . Our studies suggest that malaria may also induce heightened expression of c-myc in the GC , at least in some individuals . c-myc is a non-traditional transcription factor with a very complex regulation . High c-myc expression in the GC could account for the observed high rate of proliferation , as well as the high apoptotic tendency of GC B cells [34] , [37] . However , the question of whether c-myc is even expressed in the GC has been controversial in the past . Martinez-Valdez et al . [37] and Cutrona et al . [34] report that c-myc is highly expressed in GC B-cells , whereas Klein et al . [36] were unable to confirm these findings . Recently , Dominguez-Sola et al . [35] have presented compelling evidence that c-myc is expressed in GC cells , specifically by B cells selected for reentry into the dark zone , a conclusion supported by our work . Furthermore , for AID to target c-myc in GC B cells , c-myc must be expressed since AID deaminates transcribed substrates and acts on selected highly transcribed genes when they are over-expressed . Thus , higher than normal levels of c-myc transcription driven by malaria could further increase the risk that AID would target the c-myc gene for a translocation event . In conclusion , we have presented the first direct evidence for a mechanism to explain the link between eBL and holoendemic malaria . If correct , these observations imply that reducing the exposure to P . falciparum malaria or the development of drugs to block the ability of malaria to induce AID should dramatically reduce the incidence of eBL in young children in tropical Africa . Specifically this should act as a spur to agencies interested in reducing the incidence of exposure to P . falciparum in young children .
This study was approved by the Institutional Review Board of the Tufts Medical Center , Boston , USA and the Committee on Human Research and Ethical Publications of the School of Medical Sciences , Kwame Nkrumah University of Science and Technology ( KNUST ) , and Komfo Anokye Teaching Hospital ( KATH ) , Kumasi Ghana . The material used was deidentified , discarded tonsil tissue and was deemed exempt from informed consent by the IRB . The tonsil material was obtained indirectly either through the Pathology Department at Tufts Medical Center or the EENT Clinic at Komfo Anokye Teaching Hospital . The EBV-positive lymphoblastoid cell line IB4 ( gift of Elliott Kieff ) and Namalwa Burkitt's lymphoma cell line were used as positive controls for DNA PCR of the W-repeat region of the EBV genomes . The EBV-negative cell line CB60 , a mouse T-cell hybridoma cell line ( gift of Miguel Stadecker ) was used as a negative control in all W-PCR experiments . The Burkitt's lymphoma cell lines Raji and Rael were used as positive controls for c-myc western blot . The EBV negative BL2 Burkitt's lymphoma cell line and SP-IM 171 spontaneous EBV lymphoblastic cell line were used in hemozoin-DNA complex internalization assays . All cell lines were cultured at 37°C with 5% CO2 in RPMI 1640 supplemented with 10% fetal bovine serum , 2 mM glutamine , 2 mM sodium pyruvate , 100 IU of penicillin-streptomycin , and 10 µg/ml ciprofloxacin hydrochloride ( RPMI-complete ) . Palatine tonsils were obtained from patients 14 years or younger undergoing routine tonsillectomy . Twelve tonsil samples were obtained from patients at the EENT Clinic Komfo Anokye Teaching Hospital , Kumasi , Ghana . These were processed at the Kumasi Center for Collaborative Research in Tropical Medicine ( KCCR ) , Kumasi , Ghana , stored in liquid nitrogen and shipped to Tufts University School of Medicine , Boston , USA on dry ice ( under the supervision of Prof . Karen Duca , KNUST , Kumasi , Ghana ) . The presence of the parasite was confirmed based on detection of P . falciparum DNA and/or antigens ( see below ) . Kumasi is an area of holoendemic malaria therefore it was not surprising/unexpected that all patient samples received tested positive for the parasite . To obtain parasite free control samples we therefore also collected twenty one tonsils from a malaria negative region ( Boston MA ) through the Pathology Department at Tufts Medical Center , Boston MA , USA . The identical procedure and reagents were used for harvesting tonsils in Boston and Kumasi . The malaria infected individuals who provided the tonsils are genetically distinct from the controls and could be subject to a wider range of infection and lower level of general hygiene . To minimize this possible source of variation we collected the malaria tonsils from age and sex matched donors in an area of high socioeconomic status in Kumasi where economic and medical standards were comparable to those in Boston . Tonsil tissue was cut into very small pieces in ice-cold PBSA ( 1x PBS +0 . 5% BSA ) and then minced . Supernatants were pipetted through a cell strainer into 50 ml conical tubes to remove debris . Supernatants were centrifuged at 1 , 600 rpm at room temperature for 10 minutes and then aspirated . Pellets were re-suspended and brought to 50 ml with PBSA . About 25 ml of cells was carefully layered onto 20 mls of Ficoll-paque plus ( GE Healthcare Biosciences , Philadelphia , USA ) and then spun at 2 , 000 rpm for 30 minutes at room temperature ( with no brake ) . Mononuclear cells were collected from the interface ( buffy coat ) and cell pellet discarded . The volume of mononuclear cells was adjusted to 50 ml with PBSA ( and an aliquot taken for counting ) ; cells were then washed once by spinning at 1 , 500 rpm for 10 minutes . After counting , cells were frozen in fetal bovine serum ( FBS ) ( Sigma , St . Louis , USA ) plus 10% DMSO ( dimethyl sulfoxide ) at 1×108 cells/ml . Cells were aliquoted in cryotubes and kept on ice for about 5 minutes , stored at −80°C overnight and then transferred to liquid nitrogen for long term storage . The identical procedure and reagents were used for preparing tonsil cell suspensions in Boston and Kumasi . Tonsil mononuclear cells were thawed in medium ( RPMI/FBS ) , and spun down at 1 , 500 rpm for 5 minutes . B cells were either first purified using StemSep ( StemCell technologies , Vancouver , Canada ) according to manufacturer's instruction , or cells were re-suspended in 0 . 5% BSA in 1x PBS ( PBSA ) i . e . staining buffer . Cells were resuspended at 5×106 cells/100 µl PBSA either directly into FACs tubes or 15 ml tubes for staining . For extracellular staining the appropriate concentration of fluorochrome conjugated antibody was added to cells in appropriate tubes and incubated for 15 minutes at room temperature in the dark , after thorough mixing . Cells were washed once with PBSA , vortexed , and spun down at 1 , 500 rpm for 5 minutes . Finally cells were re-suspended in 300 µl PBSA and stored at 4°C until analyzed . For intracellular staining tonsil cells were pelleted , washed in Dulbecco's PBS− ( without calcium or magnesium ) , and fixed either with 4% formaldehyde or with BD Cytofix/Cytoperm fixation and permeabilization solution ( BD Biosciences , San Jose , USA ) and then incubated for 20 minutes at room temperature . Cells were then washed twice either with 1x BD wash/perm buffer ( BD Biosciences , San Jose , USA ) or with 0 . 04% saponin based wash buffer and spun down at 1 , 500 rpm for 5 minutes . Permeabilization buffer ( 0 . 5% saponin based buffer ) was used to re-suspend cells at 5×106 cells/100 µl . Two microliters ( 2 µl ) of normal human serum ( 60 mg/ml , Thermo Fisher Scientific Rockford , USA ) was added in order to block against non-specific antibody binding , and cells were incubated for 30 minutes at room temperature . Primary antibodies for intracellular antigens were added and incubated at room temperature for 30 minutes . Cells were then washed once and re-suspended in 0 . 5% saponin based buffer , 2 µl of normal human serum was again added and incubated for 30 minutes at room temperature . Fluorochrome conjugated secondary antibodies were then added at appropriate dilutions and incubated for 15 minutes in the dark . Cells were washed once with wash buffer , spun down at 1 , 500 rpm for 5 minutes and stored in 300 µl PBSA at 4°C until analysis . Cell sorting was performed on a MoFLo or Influx cell sorter and analysis on a FACSCalibur or Image Stream at Tufts University laser cytometry core . Sorted populations were >90% pure . A list of the antibodies and fluors used in this study is given in Table S1 . The presence of the parasite was confirmed based on detection of P . falciparum DNA and/or antigens . Nested , parasite specific DNA PCR was performed for a sequence in the 2nd exon of the Chloroquine transporter ( Pfcrt ) gene as follows: First round ( 94°C , 3 min; 94°C , 30 sec; 56°C , 30 sec; 60°C , 1 min 30 cycles , 72°C , 3 min ) forward primer CCGTTAATAATAAATACACGCAG , reverse CGGATGTTACAAAACTATAGTTACC ( 95°C , 5 min; 92°C , 30 sec; 48°C , 30 sec; 65°C , 30 sec ( 25 cycles ) ; 72°C , 3 min ) forward primer TGTGCTCATGTGTTTAAACTT reverse ACAAAATTGGTAACTATAGTTTTG . PCR product sizes were verified on 2% agarose gels . 1st round amplicon 527 bp , 2nd round 145 bp . P . falciparum antigens were detected employing a rapid diagnostic test ( RDT ) cassette ( ACON Laboratories , Inc . , San Diego , USA ) as directed by the manufacturer . For limiting dilution analysis , GC B cell ( CD19+CD10+ ) populations were isolated by flow cytometry and usually 10 replicates each of 2×105 , 1×105 , 5×104 , 2 . 5×104 , 1 . 25×104 , 0 . 625×104 , 0 . 3125×104 ( higher or lower dilutions were added as needed ) were placed in a 96 V bottom plate for subsequent EBV W-repeat DNA PCR . The plate was spun at 1 , 200 rpm for 10 minutes at 4°C and the supernatant aspirated . To each well was then added 20 µl of digestion mix ( 10X PCR buffer , 100 µl; Igepal ( NP-40 ) 100 µl; Tween-20 100 µl; Proteinase K 50 µl of 20 mg/ml stock and 650 µl of water ) . The plate was sealed air-tight and incubated at 55°C overnight . This was followed by proteinase K deactivation at 95°C for 10 minutes . Ten microliters ( 10 µl ) of water was then added to all the sample wells . The fraction of EBV positive wells was then assessed by W-repeat EBV DNA PCR for each replicate and the frequency of EBV infected cells in the starting population calculated using Poison statistics . The EBV-positive cell lines IB4 and Namalwa were used as positive controls and the EBV-negative cell line CB60 was used as a negative control . DNA real-time PCR specific for the W-repeat sequence of the EBV genome was performed as described [52] . For each reaction , a master mix was prepared , containing 12 . 5 µl IQ Supermix ( Biorad cat 170-8862 ) , 2 . 5 µl of 900 nM primers and 2 . 5 µl of 250 nM fluorogenic probe . Five microliters ( 5 µl ) of DNA was added to 20 µl of master mix with a final reaction volume of 25 µl . ( See Table S2 for primer and probe sequences ) . The PCR reactions were performed on a Biorad iCycler . The protocol was as follows: Step 1 ( 1 cycle ) : 3′ at 95°C; Step 2 ( 50 cycles ) : 15″ at 95°C , 1′ at 60°C . RNA was purified by TRIzol extraction ( Invitrogen , Life Technologies , Grand Island , USA ) and then treated with TURBO DNase ( Ambion , Life Technologies , Grand Island , USA ) to eliminate DNA prior to RNA amplification ( where necessary ) . cDNA was made from the RNA using a cDNA synthesis kit ( Invitrogen iScript cDNA synthesis kit ) . For the cDNA synthesis reaction , a master mix was prepared which included 4 µl of 5X iScript reaction mix , 1 µl of iScript reverse transcriptase , and 8 µl of nuclease-free water . Seven microliters ( 7 µl ) of purified RNA was added to 13 µl of master mix . All reactions were performed on an Applied Biosystems PCR machine ( Thermal cycler ) . The protocol was as follows: one cycle that included 5 minutes at 25°C , 30 minutes at 42°C , and 5 minutes at 85°C . For real time PCR a master mix was prepared , containing 12 . 5 µl of IQ Supermix ( Bio-Rad ) , 2 . 5 µl of 900 nM primers , and 2 . 5 µl of 250 nM fluorogenic probe , except when Taqman pre-developed assays were being used in which case , 12 . 5 µl Supermix , 1 . 25 µl 20x primer-probe mix and 6 . 25 µl water . Five microliters ( 5 µl ) of cDNA was added to 20 µl of master mix with a final reaction volume of 25 µl . All real time PCRs were performed on a Bio-Rad iCycler . The protocol was as follows: step 1 , one cycle of 3 minutes at 95°C; step 2 , 55 cycles of 15 seconds at 95°C and 1 minute at 60°C . All our real time PCR assays were optimized to detect down to the single cell level ( See Supplemental Information for a list of primers and probes ) . P . falciparum parasites ( 3D7 line ) were cultured using standard procedures as described [53] . Parasites were grown at 5% hematocrit in RPMI 1640 medium , 0 . 5% AlbuMAX II ( Invitrogen ) , 0 . 25% sodium bicarbonate , and 0 . 1 mg/ml gentamicin . Cultures were incubated at 37°C in an atmosphere of 5% oxygen , 5% carbon dioxide , and 90% nitrogen . Parasite extracts were prepared by selective lysis of the host RBC membranes through the addition of saponin . Infected RBCs were suspended in 0 . 01% saponin in PBS and incubated at room temperature for 5 min . Host cell free parasites were pelleted by centrifugation , washed twice with PBS and stored frozen at −80 degrees C . Parasite extracts were prepared by 3 freeze-thaw cycles of frozen parasites , sonicated and stored at −80 degrees Celsius until needed . The protein concentration was determined by the Bicinchoninic assay ( BCA assay – see below ) . Fresh tonsil mononuclear cells ( MNC ) were isolated as described above and re-suspended at ∼1×106 cells/ml in pre-warmed PBSA . Two microliters of 5 mM CFSE ( carboxyfluorescien diacetate , succinimidyl ester ) in DMSO ( Invitrogen , Life Technologies , Grand Island , USA ) was added per ml of cells ( final concentration 10 µM CFSE ) . The cells were then incubated for 10 minutes at 37°C . Cells were spun down and re-suspended in 2 volumes of ice-cold culture medium and incubated on ice for 5 minutes . Cells were re-pelleted and re-suspended in 2 volumes ice cold culture medium ( 2 times ) . Washing was done one more time in 2 volumes of pre-warmed medium and cells were finally re-suspended in fresh pre-warmed cell culture medium ( described above ) . A final concentration of 3 µM of CpG-2006 ( Hycult Biotech , Plymouth Meeting , USA ) , 0 . 25 µg/ml CD40 ligand ( eBioscience , San Diego , USA ) , 5 ng/ml IL-4 ( eBioscience , San Diego , USA ) , 2 . 5 or 5 µg/ml Anti-human IgG+IgM , ( Jackson ImmunoResearch , West Grove , USA ) 10 µg/ml of crude parasite extract , and hemozoin ( InvivoGen , San Diego , USA ) 50 µg/ml were used in the tonsil mononuclear cell stimulation . Cells were harvested on days 3 , 5 , 7 and 10 , stained with CD19 and then B cells were sorted with the MoFlo . Five microgram of human CpG-2006 DNA with a phosphorothionate ( PTO ) backbone bound to Alexa-488 ( Integrated DNA Technologies , Coralville , USA ) were mixed with 100 µg/ml of sonicated synthetic hemozoin [32]and co-incubated with rocking for 2 h followed by washing of the complex three times with PBS . Bound and unbound DNA were determined by measuring the DNA concentration in the collected supernatant , using the Nanodrop . BL2 and IM 171 cells were each plated at 1 ml ( with approximately 500 , 000 cells per confocal plate ) . The CpG-Alexa-488-hemozoin complex was added to the cell lines BL2 and SP-IM 171 , and incubated for 2 hours after which confocal microscopy was carried out . Confocal reflection microscopy for detection of hemozoin was combined with fluorescence microscopy to detect the Alexa-488 tagged CpG on a Leica SP2 AOBS confocal laser-scanning microscope as described in detail in [54] . Cells were spun at 1 , 500 rpm for 5 minutes and the supernatant was aspirated . 5×106 cells were re-suspended in 100 µl of RIPA buffer ( Sigma , St . Louis , USA ) with freshly added protease inhibitors ( Thermo Scientific , Rochford , USA ) . The sample was pipetted up and down to dislodge cell clumps , and then vortexed vigorously for 15 seconds then incubated on ice for ∼7 minutes , vortexed again for 15 seconds , spun at 13 , 200 rpm for 16 minutes at 4°C and the supernatant transferred to a new eppendorf . Protein concentrations were determined with either the Bio-Rad protein assay reagent ( Bio-Rad , Hercules , USA ) or the bicinchoninic acid ( BCA ) protein assay reagent mix ( Thermo Scientific , Rockford , USA ) , according to manufacturer's instruction . To each protein sample was added 12 . 5% beta mercaptoethanol and 1× SDS sample buffer ( Boston Bioproducts , Ashland , USA ) . Five microliters of protein standard ( Biorad , Hercules , USA ) was added to designated well ( s ) . Samples were heated at 95°C for 5 minutes and resolved on 4–20% Tris-Glycine gels ( Invitrogen , Life Technologies , Grand Island , USA ) . Immobilon-P polyvinylidene fluoride ( PVDF ) transfer membrane ( Millipore , Billerica , USA ) was used in a semi-dry electrophoretic transfer . The membrane was blocked with either membrane blocking solution ( Invitrogen , Life Technologies , Grand Island , USA ) or 5% milk in 1x Tris buffered saline with Tween 20 ( TBST ) ( Cell signaling , Danvers , USA ) at room temperature for 1 hr . The PVDF membrane was then incubated with primary antibody in appropriate blocking buffer at 4°C overnight . The membrane was washed in TBST or Invitrogen wash solution three times , 10 minutes each and incubated with secondary antibody in 5% milk in TBST for 45 minutes at room temperature . The membrane was then washed with Invitrogen wash solution 4 times , 10 minutes each at room temperature and then enhanced chemiluminescence ( ECL ) reagent SuperSignal west femto maximum sensitivity substrate ( Thermo Fisher scientific , Rockford , USA ) was applied to the membrane . Hyblot CL autoradiography film ( Denville scientific , Metuchen , USA ) was then exposed to the membrane and developed in a Kodak X-MAT 2000 processor . Data are expressed as mean ± SD . Differences between groups were analyzed for statistical significance with Student two-tailed , unpaired t test . Significance was considered achieved when the p value was <0 . 05 . | Endemic Burkitt's lymphoma ( eBL ) is a common tumor of young children in tropical Africa that is closely linked geographically with P . falciparum malaria . This association was noted more than 50 years ago . Since then we have learned that eBL contains the oncogenic herpes virus Epstein-Barr virus and a defining translocation that activates the c-myc oncogene . However the link to malaria has never been explained . Here we show that malaria has multiple effects that all focus on germinal center ( GC ) B cells that are known to be the origin of eBL . Together these effects of malaria act synergistically to dramatically increase the risk of developing eBL in individuals infected with the parasite . Clinical interventions that lessen the impact of malaria on GC B cells should dramatically decrease the incidence eBL . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"oncology",
"infectious",
"diseases",
"medicine",
"and",
"health",
"sciences",
"protozoans",
"malarial",
"parasites",
"epstein-barr",
"virus",
"infectious",
"mononucleosis",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"plasmodium",
"falciparum",
"parasitic"... | 2014 | A Multifactorial Role for P. falciparum Malaria in Endemic Burkitt's Lymphoma Pathogenesis |
Dengue fever is the most important arboviral disease in the tropical and sub-tropical countries of the world . Delhi , the metropolitan capital state of India , has reported many dengue outbreaks , with the last outbreak occurring in 2013 . We have recently reported predominance of dengue virus serotype 2 during 2011–2014 in Delhi . In the present study , we report molecular characterization and evolutionary analysis of dengue serotype 2 viruses which were detected in 2011–2014 in Delhi . Envelope genes of 42 DENV-2 strains were sequenced in the study . All DENV-2 strains grouped within the Cosmopolitan genotype and further clustered into three lineages; Lineage I , II and III . Lineage III replaced lineage I during dengue fever outbreak of 2013 . Further , a novel mutation Thr404Ile was detected in the stem region of the envelope protein of a single DENV-2 strain in 2014 . Nucleotide substitution rate and time to the most recent common ancestor were determined by molecular clock analysis using Bayesian methods . A change in effective population size of Indian DENV-2 viruses was investigated through Bayesian skyline plot . The study will be a vital road map for investigation of epidemiology and evolutionary pattern of dengue viruses in India .
Dengue fever ( DF ) is the most prevalent arthropod-borne disease in the tropical and sub-tropical regions of the world . It is estimated that 2 . 5 billion people are at risk of dengue infection globally , 50 million dengue virus infections occur annually and 500 , 000 people with Dengue haemorragic fever ( DHF ) require hospitalization every year [1] . Dengue virus ( Family Flaviviridae , Genus Flavivirus ) is a small enveloped RNA virus carrying a single stranded , positive-sense RNA genome ( ~10 . 6 kb ) . The genome encodes three structural and seven non-structural proteins in a single open reading frame . The structural proteins are capsid ( C ) , precursor membrane ( prM ) , and envelope ( E ) proteins . The envelope glycoprotein is essential for viral entry into the cell . Non-structural proteins are involved in viral replication within the cell , named NS1 , NS2A , NS2B , NS3 , NS4A , NS4B , and NS5 . Dengue virus ( DENV ) is transmitted primarily by Aedes aegypti mosquitoes . Common symptoms of DF are fever , fatigue , rash , headache , retro-ocular pain , arthralgia , myalgia , nausea , vomiting and low platelet count . While most infections result in asymptomatic response or mild febrile illness ( DF ) , a small percentage of cases result in the more severe and potentially fatal dengue with warning signs and severe dengue which are characterized by plasma leakage [2 , 3] . There are four antigenically distinct , closely related serotypes of the dengue virus ( DENV1–4 ) , exhibiting a 65–70% sequence homology [2] . A vaccine for dengue is not available yet because of its 4 serotypes which enhance the risk of severe disease by antibody dependent enhancement of infection [4] . Dengue fever outbreaks occur after every 3–4 years in Delhi , India . Recently , dengue fever outbreaks were reported in Delhi in 2003 [5] , 2006[6] , 2010 [7] and 2013[8] . We have studied dengue prevalence and serotypic distribution in the period 2011–2014 in Delhi [9] . We detected three dengue serotypes ( DENV-1 , 2 and 3 ) in circulation and predominance of DENV-2 in Delhi in the above mentioned study . The present study focuses on the molecular characterization and evolutionary analysis of only DENV-2 strains detected in the study . The same analysis for DENV-1 and 3 has been published elsewhere [9] .
The study was granted approval by Institutional Ethics Committee , Jamia Millia Islamia and was done in accordance with the World Medical Association Declaration of Helsinki . Written informed consent in English or Hindi was obtained from all adult subjects or a parent or guardian in case of minors . Acute phase blood samples were collected from suspected dengue patients from Dr . M . A . Ansari Health Centre , Jamia Millia Islamia , New Delhi . Sera were separated from the blood samples by centrifuging at 3000 rpm for 10 minutes at 4°C . Serum samples were stored at -80°C until further use . RNA was extracted from 140 μL serum samples using QIAamp Viral RNA Mini kit ( Qiagen , Hilden , Germany ) according to the manufacturer’s instructions . cDNA synthesis was carried out in a 25 μL reverse transcription ( RT ) reaction mixture using 20 ng of random primers ( Promega , USA ) , 1mM dNTPs ( Promega , USA ) , 8U of rRNAsin ( Promega , USA ) , 10U of Avian Myeloblastosis Virus reverse transcriptase ( Bangalore Genei , India ) and 13 . 75 μL RNA . Detection of dengue viruses were carried out by semi nested RT-PCR method as reported previously [10] with some modifications [8 , 11] . The E gene ( 574bp ) of DENV-2 viruses was amplified using another set of published primers for DNA sequencing [12] . Amplicons were run on agarose gel and the bands were cut . Gel extraction was done using QIAquick Gel Extraction Kit ( Qiagen , Germany ) as per the manufacturer’s instructions . Sequencing with both forward and reverse primers was done commercially ( Xcelris Labs , Ahmedabad , India ) . The raw sequences were subjected to similarity search using BLAST . Forward and reverse sequences were aligned and manually edited with GeneDoc ( v2 . 7 . 000 ) software . Sequences were aligned with other published sequences available in GenBank using Clustal W implemented in BioEdit ( v7 . 0 . 9 . 0 ) . The best fit substitution model for the data was determined by Akaike Information Criterion ( AIC ) using MODELTEST 3 . 7 [13] . Phylogenetic tree was constructed using Maximum likelihood method in Mega 6 . 06 software [14] . Statistical support for the nodes was assessed by bootstrapping with 1000 replicates . Rate of nucleotide substitution and Time to the most recent common ancestor ( TMRCA ) of DENV-2 strains were assessed using Bayesian inferences implemented in BEASTv1 . 8 . 1 [15] . TN93+G+I model was used as the nucleotide substitution model and among site rate variation model ( selected by MODELTEST3 . 7 ) . Both strict and relaxed ( uncorrelated exponential and uncorrelated lognormal ) molecular clocks [16] were used for the analysis . Constant size and Bayesian skyline coalescent tree prior were used in the study . The MCMC chain was run for 30 , 000 , 000 steps . The parameter values were sampled at every 3000 steps . Each analysis was performed in two separate runs and the resulting log files were combined using LogCombiner 1 . 8 . 1 ( implemented in BEAST ) with 10% burn-ins removed from each run . The resulting log files were analysed in the program Tracer 1 . 6 to ascertain convergence of the chain and to ensure that effective sample size of >200 for all parameters have been reached . The uncertainty in the parameter estimates were assessed by 95% HPD intervals . Model comparison was done by Bayes factor ( Log Marginal Likelihood ( M1 ) -Log Marginal Likelihood ( M2 ) ) . The maximum clade credibility tree was generated by Tree Annotator 1 . 8 . 1 ( available in BEAST ) , and the resulting tree file was visualized in the program FigTree 1 . 4 . 2 . Support for the node on the tree was ascertained by the Bayesian posterior probability ( BPP ) values for each node . The BEAST package was also used to infer Bayesian skyline plots for Indian cosmopolitan DENV-2 strains ( n = 24 , group size = 10 ) . This analysis enabled a graphical depiction of changing levels of relative genetic diversity ( Neτ , where Ne is the effective population size and τ the host-to-host generation time ) through time . The selection pressure acting on envelope gene codons of DENV-2 was investigated using the online facility at the web server http://www . datamonkey . org . Two datasets were used for this analysis , one dataset comprising strains from all DENV-2 genotypes ( n = 58 ) and second dataset comprising of only the Cosmopolitan genotype ( n = 30 ) strains . TN 93 model of nucleotide substitution and Neighbour Joining tree was used for the analysis . The ω ratios ( dN/dS ) were calculated using three likelihood approaches , single-likelihood ancestor counting ( SLAC ) , fixed effects likelihood ( FEL ) & Random effects likelihood ( REL ) . Sites showing evidence of positive selection by at least two of the methods with high statistical significance ( P < 0 . 1 or Bayes factor >50 ) , were considered to be under positive selection . T-cell epitopes of the envelope gene of prototype dengue 2 virus were predicted using the EpiJen online server [17] . Envelope protein sequence of DENV-2 prototype strain ( NGC 44 strain; GenBank Accession Number: AF038403 ) was submitted for epitope prediction . Appropriate proteasomal and Tap cut-off values were selected and epitopes were predicted for 18 different HLA alleles . B cell epitopes were predicted using the BCPreds prediction tool [18] . The predicted B-cell epitopes were further analyzed by VaxiJen server ( antigenicity prediction server based on auto cross variance ( ACC ) transformation of protein sequences into uniform vectors of principal amino acid properties ) . The B cell epitope having BCPreds score of >0 . 8 and VaxiJen score of >0 . 6 were selected as the predicted B- cell epitope . The structure for the partial envelope gene sequence ( 152 amino acid ) obtained in the study was modeled using I-TASSER protein structure prediction tool . The structure was visualized in Pymol and amino acid mutations ( E322 and E404 ) were mapped on the structure using Pymol . The significant number of samples to be sequenced for each lineage ( I and III ) was calculated by taking into account the total number of samples year wise with the Epi-Info software ( http://wwwn . cdc . gov/epiinfo/ ) using the epidemiologic calculator ( StatCalc ) . The statistical calculation of the Population Survey was done using the Sample size and power option in the Epi-Info software using number of samples collected during the study period ( for the years 2012 , 2013 and 2014 ) and the expected frequency of Dengue virus infection at 5% level of significance . In addition , the analysis of lineage replacement with respect to total cases taken year wise was done with chi-square test . A p-value less than 0 . 05 was considered significant .
Uncorrelated relaxed lognormal clock and Bayesian coalescent tree prior was chosen as the best fit model as they were favoured by Bayes factor ( S1 Table ) . Nucleotide substitution rate was detected to be 7 . 7 ×10−4 substitutions per site per year ( 95% HPD; 5×10−4 to 1 . 052 ×10−3 ) under the best fit model . Root of the tree was calculated to be 128 years old [86–194 years 95% HPD , 1886 ( 1820–1928 ) ] . TMRCA of the Cosmopolitan genotype was determined to be 59 years [45–81 years 95% HPD , 1955 ( 1933–1969 ) ] . TMRCA of the Indian Cosmopolitan genotype strains was determined to be 48 years [41–58 years 95% HPD; 1966 ( 1956–1973 ) ] . The Maximum Clade Credibility Tree was constructed with the best model as shown in Fig 3 . The tree showed three lineages of dengue 2 viruses in the period 2011 to 2014 . TMRCA of Lineage I was estimated to be 30 years ( 1984 ) , that of lineage II as 10 years ( 2004 ) and that of lineage III as 7 years ( 2007 ) . Bayesian skyline plot ( BSP ) for Indian Cosmopolitan strains was constructed in the study ( Fig 4 ) . The plot shows changes in the median estimate of relative genetic diversity ( Ne τ ) of the virus with time where Neτ is the product of effective population size ( Ne ) and generation time ( τ ) . The plot also shows 95% highest probability density intervals which represents both phylogenetic and coalescent uncertainty . Population size of Indian Cosmopolitan strains increased smoothly from 1974 to 2004 with a small growth rate ( Fig 4 ) . Population size remained almost constant in the period 2005 to 2009 . It decreased between 2009 & 2012 and then remained constant between 2012–2014 . Levels of Neτ in 2012–2014 were lower than in the start of the plot . The mean dN/dS ratio was found to be 0 . 05 in the codons of the E gene . There was no codon detected under positive pressure by the SLAC and FEL methods for the datasets . In the dataset comprising of all genotype strains ( N = 58 ) , three codons were found to be under positive selection pressure by the REL method i . e . E317 , E337 & E365 . The E322 site showing Valine in Lineage I and Isoleucine in Lineage III was detected as negatively selected by SLAC & REL method and as neutrally evolving site by FEL method in the dataset containing strains from all the genotypes . Site E404 was shown to be neutrally evolving by SLAC &FEL methods and as negatively selected by the REL method . In the second dataset which included strains of only the Cosmopolitan genotype strains ( n = 30 ) , no codon was detected to be under positive selection pressure by SLAC , FEL & REL method . Site 322 & 404 were detected to be neutrally evolving by the SLAC & FEL methods . Site 322 was detected to be negatively selected and site 404 as neutrally evolving by the REL method .
Delhi is a metropolitan city of India which battles dengue outbreaks after every 3–4 years . Recently we have reported a change in dengue serotype predominance in Delhi [9] . Dengue serotype 3 predominated in 2003–2006 , dengue 1 in 2008 and 2010 and dengue 2 in 2011–2014 in Delhi [9] . Dengue virus type 2 ( DENV-2 ) has been responsible for the 1996 [20] and 2013 [8] dengue fever outbreaks in Delhi , India . In the present study DENV-2 strains detected in Delhi during 2011–2014 were analyzed phylogenetically . All the DENV-2 strains sequenced in the study clustered in the Cosmopolitan genotype . This genotype has been reported from India since 1974 [19 , 21] . Earlier Indian study based upon E-NS1 gene sequencing of Delhi strains detected during 1957 , 1967 and 1996 [22] reported that before the Cosmopolitan genotype , American genotype was found to be in circulation in India . This was later confirmed by CprM gene sequencing of the same strains [23] . Kumar and colleagues [19] reported that Indian strains from 1956 to 1980 belonged to the American genotype and strains from 1974 onwards clustered in the Cosmopolitan genotype . Dash and co-workers [24] have also reported change in circulating genotype in India based upon complete genomic sequencing . Lineage replacement in dengue viruses has been reported in many studies [25–33] including India . A lineage replacement event where lineage III replaced lineage I was detected in 2012–2014 in Delhi . The statistical analysis revealed that the lineage replacement event was independent of the total number of samples analyzed year wise . We detected lineage replacement in 2013 in which a dengue outbreak occurred in Delhi [8] . It has been suggested that lineage replacement could result in increased transmission of the viral infection [32] . Thus , the lineage replacement event described in the present investigation could be related to the outbreak situation in Delhi in 2013 . Lineage replacement has been also postulated to result in improved viral fitness & increased transmissibility [31 , 32] or as a stochastic event due to virus population bottleneck effects [30] . Lineage I strains showed Valine ( V ) at E322 and Lineage III showed Isoleucine ( I ) at E322 . DENV-1 strains and DENV-4 strains have Valine at position 322 while DENV-3 strains have Isoleucine . Difference of V/I between strains might be helpful in escape of DENV-2 strains from cross neutralizing antibodies generated by DENV-1 , 3 and 4 strains which target E322 [34] . Besides this difference at E322 , the concerned lineages might also differ at other significant positions in the genomic regions not sequenced in the present study . A novel mutation Thr404Ile was also detected in the study . This residue is part of the stem region ( helix I ) of the E protein which is essential for the formation of E protein trimers in response to low pH [35] . Site 404 is variable amongst flaviviruses showing Lysine in Tick-Borne Encephalitis Virus , Alanine in DENV-1 and 3 and Serine in DENV-4 [36] . In the strain DL/DENV-2/8/14 showing this mutation , Threonine a polar amino acid was replaced by Isoleucine which is non polar but retains the property of having a chiral carbon in the side chain . This change in the type of amino acid may have impact on the structure of the E protein . Role of this mutation in the virus life cycle is not clear at present and should be explored in the site directed mutagenesis studies . The mutations at E322 and E404 were found in domain III and stem region respectively . These mutations may alter local secondary structure of the E protein which might affect the structure-function relationship of the protein . The mutation at E322 might influence attachment and fusion properties of the virus as domain III is the putative receptor binding domain of the E protein [37] . Mutation at E404 may influence the low pH induced fusion process of the E protein . Further these 2 sites were also mapped to predicted T-cell and B- cell epitopes . As a consequence the mutations I322V and T404I may influence immunogenicity of the dengue virus . Nucleotide substitution rate of DENV-2 viruses calculated in the study ( 7 . 7 ×10−4 substitutions per site per year ) is comparable to that reported in the earlier studies on complete E gene [38 , 39] and complete genome [40] . Kumar and colleagues [19] reported nucleotide substitution rate of DENV-2 viruses as 6 . 5×10−4 ( 4 . 1–8 . 7 ×104 , 95%HPD ) substitution per site per year . Our estimate is within the 95% HPD interval reported by them . Estimates for the TMRCAs of DENV-2 strains , Cosmopolitan genotype strains and Indian Cosmopolitan strains ( 128 , 58 & 48 years ) differ from those reported in the earlier Indian study [19] but they are within the 95% HPD intervals reported in the study and are supported by smaller 95% HPD intervals . Lineage I was estimated to have TMRCA of 30 years and Lineage III showed TMRCA of 7 years . The newly emerged lineage III replaced lineage I in 2013 and 2014 . Bayesian Skyline Plot of Indian Cosmopolitan DENV-2 strains have not been previously reported in literature . We show a decrease in effective population size of DENV-2 in recent years . Thus in the present study sequencing and phylogenetic analysis of DENV-2 strains revealed that the circulating strains belonged to the Cosmopolitan genotype . A lineage replacement event and a novel mutation were also detected . Nucleotide substitution rate and TMRCA of DENV-2 viruses are also reported . Past population dynamics as studied by Bayesian skyline plots showed a decrease in the population size of DENV-2 strains in recent years . The present investigation will assist in design of effective vaccine and antivirals for control of the virus . Further , it will also help us to determine the epidemiological and evolutionary pattern of Dengue outbreaks in India . | Dengue is a mosquito borne disease prevalent in more than 100 tropical and sub-tropical countries . In this study , we carried out phylogenetic and molecular clock analysis of dengue serotype 2 strains found to be circulating in Delhi , India , in 2011–2014 . All the study strains were found to belong to the Cosmopolitan genotype of dengue 2 viruses . Further we also detected a lineage replacement event in the 2013 dengue fever outbreak . We identified a novel mutation in the stem region of the Envelope protein . We also determined nucleotide substitution rate and time to the most recent common ancestor of dengue 2 viruses . Further a change in the population size of dengue 2 viruses in India as depicted by Bayesian Skyline plot is reported . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"sequencing",
"techniques",
"dengue",
"virus",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"chemical",
"compounds",
"aliphatic",
"amino",
"acids",
"pathogens",
"tropical",
"diseases",
"microbiology",
"organic",
"compounds",
"vi... | 2016 | Evolutionary Analysis of Dengue Serotype 2 Viruses Using Phylogenetic and Bayesian Methods from New Delhi, India |
With relatively few known specific transcription factors to control the abundance of specific mRNAs , Plasmodium parasites may rely more on the regulation of transcript stability and turnover to provide sufficient gene regulation . Plasmodium transmission stages impose translational repression on specific transcripts in part to accomplish this . However , few proteins are known to participate in this process , and those that are characterized primarily affect female gametocytes . We have identified and characterized Plasmodium yoelii ( Py ) CCR4-1 , a putative deadenylase , which plays a role in the development and activation of male gametocytes , regulates the abundance of specific mRNAs in gametocytes , and ultimately increases the efficiency of host-to-vector transmission . We find that when pyccr4-1 is deleted or its protein made catalytically inactive , there is a loss in the initial coordination of male gametocyte maturation and a reduction of parasite infectivity of the mosquito . Expression of only the N-terminal CAF1 domain of the essential CAF1 deadenylase leads to a similar phenotype . Comparative RNA-seq revealed that PyCCR4-1 affects transcripts important for transmission-related functions that are associated with male or female gametocytes , some of which directly associate with the immunoprecipitated complex . Finally , circular RT-PCR of one of the bound , dysregulated transcripts showed that deletion of the pyccr4-1 gene does not result in gross changes to its UTR or poly ( A ) tail length . We conclude that the two putative deadenylases of the CAF1/CCR4/NOT complex play critical and intertwined roles in gametocyte maturation and transmission .
Malaria remains one of the great global health problems today , with 216 million new infections and 445 , 000 deaths attributed to it annually [1] . Resistance to frontline drugs is spreading , and understanding the development and transmission of the malaria parasite is important to bolster efforts to reduce or eliminate deaths due to this infection . For the parasite to transmit from a vertebrate host to the mosquito vector , a small percentage of the cells will differentiate from asexual forms and develop into sexual stage gametocytes , which can persist in an infectious state until a mosquito takes a blood meal . This event allows a small number of gametocytes to be taken up into the mosquito , but with far fewer parasites productively infecting it [2] . Following two weeks of development within the mosquito , a small number of sporozoites will similarly be injected into a host by the mosquito as it takes another blood meal [3] . In the effort to develop vaccines and drugs , transmission events have been identified as prime targets to prevent the spread of the parasite because they are population bottlenecks in the parasite life cycle . We and others have focused upon the transmitted gametocyte and sporozoite stages of Plasmodium parasites to identify and exploit their weaknesses . In both cases , very few parasites are transmitted , and thus these bottlenecks are excellent points of intervention . The identification of molecular processes that are important for the transmission of the parasite in one or both of these events , and their modes-of-action , are thus top priorities for the development of new therapeutics . Recent work has shown that the parasite requires tight transcriptional and translational control to navigate its complex transmission events [4–8] . Despite these strict requirements for the effective transmission of the parasite , Plasmodium has only one known , expanded family of specific transcription factors , the ApiAP2 proteins ( 27 members , reviewed in [9] ) . However , Plasmodium expresses a larger number of RNA-binding proteins than other eukaryotes ( ~10% of its proteome ) , which likely provides expanded capabilities to regulate gene expression post-transcriptionally to control translation , and to stabilize or degrade RNA [5 , 10 , 11] . For instance , Plasmodium uses RNA-binding proteins such as DOZI ( an orthologue of the human DDX6 RNA helicase ) , CITH ( an orthologue of Lsm14A ) , and ALBA family proteins to impose translational repression and mRNA stabilization on transcripts in female gametocytes that are important for the establishment of a new infection of a mosquito [10 , 12–16] . A current model invokes these controls as a means for the parasite to always be ready to respond to external stimuli that indicate that transmission has occurred , and thus enables the “just-in-time” translation of the preserved mRNAs and establishment of the new infection [17] . Moreover , this molecular process is essential to the transmission of the parasite , as deletion of dozi or cith results in a complete arrest of development in early mosquito stage [14 , 15] . Similar regulatory events occur in the other transmitted stage ( sporozoites ) via the PUF2 RNA-binding protein , as deletion of puf2 results in the gradual loss of infectivity and subsequent premature dedifferentiation into a liver stage-like form while in the salivary gland [6–8] . As in model eukaryotes ( e . g . yeast , flies , worms ) and in humans , many of these regulatory functions of RNA metabolism occur in cytosolic granules within the parasite as well [14 , 15 , 18 , 19] . In addition to transcript stabilization , translational control can also be accomplished by the degradation of transcripts . Degradation of mRNAs is typically initiated by deadenylases , which remove the protective poly ( A ) tail . Historically , it was concluded that shortening the poly ( A ) tail to a critical length in turn promotes the subsequent decapping and complete degradation of the transcript by other factors [20 , 21] . However , recent work has shown that mRNAs with short poly ( A ) tails can remain stable and are actively translated [22] . In many eukaryotes , the main complex responsible for deadenylation is the CAF1/CCR4/NOT complex , which also participates in transcriptional elongation , translational repression , and histone modification functions , and thus acts broadly upon gene expression [21] . This complex typically contains two putative deadenylases , with CAF1 ( CCR4-Associated Factor 1 , POP2 ) serving as the major deadenylase and CCR4 ( Carbon Catabolite Repressor 1 ) playing additional or specialized roles , except for in yeast where the roles are reversed [21] . While CAF1’s role in binding to and degrading poly ( A ) tracts is best appreciated , it has been shown to bind several other poly-nucleotide tracts [23] . Recently , a cryo EM structure of the S . pombe CAF1/CCR4/NOT complex was reconstructed using immunoprecipitated material . This work confirmed previous studies that used recombinant proteins and binding assays to show that the complex is L-shaped , that NOT1 ( Negative on TATA-less 1 ) acts as the scaffold , and that CCR4 binds to the complex indirectly through bridging interactions with CAF1 [24 , 25] . While these associations and activities have been well described in many eukaryotes , little is known about the CAF1/CCR4/NOT complex’s form and function in malaria parasites . In Plasmodium , previous work confirmed that normal deadenylase activity provided by CAF1 is essential for asexual blood stage growth [26 , 27] . Interestingly , insertion of a piggyBac transposon into the coding sequence revealed that CAF1 contributes to the regulation of invasion and egress-related genes in asexual blood stage parasites [26] . It is possible that this transposon insertion still results in the production of a partially functional CAF1 protein . Multiple independent attempts to knock out caf1 in the rodent-infectious species P . berghei failed , indirectly indicating that it is essential for parasite development [26 , 28] . Moreover , previous work on Plasmodium sporozoites identified that deletion of pypuf2 led to significant changes in the transcript abundance of several members of the CAF1/CCR4/NOT complex [6 , 29] . Among the affected transcripts , two mRNAs encoding CCR4 domain-containing proteins were dysregulated . As the deadenylase proteins of the CAF1/CCR4/NOT complex have been shown to be specialized regulators in other species , we investigated the possibility that CCR4 domain-containing proteins may be acting in this capacity in Plasmodium as well [30 , 31] . Here , we demonstrate that CCR4-1 is a specialized regulator during gametocytogenesis and transmission of the rodent-infectious Plasmodium yoelii parasite from the mammalian host to the mosquito vector . Deletion of pyccr4-1 , or expression of a putatively catalytic dead variant , resulted in a loss of the initial synchronous development of male gametocytes that can activate into gametes , as well as a reduction in the total number of mature male gametocytes . Moreover , deletion of pyccr4-1 also reduced the transmissibility of the parasite to the mosquito on both peak and post-peak transmission days , indicating that PyCCR4-1’s functions extend beyond its role in the partial synchronization of gametocytes . Comparative transcriptomics of wild-type and ccr4-1- gametocytes revealed that PyCCR4-1 significantly impacts the abundance of transcripts that are translationally repressed in female gametocytes , and those that impact the transmission to and establishment of an infection in the mosquito . We found that PyCCR4-1 binds directly to some of these affected transcripts and allows for increased transcript abundance without affecting UTR or poly ( A ) tail length . Surprisingly , this effect runs counter to the major canonical role of a deadenylase . Finally , proteomic characterizations and genetic modifications of pycaf1 and pfcaf1 indicate that the C-terminal region of CAF1 is needed for proper gametocyte development and to promote host-to-vector transmission .
To first assess the importance of the CAF1/CCR4/NOT complex in Plasmodium , we bioinformatically identified the genes for all members of the canonical CAF1/CCR4/NOT complex in Plasmodium , except for not3 and caf130 . The absence of these two particular genes is not surprising , as these genes are also absent in some eukaryotes [20] . In addition , we identified four CCR4 domain-containing proteins ( PyCCR4-1 , PyCCR4-2 , PyCCR4-3 , PyCCR4-4 ) that have homology to CCR4 deadenylases in other eukaryotes ( e . g . yeast , human , mouse ) ( S1A and S1B Fig ) through BLASTp alignments [32] . The typical domain architecture of CCR4-like proteins involves a Leucine Rich Repeat Region ( LRR ) and an Endonuclease/Exonuclease/Phosphatase ( EEP ) domain . The LRR mediates the interaction of CCR4 with CAF1 and the rest of the NOT complex , while the EEP domain contains active site residues required for deadenylation activity . Of these , we found that the overall length and sequence conservation within the EEP domain of PyCCR4-1 aligns most closely with the consensus CCR4 domain-containing proteins from model eukaryotes and humans ( S1A Fig ) . However , beyond the CCR4-EEP domain , there is no significant homology between other regions from PyCCR4-1 , 2 , 3 , and 4 to each other , or to homologues from model species ( S1B Fig ) [33] . As deadenylases are also known to act as translational regulators in specific and temporal manners , we investigated the role of all four CCR4 domain-containing proteins throughout the Plasmodium life cycle [34 , 35] . Our recent RNA-sequencing data from Plasmodium yoelii shows that all four genes are expressed in asexual blood stages and in gametocytes [4] , and thus we sought to determine if any of the CCR4 domain-containing proteins played an important , stage-specific role in the parasite life cycle . To this end , we replaced the coding sequences of pyccr4-1 , pyccr4-2 , pyccr4-3 , pyccr4-4 with a GFP-expression cassette and a human dihydrofolate reductase ( HsDHFR ) -expression cassette via double homologous recombination in the Plasmodium yoelii 17XNL strain ( S1C , S1D , S1E and S1F Fig ) . These lines were cloned via limiting dilution prior to characterization and their transgenic genotypes were confirmed using PCR across both homology regions . These clonal parasites revealed that deletion of any one of these genes individually was not lethal in asexual blood stages . Moreover , deletion of pyccr4-2 , pyccr4-3 , or pyccr4-4 resulted in transgenic parasites that behaved as wild-type in all life cycle stages with respect to parasite numbers , prevalence of mosquito infection , and developmental timing/completion throughout the Plasmodium life cycle ( Fig 1A , 1B and 1C , S1 Table ) . Thus , CCR4-2 , -3 , and -4 may play redundant roles with one or more of the other CCR4-domain containing proteins . However , while deletion of pyccr4-1 had no effect upon asexual blood stage growth ( S2A Fig ) , it led to significant phenotypes during male gametocyte maturation and host-to-vector transmission ( Fig 1A , 1B and 1C ) . First , to assess gametocytogenesis and the number of mature male gametocytes present , we developed an antibody-based flow cytometry assay based in part upon the effective reporter system ( 820cl1m1cl1 ) commonly used in P . berghei [15] . We generated antibodies against a recombinant domain variant of dynein heavy chain delta ( PyDD , PY17X_0418900 , “PyDDD” = AA1845-2334 ) , and together with anti-PvBiP antibodies to counterstain cells containing a parasite , we confirmed by flow cytometry and Giemsa staining that PyDD is a marker for mature male gametocytes in P . yoelii , as it is in P . berghei ( S2B Fig ) . This was further validated using a transgenic parasite line with a PyDDprom::GFPmut2 cassette integrated in the p230p safe harbor locus , where the population positive for both anti-PyDD and anti-GFP signals highly overlapped ( S2B Fig ) . The p230p safe harbor locus has been used extensively when expression out of a dispensable locus is required [4 , 15] . Ultimately , this approach allows these measurements to be done without the need to conduct reverse genetics in a base fluorescent reporter line , and frees up the use of GFP and RFP for other purposes . Using this flow cytometric method , we found that transgenic pyccr4-1- parasites produce fewer mature male gametocytes and more immature/female gametocytes compared to wild-type parasites ( Fig 1A , S2C Fig ) . Secondly , in contrast to wild-type parasites , which have a semi-synchronous wave of gametocyte development , pyccr4-1- parasites lose this coordination and instead develop fewer male gametocytes that can form gametes , and do so in an asynchronous manner ( Fig 1B ) . Total proteomics of mixed blood stage samples of P . yoelii wild-type and pyccr4-1- parasite lines indicated that many of the bioinformatically identified members of this complex ( NOT1 , NOT4 , NOT5 , NOT Family Protein , CAF40 ) are expressed at levels that permit their detection , whereas CCR4-1 and CAF1 were not sufficiently abundant to be detected using highly stringent thresholds ( S2 Table ) . To experimentally determine the composition of this complex in Plasmodium yoelii , a transgenic PyCCR4-1::GFP parasite was created ( S2D Fig ) . The CAF1/CCR4/NOT complex was immunoprecipitated via the GFP tag from synchronized schizonts , when PyCCR4-1 is most abundant and is most prominently localized to cytoplasmic granules ( S2D Fig ) . As seen in other eukaryotes , PyCCR4-1 associates directly or indirectly through bridging interactions with most members of the canonical CAF1/CCR4/NOT complex in P . yoelii ( Table 1 , S3 Table ) [21] . Specifically , through mass spectrometric analyses we found that PyCCR4-1 associates with CAF1 , NOT1 , CAF40 , NOT2 and a NOT family protein above our most stringent SAINT ( Significance Analysis of INTeractome ) threshold ( 0 . 1 ) , and with NOT5 using a less stringent threshold ( 0 . 1 to 0 . 35 ) . A small number of peptide spectral matches for NOT4 were also observed , but were not sufficiently enriched to be confidently included . This low abundance of NOT4 is consistent with its known transient association with the CAF1/CCR4/NOT complex in other eukaryotes [36] . We also found that PyCCR4-1 interacts with proteins involved in the nuclear pore complex and RNA export ( e . g . karyopherin-beta 3 , exportin-1 , UAP56 ) , proteins involved in translation initiation ( e . g . eIF2A , EF-1 , EIF3D , PABP ) , and translational repression ( e . g . CELF2/Bruno , DOZI , CITH , PABP ) ( Table 1 , S3 Table ) [21] . All of these interactions are consistent with appreciated CAF1/CCR4/NOT protein-protein or protein-RNA-protein interactions in other eukaryotes . Recently , a proteome of stress granule components in S . cerevisiae defined several cytosolic granule regulators , several of which we also found associated with PyCCR4-1 [37] . Specifically , we identified that multiple CCT proteins of the TRiC complex ( e . g . CCT4 , CCT5 , CCT8 ) and HSP40-A associate ( SAINT score < 0 . 1 ) , and this list expands to include the remainder of the TRiC core complex ( e . g . TCP1 , CCT2 , CCT3 , CCT6 , CCT7 ) and a regulatory kinase ( CK1 ) ( SAINT scores between 0 . 1 to 0 . 35 ) . CCT proteins are known to inhibit stress granule formation in yeast and may be present with the CAF1/CCR4/NOT complex while it plays roles in active translation [37] . In addition , recent work has implicated karyopherins/nuclear import receptors in the regulation of proteins found within liquid-liquid phase separations/cytosolic granules [38] . Here , we identified that karyopherin beta 3 associates with the P . yoelii CAF1/CCR4/NOT complex , and perhaps indicates that similar regulatory processes are at work . These data indicate that the composition of the CAF1/CCR4/NOT complex , including the presence of cytosolic granule regulators , are likely conserved throughout eukaryotes , including Plasmodium . CCR4 proteins have well defined , conserved catalytic residues in other eukaryotes that are also conserved in Plasmodium species ( S1B Fig ) [39] . To determine if the putative active site residues of PyCCR4-1 contribute to its functions in male gametocyte maturation and transmission , we created transgenic parasites with alanine substituted for two of the putative catalytic residues ( D1852A , H1898A ) of PyCCR4-1 ( dCCR4-1 ) ( S1B Fig , S2E Fig ) . Like pyccr4-1- parasites , dCCR4-1 transgenic parasites also produce fewer mature male gametocytes , and also lacked a synchronous wave of male activation ( Fig 1A and 1D ) . Because some male gametocytes retained the ability to mature and become exflagellating gametes in both the pyccr4-1- and dCCR4-1 lines , we assessed whether they were transmissible to mosquitoes . In both transgenic lines , we observed a corresponding decrease of similar scale ( 2-to-10 fold ) in the number of day seven oocysts compared to wild-type parasites when transmitted to An . stephensi on the peak day of male gametocyte activation into gametes ( Fig 1C ) . Moreover , although there is no statistical difference in the number of male gametocytes that can activate between wild-type and pyccr4-1- parasites after the peak day ( Fig 1B , days six and beyond ) , a significant decrease ( ~30% of wild-type , p<0 . 05 ) in the number of oocysts in the mosquito was still observed when parasites were transmitted two days post-peak ( S2F Fig ) . These data indicate that the catalytic residues of PyCCR4-1 are required for normal male gametocyte development and host-to-vector transmission . We next sought to determine if the other deadenylase in the CAF1/CCR4/NOT complex , CAF1 , was required for these effects upon gametocyte development and host-to-vector transmission [40–42] . As CCR4 domain-containing proteins associate with the NOT1 scaffold of the CAF1/CCR4/NOT complex indirectly by binding CAF1 , genetic deletion of caf1 would theoretically dissociate CCR4-1 from its complex . However , complete deletions of the caf1 gene have been unsuccessful in both a conventional targeted attempt and in the PlasmoGEM broad-scale genetic screen in P . berghei , indicating that it is likely essential [26 , 43] . We also attempted to completely delete the P . yoelii caf1 coding sequence and similarly were unable to delete these sequences ( S3A Fig ) . Instead , as the insertion of the piggyBac transposon into the P . falciparum caf1 gene occurred in the coding sequence downstream of the CAF1 domain , we hypothesized that this 5’ portion of pfcaf1 mRNA may still be expressed and may encode the necessary portion of the protein [26] . In support of this hypothesis , transcript expression analysis of the P . falciparum CAF1 disruptant line ( PfCAF1ΔC ) bearing this transposon insertion indicated that the CAF1 domain was still transcribed up to the insertion site , but not after ( S3B Fig ) . Based upon these expression data , we created a Plasmodium yoelii transgenic line that mimics this transposon insertion by inserting a C-terminal GFP tag and stop codon in the Plasmodium yoelii caf1 gene in a comparable location following the CAF1 domain , thus creating a PyCAF1ΔC ( AA 1–335 ) variant ( S3C Fig ) [26] . We found that expression of the PyCAF1ΔC::GFP variant resulted in viable parasites , but importantly , that these parasites exhibit a similar growth attenuation as was observed for the P . falciparum PfCAF1ΔC line ( Fig 1E ) [26] . To further assess the impact of the PyCAF1ΔC variant upon parasite growth and transmission , we observed comparable , but more pronounced , effects upon the activation of male exflagellation and parasite transmission to those seen with pyccr4-1- parasites ( 10-fold decrease in male activation on peak day and >4-fold reduction in transmission to mosquitoes , respectively ) ( Fig 1C and 1F ) . These exacerbated effects may be caused by the combined effects of a reduction in total parasite numbers due to the deletion of portions of PyCAF1 and a PyCCR4-1-dependent defect in male gametocyte development . These effects upon asexual and sexual blood stage parasites could occur if this truncated form of PyCAF1 had reduced functionality due to an inability to associate with its complex . To assess this possibility , we raised specific antibodies to the N-terminus of the PyNOT1 scaffold protein and with them immunoprecipitated this complex , including PyCAF1ΔC::GFP ( S3D Fig ) . This indicates that the CAF1 domain of PyCAF1 is sufficient for association with its complex and for its required functions . However , the remainder of the PyCAF1 protein contributes to the functions of PyCCR4-1 , as this truncation phenocopies pyccr4-1- and dPyCCR4-1 parasites . To determine if the PfCAF1ΔC variant in human-infectious P . falciparum similarly impairs gametocytogenesis as was seen in rodent-infectious P . yoelii , the PfCAF1ΔC piggyBac-insertion parasite line was assessed for effects upon parasitemia , gametocytogenesis , as well as male gametocyte activation ( Fig 2 ) . The PfCAF1ΔC line exhibited significant decreases in gametocyte conversion , total gametocytemia , and exflagellation on the peak day as compared to wild-type P . falciparum NF54 strain parasites ( Fig 2 , S4 Table ) . These data support the observed P . yoelii phenotype and indicate that this conserved complex is important to sexual development across Plasmodium species . In eukaryotes , CCR4 and CAF1 function while in association with the other members of the CAF1/CCR4/NOT complex and is found in nuclear and cytosolic granular structures [20] . Because PyCCR4-1 lacks an obvious LRR domain by which it can associate with the rest of the complex , we used immunofluorescence and live fluorescence assays to further validate these interactions . First , using transgenic PyCCR4-1::GFP parasites , we observed that PyCCR4-1 localized to cytoplasmic puncta in asexual blood stage parasites , and is similarly localized in both male and female gametocytes ( Fig 3 ) . Moreover , this expression profile extends to oocysts , oocyst sporozoites , and salivary gland sporozoites , where PyCCR4-1 was seen both in cytosolic puncta and located diffusely throughout the parasite ( S4A Fig ) . However , PyCCR4-1 was not detected above background in liver stage parasites ( S4B Fig ) . Thus , the near constitutive expression and localization of PyCCR4-1 in cytoplasmic foci in Plasmodium resembles that of its orthologues in model eukaryotes . Next , using either full length PyCAF1::GFP or PyCAF1ΔC::GFP transgenic parasites , we observed a similar expression and localization pattern to that of PyCCR4-1::GFP ( S4C and S4D Fig ) . Moreover , colocalization of PyCAF1ΔC::GFP and PyNOT1 signals were observed ( S4E Fig ) . Together , these data further indicate that the truncated PyCAF1ΔC::GFP variant can remain associated with the rest of its complex and yet phenocopies these PyCCR4-1-associated effects . Because PyCCR4-1 is a putative deadenylase , we hypothesized that the defects in male gametocyte maturation and transmission observed in pyccr4-1- and dPyCCR4-1 parasites may be attributed to PyCCR4-1 acting upon specific transcripts important to gametocytogenesis , gamete activation , and/or parasite transmission to mosquitoes . To determine the role of PyCCR4-1 in the regulation of transcripts in gametocytes , total comparative RNA sequencing ( RNA-seq ) was performed . Gametocytes from a wild-type line expressing GFP from the p230p dispensable locus ( WT-GFP ) and the pyccr4-1- transgenic line were selected using sulfadiazine treatment , purified on an Accudenz gradient , and their RNA extracted for RNA-seq . Differential abundance of transcripts was assessed via DEseq2 , and the p-adjusted value was used for all analyses ( Fig 4A , S5 Table ) [44 , 45] . Nearly all ( 172 of 175 ) of the significantly affected transcripts ( P-adjusted < 0 . 05 , > 2 fold change ) between WT-GFP and pyccr4-1- parasites decreased in abundance in the pyccr4-1- parasites , while only 3 transcripts increased in overall abundance ( S6 Table ) . Many of these decreases in transcript abundance are for 55 mRNAs that encode male-enriched proteins , and thus these changes can likely be attributed to the production of fewer mature male gametocytes in the pyccr4-1- transgenic line . However , the effect upon other transcripts , including those associated with female gametocytes , cannot be explained in this way ( Fig 4B ) . Most notably , transcripts that encode for proteins involved in gamete function ( e . g . GEST ) and early mosquito stage development ( e . g . p28 , CITH , AP2-O , HMGB2 , LAP2 ) decreased in abundance significantly in the absence of PyCCR4-1 . While a catalog of translationally repressed transcripts is not available for P . yoelii , many of these transcripts are known to be translationally repressed in P . falciparum female gametocytes ( Fig 4B ) . Interestingly , an ApiAP2 protein , recently identified as AP2-G3 ( PY17X_1417400 ) , that is important for gene expression in gametocytogenesis decreased in abundance 10-fold in pyccr4-1- gametocytes [46] . Disruption of this ApiAP2 gene in P . falciparum by piggyBac transposon insertion resulted in the formation of no gametocytes and deletion of this gene in P . yoelii resulted in significantly reduced numbers of gametocytes [46 , 47] . Effects upon this gene may have broad reaching effects on gametocyte development , and may contribute to the phenotypes observed here . Other transcripts-of-interest that decreased in abundance are those that encode for multiple uncharacterized RNA-binding proteins ( PY17X_1203900 , PY17X_1457300 , PY17X_0923600 ) , a second ApiAP2 protein ( ApiAP2-O5 , PY17X_1317000 ) and BDP2 ( PY17X_1431000 ) , an uncharacterized putative transcriptional activator ( S6 Table ) [47] . Additional analysis demonstrates that many of the transcripts that decrease in abundance in pyccr4-1- gametocytes are differentially expressed in P . berghei gametocytes when compared to asexual parasites ( Fig 4C ) [48] . Together , we conclude that these differences in transcript abundance not only reflect a reduction in the number of mature male gametocytes , but also indicate that PyCCR4-1 is acting to preserve specific transcripts important for the gametocyte and early mosquito stage parasite . As transcript abundances could be affected directly by PyCCR4-1 and its complex , or indirectly through compensatory mechanisms such as gene buffering when the pyccr4-1 gene is deleted [49] , we investigated whether these dysregulated mRNAs were bound by the CAF1/CCR4/NOT complex . To this end , unfused GFP ( expressed in WT-GFP parasites ) and PyCCR4-1::GFP were immunoprecipitated from purified , transgenic gametocytes , and the association of co-precipitated transcripts was detected by RT-PCR . We found that PyCCR4-1::GFP interacted specifically with a number of selected transcripts that substantially change in abundance in pyccr4-1- parasites , including p28 ( PY17X_0515900 ) , lap2 ( PY17X_1304300 ) , and nek3 ( PY17X_0603200 ) ( Fig 5A , top row; S5A Fig ) . These transcripts are notable , as they are all important/essential for gametocytogenesis , transmission , or early mosquito stage development and include transcripts known to be expressed in male ( nek3 ) and/or female ( p28 , lap2 ) gametocytes that are important for establishment of mosquito infections [48 , 50–53] . However , not all dysregulated transcripts ( cith and ap2-o ) were found specifically associated with PyCCR4-1 ( Fig 5A , bottom row; S5A Fig ) , suggesting that these effects likely result from a combination of both direct and indirect effects . The direct effects that CCR4 can have on transcript abundance in other eukaryotes have resulted from deadenylation of a target transcript , or from translational repression by binding/tethering to the CAF1/CCR4/NOT complex [30] . To investigate whether the poly ( A ) tail length and/or UTRs were affected in the presence or absence of PyCCR4-1 , circular-RT PCR ( cRT PCR , illustrated in Fig 5B ) was used to interrogate both a control transcript ( gapdh , not affected by ccr4-1 deletion , does not interact with CCR4-1 ) , and an affected/bound transcript ( p28 ) ( Fig 5C and 5D , S5B and S5C Fig ) . In the absence of PyCCR4-1 , there were no gross effects upon UTR/poly ( A ) tail lengths of these transcripts when visualized by PCR using primers that anneal near the start and stop codons ( Fig 5B and 5C , oligonucleotides provided in S7 Table ) . Sequencing of cloned PCR products from both wild-type and pyccr4-1- samples revealed the consistent composition of the 5’ and 3’UTRs , as well as the presence of a poly ( A ) tail ( S6 Fig ) . Using the sequencing data , primers that anneal near to the poly ( A ) tail were used to assess the distribution of poly ( A ) tail lengths in the population . Using these primers , we did not observe any differences in poly ( A ) length between the wild-type and pyccr4-1- populations , and are estimated to be ~75nt long ( Fig 5D ) . These data indicate that the direct effect of PyCCR4-1 on these specific transcripts in gametocytes do not impact the poly ( A ) tail/UTR length , suggesting that the complex may be acting in other ways to preserve these transcripts .
Plasmodium encodes few known specific transcription factors and a relatively over-represented number of RNA-binding proteins ( 10% of its predicted proteome ) [11 , 29] . One model suggests that Plasmodium has adapted these complementary regulatory mechanisms to achieve its preferred RNA homeostasis . Moreover , the malaria parasite also proactively transcribes a large number of genes before transmission , but does not produce the encoded proteins until transmission has occurred . This translational repressive mechanism has been shown to be imposed by members of the DOZI/CITH/ALBA complex , as well as by PUF2 . Here , we demonstrate that the PyCCR4-1 and PyCAF1 members of the CAF1/CCR4/NOT complex play additional roles in either the preservation or expression of translationally repressed transcripts through direct and indirect means ( Fig 6 ) . Moreover , we find that PyCCR4-1 is also important for the development of the male gametocyte , as well as for the efficient transmission of gametocytes to the mosquito vector . The composition and behaviors of the CAF1/CCR4/NOT complex are well-conserved in eukaryotes , but key species-specific differences exist . Our proteomic and bioinformatic analyses of Plasmodium species showed that many of the core components of this complex are shared between Drosophila melanogaster , Saccharomyces cerevisiae , and Homo sapiens ( NOT1 , NOT2 , NOT5 ( NOT3 ) , NOT4 , CCR4 , CAF1 ( POP2 ) , and CAF40 ) , but that some proteins are apparently not encoded at all ( CNOT10 ) [29] . Additionally , while there are well-defined roles for this complex in both the nucleus and cytoplasm of other eukaryotes , we find by IFA that the vast majority of PyCCR4-1 , PyCAF1 , PyNOT1 , and by inference the entire complex , localizes to discrete cytosolic granules throughout most of the Plasmodium life cycle . For this reason , we have focused our analyses here upon the known cytoplasmic roles of the complex , but additional work to define possible nuclear functions of the CAF1/CCR4/NOT complex is certainly warranted . Previous studies of CCR4 in other eukaryotes have demonstrated that CCR4 can act as a translational repressor , and that its catalytic residues and deadenylase activity are implicated in its repressive functions [30 , 54 , 55] . Our results demonstrate that PyCCR4-1 helps to preserve translationally repressed transcripts and are suggestive of similar roles in Plasmodium ( Fig 4A ) . Furthermore , transgenic parasites that lack pyccr4-1 , that express a putatively dead catalytic variant of PyCCR4-1 , or that express a truncated variant of PyCAF1 ( PyCAF1ΔC ) demonstrate the same phenotype , and thus we conclude that the catalytic residues of the PyCCR4-1 protein and the portion of CAF1 C-terminal to the CAF1 domain are required for this regulatory function ( Figs 1 and 6 ) . As the maturation of male gametocytes is substantially promoted by the presence of catalytically active PyCCR4-1 , it is unsurprising that one third of the transcripts that PyCCR4-1 affects encode proteins that are present in a P . berghei male gamete proteome ( Fig 4 ) . It should be noted that the PyCCR4-1 catalytic mutant ( dCCR4-1 ) has a slightly stronger effect upon gametocytogenesis , gametogenesis , and parasite transmission than does the deletion of pyccr4-1 . This effect could result if dCCR4-1 is still able to bind transcripts and prevent other proteins from assembling onto them , thus acting as a dominant negative variant . Expression of a truncated CAF1 variant in P . yoelii or P . falciparum produced similar phenotypes during gametocytogenesis and male gametocyte activation ( Figs 1 and 2 ) . Immunoprecipitation and IFA analyses of PyNOT1 and PyCAF1ΔC showed that this could not be explained by the disruption of their interaction . This work also defines the CAF1 domain as the essential portion of the PfCAF1 and PyCAF1 proteins , and that sequences C-terminal of it play additional functional , but non-essential roles . During sexual development , we find that PyCCR4-1 binds to multiple transcripts essential for gametocyte development and host-to-vector transmission , which are dysregulated in pyccr4-1- , parasites . We hypothesize that PyCCR4-1 is interacting with and acting upon these transcripts to preserve them for use post-transmission . Circular RT-PCR demonstrated that UTR and poly ( A ) lengths are not affected in pyccr4-1- transgenic parasites for a dysregulated , female transcript ( p28 ) . Global analyses of poly ( A ) tail length have been demonstrated in other eukaryotes , and could provide evidence for whether this phenomenon holds for all dysregulated transcripts . Therefore , we posit that PyCCR4-1 and its complex can be utilizing functions independent of deadenylation for these regulatory functions . For instance , tethering of transcripts to the CAF1/CCR4/NOT complex can induce repression of a target transcript in other systems , even in the absence of the deadenylation of those transcripts [30] . Taken together , these data indicate that the Plasmodium CAF1/CCR4/NOT complex provides key functions in the regulation of specific transcripts to promote coordinated male gametocyte maturation and parasite transmission .
All animal care strictly followed the Association for Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) guidelines and was approved by the Pennsylvania State University Institutional Animal Care and Use Committee ( IACUC# 42678–01 ) . All procedures involving vertebrate animals were conducted in strict accordance with the recommendations in the Guide for Care and Use of Laboratory Animals of the National Institutes of Health with approved Office for Laboratory Animal Welfare ( OLAW ) assurance . Six-to-eight week old female Swiss Webster mice from Harlan ( recently acquired by Envigo ) were used for all of the experiments in this work . Anopheles stephensi mosquitoes ( obtained from the Center for Infectious Disease Research; Seattle , WA ) were reared at 24C and 70% humidity and were used to cycle Plasmodium yoelii ( 17XNL strain ) parasites . Transgenic Plasmodium yoelii ( 17XNL strain ) parasites were created using targeting sequences to incorporate sequence into the target gene using double homologous recombination using standard procedures [56] . Parasite genomic DNA was purified ( QIAamp DNA Blood Kit , Qiagen , Cat# 51106 ) and genotyping PCR was performed to assess the ratio of WT to transgenic parasites present . Clonal parasite populations were produced using limiting dilution cloning . The PfCAF1ΔC line was previously generated as described [26] . Validation of CAF1 transcript expression was performed via RT-PCR on 100ng DNase-treated RNA from PfCAF1ΔC and NF54-control parasites using primer sets supplied in S7 Table . To produce schizonts in culture , infected Swiss Webster mice were exsanguinated by cardiac puncture and the blood was collected into complete RPMI ( cRPMI ) , spun at 200 xg for 8 min to remove the serum , and then cultured in 30 ml cRPMI in a 5% CO2 , 10% O , 85% N gas mixture for 12 hours at 37C . Cultures were underlayed with 10 ml of 17% w/v Accudenz in 5 mM Tris-HCl ( pH 7 . 5@RT ) , 3 mM KCl , 0 . 3mM disodium EDTA , 0 . 4x PBS ( without calcium and magnesium ) and spun at 200 xg for 20 min with no brake [56] . Parasites were collected from the interface between the Accudenz and cRPMI layers , transferred to a fresh conical tube , supplemented with an equal volume of additional cRPMI , and spun for 10 min at 200 xg . The supernatant was then removed and the parasite pellet processed for downstream applications . Gametocytes were produced by treatment of the mice at 1% parasitemia with 10 mg/L sulfadiazine ( VWR , Cat# AAA12370-30 ) in their drinking water for two days prior to exsanguination . The blood was maintained in warm ( 37C ) cRPMI to prevent activation of gametocytes and parasites were purified as described above . Gametocyte-producing cultures were established as described previously [57] with some modification . Briefly , starter cultures of wild-type P . falciparum NF54 and PfCAF1ΔC were grown to ~5% parasitemia in standard culture conditions in cRPMI supplemented with 25 mM HEPES , 0 . 2% D-glucose , 200 uM hypoxanthine , 0 . 2% w/v sodium bicarbonate , and 10% v/v heat-inactivated human serum at 6% hematocrit in a tri-gas incubator ( 5% CO2 , 5% O2 ) at 37C . On Day 0 , starter cultures were then used to inoculate 75 cm2 flasks ( 15ml culture volume at 6% hematocrit ) in technical duplicate for each line at 0 . 5% parasitemia . Parasites were cultured for 17 days with daily media changes and no fresh addition of blood . Samples were taken to monitor parasite development starting at Day 3 post-infection and then every 48 hours until Day 13 post-infection , determined through Giemsa-stained smears . At seven days post-infection , technical replicate flasks were combined into one flask , which was maintained for the duration of the experiment . Samples were assessed in biological triplicate . Coding sequence for a domain of PyDD ( PY17X_0418900 , “PyDDD” = AA1845-2334 ) was generated by IDT as a codon-optimized gene block ( gBlock ) for expression from a modified pET28b+ vector ( pSL0220 ) in E . coli BL21 ( DE3 ) pLysS CodonPlus bacteria . Protein was purified first by standard Ni-NTA and then glutathione resin approaches ( details provided in S1 File ) . Purity was confirmed to be >90% by SDS-PAGE by Coomassie Blue staining . Antibodies were generated in rabbits ( screened for pre-immune sera with minimal background reactivity ) by Pocono Rabbit Farm and Laboratory ( Canadensis , PA ) . PyCCR4-1 and PyCAF1 expression in blood stages , oocyst sporozoites , salivary gland sporozoites and liver stages was observed by an indirect immunofluorescence assay ( IFA ) , and expression in day seven oocysts was observed by live fluorescence . All samples for IFA were prepared as previously described , with all details provided in S1 File [58] . Fluorescence and DIC images were taken using a Zeiss fluorescence/phase contrast microscope ( Zeiss Axioscope A1 with 8-bit AxioCam ICc1 camera ) using a 40X , 63X , or 100X oil objective and processed by Zen imaging software . Cryopreserved blood infected with either wild-type ( Py17XNL ) , pyccr4-1- , dCCR4-1 , or PyCAF1ΔC parasites were injected intraperitoneally into Swiss Webster starter mice and parasitemia was allowed to increase to 1% . This blood was extracted via cardiac puncture and diluted in RPMI to 10 , 000 parasites per 100 ul ( CCR4-1 , CAF1ΔC ) or 1 , 000 parasites per 100ul microliter ( dCCR4-1 ) . One hundred microliters was injected intravenously ( IV ) into three mice per replicate for each parasite line . Three biological replicates were conducted , each with three technical replicates . Parasitemia was measured daily by giemsa-stained thin blood smears . Centers of movement/exflagellation centers were also measured daily via wet mount of the blood incubated at room temperature for 10 min by counting the number of exflagellating male gametocytes in a confluent monolayer per 400x field ( 40x objective x 10x eyepiece ) . P . falciparum ring stage parasitemia and total gametocytemia were calculated every two days starting on Day 3 post-infection by averaging counts in 10 , 000 RBCs across a minimum of two biological replicates ( provided in S6 Table ) . Sexual conversion was calculated as described previously [59] by taking the stage II-gametocytemia on Day T and dividing by ring stage parasitemia on Day T-2 . Samples were taken for exflagellation assays on days 13 , 14 , 15 , and 16 post-infection . Two-hundred microliter samples were taken from each flask and spun down at 300 xg for 30 seconds . Supernatant was removed and a 20 ul aliquot of remaining blood pellet was mixed with 20 ul of heat-inactivated human serum previously warmed to 37°C . The mixture was then allowed to incubate at room temperature for 15 min , after which exflagellation events were counted under 40x magnification for 10 fields-of-view . Cryopreserved blood infected with either wild- type ( Py17XNL ) , ccr4-1- , dCCR4-1 , or CAF1ΔC parasites was injected intraperitoneally into starter mice and transferred as above ( 10 , 000 parasites/100ul ) . On Day 5 , gametocytes were produced by treatment of the mice with 10 mg/L sulfadiazine ( VWR , Cat# AAA12370-30 ) in their drinking water for two days . Blood was collected by cardiac puncture and maintained in warm cRPMI to prevent activation of gametocytes and spun at 37°C . Blood was then fixed , passed through a cellulose column and stained as described above for IFA . Parasites were stained with the following primary antibodies: mouse anti-PvBIP Clone 7C6B4 ( 1:1000; [60] ) and rabbit anti-PyDynein Heavy Chain Delta Domain ( “PyDDD” , PY17X_0418900 AA: 1845 to 2335 ) ) ( 1:1000 , Pocono Rabbit Farm & Laboratory , Custom PAb ) , along with goat anti-mouse conjugated to AF594 ( Fisher Scientific , A11012 ) and goat anti-rabbit conjugated to AF647 ( Fisher Scientific , PIA32733 ) secondary antibodies . These were then analyzed on a LSR Fortessa ( BD ) in tube mode and collected samples were analyzed in FlowJo . Cryopreserved blood infected with either wild-type ( Py17XNL ) , ccr4-1- , dCCR4-1 , or CAF1ΔC parasites was injected intraperitoneally into starter mice and transferred as above . Centers of movement were checked daily as above and mice were fed to mosquitoes on the peak day of exflagellation ( day 5 ) . Mosquito midguts were dissected at D7 post feed and analyzed for the prevalence of infection and oocyst numbers by microscopy . Mosquito midguts ( day 10 ) or salivary glands ( day 14 ) were dissected , ground , and sporozoite numbers counted . Parasite pellets ( schizonts ) were crosslinked in 1% v/v formaldehyde and lysed using RIPA lysis buffer with a 1x protease inhibitor cocktail and 0 . 5% v/v SUPERase In , dounce homogenization with a tight pestle , and sonication . The parasite lysate was then precleared using streptavidin-coated dynabeads was immunoprecipitated using a biotin-conjugated antibody ( anti-GFP or anti-PyNOT1 ) loaded on streptavidin-coated dynabeads for three hours at 4C with rotation . The beads were washed with modified RIPA wash buffer ( 50 mM Tris-HCl ( pH 8 . 0@RT ) , 1 mM EDTA , 150 mM NaCl , 1% v/v NP40 ) once and then transferred to a new tube . The beads were washed 3 more times with modified RIPA wash buffer and then eluted at 45C overnight in a heat block . Samples were quality controlled by western blotting , and then subjected to tryptic digest and LC/MS/MS identification ( Harvard Proteomics Core , run parameters listed in S1 File ) . The data was processed using the Trans-Proteomic Pipeline ( TPP ) [61] as described previously with few modifications [19] . Spectra were searched against reference sequences downloaded in February 2016 from Plasmodium yoelii 17X ( PlasmoDB , v27 ) , mouse ( Uniprot ) , and common contaminants ( Common repository of adventitious protein sequences , [62] and randomized decoys generated through TPP . X ! Tandem and Comet searches were combined in iProphet [63] and protein identifications were determined by Peptide Prophet . Only proteins with a highly stringent false positive error rate of less than 1% are reported . To combine replicate proteomics datasets , SAINT version 2 . 5 . 0 was used [64] . Only proteins with SAINT scores below 0 . 1 ( most stringent ) or 0 . 35 ( stringent ) were considered significant hits and included in the analyses , as used previously [2 , 4 , 6] . The total proteome of Py17XNL mixed blood stages was determined using the same workflow ( Penn State Proteomics Core ) . Gametocytes were produced , collected , and purified by an Accudenz gradient , as above . Infected RBCs were lysed with saponin , washed with 1xPBS , and released parasites were then lysed immediately using the QIAgen RNeasy Kit using the manufacturer’s protocol with the additional on-column DNaseI digestion . RNA yields were quantified spectrophotometrically by NanoDrop , and RNA samples were further quality controlled ( BioAnalyzer ) and used to create barcoded libraries ( Illumina TruSeq Stranded mRNA Library ) . An equimolar pool of all samples was made and 100 nt single end read sequencing was performed on an Illumina HiSeq 2500 in Rapid Run mode . The resulting data was mapped to the P . yoelii 17XNL strain reference genome ( plasmodb . org , v32 using Tophat2 in a local Galaxy instance ( version . 9 ) . Gene and transcript expression profiles for both WT-GFP and ccr4-1- assemblies were generated using htseq-count ( Galaxy version 0 . 6 . 1galaxy3 ) [65] using the union mode for read overlaps . Count files were merged and compared using DESeq2 ( Galaxy version 2 . 11 . 39 [66] ) . Six biological replicates were used for the WT transcriptomic profile , while four replicates were used in for the ccr4-1- profiles . These were analyzed by a mean fit type with outlier replacement turned on to normalize the variance between the count files . The P-adjusted value was used for all analyses . RNA was isolated from purified P . yoelii wild type or pyccr4-1- gametocytes by TRIzol/chloroform extraction and extensive DNaseI digestion . The 7-methylguanosine cap was removed from 10ug of total RNA using 2 . 5U Cap-Clip Acid Pyrophoshatase in 1xCap-Clip Buffer supplemented with 10U Murine RNase Inhibitor at 37C for 1 hour . Treated RNA was TRIzol extracted , precipitated , dried , and then circularized with T4 RNA Ligase in T4 DNA Ligase buffer supplemented with 10% w/v PEG8000 and 10U Murine RNase Inhibitor at 16C for 24 hours . RNA was purified , precipitated , dried , and then subjected to reverse transcription using SuperScript IV and gene-specific primers ( S7 Table ) . Specific PCR amplification of gapdh and p28 sequences from the resulting cDNA was conducted using Phusion polymerase ( NEB ) and gene specific primers ( S7 Table ) . Statistical differences between P . yoelii wild-type and transgenic parasites were assessed via a two-tailed t-test on Graphpad Prism . Statistical differences between P . falciparum wild-type and PfCAF1ΔC parasites were assessed via a paired Wilcoxon test using R v . 3 . 3 . 1 [67] with p < 0 . 05 indicating statistical significance . All data is publically available on common data repositories . Proteomics data is accessible at the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD007042 [68] . Transcriptomics data ( both RAW and processed files ) is accessible at the GEO repository ( Accession #GSE101484 ) . Details of datasets and identifiers are available in S1 File . | Malaria is a disease caused by Plasmodium parasites , which are transmitted during an infectious blood meal by anopheline mosquitoes . Transmission of the sexual stages of the parasite to mosquitoes requires the proper regulation of specific mRNAs . While much work has been done to characterize regulation of mRNAs in female gametocytes , little has been done to assess this regulation in male gametocytes . Here , we demonstrate that PyCCR4-1 , a member of the CAF1/CCR4/NOT RNA metabolic complex , acts upon transcripts both directly and indirectly in gametocytes , and results in a reduction of male gametocytemia . In gametocytes lacking PyCCR4-1 , as well as those expressing a catalytically dead variant , the initial coordinated wave of male gametocyte activation is lost , and these parasites are ~4-fold less able to productively infect mosquitoes . We find that the deletion of the C-terminal portion of CAF1 in both Plasmodium yoelii and Plasmodium falciparum phenocopies the deletion of pyccr4-1 . We also find that the CAF1/CCR4/NOT complex is directly binding some of these transcripts and is likely acting both directly upon mRNAs and indirectly to modulate transcript abundance . These findings demonstrate that the combined effects of the CAF1/CCR4/NOT complex upon specific mRNAs are important for both male and female gametocytes , and that this regulation is required for efficient transmission to the mosquito vector . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"parasite",
"groups",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"plasmodium",
"plasmodium",
"yoelii",
"gametocytes",
"messenger",
"rna",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"parasitology",
"germ",
"cells",
"apicomplexa",
"protozoans",
"m... | 2019 | Plasmodium male gametocyte development and transmission are critically regulated by the two putative deadenylases of the CAF1/CCR4/NOT complex |
Recent studies have shown that the transcriptional functions of REST are much broader than repressing neuronal genes in non-neuronal systems . Whether REST occupies similar chromatin regions in different cell types and how it interacts with other transcriptional regulators to execute its functions in a context-dependent manner has not been adequately investigated . We have applied ChIP-seq analysis to identify the REST cistrome in human CD4+ T cells and compared it with published data from 15 other cell types . We found that REST cistromes were distinct among cell types , with REST binding to several tumor suppressors specifically in cancer cells , whereas 7% of the REST peaks in non-neuronal cells were ubiquitously called and <25% were identified for ≥5 cell types . Nevertheless , using a quantitative metric directly comparing raw ChIP-seq signals , we found the majority ( ∼80% ) was shared by ≥2 cell types . Integration with RNA-seq data showed that REST binding was generally correlated with low gene expression . Close examination revealed that multiple contexts were correlated with reduced expression of REST targets , e . g . , the presence of a cognate RE1 motif and cellular specificity of REST binding . These contexts were shown to play a role in differential corepressor recruitment . Furthermore , transcriptional outcome was highly influenced by REST cofactors , e . g . , SIN3 and EZH2 co-occupancy marked higher and lower expression of REST targets , respectively . Unexpectedly , the REST cistrome in differentiated neurons exhibited unique features not observed in non-neuronal cells , e . g . , the lack of RE1 motifs and an association with active gene expression . Finally , our analysis demonstrated how REST could differentially regulate a transcription network constituted of miRNAs , REST complex and neuronal factors . Overall , our findings of contexts playing critical roles in REST occupancy and regulatory outcome provide insights into the molecular interactions underlying REST's diverse functions , and point to novel roles of REST in differentiated neurons .
The REST ( RE1-silencing transcription factor ) [1] , also known as NRSF ( Neural Restrictive Silencing Factor ) [2] and XBR ( X2 Box Repressor ) [3] , encodes a zinc-finger transcription factor that was initially shown to repress neuronal genes in non-neuronal tissues and neural progenitors . It has since been shown to play a broad range of roles in neuronal differentiation and development [4]–[6] , such as fine-tuning neural gene expression [7] and modulating synaptic plasticity [8] . REST is necessary for the maintenance of self-renewal capacity of neural stem cells ( NSCs ) , as its knockdown led to a lower mitotic index and a higher rate of early neuronal differentiation [5] . REST has also been implicated as a tumor suppressor in breast cancer , colorectal cancer and small cell lung cancer , and as an oncogene in neuroblastomas , medulloblastomas and pheochromocytomas , which are associated with von Hippel-Lindau syndrome [9] , [10] . These findings show that REST plays diverse roles in multiple cellular processes . In addition to the 21-bp DNA sequence bound by REST ( termed the RE1 motif ) , an array of cofactors have been found to interact and cooperate with REST , including SIN3 , CoREST , Polycomb Repressive Complexes ( PRCs ) , and various histone deacetylases ( HDACs ) [9] , [11] , [12] . Many of these cofactors are chromatin modifiers or are associated with enzymes that have effects on post-translational histone modifications , suggesting that at the molecular level REST functions as a platform for the recruitment of multiple chromatin modifiers and that together they orchestrate gene regulation [9] , [11] , [12] . In fact , REST occupancy has been found to correlate with an increase of repressive and a decrease of active histone modifications [13] . Not all of the REST cofactors , however , are recruited to each of the REST-bound loci concomitantly . For example , a study of REST occupancy in eight RE1 loci in mouse NSCs found the existence of four distinct configurations of REST and its cofactors: REST-Sin3b-CoREST-HDAC1/2 , REST- Sin3b-CoREST , REST-CoREST , and REST-Sin3b-HDAC1/2 [14] . An independent genome-wide study showed that approximately half of the REST-bound sites in mouse ESCs were associated with REST cofactors ( in various combinations of the Sin3 and CoREST family members ) and that genes targeted by REST together with its cofactors showed more repression than genes bound by REST alone [15] . These studies indicate that differential recruitment of REST cofactors can potentially orchestrate distinct transcriptional outcomes . Moreover , REST acts as a repressor at only a subset of RE1 containing genes [16] , [17]; in particular , REST and its splicing variants have been reported to activate a variety of genes in certain cell types and conditions , such as CHRNB2 , CRH , PAX4 , and OPRM1 [18]–[23] . Previous genome-wide studies have characterized REST targets in human Jurkat cells [24] , K562 cells [25] , APL blasts [26] , mouse embryonic [15] , [27]–[29] and neural stem cells [27] , [28] . Widespread switches in REST targeting during mouse neuronal and glial differentiation have also been reported [7] , [30] , [31] . Moreover , analysis of 1% of the human genome ( the ENCODE pilot regions [32] ) using ChIP-chip technology has found interesting context-dependent REST functions among cell types [33] . While these previous studies have provided important information and principles about REST chromatin interaction and gene regulation , they have not systematically addressed how REST cistromes differ across diverse human cell types . In our current study we set out to address this by integrated analysis of ChIP-seq and RNA-seq data , including primary cells and differentiated neurons . By a comprehensive analysis of ChIP-seq data for 16 cell types , including one collected for this study for CD4+ T cells and 15 cell types from the ENCODE project [34] , we have identified a total of 21 , 134 non-redundant REST binding sites ( i . e . , peaks ) in the human genome . Among the REST sites in 15 “non-neuronal” cell types , only 7% were common to all . We then studied how RE1 motif status , genic location and the cellular context were related to REST binding and gene expression , as well as how these contexts were related to co-factor colocalization . Finally we compared REST occupancy between neurons and non-neuronal cells and found that REST bound to a distinct set of targets in neurons . Moreover , in contrast to other cell types , REST binding was largely localized to genes that were activated in differentiated neurons , as indicated by high levels of gene expression and active histone modifications . Our study provides valuable insights into the dynamic landscape of REST-chromatin interactions in the human genome and the importance of genomic and cellular contexts in modulating the outcome of REST regulation .
To investigate the roles of REST occupancy in cell types relevant to normal physiology , we have performed a ChIP-seq analysis on human CD4+ primary T cells and compared the results with REST ChIP-seq data for 28-day-old differentiated human neurons ( derived from the H1-ESC line ) , which were obtained from the ENCODE project [34] ( Table S1 ) . In order to make the data uniform and comparable across cell types , we called REST binding sites ( i . e . , ChIP-seq peaks ) by the same pipeline and with consistent parameters across datasets: using the program SPP [35] and the IDR methodology recommended by the ENCODE project [34] ( see Methods ) . As a result , we called 4 , 404 REST binding sites for the T cells and 5 , 387 for the neurons ( Table 1 ) . From the T cell REST peaks , we selected 10 sites and confirmed the binding of nine by ChIP-qPCR; the qChIP enrichments were consistent with the peak enrichment scores provided by SPP ( Fig . S1 ) . 38% of the T cell REST peaks had the canonical 21 bp RE1 motif ( cRE1 ) ( Fig . 1A ) , while an additional 3% contained a non-canonical RE1 motif ( ncRE1 ) , in which two halves of the cRE1 motif were separated by a 1–10 nucleotide insertion [24] , [27] ( Fig . 1A , bottom ) . Consistent with previous reports , we also found that a large fraction ( 22% ) of the REST peaks had only one of the two half-sites . In total , 63% of REST peaks in T cells had the cRE1 motif or one of its variants . For the neuron peaks , motif analysis revealed an unexpected and different picture . Less than 10% of the REST peaks in neurons contained either the cRE1 or ncRE1 motifs ( Fig . 1B ) . Even the enrichment of RE1 motifs in the top 600 neuronal REST peaks was marginal , occurring in 22 sites ( p = 4 . 9E-64 from MEME [36] ) . A second enriched motif , GGAAA/TA , was detected among these peaks ( n = 180 , p = 1 . 6E-50 ) ( Fig . S2 ) . It is similar to the DNA motifs recognized by transcription factors NFATC2 , dl_2 and EDS1 , and it was found in 12% of the total H1-derived neuron REST peaks , compared to 3% for the T cell REST peaks and 9% of randomly selected genomic sequences . A comparison of the REST-bound genes further demonstrated the distinction between REST targeting in T cells and neurons . The 4 , 404 peaks identified in T cells were associated with 3 , 307 Refseq [37] genes and miRNAs [38] , while the 5 , 387 peaks in H1-derived neurons were associate with 3 , 389 genes/miRNAs . Despite a previous report that RE1 motifs were prominently distributed in introns [24] , 51% and 41% of the REST peaks were localized to promoter regions ( −5 kb to +1 kb from transcription start sites , TSSs ) in T cells and neurons , respectively . Other than promoter regions , 28% and 25% of the REST peaks were found within intragenic regions and another 14% and 12% were within 50 kb of genes ( Fig . 1C ) . In total , 93% and 78% of the T-cell and neuronal peaks were assigned to one or more genes , respectively . Furthermore , functional analysis of the REST-bound genes using GREAT [39] revealed a dramatic disparity in the enriched pathways between these two cell types . One of the pathways defined by Pathway Commons [40] that was enriched in T cells showed clear involvement in neuronal function: neuronal system ( p = 1 . 9E-5 ) ( Fig . 1D ) , the other top enriched pathways , however , were related to functions important for lymphocytic cells . We found different categories of REST targets in neurons; the top four enriched pathways were involved in general gene expression ( p = 1 . 1E-38 ) ( Fig . 1D ) . The primary known REST function is to repress neuronal genes in non-neuronal cells [41]–[43] , our data agreed with this , as this function was only found to be enriched in the T-cell REST targets . Only 487 ( 11% ) of the T cell peaks overlapped with those from neurons , and the majority ( n = 259; 53% ) of these contained either a cRE1 or ncRE1 motif . Notably , 655 ( 73% ) of the 894 genes targeted in both cell types had at least one peak that was detected in only one of the two cell types . Those common REST targets were enriched for neuronal functions , so were the T cell only REST targets , but not the neuron-only REST targets . These results suggest that REST cistromes of non-neuronal and neuronal systems may share limited overlap . To extend our observation of distinct REST occupancy between neurons and non-neuronal cells , and also to gain insight into the dynamics of REST cistromes across human cell types , we decided to explore more publicly available data from the ENCODE project [34] and expand our comparison to include fourteen additional human cell lines: alveolar adenocarcinoma cells ( A549 ) , endometrial carcinoma cells ( ECC1 ) lymphoblastoid cells ( GM12878 ) , embryonic stem cells ( H1 ) , colon carcinoma cells ( HCT-116 ) , cervical adenocarcinoma cells ( HeLa S3 ) , liver hepatocellular carcinoma cells ( Hep G2 ) , promyelocytic leukemia cells ( HL-60 ) , erythroleukemia cells ( K562 ) , breast adenocarcinoma cells ( MCF-7 ) , pancreatic carcinoma cells ( PANC-1 ) , primitive neuroectodermal tumor cells ( PFSK-1 ) , neuroblastoma cells ( SK-N-SH ) , and glioblastoma cells ( U87 ) ( Table 1 and Table S1 ) . These cell types represent a number of lineages . The SK-N-SH cell line is particularly interesting as it allows us to perform a comparison between normal neurons and tumorgenetic neuroblastoma cells , which have been reported to be associated with increased REST expression [44] . The resulting peak numbers from our ChIP-seq analysis pipeline are shown in Table 1 , and range from 2 , 048 ( in U87 ) to 8 , 199 ( in H1 ESCs ) . ( See Table S2 for list of peaks ) . We next evaluated REST binding sites for cell specificity . Based on the above described difference in REST binding sites between T cells and neurons and the expectation of REST repression of neuronal genes in non-neuronal cells , we decided to compare REST cistromes across all 15 non-neuronal cells first , and then brought in neurons for a final comparison . Noted that we considered those tumor cell lines derived from neural tissues ( e . g . , SK-N-SH ) as “non-neuronal” in this report . After overlapping peaks were merged , we obtained a set of 16 , 913 non-redundant REST binding regions from the total of 61 , 801 peaks in the 15 non-neuronal cell types . Analysis of the ChIP-seq signals showed different levels of REST enrichment across these peaks among the 15 cell types ( Fig . 2A ) , and to our surprise , only 1 , 116 ( 7% ) of the merged REST peaks were consistently called by SPP in all these cell types ( referred to as “common” peaks ) . Nevertheless , 7% is much more than expected by chance , since we obtained 0 in common when we randomly picked genomic regions , with total number and size distribution matching to those of the REST peaks in individual cell lines , and performed the same merging procedure . A similar small fraction of REST peaks were found to be common in a previous analysis of REST bindings in 1% of the human genome [33] . Interestingly , these common peaks indeed exhibited the greatest enrichment of REST ChIP-seq signals in all cell types ( Fig . 2A/B ) . To study cellular specificity of REST occupancy more robustly , we compared several quantitative metrics for evaluating REST ChIP-seq signals at individual peaks for differential binding across cell types ( see Supplementary Methods for more details ) . In brief , we analyzed the number of ChIP-seq reads at the summits of the non-redundant peaks using the program seqMiner [45] and then computed Z-scores to detect cell types with significantly more ChIP-seq reads than the rest . The results showed that ChIP-seq signals for 2 , 690 ( 16% ) of those non-neuronal REST peaks were significantly stronger in one cell type than in any other cell type; these were termed “cell-specific” peaks ( Fig . 2A/D and Table 1 ) . Nevertheless , a large fraction ( 77% ) of non-neuronal REST binding sites were shared by at least two or more cell types ( referred to as “shared” peaks ) , as illustrated in Fig . 2A . In a comparison of the common , shared ( i . e . , 2–14 cell types ) and cell-specific peaks , we found that common peaks were more enriched with the cRE1 motif ( 86% ) than non-common ones ( 53% ) . The GGAAA/TA motif identified in neurons was not particularly enriched in any of the other cell types , as it occurred in 5 . 4% of the combined non-neuronal REST-binding sites ( binomial test , p = 1 . 7E-133 ) . Although 78% ( n = 4 , 240 ) of the neuron peaks were called only for this cell type by SPP , 22% of them exhibited significant ChIP-seq signals in other cell types as well , thus resulting in 62% of neuronal REST peaks specific to neurons by our definition . This change indicates that comparison of transcription factor occupancy across samples by simple intersection of the genomic coordinates of the ChIP-seq peaks could exaggerate the true difference significantly . Interestingly , even for the neuronal REST peaks overlapping with peaks in other cell types , the peak summits were often shifted slightly to a new position in neuronal chromatin . While the average distance between the summits of overlapping peaks found in pairs of non-neuronal cells ranged from 6 bp ( GM12878 vs . Hep G2 , MCF-7 vs . HL-60 , and MCF-7 vs . K562 ) to 24 bp ( A549 vs . T cell ) , the mean distance of REST peak summits between neurons and other cell types ranged from 26 bp ( vs HCT-116 ) to 79 bp ( vs T cell ) ( Fig . S3 ) . This observation again reveals the distinction of REST occupancy in neuronal cells . To address if chromatin factors may contribute to cell-specific REST binding , we analyzed available DNase-seq data from the ENCODE project and related them to REST binding in A549 , GM12878 , Hep G2 , H1 ES , K562 , MCF-7 and T cells . We had expected that chromatin regions bound by REST would have greater DNAse-seq signals than “potential” but unbound REST candidate sites ( i . e , REST-bound in other cell lines ) . While this was true in four of the cell types ( A549 , K562 , MCF7 and T cells ) , with REST-bound regions showing 1 . 5–3 . 0× more overlapping with DNAse-seq peaks , no difference was observed for GM12878 , Hep G2 and H1 cell lines . Nevertheless , we found that DNase-seq signals ( measured by read densities ) at the sites with stronger REST occupancy were generally higher than sites with weaker REST occupancy in all of these seven cells ( data not shown ) . On the other hand , DNA-seq signals at many unbound REST candidate sites still showed much greater DNase-seq read enrichment in comparison to adjacent genomic regions . Taken together , these results indicate that chromatin accessibility is not the critical factor determining REST occupancy and thus the dynamics of REST cistromes . This observation is consistent with previous finding that neither DNase hypersensitivity nor chromatin features were a good predictor of REST binding [46] . Next , we compared the genes and miRNAs [37] , [38] that were bound and thus potentially regulated by REST . Similar to REST peaks , the numbers of REST targets varied from one cell type to another ( Table 1 ) , with a total of 10 , 286 genes and miRNAs bound in at least one of the 16 cell types . There was approximately a 3-fold difference between the cell type with the highest ( H1 ESCs , 4 , 509 ) and the one with the lowest ( U87 cells , 1 , 682 ) number of REST targets . Five pathways: neuronal system , GPCR ligand-binding , potassium channels , transmission across chemical synapses , voltage-gated potassium channels were identified as significantly enriched in the REST targets for >10 cell types ( Table S3 ) . All of these pathways are important for neuronal function . Interestingly , pathways involved in translation ( e . g . , peptide chain elongation ) were significantly enriched in REST targets in A549 , HL-60 , PFSK-1 , and SK-N-SH cells , along with neuronal pathways , and in REST targets in neurons , but to the exclusion of top neuronal pathways ( Table S4 ) . Notably , REST-bound genes in these pathways were predominantly the same set of targets shared by these cell types . In addition , nearly all of the genes ( n = 856 ) targeted by REST in all 15 non-neuronal cells contained a common REST peak ( i . e . , called in all cells ) . Among these common genes , only 1 . 2% ( n = 10 ) were bound by REST at different genomic sites in any of the 15 non-neuronal cell types and 41% ( n = 353 ) were also bound by REST in neurons . Interestingly , one of those genes targeted by REST in all cell types except neurons was REST itself , for which negative auto-regulatory feedback has been proposed [24] . REST was found to bind proximally to many of the well-characterized neuronal genes in various brain cancer cell lines: PFSK-1 , SK-N-SH , U87 and in H1-derived neurons , a phenomenon previously reported [41]–[43] . We compiled a list of 15 known REST target genes from the literature , including BDNF [43] , [47] , CALB1 [48] , L1CAM [49] , CHAT [50] , GRIA2 [51] , CHRM4 [52] , NRCAM [53] , GRIN1 [54] , STMN2 [52] , SCG2 [55] , SYN1 [52] , SYP [56] , SYT4 [48] , GLRA1 [52] , CHRNB2 [52] . All of these genes were REST-bound in 14 or more cell types . Among these , 6 of them ( GLRA1 , GRIA2 , SCG2 , CALB1 , STMN2 , CHRNB2 ) were bound in all 16 cell types , including neurons . The remaining nine genes displayed variable binding , with any lack of REST occupancy occurring in only four cell types: neurons , HCT-116 cells , PANC-1 cells , or Hep G2 cells . This result indicates that our observation of neurons as an “outlier” is unlikely due to some experimental technical biases ( e . g . , off target immunoprecipitation ) in ChIP-seq analysis . Instead , it suggests that the prevalent view of REST having similar functions in non-neuronal systems through the repression of neuronal genes may have arisen from a systematic experimental bias that the same small set of genes has been examined repeatedly in previous studies . In addition , of the 52 genes that were upregulated upon REST knockdown in HEK293 cells [57] , 54% ( n = 28 ) of them showed variable REST binding among the 16 cells . 11 of these genes exhibited differential REST occupancy in neurons , with 10% ( n = 5 ) of them bound by REST exclusively in neurons and 12% ( n = 6 ) bound in other cell types , but not in neurons . This result suggests that a large fraction of genes repressed by REST are direct REST targets in non-neuronal cells . As REST has been previously implicated in a variety of cancers , we decided to look into whether there were any cancer specific REST targets . A comparison of the REST occupancy in differentiated cell types ( T cells and neurons ) with the 13 cancer-derived cell lines revealed that several tumor suppressor genes ( OSMR , MYO1A , THRB , FRMD3 , LOXL4 , CEACAM3 , TRH ) were bound by REST in all of the cancer cell lines , but in neither neurons nor T cells . Conversely , IL-7 receptor ( IL7R ) and LAMA2 , two genes that are upregulated in a number of cancers [58] , [59] , were targeted by REST only in the two non-tumorigenic differentiated cell types . Notably , in H1 ESCs the REST binding patterns at all of these genes ( except LAMA2 ) matched with those in the tumor cell lines , suggesting that REST regulation of these genes may have a role in cell proliferation , since active growth is a common feature of ESCs and tumor cells . We next sought to better understand the transcriptional effect of REST occupancy , by integrating REST cistromics data and transcriptomics data . We utilized RNA-seq data ( Table S5 ) and the TopHat/Cufflinks software suite [60] to determine gene expression levels in all of the cell types ( Fig . S4 ) . Hierarchical clustering analysis of gene expression showed that cell types of similar lineages and functions were grouped together , affirming the quality of our RNA-seq data ( Fig . S4 ) . We took the combined REST targets ( n = 10 , 286 ) in all 16 cells and then for a particular cell type we compared the transcription of the bound ( b ) to the unbound REST candidates ( ubrc = 10 , 286-b; approximate for potential REST targets ) as well as to the transcription of all genes . For the majority ( 13 ) of the 16 cell types , REST bound genes were transcribed at a level significantly lower ( 2 . 4–36 fold lower; p<2E-16 ) than the unbound REST candidates ( Fig . 3A ) , congruent with REST's primary role as a repressor [9] . In addition , in the majority ( 11 ) of the cell types lower expression of REST-bound genes compared to all genes was observed ( significant in 8; p<2E-5 ) . In A549 cells , T cells and neurons , REST-bound targets exhibited a significantly higher expression ( Fig . 3A and Table S6 ) . The predominantly repressive function of REST was further supported by very low expression of the 15 known REST targets described above ( medians of FPKMs were 0 . 02–1 . 1 in all non-neuronal cell types but 16 . 2 in neurons ) . We were not surprised to find that REST targets were highly expressed in neurons as it has been reported that REST could bind and activate neuronal genes [9] . We were , however , surprised to see that REST-associated genes also exhibited greater expression than unbound REST candidate targets in A549 cells and T cells . Out of the top 200 most expressed genes in individual cell types , 37%–59% were REST-bound in A549 cells , neurons or HL-60 cells , while <23% were bound by REST in all other cell types . Interestingly , neurons , T cells and A549 cells had high proportions of REST peaks located to promoters and a higher percentage of REST peaks without an RE1 motif than the other cell types , the significance of which needs further investigation . Current data also allows us to make an interesting comparison between the SK-N-SH neuroblastoma cell line and neurons . It has been reported that elevated expression of REST is associated with neuroblastomas [44] . RNA-seq data showed that REST was indeed more highly expressed in the SK-N-SH line ( 6 . 3 FPKM ) than neurons ( 2 . 7 FPKM ) . Relatively few REST peaks ( n = 386 ) overlapped in these two cell types . We found that a variety of neuronal pathways were enriched among the genes that were expressed at a lower level in SK-N-SH and bound by REST in SK-N-SH cells but not in neurons , while many genes involved with translation and RNA processing were uniquely bound and more highly expressed in neurons . As expected , genes in neuronal system pathways had higher expression in neurons than in SK-N-SH cells ( median FPKM of 5 . 6 vs . 3 . 0 ) , but even those bound by REST only in neurons ( n = 17 ) were highly expressed ( median FPKM 34 . 7 vs . 0 . 3 FKPM for those bound in SK-N-SH cells only ) . Interestingly , the tumor suppressors TRH and MTSS1 were not expressed in SK-N-SH cells ( 0 and 0 . 27 FPKM , respectively ) but were in neurons ( 3 . 7 and 20 FPKM , respectively ) . In contrast , the oncogene β-catenin and proto-oncogene IL-7 were much more highly expressed in SK-N-SH cells ( 99 and 4 . 7 FPKM , respectively ) than neurons ( 61 and 0 . 4 FPKM , respectively ) . These data support the view that REST may have a direct functional role in cancer progression by regulating oncogenes and tumor suppressors and the point that REST does not always repress its targets . Previous studies show that REST binding does not always lead to gene repression and that in some cases it is conversely correlated with gene activation . In addition , it has been shown that the sequence bound by transcription factors can determine cofactor specificity [61]–[64] for a number of proteins such as for the transcription factor NF-κB [61] , hormone activated estrogen receptors [62] , and the glucocorticoid receptor [63] , [64]; thus , we wondered if the context of REST binding plays a role in gene regulation . By analyzing the REST peak numbers , we found that genes with more REST peaks generally exhibited lower levels of expression than those with a single peak . This observation persisted in all 16 of the cell types and was statistically significant in 15 ( Fig . 3B shows data for GM12878 , Hep G2 , T cells and neurons; data for all cell types in Table S6; about 1 . 5–68 fold lower ) . This finding suggests that there may be an additive dosage effect of REST occupancy on the repression of its targets . We also found that genes with REST peaks containing RE1 motifs ( either cRE1 or ncRE1 ) generally exhibited lower expression levels than those without RE1 motifs or with half-sites , consistently across all cell types ( Fig . 3C , Fig . S5 , and Table S6; about 1 . 4–164 fold lower ) . Intriguingly , despite the low occurrence of RE1 motifs in neuron peaks , this trends held . However , even the REST bound genes with RE1 motifs had a higher expression level in neurons than in any other cell types ( median of 5 . 6 FPKM in neurons vs . <1 . 3 FPKM in others , Table S6 ) . Interestingly , genes with ncRE1-peaks tended to exhibit even lower levels of expression than those with cRE1 , in agreement with a previous report [13] . This also persisted in 14 of the cell types and was statistically significant in 10 ( Fig . 3C , Fig . S5 , and Table S6 ) . Notably , REST genes with RE1 motifs had lower expression than those without RE1 motifs and overall stronger peaks were associated with lower gene expression , except in neurons ( data not shown ) . Next , we examined whether genes consistently bound by REST in all cell types were as lowly expressed as genes that were bound variably ( shared or cell-specific ) . Indeed , we found that genes with a common REST binding site exhibited lower levels of expression than those with shared or cell-specific sites ( Fig . 3D and Table S6; about 1 . 7–195 fold lower ) . Interestingly , the highest levels of expression for these common genes occurred in neurons exclusively ( median expression of 4 . 4 FPKM ) . Finally , we compared the expression of different groups of REST targets that were separated based on the relative locations of REST binding . In 11 of the cell lines , genes bound by REST in their bodies ( introns or exons ) exhibited significantly lower expression levels than those with REST in their promoters ( data for all cell types in Table S6; about 1 . 4–69 fold lower ) . This is quite interesting , since the effect of REST on gene expression has been mostly studied through its binding at promoter regions . On the other hand , it has been reported that REST binding within 50 bp of the TATA box in neuronal cells ( but not in non-neuronal cells ) was correlated with gene activation [18] . Indeed , those genes bound by REST at their promoters with peak summits located <50 bp from the TSS exhibited higher expression levels ( Table S6; 1 . 8–5 . 4 fold higher ) than even unbound REST candidate genes . This difference was found in 9 of the cell types , and was statistically significant in 7 . This finding is probably related to the emerging view that transcription factor binding at enhancers has a greater effect on gene expression than binding at promoters , where many factors likely act competitively or coordinately . As previously mentioned , data from A549 cells , T cells and neurons did not indicate repression of REST targets . No dosage-associated expression difference was detected for REST binding in neurons . Nevertheless the presence of RE1 motifs , cell specificity , and location of REST peaks still made a difference in terms of the extent of REST repression ( Table S6 ) . These three cell types had the largest percentages of REST peaks in the promoter regions ( 34 . 6% , 50 . 6% , and 41 . 2% for A549 cells , T cells , and neurons , respectively , in comparison with 14 . 2–28 . 1% of peaks in other cell types ) and the smallest percentages of their peaks contained a cRE1 or ncRE1 motif ( 48 . 2% , 40 . 6% , and 8 . 5% A549 cells , T cells , and neurons , respectively , in comparison with 51 . 3%–83 . 8% in other cell types ) . Both of these two features showed a strong bias towards higher expression of the REST targets . Since these aforementioned factors were not independent , we performed a logistic regression to determine the individual contributions of these factors to predict the transcription outcome of a REST target as expressed ( FPKM≥1 ) or not expressed ( FPKM<1 ) . The results indicated that genes next to RE1 motif peaks , common REST peaks , and intragenic/distal peaks were 2 . 6 , 1 . 9 and 1 . 3 times more likely to not be expressed than genes next to RE1-free peaks , non-common REST peaks , and promoter peaks , respectively , suggesting that motif status and cellular context were the two primary factors . The above analyses show that context plays a role in REST regulation of its targets , which brings up a question as to whether this is due to different sets of REST cofactors being recruited . To address this , we investigated the co-occupancy of REST with CoREST , SIN3 , and EZH2 ( a core component of PRC2 ) using the ENCODE project ChIP-seq data from the two cell types: GM12878 and Hep G2 [34] . There were 17 , 590 SIN3 , 44 , 065 CoREST , and 64 , 277 EZH2 peaks in GM12878 . The corresponding numbers in Hep G2 cells were 32 , 019 SIN3 , 51 , 883 CoREST and 79 , 275 EZH2 peaks . Note that we merged the two sets of SIN3 peaks in Hep G2 . First , we found that 14% , 29% and 43% of the GM12878 REST peaks overlapped with SIN3 , CoREST and EZH2 , respectively . Interestingly , a previous analysis also showed that RE1 motifs were highly enriched in the SIN3A-occupied genomic sites in H1ESCs [65] . Further analysis showed that only a small fraction ( 7% ) of REST-bound regions had all three cofactors , while a large percent ( 42% ) were not occupied by any of these cofactors at all ( Fig . 4A ) . In addition , most ( 85% ) of the SIN3-REST co-binding regions were localized to sites bound by CoREST , and a large fraction ( 36% ) of the REST-EZH2 sites were co-occupied by CoREST . Next we determined the enrichment of the contexts in each of the groups with different combinations of REST and its cofactors . We found that REST-SIN3 peaks were 2 . 7 times more likely than expected ( in relation to all REST peaks ) to be in promoters , 1 . 7 times more likely to contain either a half RE1 motif or no RE1 motifs . It should be noted that the TSS proximity of SIN3 has been reported previously [66] . REST-CoREST peaks were 1 . 5 times more likely than expected to be REST-bound in all 15 cell types ( i . e . common REST peaks ) . REST-EZH2 peaks did not show an increased association with any of the examined features . Cofactor-free REST sites showed a depletion of both promoter and common REST binding ( Fig . 4B ) . These observations were generally reproduced with data from the Hep G2 cell line ( Fig . S6 ) . We then asked how variable cofactor association played into the transcriptional regulation on REST targets . A previous study [15] has shown that in mouse ESCs , REST-binding sites colocalized with SIN3 or CoREST occurred more frequently in genes whose transcription was directly repressed by REST , as measured by their upregulation upon shRNA knockdown of REST expression . Since we did not have REST knockdown data in any of the cell types studied here , we contrasted REST targets that were bound by cofactors to all REST targets . In contrast to the previous finding [15] , we found that REST-bound sites with SIN3 and CoREST were located to genes exhibiting 28 and 4 times higher transcription , respectively , when compared to all REST targets in GM12878 ( Fig . 4C; data for Hep G2 in Table S7 ) . This was due to the strong positive association between SIN3 occupancy and active transcription , since REST-SIN3 only targets were expressed 47 times higher than REST targets ( Table S8 ) and SIN3 was the transcription factor most associated with active expression among the 3 cofactors assessed . This result is surprising because SIN3 interacts with HDACs and reduction of histone acetylation is often associated with gene repression , although a previous study has reported that HDACs were localized to many active genes [67] . Moreover , REST sites colocalized with SIN3 , CoREST and EZH2 were actually found at genes ( e . g . , CHRNB2 ) that were expressed 26 times more highly than the median of all REST targets . REST sites colocalized with EZH2 and CoREST , either alone or together , were associated with genes ( e . g . , DRD3 , HTR3A , and BDNF , respectively ) expressed at a level of 2 . 0 , 3 . 6 , and 1 . 9 times lower , respectively , in relation to all REST targets . REST targets without any of the three cofactors were expressed , unexpectedly , at 2 . 5 times lower levels than all targets in GM12878 . The expression difference associated with differential cofactor binding generally held true by examining either all REST peaks or only promoter peaks . In addition , similar results were obtained with data from Hep G2 cells ( Table S7 ) , except that in Hep G2 the REST targets not exhibiting colocalization with any of the three cofactors were actually expressed at levels 3 . 5 times higher than all REST-bound genes . In Hep G2 cells , the REST sites that colocalized only with CoREST were also more highly expressed . The opposite trends observed for REST-alone targets and the REST-CoREST “only” targets between GM12878 and Hep G2 cells suggest that a different set of other REST cofactors may have been recruited to those targets , also in a cell-specific manner . It would be interesting to study in the future if G9A , CTBP , MECP2 , LSD1 or other yet-to-be-identified REST interactors are involved . Notably , we did not observe pathways specifically enriched in any group of REST targets with different cofactor occupancies . In summary , our results indicate that SIN3 co-localization was correlated with higher expression while EZH2/CoREST co-occupancy was associated with lower levels of expression of REST targets , with SIN3 seemingly dominant over EZH2/CoREST . In the future , it would be interesting to test experimentally if SIN3 and EZH2/CoREST indeed confer opposite regulatory roles in some REST targets by knocking down these chromatin factors . The fact that the outcome of REST regulation is largely dependent on its cofactors is not entirely surprising , but it reinforces the view that REST is a molecular platform for recruiting chromatin modifiers , which ultimately determine the transcription activity . To address this computationally , we combined the cofactor colocalization information with histone modification data in GM12878 cells from a previous study [68] . We observed that 82% of the REST-EZH2-only , 74% of the REST-CoREST-only , and 19% of REST-SIN3-only sites were located to regions enriched with H3K27me3 ( a repressive histone mark ) . On the other hand , 74% and 48% of the REST-SIN3-only , 6% and 2% of REST-CoREST-only and 12% and 3% of REST-EZH2-only sites were located to either H3K4me3 ( an active histone mark ) or H3K27ac ( an active histone acetylation mark ) enriched regions , respectively . These enrichments are congruent with the known addition of H3K27me3 by EZH2 and removal of H3K4me by LSD1/CoREST . SIN3 acts on the chromatin through the recruitment of HDACs . Since there was no HDAC ChIP-seq data in GM12878 cells , we only studied the colocalization of SIN3 and HDAC2 in Hep G2 cells . A total of 32 , 895 HDAC2 peaks were called for Hep G2 . We found that 69% of the REST sites that colocalized with HDAC2 ( n = 374 ) were also enriched with SIN3 ( binomial test , p = 7 . 5E-129 ) , supporting SIN3-HDAC interaction at REST-bound sites . All together , our study demonstrates that distinct combinations of chromatin modifying cofactors are recruited to different REST-binding regions , and that they likely contribute to the transcriptional outcome of REST regulation . Whether and how these cofactors work together with REST to activate or repress gene expression cannot be directly addressed here , due to the limitation of computational work , and thus requires more study in the future . Throughout our analysis the H1-derived neurons stood out from the other cell types . For instance , neuronal REST peaks were enriched with a different motif and REST targets in neurons were overall highly transcribed . This raises the question of whether different cofactors are recruited to the REST-bound sites in neurons . Unfortunately , TAF1 and RNA polymerase II were the only DNA-binding proteins that had been analyzed by ChIP-seq in the same neuron samples , both of which are unlikely to play a deterministic role in REST-specific gene regulation . We have , however , analyzed several available genomic and epigenomic data sets in order to gain a glimpse at the neuron-specific REST function . As it has been suggested that a REST isoform ( REST4 ) could inhibit REST function and activate REST targets [9] and that REST4 expression is neuron-specific [44] , we examined whether this isoform was more abundant in neurons . Our analysis of RNA-seq reads mapped to REST4 specific exon , however , did not find evidence that REST4 was the dominant REST isoform in neurons . In fact , in comparison to the full length REST ( 2 . 7 FPKM ) , REST4 transcription was 7-fold lower at 0 . 39 FPKM ( Fig . S4 ) . Next , we asked if the presence of small RNAs within REST-binding sites might have altered REST-mediated gene regulation , since it has been shown that the transcription of enhancer RNAs ( eRNAs ) is often correlated with gene activation [69] , [70] and that a double stranded small RNA was shown to activate REST targets [71] . We utilized small RNA sequencing data from the ENCODE project for H1-derived neurons , GM12878 , and H1 ESCs [72] , and determined small RNA read abundance at individual REST peaks using the algorithm HTSeq [73] . Interestingly , we found in neurons that ( i ) 804 of the REST sites ( 15% ) had small RNAs mapped to them ( ≥1 reads per kb per million small RNA-seq reads ( RPKM ) ; binomial test , p<2 . 2E-16 ) , and ( ii ) genes associated with these REST peaks ( n = 761 ) were expressed at significantly higher levels ( median FPKM = 15 . 0 ) than all REST bound genes ( p = 2 . 8E-39 ) . In contrast , very few genes were associated with REST peaks that could potentially produce eRNAs by the same criterion in both GM12878 ( n = 33 ) and H1 ESC ( n = 69 ) . This difference remained when the threshold for small RNA presence was set to 0 . 1 RPKM ( data not shown ) . Next , we examined the chromatin modifications at REST binding sites . We obtained chromatin regions that were determined to be enriched with H3K27me3 , H3K36me3 , H3K4me1 , H3K4me3 , and H3K9me3 from a previous report , in which these modifications in H9-derived neurons , GM12878 cells , H1 ESCs , and many other tissues were studied [68] . Intersecting those histone modification regions with the REST peaks in neurons , GM12878 , and H1 ESCs demonstrated that a much greater percentage ( 49% ) of neuron REST peaks overlapped with H3K4me3 , an active histone modification , than did GM12878 ( 20% ) or H1 ESC ( 18% ) REST peaks . This finding was further supported by the high enrichment of H3K4me3 ChIP-seq signal at the center of REST peaks in neurons but not in GM12878 cells , H1 ESCs , A549 , HeLa S3 , Hep G2 , or K562 cells ( Fig . 5A ) . The overlaps of REST peaks with chromatin regions enriched with two additional active histone marks: H3K4me1 and H3K36me3 , were also 2× higher in neurons than in either GM12878 or H1 ESCs ( Table 2 ) . Conversely , a significantly greater percentage ( 69% ) of GM12878 REST peaks overlapped with H3K27me3 regions , a repressive histone modification , than did neuron REST peaks ( 27% ) . This finding was further supported by the enrichment of H3K27me3 ChIP-seq signal at the center of REST peaks in GM12878 , Hep G2 and other cell lines , but not in H1-derived neurons ( Fig . 5B ) . Although only 18% of the REST peaks in ESCs overlapped with H3K27me3 regions ( Table 2 ) , H3K27me3 ChIP-seq signals showed significant enrichment at REST peaks in ESCs ( Fig . 5B ) , which was upon further examination due to strong H3K27me3 signals in these 18% overlapping peaks ( no enriched profile detected after their removal , data not shown ) . Similarly , a higher overlap of REST peaks with H3K9me3 regions was observed in GM12878 cells compared to either neurons or ESCs . It is possible that lower cofactor colocalization at these neuron REST sites , could explain the higher levels of H3K4me1/3 and lower levels of H3K27me3 and H3K9me3 . Likewise , for the 15 known REST target genes , we found much higher association of their REST sites with repressive histone marks outside neuronal cells , 14 of the 15 genes with H3K27me3 , and 4 of the 15 with H3K9me3 in GM12878 compared with 6 of the 10 genes with H3K27me3 and none with H3K9me3 in neurons . These differential histone modification colocalizations go hand in hand with the higher expression of these genes in neurons based on RNA-seq data . This finding underscores the hypothesis that distinct sets of REST cofactors are recruited to REST-bound regions in a context-dependent manner , as these 15 genes are occupied by REST across cell types but exhibit differential enrichment of histone marks . We next examined how REST differentially regulated a mini regulatory circuitry that has previously been suggested to be important for neurogenesis . The circuitry includes REST cofactors , several miRNAs , and neurogenic factors critical for neuronal development ( Fig . 6 ) . Among the known components of the REST complex [9] , [11] , [12] , [48] , LSD1 , BRG1 , HDAC4 , and HSPC1 ( a component of PRC1 ) were bound by REST in 5 or more cells , frequently including ECC-1 , H1 ESCs , HL-60 , PANC-1 and SK-N-SH cells . EZH2 showed REST binding at its promoter in neurons but not other cells ( see Table S2 ) . Finally , we examined the expression of and REST-binding at several miRNAs that have been reported to control neural development [74]–[77] . The expression levels of miRNAs were determined from small RNA-seq datasets . Of the 939 miRNAs annotated in miRBase [38] 134 of them ( 14% ) were bound by REST in at least one of the 16 cell types and 39 of them ( 4 . 2% ) were differentially bound in neurons and non-neuronal cell types . 32 of the 39 REST-bound miRNAs in neurons ( 82% ) were bound either only in neurons ( n = 20 ) or at a different genomic position in the non-neuronal cells ( n = 12 ) . Five miRNAs have been extensively studied for their interaction with REST and their roles in promoting neurogenesis: miR-9 [9] , [78] , [79] , miR-124 [9] , [79] , miR-132 [9] , [79] , [80] , miR-135b [78] and miR-212 [80] , many of which directly target neuronal genes . Interestingly , each of these miRNAs showed a different pattern of REST binding in neurons and non-neuronal cells ( Fig . S7 ) , with miRs 124-3 , and 132/212 targeted by REST in all cell types but neurons , with miRs 9-3 , 124-1 , 124-2 , and 135b targeted at a position in neurons different from other cells , and miR-9-2 bound only in neurons . Compared to their expression in H1 ESCs and GM12878 cells , all of these miRNAs were more highly expressed in neurons ( Table 3 ) , suggesting that REST may play a role in activating instead of repressing these miRNAs in neurons . In support of this hypothesis , the expression of miR-9 was indeed downregulated in mouse neuronal stem cells upon conditional knockdown of REST expression [81] .
In this study , we have characterized genome-wide REST occupancy across multiple cell types and related that to other transcription factor binding and transcriptional outcome . With REST ChIP-seq data from 16 different human cells , we made an attempt to estimate the number of potential in vivo REST binding sites in the human genome . Although we called a total of ∼62 , 000 REST peaks , they were highly redundant . 63% of the total bindings in the 15 non-neuronal cell types could be identified using data from just three cell types with the most peaks ( Fig . S8 ) . Starting from the cell type ( H1 ESC ) with the most peaks , we computed the number of new REST peaks that were identified by including data from additional cells . As shown in Fig . S8 , the number of additional peaks added with each cell line decreases and approaches a plateau at about 7 cell types . The extrapolation of our data indicates that there are ∼18 , 000 potential REST-bound regions in the human genome . This estimation , however , excludes neurons , because our analysis shows that REST binding in neurons is quite distinct from non-neuronal cell types and much is yet to be learnt for the diversity of REST bindings among neuronal subtypes [7] , [30] . We should note that of the genomic sites previously predicted to contain cRE1 motifs [13] , almost all ( 234/235 ) of the most highly confident ones were included in our REST peaks and additionally 85% ( 669/783 ) of those lower confidence peaks in non-repetitive regions were also covered by our REST peaks . Our findings indicate that the REST cistrome in differentiated human neurons is dramatically different from those in non-neuronal cells . This difference , as well as differences in the dynamics of REST targeting , was also observed in previous studies of REST occupancy at promoters during the development of a variety of mouse neural lineages using microarrays [27] , [30] , [31] . The limitation of the current study is that the samples used for ChIP assays contained a mixture of neuronal subtypes , which may have differences in REST occupancy . The maturation status of the cultured neurons could be another important factor . From that perspective , the differences between the neuronal REST cistrome and that of non-neuronal cells may actually be larger than what we report here , since the limited overlapping could be due to the presence of a small number of neuronal progenitor cells in the ChIP samples . We should mention that RNA-seq data for our analysis of gene expression in neurons were not derived from the same sample used for ChIP-seq experiment: H1 ESC derived neurons for ChIP and iPSC derived neurons for RNA-seq . The differentiation protocols , however , were essentially the same and promoted the generation of GABAergic and Glutamatergic neurons , and both samples were harvested after four weeks of culture . The enrichment of neurons in our cell culture was supported by the high expression of neuronal and synapses genes [82]: ( TUBB3 , 431 FPKM; MAP2 , 141 FPKM; MAPT , 86 FPKM; SYN1 , 16 FPKM; DLG4 , 41 FPKM ) , low expression of genes marking astrocyte and oligodendrocytic glia [82] ( GFAP , 6 FPKM; MBP , 0 . 5 FPKM; OLIG2 , 0 . 7 FPKM ) , as well as low expression of NSC/NPC markers [83] . We should point out that REST expression in Glutamatergic and GABAergic neurons has been demonstrated previously by co-staining of REST and neuronal specific marks ( see Fig . S2 in ref [30] ) and recently mRNA and protein expression in neurons in the prefrontal cortex has been reported [84] . In addition , 21% and 25% of REST promoter targets previously found in mouse GABAergic and Glutamatergic neurons [30] were also identified in our neuron ChIP-seq data , respectively . We do not think the sample “mismatch” undermines our finding that REST targets exhibited active expression in neurons but lower expression in non-neuronal cells , because results from our analysis of histone modification ChIP-seq data collected for another neuronal sample ( derived from H9-ESC line ) also support our conclusion . Nevertheless , it will be important to revisit this issue when the RNA-seq data for H1-derived neurons becomes available from the ENCODE project . The lack of RE1 motif enrichment in the neuronal REST peaks is quite surprising and intriguing . Although motif analysis identified GGAAA/TA as a potential alternative binding sequence , we do not think it represents true REST recognition motif in neurons , because it is a rather common motif and present in 9% of randomly selected genomic sequences . Instead , a more likely scenario is that REST loses its direct interaction with neuronal chromatin and becomes a co-factor for other transcription factors . Our study showed that small RNAs were enriched in the neuronal REST peaks , but more direct experimental assays are needed to investigate if small RNAs or long non-coding RNAs have a role in altering REST interaction with chromatin and its targets . REST4 has been associated with active gene expression in neurons . Although we did not uncover evidence that REST4 was the dominant isoform in neurons , we did observe that REST4 transcript was increased 3-fold in the transition of day-14 to day-27 neurons ( data not shown ) . Furthermore , our RNA-seq data indicated that SRRM4 , shown to promote alternative splicing to generate REST4 transcript [85] , was transcribed only in neurons , with 14 . 0 FPKM in neurons but <0 . 95 FPKM in all non-neuronal samples . This suggests that REST4 could be the dominant REST protein product in our neuron cultures . Nevertheless , REST4-bound chromatin sites would still be expected to be enriched with variants of RE1 motifs . Therefore , it remains a mystery how REST changes its DNA recognition specificity in neurons . Only with further study can we address if REST interaction with the neuronal genome is mediated by other factors , like in the case of recently reported cellular dependent chromatin interaction of TCF7L2 [86] . We need to point out that the lack of RE1 motif enrichment has also been reported in a previous analysis of REST regulation in mouse neurons [30] . We should also emphasize that two ChIP-seq replicates from the H1-derived neurons showed consistent REST binding and the same lab produced all the REST ChIP-seq data with the same antibody , except those from CD4+ T cells . Thus , we do not believe the unique features of neuronal REST cistrome is due to any experimental issues . Our study sheds new insight into REST regulation of many genes that have critical roles in neuronal development and function ( Fig . 6 ) , including miR-9 and miR-124 ( Fig . S7 ) . These miRNAs contribute to neural differentiation , neural fate determination and cell cycle exit through the repression of a number of neural transcription factors including TLX , FOXG1 , HES-1 ( all repressed by miR-9 ) , REST , and some of its cofactors ( Fig . 6 ) . TLX is critical for maintaining neural progenitor cells ( NPCs ) in their undifferentiated state [87] . HES-1 is required for NSC homeostasis/maintenance [88] , as its repression accelerates , while its overexpression inhibits , neurogenesis [89] . FoxG1 maintains NPC self-renewal [90] and suppresses the formation of early-born neurons [91] . Our analysis showed that REST bound to HES1 and TLX in nearly all cell types but neurons , while REST occupied FOXG1 only in neurons . There is more than one locus encoding miR-9 and miR-124 in the human genome . The neuronal expression of miR-9 and miR-124 was likely promoted by the differential promoter REST binding in neurons , as well as certain loci not bound by REST altogether ( Fig . S7 ) . Together , these suggest that REST is involved in miR-9 and miR-124 transcription and consequently that their repression of neuronal factors that are critical for maintaining the fate of neural stem cells . Simultaneously , REST also directly represses those factors in non-neuronal cells . For two transcription factors that promote neuronal differentiation , NEUROD1 and BDNF , expression is also controlled by REST; both exhibit similar regulation by REST in all of our analyzed cell types except neurons ( NEUROD1 lacks upstream REST binding in both neurons and T cells ) . Their expression in neurons , however , is also modulated by miR-124 , likely to strike a balance between factors that activate and inhibit neurogenesis . Several other key neuronal factors are also regulated by the REST/miRNA regulatory circuitry ( see Table S2 ) , including BAF53a/b and EFNB1 , as well as two ( MYT1L and POU3F2 ) of the three genes sufficient to convert fibroblasts to neurons [92] and a transcription factor ( NeuroD2 ) that speeds their miRNA-mediated conversion [93] . These results , along with previous findings in the literature , indicate that REST plays an important role in neurogenesis by both directly targeting key neuronal transcription factors and regulating the transcription of neuronal miRNAs . Together , the mini REST/miRNA regulatory network controls neurogenesis synergistically by fine-tuning the expression of individual components to maintain a balance , which is necessary for the proper development of multiple neuronal lineages and for maintaining some level of developmental plasticity . We were intrigued to discover that REST bound to its own promoter in all 15 cell types but not in neurons . It may be that in neurons , the dynamics of REST production and degradation are different , for example , perhaps more post-translational control is taking place . It is known that REST is a protein with high turnover [10] and that negative auto-regulation speeds response times of transcriptional networks [94] . It could be that less steady-state transcription of REST leads to phases of lower levels of available REST despite the fact that REST expression levels are similar ( ranging from 2 . 6–10 . 2 FPKM ) in neurons and other cell types ( Fig . S4 ) . REST , itself , along with other members of the REST complex , such as SCP1 , EZH2 , CoREST ( RCOR1 ) and MECP2 , are all regulated by REST-bound and brain-specific miRNAs , including miR-9 and miR-124 . Many of the REST complex components , miRNAs and REST targets are also regulated by CREB [95] , a potential positive regulator of all the REST-regulated miRNAs . Therefore , all of these cofactors are controlled at a number of different levels through REST and its downstream miRNAs . It will be important to study how these interconnected regulatory factors are involved in the formation and function of the REST complexes , especially during neurogenesis , neuronal differentiation and maturation .
The protocols for ChIP-seq and ChIP-qPCR have been described [96] . In brief , human primary CD4+ T cells were purified from blood as previously described [97] . This cell type was chosen because it is an easily accessible primary cell type and known to express REST [3] . Twenty million cells were cross-linked by formaldehyde . Following sonication , chromatin fragments were immunoprecipitated with an anti-REST antibody [Millipore 07-579] and then prepared for sequencing . Sequencing was performed on an Illumina HiSeq 2000 machine . ChIP-seq reads were processed by the Illumina Analyzer Pipeline and aligned to the human genome ( GRCh37/hg19 ) using Bowtie [98] . Unique reads mapped to a single genomic location ( allowing up to three mismatches ) were kept for peak identification . The primers used for qChIP analysis were listed in Table S9 . The ENCODE ChIP-seq data ( Table S1 ) , including the one for H1-derived neurons , were collected with a REST antibody provided by Dr . David Anderson at Caltech . Sample information for the H1-derived neurons ( including its purity ) can be found at ( http://genome . cse . ucsc . edu/ENCODE/protocols/cell/human/H1_Neurons_Round1 . pdf ) . Peaks were called using the SPP pipeline [35] , following the guidelines of controlling Irreproducible Discover Rate ( IDR ) [99] by the ENCODE project [34] , with a relatively strict IDR threshold ( 0 . 001 ) . Where multiple ChIP-seq replicates were available , reads from all replicates were combined for peak calling . For T cell data , which did not have a replicate , we divided the ChIP-seq data into two halves to generate pseudoreplicates , and the IDR threshold was calibrated from analysis of the pooled-pseudoreplicated data of the other 15 cell types . We also filtered out REST peaks that mapped to genomic regions red-flagged by the ENCODE ( both Duke mappability regions and ENCODE Dac mappability consensus regions ) [34] , as they were likely a result of experimental artifacts . We used MEME 4 . 6 . 1 [36] to find enriched motifs within 200 bp sequences centered on the summits of all the REST peaks in the CD4+ T cell data . The resultant RE1 motifs were used to identify peaks with RE1 motifs by the program MAST in MEME suite , in all cases by scanning 200 bp sequences around peak summits and using default parameters . The genes containing peaks were identified using in-house scripts . We used RefSeq [37] and miRBase [38] annotations from the UCSC Table Browser ( http://genome . ucsc . edu ) , which provides the precursor forms of miRNAs; in cases of overlapping transcripts from the same gene we picked the longest one for both ChIP-seq and RNA-seq analysis , resulting in 23 , 010 genes and 928 miRNAs . The script assigned peaks to genes in the following step-wise manner: to promoter regions ( −5 kb to +1 kb of TSSs ( transcriptional start sites ) ) , to exons , to introns , to distal regulatory regions ( −50 kb of transcription starts to +50 kb of transcription ends ) . When mapping peaks to either promoter regions or distal regions , only the gene with the closest TSS was selected . A single base overlap was used as a rule for these assignments . A peak can be mapped to multiple genes , but only it is equidistant from the TSS of multiple genes , or if it is located to exons or introns shared by multiple genes . Notably , the definition of exonic and intronic REST binding was not applicable to miRNAs and their precursors , as they are short . The merged and non-redundant REST peaks were compared to the peaks originally called for each cell type by the SPP program , and those overlapping with peaks from all cells were defined as “common peaks . ” For the rest of the merged peaks , we computed a sequencing-depth-normalized maximal read coverage at the 300-bp surrounding the peak summit ( averaged from all cells ) of each peak in each cell type , and then inferred cell-specific REST binding based on significantly differential read coverage . Each ChIP-seq read was extended to 200 bp for this analysis . For each peak at each cell , we obtained a REST ChIP-enrichment score ( Ej , j = 1 , 2 , … , 16 ) that was determined from difference in maximal read coverage between REST-ChIP and input samples and normalized by a scaling factor that quantified the genome-wide background noise . The enrichment scores for all peaks were subject to quantile normalization across cell types before used for Z-score statistics . In the end , we defined cell-specific peaks as those having a high Z-score ( >3 ) in one cell type but low Z-scores ( <1 ) in the rest . Further information on this procedure and more details of the methodology development can be found in the Supplemental Methods ( Text S1 ) , where we also presented our exploration of several computational schemes to account for different chromatin structure , different immunoprecipitation efficiency/enrichment and other factors in ChIP-seq experiments of different cell types . All RNA-seq data have been published ( Table S5 ) except the one from neurons , which was collected from 27-day differentiating human neurons derived from induced pluripotent stem cells ( iPSCs ) , which were made from a healthy male . The derivation of the iPSC line ( iPSC-2 ) , neuronal differentiation , and RNA-seq sample preparation have been described in previous publications [100] , [101] . RNA-seq reads ( from polyA+ RNAs ) from replicates ( when available ) were merged and were subsequently aligned to the human genome ( version hg19 ) by TopHat ( v2 ) [60] . Transcripts from Refseq [37] ( with micro- and sno-RNAs removed ) were used to determine gene expression by the Cuffdiff tool in Cufflinks package ( v2 ) [102] , using options for correcting sequence bias and multiple hits , as well as the default geometric mean normalization . All expression comparisons were carried out by the Wilcoxon test and fold changes between groups were based on the medians of FPKMs , unless stated otherwise . Small RNA-seq data were downloaded from GSE24565 [72] and reads were aligned to the human genome by Bowtie ( v2 ) [103] , using the options: –local –sensitive-local –score-min G , 0 , 2 . We then counted the number of reads at individual REST peaks or miRNAs [38] using HTseq [73] , which only uses uniquely mapped reads , with normalization by peak sizes and the number of total aligned reads to yield a RPKM ( Reads Per Kb per Million mapped reads ) value . The ChIP-seq read density was calculated using the program seqMiner [45] , which yielded an array that consisted of the maximal number of overlapping ChIP-seq reads ( extended to 200-bp ) in 300-bp bins from −3 . 15 kb to +3 . 15 kb of the REST peak summits . The enrichment of the sequencing-depth-normalized reads over those of input experiments for each cell type was calculated and the enrichment values were subject to both background normalization [104] as described above . This matrix of enrichment values were finally used to generate heatmaps in Figure 2 with the gplots package [105] in R . All data are publicly available and they can be accessed in the Gene Expression Omnibus ( see Table S1 and S5 ) . | The RE-1 silencing transcription factor ( REST ) binds to DNA and has been shown to repress neuronal genes in non-neuronal systems , but more recent studies have expanded its functions much beyond this . At the molecular level , REST acts cooperatively with other proteins to execute its transcriptional regulatory roles . The dynamics of REST binding and cofactor recruitment and its association with the underlying DNA sequence remain unclear . Here , we have applied chromatin immunoprecipitation and deep sequencing to identify REST binding across 16 different cell types , including neurons . Our results demonstrate that REST binding events are dynamic and quite distinct among cells and that REST binding is generally associated with low gene expression . Closer examination finds that the context of the DNA sequence at REST bound sites is associated with the lower expression of REST-associated targets and that different contexts correlate with different cofactor recruitment . These in turn have an effect on the expression of REST targets . REST targets in human neurons , however , are drastically different from those in other cell types . These findings provide insights into the effect of genomic and cellular contexts on REST's diverse functions and point to distinct and novel roles for REST in neurons . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"systems",
"biology",
"genome",
"expression",
"analysis",
"genomics",
"functional",
"genomics",
"genome",
"analysis",
"transcriptome",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology"
] | 2014 | Comparison of REST Cistromes across Human Cell Types Reveals Common and Context-Specific Functions |
In many organisms , transcription of the zygotic genome begins during the maternal-to-zygotic transition ( MZT ) , which is characterized by a dramatic increase in global transcriptional activities and coincides with embryonic stem cell differentiation . In Drosophila , it has been shown that maternal morphogen gradients and ubiquitously distributed general transcription factors may cooperate to upregulate zygotic genes that are essential for pattern formation in the early embryo . Here , we show that Drosophila STAT ( STAT92E ) functions as a general transcription factor that , together with the transcription factor Zelda , induces transcription of a large number of early-transcribed zygotic genes during the MZT . STAT92E is present in the early embryo as a maternal product and is active around the MZT . DNA–binding motifs for STAT and Zelda are highly enriched in promoters of early zygotic genes but not in housekeeping genes . Loss of Stat92E in the early embryo , similarly to loss of zelda , preferentially down-regulates early zygotic genes important for pattern formation . We further show that STAT92E and Zelda synergistically regulate transcription . We conclude that STAT92E , in conjunction with Zelda , plays an important role in transcription of the zygotic genome at the onset of embryonic development .
Embryonic pattern formation is a complex and progressive process . In many multicellular organisms , the initial period of embryogenesis relies on gene products inherited from the mother . In Drosophila , maternally derived morphogen proteins form broad gradients along the major body axes to define body polarities [1]–[3] . Zygotic transcription begins during the maternal-to-zygotic transition ( MZT ) , which is characterized by a decline in maternal mRNA levels and a dramatic increase in a large number of zygotic transcripts [4] , [5] . Many of the zygotic genes transcribed the earliest , exhibit region-specific patterns . For instance , the “gap genes” , such as zygotic hunchback ( hb ) , Krüppel ( Kr ) , knirps ( kni ) , and tailless ( tll ) are transcribed zygotically in broad and mostly non-overlapping domains along the anteroposterior ( A/P ) body axis . The boundaries of these zygotic genes are determined by morphogen gradients that are set up by maternal gene products , such as Bicoid ( Bcd ) and maternal Hb [2] , [3] . Additional zygotic genes , mostly transcription factors , are induced in more refined embryonic regions as a result of cooperation between the maternal morphogens and gap gene products . The combinatorial input of different transcription factors at different positional coordinates results in expression of thousands of zygotic genes in an increasingly refined pattern , leading to cell fate determination and differentiation [1]–[3] , [6] . To date , only a few transcription factors have been implicated in transcription of the zygotic genome during the MZT . For example , the maternal morphogens Bcd and Dorsal activate target genes along the anteroposterior ( A/P ) and dorsoventral ( D/V ) axis , respectively [7] , [8] . The dramatic increase in gene expression that occurs during the MZT raises the possibility that additional unidentified transcription factors are involved in the rapid initiation and maintenance of the heightened levels of zygotic gene transcription that characterize the MZT . It has been proposed that the few known regionally localized transcription factors , such as Bcd and Dorsal , act in conjunction with ubiquitously present factors to induce and maintain expression of a large number of zygotic genes in cell type-specific patterns . This idea is supported by the identification of a ubiquitous factor encoded by zelda ( zld; a . k . a . vielfaltig or vlf ) [9] , and further by the demonstration that combining Dorsal with Zelda- or STAT-binding sites supports transcription in a broad domain in the embryo [10] . To identify additional ubiquitous transcription factors that are important for transcription of the zygotic genome during the MZT , we first conducted in silico analyses , taking advantage of the large amount of information available in public databases on transcriptional regulation of zygotic genes expressed during early embryogenesis in Drosophila . This approach led to the identification of STAT92E , in addition to Zelda , as a plausible transcription factor important for the upregulation of multiple genes during the MZT . Global expression profiling studies indicate that loss of STAT92E , similarly to loss of Zelda , preferentially causes down-regulation of zygotic genes essential for early embryogenesis . We further demonstrate that STAT92E is indeed involved in transcription of the developmentally important genes dpp , tailless ( tll ) , and Kr during early embryogenesis . Our results suggest that STAT92E is essential for upregulation of a multitude of zygotically transcribed genes during the MZT , and thus is important for transition of the early embryo from a totipotent embryonic stem cell state to a state of cellular differentiation .
To identify general transcription factors that are required for transcription of a large number of zygotic genes at early embryonic stages , or during the MZT , we performed a meta-analysis to search for candidate transcription factors required for activation of multiple zygotic genes . To this end , we first selected a list of developmentally important zygotic genes transcribed during the MZT ( referred to as “zygotic genes” ) , whose expression patterns altogether cover the entire embryo , and whose transcriptional activation has previously been studied . We analyzed a total of 21 early zygotic genes , including the gap genes: hunchback ( hb ) , huckebein ( hkb ) , Giant ( Gt ) , Krüppel ( Kr ) , knirps ( kni ) , and tailless ( tll ) ; the pair-rule genes: even skipped ( eve ) , fushi tarazu ( ftz ) , hairy ( h ) , odd paired ( opa ) , paired ( prd ) , sloppy paired 1 ( slp1 ) , and runt ( run ) ; the segmental polarity and other genes: engrailed ( en ) and Sex lethal ( Sxl ) , as well as genes expressed along the D/V axis: decapentaplegic ( dpp ) , zerknüllt ( zen ) , rhomboid ( rho ) , short gastrulation ( sog ) , snail ( sna ) , and twist ( twi ) . As a second step , for each of these genes , we searched Flybase ( http://flybase . org ) and PubMed ( http://www . ncbi . nlm . nih . gov ) , and compiled a list of all currently known or potential transcriptional activators or signaling pathways involved in their transcriptional induction ( Table S1 ) . We used the RedFly database ( http://redfly . ccr . buffalo . edu ) [11] to obtain a list of experimentally verified transcription factor binding sites for each target gene , and the FlyEnhancer program ( http://genomeenhancer . org/fly ) [12] to search for the presence of particular transcription factor binding sites in the promoter region ( defined as 4 kb upstream of the transcriptional start site ) of all the target genes . Based on these search results , we assigned activation scores to the putative or known transcriptional activators to reflect their importance in the expression of a particular zygotic gene ( Table S1 ) . These scores were added to obtain a cumulative score for each activator ( Figure 1A; Table S2 ) . The connections between activators and their target genes are represented in an activation map ( Figure 1B ) . The top seven activators identified , in descending order of cumulative interaction score , were Zelda ( Zld ) , Bicoid ( Bcd ) , STAT92E , Torso , Caudal ( Cad ) , Dorsal , and Twist ( Twi ) ( Figure 1A; Table S2 ) . Zelda has previously been shown to be a key transcription activator of the early zygotic genome [9] , validating our bioinformatic approach . Both Bcd and Cad are maternal-effect gene products that form gradients along the A/P axis in the early embryo [7] , [13] , [14]; Torso signaling is activated only at the anterior and posterior poles , and the specific transcriptional activators that it regulates remain unidentified [15]–[17]; Dorsal and Twi are active only in the ventral region of the embryo [18] . On the other hand , STAT92E is ubiquitously distributed in the early embryo as a maternal product [19] and is activated early [20] , and thus has the potential to act more universally . STAT92E is the transcriptional activator mediating the JAK/STAT ( Hop/STAT92E ) pathway [19] , [21] , [22] , and also participates in Torso signaling [23]–[25] . Thus , we decided to investigate whether STAT92E acts as a general transcriptional regulator during early embryogenesis , similar to Zelda . To test whether STAT92E is important for transcription of early “zygotic genes” , we first assessed the occurrence of consensus STAT92E binding sites ( TTCnnnGAA ) in the promoter region , defined as 4 kb genomic sequence upstream of the transcription start site , of the 21 zygotic genes in this study . The Drosophila genome is slightly AT-rich , with 57 . 4% AT and 42 . 6% GC base pairs [12] . Thus the probability for A or T to occur at any position is 0 . 287 , and for G or C is 0 . 213 , and the probability ( p ) for random occurrence of one STAT binding site ( with 6 fixed nucleotides ) at any position is 3 . 08x10−4 ( 0 . 2874x0 . 2132 ) , and its frequency of occurrence within the 4 kb upstream regulatory regions of 21 genes ( n = 84 , 000 bp ) at random is 25 . 9 ( np; expected value ) . However , when we searched for STAT binding sites within the 4 kb upstream region of the 21 zygotic genes , we found 43 in total ( observed value ) ( Figure 1C ) . Assuming the actual occurrence of STAT-binding sites exhibits Binomial distribution with a probability of 3 . 08x10−4 , the standard deviation ( σ ) should be 5 . 1 . The difference between the observed ( 43 ) and expected ( 25 . 9 ) values is 17 . 1 , which is beyond three standard deviations ( Z = 3 . 29; p = 0 . 001 ) . In contrast , when we searched for STAT-binding sites within a 4 kb window upstream of the transcription start site of 21 housekeeping genes ( defined as ubiquitously expressed , both maternally and zygotically , with generally cellular metabolic or structural functions ) , including rp49 , GAPDH , Actin5C , and those encoding ribosomal proteins and RNA polymerases , we found a total of 13 STAT-binding sites ( Figure 1D ) , which is significantly lower than the expected 25 . 9 sites ( Z = 2 . 48; p = 0 . 013 ) . ( A total of 78 housekeeping genes and the numbers of STAT-binding sites in their upstream regions are listed in Table S3 . ) Moreover , many of the STAT-binding sites in the upstream regions of the 21 zygotic genes are clustered ( defined by two sites occurring within 500 bp ) , which is characteristic of functional transcription factor binding sequences [12] , [19] , [25] , [26] ( Figure 1C ) , whereas in the promoter regions of the 21 housekeeping genes , the STAT-binding sites occur as single sites ( Figure 1D; Table S3 ) . It has been shown that Zelda-binding sites ( the TAGteam motif ) are enriched in the promoter regions of “zygotic genes” [9] , [27] . We examined the distribution of Zelda-binding sites in the promoter regions of the 21 zygotic and housekeeping genes , respectively . Consistent with the previous report [9] , [27] and similar to STAT-binding sites , we found that Zelda-binding sites are similarly enriched in the promoters of the zygotic and very infrequently in the housekeeping genes ( Figure 1C , 1D ) . Since the enhancers for many of the early zygotic genes are not localized in the upstream promoter regions , we also searched for STAT and Zelda-binding sites in the promoter-distal enhancers for these 21 zygotic genes , and found that promoter-distal enhancers are not enriched for STAT-binding sites ( Z = 0 . 63; p = 0 . 736 ) , but are significantly enriched for Zelda-binding sites ( Z = 3 . 13; p = 0 . 0017 ) ( Figure S1 ) . Such a result suggests that STAT92E might differ from Zelda and might not be important for regulating promoter-distal enhancers , which usually control spatial expression patterns . Nonetheless , our studies indicate that DNA-binding sites for both STAT and Zelda are enriched in the upstream promoter regions of the 21 zygotic genes that are highly transcribed during the MZT , but are underrepresented in the housekeeping genes that are ubiquitously transcribed . This observation is consistent with the finding that Zelda is required specifically for expression of “zygotic genes” at the MZT [9] , raising the possibility that STAT may play a similar role . To determine whether STAT92E functions as a general transcriptional activator of the zygotically expressed genes in the early embryo , we determined the expression profiles of early stage embryos ( corresponding to nuclear division cycle 8–14 , a time window for the MZT ) of wild-type control and of those lacking the maternal Stat92E gene products ( referred to as Stat92Emat–; see Methods ) at the same stage . We found that in Stat92Emat– embryos , 657 genes were down regulated and 558 genes up-regulated by at least 1 . 5 fold , compared with wild-type control ( Figure 2A ) . In Stat92Emat– embryos , genes exhibiting >1 . 5 fold change in expression constituted 8 . 9% of all genes ( n = 13 , 615 ) on the Gene Chip , while the majority ( 91 . 1% ) of the genes exhibited no significant changes ( Figure S2 ) . Consistent with the idea that STAT92E is preferentially required for expression of “zygotic genes” , the vast majority ( 78 . 2% ) of the down-regulated genes in Stat92Emat– embryos were “zygotic genes” ( Figure 2B , left; Table S4 ) . In contrast , the up-regulated genes contained more maternally expressed than zygotically expressed genes ( Figure 2B , right; Table S5 ) . This observation is reminiscent of gene expression profiles of zld mutant embryos at the same stage , in which more “zygotic genes” than maternal genes are down-regulated [9] . By comparing the two sets of genes , we found that >50% of the “zygotic genes” that were down-regulated in zldmat– embryos ( 67/120 ) were also down-regulated in Stat92Emat–embryos , suggesting that these genes might be co-regulated by STAT and Zelda ( Table S4 ) . Consistent with the observed difference in the abundance of STAT-binding sites present in their promoter regions , the 21 zygotic genes ( except for hb ) were all significantly down-regulated , with a 4 . 3 fold down-regulation on average , whereas the 21 housekeeping genes showed no significant changes in expression , with the exception of DNase II ( Figure 2C ) , in Stat92Emat– embryos . Similar to Stat92Emat– embryos , in zldmat– embryos , many of these 21 zygotic genes were also significantly down-regulated , whereas the housekeeping genes were not significantly changed [9] , suggesting that STAT92E and Zelda may both be important for transcription of early zygotic genes . Expression profiling experiments indicate that STAT92E and Zelda do not transcriptionally regulate each other ( Liang et al . , 2008; this study ) . We further performed qRT-PCR experiments and found that Zelda mRNA levels were indeed not significantly changed in Stat92E loss-of-function or hop gain-of-function mutants ( Figure S3 ) , suggesting that STAT92E does not indirectly control zygotic gene activation by affecting Zelda levels . Finally , we tested expanded sets of zygotic and housekeeping genes to include >40 genes in each set ( Table S6 ) using the Gene Set Enrichment Analysis ( GSEA ) software ( http://www . broadinstitute . org/gsea/index . jsp ) , which is a computational method that determines whether an a priori defined set of genes shows statistically significant , concordant differences between two biological states ( e . g . , mutant versus wild-type ) [28] . Indeed , by subjecting our microarray data to GSEA analysis , we found that the “zygotic genes” were highly significantly down regulated ( p = 0 . 00 ) , whereas the housekeeping genes were insignificantly changed ( p = 0 . 44 ) , in Stat92Emat– embryos when compared with wild-type control ( Figure 2D ) . Thus , similar to Zelda , STAT92E is preferentially required for transcription of “zygotic genes” . To validate our gene profiling results from the microarray studies , we investigated the effects of over-activation and loss of STAT92E on transcript levels of a number of early “zygotic genes” . We chose to examine expression levels of dpp , Kr , tll , and eve , four early zygotic genes whose promoter regions contain STAT-binding sites and whose expression domains span broad and distinct regions of the early embryo ( see below ) . We first examined mRNA levels of dpp , Kr , tll , and eve in the early embryo ( 1–2 h after egg laying ) using semi-quantitative reverse-transcription polymerase chain reaction ( RT-PCR ) in Stat92E gain- or loss-of-function genetic backgrounds . We found that in hopGOF embryos , in which STAT92E is overactivated [29]–[31] , mRNA of these four genes were all expressed at significantly higher levels relative to wild-type; whereas in Stat92Emat– embryos , these four genes were expressed at approximately 50% of the wild-type levels ( Figure 3A , 3B ) . Moreover , reducing the dosage of zelda by half in Stat92Emat– embryos caused further reductions in the transcript levels of dpp , Kr , tll , and eve ( zelda+/–; Stat92Emat– in Figure 3A , 3B ) . We examined zelda+/–; Stat92Emat– embryos only , because it was technically not possible to examine embryos lacking both Zelda and Stat92E . We further confirmed the expression results by quantitative real-time PCR ( Figure 3C ) . These results were consistent with the microarray data , which suggested that Stat92E and Zelda may co-regulate transcription of many “zygotic genes” . We next investigated whether STAT92E binds to the putative STAT-binding sites in the respective promoter regions of dpp , Kr , and tll using chromatin immunoprecipitation ( ChIP ) experiments with early embryo extracts using anti-STAT92E antisera . Binding of STAT92E to the eve enhancer and of Zelda to the TAGteam sequences enriched in “zygotic genes” have been previously shown [9] , [19] , [21] . Using primers flanking the putative STAT-binding sites in these promoter regions , we detected STAT92E binding to the promoter regions dpp , Kr , and tll ( Figure 3D ) . The results from RT-PCR and ChIP studies were consistent with the bioinformatic and gene profiling studies shown above , suggesting that STAT92E , likely together with Zelda , regulates the transcription of early “zygotic genes” in vivo . Having shown that STAT92E regulates expression levels of early “zygotic genes” , and that STAT92E binds to the consensus STAT-binding sites present in the promoter regions of dpp , Kr , and tll , we next investigated whether these consensus STAT-binding sites are indeed essential for mediating STAT92E transcriptional activation , and whether STAT92E and Zelda cooperate to regulate “zygotic genes” , as it has previously been shown that Zelda is essential for expression of dpp , Kr , tll , and eve , among others , in the early embryo [9] . We carried out reporter gene assays in Drosophila S2 cells ( Figure 4A ) . We first tested whether activated STAT92E binds to the promoter regions of dpp , Kr , tll , and eve in S2 cells as it does in early embryos ( see Figure 3C ) . We transfected a V5-tagged STAT92E into S2 cells and performed ChIP assays . STAT92E activation in S2 cells was achieved by co-expressing Hop , which phosphorylates and activates STAT92E when over-expressed ( Figure 4B ) . By immunoprecipitation with anti-V5 antibody , we found that co-transfection with Hop leads to an enrichment of STAT92E binding to the endogenous dpp promoter ( Figure 4C , lane 3 ) . Activation of JAK/STAT signaling thus induces a stronger association of STAT92E with the dpp promoter , consistent with the idea that STAT92E directly regulates dpp expression . However , the same ChIP experiments failed to detect association of STAT92E with the Kr , tll , or eve promoter in S2 cells , in contrast to the ChIP results in early embryos ( see Figure 3C ) , suggesting that the epigenetic states of these promoter sequences may be different in S2 cells than in early embryos . We thus focused on the dpp promoter for reporter gene analysis . To this end , we isolated a 1 . 3 Kb dpp promoter fragment ( Figure 4A; Figure S4 ) , which contains the two clustered STAT92E binding sites we had tested in ChIP experiments ( see Figure 3C , Figure 4C ) . To test whether the STAT-binding sites in the dpp promoter are important for JAK/STAT-induced dpp expression , we made reporter genes by fusing a wild-type dpp promoter fragment ( WT ) , or a mutant version with both STAT-binding sites mutated ( DM ) , with an enhanced yellow fluorescent protein ( EYFP ) , and transfected S2 cells ( Figure 4A ) . In order to activate reporter gene expression , we first treated the cells with H2O2/vanadate ( pervanadate ) , which causes rapid and efficient STAT92E phosphorylation [32] , [33] ( Figure S5A ) and is more efficient than transient transfection of hop in activating STAT . We found that , indeed , EYFP was expressed 1 . 5 hours after pervanadate treatment in S2 cells transfected with the wild-type ( WT ) , but not the double mutant ( DM ) construct ( Figure S5B ) , indicating that these STAT92E-binding sites are important for phosphorylated STAT92E-induced reporter gene expression . To more accurately quantify transcription from the dpp promoter with or without the two STAT-binding sites , we replaced EYFP with luciferase in the reporter constructs to obtain dppWT-luc and dppDM-luc , respectively . In addition , we used Hop and STAT92E co-transfection , instead of pervanadate , to ensure specific activation of STAT92E . In the presence of co-transfected Hop and STAT92E , we detected an increase in luciferase activity in S2 cells tranfected with dppWT-luc to more than 20 fold when measured 72 hours after transgene expression , and this increase was abolished when dppDM-luc was used in the assay , which showed much less pronounced increase ( Figure 4D ) . These results further substantiate our finding that STAT92E-mediated activation of dpp requires the two STAT92E binding sites . It has previously been shown that transcription of dpp is significantly down-regulated in the absence of Zelda [9] , and that Zelda-binding sites are present in the dpp promoter region ( Figure 1C; Figure 4A; also see [9] ) . To test whether Zelda binds to the putative site in the dpp reporter gene , we carried out ChIP assays in S2 cells after transfecting a Zelda-Flag plasmid . Indeed , we detected Zelda binding to the dpp promoter region using an anti-Flag antibody and ChIP assay ( Figure 4E ) . We next investigated the role of Zelda in dpp transcription using dppWT-luc and a mutant promoter fragment with the Zelda-binding site and the two STAT-binding sites mutated ( designated as dpp™-luc as it bears triple mutations; Figure 4A ) . To evaluate whether Zelda and STAT cooperate in regulating dpp transcription , we co-transfected S2 cells with STAT92E ( together with Hop to achieve STAT activation ) or Zelda , or both STAT92E ( with Hop ) and Zelda , in the presence of dppWT-luc or dpp™-luc , and carried out luciferase assays . When assayed at 72 h after induction of transgene expression , we found that STAT activation alone induced dppWT-luc transcription by 22 fold , and Zelda alone caused upregulation of dppWT-luc by 48 fold , whereas in the presence of both Zelda and activated STAT , dppWT-luc was up-regulated by 230 fold ( Figure 4F ) . Mutating STAT and Zelda binding sites prevented the dramatic increase in transcription as measured by luciferase activity ( Figure 4F ) . These results suggest that Zelda and STAT have synergistic effects on dppWT-luc transcription . Interestingly , an increase in luciferase activity was observed even when binding sites for STAT or Zelda , or both , were mutated , albeit to a much less pronounced level than with the wild-type promoter ( Figure 4D , 4F ) , suggesting that there might be other cryptic binding sites present in the promoter , or that other molecules were activated by over-expressed JAK or Zelda . The apparent synergy between STAT92E and Zelda could explain the results from the gene profiling experiments . Microarray results show that embryos without STAT92E ( in which Zelda presumably remains active ) exhibit a 3 . 1 fold decrease in dpp expression ( Figure 2B ) , and that Zld mutant embryos ( in which presumably STAT92E is still active ) have reduced dpp expression by 5 . 7 fold [9] . These data suggest that in the early embryo either Zelda or STAT activation could induce dpp transcription to a limited extent , whereas the presence of both Zelda and STAT activation synergistically promote dpp transcription . Having shown that STAT92E , possibly acting synergistically with Zelda , is important for expression levels of many early “zygotic genes” , we next investigated whether loss of STAT92E also affects the spatial expression patterns of the early “zygotic genes” . We examined the expression of dpp , Kr , and tll in the early embryo , by in situ hybridization , while the effects of Stat92E mutation on eve expression have previously been documented [19] , [21] . These genes are expressed in distinct spatial domains that altogether cover nearly the entire early embryo ( see below ) . The dpp expression domain spans nearly the entire A/P axis in the dorsal regions of the early embryo [34]-[37] ( Figure 5A ) . It has been shown that dpp transcription in the ventral region is repressed by Dorsal , a Rel family transcription factor [38] , and that general transcription factors , such as Zelda and STAT , are responsible for dpp expression in the dorsal region ( [9]; this study ) . By employing in situ hybridization , we found that compared to wild type , the overall level of dpp mRNA is much reduced in Stat92Emat– embryos , especially in the posterior pole region ( Figure 5B ) . Moreover , we found that JAK/STAT signaling also regulates dpp expression during late embryogenesis ( Figure S6 ) . These results are consistent with previous findings in other developmental contexts [39] , [40] as well as with the above microarray results and mRNA measurements ( Figure 2 , Figure 3A–3C ) . Kr is expressed in the central region of the early embryo [41] ( Figure 5C ) . Other than the maternal morphogens Bcd and Hb , it is not known whether additional factors contribute to Kr transcriptional activation . We found that in Stat92Emat– embryos , although the overall expression pattern of Kr mRNA was little changed , its levels were reduced ( Figure 5D ) , consistent with the microarray and qPCR results . tll is expressed in two domains along the A/P axis-the anterior and posterior pole regions [42] ( Figure 5E ) . The Torso pathway controls tll expression by antagonizing its repressors [17] , [43]; the identity of transcriptional activators of tll remains obscure , although STAT92E has been speculated to contribute to tll expression [25] . We have previously shown that STAT92E is essential for the expansion of tll expression domains caused by Torso , over-activation , but not for the extent of tll spatial expression domains under normal conditions [25] . In addition , we have previously shown that there are two consensus STAT binding sites in the tll promoter region that are particularly important for Torso overactionvation-induced ectopic tll expression [25] . In light of our finding that STAT92E is important for the expression levels of dpp , Kr , and tll , we reexamined the role of STAT92E in endogenous tll expression in Stat92Emat– and wild-type control embryos by in situ hybridization done under identical conditions . We found that , similar to dpp and Kr mRNA , while the spatial patterns of tll expression were not dramatically changed as previously shown [25] , the overall levels of tll mRNA were significantly reduced in Stat92Emat– embryos ( Figure 5F ) . Taken together , the above results indicate that loss of STAT92E led to much reduced expression levels of dpp , Kr , and tll , without affecting their spatial expression domains . Similarly , it has been shown that loss of STAT92E results in reductions , but not complete loss of , eve stripe 3 and 5 , without affecting the overall spatial expression pattern of eve [19] , [21] . Thus , STAT92E is likely required for regulating the expression levels of early “zygotic genes” , but not for controlling their spatial patterns . Finally , we investigated the biological consequences of reducing expression levels , without altering spatial domains , of multiple zygotically expressed early genes , as with loss of STAT92E . The correct expression of the early zygotic genes during the MZT is essential for formation of different tissues and body parts at the correct positions , i . e . , pattern formation [1]–[3] . Pattern formation in Drosophila can be conveniently visualized by examining the exoskeleton ( cuticle ) morphology of the larva or late embryo [1]–[3] . In the wild-type cuticle ( Figure 6A ) , anteroposterior ( A/P ) polarity is defined by the head skeleton and three thoracic segments in the anterior , followed by the abdominal segments , and the posterior and terminal structure , consisting of the 8th abdominal segment and the Filzkörper ( Figure 6A; Arrow ) . Dorsoventral ( D/V ) polarity can easily be seen by the positions of the eight abdominal denticle belts , which form in the ventral region , while bare cuticle marks the dorsal region ( Figure 6A ) . Removal of STAT92E from the early embryo resulted in heterogeneous defects , mostly notably along the A/P axis as seen in the larval cuticles , which were missing part or all of A3 , A4 , A5 , and A8 to various degrees ( Figure 6B; also see [19] , [25] ) . Thus , loss of STAT92E , which significantly reduces multiple early “zygotic genes” but does not completely eliminate their expression ( see Figure 5 ) , leads to heterogeneous patterning defects , consistent with defects in multiple pathways . To understand the role of STAT92E in individual signaling pathways important for pattern formation , we investigated whether loss of STAT92E could further compromise pattern formation in sensitized genetic backgrounds . To this end , we examined cuticles of Stat92Emat– embryos that were also heterozygous for tll , Kr , or dpp , and indeed found patterning defects ( see below ) . The gap gene tll is essential for the development of terminal structures [17] , [42] , and tll mutant homozygous embryos do not have A8 and the Filzkörper ( Figure 6C ) . tll heterozygous flies , in contrast , are perfectly viable and normal , with cuticles indistinguishable from wild-type controls , according to our own observation . In the absence of STAT92E , however , we found that tll+/– embryos were missing the terminal structures ( A8 and Filzkörper ) ( Figure 6D ) . This suggests that without STAT92E , a half dose of tll+ is no longer sufficient for development , consistent with the idea that STAT92E is partially required for tll transcriptional output . Kr is required for development of the thoracic and anterior segments , and these segments are missing in Kr–/– embryos ( Figure 6E; also see [44] ) . Kr+/– embryos are mostly normal but have subtle anterior defects ( Figure S7; also see [44] ) . In the absence of STAT92E , however , we found that Kr+/– embryos were missing a large area of the thoracic and anterior regions ( Figure 6F ) , suggesting a haploinsufficiency in the absence of STAT92E , similar to what we observed for tll . The dorsally expressed dpp specifies dorsal cell fates and is crucial for the dorsoventral polarity of the embryo , which is reflected in the cuticle by the presence of naked cuticles in the dorsal region and eight abdominal denticle belts in the ventral region ( Figure 6A ) [37] . Notably , although dpp expression was significantly reduced in Stat92Emat– embryos ( Figure 2 , Figure 3A–3C , Figure 5B ) , they did not exhibit gross D/V polarity defects ( Figure 6B ) , suggesting that the residual dpp transcripts present in Stat92Emat– embryos are sufficient for specifying dorsal cell fates , or that the reduction in dpp expression is compensated for by a reduction in a dpp antagonist that is also regulated by STAT92E . Despite the fact that dpp is haploinsufficient for viability , dpp heterozygous embryos exhibit normal D/V polarity , with clearly discernable ventral denticle belts and bare dorsal cuticles ( Figure S7 ) , suggesting that a half dose of dpp+ suffices for D/V patterning ( also see [37] . Embryos homozygous for dpp , nonetheless , are completely “ventralized , ” having denticle belts that extend into the dorsal region to surround the entire D/V axis ( Figure 6G; also see [36] , [37] ) . The combination of Stat92Emat– and dpp heterozygosity caused partial ventralization of the embryo; in 13% of Stat92Emat–; dpp+/– embryos ( n = 11/86 ) , the posterior-most denticle belt extended significantly dorsally to cover approximately 80% of the circumference ( Figure 6H , arrow ) . Similar ventralization defects were never observed in Stat92Emat– and dpp+/– embryos ( n>500 ) . Thus , in the absence of STAT92E , a half dose of dpp is no longer sufficient for dorsoventral patterning , consistent with the notion that STAT92E normally regulates dpp expression levels . In summary , loss of STAT92E caused heterogeneous patterning defects , as revealed by varying cuticle defects , consistent with an insufficiency of multiple pathways . A further reduction in the dosage of genes in different pathways , such as tll , Kr , and dpp , uncovered the role of STAT92E in regulation of specific early zygotic genes important for pattern formation .
We have undertaken a bioinformatics approach to investigating the mechanisms controlling transcription of the zygotic genome that occurs during the MZT , and have identified STAT92E as an important general transcription factor essential for up-regulation of a large number of early “zygotic genes” . We have further investigated the role of STAT92E in controlling transcription of a few representative early zygotic genes , such as dpp , Kr , and tll , that are important for pattern formation and/or cell fate specification in the early embryo . Our studies suggest that STAT92E cooperate with Zelda to control transcription of many “zygotic genes” expressed during the MZT . While STAT mainly regulates transcription levels , but not spatial patterns , of dpp , tll , and Kr , and possibly also other “zygotic genes” , Zelda is essential for both levels and expression patterns of these genes [9] . The transcriptional network that controls the onset of zygotic gene expression during the MZT has remained incompletely understood . It has been proposed that transcription of the zygotic genome depends on the combined input from maternally derived morphogens and general transcription factors . The former are distributed in broad gradients in the early embryo and directly control positional information ( e . g . , Bicoid , Caudal , and Dorsal ) , whereas the latter are presumably uniformly distributed regulators that augment the upregulation of a large number of “zygotic genes” . Other than Zelda , which plays a key role as a general regulator of early zygotic expression [9] , the identities of these general transcriptional activators have remained largely elusive . It has been shown that combining Dorsal with Zelda- or STAT-binding sites supports transcription in a broad domain in the embryo [10] . The demonstration of STAT92E as another general transcription factor sheds light on the components and mechanisms of the controlling network in the early embryo . Moreover , we have found that STAT92E and Zelda may cooperate to synergistically regulate “zygotic genes” . Our results thus validate the bioinformatics approach as useful in identifying ubiquitously expressed transcription factors that may play redundant roles with other factors and thus might otherwise be difficult to identify . Our conclusion that STAT92E is important for the levels but not the spatial domains of target gene expression in the early embryo is consistent with several previous reports . It has been shown that in Stat92E or hop mutant embryos , expression of eve stripes 3 and 5 are significantly reduced but not completely abolished [19] , [21] . In addition , JAK/STAT activation is required for the maintenance of high levels , but not initiation , of Sxl expression during the MZT [45] , [46] . Moreover , it has previously been shown that STAT92E is particularly important for TorsoGOF-induced ectopic tll expression but not essential for the spatial domains of tll expression in wild-type embryos under normal conditions [25] . On the other hand , Zelda may be important for both levels and spatial patterns of gene expression . This idea is consistent with our finding that Zelda-binding sites are enriched in both promoter and promoter-distal enhancers regions , whereas STAT-binding sites are enriched in promoter regions only . It has been reported that pausing of RNA polymerase II is prominently detected at promoters of highly regulated genes , but not in those of housekeeping genes [47] . In light of our results that STAT and Zelda sites are highly enriched in the early zygotic gene promoters , we suggest that these transcription factors might contribute to chromatin remodeling that favors RNA polymerase II pausing at these promoters . Finally , the MZT marks the transition from a totipotent state to that of differentiation of the early embryo . As a general transcription factor at this transition , STAT , together with additional factors ( such as Zelda [9] ) , is important for embryonic stem cell differentiation . Further investigation is required to understand the molecular mechanism by which STAT and Zelda [9] cooperate in controlling zygotic transcription in the early Drosophila embryo . Moreover , it would be interesting to investigate whether STAT plays similar roles in embryonic stem cell differentiation in other animals .
All crosses were carried out at 25°C on standard cornmeal/agar medium unless otherwise specified . Fly stocks of hopTum-l , Stat92E6346 , and dppH46 were from the Bloomington Drosophila Stock Center ( Bloomington , IN ) . To generate Stat92Emat– embryos , hsp70-flp; FRT82B Stat92E6346/TM3 females were crossed to hsp70-Flp; FRT82B [ovoD1 , w+]/TM3 males . Their 3rd instar larval progeny were heat-shocked at 37°C for 2 hrs daily for 3–4 days , and resulting adult females of the genotype hsp70-flp; FRT82B Stat92E6346/FRT82B [ovoD1 , w+] were used to produce embryos that lack maternal Stat92E gene products , as described in the dominant female-sterile “germline clone” technique [48] . The following rules were used for assigning a score to known or putative activators of each of the “zygotic genes” . We placed top importance on genetically demonstrated activation during early embryogenesis , with such an activator receiving an activation score of 10 . For instance , Torso was assigned a score of 10 as an activator of tll transcription based on the reports that tll is not expressed in torso loss-of-function mutants and is overexpressed in torso gain-of-function mutants [17] , [49] . Activators identified by biochemical/promoter studies in early embryos or by genetic studies at other developmental stages were assigned a score of 5 . Lower scores were assigned to other less stringent evidence of interaction , such as unconfirmed genetic screen results ( 5 ) , in vitro biochemical assays ( 2 ) , or bioinformatics studies ( 1 ) ( Table S1 ) . Databases and programs used in this study: Flybase ( http://flybase . org ) ; PubMed ( http://www . ncbi . nlm . nih . gov ) ; RedFly ( http://redfly . ccr . buffalo . edu/ ) ; FlyEnhancer ( http://genomeenhancer . org/fly ) . The dpp promoter used in this study was a 1 . 3 kb genomic DNA fragment including the upstream regulatory sequences and the non-coding exon 1 of the of dpp transcript A ( Figure S2 ) . This genomic region has previously been shown to be the core promoter of dpp [38] . Standard cloning was used to generate transcription fusions between the dpp promoter and cDNAs of reporter genes , such as enhanced yellow fluorescent protein ( EYFP ) and luciferase . Mutagenesis of two STAT92E binding sites within the dpp promoter was done by PCR , and was verified by sequencing . V5-Hop and V5-STAT92E are gifts from S . X . Hou [50] . Cuticle preparations were performed according to a standard protocol with minor modifications . Embryos were dechorionated with 50% Clorox , washed extensively with 0 . 1% Triton , mounted in Hoyer's , and photographed using dark-field optics . In situ hybridization for detecting dpp , Kr , and tll mRNA was performed according to a standard protocol using digoxigenin-incorporated antisense RNA probes made from dpp , Kr , and tll cDNA , respectively , according to the supplier's protocol . A standard protocol was used for antibody staining of embryos , and a biotinylated secondary antibody and the Vectastain ABC kit ( Vector Laboratories , Inc . ) were used according to the manufacturer's instructions . Stained embryos were mounted in DAPI-containing mounting medium for accurate staging , when necessary . Mounted embryos were photographed using Normaski optics on a Zeiss Axioscope and images were analyzed using Photoshop or ImageJ software . Total RNA was isolated from embryos ( from flies raised at 25°C ) collected at 1–2 h after egg laying ( corresponding to nuclear division cycles 8–14 ) using trizol ( Invitrogen ) or the RNeasy Kit ( QIAGEN ) according to the manufacturer's instructions . RNA quality was assessed using the Agilent 2100 Bioanalyzer and the RNA 6000 Nano kit ( Agilent Technologies Inc . , Palo Alto , CA ) . For RT-PCR analysis , first strand complementary DNA ( cDNA ) was generated from 5 µg of purified total RNA using Superscript III reverse transcriptase ( Invitrogen ) and oligo ( dT ) 12–18 in 50 µl total reaction volume . The cDNA ( at 1∶100 dilution ) was used as template for either semi-quantitative PCR reactions or real time PCR analysis using SYBR green based detection on a BioRad iCycler . Reactions were carried out in triplicate , and melting curves were examined to ensure single products . Results were quantified using the “delta-delta Ct” method to normalize to rp49 transcript levels and to control genotypes . Data shown are averages and standard deviations from at least three independent experiments . The following primer pairs were used . rp49: TCCTACCAGCTTCAAGATGAC , CACGTTGTGCACCAGGAACT . dpp: AATCAATCTTCGTGGAGGAGCCGA , TTGGTGTCCAACAGCAGATAGCTC . eve: TGCACGGATACCGAACCTACAACA , GTTCTGGAACCACACCTTGATCGT . Kr: CAAGACGCACAAACGCGAACCTTA , TTGACGGTTTGCAGCCAGAAGTTG . tll: AATACAACAGCGTGCGTCTTTCGC , ACATTGGTTCCTGTGCGTCTTGTC . For microarray analysis , 200 ng of total RNA was used to prepare biotin-labeled RNA using Ambion MessageAmp Premier RNA Amplification Kit ( Applied Biosystems , Foster City , CA ) . Briefly , the first strand of cDNA was synthesized using ArrayScript reverse transcriptase and an oligo ( dT ) primer bearing a T7 promoter . Then DNA polymerase I was used ( in the presence of E . coli RNase H and DNA ligase ) to convert single-stranded cDNA into a double-stranded DNA ( dsDNA ) . The dsDNA was then used as a template for in vitro transcription in a reaction containing biotin-labeled UTP and T7 RNA Polymerase to generate biotin-labeled antisense RNA ( aRNA ) . Twenty µg of labeled aRNA was fragmented and fifteen µg of the fragmented aRNA was hybridized to Affymetrix Drosophila Genome 2 . 0 Array Chips according to the manufacterer's Manual ( Affymetrix , Santa Clara , CA ) . Array Chips were stained with streptavidin-phycoerythrin , followed by an antibody solution ( anti-streptavidin ) and a second streptavidin-phycoerythrin solution , performed by a GeneChip Fluidics Station 450 . The Array Chips were then scanned with the Affymetrix GeneChip Scanner 3000 . The microarray image data were converted to numerical data with Genespring software ( Agilent Technologies Inc . , Palo Alto , CA ) and normalized using the recommended defaults . The signals from 11 perfect matched oligonucleotides for a specific gene and 11 mis-matched oligonucleotides were used to make comparisons of signals . Genes were identified as present when the present ( P ) assignment was significant ( p<0 . 05 ) . The Gene Set Enrichment Analysis ( GSEA ) online software ( http://www . broadinstitute . org/gsea ) was used to determine whether the predetermined gene sets ( e . g . , zygotic versus housekeeping; see Figure S6 ) show statistically significant , concordant differences between wild-type and Stat92Emat– embryos . Primary antibodies used in this study include mouse anti-V5 ( Invitrogen; 1∶500 for Western blots ) , Rabbit anti-V5 ( QED; 1∶200 for immunoprecipitation ) , goat anti-STAT92E ( Santa Cruz; Cat# sc-15708; affinity-purified against the N-terminus of STAT92E; 1∶200 ) , rabbit anti-Kr ( 1∶5000; a kind gift from C . Rushlow ) , and anti-phospho-STAT92E ( Cell Signaling Technology; 1∶1000 ) . Common secondary antibodies were used in whole-mount immunostaining or Western blots . Drosophila Schneider L2 ( S2 ) cells were cultured at 25°C in Drosophila Serum-Free Medium ( SFM; Invitrogen ) supplemented with 10% Fetal Bovine Serum ( FBS; Invitrogen ) and 0 . 5x Antibiotic-Antimycotic ( Invitrogen ) . Cells were cultured at 2 . 5×106/ml prior to transfection . Transfections were performed with FuGene 6 ( Roche ) according to the manufacturer's instructions . Cu2SO4 ( Sigma ) was added to the medium at a final concentration of 0 . 5 mM 16 hours after transfection , and cells were harvested 48 hours after induction . To stimulate JAK/STAT activation in S2 cells , 2 mM H2O2 and 1 mM sodium vanadate ( pervanadate ) were added to the medium and cells were harvested at desired times after treatment . Treated S2 cells were harvested in cell lysis buffer ( from Cell Signaling Tech . ) for Western blotting or ChIP experiments . ChIP experiments were carried out according to standard protocols with the following modifications . 200 µl of early embryos ( 1–2 h AEL ) or 1×107 S2 cells were treated with 1% formaldehyde at room temperature for 20 min ( embryos ) or 10 min ( cells ) to crosslink protein with chromatin/genomic DNA . Embryos or cells were homogenized and lysed in 300 µl of RIPA lysis buffer with 2 mM EDTA and protease inhibitors on ice . The lysate was sonicated to shear the genomic DNA to lengths between 500 and 1000 bp . An aliquot ( 50 µl ) of sonicated sample was saved as the input control . 5 µg goat anti-STAT92E ( Santa Cruz , CA ) or rabbit anti-V5 antibodies were added to 200 µl experimental samples in RIPA buffer with 2 mM EDTA and protease inhibitors , and the mixture was incubated overnight at 4°C with rotation . Protein G beads ( Sigma ) , pre-blocked with sonicated salmon sperm DNA ( Stratagene ) , were added to precipitate the antibody-bound chromatin and the precipitate was washed extensively . After reversing the crosslink , DNA was recovered by using a Qiagen PCR purification kit and quantified by PCR . The following forward and reverse primers ( flanking two STAT-binding sites in the respective promoter regions ) were used for PCR reactions . dpp: AATTCCGGATAGCGCCTGG , AAAGATGGCACACGCTGGG . Kr: CATGCGTTTGCATACTGGAG , CTATTCGAATCGCCCTTGTC . tll: AGTGCTTTGAGGTCGGAATG , AAGAAACCGTGGTGTCCTTG . Stat92E: TGACTGCCCGCTTTTATACC , CAAACGGCGGTCAATAGTTT . | In the initial phase of the early embryo , transcription is inactive and development is supported by maternally derived gene products . During a time window termed the maternal-to-zygotic transition ( MZT ) , the maternal gene products are degraded and the zygotically expressed genes required for embryogenesis initiate their transcription . How the dramatic upregulation of zygotic genes during the MZT is achieved is not completely understood , although it has been shown that the transcription factor Zelda plays a critical role . In this manuscript , we show that Drosophila STAT ( STAT92E ) functions as a general transcription factor that , together with Zelda , induces transcription of a large number of early-transcribed zygotic genes during the MZT . We further show that STAT92E and Zelda synergistically regulate transcription . Thus , multiple transcription factors , such as STAT92E and Zelda , cooperate to control transcription of the zygotic genome at the onset of embryonic development . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology",
"model",
"organisms",
"biology"
] | 2011 | STAT Is an Essential Activator of the Zygotic Genome in the Early
Drosophila Embryo |
To escape immune recognition in previously infected hosts , viruses evolve genetically in immunologically important regions . The host’s immune system responds by generating new memory cells recognizing the mutated viral strains . Despite recent advances in data collection and analysis , it remains conceptually unclear how epidemiology , immune response , and evolutionary factors interact to produce the observed speed of evolution and the incidence of infection . Here we establish a general and simple relationship between long-term cross-immunity , genetic diversity , speed of evolution , and incidence . We develop an analytic method fusing the standard epidemiological susceptible-infected-recovered approach and the modern virus evolution theory . The model includes the factors of strain selection due to immune memory cells , random genetic drift , and clonal interference effects . We predict that the distribution of recovered individuals in memory serotypes creates a moving fitness landscape for the circulating strains which drives antigenic escape . The fitness slope ( effective selection coefficient ) is proportional to the reproductive number in the absence of immunity R0 and inversely proportional to the cross-immunity distance a , defined as the genetic distance of a virus strain from a previously infecting strain conferring 50% decrease in infection probability . Analysis predicts that the evolution rate increases linearly with the fitness slope and logarithmically with the genomic mutation rate and the host population size . Fitting our analytic model to data obtained for influenza A H3N2 and H1N1 , we predict the annual infection incidence within a previously estimated range , ( 4-7 ) % , and the antigenic mutation rate of Ub = ( 5 − 8 ) ⋅ 10−4 per transmission event per genome . Our prediction of the cross-immunity distance of a = ( 14 − 15 ) aminoacid substitutions agrees with independent data for equine influenza .
Spread of many RNA viruses occurs as a race between host immune responses and rapid viral evolution . The development of treatment and effective preventive measures such as vaccines and therapeutic interference particles [1–3] requires understanding of the mechanics of viral evolution at the scale of a population . To evade immune recognition by hosts previously exposed to infection , in a never-ending chase , viruses accumulate mutations in immunologically relevant regions of the genome [4] . Despite advances in the collection and analysis of epidemiological and genomic data , it remains conceptually unclear how epidemiology , immune response , and evolutionary factors interact to produce the observed evolution speed and the incidence of infection . Influenza virus infects 5-15% of the world population . The global spread and reinfection of the same individuals is caused by rapid evolution of antibody-binding regions [4] . A large amount of information has been obtained , including world-wide circulation [5–7] , genetic maps of virus variants and antibodies , molecular mechanisms , and fitness effect of specific mutations [4 , 8–10] . Vigorous data analysis and computer simulation helped to understand many features of influenza virus evolution [7 , 11–15] . In particular , recent work [15] offers an inference model to predict short-term evolution of influenza , which is helpful for optimization of vaccination strategy . However , the more general connection between the population-scale viral parameters and its evolutionary behavior is still lacking . The aim of this work is to establish general and simple relationships for the speed of virus evolution , genetic diversity , and annual incidence in terms of population parameters , and to train them on the available data for influenza virus . We propose a general analytic approach combining a susceptible-infected-recovered ( SIR ) framework [11 , 16] with the stochastic evolution theory [17–25] . Using the experimental observation that phylogenetic trees of influenza virus have a vine-like structure with short branches [4] , we focus on virus evolution along the one-dimensional trunk . Analysis demonstrates that the evolution under immune memory occurs in the form of a traveling wave in antigenic space , with fitness landscape moving together with the wave . The fitness slope ( effective selection coefficient ) can be expressed in terms of the cross-immunity distance . We provide analytic predictions for the speed , incidence , and the average time to most recent common ancestor in terms of population parameters , including reproduction number , population size , and cross-immunity distance . Then we discuss how the punctuated nature of influenza evolution alternating small-effect and large mutations [4 , 14] may be interpreted from the stochasticity of evolution .
We start by describing briefly our model and approach . The details are given in S1 Appendix . Standard models of evolution focus on the dynamics of virus strains ( variants ) , while standard epidemiological models study the transmission of a virus in a host population . For viruses that evolve to evade immune memory of previously infected hosts , evolutionary and epidemiological dynamics are tightly coupled [26] . Here we adopt a strain-based formulation of epidemiological models , in which all individuals are infected or recovered . Recovered individuals are classified according to their current ability to respond to various viral strains which represent genetic variants of an antibody-binding region of the virus ( e . g . , hemaglutinin gene for influenza virus ) . Each infected individual is assumed to be infected with a single strain denoted by x . We measure the “antigenic coordinate” x as genetic distance in terms of non-synonymous nucleotide substitutions . Infection by a viral strain is cleared in several days or weeks leaving in the recovered individual immunological memory that provides full protection against the same strain and partial protection against infection by genetically close strains . We assume one-dimensional space , x , that represents the trunk of the phylogenetic tree . For each recovered individual , we keep track only of the memory of the most recent infection [11 , 12] . In S1 Appendix , Section 1 . 3 . 3 , we show that this approximation has a modest effect on the final results . Let i ( x , t ) denote the density of individuals currently infected with strain x , and r ( x , t ) be the density of individuals whose last infection was with strain x and who then recovered . The model is represented by a system of differential equations that describe the dynamics of the distributions i ( x , t ) and r ( x , t ) : d r ( x , t ) d t = - r ( x , t ) R 0 ∫ x ∞ K ( x - y ) i ( y , t ) d y + i ( x , t ) , d i ( x , t ) d t = i ( x , t ) [ R 0 ∫ - ∞ x K ( y - x ) r ( y , t ) d y - 1 ] + ( mutation term ) ( 1 ) We assume that each individual is either infected or recovered , as given by the normalization condition ∫ - ∞ + ∞ [ r ( x , t ) + i ( x , t ) ] d x = 1 . The treatment of mutations , which are assumed to be rare , will be described below in subsection Mutation . Eq 1 describe the following epidemiological processes . Firstly , recovered individuals from strain x can be infected with strain y and their susceptibility is proportional to the cross-immunity function K ( x − y ) , which depends on the genetic distance between strains x and y , so that K ( x − y ) > 0 , y > x; K ( x − y ) ≡ 0 , y < x; K ( −∞ ) = 1 . Here we assume that individuals recovered from strain x can be infected only by strains y ahead of x in time , y > x , so that K ( u ) is zero when its argument u is zero or positive ( Fig 1 , blue curve ) . In fact , there is a narrow region at the leading edge , where the backward infection could be possible . However , since the edge region is very narrow in the parameter range of interest , this process has a minor effect on the results ( see the details in S1 Appendix , Section 1 . 3 . 2 ) . Secondly , infected individuals with the density i ( x , t ) recover . Thirdly , individuals infected with a strain x may produce a mutant strain x′ with a small probability , as explained below ( Mutation ) . We measure time in the units of infectious period , trec , so that recovery rate is 1 , and transmission rate equals the basic reproduction number , R0 , defined as the reproduction number in a population of previously uninfected individuals . So far we have considered only dynamics of strains x which already exist . What drives the antigenic evolution is the emergence of new viral strains . Each strain x occasionally and accidentally undergoes a mutation event which changes its ability to be recognized by antibodies ( phenotype ) . We describe this as a variable change in its antigenic coordinate Δx > 0 . The new influenza strain with a new antigenic coordinate , x + Δx , is either cleared from the individual or ( with small probability ) transmitted to another person . The model parameters describing random mutations are the average rate Ub per genome per infectious period ( Table 1 ) and the distribution of Δx . The actual distribution may be quite complex [27]; here , we consider a class of exponential distributions [23] . Specifically , we assume that with each mutation , the value of Δx is drawn randomly with the following probability density ρ ( Δ x ) = e - ( Δ x ) β Γ ( 1 + 1 / β ) , ( 2 ) where β is a fixed parameter . Below in Results , we introduce the critically important factor of random genetic drift [28 , 29] by allowing the number of new infections to vary randomly among the sources of transmission . The model parameters and their estimates used in the analysis are summarized in Table 1 .
Below we analyze epidemiological dynamics in two steps . First , we assume that , in the realistic parameter range , a ≫ 1 , the frequency of infected individuals , i ( x , t ) represents a solitary peak , much more narrow in genetic distance x than the frequency of recovered individuals , r ( x , t ) . Using this fact , we find analytically the form of r ( x , t ) . Second , we apply the well-developed theory of asexual evolution [18–21 , 23] to obtain parameters of the distribution of infected individuals i ( x , t ) . Details are given in S1 Appendix; here we present the main steps of the derivation . We start our analytic derivation by noting that , in the limit of small mutation rates , the main role of mutation is to form new strains with antigenic coordinate x larger than for already existing strains . For already existing strains , mutation is negligible . This assumption is intuitively clear and is verified in the relevant parameter range , using estimates of mutation rate Ub ( Table 1 ) . Neglecting the mutation term in Eq 1 , we seek for a traveling wave solution of the form r ( x , t ) = r ( x − ct ) and i ( x , t ) = i ( x − ct ) where x − ct ≡ u is the relative antigenic coordinate of a strain and c = d 〈x〉/d t is the wave speed defined as the average number of non-synonymous nucleotide substitutions per year . Without loss of generality , we choose the peak of the infected wave i ( u ) to be at u = 0 , [di ( u ) /du]u=0 = 0 . The traveling wave solution of Eq 1 for infected and recovered individuals then reads i ( u ) = A c f ( u ) , r ( u ) ≈ { A exp [ - A R 0 ∫ u 0 K ( v ) d v ] , u < 0 , 0 , u > 0 , ( 3 ) where A is a constant found from the normalization condition ∫ - ∞ + ∞ [ r ( u ) + i ( u ) ] d u = 1 , and f ( u ) is a narrow peak with unit area and a width much less than the width of the recovered distribution , r ( u ) . The wave speed c and the shape of the infected density f ( u ) are to be determined later on . At large R0 , K ( v ) in Eq 3 can be expanded linearly near zero , so that density of the recovered becomes a half of a Gaussian r ( u ) ≈ 2 R 0 π a e - ( R 0 u a π ) 2 , u < 0 ; 0 , u > 0 ( 4 ) and A = 2R0/ ( πa ) . The fraction of infected individuals in population N inf N = ∫ - ∞ ∞ i ( u ) d u = A c = 2 R 0 c π a ( 5 ) is assumed to be much smaller than 1 . Then the annual incidence of infection is expressed in terms of cross-immunity distance , evolution speed , and basic reproduction number as Annual incidence = 2 R 0 c π a 365 t rec , ( 6 ) which is a directly testable prediction . Analytic solution , Eqs 3 and 4 , is based on the assumption that the infected wave i ( u ) is much more narrow than the recovered wave r ( u ) . To verify the validity of this approximation , we compare the Eq 3 with Monte-Carlo simulation based on Eq 1 . The simulation confirms the existence of a steady traveling wave with two linked components moving to the right in antigenic coordinate ( Fig 1 ) . Infected wave i ( u ) is , indeed , a narrow peak . The time-averaged solution for recovered individuals obtained from simulation agrees fairly well with the analytic prediction ( black line ) . Recovered wave r ( u ) displays a sharp increase near the maximum of i ( u ) and a slowly decaying tail at u < 0 . The sharp increase is due to continuous recovery of infected individuals . The decaying tail is caused by reinfection of recovered individuals once they become genetically remote from the moving front of wave r ( u ) . This derivation captures only the shape of the recovered peak leaving the narrow infected peak undefined . In order to determine the infected individual distribution , i ( u ) , we use standard traveling wave theory [18–23] . The interesting feature of the selection due to immune escape is that the fitness landscape which controls the traveling wave travels with the wave . Moreover , it is the wave itself which creates its own landscape , as follows: the recovered create a landscape for the infected evolution , which moves the recovered distribution forward in x , and so on . To derive the form of landscape on the human population level , we use the standard definition of viral fitness as the average number of secondary infections caused by an infected individual [28 , 32–34] . ( The reproductive number must not to be confused with the basic reproductive number R0 , which is its maximum value , i . e . the value in a totally susceptible population . ) Here we choose to define fitness w ( x , t ) as the log of R0 − 1 , i . e . , the exponential expansion rate of the density of infected individuals i ( x , t ) measured per infectious period: w ( x , t ) = ∂ ln i ( x , t ) ∂ t = R 0 ∫ - ∞ x K ( y - x ) r ( y , t ) d y - 1 . ( 7 ) The form of w ( u ) obtained from Eqs 7 and 3 is shown in Fig 2 ( red line ) . The asymptotic cases of the fitness landscape w ( u ) are w ( u ) ≈ { R 0 - 1 , u ≫ a , σ u , | u | ≪ a , - 1 , u < 0 , | u | ≫ a . ( 8 ) where σ = - R 0 ∫ - ∞ 0 d K d u r ( u ) d u ( 9 ) has the meaning of the fitness landscape slope , or the average selection coefficient . According to Eq 8 , w ( u ) is positive for u > 0 and negative for u < 0 , indicating that viruses are selected for in front of the infected peak and selected against in the wake of the wave . For large positive or negative u , |u| ≫ a , we predict saturation of w ( u ) . At u = 0 , w ( 0 ) = 0 , which is equivalent to the fact that the actual reproduction number is exactly 1 at the peak of the wave . Within the range |u| ≪ a , where the narow peak of the infected individuals is located , fitness landscape can be expanded linearly with slope σ > 0 which represents the average selection coefficient of a mutation event . For sufficiently large R0 , from Eqs 4 and 9 , σ can be approximated by a series in 1/R0 σ ( a , R 0 ) = 1 a [ R 0 - 2 + 3 π 2 R 0 + O ( 1 R 0 2 ) ] , ( 10 ) where a ≡ 1/|K′ ( 0 ) | , and the second and third terms are small corrections to the first term . Thus , the average selection coefficient σ of the traveling fitness landscape is inversely proportional to the cross-immunity distance a . It also increases linearly with the basic reproduction ratio R0 when R0 is large . The two correction terms in Eq 10 depend on the form of cross-immunity function in Table 1 . For an alternative form K ( x ) = 1 − exp ( −x/a ) , they are smaller by factors of 2 and 6 , respectively . The overall agreement for the entire landscape w ( u ) between the analytic prediction and simulation is quite good ( Fig 3 ) . We get further insight into the dynamics of the model by predicting the speed of viral evolution c . So far , we have left this value undetermined because it weakly affects the shape of the density of recovered individuals r ( x , t ) , Eq 3 . In contrast , the density of infected individuals i ( u ) , which is much more narrow , needs to be determined simultaneously with c . Our result for the average selection coefficient σ , Eq 10 , reduces the problem of epidemiological evolution to models of asexual populations with many diverse sites where the speed was derived previously in terms of population size , selection coefficient and mutation rate ( [18–23] ) . We consider a case with randomly distributed selection coefficient s = σΔx , where mutational distance Δx is sampled from distribution in Eq 2 with large parameter β . This section contains the central result of our analysis: Antigenic diversity Var[x] = < ( Δx ) 2 > and adaptation rate v defined as the average rate of fitness increase ( “fitness flux” ) depend on crossimmunity range a and other parameters [23] V a r [ x ] = 2logN inf σ log ( β σ / U b ) ( 11 ) v = σ 2 V a r [ x ] ( 12 ) Another measure of evolution rate is the average substitution rate c c = ( σ 2 /s* ) V a r [ x ] ( 13 ) s * = σ [ 2 βlog σ U b ]1 β - 1 ( 14 ) where s* represents the most probable fitness gain of a mutation established in a population [23] . Note that s* is larger than the average selection coefficient σ . The expressions for Var[x] and s* are approximate , within the accuracy of logarithms inside the large logarithms . For more accurate expressions , see S1 Appendix . To apply these results to our case of antigenic evolution , we substitute average selection coefficient σ from Eq 10 and infected population size Ninf from Eq 5 . Then the metrics of evolution speed c , v are expressed in terms of a and epidemiological parameters ( Table 1 ) . In the limit of very large β , Eqs 11–14 match results of a model with constant selection coefficient σ [18 , 20] . We verified analytic results for wave speed c by Monte-Carlo simulation in a wide range of N and Ub ( Fig 3 ) . We used two methods: full simulation of initial Eq 1 with randomly distributed mutational effects , and a reduced Moran algorithm with linearized fitness landscape ( symbols in Fig 3 ) . We observe that our analytic prediction of a logarithmic increase of c with N and Ub follows simulation quite well , except at smallest Ub and N explored in our study . Logarithmic dependencies are characteristic for asexual evolution models ( [18–23 , 35 , 36] ) . Abbreviations IS , CI , MM near symbols indicate different regimes regarding the number of genomic sites evolving within the same time frame: selection sweeps at isolated sites ( IS ) , pairwise clonal interference ( CI ) [23 , 35 , 36] , and multiple-mutation regime ( MM ) [18–21 , 23] . The traveling wave models are designed for MM regime , which explains the discrepancy at smallest Ub and N . We also observe that the steepness of the selection coefficient distribution , β , weakly affects the predicted speed . Our analysis predicts that substitution rate of antigenic mutations c , Eq 13 , is inversely proportional to the cross-immunity distance a and increases logarithmically with host population size and mutation rate . The average selection coefficient at the population level , σ , is also inversely proportional to a , Eq 10 . An alternative measure of the evolution speed , the adaptation rate v , Eq 12 , is inversely proportional to a2 . The annual incidence of infection , Eq 6 also scales as 1/a2 . Taking advantage of recent theoretical progress in asexual phylogeny [24 , 25 , 38] , we also calculated an important observable quantity , the time to the most recent common ancestor of two co-existing viruses ( S1 Appendix , Eqs S20-S21 ) . T MRCA2 = z 2log ( N σ ) v ( 15 ) Here numeric factor z depends on the distribution of mutational effect Δ[x] [24 , 25] . The predicted values are z = 1 . 5 in the case of fixed mutational effect Δ[x] , and z = 3 in the case of the Gaussian distribution of Δ[x] ( Eq 2 with β = 2 ) . Because the Gaussian case is more realistic , and because we are not aware of any results for TMRCA2 for other forms of distribution , below we choose the value β = 2 for data fitting . To test the model , we compared its predictions with available data on influenza A H3N2 and H1N1 , as follows . The input parameters of the model and the output ( predicted ) parameters are summarized in Table 1 . The values of input parameters such as population size N , reproduction ratio in the absence of immune recognition R0 ( during a major pandemic caused by antigenic shift ) , and recovery time trec have been measured [7 , 13 , 30 , 31] . In contrast , parameters a and Ub result from biological interactions at multiple biological scales ( cell , host , population ) and are hard to come by . On the other hand , data on two parameters predicted by the model , TMRCA2 and substitution rate c , are available . Therefore , we opted to adjust the unknown input parameters a and Ub to fit available data for the two predicted parameters ( Fig 4A ) . We assumed a total susceptible population of N = 108 individuals , which corresponds to a large country . It is evident that strain H2N3 has a faster evolution rate and a shorter time TMRCA2 than strain H1N1 due to a larger value of R0 causing , in turn , a larger average selection coefficient σ . The values of Ub and a for the two strains are similar ( Fig 4a ) . The best-fit values for the cross-immunity distance , a = 14 − 15 , agree very well with independent data on equine influenza [37] , which represents a direct confirmation of the model . The predicted annual incidence in humans of ( 4 − 7 ) % also falls within the experimentally observed range and previous modeling estimates [12 , 13 , 15] . Interestingly , the model explains the inverse correlation between TMRCA2 and evolution rate c reported previously for H2N3 , H1N1 and two strains of influenza B [7] . Indeed , the predicted evolution rate c is linearly proportional to the effective selection coefficient σ ∝ R0/a , while TMRCA2 is inversely proportional to σ . The dependence of c and TMRCA2 on the other parameters , Ub and N , is logarithmically slow . To generalize our results for epidemics occurring on larger or smaller scales , we calculated the dependence of c , TMRCA2 , and the annual incidence on population size N ( Fig 4B ) . The sensitivity of our predictions to input parameters Ub , a , and R0 has also been tested ( S1 Appendix , S3 and S4 Figs ) . Thus , traveling wave theory with modest selection predicts logarithmic dependence of the speed on population size ( Fig 4B ) . Epidemiological data demonstrate that , a priori , antigenic space is not one-dimensional but has fractal nature and fractal dimensionality more than 1 [8 , 31] . To demonstrate the weak sensitivity of our model to the existence of additional dimensions , we extended our model to a discrete random tree of epitope variants and solved it numerically ( S1 Appendix , S6 Fig ) . Phylogeny demonstrates quasi-1D behavior comprising a long trunk of permanently fixed mutations and short branches representing transient virus variants and resembling the actual influenza H3N2 phylogeny [4 , 12 , 13 , 15] . We also confirmed the formation of a 1D traveling wave for two-dimensional genetic space ( S5 Fig ) .
We investigated stochastic evolutionary dynamics of a virus driven by the pressure to escape immune recognition in previously infected individuals . We mapped this problem to an evolutionary model with fitness landscape expressed in terms of the cross-immunity function K ( x ) ( Fig 2 ) . Stochastic evolution occurs as a traveling wave with two population components structured in the antigenic variant space x , recovered individuals and the currently infected individuals , with different widths and total counts ( Fig 1 ) . The recovered distribution is broad and large . The infected distribution represents a narrow and small peak at the recovered distribution front . We expressed several observable parameters including the speed of viral evolution , the annual incidence of infection , and the average time to the most recent ancestor in terms of model parameters N , Ub , R0 , K ( x ) ( Table 1 ) . The analytic predictions agree with simulation and are able to estimate correctly important parameters of viral evolution in host populations , as we illustrated using genomic data on influenza . One of the puzzling aspects of influenza virus evolution is is punctuated nature [4] . While most mutations are almost neutral or have a modest phenotypic ( fitness ) effect , some represent large jumps in antibody recognition [14] . Our results interpret these jumps as a natural consequence of the stochastic nature of the traveling wave models . The extension of the leading edge of a wave occurs due to adding rare , best available escape alleles . Asexual evolution theory with variable fitness effect of mutations demonstrates that most fixed mutations have a fitness effect in excess of average fitness effect [23] . Good et al show that the most likely selection coefficient s* that drives the wave depends on model parameters σ , N , Ub , mapping the results either onto the multiple-mutation ( MM ) model with fixed s [18–21] or the two-site clonal interference ( CI ) model [35 , 36] . Present work demonstrates that influenza virus evolves within MM regime near the border with CI regime ( Fig 3 ) . In this region , the fitness effect of a fixed allele is predicted to fluctuate strongly around the most likely value s* , which represents a possible explanation of the punctuated effect . An SIR model with immune memory and 1D antigenic space ( Eq 1 ) has been previously proposed by Lin et al [11] . Their analysis differs from ours in two critical aspects . Firstly , their approach to viral evolution was completely deterministic , i . e . assumes infinite population size . In fact , the effect of clonal interference acting in finite population diminishes antigenic return on additional mutations . Secondly , their mutation term in Eq 1 had a diffusion form proportional to the second derivative of the infected individual density , ∂2i ( x , t ) /∂x2 . This approximation would be correct if the front edge of the wave was smooth . As we discuss in S1 Appendix , neither approximation holds at low mutation rates , Ub ∼ 10−4 . As a result , the approach of Lin et al predicts evolution speeds far below simulation results . The traveling wave approach employed here naturally accounts for both the stochastic effects and the steepness of the leading edge . Future development of this model requires inclusion of finite mutation cost [39] . Our analytic results agree with the numeric results of a previous simulation by Bedford et al [12] . Using a similar model , they predicted the same incidence range for influenza A , the same range for the evolution speed , and interpreted the quasi-one-dimensional trajectory in the genetic space we have also observed ( S5 and S6 Figs ) . As starting parameters , they assumed mutation rate Ub ∼ 10−4 and set the cross-immunity distance to be a = 1/0 . 07 based on equine flu data [37] . By comparison , here we determine Ub and a a posteriori from fitting human H3N2 and H1N1 data on c and TMRCA from the cited work [7] . We test the model by comparing our prediction with the experimental value of a [37] .
Merging the standard epidemiological approach and the modern traveling wave theory , we develop a general analytic approach that connects epidemiological and immunological parameters to the observed parameters of influenza evolution . We demonstrate that the distribution of recovered individuals in the genetic space effectively creates a fitness landscape for the infected individual distribution , and both distributions move together along quasi-one-dimensional path . Our predictions demonstrate a good experimental agreement with data on influenza A H3N2 . | Spread of many RNA viruses in a population represents a competition between host immune responses and viral evolution . RNA viruses accumulate mutations in immunologically important regions to escape immune recognition in hosts previously exposed to infection , while the immune system responds by producing new memory cells . Despite recent advances in data collection and their analysis , it remains conceptually unclear how epidemiology , immune response , and evolutionary factors interact to produce the observed speed of evolution and its incidence . By combining the standard epidemiological approach with the modern theory of viral evolution , we predict a general relationship between long-term cross-immunity , antigenic diversity of virus , its evolution speed , infection incidence , and the time to the most recent common ancestor . We apply these theoretical findings to available data on influenza virus to determine two important parameters of its evolution and confirm the model . Current strategies of vaccination against influenza should take into account stochastic fluctuations in fitness effect of mutations predicted by the theory . | [
"Abstract",
"Introduction",
"Model",
"and",
"methods",
"Results",
"Discussion",
"Conclusion"
] | [
"organismal",
"evolution",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"influenza",
"pathogens",
"microbiology",
"orthomyxoviruses",
"viruses",
"rna",
"viruses",
"microbial",
"evolution",
"waves",
"traveling",
"waves",
"genetic",... | 2018 | Antigenic evolution of viruses in host populations |
Plasmodium vivax invasion of human erythrocytes depends on the Duffy Binding Protein ( PvDBP ) which interacts with the Duffy antigen . PvDBP copy number has been recently shown to vary between P . vivax isolates in Sub-Saharan Africa . However , the extent of PvDBP copy number variation , the type of PvDBP multiplications , as well as its significance across broad samples are still unclear . We determined the prevalence and type of PvDBP duplications , as well as PvDBP copy number variation among 178 Ethiopian P . vivax isolates using a PCR-based diagnostic method , a novel quantitative real-time PCR assay and whole genome sequencing . For the 145 symptomatic samples , PvDBP duplications were detected in 95 isolates , of which 81 had the Cambodian and 14 Malagasy-type PvDBP duplications . PvDBP varied from 1 to >4 copies . Isolates with multiple PvDBP copies were found to be higher in symptomatic than asymptomatic infections . For the 33 asymptomatic samples , PvDBP was detected with two copies in two of the isolates , and both were the Cambodian-type PvDBP duplication . PvDBP copy number in Duffy-negative heterozygotes was not significantly different from that in Duffy-positives , providing no support for the hypothesis that increased copy number is a specific association with Duffy-negativity , although the number of Duffy-negatives was small and further sampling is required to test this association thoroughly .
Plasmodium vivax and P . falciparum are the two major parasite species that cause clinical malaria worldwide . While P . falciparum causes most malaria mortality , P . vivax can also cause severe disease and , more rarely death [1 , 2] . P . vivax is prominent in India , Southeast Asia , and South America . By contrast , vivax malaria is rare in sub-Saharan Africa , with significant rates of infection only found in a few countries [3] . The rarity of this infection is tightly associated with high prevalence of Duffy negativity in West and Central Africa [4] . This association represents a classic example of innate resistance to a pathogen resulting from the absence of a receptor for the pathogen [5] . A point mutation in the GATA-1 transcription factor binding site of the Duffy antigen/receptor for chemokines ( DARC ) gene promoter eliminates Duffy antigen ( Fy ) expression on the RBC surface required for P . vivax invasion [6] . This phenomenon is thought to account for the lack of P . vivax malaria in sub-Saharan Africa . In Ethiopia , P . vivax and P . falciparum account for approximately 40% and 60% of the malaria cases , respectively [7 , 8] . The proportion of Duffy negatives in Ethiopia is about 35% [9 , 10] , considerably lower than that documented in West and Central Africa ( >97% ) [4] . About 20% of hospital patients presented with malaria symptoms are Duffy-negatives [10–11] . While Duffy negativity was previously thought to be associated with complete resistance to P . vivax , P . vivax infection in Duffy-negative individuals was confirmed unequivocally in Madagascar [12] , and has also been observed in multiple other parts of Africa as well as other parts of the world [13–18] . Plasmodium vivax invasion of human erythrocytes relies on interaction between the Duffy antigen and P . vivax Duffy Binding Protein ( PvDBP ) [19–20] , though other reticulocyte ligand-receptor have also been recently identified [21] . Our recent study has shown that mutations in PvDBP do not explain the ability of P . vivax parasites to infect Duffy-negative individuals [22] , but there could be other molecular mechanisms at play . Whole genome sequences from P . vivax field isolates in Madagascar identified a duplication of the PvDBP gene [23] and PvDBP duplication has also been detected in non-African P . vivax-endemic countries [24 , 25] . In Ethiopia , PvDBP expansion has been reported with a 79% prevalence in a recent analysis of 24 P . vivax genomes [26] . However , sample size was relatively limited and the extent of copy number variation among broad samples is still unclear . The high prevalence of PvDBP duplications raises the possibility that it is linked to the ability of P . vivax to infect Duffy-negative individuals . Two types of PvDBP duplications have been reported , termed Cambodian and Malagasy-type duplications based on the samples in which they were first described [23–24] . Both create a complete duplication of the PvDBP gene , but they differ in the sequence and length of the 3’-boundary of the linker region between the tandem copies of PvDBP [24] . In this study , we used whole genome sequencing and PCR-based diagnostic methods to determine the prevalence of both the Madagascar- and Cambodia-type PvDBP duplications in southwestern Ethiopia where a high incidence of P . vivax infections and a modest proportion of Duffy-null individuals co-occur . We compared PvDBP copy number between Duffy-null heterozygotes ( FyA/BES or FyB/BES ) and homozygotes ( FyBES/BES ) , between symptomatic and asymptomatic infections , and explored the correlation of gene copy number with parasitemia , age , gender , ethnicity , and malaria symptoms in order to shed light on the epidemiological significance of PvDBP duplications .
Scientific and ethical clearance was given by the institutional scientific and ethical review boards of the Jimma University , Ethiopia , and University of North Carolina at Charlotte , USA . Written informed consent/assent for study participation was obtained from all consenting heads of households , parents/guardians ( for minors under age of 18 ) , and each individual who was willing to participate in the study . All experimental protocols were reviewed and approved by the institutional review board ( IRB ) of the University of North Carolina at Charlotte , USA and performed in accordance with the IRB guidelines and regulations . A total of 178 P . vivax samples were included in this study . They were isolated from 145 symptomatic patients and 33 asymptomatic volunteers in Jimma , Ethiopia that had previously been confirmed by microscopy and PCR to be infected with P . vivax [9 , 10] . Symptomatic samples were obtained from patients who visited the health centers/hospitals in the Jimma town and presented with malaria signs/symptoms including axillary temperature ≥37°C , vomiting , diarrhea , nausea , abdominal pain , chills , fatigue , muscle pain , headache , malaise or loss of appetite ( S1 Table ) . The asymptomatic samples were obtained from the community in the same area and had no fever or other malaria symptoms at the time of collection . Thick and thin blood smears were prepared for each subject to screen for P . vivax by microscopy . Parasites were counted against 200 leukocytes and a slide was considered negative when no parasites were observed after counting over 100 microscopic fields . All slides were read in duplicate by two microscopists at the time of sample collection . The density of parasitemia was expressed as the number of asexual P . vivax per microliter of blood , assuming a leukocyte count of 8000 per microliter . Three to four blood spots , each equivalent to ~50μl of blood , were blotted on Whatman 3MM filter paper from each participating individual . Parasite DNA was extracted from dried blood spots by the Saponin/Chelex method [27] and genomic DNA was eluted in a total volume of 200 μl TE buffer . Eluted DNA was confirmed with P . vivax by nested and quantitative PCR using the 18S rRNA P . vivax-specific primers [10] , prior to PCR diagnosis of PvDBP duplications and quantification of copy number . For a subset of 20 symptomatic samples that were confirmed with P . vivax , 6ml whole blood was collected from the patients . We used the Lymphoprep/Plasmodipur-based protocol to deplete white cells and enrich red cells before DNA extraction [28] . DNA was extracted from the enriched RBCs using the ZymoBead Genomic DNA kit ( Zymo Research ) following the manufacturer’s procedures . Whole genome sequences were obtained for these samples on a HiSeq Illumina Sequencing Platform at the Wellcome Trust Sanger Institute ( ENA accession number of each sample in Table 1 ) . The generated sequence reads were mapped individually to the P . vivax PVP01 reference genome [29] using bwa version 0 . 5 . 9-r16 with default parameters [30] . The resulting assembly bam files were reviewed in the region containing PvDBP ( chromosome 6: 976329–980090 ) using the Artemis genome viewer [31] . Using a “non-proper pair” read filter in Artemis , mate pairs that were oriented tail-to-tail indicated PvDBP duplications [24] . The amount of P . vivax DNA in each sample was estimated using the SYBR Green quantitative PCR detection method using primers that targeted the 18S rRNA genes of P . vivax [10] . Amplification was conducted in a 20 μl reaction mixture containing 1 μL of genomic DNA , 10 μl 2×SYBR Green qPCR Master Mix ( Thermo Scientific ) , and 0 . 5 uM primer . The reactions were performed in CFX96 TouchTM Real-Time PCR Detection System ( BIORAD ) , with an initial denaturation at 95°C for 3 min , followed by 45 cycles at 94°C for 30 sec , 55°C for 30 sec , and 68°C for 1 min with a final 95°C for 10 sec . This was followed by a melting curve step of temperature ranging from 65°C to 95°C with 0 . 5°C increments to determine the melting temperature of the amplified product . Each assay included positive controls of P . vivax Pakchong ( MRA-342G ) and Nicaragua ( MRA-340G ) isolates , in addition to negative controls including uninfected samples and water . A standard curve was produced from a ten-fold dilution series of the control plasmids to determine the amplification efficiency . Melting curve analyses were performed for each amplified sample to confirm specific amplifications of the target sequence . A cut-off threshold of 0 . 02 fluorescence units that robustly represented the threshold cycle at the log-linear phase of the amplification and above the background noise was set to determine Ct value for each assay . The mean Ct value was calculated from three independent assays of each sample . Samples yielding Ct values higher than 40 ( as indicated in the negative controls ) were considered negative for Plasmodium species . The amount of parasite DNA in a sample was quantified using the follow equation: Parasite DNA ( /uL ) = [2 E× ( 40-Ctsample ) /10]; where Ct for the threshold cycle of the sample , and E for amplification efficiency; assuming 10 , 000 18S rRNA copies/mL ( i . e . , 10 copies/uL ) in each Plasmodium genome [32] . Two sets of primers were used to examine the prevalence and type of PvDBP duplications following published protocols [23 , 24] . The first set of primers includes AF+AR , BF+BR , and BF+AR [23] . Primers BF+AR amplify a 612-bp product that contains the junction between the PvDBP copies in isolates with the Malagasy-type duplications . The second set of primers includes AF2+AR2 , BF+BR , and BF+AR2 [24] . Primers BF+AR2 amplify a 736-bp product that contains the junction between the PvDBP copies in isolates with the Cambodia-type duplication . This primer pair , in theory , should also amplify a 1584-bp product should there be Malagasy-type duplications . Without duplications , BF+AR and BF+AR2 primers are opposite-facing in the samples and thus do not amplify a product [24] . PvDBP copy number was measured with a SYBR Green qPCR detection method using primers ( forward: 5’-AGGTGGCTTTTGAGAATGAA-3’; reverse: 5’-GAATCTCCTGGAACCTTCTC-3’ ) designed between region II to III of PvDBP ( PVX_110810 ) . Plasmodium vivax aldolase , which is known to be a single-copy gene , was used as an internal reference to calculate gene copy number [33] . Two Cambodian isolates , Pv0430 and Pv0431 , which were confirmed previously by whole genome sequencing to contain a single and two PvDBP copies [24] , respectively , were included in each run as positive controls . Water was used as a no DNA control . Amplification was conducted in a 20μL reaction mixture containing 1μL of genomic DNA ( all samples was standardized to ~50 genomes/μL in each reaction based on 18S qPCR assay; S2 Table ) , 10μL 2×SYBR Green qPCR Master Mix ( Thermo Scientific , USA ) , and 0 . 5μM of each primer . Reaction was performed in CFX96 TouchTM Real-Time PCR System ( Bio-Rad ) with an initial denaturation at 95°C for 3 min , followed by 45 cycles at 94°C for 30 sec , 55°C for 30 sec , and 68°C for 1 min with a final 95°C for 10 sec [22] . This was followed by a melting curve step of temperature ranging from 65°C to 95°C with 0 . 5°C increments to determine the melting temperature of the amplified product . Each assay included an internal reference P . vivax aldolase as well as the negative controls ( uninfected samples and water ) . Melting curve analyses were performed for each amplified sample to confirm specific amplifications of the target sequence . A cut-off threshold of 0 . 02 fluorescence units that robustly represented the threshold cycle at the log-linear phase of the amplification and above the background noise was set to determine Ct value for each assay . The amplification of the P . vivax aldolase and PvDBP gene fragments was optimized to obtain similar amplification efficiency . The mean Ct value and standard error was calculated from three independent assays of each sample ( S3 Table ) . PvDBP copy number ( N ) was calculated with the equation previously used for Pvmdr1 copy number estimation [33] as follow: N = 2ΔΔCt±SD , where ΔΔCt = ( Ctpvaldo- Ctpvdbp ) - ( Ct pvaldo cal-Ct pvdbp cal ) . The Ctpvaldo and Ctpvdbp are threshold cycle values for the P . vivax aldolase and PvDBP genes respectively , whereas Ctcal is an average difference between Ctpvaldo and Ctpvdbp obtained for the positive control containing a single copy of the P . vivax aldolase and PvDBP gene fragments ( i . e . , the Cambodian isolate Pv0430 in this case ) . The SD value is a standard deviation calculated as follow: SD = √ ( S2pvdbp+S2pvaldo+S2cal ) where Spvdbp and Spvaldo are the standard deviations from the average Ct calculated for three replicates in the P . vivax aldolase and PvDBP gene amplifications . Scal is an average standard deviation of the ΔCt values for the calibrator . The PvDBP copy number of each sample was assessed twice for validation . For all P . vivax positive samples , a 1 , 100-bp fragment of the human Duffy blood group antigen gene that encompasses the GATA1 transcription factor-binding site of the gene promoter was amplified using published primers [12] . PCR reaction contained 20μl DreamTaq PCR Mastermix , 1μl DNA template , and 0 . 5μl each primer . PCR conditions were: 94°C for 2-min , followed by 35 cycles of 94°C for 20s , 58°C for 30s , and 68°C for 60s , followed by a 4-min extension . PCR products were sequenced and the chromatograms were visually inspected to determine the Fy genotypes ( see Duffy blood group nomenclature in [34] ) . A one-tailed t-test was used to test for the significance of differences in PvDBP copy number between symptomatic and asymptomatic P . vivax infections , as well as among Duffy genotypes . In addition , we calculated Pearson's correlation coefficient ( r2 ) in R for the associations of PvDBP copy number with parasitemia and age among the clinical samples . Samples were also stratified by gender and ethnicity to test if there was a significant difference in PvDBP copy number . Malaria symptoms including fever , chills , malaise , fatigue , muscle/joint pain , headache , nausea , vomit , diarrhea , abdomen pain , loss of appetite , and respiratory difficulty ( dependent variables ) were recorded as presence or absence for each of the patients . Principle component analyses were performed to show the variation of all variables among the samples with different PvDBP copy number using the logistic PCA function in R . A 3-dimensonal PCA plot was constructed with the plot3d function from the rgl package .
To establish whether Cambodian or Malagasy-type PvDBP duplications were present in the sample population , whole genome sequences were obtained for a subset of 20 Ethiopian P . vivax samples using Illumina sequencing ( Table 1 ) . For these 20 samples , 13 to 381 million 150 bp pair-end reads were generated and 10–96% of these reads was mapped to the P . vivax reference genome PVP01 . The average P . vivax genome coverage was 16–715× with over 95% of the genome covered by at least 15 reads in the majority of the samples . The average PvDBP coverage was 27–1746× , of which the sequence coverage of the PvDBP region was clearly higher than the flanking regions in some samples ( S1 Fig ) . We further confirmed and identified the type of PvDBP duplications by mapping the paired-end reads in a tail-to-tail orientation , respectively , on the Malagasy M15 [23] and Cambodian PV0431 [24] reference genomes . Based on whole genome sequences of the duplication break points , two out of the 20 P . vivax samples contained the Malagasy-type and 11 contained the Cambodian-type duplications . The remaining had a single copy PvDBP ( Table 1 ) . To test more extensively for PvDBP duplications , we used recently published diagnostic PCR primers [24] . These primers had previously been used to genotype 20 Ethiopian P . vivax symptomatic samples [24] . We genotyped an additional 158 ( 125 symptomatic and 33 asymptomatic ) P . vivax samples , bringing the total tested samples to 178 across this and the previous study . All samples produced bands of the expected size of 650-700bp with the control primers AF/AR , BF/BR , and AF2/AR2 ( Fig 1 ) , indicating the presence of PvDBP in all isolates . Among them , 14 ( 9 . 7% ) showed a fragment of ~600bp with the Malagasy duplication primers BF/AR ( e . g . , sample SGH ( 1 ) -357 in Fig 1 ) , 81 ( 55 . 9% ) showed a fragment of ~800bp with the Cambodian duplication primers BF/AR2 ( e . g . , BBH ( 1 ) -125 in Fig 1 ) , and 50 ( 34 . 5% ) showed no amplifications with either pair of primers ( e . g . , BBH ( 1 ) -132 in Fig 1 ) , suggesting either they only contained a single copy of PvDBP or a duplication not detected by either primer pair . For the 14 samples that showed amplification with the Malagasy duplication primers BF/AR , two were amplified with the Cambodian primers BF/AR2 giving a ~1 , 500bp band . To test whether additional duplication types not detected by these primer pairs were present , we developed a qPCR assay that compared the quantity of PvDBP products to that of a known single copy gene control , P . vivax aldolase . To validate the assay , we compared PvDBP copy numbers estimated by qPCR with those estimated by PCR diagnostic primers ( BF/AR and BF/AR2 ) and from whole genome sequencing for the 20 samples that had been whole genome sequenced ( Table 1 ) . There was a significant correlation between these two metrics ( Fig 2A ) , confirming that the qPCR assay was measuring changes in copy number . For 50 samples that showed no amplification with primers BF/AR and BF/AR2 , indicative of a single copy gene , qPCR copy number estimates ranged from 0 . 85–1 . 77 ( a median value of 1 . 17±0 . 26; Fig 2B ) . Six of these samples had also been whole genome sequenced , and had fold-coverage of PvDBP ranging from 0 . 88–1 . 42 , confirming the presence of only a single PvDBP copy ( Table 1; Fig 2B ) . We therefore used qPCR estimates of >1 . 9 to score samples containing multiple PvDBP copies . Among the 145 symptomatic samples , there were 95 samples with PvDBP qPCR copy number estimates >1 . 9 , and all of these also showed amplifications with BF/AR or BF/AR2 primers , confirming the presence of more than one PvDBP copy . qPCR estimates for these samples ranged from 1 . 90–6 . 91 ( Fig 2B ) , suggesting that some samples might contain higher order amplifications . We subdivided these samples into two categories: ( 1 ) samples with value < 3 . 5 were defined as 2–3 PvDBP copies and ( 2 ) samples with value ≥3 . 5 were defined as ≥4 PvDBP copies . By this definition , 71 samples had 2–3 PvDBP copies ( 48 . 9%; median value of 2 . 42 ± 0 . 49 ) and 24 had ≥ 4 PvDBP copies ( 16 . 5%; median value of 4 . 40 ± 0 . 89; Fig 2B ) . This is to our knowledge the first report of higher order PvDBP copy numbers in a large number of P . vivax isolates . PvDBP fold coverage estimated from whole genome sequences corroborated the qPCR results in 16 out of the 20 samples . In four samples , qPCR data estimated >4 PvDBP copies whereas whole genome sequencing data estimated 2–3 PvDBP copies ( S4A Table ) . For these four samples , we used the fold-coverage based on the whole genome sequences to score their copy number category . In total , 95 of 145 symptomatic samples tested ( 65 . 5% ) had more than a single copy of PvDBP . For the 33 asymptomatic samples , two ( 6 . 1% ) were found with multiple PvDBP copies ( Fig 3; S4B Table ) . The qPCR data was further confirmed with the PCR diagnostic primers BF/AR and BF/AR2 for all the 33 asymptomatic samples ( S4B Table ) . No amplification was observed in 31 samples , indicative of the absence of duplication . In two asymptomatic samples , amplification was observed with primers BF/AR2 , indicative of the Cambodian-type duplication . Two PvDBP copies were detected by qPCR in these two asymptomatic samples . A ten-fold higher rate of multi-copy PvDBP infections was observed in symptomatic patients compared to the asymptomatic volunteers ( P < 0 . 001 ) . Based on DARC gene sequences , 101 of 145 ( 69 . 7% ) symptomatic P . vivax-infected samples were Duffy-null heterozygote ( FyA/BES or FyB/BES ) , two ( 1 . 4% ) were Duffy-null homozygote ( FyBES/BES ) , 17 ( 11 . 7% ) were Duffy-positive homozygote ( FyA/A or FyB/B ) and 25 ( 17 . 2% ) were Duffy-positive heterozygote ( FyA/B or FyB/X2; S5 Table ) . Significant differences were observed in PvDBP copy number between FyA/BES and FyB/BES , FyB/B and FyA/B , FyB/BES and FyA/B , as well as FyA/BES and FyB/B ( Fig 4 ) . While both being Duffy-null heterozygotes , FyA/BES indicated a significantly higher PvDBP copy number than FyB/BES . Among the Duffy-positives , heterozygote FyA/B had a significantly higher PvDBP copy number than homozygote FyB/B . The PvDBP copy number of Duffy-null homozygote FyBES/BES and Duffy-positive homozygote FyA/A was not compared with other Duffy genotypes because their sample size was too small to generate significant comparisons ( Fig 4 ) . The two asymptomatic samples detected with two PvDBP copies were both Duffy-null heterozygote ( FyB/BES ) . The remaining samples with a single PvDBP copy comprised both Duffy-null heterozygote and Duffy-positive homozygote and heterozygote ( S5 Table ) . To investigate whether high PvDBP copy number increases invasion efficiency of the parasite , the association of parasite densities with PvDBP copy number was examined . No significant association was detected between PvDBP copy number and parasite densities in the symptomatic samples ( S2A Fig ) . We did not examine this correlation in the asymptomatic samples because of the small sample size . Also , we do not have the parasitemia and demographic information of the asymptomatic samples . For the symptomatic samples , no significant association was detected between PvDBP copy number and age , gender , and ethnicity ( S2B–S2D Fig; S6 Table ) . The PCA plot based on the first three principle components reflected 99 . 5% of the total variation from 12 recorded malaria symptoms ( Fig 5; S1 Table ) . No clear cluster was observed among samples with a single or multiple PvDBP copies , suggesting that the symptoms of these patients did not relate to PvDBP copy number .
PvDBP copy number variation has previously been studied primarily using PCR genotyping . In this study , we used a quantitative PCR method for estimating copy number , an approach used in several studies particularly related to human diseases [35] . The qPCR assay outcome was validated against PCR genotyping and whole genome sequencing methods . We are aware that difference in the amount of parasite DNA among samples , particularly that between symptomatic and asymptomatic infections , may influence the quantification of PvDBP copy number . Therefore , we standardized the amount of parasite DNA in each reaction based on the results of 18S rRNA quantification prior to PvDBP qPCR assay . Also , for each sample we used the single-copy P . vivax aldolase gene as internal standard to calibrate and calculate the PvDBP gene copy number . Given that the results of PvDBP qPCR were consistent with those by PCR genotyping and whole genome sequencing , we are confident that the estimated PvDBP copy number was not biased by the amount of parasite DNA in our samples . Quantitative PCR offers a time- and cost-effective approach to analyze a large number of samples , as unlike whole genome sequencing , it can be performed with relatively little DNA , such as dried blood samples routinely taken in malaria studies . The genotyping PCR method used in previous studies of PvDBP copy number has the advantage of providing better detection of PvDBP duplications in polyclonal infections [23 , 24] . However , this method is limited to identifying only presence or absence of PvDBP duplications rather than copy number variants as by qPCR method [36] . PvDBP duplications detected with specific PCR primers correlated strongly with qPCR estimation of copy number . Samples that showed no amplifications by primers BF/AR ( Malagasy-type ) [23] or BF/AR2 ( Cambodian-type ) [24] were all estimated to contain a single copy of PvDBP by qPCR . A previous small scale test of only 25 Ethiopian P . vivax samples suggested that multi-copy PvDBP infections are more common in Ethiopia than in other parts of the world [24 , 26]; increasing to 178 samples in this study confirms this finding . Genotyping 178 samples in total revealed that 65% of the Ethiopian isolates contained multiple PvDBP copies , which is considerably more than any other P . vivax location , as indicated previously in a worldwide study that included only very few samples from Ethiopia [24 , 26]—this study provides more statistical robustness to that finding . It is formally possible that qPCR estimates of copy number could be complicated by the presence of mixed infections , where qPCR would generate an estimate that averages all clones present . However , in a previous study , P . vivax infections in Ethiopia including Jimma were found to have a relatively low polyclonality rate ( 4 . 3% ) based on microsatellite typing [9] . The twenty samples in this study were confirmed as monoclonal by microsatellites prior to whole genome sequencing . In P . vivax-endemic area like Papua , Indonesia , asymptomatic samples have been showed with higher multiplicity of infection and percentage of polyclonality than symptomatic samples based on microsatellite markers [37] . Although P . vivax is not highly endemic in our study sites in Jimma , Ethiopia , we cannot rule out the possibility that such differences may occur between our symptomatic and asymptomatic samples and affect the comparison of PvDBP amplifications . Cambodian-type duplications were five-times more common than the Malagasy-type . Although primers BF/AR2 were used to diagnose the presence of the Cambodian-type PvDBP duplications , this primer pair can also produce a 1 , 584-bp amplicon should there be Malagasy-type duplications [24] . Among the 14 samples that were diagnosed with the Malagasy-type duplications based on BF/AR , only two were detected with a 1 , 584-bp amplicon by BF/AR2 , likely due to limited quality of DNA extracted from dried blood on filter papers . The preponderance of the Cambodian-type duplications among our samples corroborated the finding by Auburn et al . [26] that indicated a 63% prevalence of the Cambodian-type duplications among 24 P . vivax isolates in Ethiopia . This finding suggested that PvDBP duplications observed in the Ethiopian isolates may have arisen independently through local selection and adaptation , or derived from Southeast Asia after P . vivax was re-introduced back to Africa [38 , 39] . It is possible that P . vivax with expanded PvDBP acquired better fitness and spread through Africa where the majority of populations are Duffy-negative [3] . Though less common , the presence of the Malagasy-type duplications in Ethiopian P . vivax may reflect a contemporary gene exchange between populations via human movement . More than four copies were detected in some of the Ethiopian isolates , higher than that reported in Madagascar ( up to two copies ) [23] and Southeast Asia ( two and three copies ) [24 , 25] . Our findings were consistent with earlier studies that reported high-order PvDBP copies among a smaller number of P . vivax samples in Ethiopia [22 , 26] . Gene duplication can generate new gene functions or alter gene expression patterns [40] . For examples , in P . knowlesi duplication of PkDBP-alpha and deletion of PkDBP-gamma allow the parasite to invade human erythrocytes that lack surface Neu5Gc , a form of sialic acid P . knowlesi requires for binding [41] . In P . falciparum , duplications of the Pfmdr1 gene resulted in increased resistance to antimalarial drug mefloquine [42–45] . While we did not detect any formal association between PvDBP copy number and Duffy negativity , it is worth noting that in other parts of the world such as Cambodia , India , and Brazil where only a small proportion of Duffy-negative individuals live , PvDBP expansion was observed with much lower frequency [24] , with the exception of such in Thailand where a moderate rate ( 30% ) of PvDBP duplications was observed [26] . Apart from PvDBP , another gene PvEBP , which harbors all the hallmarks of a Plasmodium red blood cell invasion protein , including conserved Duffy-binding like and C-terminal cysteine-rich domains [46] , has been recently shown to be variable in copy number among the Malagasy P . vivax [47] . Functional analyses indicated that region II of this gene bound to both Duffy-positive and Duffy-negative reticulocytes , although at a lower frequency compared to PvDBP , suggestive of its role in erythrocyte invasion [48] . While we detected only a single copy of PvEBP in the 20 Ethiopian P . vivax samples based on whole genome sequences , further investigation is needed on a larger sample . The functional significance of PvDBP expansion merits further investigations through comparison of gene expression patterns and in-vitro binding assay of varying PvDBP dosage , and study of P . vivax isolates from Duffy negative individuals is clearly a high priority . While we observed no association with Duffy negativity , with the clear caveat that Duffy negative sample numbers were limited , we did observe a statistically significant higher number of P . vivax isolates with multiple PvDBP copies in FyA/FyBES compared to FyB/FyBES and FyB/FyB , as well as FyA/B compared to FyB/FyBES and FyB/B . Given that PvDBP has a higher binding affinity with erythrocytes expressing the FyB than the FyA antigens [49] , it is possible that P . vivax expanded the PvDBP gene into multiple copies in order to increase binding affinity to erythrocytes that express FyA antigen . The observation of a low number of P . vivax infections in the FyA/A patients in this study lends support to the association between FyA antigen and a reduced risk of clinical vivax malaria in humans [49 , 50] . While Duffy-negative heterozygotes in Papua New Guinea and the Brazilian Amazon region were previously shown with reduced erythrocyte susceptibility to P . vivax infection [51 , 52] and a higher level of anti-PvDBP antibodies compared to Duffy-positives [53] , the relative high proportion of Duffy-null heterozygotes ( FyA/FyBES and FyB/FyBES ) among our P . vivax-infected patients suggested that some lineages of P . vivax ( such as those in Ethiopia ) may have evolved with higher binding or invasion efficiency to erythrocytes with reduced Duffy antigen expression , perhaps through PvDBP expansion . In addition , we also observed a statistically significant higher number of P . vivax isolates with multiple PvDBP copies in symptomatic infections compared to asymptomatic infections . It is therefore possible that increased expression of PvDBP as a result of gene expansion may play a role in evading the immune response developed by the infected individuals , leading to symptoms . Further study with higher numbers of asymptomatic samples is needed to draw a definite conclusion . Moreover , symptoms and parasitemia of the samples were measured only at the initial stage of the infection rather than at various follow-up time intervals . Thus , it is yet unclear if high PvDBP copies of the parasites will increase the duration and/or severity of symptoms in the infected individuals over time . Given that age , gender and ethnicity did not correlate with PvDBP copy number , parasites with single or higher PvDBP copies could all cause infection equally within the host population . To conclude , an exceptionally high prevalence of PvDBP expansion was observed among Ethiopian P . vivax isolates , of which the majority were Cambodian-type duplications , and higher order PvDBP copy number variants were identified by both qPCR and whole genome sequencing . Duffy-negative heterozygotes did not show a significantly higher PvDBP copy number than the Duffy-positives , but symptomatic infections had a significantly higher copy number than the asymptomatic ones . For the symptomatic samples , PvDBP copy number was not significantly associated with parasite density , age , gender and ethnicity . The functional significance of common PvDBP expansion and the presence of high copy number variants among the Ethiopian P . vivax are unclear . Our ongoing investigations focus on PvDBP copy number variation in expanded Duffy-negative homozygote individuals . PvDBP copy number in homozygous Duffy-negative infections and its correlation with symptoms are yet to be explored . | Plasmodium vivax invasion of human erythrocytes relies on interaction between the Duffy antigen and P . vivax Duffy Binding Protein ( PvDBP ) . Whole genome sequences from P . vivax field isolates in Madagascar identified a duplication of the PvDBP gene and PvDBP duplication has also been detected in non-African P . vivax-endemic countries . Two types of PvDBP duplications have been reported , termed Cambodian and Malagasy-type duplications . Our study used a combination of PCR-based diagnostic method , a novel quantitative real-time PCR assay , and whole genome sequencing to determine the prevalence and type of PvDBP duplications , as well as PvDBP copy number on a broad number of P . vivax samples in Ethiopia . We found that over 65% of P . vivax isolated from the symptomatic infections were detected with PvDBP duplications and PvDBP varied from 1 to >4 copies . The majority of PvDBP duplications belongs to the Cambodian-type while the Malagasy-type duplications was also detected . For the asymptomatic infections , despite a small sample size , the majority of P . vivax were detected with a single-copy based on both PCR and qPCR assays . There was no significant difference in PvDBP copy number between Duffy-null heterozygote and Duffy-positive homozygote/heterozygote . Further investigation is needed with expanded Duffy-null homozygotes to examine the functional significance of PvDBP expansion . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"blood",
"cells",
"sequencing",
"techniques",
"parasite",
"groups",
"medicine",
"and",
"health",
"sciences",
"plasmodium",
"tropical",
"diseases",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"parasitology",
"organisms",
"genome",
"sequencing",
"apicomplexa",
"plas... | 2019 | Frequent expansion of Plasmodium vivax Duffy Binding Protein in Ethiopia and its epidemiological significance |
The ubiquitous second messenger c-di-GMP promotes bacterial biofilm formation by playing diverse roles in the underlying regulatory networks . This is reflected in the multiplicity of diguanylate cyclases ( DGC ) and phosphodiesterases ( PDE ) that synthesize and degrade c-di-GMP , respectively , in most bacterial species . One of the 12 DGCs of Escherichia coli , DgcE , serves as the top-level trigger for extracellular matrix production during macrocolony biofilm formation . Its multi-domain architecture–a N-terminal membrane-inserted MASE1 domain followed by three PAS , a GGDEF and a degenerate EAL domain–suggested complex signal integration and transmission through DgcE . Genetic dissection of DgcE revealed activating roles for the MASE1 domain and the dimerization-proficient PAS3 region , whereas the inhibitory EALdeg domain counteracts the formation of DgcE oligomers . The MASE1 domain is directly targeted by the GTPase RdcA ( YjdA ) , a dimer or oligomer that together with its partner protein RdcB ( YjcZ ) activates DgcE , probably by aligning and promoting dimerization of the PAS3 and GGDEF domains . This activation and RdcA/DgcE interaction depend on GTP hydrolysis by RdcA , suggesting GTP as an inhibitor and the pronounced decrease of the cellular GTP pool during entry into stationary phase , which correlates with DgcE-dependent activation of matrix production , as a possible input signal sensed by RdcA . Furthermore , DgcE exhibits rapid , continuous and processive proteolytic turnover that also depends on the relatively disordered transmembrane MASE1 domain . Overall , our study reveals a novel GTP/c-di-GMP-connecting signaling pathway through the multi-domain DGC DgcE with a dual role for the previously uncharacterized MASE1 signaling domain .
The discovery and functional analysis of bacterial cyclic dinucleotides , in particular bis- ( 3´ , 5´ ) -cyclic diguanosine monophosphate ( c-di-GMP ) , has opened a new era in the study of second messenger signaling in prokaryotes , revealing a complexity that matches that in eukaryotes [1–4] . C-di-GMP is synthesized by diguanylate cyclases ( DGCs , with this activity provided by their GGDEF domains ) and degraded by specific phosphodiesterases ( PDEs , with either EAL or HD-GYP domains ) [5 , 6] . These signaling enzymes usually exhibit a modular domain architecture . Diverse N-terminal domains , which in many cases also anchor the enzymes in the cytoplasmic membrane , act as signal input devices responding to often still unknown signals . These DGCs and PDEs antagonistically negotiate levels of c-di-GMP which is bound by a large diversity of effector components . These can be different types of proteins that control various targets by direct interaction or riboswitches in the upstream regions of mRNAs that affect transcriptional elongation or translation [7 , 8] . These effector-target systems can simply respond to global changes in the cellular pool of c-di-GMP . However , also local c-di-GMP signaling has been observed , especially in bacterial species that have multiple DGCs and PDEs , some of which can exert surprisingly specific functions based on their direct association with effector/target systems [9–14] . c-di-GMP-controlled targets are involved in fundamentally important cellular and physiological functions , such as the formation of highly antibiotic-tolerant biofilms , motility , virulence , cell cycle progression and development [3 , 15–19] . A major c-di-GMP target in Escherichia coli–an enteric bacterium existing as a commensal as well as in various types of pathogens–is the production of extracellular matrix polymers during biofilm formation [20] . In macrocolony or pellicle biofilms of E . coli , these matrix components are amyloid fibres termed curli [21] as well as phosphoethanolamin-modified cellulose ( pEtN-cellulose ) [22] . Their production occurs in slowly growing cells entering into stationary phase and is under the control of a transcription factor cascade consisting of the stationary phase sigma factor σS ( RpoS ) , the MerR-like regulator MlrA and the transcription factor CsgD , with the latter directly driving the expression of the curli subunits CsgA and CsgB and indirectly activating cellulose production ( summarized in [20] ) . Positive input by c-di-GMP is required at two positions in this hierarchical control network ( see also Fig 1 below ) : ( i ) in a circuit involving DgcE , PdeR and DgcM , the target is the activity of MlrA , with the mutually interacting proteins PdeR , DgcM and MlrA representing a case of local c-di-GMP signaling [10 , 12] , and ( ii ) further downstream , c-di-GMP is required for directly activating cellulose synthase ( BcsAB ) [23] as well as the co-localized cellulose modifying system ( BcsEFG ) [22 , 24 , 25] . This seems yet another process of local signaling as it requires specifically DgcC , which in turn depends on CsgD for being expressed [26] . A major question mark in this regulatory network is associated with the role of the top level diguanylate cyclase DgcE , which provides for the key trigger that activates the entire cascade thereby leading to CsgD expression and biofilm matrix production . What are the environmental and/or cellular signals that DgcE responds to and how does it do so at the molecular level ? With its six-domain architecture ( Fig 1A ) , DgcE is the most complex among the twelve DGCs of E . coli K-12 [27 , 28] . Its N-terminal part consists of a MASE1 domain , a putative sensory domain originally described to have eight transmembrane ( TM ) segments that also occurs at the N-termini of PDEs and histidine sensor kinases and is found in gamma- , beta- and alpha-proteobacteria as well as in cyanobacteria [29] . The complete hydrophobic N-terminal region in MASE1 domain proteins typically has a total length of about 300 amino acids and–based on hydrophobicity patterns and the distribution of charge ( S1 Fig ) –contains ten rather than eight transmembrane segments , with this entire region now being annotated as the MASE1 domain . In DgcE this domain is followed by three cytoplasmic PAS/PAC domains , the GGDEF domain and an EAL domain that is degenerate , i . e . it lacks all the key amino acid residues required for c-di-GMP binding and catalysis of the PDE reaction [28] . What is the role of all these domains in signal integration and transduction into and through DgcE ? Can an analysis of DgcE lead us to an understanding of the sensory function of the widespread membrane-integral MASE1 domain ? Are any other factors involved in signal input into DgcE ? To elucidate the molecular functions of DgcE , we focussed on a detailed genetic analysis since DgcE is not only a complex membrane-associated protein , but in the course of our experiments we also discovered a constitutive in-vivo turnover of DgcE , i . e . DgcE has properties that efficiently prevent purification and in-vitro analysis . This genetic approach allowed us to assign specific roles to the different domains of DgcE and to demonstrate signal input by a GTPase system that directly interacts with the MASE1 domain of DgcE and thereby triggers downstream signaling .
In order to elucidate the function of particular sites and domains in DgcE , we started by introducing specific point and deletion mutations . We chose to do so in the single chromosomal copy of the dgcE gene , as DgcE controls a local c-di-GMP signaling pathway , i . e . PdeR-DgcM-MlrA/csgD , which involves multiple protein-protein interactions [10 , 13] for which stoichiometry of proteins matters . As an experimental readout , we used ( i ) macrocolony biofilm morphology , which provides for a semi-quantitative assay for curli and pEtN-cellulose production ( as further explained in S2 Fig ) , ( ii ) direct visualization of cellular CsgD in liquid cultures by immunoblotting , and ( iii ) a single copy csgB::lacZ reporter fusion to monitor the activity of the csgBAC operon , which encodes the curli subunits . Two amino acid exchanges ( D763A/E764A ) in the active site motif of DgcE ( GGDEF to GGAAF ) resulted in the same strong reduction in macrocolony wrinkling as the complete deletion of dgcE ( Fig 1C ) , i . e . a phenotype corresponding to a strong reduction in both curli fibres and pEtN-cellulose [13] . This is also reflected in reduced CsgD levels ( Fig 1D ) and csgB::lacZ expression ( Fig 1E ) . Thus , the input of DgcE into the PdeR-DgcM-MlrA/csgD pathway relies on its ability to generate c-di-GMP . Although intuitively taken for granted , this has to be clarified , since e . g . the ability of DgcM to support MlrA activity does not require its c-di-GMP synthesis but relies on its direct interaction with MlrA [10] . Internal deletions of the entire transmembrane MASE1 domain or of all three PAS domains ( ΔPAS3 ) generated the same null phenotype , i . e . these regions of DgcE play an essential positive role in its signaling into CsgD expression and extracellular matrix production . In order to control for protein expression levels , all the internally deleted variants of DgcE were also generated as Flag-tagged variants ( with the tag inserted at the 3´ end of the chromosomal dgcE alleles ) . The presence or absence of the Flag-tag did not affect the macrocolony phenotypes and the Flag-tagged constructs showed that wildtype DgcE and internally deleted DgcE variants were expressed at levels comparable to full size DgcE ( see below ) . In contrast to eliminating the MASE1 and PAS3 regions , deleting the C-terminal EALdeg domain resulted in the opposite phenotype , i . e . increased matrix production , which became apparent as even larger , flatter and stiffer macrocolonies that fold into fewer , but higher ridges ( Fig 1C , compare also to S2 Fig ) . A corresponding increase in CsgD levels and csgB::lacZ was also observed ( Fig 1D and 1E ) . Since the latter assay was done in liquid culture using non-cellulose-producing , but otherwise identical strains ( to avoid cell clustering since matrix components are also produced in stationary phase liquid cultures ) , we also tested the inactivating or activating effects of the various internal deletions in DgcE on macrocolonies of these curli-only producing strains and found them to be the same as in the strains that synthesize both matrix components ( S3 Fig ) . The macrocolony phenotype of the dgcEΔEAL strain ( Fig 1C ) is in fact similar to that of a matrix-overproducing pdeR deletion strain ( S2 Fig ) , i . e . the EALdeg domain plays an inhibitory role . This full activation of DgcEΔEAL could be eliminated by also deleting the MASE1 domain , indicating that the MASE1 domain does not antagonize the inhibitory EALdeg domain , but that its essential activating role is independent of the EALdeg domain . In a series of epistasis experiments , we combined all these mutations in dgcE with full deletion mutations in pdeR and in pdeH . The macrocolony morphotypes of the double mutants ( Fig 1C ) confirmed that DgcE acts upstream of PdeR ( mutations in dgcE had no phenotypic effects in a pdeR mutant background ) , but at the same level as PdeH ( all double mutants carrying both the pdeH and inactivating deletion mutations in dgcE showed an intermediate phenotype , i . e . less wrinkling than pdeH alone , but more wrinkling than the corresponding dgcE mutations alone; Fig 1C ) . PdeH is the master PDE that maintains very low global c-di-GMP levels in E . coli , even under conditions where DgcE is active as a DGC and drives biofilm matrix production [13] . c-di-GMP synthesis depends on the dimerization of DGCs , with each monomer binding one of the two GTP substrate molecules , which allows to generate two phosphodiester bonds in a symmetric manner . DGCs are thus thought to be activated by dimerization promoted by their sensory input or other domains [6] . The isolated GGDEF domains of ten of the twelve DGCs of E . coli K-12 –including that of DgcE–indeed do not dimerize on their own [13] . This prompted us to test which of the other domains of DgcE can dimerize and may thus promote the dimerization of the GGDEF domain and thereby activation of DgcE . To do so , we used a bacterial two-hybrid system in which an interaction between bait and target proteins reconstitutes a functional two-domain adenylate cyclase ( Cya ) , which allows a Δcya mutant of E . coli to utilize maltose as a carbon source [30 , 31] . Testing the isolated domains of DgcE in this system confirmed the inability of the GGDEF domain to dimerize on its own [13] , whereas the PAS3 region generated a clear dimerization signal ( Fig 2A ) . An apparent dimerization of the isolated EALdeg domain was experimentally not conclusive , since the control assays showed a similar signal . We therefore used a second two-hybrid technology , the Bacterio-Match II system , which is based on transcription initiation of a reporter gene enabled by interacting proteins or domains ( which have to be cytoplasmic ) fused to lambda cI and α-NTD of RNA polymerase [32] . Here , the entire cytoplasmic region of DgcE ( i . e . PAS3-GGDEF-EALdeg ) as well as the isolated PAS3 and EALdeg domains , but neither the isolated GGDEF domain nor the control combinations yielded a dimerization signal ( Fig 2C ) . With the MASE1 domain alone , the Cya-based two-hybrid assay could not be performed due to overproduction toxicity , i . e . the combination of the plasmid-encoded MASE1 domain fused to the two adenylate cyclase domains affected growth . In a complementary 'subtractive' approach , we tested dimerization of DgcE variants lacking distinct domains ( Fig 2B ) . The relatively weak dimerization of full-size DgcE was not significantly reduced by eliminating single domains , consistent with more than one domain being able to promote dimerization ( as shown above for the isolated domains ) . Deleting the MASE1 domain even improved dimerization , possibly because the resulting soluble DgcEΔMASE1 is freely diffusible in the cytosol , which may facilitate dimerization in comparison to the sterically more constrained conditions for the membrane-attached full size DgcE . Overall , these two-hybrid interaction data show that both the activating PAS3 as well as the inhibitory EALdeg domain of DgcE have a potential for dimerization . The observation of different in-vivo activities of the various DgcE variants ( Fig 1 ) required to experimentally control for the actual expression of these variants . Using both FLAG-tagged chromosomal variants ( as mentioned already above ) as well as plasmid-encoded variants carrying a 6His tag , immunoblotting showed overall similar levels of expression . In addition , however , these experiments revealed clear degradation patterns for all variants except for the soluble cytoplasmic DgcEΔMASE1 variant ( Fig 3B; for wildtype DgcE , this degradation was already previously noticed by Sarenko et al . ( 2017 ) ) . As the N-terminal transmembrane region of DgcE is required for proteolysis and the tags for visualization are C-terminally located , the ladder-like pattern indicates processive proteolysis from the N- to the C-terminus of DgcE , with the protease pausing or slowing down between domain boundaries as suggested by the sizes of the fragments generated . When we replaced the MASE1 region by the first two transmembrane segments of the lactose carrier LacY ( TM1+2LacY ) , the presence of this new membrane-insertion region also allowed degradation ( Fig 3B ) . This suggests that DgcE is attacked by a membrane-associated protease , with the transmembrane MASE1 domain providing for appropriate localization of DgcE rather than for sequence-specific recognition for proteolysis . Proteolysis seems constitutive as it occurred in growing as well as in stationary phase cells ( Fig 3B and also Fig 4F below ) . Since the MASE1 domain is required for activity of DgcE ( Fig 1 ) , but it is also the first region to be eliminated by this processive proteolysis , this turnover represents an inactivating mechanism . Initiation of DgcE proteolysis seems highly efficient since we could not detect chromosomally expressed full-size DgcE protein , i . e . the largest visible band corresponded in size to that of the DgcEΔMASE1 variant ( Fig 3A and 3B ) . Upon overproduction , however , larger bands were detected , in particular an oligomeric form of DgcE , which was not monomerized by SDS treatment during sample preparation and was also larger than expected for a dimer . Interestingly , when DgcE lacked its EAL domain , these oligomers could be observed even when expressed from the chromosomal gene ( in stationary phase cells; Fig 3B ) and , when plasmid-expressed , they accumulated to higher levels than for wildtype DgcE ( Fig 3C ) . This suggests that the inhibitory function of the EAL domain may involve a destabilization of this oligomeric complex and that this oligomer represents the active form of DgcE . Interestingly , plasmid-encoded DgcE was found to affect in trans the chromosomally encoded FLAG-tagged DgcE . In this configuration , the latter accumulated partially as the oligomeric DgcE complex ( Fig 3D ) , i . e . it became better protected against proteolytic attack which most likely reflected a titration of the protease involved . All this raised the question of the identity of the protease and we therefore tested DgcE turnover in a series of protease knockout mutants ( S4 Fig ) . Processive degradation suggested a general ATP-dependent protease , which either could be itself membrane-associated ( such as FtsH ) or could be cytoplasmic and may proceed with proteolysis after a membrane-bound protease has generated an initial cut in DgcE . None of the mutations affecting the cytoplasmic ATP-dependent AAA+ proteases ClpXP , ClpAP , Lon or HslUV [33] altered the DgcE degradation pattern . The same was observed for the Lon homolog YcbZ , which lacks ATPase activity . Testing a putative involvement of FtsH was less straightforward since FtsH is essential and its degradation of certain substrates is modulated by the two accessory factors HflK and HflC [34] . However , a ftsH deletion is available that is viable due to the presence of a suppressor mutation [35] . When plasmid-encoded DgcE was expressed in strains carrying the ftsH deletion and the suppressor or the suppressor alone , it was still degraded ( S5 Fig ) and also knocking out hflK and hflC in our standard background did not show any effect ( S4 Fig ) . Furthermore , knocking out HtpX , a membrane-associated protease that is related to FtsH but does not hydrolyze ATP [34] , did not alter DgcE proteolysis . Also the rhomboid protease GlpG , which cleaves its substrates within the membrane [34] , showed no evidence of being involved in DgcE degradation ( S4 Fig ) . Taken together , these data suggest that DgcE may be degraded either by a very specific still unknown membrane-associated protease or , alternatively , several proteases may be redundantly involved such that knocking out individual proteases alone does not produce a significant effect . Overall , DgcE is subject to efficient and continuous , i . e . apparently not further regulated , proteolytic turnover by a still unknown protease ( s ) , which proceeds from the membrane-inserted N-terminal MASE1 domain to the C-terminal EALdeg domain of DgcE . This also means that cellular DgcE activity depends on continuous de-novo synthesis , i . e . DgcE exhibits a highly dynamic regulation which allows for rapid shut down by proteolysis when expression of DgcE ceases . The N-terminal MASE1 domain as well as the PAS3 region of DgcE are required for DgcE activity and thus may also perceive positive input signals . A mutation in dgcE has long been known to partially suppress the motility defect of a pdeH mutant [36 , 37] . Along with the suppressor in dgcE ( termed yegE at the time ) , also several other suppressor mutations were described in the study by Girgis et al . ( 2007 ) . Two of these exactly phenocopied the dgcE suppressor mutation and were mapped to yjdA and yjcZ , which are located in an operon controlled by the flagellar FlhDC/σFliA transcriptional cascade [36] . This raised the possibility that YjdA and YjcZ might operate in the DgcE pathway , possibly by activating DgcE . Since this turned out to be the case ( see below ) , we propose to rename the two genes as rdcA and rdcB , respectively ( regulators of a diguanylate cyclase; see also Discussion ) . Deleting rdcA and/or rdcB phenocopied a dgcE knockout also with respect to macrocolony biofilm formation , with the effects being non-additive , no matter whether combined with deletions of the entire dgcE gene ( Fig 4A ) or with internal deletions in dgcE affecting the activating domains only ( S6 Fig ) . Epistasis experiments in double mutants also carrying either pdeR or pdeH mutations placed the activity of RdcA and RdcB upstream of PdeR , but at the same level as PdeH ( Fig 4A ) , just as shown above for DgcE itself ( Fig 1C ) . Also with respect to CsgD levels ( Fig 4B ) , curli expression as assayed with the csgB::lacZ reporter fusion ( Fig 4C ) and cellular c-di-GMP levels ( Fig 4D ) , the rdcA and rdcB mutations showed precisely the same effects as the dgcE mutation . All these data indicated that RdcA and RdcB activate the production of CsgD , curli and pEtN cellulose and that these proteins act at the level of DgcE . One possibility seemed that they counteract the inhibitory role of the EAL domain of DgcE , but this was excluded by showing that the mutations in rdcA and/or rdcB still reduced matrix production in macrocolonies of the dgcEΔEAL strain ( Fig 4E ) . Thus , the activating RdcA/RdcB system and the inhibitory EALdeg domain affect the output of DgcE independently . The presence or absence of the RdcA/RdcB system also had no effect on the proteolytic turnover of DgcE ( S7 Fig ) . Expression of the rdcAB operon–assayed with a FLAG-tagged RdcA expressed from the chromosomal gene–showed expression throughout the growth cycle ( Fig 4F ) , with a minor increase in post-exponential phase , where the expression of flagellar genes is known to transiently increase [38 , 39] . This assures co-expression with DgcE , which is further induced in the post-exponential growth phase ( Fig 4F ) as it is under σS control [13 , 40] . We then assayed DgcE , RdcA and RdcB for potential direct interactions using the Cya-based bacterial two-hybrid system already described above . RdcA was found to strongly interact with DgcE ( Fig 5A ) . In addition , RdcA could dimerize ( or oligomerize ) and tightly interacted with RdcB , while RdcB itself did not show any evidence of dimerizing or interacting with DgcE . Two-hybrid testing of RdcA against the series of DgcE variants lacking particular domains , assigned the interaction of RdcA to the N-terminal transmembrane MASE 1 domain of DgcE ( Fig 5B ) . We conclude that DgcE and RdcA–both with the ability to dimerize–can form a complex , with RdcA also binding to RdcB , indicating that the RdcA/RdcB system tightly cooperates with DgcE . The knockouts of all three genes generate exactly the same downstream effects–including a small but significant reduction in the cellular concentration of c-di-GMP–and RdcA directly targets the N-terminal MASE1 domain of DgcE , which is required for DgcE activity . Taken together , these data indicate that the RdcA/RdcB system activates DgcE to act on the PdeR-DgcM-MlrA pathway which stimulates csgD expression . How can the RdcA/RdcB system act on DgcE and could this process in turn be controlled by some input signal ? While the C-terminal half of RdcA ( a 84 . 4 kDa protein ) as well as the smaller RdcB protein ( 32 . 9 kDa ) do not exhibit any similarities to known protein families , the N-terminal region of RdcA contains a P-loop NTPase region or G domain with similarity to dynamin-like proteins . Dynamins are GTPases [41 , 42] and such an activity has also been observed for RdcA in a previous study [43] . The G domain features Walker A and B motifs ( also termed G1 and G2 motifs , respectively ) involved in GTP binding and hydrolysis , respectively , which is crucial for the diverse functions of dynamin-like proteins [41 , 42] . Similar to eukaryotic dynamin , which is involved in clathrin-coated vesicle formation during endocytosis [44] , and several prokaryotic GTPases of diverse functions [45] , RdcA contains the conserved lysine residue ( K82 ) in its G1 motif ( the entire consensus sequence GxxxxGKS is fully conserved in RdcA ) and the conserved threonine residue ( T103 ) in its G2 motif . In the dynamin crystal structure the corresponding residues are K44 , which is crucial for GTP binding , and T65 , which was shown to be part of the active site and to be directly involved in catalysis [46] . Several amino acid replacements were isolated for T65 , with the T65D variant of dynamin still binding GTP with the same affinity as the wildtype dynamin , whereas its GTPase activity was >100fold reduced [47] . In order to test whether GTP binding and GTPase activity are involved in RdcA´s function in the DgcE-controlled pathway , we generated the corresponding amino acid exchanges in RdcA , i . e . K82A and T103D . Plasmids carrying the entire rdcA-rdcB operon with these mutations being present in rdcA were then tested for complementation of a chromosomal deletion of the rdcA-rdcB operon ( Fig 6A ) . The rdcAK82A allele was found to complement the rdcA deletion just as well as the wildtype rdcA allele , i . e . RdcAK82A was able to activate DgcE . By contrast , the rdcAT103D allele was found to be unable to complement . This suggested that GTP per se might not be required for RdcA to activate DgcE , but might rather be inhibitory , with stable GTP binding by RdcAT103D –due to its inability to cleave GTP–'freezing' RdcA in an inactive state unable to activate DgcE . In order to exclude the possibility that these phenotypes were due to plasmid-mediated overexpression of RdcAK82A and/or RdcAT103D , we also generated these mutations directly in the chromosomal rdcA gene . Also in this single copy configuration , the Walker A mutant version RdcAK82A was fully active , i . e . macrocolony morphology and thus matrix production of the mutant is identical to that of the parental strain ( Fig 6B ) , indicating that GTP binding is dispensible for RdcA-mediated activation of DgcE . By contrast , the chromosomal rdcAT103D mutant was as defective as the rdcA full deletion mutant in terms of macrocolony morphology ( Fig 6B ) , the cellular CsgD level ( Fig 6C ) and curli expression as tested by the csgB::lacZ reporter fusion ( Fig 6D ) . In order to exclude that these phenotypes might be caused by a lack of expression , we also Flag-tagged both the wildtype rdcA and the rdcAT103D alleles in the chromosome . The RdcAT103D variant was indeed expressed just like the wildtype RdcA protein ( S8A Fig ) . Furthermore , the two Flag-tagged strains produced the same macrocolony phenotypes ( S8B Fig ) as the corresponding non-Flag-tagged strains ( compare to Fig 6B ) . Thus , GTPase activity is essential for RdcA to exert its activating function in DgcE-mediated signaling . This raised the question whether the GTPase-defective RdcAT103D variant could be altered in its direct interaction with DgcE . As shown by two-hybrid analyses , interaction patterns of RdcA were indeed strongly affected by the T103D mutation . Not only was its ability to dimerize ( or oligomerize ) reduced , but its interaction with DgcE was completely abolished ( Fig 6E ) . By contrast , the T103D mutation in RdcA had no effect at all on the ability of RdcA to interact with RdcB , which also confirmed that the T103D mutation does not alter the overall structure of RdcA . In conclusion , these data support the concept that GTP hydrolysis by RdcA is crucial to allow its direct interaction with DgcE and therefore for its ability to activate the DgcE-triggered downstream pathway that stimulates the production of CsgD and therefore the synthesis of the biofilm matrix components curli and pEtN cellulose .
With its MASE1-PAS3-GGDEF-EAL domain architecture , DgcE is a prototype of c-di-GMP signaling enzymes in more than one respects . On the one hand , the function of the membrane-intrinsic putative sensory MASE1 domain , which also occurs e . g . in two-component histidine kinases [29] has remained unknown; on the other , DgcE is a representative of a large subgroup of c-di-GMP signaling enzymes with a PAS-GGDEF-EAL domain architecture . The latter usually have either GGDEF domain-mediated DGC activity or EAL domain-dependent PDE activity , with the respective other domain being degenerate and in some cases exerting a regulatory function . Already the very first DGCs and PDEs identified in Komatagaeibacter xylinus ( formerly Acetobacter xylinum ) belonged to this group of PAS-GGDEF-EAL proteins [48] . Since then numerous such enzymes have been found in various bacteria , e . g . the PDEs RbdA and RmcA in Pseudomonas aeruginosa [49 , 50] , PdeB in Shewanella oneidensis [51] and the already mentioned trigger PDE PdeR in E . coli [10 , 12] . DgcE clearly is a DGC , since its GGDEF motif ( the active site in DGCs ) is required for its output function ( Fig 1C ) and essentially all specific residues that in EAL domains are involved in c-di-GMP and/or metal binding as well as in catalysis [52 , 53] are not conserved in the degenerate EAL domain ( EALdeg ) of DgcE , which indicated a regulatory role for this domain . In contrast to many soluble cytoplasmic PAS-GGDEF-EAL domain proteins , DgcE is membrane-attached , suggesting a complex signal transduction that involves both the additional N-terminal transmembrane region , i . e . the MASE1 domain , and the PAS3 region . The top position of DgcE in the hierarchical network ( Fig 1B ) that controls the production of biofilm matrix components , i . e . amyloid curli fibres and pEtN cellulose , also pointed to a role as a key trigger of this network [10 , 37] . The well-studied output , i . e . matrix production , provided us with clear phenotypes and other experimental readouts for studying this function of DgcE . However , we also discovered an efficient and continuous turnover of DgcE , which does not seem to involve a specific proteolytic recognition site in DgcE nor a single specific protease that could be knocked out . As a consequence , this efficient degradation of DgcE could not be experimentally eliminated and therefore purification and biochemical analyses of full-size DgcE were unfeasible , i . e . our study had to built on genetics and in-vivo analyses . PAS domains ( named after the prototypical proteins Per , Arnt and Sim ) are ubiquitous in all kingdoms of life and , despite high sequence divergence , exhibit similar structure [54 , 55] . Their ability to bind diverse small molecules and cofactors makes them prominent sensor domains with a potential to react for instance to oxygen , redox changes or light . In bacteria they are found in many signaling proteins such as histidine kinases , chemotaxis receptor proteins and c-di-GMP signaling enzymes . PAS domains often act as mediators of the dimerization of their downstream domains , which explains the prominence of the PAS-GGDEF arrangement , as DGC activation requires dimerization of the GGDEF domain [6] . Our data indicate that this is also the role of the PAS domains in DgcE , since the PAS3 region is both required for DgcE activity ( Fig 1 ) and can dimerize or oligomerize ( Fig 2 ) . The finding that also the RdcA/RdcB system as well as the N-terminal transmembrane MASE1 domain of DgcE–which interacts with RdcA–are essential for activation of DgcE ( as discussed below ) , suggests that the PAS3 region is important for signal transmission rather than for initial signal perception . Nevertheless , additional and possibly differential signal input via the three PAS domains , perhaps by sensing oxygen or cellular redox signals , is not excluded and will have to be clarified in future structural studies . Deleting the C-terminal EALdeg domain of DgcE stimulated the output activity of the DgcE pathway ( Fig 1 ) , i . e . this domain performs an inhibitory function . Since it is degenerate in the crucial amino acid positions involved in c-di-GMP binding and catalysis [52 , 53] , this inhibition cannot be due to degradation of the c-di-GMP generated by the adjacent GGDEF domain of DgcE . EAL domains in general can form dimers [3] and this is so for the EALdeg domain of DgcE as well ( Fig 2 ) . Importantly , we observed higher order complexes of the DgcE variant that lacks the EALdeg domain ( Fig 3 ) . These complexes were clearly larger than dimers and may be tetramers , which are quite common among DGCs [6] . Since DgcEΔEAL is also highly active , these oligomers may represent the active form of DgcE , i . e . include the productive GGDEF domain dimers . In addition , despite identical transcriptional and translational control , DgcEΔEAL was always present at higher levels than wildtype or other variants of DgcE , indicating a lower rate of degradation . As proteolysis initiates from the N-terminus of DgcE , the C-terminal EALdeg domain is unlikely to promote proteolysis directly . Rather , the EALdeg domain may do so indirectly by interfering with the formation of the larger protein complex , which could afford some protection against proteolytic attack–actually a common mechanism to control proteolysis [56] . Taking together all these evidences , an attractive hypothesis to explain the inhibitory role of the EAL domain in DgcE could be that this domain disturbs the functionally productive oligomerization of DgcE , possibly by dimerizing itself in a sterically interfering manner . By contrast , productive oligomerization would be promoted by the interaction of the MASE1 domain of DgcE with RdcA/RdcB and dimerization of the PAS3 region which stimulates GGDEF domain dimerization . In this scenario , DgcE would not just show simple activation , but it rather seems the balance between antagonistic negative and positive mechanisms–all based on various interactions of its domains–that would determine the actual regulatory outcome . As large N-terminal transmembrane MASE1 domains were found at the N-termini of c-di-GMP signaling enzymes as well as histidine sensor kinases [29] , they obviously are somehow involved in signal input but their actual molecular function had remained enigmatic . Here we show that the MASE1 domain of DgcE plays crucial roles in membrane localization , activity and proteolytic turnover of DgcE . In DgcE , the MASE1 domain is crucial for proteolytic turnover–which proceeds from the N-terminus to the C-terminus of DgcE–since its deletion resulted in stabilization of DgcE ( Fig 3 ) . However , proteolytic turnover was still observed , when the MASE1 domain was replaced by the first two transmembrane segments of the lactose carrier LacY . Thus , DgcE proteolysis requires membrane localization indicating the involvement of a membrane-localized protease ( s ) . Moreover , the MASE1 domain of DgcE does not contain a specific 'degron' , i . e . a proteolytic recognition motif [56 , 57] , but may be recognized due to a partially unfolded structure . The MASE1 domain features a striking number of glycine and proline residues in its transmembrane segments ( S1 Fig ) , i . e . alpha helix-disrupting amino acids that usually are underrepresented in functional transmembrane proteins . This should result in a relatively disordered structure [29] , that is likely to be prone to recognition by quality control proteases . Similarly , the first two transmembrane segments of LacY are probably disordered when deprived of the additional transmembrane segments present in the native structure of the whole carrier protein . In E . coli , MASE1 domains are also present in two c-di-GMP-specific PDEs , PdeA ( YfgA ) and PdeF ( YfgF ) , which have a MASE1-GGDEFdeg-EAL domain architecture . In immunoblots , PdeA generates a proteolytic degradation pattern similar to that of DgcE [13] . PdeF is expressed only under anaerobic conditions [58] , but ectopic expression under aerobic conditions also results in protein bands on immunoblots that are smaller than expected for the full-size protein [59] . Thus , MASE1 domains may generally confer proteolytic turnover to membrane-associated signaling proteins that carry such domains at their N-termini . Taken together , all this points to one or more protein quality control proteases that should be generally conserved , membrane-associated and operating in a processive manner on DgcE . Based on these criteria , the most likely protease candidates seem FtsH and HtpX , two related membrane-bound protein quality control proteases that use HflK , HflC and QmcA as accessory factors [60] . While none of the single gene knockouts in this system stabilized DgcE ( S4 Fig and S5 Fig ) , FtsH and HtpX are known to operate in a substrate-overlapping redundant manner , with HtpX becoming essential in a ftsH mutant background , which does not allow to generate double mutants [35 , 60 , 61] . Instead of recognizing specific amino acid motifs , this system initiates proteolysis upon detecting partially unfolded regions in target membrane proteins [60] . Among the ATP-driven general proteases , FtsH also generates the lowest force in unfolding substrate proteins [62] and HtpX even lacks ATPase activity [60] , which would be consistent with the pausing of processive proteolysis between the stably folded DgcE domains that generates the observed proteolytic 'ladders' ( Fig 3 ) . Initiation of proteolysis at the MASE1 domain , however , seems highly efficient , since the full size protein was hardly detectable on immunoblots , when expressed at the normal rate from the chromosomal copy of the dgcE gene . Overall , our data suggest that two or even more proteases may operate efficiently on DgcE in a redundant manner . This also means that DgcE is subject to a highly dynamic control , which allows for rapid shut-off if required , since ongoing synthesis of DgcE is required for its activity in controlling CsgD and matrix production . Besides its role in proteolysis , the MASE1 domain is also crucial for the function of DgcE in controlling CsgD and biofilm matrix production ( Fig 1 ) as the MASE1 domain provides the site of interaction between DgcE and the activating GTPase RdcA ( Fig 5B; as discussed further below ) . These two functions of the MASE1 domain seem to be independent from each other ( apart from proteolysis limiting DgcE activation by RdcA in time ) , since the presence or absence of RdcA does not affect the turnover of DgcE ( S7 Fig ) , which seems to proceed continuously . Thus , both activation and proteolytic inactivation of DgcE crucially depend on its MASE1 domain . Moreover , this raises the possibility that N-terminal MASE1 domains in other signaling proteins may also not only promote degradation but could also serve as a docking site for additional proteins that provide for signal input . In the case of DgcE , the signaling protein that conditionally docks onto the MASE1 domain , is RdcA , which in turn binds the RdcB protein . RdcA consists of an N-terminal dynamin-like GTPase domain followed by an uncharacterized C-terminal region of about equal size , but in contrast to classical membrane-associated dynamins [44] , RdcA is a soluble cytoplasmic protein . The smaller and soluble RdcB protein shows no similarities to proteins of known function . More than ten years ago , mutations in these genes ( named yjdA and yjcZ at the time ) , which belong to the large FlhDC-controlled flagellar regulon , were found to suppress the motility defect of a pdeH mutant , just as a mutation in dgcE ( yegE ) did [36] . This provided for a rationale for investigating a putative role of these proteins in DgcE-mediated c-di-GMP signaling , which is crucial to turn down motility and induce biofilm functions during entry into stationary phase [37 , 63] . In parallel to our study , a role in submerged biofilms in a microfluidic device and interactions between DgcE , YjdA and YjcZ were also found independently by other researchers [64] . Furthermore , YjdA was also proposed to be involved in chromosome segregation as it was found to interact in vitro with clamp protein ( DnaN ) , i . e . a protein with a key role at the replication fork . Therefore , CrfC ( colocalization at the replication fork with clamp ) was proposed as a new name for YjdA [43] . Unfortunately , however , most of the in-vivo experiments reported in this study designed to show a function in DNA replication were done with either YjdA or fluorescent reporters expressed from plasmids . Notably , its interaction with DnaN was also not affected by the presence or absence of GTP [43] . In addition , knocking out yjdA has no effect on the growth rate , which clearly excludes a vital mechanistic role in the essential processes of DNA replication and segregation . Thus , it seems more likely that YjdA may play some uncharacterized regulatory role in DNA replication and thus possibly represents a 'moonlighting‘ enzyme , i . e . a protein recruited by evolution to serve in two independent functional contexts [65] . By contrast , knocking out YjdA generates a clear DgcE-mediated biofilm-related phenotype , which depends on its GTP binding/GTPase status and also requires YjcZ , i . e . the second gene product of the yjdA-yjcZ operon . Thus , we consider its function in DgcE-dependent c-di-GMP signaling the major function of this system , for which we therefore propose RdcA and RdcB as novel names ( for regulators of a diguanylate cyclase; this also opens the possibility of a systematic nomenclature for additional DGC regulators to be identified in the future ) . As shown here , RdcA and RdcB are both required for DgcE to stimulate–via the PdeR-DgcM-MlrA pathway–the expression of the biofilm regulator CsgD and thus the production of curli fibres and pEtN-cellulose ( Fig 4 ) . RdcA , which is able to dimerize and also binds RdcB , interacts specifically with the N-terminal MASE1 domain of DgcE ( Fig 5 ) , i . e . the putative sensory input domain , which is essential for the output activity of DgcE ( Fig 1 ) . Thus , the RdcA/RdcB system seems to activate DgcE by direct interaction , with the RdcA dimer ( or oligomer ) possibly bringing together two DgcE molecules and thereby promoting the alignment and dimerization of the PAS3 and GGDEF domains of DgcE , which allows the two GGDEF domains to synthesize c-di-GMP ( Fig 7 ) . For this activating function of RdcA , its ability to hydrolyze GTP is essential ( Fig 6 ) . This was shown with the RdcAT103D variant , which lacks the essential catalytic threonine residue in the G2 region that is highly conserved in dynamin-like GTPases . RdcAT103D still interacted with its partner protein RdcB in vivo , but its ability to dimerize ( or oligomerize ) was reduced and its interaction with DgcE was completely abolished ( Fig 6E ) . This could mean that GTP hydrolysis might provide a "powerstroke" for a large conformational change of RdcA that in turn may affect its quaternary structure and thus activation of DgcE in a manner similar to the function of more canonical dynamin-like proteins in driving membrane restructuring events [45 , 66–68] . Alternatively , the GTP-bound form of RdcA might be unable to activate DgcE because GTP may act as an allosteric inhibitor . Our data obtained with the RdcAK82A variant ( Fig 6A and 6B ) are clearly in favor of this second scenario . The K82A mutation eliminates the conserved lysine residue in the Walker A/P-loop motif of dynamin-like protein that is essential for GTP binding [42 , 69] . Our finding that this rdcAK82A allele is fully functional–no matter whether expressed in single copy from the chromosome or from a low copy number plasmid–strongly indicates that the DgcE-activating form of RdcA is the nucleotide-free protein and that GTPase activity is required to rid RdcA of GTP , which acts as an inhibitor . In contrast to other types of GTPases , dynamin-like GTPases have low affinities for GTP [42 , 47] . Since the cellular GTP level is around 1 mM in rapidly growing E . coli cells and drops five-to-tenfold during transition into stationary phase [70 , 71] , an inhibitory role of GTP–preventing RdcA from interacting with DgcE ( Fig 6E ) –could serve as a sensory input into the system . Thus , the system may be fine-tuned in a way that the cellular fraction of RdcA in a non-GTP-bound and therefore DgcE-activating state may increase during entry into stationary phase . Notably , both the RdcA/RdcB system as well as DgcE are present in cells already during earlier phases of the growth cycle ( Fig 4F ) , but their activation–detectable as the beginning of CsgD expression–occurs during the period of entry into stationary phase [37 , 72] , i . e . coincides with the decrease in the cellular GTP pool during this phase of the growth cycle . Future studies of the mechanistic role of RdcB as a co-activator for RdcA , which could act as a GTPase-activating factor or as a nucleotide release factor , will shed more light on these mechanistic details . The importance of DgcE as the 'master diguanylate cyclase' for turning down motility and switching on the production of biofilm matrix components is matched by its functional complexity as already suggested by its six domains , which make DgcE one of the largest proteins in E . coli . Taken together , DgcE seems to make use of the entire tool box of molecular control , including ( i ) small molecule signaling , with GTP exerting a dual function as a potential sensory input as well as a substrate for generating the output molecule c-di-GMP , ( ii ) a balance of antagonistic regulatory protein-protein interactions rather than simple activation , ( iii ) oligomerization in the active state and ( iv ) proteolysis as a counteracting inactivating process . The integration of all these mechanisms allows for highly dynamic and fine-tuned output control . Overall , our study also presents a framework for future studies of this key node in the regulatory network that controls E . coli biofilm formation . Thus , the molecular details and structural consequences of GTP binding and hydrolysis by RdcA , the molecular function of RdcB as a regulatory factor for RdcA and the regulated interaction between RdcA and the MASE1 domain of DgcE will have to be worked out in future genetic and biochemical studies . Also whether and how DgcE and possibly other signaling proteins with N-terminal MASE1 domains are degraded by redundantly acting quality control proteases should be explored further . Finally , DgcE also interacts with its downstream target , the trigger PDE PdeR , and–despite the key role of DgcE-synthesized c-di-GMP in triggering this trigger–has little , if any effect on the constantly very low cellular c-di-GMP pool [13] . Thus , the molecular details of this apparently local downstream signaling by DgcE also deserve further investigation .
The strains used are derivatives of the E . coli K-12 strains W3110 [73] or AR3110 , which is a direct derivative of W3110 , in which codon 6 ( the stop codon TAG ) in the chromosomal copy of bcsQ was changed to the sense codon TTG [74] . Knockout mutations in dgcE , pdeH , pdeR , csgD , clpP , clpA , clpX and lon are full open reading frame deletions/antibiotic resistance cassette insertions previously described [10 , 37 , 40 , 74 , 75] . The newly constructed mutant alleles glpG::kan , hflC::kan , hflK::kan , hslVU::kan , htpX::kan , qmcA::kan , rdcA::kan . rdcB::kan , rdcA-rdcB::kan and ycbZ::kan are deletion/insertion mutations generated by one-step inactivation [76] using the oligonucleotide primers listed in S1 Table . When required , cassettes were flipped out using the protocol of Datsenko and Wanner [76] . Mutations were transferred using P1 transduction [77] . Strains carrying the single copy csgB::lacZ reporter fusion also carry a Δ ( lacI-A ) ::scar deletion as previously described [13 , 75] . In order to test effects of knocking out FtsH protease , a derivative of W3110 carrying ΔftsH3::kan sfhC21 zad220::Tn10 as well as an otherwise isogenic ftsH+ strain carrying only the suppressor ( sfhC21 ) and the co-transducible zad220::Tn10 were used [78] . Chromosomally encoded C-terminally 3xFLAG-tagged constructs were generated using plasmid pSUB11 as a PCR template and the oligonucleotide primers listed in S1 Table following a procedure based on lambda-RED technology [79] , with wildtype dgcE::3xFlAG being previously described [13] . Cells were grown in liquid LB medium under aeration at 28 or 37°C . Antibiotics were added as recommended [77] . Liquid culture growth was followed as optical density at 578 nm ( OD578 ) . Growth of macrocolony biofilms was previously described [74 , 80] . Briefly , 5 μl of the overnight cultures ( free of extracellular matrix , since grown in liquid LB at 37°C ) were spotted on salt-free LB agar plates supplemented with Congo red and Coomassie brilliant blue ( 40 μg/ml and 20 μg/ml , respectively; referred to as 'Congo red plates' ) for the detection of Congo-Red binding ( indicative of curli and cellulose production ) . For growing plasmid-containing strains , 100 μg ml-1 ampicillin was included in the agar plates . In order to achieve reproducible colony morphology , the volume and water content of the agar-containing medium has to be precisely controlled , i . e . plates have to be prepared under exactly identical conditions . All macrocolonies that had to be compared in a given experiment were grown on a single agar plate . Since cellulose and curli fibre expression occurs only below 30°C in E . coli K-12 strains , cultures were grown at 28°C . All oligonucleotide primers used for mutagenesis are listed in S1 Table . All mutated alleles were verified by PCR and DNA sequencing . Point mutations within dgcE ( dgcEGGAAF ) and rdcA ( rdcAK82A , rdcAT103D ) were generated using a four-primer/two-step PCR protocol [81] . Two primers carried the mutation and were complementary to each other ( mutagenic primers , MP , in S1 Table ) , whereas the other two primers bound up- and downstream of the region of the mutation . In a first step , two separate PCR fragments were generated ( primer combinations A and B ) that overlapped at the region of the mutation and covered either the up- or downstream region . In the second step , these fragments were mixed together as DNA templates for a PCR reaction using the up- and downstream primers only in order to generate the final PCR product . This fragment contained the mutated gene region flanked by up- and downstream regions necessary for homologous recombination into the chromosome ( see below ) . For generating dgcE alleles with deletions eliminating internal DgcE domains ( dgcEΔPAS3 , dgcEΔGGDEF ) the reverse MP primer were used that bind directly upstream of the domain sequence that needed to be deleted and were combined with a forward primer binding further upstream of that domain . The forward MP primer contained a sequence that was complementary to the reverse MP primer plus a ~20 nt sequence which bound directly downstream of the domain sequence that needed to be deleted . By combining the forward MP primer with a further downstream binding reverse primer , a PCR product was generated , that contained the nucleotide sequence up- and downstream of the domain that needed to be deleted . After the second step , in which both PCR products were used together as DNA-templates for a PCR reaction using the up- and downstream primers only , the final DNA fragment lacking the single domain was generated and could be used for homologous recombination . For generating the dgcEΔMASE1 and dgcEΔEAL alleles , two-primer combinations were sufficient . dgcEΔMASE1 was generated by a forward mutagenic primer containing an upstream region and the start of dgcE fused to a site directly downstream of the transmembrane ( TM ) domain-encoding region and a reverse primer that binds further downstream in the dgcE gene . dgcEΔEAL was generated by a reverse MP primer containing the DNA sequence directly downstream of the dgcE sequence fused to a ~ 20 nt segment directly upstream of the region encoding the EAL domain and a forward primer that binds further upstream in the dgcE gene . When generating mutated dgcE alleles for cloning into a vector , the whole dgcE gene containing the mutation was amplified carrying appropriate restriction sites . For introducing the dgcEGGAAF , rdcAK82A and rdcAT103D alleles as well as deletion alleles lacking the coding regions for single DgcE domains into the chromosomal background , a two-step method related to the one-step-inactivation protocol [76] was applied . All oligonucleotides for this procedure are listed in the S1 Table . The procedure uses a fragment of plasmid pKD45 [76] , encoding a kanamycin-resistance cassette and the ccdB toxin cassette under the control of a rhamnose-inducible promoter , which is first introduced into the target locus in the chromosome ( step 1 ) , followed by its recombinatorial replacement using a PCR fragment with the desired allele ( step 2 ) , which is selected for by growth in the presence of rhamnose [81] . For generating the ( TM1+2 ) LacY::dgcEΔTM mutant allele , a longer synthetic DNA fragment ( obtained from GeneArt® Strings , Invitrogen ) was used ( for sequence , see S1 Table ) . All other dgcE mutant alleles , rdcAK82A and rdcAT103D were generated using a four-primer/two-step PCR protocol ( see above ) . Correct allelic states of the resulting transformants were verified by PCR and DNA sequencing . Plasmids were constructed using oligonucleotide primers listed in S1 Table . The relevant sequences of the generated plasmids were verified by PCR and DNA sequencing . Unless otherwise indicated , genomic DNA from the E . coli K12 strain W3110 was used as a template for PCR . For in-vivo two-hybrid assays , relevant genes were cloned into pKT25/pKNT25 as well as into pUT18/pUT18C ( Euromedex ) , which generates hybrid proteins with fragments of B . pertussis adenylate cyclase fused to the N-termini/C-termini of relevant proteins . For cloning dgcEΔPAS3 and rdcAT103D , DNA fragments generated by PCR from the respective chromosomal mutant strains were used . Cloning of dgcEΔGGDEF was performed by a four-primer protocol ( see above for mutagenesis ) . For complementation and overexpression assays performed with mutant derivatives of the E . coli K12 strains AR3110 and W3110 , dgcE alleles , rdcA-rdcB , rdcAK82A-rdcB and rdcAT103D-rdcB were cloned into the low copy number lacIq tac promoter vector pCAB18 [38] . C-terminal 6His tag codons fused to dgcE alleles were introduced via the oligonucleotide primers . For cloning dgcEΔPAS::6His , dgcEGGAAF::6His , rdcAK82A-rdcB and rdcAT103D-rdcB into pCAB18 genomic DNA from the respective mutant strains was used ( see above for mutagenesis ) . Protein-protein interactions were generally assayed using an adenylate cyclase-based bacterial two-hybrid system [30] . The construction of the hybrid plasmids from pUT18/pUT18C and pKT25/pKNT25 vectors is described above . Plasmids were co-transformed into a Δcya derivative of the K12 strain W3110 and incubated for 24 h at 28°C on MacConkey agar plates supplemented with maltose ( 1% ) , ampicillin ( 100 μg/ml ) and kanamycin ( 50 μg/ml ) . Single colonies were resuspended in 50 μl sterile water , 5 μl of this suspension was spotted onto fresh MacConkey agar plates and incubated at 28°C for up to 2 d . All colonies to be compared were grown together on a single agar plate , but separate close-up photographs were taken to obtain higher photographic resolution . Red colonies indicate utilization of maltose which depends on cAMP which indicates reconstitution of adenylate cyclase via direct interaction of the proteins fused to the otherwise separate adenylate cyclase domains . The stably dimerizing leucin zipper part ( 'zip' ) domain of the yeast transcription factor GCN4 ( cloned onto pKT25 and pUT18C ) was used as a positive control [30] . For testing interactions between soluble proteins or domains , also the Bacterio-Match II two-hybrid system ( Agilent Technologies ) was used , in which candidate proteins or domains are linked to the NTD of lambda cI ( expressed from pBT ) and to the bacterial RNA polymerase alpha-NTD ( expressed from pTRG ) , with co-expression of interacting proteins leading to expression of the HIS3 gene . This suppresses histidin auxotrophy of the E . coli reporter strain ( a derivative of XL1-Blue MRF’ ) in a manner that can be fine-tuned by adding the His3 inhibitor 3-amino-1 , 2 , 4-triazole ( 3-AT ) [32] . The detailed procedure as well as the two-hybrid plasmids expressing hybrid proteins containing different DgcE domains have been described previously [13] . For immunoblot analyses of 3xFLAG- and 6His-tagged proteins , samples were taken at different time points during growth in LB medium . For CsgD analysis samples were taken at an OD578 of 3 , 5–4 ( CsgD expression starts at an OD578 of 2 . 5 and is shut off again later in stationary phase; [72] ) . Samples corresponding up to 100 μg of total cellular protein were re-suspended in SDS-PAGE sample buffer and incubated for 10 min at 70°C and 15 min at 100°C . Samples were adjusted such that all samples to be compared contained similar amounts of total protein . These were then loaded along with a protein size marker and run on SDS-polyacrylamid ( 10% ) gels . Proteins were detected by immunoblotting as previously described ( Lange and Hengge-Aronis , 1994 ) using antibodies against CsgD ( custom-made by Pineda Antikörper Service; used at 1:10000 dilution ) , 6His-tag ( Bethyl Laboratories , Inc . ) or 3xFLAG-tag ( Sigma ) . Anti-rabbit or anti-mouse IgG horseradish-peroxidase conjugate from donkey ( GE Healthcare ) was used ( at 1:20000 dilution ) for protein visualization in the presence of Western Lightning® Plus-ECL enhanced chemiluminescence substrate ( PerkinElmer ) . Use of the WesternC Precision Plus Marker ( Biorad ) required also the addition of Streptactin-HRP conjugate . Non-specific protein bands were identified by comparison of the band patterns of recombinant strain samples with those of wildtype E . coli K-12 samples with similar amounts of cellular protein loaded on the same gels . Strains were grown at 28°C under aeration in LB medium . At an OD578 of 3 , 10 ml culture volume was pelleted ( 4°C , 5000 rpm , 30 min ) and stored at -80°C . Sample extraction and analysis of c-di-GMP by LC-MS/MS was performed as described previously [82] . Intracellular concentrations of c-di-GMP were calculated by using the standard OD/cell mass ratio [77] . Extractions were performed in biological triplicates . β-galactosidase activity was assayed by use of o-nitrophenyl-β-D-galactopyranoside ( ONPG ) as a substrate and is reported as mmol of o-nitrophenol per min per mg of cellular protein [77] . Experiments showing the averaged expression of lacZ fusions were done in three or more biological replicates . Macrocolonies were visualized at 10-fold magnification with a Stemi 2000-C stereomicroscope ( Zeiss; Oberkochen , Germany ) . Digital photographs were taken with an AxioCam ICC3 digital camera coupled to the stereomicroscope , which was operated using AxioVision 4 . 8 software ( Zeiss ) . | Biofilms represent a multicellular life form of bacteria , in which large numbers of cells live in communities surrounded and protected by a self-generated extracellular polymeric matrix . As biofilms tolerate antibiotics and host immune systems , they are causally associated with chronic infections . Biofilm formation is generally promoted by the ubiquitous bacterial second messenger c-di-GMP . DgcE , one of the 12 diguanylate cyclases that produce c-di-GMP in E . coli , was previously shown to specifically act as a top level trigger in the regulatory network that drives biofilm matrix production in this bacterium . However , signal input into DgcE itself , which is a large six-domain protein , had remained unknown . Here we demonstrate that DgcE activity is controlled by a novel type of dynamin-like GTPase that directly interacts with the N-terminal membrane-intrinsic MASE1 domain of DgcE . Our finding of a dual function of this MASE1 domain , which is essential for both activation and continuous proteolysis of DgcE , is the first characterization of this widespread bacterial signaling domain . Signal input via the dynamin-like GTPase system suggests that c-di-GMP production by DgcE might be stimulated by the decreasing cellular GTP level during entry into stationary phase , which is precisely the time when biofilm matrix production is turned on . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"bacteriology",
"biofilms",
"deletion",
"mutation",
"enzymes",
"metabolic",
"processes",
"enzymology",
"microbiology",
"mutation",
"molecular",
"biology",
"techniques",
"physical",
"chemistry",
"chemical",
"properties",
"research",
"and",
"analysis",
"methods",
"dimerizatio... | 2019 | Genetic dissection of Escherichia coli's master diguanylate cyclase DgcE: Role of the N-terminal MASE1 domain and direct signal input from a GTPase partner system |
Soil-transmitted Helminths and Anemia potentially reduce and retard cognitive and physical growth in school-age children with great implications for national control programs in Africa . After 13 years of deworming and limited health education campaigns , a study was undertaken to evaluate the impact of deworming interventions on the prevalence and intensity of soil-transmitted helminthic infections in school-age children in Uganda . A cross-sectional study was carried out in six regions of Uganda , where two districts were randomly selected per region based on the ecological zones in the country . Included in the study were the districts; Mpigi and Nakasongola from the Central; Nakapiripirit and Kotido from Karamoja; Arua and Yumbe from West Nile; Gulu and Alebtong from the North; Kaliro and Mbale from the East; Hoima and Bundibugyo in the West . Five schools were randomly selected from each district and in each school 50 children aged 6–14 years were randomly selected . Stool samples were taken each child and examined for the presence of helminthic infections . A short pretested questionnaire was administered to each participant to obtain their knowledge , attitude , and practice relating to STH infections , their control . General observations were made on environmental sanitation in the schools . The location of each school was geo-referenced using a GPS machine ( Garmin®GPSMAP62 , Garmin Ltd , Southampton , UK ) . In total , 4 , 285 children were assessed including 719 ( 16 . 82% ) from central region , 718 ( 16 . 80% ) from eastern region , 719 ( 16 . 82% ) from northern region , 689 ( 18 . 82% ) from Karamoja region , 717 ( 16 . 77% ) from West Nile region and 723 ( 16 . 91% ) from western region . The average age of the children was 12 . 6 years with a standard deviation , SD 1 . 8 years and the minimum age was 6 years and upper age limit of 12 years . The percentage of boys ( 50 . 1% ) and girls ( 49 . 9% ) was comparable . 8 . 8% ( 95% CI; 8 . 0–9 . 7 ) were infected with at least any one STH species . Hookworm was the most prevalent ( 7 . 7%; 95% CI; 6 . 9–8 . 5 ) followed by whipworms ( Trichuris trichiura ) ( 1 . 3%; 95% CI; 1 . 0–1 . 7 ) and roundworms ( Ascaris lumbricoides ) ( 0 . 5%; 95% CI; 0 . 3–0 . 7 ) . Some children had Schistosoma mansoni , 13 . 0% ( 95% CI; 12 . 0–14 . 0 ) . All the children knew what soil transmitted helminths were ( 62 . 8% , 95% CI: 61 . 3–64 . 2 ) and most common knowledge of information were from; home ( 39% , 95% CI: 37 . 1–40 . 8 ) , media ( radio& newspaper ) ( 11% , 95% CI: 9 . 8–12 . 2 ) , school ( 65 . 7% , 95% CI: 63 . 9–67 . 5 ) and friends ( 11 . 5% , 95% CI: 10 . 3–12 . 7 ) . Majority were aware of how one gets infected with soil transmitted helminths through; eating contaminated food ( 77 . 5% , 95% CI: 76 . 0–79 . 1 ) , walking barefoot ( 59 . 6% , 95% CI: 57 . 8–61 . 5 ) , drinking contaminated water ( 52 . 9% , 95% CI: 51 . 0–54 . 8 ) , playing in dirty places ( 21 . 8% , 95% CI: 20 . 2–23 . 3 ) and dirty hands ( 2 . 3% , 95% CI: 1 . 7–2 . 9 ) . Semi-annual deworming campaigns have proved effective in significantly reducing helminthic infections in most of the districts in Uganda . Regular evaluations are vital to assess impact of the interventions and guide programme implementation . Our data shows that the prevalence of infection has been reduced to a level where STH morbidity is no longer of public health importance in most districts surveyed .
In terms of the disease burden in school-age populations in developing countries including Uganda [7] , intestinal helminths infections rank first among the causes of all communicable and non-communicable diseases [8] . Field studies of Schistosomes and the major intestinal nematodes Trichuris trichiura and Ascaris lumbricoides repeatedly demonstrate that the intensity and prevalence of infection exhibit marked dependency on host age[9] . Peak levels of infection typically occur in hosts aged between 10 and 14 years in endemically infected communities [10] . Age-dependent patterns of infection prevalence are generally similar among the major helminth species , exhibiting a rise in childhood to a relatively stable asymptote in adulthood [11] . Epidemiological studies of STH infections have shown that the prevalence and intensity of infection are highest among children 4–15 years of age[3] . For the vast part of Uganda , a uniform treatment strategy involving school-based ( for school age children ) and community-based treatment ( for under-fives ) is implemented twice yearly during the child health days in April and October . However , in schistosomiasis high-risk villages and in lymphatic filariasis ( LF ) endemic districts , the first round of mass drug administration ( MDA ) targets the whole population while the second round of treatment is limited to under-fives and school-age children . No study has been done to assess the impact of school-based deworming which has been ongoing since 2004 . Program monitoring has been limited to reported treatment coverage , but the reports have never been validated and it is possible that the data are exaggerated . Soil-transmitted helminths frequently cause chronic and debilitating diseases , mainly in infants , preschool and school-aged children , adolescent girls and pregnant women [12–13] . Global burden is typically expressed in DALYs estimated as high as 39 million , similar to malaria or tuberculosis[11 , 14] . Globally an estimated 807–1221 million people are infected with A . lumbricoides , 604–795 million with T . trichiura , 576–740 million with hookworms [12 , 15] . Concurrent infections with multiple helminth species are common [16–19] a known cause of malnutrition , intestinal obstruction , biliary colic and pancreatitis[20] . T . trichiura infections can induce Trichuris dysentery syndrome , whose symptoms include rectal prolapse , anemia , and clubbing of fingers[21] . Hookworm is implicated as the causative factor in more than 50% of cases of iron deficiency anemia in Asia and Africa [22] . Generally , infections with STH have a negative impact on pregnancy and birth outcomes , hamper children’s cognitive and physical development [23–24] , resulting in reduced work capacity , and therefore compromise the social and economic development of communities and entire nations [11 , 25] . An infection level of A . lumbricoides and T . trichiura are highest in children aged between 5 and 14 years [14]with decline in frequency and intensity in adulthood [26] . This age dependency might be due to changes in exposure and/or acquired immunity [10] and morbidity associated with the number of worms harbored[27] . The aim of this study was to evaluate the impact of a national deworming campaign on the prevalence and intensity of soil-transmitted helminthic infections in Uganda .
The selection criteria for the districts was based on the six regions in which two districts were selected per region with Central region ( Mpigi and Nakasongola ) , Karamoja region ( Nakapiripirit and Kotido ) , West Nile region ( Arua and Yumbe ) , Northern Region ( Alebtong and Gulu ) , Eastern region ( Mbale and Kaliro ) and Western region ( Bundibugyo and Hoima ) . The Fig 1 below shows the list districts that were selected for the surveys . The study was descriptive and cross-sectional . The major objective of the study was to assess the impact of deworming on STH and guided by the following objectives: From each district , five schools were randomly selected . In each of the selected schools , children aged 6–14 years ( 30 boys and 30 girls ) were randomly selected , whereas , in communities , 60 participants aged 5 years and above were selected and conducted during June-July 2016 period . Duplicate Kato-Katz thick smears [2] were prepared in the field upon receipt of the fecalsamples and were read under a microscope within 60 minutes of slide preparation to determine hookworm ova . To avoid inter-observer errors , the two slides from one specimen were read by different technicians . For quality control , 10% sub-sample of the slides was re-read by an experienced technician . There was no significant difference between the technicians’ readings and those of the sampled slides . Haemoglobin ( Hb ) levels were measured using finger prick blood taken from each child for the assessment of Hb as proxy to measuring anaemia levels . The Hb was measured in the field using a portable hamoglobinometer ( HemoCue Hb301 , LTD Angelholm , Sweden ) . Determination of anaemia levels followed WHO references defined as: Hb< 115 g/L for children 5–11 years; Hb<120 g/L for children 12–14 years; Hb<120 g/L for non-pregnant women ≥ 15 years; Hb<110 g/L for pregnant women; Hb<130 g/L for men ≥ 15 years [28–29] . A short pretested questionnaire was administered to each participant to obtain their knowledge , attitudes , and practices relating to infection and control of STH . General observations were made on the sanitations of their environments . Observations were made on school and community environment including availability and conditions of latrines and hand washing facilities . Coordinates of each school were taken using a GPS machine ( GarminGPSMAP62 , Garmin Ltd , Southhampton , UK ) . All children found positive for intestinal schistosomiasis ( egg-patency ) were treated with praziquantel ( Distocide , Shin Poong Pharmaceuticals , Seoul Republic of Korea ) at 40 mg/kg body weight . Regardless of infection status , a tablet of albendazole ( 400mg ) was given to each child for soil-transmitted helminthiasis . This study was approved by the Ugandan National Council of Science and Technology and formed part of the monitoring and surveillance activities of the Ugandan National Bilharzia & Worm Control Programme . Permission from school authorities was sought and children consented before participation in the study . A descriptive analysis of the data on prevalence and intensity of infection from the national registry of the Ministry of Health in 2004 was done and compared with the recent study for the selected districts . Besides the analysis of frequency and proportion distributions , the geographic locations of prevalence and intensity of infection data points for school-age children were mapped , as this was useful to visualize gaps in data . The database was made with MS-Excel 2013 and the analysis with STATAVersion10 . All the data were double entered into the computer using excel programme by different clerks . The impact on the intensity of infection was determined using student’s T-test . Binary logistic regression analysis was utilized to determine WASH factors as associated with STH infections . Multivariable models for infection with each type of infection was adjusted for gender , residence in a treatment area , playing barefoot; and reported the presence of latrine and any other purported risk factor under study . None overlapping 95% confidence intervals and p-values <0 . 05 were considered as significant levels .
Table 2 above summarizes the prevalence and intensities of infection by the district . Generally , the pattern of Soil-transmitted helminths ( STH ) varied markedly with an overall 8 . 8% , ( 95% CI; 8 . 0–9 . 7 ) infected with at least any one STH species . The most common soil-transmitted helminths were hookworms ( 7 . 7% 95% CI; 6 . 9–8 . 5 ) , Trichuris trichiura ( 1 . 3% , 95% CI; 1 . 0–1 . 7 ) and Ascaris lumbricoides ( 0 . 5%; 95% CI; 0 . 3–0 . 7 ) respectively . The highest prevalence infection was observed in Bundibugyo district with the prevalence of 24 . 9% [20 . 5–29 . 4] , 4 . 1% [2 . 1–6 . 1] , 4 . 1% [2 . 1–6 . 1] and 28 . 2% [23 . 6–32 . 8] for Hookworm , Ascaris lumbricoides , Trichuris trichiura and any infections of STH respectively . The prevalence of infection for Hookworm infections varied from 0 . 0% in Kotido to -24 . 9% in Bundibugyo district , while for Ascaris lumbricoides ranged from 0 . 0% in Arua , Gulu , Kaliro , Kotido , Mbale , Nakasongola and Yumbe to 4 . 1% in Bundibugyo district . A similar trend to Ascaris lumbricoides was observed for Trichuris trichiura infections . Generally , the majority of infections observed across all the districts surveyed had a light intensity of infections with the greatest number of light infection category observed at 57 ( 98 . 3% ) , 329 ( 97 . 1% ) and 19 ( 86 . 4% ) for Trichuris trichiura infections , Hookworm infections , and Ascaris lumbricoides respectively . The prevalence of light intensity of infections for Hookworm infections ranged from 0 . 0% to 100% with majority ( 97 . 1% ) of infections in this category however; there was a marked reduction in the prevalence of intensities of infections from moderate ( 1 . 2% ) to heavy ( 1 . 8% ) intensities of infections observed across all surveyed districts . The prevalence of light intensities of infections for Ascaris lumbricoides varied from 0 . 0% to 100% with an overall 86 . 4% while the prevalence of moderate intensities of infections ranged from 0 . 0% to 100% with an overall 13 . 6% . There were no individuals observed with heavy intensities of infection . The prevalence of light intensities of infections for Trichuris trichiura infections ranged from 0 . 0% to 100% with an overall 98 . 3% while Mpigi district recorded 4 . 2% prevalence for moderate intensities of infections . There were no individuals observed with heavy intensities of infection . Overall , the study results show that the prevalence of STH was generally high at baseline ( 2002 ) compared to the findings of this survey in 2016 . For example; in Yumbe district that had the highest burden of STH among the districts surveyed in 2002 versus 2016 showed a tremendous reduction from 62 . 5% to 0 . 3% ( 99 . 5% reduction in prevalence ) . A similar trend was observed in other districts like; Bundibugyo ( 56 . 8% vs . 27 . 7% ) , Gulu ( 55 . 3% vs . 1 . 1% ) , Nakasongola ( 54 . 3% vs . 6 . 6% ) , Arua ( 54 . 3% vs . 2 . 5% ) , Mbale ( 54 . 1% vs . 6 . 9% ) , Alebtong ( 45 . 4% vs . 3 . 6% ) , Hoima ( 27 . 9% vs . 13 . 2% ) , Kotido ( 24 . 7% vs . 0 . 3% ) and Nakapiripirit ( 12 . 2% vs . 3 . 9% ) . However , there was an increase in STH prevalence observed for Kaliro ( 16 . 4% vs . 21 . 9% ) as shown in Fig 2A . Overall 22 . 2% were anaemic with 0 . 4% having severe anaemia . Gulu district had the highest percentage of individuals with severe anaemia at ( 56 . 0% any anaemia vs . 1 . 7% severe anaemia , 95% CI; 113 . 2–116 ) , followed by Kaliro district with 2 . 4% for any anaemia vs . 1 . 1% severe anaemia , 95% CI; 120 . 9–132 ) . Other districts with severe cases were Yumbe ( 37 . 3% any anaemia vs . 0 . 3% severe anaemia , 95% CI; 120 . 3–137 . 2 ) , Nakasongola ( 15 . 6% any anaemia vs . 0 . 3% severe anaemia , 95% CI; 126–129 ) , Mpigi ( 6 . 7% moderate anaemia vs . 0 . 3% severe anaemia , 95% CI; 131 . 8–134 ) , Hoima ( 26 . 1% any anaemia vs . 0 . 3% severe anaemia , 95% CI; 123 . 4–126 ) , Bundibugyo ( 33 . 7% any anaemia Vs 0 . 3% severe anaemia , 95% CI; 118 . 9–131 ) and Alebtong ( 0 . 6% any anaemia Vs 0 . 3% severe anaemia , 95% CI; 168 . 4–172 ) as shown in Table 3 below . The limitation here is lack of baseline anaemia data for these districts for comparison . Mass drug administration ( MDA ) is the mainstay of morbidity control for schistosomiasis and soil-transmitted helminths ( STHs ) [5]; hookworm , roundworm , and whipworm are the three main species of STHs that infect humans . MDA is the delivery of free single-treatment preventive chemotherapies at regular intervals to endemic populations . STHs are treated with albendazole ( ALB ) or mebendazole ( MBZ ) . Repeated annual or biannual treatments are necessary mainly due to susceptibility to reinfection after treatment [30–31] . The central challenges identified by the WHO to increase MDA coverage are sustaining financial support for MDA , improving monitoring and evaluation , expanding local administrative capacity , and increasing access to preventive chemotherapies during MDA[32] . Implemented by districts using school teachers and volunteers known as Village Health Teams ( VHTs ) and Healthcare workers during Child Days plus ( CDP ) with the aim to control morbidity . The delivery strategy is through mass annual anthelmintic treatment targeted at school-aged children and high-risk groups in the endemic areas using Albendazole ( ALB ) /Mebendazole ( MBZ ) to treat STH infection [33–34] . Generally , there has been a tremendous improvement in the STH worm burden over the years . However , there is still a challenge to STH elimination as some districts continue to register lower MDA coverage than recommended >75% by WHO as indicated in Fig 3 below . All the children interviewed knew what STH were ( 62 . 8% , 95% CI: 61 . 3–64 . 2 ) , Table 4 . The most common knowledge of information were; home ( 39 . 0% , 95% CI: 37 . 1–40 . 8 ) , media ( radio& newspaper ) ( 11 . 0% , 95% CI: 9 . 8–12 . 2 ) , school ( 65 . 7% , 95% CI: 63 . 9–67 . 5 ) and friends ( 11 . 5% , 95% CI: 10 . 3–12 . 7 ) . Majority of the study participants interviewed were aware of how one gets infected with STH through; eating contaminated food ( 77 . 5% , 95% CI: 76 . 0–79 . 1 ) , walking barefoot ( 59 . 6% , 95% CI: 57 . 8–61 . 5 ) , drinking contaminated water ( 52 . 9% , 95% CI: 51 . 0–54 . 8 ) , playing in dirty places ( 21 . 8% , 95% CI: 20 . 2–23 . 3 ) and dirty hands ( 2 . 3% , 95% CI: 1 . 7–2 . 9 ) .
The target for STH in Uganda is control whereby we aim at eliminating the associated morbidity to levels of no public health importance ( WHO 2012 Roadmap for NTDs Implementation towards 2020 ) . Treatment and limited health education alone cannot sustain STH morbidity elimination . The high reproductive capacity of soil transmitted helminthes means that in the absence of additional interventions , transmission cannot be interrupted even if the infection intensity is greatly reduced [40] and this can lead to morbidity recrudescence . There is need for stable provision and use of adequate sanitation facilities to end open defecation , improved personal and food hygiene behavior and access to clean water [41] . The Uganda sanitation and water supply have made substantial progress since 1990s especially in urban centers ( http://en . wikipedia . 0rg/wiki/water ) . The government with support from multilateral and bilateral agencies , NGOs and the private sector has been vigorously supporting programmes aimed at improving safe water supply for now over 3 decades . In early 1990s , only 10% of urban dwellers were supplied with safe water but the figure had risen to 81% by 2006 . However , most households and schools rural areas in Uganda still have inadequate sanitation and safe water supply . When the whole population estimated at 37 . 5 million people is considered , access to safe water was still just 62% [42] and has improved a little since then . Despite efforts punt into improving sanitation , it has been accorded low priority with little resources and inadequate collaboration and coordination among stakeholders [43] . According to Ministry of Health records , the national average toilet coverage is still at around 68% but it varies; 80% in urban areas , 60% in rural areas and about 40% in slums [43] . However , latrine coverage in some communities can be as low as 10% especially in fish landing villages [44] . In some rural areas like Karamoja and the Lake Victoria Islands , superstitions against using toilets result in extensive open air defecation [43] . There is almost complete absence of hand washing facilities [43–44] and all these factors promote STH transmission . The situation as it is today shows reduced levels of STH morbidity . This status cannot be guaranteed unless sanitation issues are well addressed . Without significantly improving WASH , it will be difficult to interrupt STH transmission in most parts of the country . It is also necessary to prioritize operational research agenda that includes NTD monitoring within the local government health sector . Nationally , there is need to monitor STH programme every 5 years to guide the MoH on feasible interventions . Routine evaluations should not be limited to impact on infection rates but should also aim at evaluating impact of the programme on nutrition and educational achievements . This would leverage proper resource allocation and utilization . Combating the burden of STH requires a holistic approach and it should be a responsibility of all stakeholders including the affected communities . | Soil-transmitted Helminths potentially reduce physical growth and retard cognitive development in school-age children ( SAC ) with great implications for national control programs in Africa . In Uganda , baseline investigations between 1998 and 2002 , indicated STH prevalence was over 60 . 0% in most districts , the commonest worms infections were Hookworms , Ascaris and Trichuris . Twice a year national deworming campaign was initiated in 2003 targeting aged 1–14 years . Over ten years of deworming campaigns , has reduced the overall STH prevalence to 8 . 8% in 2016 . The findings suggest routine deworming campaigns reduce STH exposure and infections . Periodic program evaluations are key to determining the progress made in order to achieve the elimination targets by 2020 . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"education",
"helminths",
"sociology",
"tropical",
"diseases",
"hookworms",
"social",
"sciences",
"geographical",
"locations",
"uganda",
"parasitic",
"diseases",
"animals",
"anemia",
"health",
"care",
"ascaris",
... | 2018 | Impact of a national deworming campaign on the prevalence of soil-transmitted helminthiasis in Uganda (2004-2016): Implications for national control programs |
START-dependent transcription in Saccharomyces cerevisiae is regulated by two transcription factors SBF and MBF , whose activity is controlled by the binding of the repressor Whi5 . Phosphorylation and removal of Whi5 by the cyclin-dependent kinase ( CDK ) Cln3-Cdc28 alleviates the Whi5-dependent repression on SBF and MBF , initiating entry into a new cell cycle . This Whi5-SBF/MBF transcriptional circuit is analogous to the regulatory pathway in mammalian cells that features the E2F family of G1 transcription factors and the retinoblastoma tumor suppressor protein ( Rb ) . Here we describe genetic and biochemical evidence for the involvement of another CDK , Pcl-Pho85 , in regulating G1 transcription , via phosphorylation and inhibition of Whi5 . We show that a strain deleted for both PHO85 and CLN3 has a slow growth phenotype , a G1 delay , and is severely compromised for SBF-dependent reporter gene expression , yet all of these defects are alleviated by deletion of WHI5 . Our biochemical and genetic tests suggest Whi5 mediates repression in part through interaction with two histone deacetylases ( HDACs ) , Hos3 and Rpd3 . In a manner analogous to cyclin D/CDK4/6 , which phosphorylates Rb in mammalian cells disrupting its association with HDACs , phosphorylation by the early G1 CDKs Cln3-Cdc28 and Pcl9-Pho85 inhibits association of Whi5 with the HDACs . Contributions from multiple CDKs may provide the precision and accuracy necessary to activate G1 transcription when both internal and external cues are optimal .
Cyclin-dependent protein kinases ( CDKs ) act as molecular machines that drive cell division , and cell cycle progression is dependent on oscillation between CDK active and inactive states . In S . cerevisiae , the CDK Cdc28 associates with nine different cyclin subunits to promote and coordinate a complex network of events necessary for smooth cell cycle transitions [1] . Irreversible commitment to a new round of cell division occurs toward the end of G1 phase in a process called Start in yeast . The analogous regulatory event is called the restriction point in mammalian cells [2] , [3] . In yeast , three G1 cyclins , Cln1 , Cln2 , and Cln3 , associate with Cdc28 to initiate events required for progression through Start . Passage through Start catalyzes a defined molecular program that initiates DNA replication , budding , spindle maturation , and chromosome segregation [3] . One key feature of Start in yeast , and G1 progression in other eukaryotic cells , is the induction of a transcriptional program involving over 200 genes , including those encoding the G1 ( CLN1 , CLN2 , PCL1 , and PCL2 ) and B-type cyclins ( CLB5 and CLB6 ) [4] , [5] . G1/S phase-specific transcription depends on two heterodimeric transcription factors called SBF ( Swi4 , 6 cell cycle box binding factor ) and MBF ( MluI binding factor ) . These complexes share a common regulatory subunit , Swi6 , which is tethered to DNA via its binding partners , encoded by SWI4 in SBF and MBP1 in MBF [5] . At the well-studied HO locus , binding of the zinc-finger transcription factor Swi5 is followed by recruitment of the Swi/Snf chromatin remodeling complex and the SAGA histone acetyltransferase complex [6]–[8] . These events set the stage for SBF binding and recruitment of the SRB/mediator complex [6] . Importantly , subsequent recruitment of PolII and transcription initiation is dependent on CDK activity [9] . Although any one of the three G1 cyclins is sufficient to drive Start , genetic studies indicate a key role for Cln3-Cdc28 in activating SBF and MBF . At the same time Cln1 and Cln2 are required for the proper execution of other Start-related events such as budding and DNA synthesis . Cells lacking CLN3 are large and severely delayed for onset of G1/S transcription , while ectopic induction of CLN3 in small G1 cells activates transcription and accelerates passage through Start [10] . Start does not occur until cells have passed a critical cell size threshold , a barrier modulated by nutrient conditions , among other regulatory inputs [11] . A systematic analysis of cell size profiles for the entire set of yeast deletion mutants uncovered many new regulators of Start including Whi5 and implicated it as an inhibitor of G1/S-specific transcription [12] , [13] . Whi5 occupies specific promoters early in G1 phase when CDK activity is low . However , Cdc28-dependent phosphorylation of both Whi5 and SBF/MBF late in G1 phase results in disengagement from SBF and nuclear export of Whi5 consequently leading to activation of SBF- and MBF-dependent transcription [12] , [13] . Whi5 is proposed to function in a manner analogous to the well-characterized Rb family proteins in metazoans . E2F , the functional analog of SBF/MBF , regulates G1-specific gene expression required for passage through the restriction point [14] . E2F activity is restricted to late G1 phase because of inhibition by the retinoblastoma protein ( Rb ) . Rb associates with E2F to restrain its activity until late G1 , at which point stepwise phosphorylation of Rb by two CDKs , cyclin D-Cdk4/6 and cyclin E-Cdk2 , causes the dissociation of Rb from E2F [15] . This process appears to be regulated by a positive feedback loop in which Rb phosphorylation by cyclinE-Cdk2 leads to further dissociation of Rb from promoters and enhancement of G1-transcription . At the molecular level , Rb interacts with both E2F and chromatin remodeling complexes such as histone deacetylases ( HDACs ) [16]–[18] . Rb appears to repress transcription through at least three distinct mechanisms: ( 1 ) Rb can bind directly to the activation domain of E2F thereby blocking its activity [19]; ( 2 ) recruitment of Rb can block the assembly of the pre-initiation complex thus inhibiting the activity of adjacent transcription factors [20] and; ( 3 ) Rb can recruit remodelers such as HDAC1 and BRG1 to modify chromatin structure . BRG1 is one of the human Swi/Snf adenosine triphosphatases ( ATPases ) that remodel nucleosomes by utilizing ATP to weaken the interactions between DNA and histones [16] , [17] . The specific roles of different CDKs in regulating E2F-Rb function , however , remain unclear . Another yeast CDK Pho85 was originally discovered as a regulator of phosphate metabolism , but has since been shown to play numerous roles in the regulation of cell division and other processes [21]–[23] . Ten genes encoding Pho85 cyclins ( Pcls ) have been identified and they appear to dictate substrate and functional specificity of Pho85 [24]–[26] . Expression of three Pcls , PCL1 , PCL2 , and PCL9 , is restricted to G1 phase of the cell cycle [25] . Specifically , PCL9 expression peaks early in G1 , whereas maximal expression of PCL1 and PCL2 is observed at Start and is dependent largely on SBF [27]–[29] . Although Pho85 is not essential for viability , it is required for cell cycle progression in the absence of the Cdc28 cyclins CLN1 and CLN2 [29] , and its absence leads to catastrophic morphogenic changes that culminate in a G2 arrest [30] . Consistent with this observation , inactivation of both Cdc28 and Pho85 CDKs specifically inhibits expression of G1-regulated genes involved in polarized growth [31] . As noted above , transcriptional repression by Rb has been linked to its interaction with histone modification complexes , in particular HDACs . Recent work highlights the importance of post-translational modifications of histone and nucleosome positioning in regulating gene expression [32] , [33] . Histone acetylation neutralizes the positive charge generated by lysine-rich regions present in the N-terminal tails of histones , thereby disrupting nucleosome structure and increasing promoter accessibility [34] . As a result , many transcription activators have been shown to interact with histone acetyltransferases , whereas transcriptional repressors often associate with HDACs to promote nucleosome formation to occlude transcription factor binding [35] , [36] . Histone deacetylation in S . cerevisae is mediated by a family of HDACs including Rpd3 , Hda1 , Hda2 , Hos1 , Hos2 , and Hos3 [37] . Similar to their mammalian counterparts , some yeast HDACs are recruited to promoters by sequence-specific regulatory factors to repress gene expression . For example , the Rpd3 deacetylase complex is recruited to the INO1 promoter by the DNA binding protein Ume6 [35] , [38]–[40] . This recruitment results in local histone deacetylation and repression of INO1 gene expression [41] . Hda1 is another example of this type of regulation , and is recruited to its target promoters by the repressor Tup1 [42] . In this study , we provide detailed mechanistic insights into Whi5-dependent regulation of G1-specific transcription and cell cycle progression . Specifically , we identify Whi5 , to our knowledge , as the first demonstrated physiological substrate for the G1-specific Pcl9-Pho85 CDK and provide genetic and biochemical evidence supporting a direct role for Pho85 at Start . Furthermore , we show that in a manner similar to Rb in mammalian cells , Whi5-mediated repression involves the HDACs Rpd3 and Hos3 . Dual phosphorylation of Whi5 by Cdc28 and Pho85 inhibits Whi5 activity in at least two ways . Both kinases appear to regulate interaction of Whi5 with different HDACs , whereas Cdc28 is also involved in disrupting Whi5 association with SBF and promoting its nuclear export [12] , [13] . G1-specific CDKs thus are specialized to regulate different aspects of the same critical cell cycle event—inhibition of Whi5—resulting in definitive inactivation of the Whi5 repressor .
Synthetic dosage lethality ( SDL ) is a genetic assay that is based on the rationale that increasing levels of a protein may have no effect on the growth of an otherwise wild-type ( wt ) strain but may cause a measurable phenotype—such as lethality—in a mutant strain with reduced activity of an interacting protein [43] , [44] . Previous studies suggest that SDL can be used effectively to identify novel enzyme targets and a genome-wide SDL screen in cells lacking Pho85 identified known targets of the CDK [23] . In addition to known substrates , several putative Pho85 targets were also identified , including the G1-specific transcription repressor Whi5 [24] . To further explore the role of Pho85 in G1 phase-specific transcription we examined the WHI5-PHO85 SDL interaction in greater detail . As noted previously , Pho85 activity and substrate specificity depends on its interaction with cyclin subunits known as Pcls [25] . To implicate specific Pcl-Pho85 complexes in modulating Whi5 function we examined the effects of WHI5 overexpression in cells lacking different Pcls ( Figure 1 ) . Similar to effects observed in cln3Δ and cln1Δ cln2Δ mutants [13] , overexpression of WHI5 resulted in growth inhibition of pcl1Δ and pcl9Δ deletion strains and this growth defect was exacerbated in a pcl1Δ pcl9Δ double mutant ( Figure 1 ) . Unlike pcl1Δ or pcl9Δ mutants , strains lacking PCL2 or PHO80 cyclins were not adversely affected by increased WHI5 dosage suggesting that the WHI5-PHO85 genetic interaction is dependent on the PCL1 , 2 cyclin subfamily and more specifically on PCL1 and PCL9 ( Figure 1 ) . This observation is consistent with the fact that Pcl1 and Pcl9 ( but not Pcl2 ) are the two G1-specific cyclins that localize to the nucleus [30] , [45] . The growth phenotype seen in the plating assay was confirmed by measuring growth rates in liquid culture ( unpublished data ) . On the basis of these results , Pcl1/9-Pho85 may contribute to Whi5 regulation in a manner similar to Cln3-Cdc28 . The genetic interactions described above suggest Whi5 may be a direct target of Pho85 . Evidence supporting this hypothesis is provided by protein microarray assays where Whi5 is phosphorylated in vitro by Pcl1-Pho85 [46] . We characterized the Whi5-Pho85 interaction biochemically by performing in vitro kinase assays using recombinant Pcl-Pho85 CDK complexes and purified Whi5 as substrate ( Figure 2A ) . Incorporation of [32P] into Whi5 was not detected in the absence of CDKs ( Figure 2A , lane 4 ) . However , Whi5 phosphorylation was observed in the presence of Pcl1- and Pcl9-Pho85 ( Figure 2A , lanes 1 , 2 ) and when compared to Cln2-Cdc28 kinase activity , Pho85 and Cdc28 phosphorylated Whi5 at similar levels in vitro ( Figure 2A , lanes 1–3 ) . Previous studies revealed multiple Whi5 slow-migrating isoforms that correlate with its phosphorylation state [12] , [47] . We examined the effect of various cyclin or CDK mutants on Whi5 mobility ( Figure 2B ) . Because of genetic redundancy of Pcl cyclins [27] , we were unable to reproducibly detect changes in Whi5 phosphoforms in cyclin mutant strains . Therefore , a Pho85 mutant was used to asses the phosphorylation status of Whi5 . Consistent with previous findings [12] , [13] , slow migrating Whi5 isoforms present in asynchronous wt extracts ( Figure 2B , lane 1 ) were modestly reduced in cells lacking CLN3 ( Figure 2B , lane 7 ) and completely absent in a cln1Δ cln2Δ double mutant ( Figure 2B , lane 6 ) , confirming that Whi5 phosphorylation depends on Cln-Cdc28 kinase complexes . Consistent with our SDL results and in vitro kinase assays , we observed a significant reduction in Whi5 mobility in extracts from a pho85 mutant strain ( Figure 2B , lane 2 ) . Thus , similar to Cdc28 , phosphorylation of Whi5 also depends on Pho85 in vivo . To determine if Whi5 physically associates with Pho85 in yeast , we first assayed Whi5FLAG immune complexes for kinase activity . A robust autophosphorylation activity was recovered from Whi5FLAG immunoprecipitates derived from wt cell extracts when radiolabeled ATP was added to the immunoprecipitated sample ( Figure 2C , lane 2 ) . This activity was partially dependent on both CDC28 and PHO85 ( Figure 2C , lanes 3–5 ) . We also confirmed a physical interaction between Whi5 and Pcls using a co-immunoprecipitation assay ( Figure 2D ) . Immunoprecipitation of Whi5MYC from epitope-tagged cyclin extracts revealed a specific association between Pcl9 and Whi5 ( Figure 2D , lane 4 ) . We failed to reproducibly detect a physical interaction between Whi5 and Pcl1 ( Figure 2D , lane 2 ) suggesting that Pcl9-Pho85 is the primary Whi5 CDK . Taken together , the phosphorylation and co-immunoprecipitation assays strongly suggest that , in addition to Cdc28 , Pho85 also phosphorylates Whi5 . Furthermore these results identify Whi5 as the first reported substrate for Pcl9-Pho85 , one of two Pcls whose activity is restricted to early G1 phase . Whi5 associates indirectly with G1 phase-regulated promoters through interaction with SBF and MBF . Interactions with these transcription factors and subsequent promoter binding are disrupted by CDK-dependent phosphorylation [12] , [13] . Because Whi5 appears to be a Pho85 substrate , we assessed the occupancy of SBF promoters by Pcl9 . To date , cyclins have not been detected at yeast promoters . Pcl9 is normally an unstable short-lived protein [27]; however , similar to other cyclins , Pcl9 turnover appears to be catalyzed in part by its cognate CDK , Pho85 ( Figure 3A ) [48] . Therefore , to test Pcl9 promoter localization in a more sensitive genetic background , we performed ChIP ( Chromatin immunoprecipitation ) experiments in a pho85Δ strain ( Figure 3B ) . The highest levels of CLN2 promoter DNA were detected in Pcl9MYC immune complexes 30 min following release from a metaphase-anaphase arrest ( Figure 3B ) . The Pcl9-chromatin association was no longer detectable 45 min after GAL-CDC20 induction indicating that the interaction is short-lived and transient as predicted for a regulator of Start . The association was Whi5-dependent since Pcl9 was not detected at the CLN2 promoter in a strain lacking Whi5 ( Figure 3C ) . The localization of Pcl9 to CLN2 , a G1 promoter , is consistent with a direct role for Pcl9-Pho85 in regulating G1 transcription . As mentioned above , cln3Δ mutants arrest in G1 phase as large unbudded cells in response to increased WHI5 dosage , indicating that Whi5 is a dose-dependent regulator of Start . Therefore , if Pho85 and Cdc28 function analogously to inhibit Whi5 activity , we predict that elevated Pho85 kinase activity would antagonize the toxic effects of WHI5 overexpression and suppress the growth defects observed in a cln3Δ mutant . To test this prediction , high copy plasmids expressing PCL1 , PCL2 , PCL9 , or PHO80 were introduced into a cln3Δ strain expressing WHI5 from a conditional MET25 promoter ( Figure 4A ) . Plasmid-based expression of Pcls and Whi5 was confirmed by immunoblotting ( Figure S1 ) . Induction of WHI5 expression in a cln3Δ mutant resulted in cell death whereas overexpression of PCL1 or PCL9 partially suppressed this toxicity and restored growth ( Figure 4A ) . Consistent with results from SDL analyses ( Figure 1 ) , this suppression was specific to PCL1 and PCL9 since neither PCL2 nor PHO80 were able to function effectively in the assay ( Figure 4A ) . Furthermore , PCL1/9-mediated suppression was dependent on phosphorylation since growth of a cln3Δ mutant expressing a nonphosphorylatable form of WHI5 ( Whi512A ) [13] could not be restored ( Figure 4A ) . These genetic results corroborate the biochemical evidence that Pcl-Pho85 regulates Whi5 activity through phosphorylation . Given its effect on WHI5 overexpression , we next examined PCL effects on other CLN3-associated phenotypes . CLN3 is required to activate G1-specific transcription once cells have achieved a critical size [49]–[51] . A cln3Δ mutant exhibits a large cell size phenotype because of its inability to inhibit Whi5 and activate Start-specific transcription [12] , [13] . Ectopic expression of PCL1 or PCL9 reduced cln3Δ cell size to an intermediate level between that of wt and cln3Δ cells ( Figure 4B ) . Conversely , deletion of PCL9 , PCL1 , and the partially redundant cyclin PCL2 resulted in a cell size increase ( Figure 4C ) . These results suggest that Pcl-Pho85 and Cln3-Cdc28 share a common role in cell cycle progression to regulate Whi5 activity and promote passage through Start . To determine if Pcl-Pho85 and Cln3-Cdc28 might function in parallel to regulate Start , we first tried to test whether pcl9Δ cln3Δ or pcl1Δ pcl9Δ cln3Δ strains showed any synthetic growth defects . As expected , no growth defects were observed , probably because of the redundant effects of other Pcls [27] . Unlike the Cdc28 cyclins , which shows distinct cell cycle expression patterns , most Pcls are expressed at all cell cycle stages [25] . We then examined the phenotype of a pho85Δ cln3Δ double mutant . Cells lacking cln3Δ are larger than wt cells but do not display overt defects in growth rate while pho85Δ mutants are slow growing ( Figure 5A ) . However , pho85Δcln3Δ double mutants exhibited a more pronounced growth defect compared to single mutants and analysis of DNA content revealed that the pho85Δ cln3Δ double mutant cells accumulated in G1 phase with predominantly unreplicated DNA ( Figure 5A ) . Importantly , deleting WHI5 overcame both the cell cycle progression and growth defects observed in the absence of both CLN3 and PHO85 . Notably , a pho85Δ cln3Δ whi5Δ triple mutant exhibited a growth rate similar to a cln3Δ single mutant indicating that Pcl-Pho85 and Cln3-Cdc28 function in separate yet converging pathways to regulate Whi5 function and , by extension , G1 cell cycle progression ( Figure 5A ) . These observations also hold true under liquid growth conditions as shown . WHI5-dependent suppression appears to be specific to the pho85Δ cln3Δ phenotype because WHI5 deletion was unable to rescue 53 additional synthetic lethal interactions involving PHO85 ( Table S1; D . Q . Huang and B . J . Andrews , unpublished data ) . Given that Whi5 represses SBF- and MBF-specific transcription , we asked whether PHO85 affects SBF-driven reporter gene expression . A reporter gene consisting of tandem SCB consensus element repeats fused upstream of the HIS3 coding region was constructed and integrated into wt , cln3Δ , and pho85Δ strains . Previous work has shown that this reporter provides a highly specific read-out for SBF-dependent transcription [13] , [52] . Growth on medium lacking histidine supplemented with 3-aminotriazole ( 3-AT ) was used to assess SBF transcriptional activity ( Figure 5B ) . Even though cells lacking PHO85 were moderately sensitive to higher concentration ( 5 mM ) of 3-AT ( unpublished data ) , both cln3Δ and pho85Δ mutants showed no growth in media containing 30 mM 3-AT indicating that SBF transcription is impaired in these mutants , whereas growth of wt cells was unaffected [13] . Furthermore , defects in SCB-driven gene expression were more pronounced in the pho85Δ cln3Δ double mutant ( at 10 mM 3-AT , Figure 5B ) . Consistent with the genetic interactions described above ( Figure 5A ) , SBF-dependent reporter activity was restored in pho85Δ cln3Δ mutants when WHI5 was deleted ( Figure 5B ) . However , WHI5 deletion only partially rescued the growth defect in pho85Δ cells at 30 mM of 3-AT ( Figure 5B ) . The Whi5-independent 3-AT sensitivity of pho85Δ cells may be due to unregulated Gcn4 in the absence of PHO85 , since GCN4 is induced by 3-AT and Pho85 has been shown to regulate Gcn4 stability [53] , [54] . Nonetheless , these data suggest that , like Cln3-Cdc28 , Pcl-Pho85 modulates SBF activity through Whi5 . We next interrogated the effects of CDK activity on Whi5-mediated transcriptional repression ( Figure 6 ) . A construct expressing a LexA DNA binding domain fused to WHI5 was introduced into a strain harboring a LacZ reporter gene containing LexA binding sites in its promoter ( Figure 6 ) . Consistent with its role as a negative regulator of G1-specific transcription , a ∼10-fold reduction in β-galactosidase activity was observed in cells expressing the LexA-Whi5 fusion protein compared to a vector control ( Figure 6 ) . Overexpression of PCL9 , CLN3 , or CLN2 restored LacZ expression to intermediate levels indicating that activation of either CDC28 or PHO85 was capable of antagonizing Whi5 function in this assay ( Figure 6 ) . Consistent with suppression of WHI5-mediated growth defects ( Figure 4 ) , inhibition of Whi5 activity was dependent on phosphorylation since LacZ expression could not be restored in cells harboring an unphosphorylatable LexA-Whi512A fusion protein ( Figure 6 ) . Cln2-Cdc28 activity was previously shown to disrupt recombinant Whi5-SBF complexes in vitro [13] , but Cln3-Cdc28 and Pho85 kinases had not been assessed for this activity . A preassembled recombinant Whi5-Swi4FLAG-Swi6 complex bound to anti-FLAG resin was incubated with purified kinases in the presence of radiolabeled ATP and separated into soluble ( Figure 7B , labeled “S” ) and bound fractions ( Figure 7B , labeled “B” ) . Equivalent amounts of kinase were approximated on the basis of in vitro kinase activity ( Figure 7A , and Materials and Methods ) . As expected , Cln2-Cdc28 phosphorylation caused most of the SBF-bound Whi5 to be released into the soluble fraction ( Figure 7B , lanes 3 and 4 ) . In contrast , we failed to observe dissociation of Whi5 from SBF in the presence of Cln3- or Pcl9-CDK complexes ( Figure 7B , lanes 5–10 ) . In addition to negatively regulating the interaction of Whi5 with SBF , Cdc28 also controls its localization [13] . Unlike Cln-Cdc28 phosphorylation , which promotes Whi5 export from the nucleus , deletion of PHO85 did not dramatically affect the subcellular localization of Whi5 ( Figure 7C ) . Together , these results suggest that Pho85 must regulate Whi5 function through alternate mechanisms . We next explored what additional mechanism might explain Pcl- and Cln3-mediated regulation of Whi5 activity . Functional conservation clearly extends to Whi5 and its metazoan analogue Rb [14] . Since Rb represses transcription , in part , through recruitment of HDACs , we used a batch affinity chromatography assay to test for physical interactions between a Whi5GST ligand and tandem affinity tagged HDACs ( Figure 8A ) . Specific interactions between Whi5 and Hos3 , Rpd3 , and , to a lesser extent , Hos1 were identified ( Figure 8A , lanes 1 , 5 , 13 ) suggesting that , like Rb , Whi5-dependent transcriptional repression involves recruitment of HDACs . This observation is consistent with previous work that detected Rpd3 at the PCL1 promoter using a ChIP assay [55] . Furthermore , HOS3 and RPD3 were required for WHI5 dose-dependent effects on cell size . Like wt cells , strains lacking either HOS3 ( Figure 8B , panel 1 ) or RPD3 ( Figure 8B , panel 2 ) also exhibited a dose-dependent increase in cell size in response to WHI5 overexpression . However , additional cell size effects were not observed in strains lacking both HDACs , suggesting that Hos3 and Rpd3 regulate Whi5 function synergistically ( Figure 8B , panel 3 ) . If HDACs are required for Whi5 function , then strains lacking HDAC function should be resistant to toxic effects associated with WHI5 overexpression . Consistent with this prediction , the growth defect caused by WHI5 overproduction in a cln3Δ was alleviated by the deletion of HOS3 and RPD3 ( Figure 9A ) . Deletion of HOS3 alone rescued WHI5 toxicity in a pho85Δ strain while a cln3Δ mutant required deletion of both HOS3 and RPD3 in order to tolerate increased dosage of WHI5 ( Figure 9A ) . Given that Whi5 appears to be acting through HDACs , we predicted that deletion of HOS3 and RPD3 should phenocopy those genetic interactions seen in whi5Δ mutants . We first tested various HDAC deletion strains for suppression of the slow growth phenotype of a pho85Δcln3Δ mutant . As for WHI5 , deletion of HOS3 and RPD3 partially suppressed the growth defect seen in the pho85Δcln3Δ double mutant strain ( Figure 9B ) . Suppression was specific to HOS3 and RPD3 because deletion of other HDACs showed no suppression , and the growth rate of the pho85Δcln3Δhos3Δ strain was not improved by subsequent deletion of RPD3 and vice versa ( Figure 9B ) . We next asked if deletion of HDACs might overcome the Start arrest seen in cells lacking both CLN3 and BCK2 , another regulator of G1 transcription that functions in parallel with CLN3 [56] . A cln3Δbck2Δwhi5Δ triple mutant grows as vigorously as wt , placing WHI5 downstream of both upstream activators of G1 transcription [13] . Interestingly , deletion of RPD3 partially restored growth in the cln3Δbck2Δ strain providing further evidence for an HDAC requirement in Whi5-mediated transcriptional repression ( Figure 9C ) . Neither subsequent deletion of HOS3 nor deletion of other HDACs affected growth appreciably ( Figure 9C ) . We also employed the SCB-HIS3 assays used above to explore SBF-driven reporter gene expression in the HDAC mutants ( Figure 10 ) . As expected , deletion of RPD3 rescued the growth defects of cln3Δ SCB-HIS3 cells in the presence of both 10 mM and 30 mM of 3-AT , whereas HOS3 gene knockout had a marginal but additive effect . In contrast , the growth of pho85Δ cells was slightly rescued by deletion of HOS3 but not RPD3 providing further evidence for Pho85 acting specifically through Hos3 . Because of difficulties in detecting HDACs at promoters , we were unable to confirm these observations in vivo . We also performed co-immunoprecipitation assays using affinity tagged RPD3 and HOS3 strains and observed an obvious decrease in Rpd3 and Hos3 in Whi5 precipitates from strains harboring increased levels of Pcl9 , Cln2 , or Cln3 cyclins ( Figure 11A and 11B ) . Together , our genetic and biochemical results suggest that Pho85 may preferentially influence Whi5-Hos3 activity , whereas Cln3-Cdc28 is required for inhibition of both Rpd3 and Hos3 .
Whi5 is a critical cell cycle regulator that links CDK activity in G1 phase to the broad transcriptional program that accompanies commitment to cell division . We provide substantial evidence that the multifunctional Pho85 CDK is an important regulator of Whi5 activity and G1 phase-specific transcription including: ( 1 ) Whi5 is phosphorylated and antagonized by Pho85 and is the first reported substrate for the G1-specific CDK complex , Pcl9-Pho85; ( 2 ) the activity of an SBF-dependent promoter is influenced by PHO85; ( 3 ) the Pcl9 cyclin binds to SBF-regulated promoters; ( 4 ) the repressor function of Whi5 is mediated through the HDACs Hos3 and Rpd3; and ( 5 ) HDAC-Whi5 association is regulated by G1-specific forms of both the Pho85 and Cdc28 CDKs . We therefore conclude that timely and efficient release from Whi5 inhibition and subsequent G1/S cell cycle progression requires the concerted activity of both Cdc28 and Pho85 . Several lines of evidence point to common roles for Pho85 and Cdc28 . For example , a burst of both G1-specific Cdc28 and Pho85 activity is essential for cellular morphogenesis . A strain lacking the G1-specific cyclins , CLN1 , CLN2 , PCL1 , and PCL2 , undergoes a catastrophic morphogenic change and fails to establish polarized cell growth and cytokinesis [30] . Consistent with these observations , a chemical genomic analysis demonstrated that expression of genes involved in polarized cell growth was sensitive to simultaneous inhibition of both kinases , but not either single kinase [31] . A functional connection between Pho85 and Cdc28 is further supported by independent genetic and biochemical analyses that identify common targets phosphorylated by both kinases [45] , [46] , [48] , [57]–[59] . Despite the clear functional overlap for G1-specific forms of Cdc28 and Pho85 in controlling morphogenesis , up to now , a direct role for Pho85 in cell cycle commitment and G1 phase-specific transcription has remained unclear . We discovered that , like Cdc28 , Pho85 activates G1 transcription through inhibition of the Whi5 repressor . While the two kinases collaborate to control certain facets of Whi5 regulation , they are also specialized to modulate Whi5 function by distinct mechanisms . We have defined a novel HDAC-dependent mechanism that impinges on Whi5 function and implicates both Pho85 and Cdc28 as regulators of this process . On the basis of these and other observations , we propose that Whi5 functional regulation involves perturbation of specific HDAC-Whi5 interactions and requires the concerted activity of both Cdc28 and Pho85 ( summarized in Figure 12 ) . Interestingly , our genetic observations support a model whereby Pcl-Pho85 preferentially targets the Hos3-Whi5 interaction illustrating a functional distinction between the two CDKs . While Pho85 associates with several cyclin subunits , only Pcl9 exhibits temporal expression and localization patterns compatible with such a function . PCL9 is expressed at the M/G1 phase transition and encodes a short-lived protein localized exclusively to the nucleus in early G1 phase [27] , [60] , [61] . Cln3 is also present in early G1 cells , but shows a complex localization pattern , with significant retention to the ER in early G1 cells , followed by chaperone-mediated release into the nucleus in late G1 phase [62] . How the specific features of Pcl9 and Cln3 localization might influence the timing of HDAC inhibition remains to be explored . The second component of Whi5 regulation is predicated on previous studies indicating that G1/S gene expression is preceded by Whi5-SBF complex dissociation and subsequent nuclear export of Whi5 ( Figure 12 ) [13] . Unlike early regulatory events , Cdc28 activity is both necessary and sufficient to drive these events since neither SBF binding to Whi5 nor nuclear localization of Whi5 was adversely affected in a pho85Δ mutant ( Figure 7 ) . Also , we are able to detect binding of SBF in vivo to CLN2 promoters when PHO85 is deleted ( Figure 3C ) . However , both purified Cln3-Cdc28 and Pcl9-Pho85 failed to affect Whi5-SBF stability in vitro , while complex disruption was effectively achieved in the presence of Cln2-Cdc28 kinases ( Figure 7 ) . Cln3-Cdc28 and Pcl9-Pho85 may have a more pronounced effect on the Whi5-SBF complex in vivo . Alternatively , Cln3- and Pcl9-CDKs may act primarily as agonists of HDAC interactions while physical interactions with SBF and nuclear export are optimally mediated by the late G1 CDKs , Cln1- and Cln2-Cdc28 . Indeed , recent work reveals activation of CLN2 expression while Whi5 remains bound to the promoter ( H . Wang , L . B . Carey , Y . Cai , H . Wijnen , and B . Futcher , personal communication ) . Such a mechanism may serve to sharpen the onset , as opposed to the timing , of G1/S gene expression thus ensuring a sustained transcriptional burst and irreversible commitment to cell division [13] . Consistent with this idea , recent analysis of cyclin gene expression using a single cell assay affirms that positive feedback involving the Cln1 and Cln2 cyclins induces the G1/S regulon , and that this regulatory feedback is important for maintaining coherence of gene expression at Start [63] . SBF promoter recruitment depends on a series of well-organized chromatin remodeling events [7] , [36] . SBF , in turn , regulates the recruitment of the general transcription machinery via a two-step process beginning with the mediator complex followed by CDK-dependent recruitment of RNA PolII , TFIIB , and TFIIH [9] . Previous studies suggested that this CDK requirement stems from Whi5 , which in its unphosphorylated state , remains bound to SBF and occludes the basal transcription machinery from binding specific promoters [13] . We have extended this model to include a role for HDAC activity . We predict that Hos3 and Rpd3 contribute to Whi5 repression by preventing holoenzyme access to chromatin . During states of high CDK activity , Cdc28 and Pho85 abrogate Whi5-HDAC and Whi5-SBF interactions and initiate transcription . Consistent with our model , Pcl9 and Cln3 cyclins localize to G1 promoters and Whi5 remains associated with G1-specific promoters in the absence of HDAC-promoter interactions ( Figure 3; H . Wang , L . B . Carey , Y . Cai , H . Wijnen and B . Futcher , personal communication ) . However , Whi5 may also repress transcription by additional mechanisms since its activity is partially retained in hos3Δ rpd3Δ mutants ( Figure 9 ) . Rpd3 is a well-characterized HDAC that accomplishes most of its functions as part of a large protein complex [37] . The Rpd3-Sin3 deacetylase complex has long been implicated as a cell cycle regulator required for silencing HO gene expression to prevent mating type switching in newly budded cells [64] , [65] . Our observations that Whi5 associates with Rpd3 and our genetic data linking G1 Cdks , Whi5 , and Rpd3 reveal a more general role for Rpd3 in G1/S-phase specific transcription . These data are consistent with observations from Futcher and colleagues that the Rpd3 protein can be detected at the CLN2 promoter and that the amount of Rpd3 at the promoter is decreased when CLN3 is induced ( H . Wang , L . B . Carey , Y . Cai , H . Wijnen and B . Futcher , personal communication ) . The Rpd3-Sin3 HDAC has also been connected to G1 transcription factors through the interaction of Sin3 with Stb1 , a Swi6-binding protein [66]–[68] . Both Stb1 and Sin3 are required for repression of G1 transcription early in G1 phase [68] . Unlike Rpd3 , Hos3 is largely uncharacterized , although a recent study suggests a role for Hos3 in yeast apoptosis upon exposure to oxidative radicals [69] . We have uncovered an additional role for Hos3 in Whi5-mediated transcriptional repression . A question that arises from our observations is what advantage does combinatorial kinase regulation impart on specific biological processes such as G1/S cell cycle progression ? Contributions from multiple CDKs may provide the precision and accuracy necessary for rapid definitive decisions that irreversibly affect cellular fate . Indeed , distributive multisite phosphorylation mechanisms exhibit ultrasensitivity with respect to kinase concentration , thereby creating a “switch-like” behavior in biological circuits [70] . Since cell cycle transitions typically display switch-like attributes , multisite phosphorylation by various kinase combinations may prove to be a rule rather than the exception amongst CDK targets , including key cell cycle regulators such as Whi5 . In fact , a recent computational analysis showed enrichment of multiple closely spaced consensus sites for Cdc28 substrates in yeast , a pattern that proved predictive of likely CDK targets [71] . Although kinase combinations are likely necessary for cell cycle regulation , the contribution of each individual kinase may vary depending on specific signals and environmental stimuli . In certain environments , Pcl-Pho85 may have more dramatic , condition-specific effects on Whi5 function than Cdc28 analogous , perhaps , to the regulation of Rb that is required for quiescence and prevention of apoptosis [72] , [73] . Previous studies indicate that Whi5 localizes to nuclei in stationary phase cells suggesting that Whi5 may also play a role in G0 [13] . Interestingly , Pho85 is required for survival in starvation conditions and plays an important role during stationary phase [74]–[76] . Furthermore , CDK5 , the mammalian Pho85 homolog , induces apoptosis in neuronal cells via Rb phosphorylation [77] . Whether Whi5 activity is more prominently affected by Pcl-Pho85 in response to stationary or stress conditions requires additional investigation . Similarities between metazoan and yeast cell cycle regulation are increasingly evident as we continue to characterize Whi5 function . For example , similar to proposed Pcl9/Cln3 “early” phase regulation ( Figure 12 ) , cyclinD-CDK4/6 phosphorylates Rb to promote HDAC dissociation and E2F transcriptional activation . E2F activation then leads to cyclin E expression , which , similar to Cln1/2 “late” phase regulation ( Figure 12 ) , may establish a positive feedback loop whereby cyclinE-CDK2 activity disrupts Rb-promoter interactions and stimulates G1-transcription further [15] . Despite these similarities , the importance of multiple regulatory components in both yeast and mammalian systems remains poorly understood and may be most fruitfully dissected using the yeast model .
The S . cerevisiae strains used are listed in Table 1 . All gene disruptions and integrations were achieved by homologous recombination at their chromosomal loci by standard PCR-based methods and confirmed by PCR with flanking primers [78] . Standard methods and media were used for yeast growth and transformation . Two percent of galactose in the media was used to induce the expression of genes under the GAL1 promoter . Synthetic minimal medium with appropriate amino acid supplements was used for cells containing plasmids . Appropriate amounts of 3-AT were added to SD-HIS plates to assess the expression of HIS3 reporter genes . 10-fold serial dilutions ( 5–10 µl ) of yeast cells were spotted onto plates with appropriate nutrition conditions to assess growth . Plasmids used in this study are listed in Table 2 . In most cases , a DNA insert was amplified by PCR and inserted into a linearized vector by homologous recombination in yeast . Details of construction will be provided upon request . The in vitro protein kinase assays monitored the incorporation of [32P] transferred from γ-32P-ATP to purified recombinant GST-Whi5 . The reaction mixture for assays shown in Figure 2A contained 50 mM Tris-HCl ( pH 7 . 5 ) , 1 mM DTT , 10 mM MgCl2 , and 1 µM ATP ( including 20 µCi γ-32P-ATP ) and 0 . 2 µg GST-Whi5 in 20 µl of total volume . 2 µl of a purified recombinant kinase ( 0 . 4 µg–0 . 8 µg ) was added to the mixture and incubated at 30°C for 30 min . Purification of Cln and Pcl CDKs from insect cell expression systems have been previously described [13] , [46] . Whi5 was then analyzed by SDS-PAGE and autoradiography . Kinase assays on immunoprecipitated proteins from yeast cell extracts were performed as described [13] . Kinase assays preceding the Whi5-SBF dissociation assay ( Figure 7 ) were performed as described above except that 200 µM γ-32P-ATP was used instead of 1 µM . The final concentration of Cln3 and Pcl9 was 3 µM , and the final concentration of Cln2 was 60 nM ( 50-fold less ) . Liquid β-galactosidase assays were performed as described [29] . Strains carrying appropriate plasmids were grown in synthetic minimal medium to mid-log phase , transferred to synthetic galactose medium , and incubated for 4 h . Cells were harvested and broken in lysis buffer ( 100 mM Tris-HCl [pH 8 . 0] , 1 mM DTT , and 20% glycerol with protease inhibitors ) with glass beads . The β-galactosidase activity was determined by adding 100 µl of total cell extract to 0 . 9 ml of Z buffer ( 100 mM Na2PO4 , 40 mM NaHPO4 , 10 mM KCl , 1 mM MgSO4 , and 0 . 027% β-mercaptoethanol ) and 200 µl ONPG ( 4 mg/ml ) ( Sigma ) . Units of β-galactosidase activity were determined as described [29] . The protein binding assay essentially followed the procedures described previously [13] . Briefly , 1 µl of insect cell lysate expressing SBF ( Swi6-Swi4FLAG ) was mixed with 1 µl of purified GST-Whi5 ( ∼0 . 1 µg ) and 7 µl of M2 anti-FLAG resin ( Sigma ) in 8 µl of kinase buffer ( 50 mM Tris-HCl [pH 7 . 5] , 1 mM DTT , and 10 mM MgCl2 ) . The mixture was incubated at 4°C for 1 h with mixing . The beads bound to the SBF-Whi5 complex were then washed three times with kinase buffer , and mixed with various cyclin dependent kinases in kinase buffer with 0 . 2 mM ATP in a 20 µl volume . The kinase reaction was incubated at 30°C for 1 h . The soluble portion was taken out and mixed with 20 µl of 2×SDS-PAGE loading buffer . The beads in the tube were washed three times with kinase buffer before mixing with 15 µl of 2×SDS-PAGE loading buffer . Strains containing galactose-inducible plasmids were grown to saturation in 2% raffinose media for 48 h . Expression of plasmids were induced by transferring into 2% raffinose 2% galactose media and liquid growth assays were performed as previously described over 36 h using a Tecan GENios microplate reader ( Tecan ) [79] . Average doubling ( AveG ) for each culture was calculated as previously described [79] . Growth rate for each mutant was calculated relative to the AvgG of the wt strain . The localization of Whi5-GFP was monitored in wt , cdc28-4 , and pho85Δ strains . Cells expressing pMET-GFP-WHI5 were grown to log phase in synthetic glucose medium without methionine . Cells were observed at a magnification of 1 , 000× using Nomarski optics and fluorescence microscopy and photographed by a Cascade 512B high-speed digital camera ( Roeper Scientific ) mounted on a Leica DM-LB microscope . Images were captured and analyzed by MetaMorph software ( Universal Imaging Media ) . The pho85Δ PCL9MYC GALpr-CDC20 and pho85Δwhi5ΔPCL9MYC GALpr-CDC20 cells were grown in YP-Galactose ( YPG ) medium to an optical density ( OD600 ) of 0 . 4 , blocked at M phase by growing in YPED medium for 3 h , and released into YPG medium . Samples were taken every 15 min after release and cross-linked with a final concentration of 1% formaldehyde . Wt and swi4Δ strains ( for controls ) were grown to OD600 of 0 . 6 in YPD . Formaldehyde cross-linking and preparation of whole-cell extracts were performed as previously described [80] . Immunoprecipitation were performed using 1∶200 dilution of α-myc monoclonal antibody ( 9E10 ) , α-Swi6 or α-Swi4 polyclonal antibodies . The precipitates were washed twice with lysis buffer , once with LiCl detergent and once with Tris-buffered saline and processed for DNA purification . Enrichment at the CLN2 promoter sequence was quantified with real-time PCR , using a dual fluorogenic reporter TaqMan assay in an ABI PRISM 7500HT Sequence Detection System as previously described [13] . Recombinant GST-Pcl1 and GST-Whi5 were produced in a BL21 bacterial expression strain; other recombinant proteins were produced in insect cells infected with Baculovirus expression vectors [11] , [46] , [73] . Proteins were detected with 9E10 anti-Myc , 12C5 anti-HA , and M2 anti-FLAG monoclonal antibodies . FACS analysis of DNA content and cell size measurements were described previously [81] . | Eukaryotic cells grow and divide by progressing through carefully orchestrated stages of the cell cycle characterized by stage-specific patterns of gene expression , DNA replication , and scission . How stage-specific gene expression is coordinated with cell cycle progression is only partially understood . The phase known as G1 marks the initiation of the cell cycle ( called START in yeast ) and involves the coordinated expression of more than 200 genes regulated by two transcription factors , SBF and MBF . The activity of SBF and MBF is restrained by binding of the repressor protein Whi5 to the two transcription factors early in G1 phase . Phosphorylation of Whi5 by G1-specific forms of the cyclin-dependent kinase ( CDK ) Cdc28 promotes dissociation of Whi5 from SBF and its export from the nucleus; this , in turn , releases SBF to activate G1-specific transcription . This G1 transcriptional circuit is analogous to that defined in mammals by the E2F family of transcription factors and the retinoblastoma ( Rb ) tumor suppressor protein . Rb further contributes to the repression of G1-specific transcription in mammals by recruiting histone deacetylases ( HDACs ) , which are chromatin remodeling complexes that regulate promoter accessibility . Here , we show that regulation of G1-specific transcription in yeast also involves repressor-mediated recruitment of HDACs . We demonstrate that repression by Whi5 is modulated by both Cln-Cdc28 and a second G1-specific CDK , Pcl-Pho85 , and further show that both kinases regulate the interaction of Whi5 with HDACs . We propose that regulation of the repressor by more than one G1-specific CDK ensures definitive inactivation of Whi5 , a critical event for appropriate cell cycle initiation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/gene",
"expression",
"biochemistry/transcription",
"and",
"translation",
"genetics",
"and",
"genomics/epigenetics",
"genetics",
"and",
"genomics",
"cell",
"biology/gene",
"expression"
] | 2009 | Dual Regulation by Pairs of Cyclin-Dependent Protein Kinases and Histone Deacetylases Controls G1 Transcription in Budding Yeast |
The filamentous fungus Magnaporthe oryzae is the causal agent of rice blast disease . Here we show that glycogen metabolic genes play an important role in plant infection by M . oryzae . Targeted deletion of AGL1 and GPH1 , which encode amyloglucosidase and glycogen phosphorylase , respectively , prevented mobilisation of glycogen stores during appressorium development and caused a significant reduction in the ability of M . oryzae to cause rice blast disease . By contrast , targeted mutation of GSN1 , which encodes glycogen synthase , significantly reduced the synthesis of intracellular glycogen , but had no effect on fungal pathogenicity . We found that loss of AGL1 and GPH1 led to a reduction in expression of TPS1 and TPS3 , which encode components of the trehalose-6-phosphate synthase complex , that acts as a genetic switch in M . oryzae . Tps1 responds to glucose-6-phosphate levels and the balance of NADP/NADPH to regulate virulence-associated gene expression , in association with Nmr transcriptional inhibitors . We show that deletion of the NMR3 transcriptional inhibitor gene partially restores virulence to a Δagl1Δgph1 mutant , suggesting that glycogen metabolic genes are necessary for operation of the NADPH-dependent genetic switch in M . oryzae .
Rice blast disease is the most serious disease of cultivated rice and in recent years has caused epidemics in South Korea , Japan , Bhutan and China [1] , [2] , resulting in severe harvest losses . Understanding the biology of rice blast disease is therefore important , if durable control strategies for the disease are to be developed [2] . The rice blast fungus , Magnaporthe oryzae , brings about infection of rice leaves by forming a specialized infection structure called an appressorium [2] . This is a unicellular , dome-shaped structure that develops from the end of a fungal germ tube and is responsible for mechanically breaching the rice cuticle , allowing the fungus entry to plant cells . Appressorium development is cell cycle-regulated in M . oryzae with checkpoints regulating initiation and maturation of the appressorium [3] , [4] . Differentiation of the appressorium is accompanied by autophagy in the conidium leading to programmed cell death and mobilisation of the contents of the three-celled spore to the infection cell . Prevention of autophagy by deletion of any of the core genes associated with non-selective macroautophagy , renders the fungus non-pathogenic , demonstrating that re-cycling of the contents of the conidium is essential for the appressorium to function correctly [3] , [5] . The enormous turgor generated by the M . oryzae appressorium is the result of glycerol accumulation , which acts as a compatible solute , causing influx of water into the cell to create hydrostatic pressure [6] . Efflux of glycerol is prevented by a layer of melanin in the appressorium cell wall and mutants unable to synthesize melanin cannot generate turgor and are consequently non-pathogenic . Previously , glycogen reserves and lipid bodies were shown to move from the conidium to the M . oryzae appressorium prior to turgor generation [7]–[9] . This process is controlled by the Pmk1 MAP kinase pathway , which regulates appressorium morphogenesis [10] and is likely to be linked to the onset of autophagy in the conidium [3] . Lipid and glycogen breakdown in the appressorium are controlled by the cAMP response pathway and cpkA mutants , which lack protein kinase A activity , show significant delays in lipid and glycogen breakdown [7] . The rapid changes in primary metabolism during appressorium maturation appear to be regulated in part by a trehalose-6-phosphate synthase ( Tps ) -mediated genetic switch , which responds to levels of glucose-6-phosphate ( G6P ) and the NADPH/NADP balance in cells [11] . The Tps-mediated gene switch interacts with three transcriptional inhibitors which regulate virulence-associated gene expression in response to prevailing metabolic conditions [11] . In this study , we investigated the role of glycogen metabolism in the function of the M . oryzae appressorium . We show that glycogen reserves in the spore are broken down rapidly during spore germination , in a process regulated by the cAMP response pathway . We demonstrate that the GPH1 glycogen phosphorylase and AGL1 amyloglucosidase genes , which encode enzymes required for cytosolic glycogen breakdown , are virulence factors involved in plant infection . Surprisingly , however , we also show that glycogen synthase , which is encoded by the GSN1 gene in M . oryzae , is dispensable for pathogenicity . To explain this apparent paradox , we provide evidence that loss of amyloglucosidase or glycogen phosphorylase activity in M . oryzae leads to a reduction in the expression of TPS1 , thereby affecting G6P/NADPH-dependent gene regulation [11]–[13] . Consistent with this idea , we show that deletion of NMR3 , one of the transcriptional inhibitors associated with Tps1 , partially restores virulence to a Δgph1Δagl1 mutant . Our results suggest that glycogen breakdown in the appressorium is a significant factor in regulating virulence-associated gene expression .
To investigate the role of glycogen mobilisation during appressorium formation , we first studied the cellular distribution of glycogen granules in a M . oryzae wild-type strain , Guy-11 and regulatory mutants affected in appressorium morphogenesis . In Guy-11 , un-germinated conidia ( 0 h incubation ) were glycogen-rich , indicated by a dark precipitate in each of the three conidial cells after incubation in iodine solution ( Figure 1A ) , as previously described [7] . During germination and early appressorium formation ( 2–4 h ) , glycogen was degraded , with residual glycogen located only within the central cell of the conidium . After 6 h germination , glycogen appeared in the nascent appressorium , but was rapidly depleted during appressorium maturation , until at 24 h only the dark melanin ring around the appressorium and small number of glycogen grains were visible ( Figure 1A , [7] ) . In order to investigate the role of the cAMP pathway in glycogen mobilisation , we investigated a ΔcpkA mutant ( Figure 1B ) . CPKA encodes the catalytic subunit of cAMP-dependent protein kinase A [14] , [15] . Conidia were initially rich in glycogen , but glycogen degradation in ΔcpkA mutants was delayed with glycogen still present in conidia at 8 h . Glycogen was then deposited in misshapen ΔcpkA appressoria , but these cells did not fully develop and glycogen failed to be degraded ( Figure 1B ) . We conclude that the cAMP pathway regulates efficient degradation of glycogen during germination and appressorium formation . Consistent with this idea , the cAMP bypass suppressor mutant Δmac1 sum1-99 [16] , exhibits accelerated glycogen breakdown in appressoria , as previously reported [7] . To determine whether glycogen mobilisation is associated with appressorium development , Guy-11 conidia were germinated under high nutrient conditions ( 2% yeast extract ) which suppress formation of appressoria [16] . Limited degradation of glycogen deposits was observed and after 48 h glycogen granules could be seen within branched germ tubes ( Figure 1C ) . Mobilisation of glycogen from the conidium is therefore associated with appressorium morphogenesis in nutrient-free conditions . We next developed an assay to allow quantification of glycogen reserves and found that un-germinated Guy11 conidia contain around 1200 pg of glycogen ( Figure 1D ) . We observed 1500 pg glycogen germling−1 after 2 h germination , during germ tube emergence , but between 2 and 12 h after germination , the amount of glycogen fell to 500 pg germling−1 and remained at this level during appressorium maturation , representing an average decrease of 988 pg with 95% family wise confidence interval ( 667 , 1310 ) , p<0 . 001 . Over the entire 48 hour time course average glycogen levels dropped by 751 pg ( 95% CI 1073 , 430 p<0 . 001 ) . To compare these results with previous measurements of glycogen levels in budding yeast Saccharomyces cerevisiae [17] , they must be converted to femto moles ( fmol ) per cell . The number of M . oryzae cells can only be easily defined at 0 h ( 3-celled conidium ) and 48 h ( one-celled appressorium ) . An un-germinated M . oryzae conidium therefore contains an average of 2363 fmol glycogen cell−1 , while an appressorium contains 2450 fmol glycogen cell−1 . The significant glycogen storage capacity of wild-type M . oryzae cells is emphasised when these values are compared to the 19 . 5 fmol glycogen measured in an S . cerevisiae yeast cell [17] . The ΔcpkA mutant contained lower levels of glycogen than Guy11 ( Figure 1E ) . At 0 h we detected 800 pg glycogen germling−1 . There were no significant changes in ΔcpkA glycogen levels when considered over the entire period of the time course ( overall average decrease 13 pg [95% CI 354 , 328 p = 1] ) . In the cAMP-independent PKA mutant strain Δmac1 sum1-99 the glycogen assay revealed initially high levels of glycogen of 1700 pg germling−1 ( Figure 1F ) . This continued to increase to a peak at 2100 pg germling−1 after 4 h ( average increase from 0 h to 4 h of 450 pg [95% CI −265 , 1164 p = 0 . 545] ) . Between 4 and 6 h after germination , at the time of appressorium formation , glycogen levels decreased rapidly by an average of 986 pg germling−1 ( 95% CI 1700 , 272 p<0 . 001 ) . Levels appeared to rise again until 12 h , peaking at the lower value of 1800 pg germling−1 ( average increase 646 pg [95% CI −152 , 1445 p = 0 . 2156] . Glycogen then decreased gradually by an average of 1340 pg over the remaining 36 hours ( 95% CI 2139 , 541 p<0 . 001 ) to a final value of approximately 400 pg germling−1 by 48 h . Over the entire time course glycogen levels in the Δmac1 sum1-99 mutant decreased by an average 1230 pg ( 95% CI 1944 , 516 p<0 . 001 ) . We conclude that the cAMP response pathway regulates glycogen mobilisation during appressorium development in M . oryzae . To determine whether glycogen mobilisation is necessary for pathogenicity of M . oryzae we identified the major enzymes associated with its breakdown . Glycogen is comprised of chains of glucose linked by α-1 , 4-glycosidic bonds and at approximately every tenth residue , a branch point is formed by an α-1 , 6-glycosidic bond . Amyloglucosidase ( glycogen de-branching enzyme ) removes α-1 , 6-glycosidic bonds from the highly branched polymer , to release glucose . A putative amyloglucosidase-encoding gene , AGL1 ( MGG_00063 . 6 ) , was identified in the genome sequence of M . oryzae [18] . AGL1 has a 4749 bp open reading frame interrupted by two introns of 134 bp and 91 bp and encodes a putative 1583 aa protein . Alignment of the potential M . oryzae Agl1 protein with amyloglucosidases from other organisms ( Figure S1 ) revealed 72% identity to a hypothetical protein from Neurospora crassa and 47% identity to the glycogen de-branching enzyme Gdb1 of S . cerevisiae [19] . Further analysis of the protein sequence revealed four conserved stretches of amino acids present in the α/β barrel domain of all members of the α-amylase superfamily ( data not shown ) . Breakdown of glycogen also requires the action of glycogen phosphorylase . This enzyme complements the α-1 , 6-glycosidic activity of amyloglucosidase , by attacking the exoglycosidic α-1 , 4-glycogen bonds of glycogen , cleaving and phosphorylating sequentially from non-reducing ends of the polymer . The M . oryzae GPH1 ( MGG_01819 . 6 ) gene was identified and revealed a 2664 bp ORF encoding a putative 888 aa protein . Two in-frame ATG start sites were present in the sequence , the most 5′ of these has the start site indicator nucleotide , A , at −3 bp [20] which is likely to be the translation initiation site for GPH1 [21] . The protein showed 78% identity to a hypothetical protein in N . crassa and 74% identity to a putative glycogen phosphorylase from Aspergillus fumigatus ( Figure S2 ) . In S . cerevisiae a phosphorylation site is present in the N-terminal region of the protein [22] . The M . oryzae protein shows a high degree of similarity to this sequence and the threonine residue , at which phosphate is bound in the yeast glycogen phosphorylase , is conserved . In addition , all glycogen phosphorylases possess a binding site for pyridoxal-5′-phosphate , a co-factor involved in phosphorylation . The aldehyde group of this pyridoxine derivative forms a Schiff base with a specific lysine side chain of glycogen phosphorylase [23] . The amino acid sequence surrounding this site has been identified in yeast glycogen phosphorylase [22] and is highly conserved in M . oryzae GPH1 with 17 out of 18 residues identical between the two organisms ( data not shown ) . The 3021 bp genomic DNA sequence of M . oryzae GPH1 was also identified from the published genome sequence [18] . Comparison of this genomic sequence with the GPH1 cDNA sequence identified the presence of four introns . To examine the role of AGL1 and GPH1 in pathogenicity of M . oryzae , we performed targeted gene replacement ( Figure S3A ) . A 2 . 8 kb HindIII fragment ( Figure S3B ) , containing the majority of the 5′ region of AGL1 , was removed and replaced with the hygromycin phosphotransferase cassette [24] conferring resistance to hygromycin B . The resulting construct ( Figure S3B ) was introduced into Guy-11 and transformants verified by DNA gel blot ( Figure S3C ) . To delete GPH1 , a 2 . 2 kb portion of the 3 kb coding sequence was subsequently replaced with the 1 . 4 kb hygromycin phosphotransferase resistance cassette ( Figure S3D ) . The resulting cassette was introduced into Guy-11 and transformants selected ( Figure S3E ) . Deletion of each gene was confirmed by measurement of Agl1 and Gph1 enzyme activities ( Figure 2A and B ) . Amyloglucosidase activity was not detectable in the Δagl1 strain and glycogen phosphorylase activity was not detected in Δgph1 mutants . We also observed , however , that amyloglucosidase activity could not be detected in Δgph1 deletion strains . Amyloglucosidase cleaves glycogen branch points only when they have been made accessible by the action of glycogen phosphorylase and was therefore not measurable in this assay . To generate a Δagl1Δgph1 double mutant , we carried out targeted deletion of AGL1 in a Δgph1 mutant using the split marker gene deletion method [25] . Gene deletion constructs were transformed into Δgph1 using an allele of acetolactate synthase gene conferring resistance to sulfonylurea . The resulting transformants were selected and analysed by DNA gel blots ( Figure S4 ) . Additionally , RT-PCR was carried out to confirm successful gene deletion . In Guy11 and Δgph1 , a 789 bp amplicon corresponding to the Agl1 transcript was amplified but this was not detected in Δagl1 mutants or the Δagl1Δgph1 double mutant ( Figure S4D ) . To investigate whether cellular distribution of glycogen is affected in mutants carrying targeted deletions of AGL1 and GPH1 , we analysed glycogen mobilisation during appressorium formation in Δagl1 , Δgph1 and Δagl1Δgph1 mutants ( Figure 3 ) . Mobilisation of glycogen was retarded in Δagl1 and Δagl1Δgph1 mutants and glycogen was not depleted from conidia until 24 h compared with 8–12 h in Guy11 . By contrast , Δgph1 mutants appeared to less compromised in glycogen mobilisation . These results suggest that amyloglucosidase is more important for efficient degradation of glycogen during infection-related development than glycogen phosphorylase . This may reflect the ability of a recently described glucoamylase to perform the α-1 , 4-glycosidic activity of glycogen phosphorylase and thus potentially compensate for loss of this activity [25] . Glycogen phosphorylase is important for mycelial glycogen breakdown , however , as glycogen accumulated in Δgph1 strains during growth on minimal medium ( Figure 3B ) , but not in Δagl1 strains . To determine whether Δagl1 , Δgph1 and Δagl1Δgph1 mutants are able to utilize external glycogen as sole carbon source , we carried out growth tests on a variety of carbon sources . Both Δagl1 and Δgph1 mutants , together with the Δagl1Δgph1 double mutant , were able to utilize glucose , but were compromised in growth on glycogen and starch ( Figure 3C ) . Re-introduction of AGL1 and GPH1 into corresponding null mutants complemented the mutant phenotype ( Figure S5 ) . We conclude that Δagl1 and Δgph1 mutants are impaired in their ability to utilise both starch and glycogen as sole carbon source . To determine the sub-cellular location of Agl1p and Gph1p during appressorium formation , AGL1:GFP and GPH1:GFP gene fusions were expressed in Guy11 . Agl1-GFP and Gph1-GFP fusions localized throughout the cytoplasm in conidia of Guy11 and within germ tubes and nascent appressoria by 2 h and 4 h , respectively ( Figure 4A ) . After 8 h , the fluorescent signal was observed predominantly in the developing appressorium and by 24 h , Agl1-GFP and Gph1-GFP were found exclusively in the cytoplasm of the appressorium . Interestingly , Agl1-Gfp and Gph1-Gfp were also highly expressed during plant infection after 36 h and localized to the cytoplasm of the intracellular invasive fungal hyphae ( Figure 4B ) . To determine the role of AGL1 and GPH1 in fungal pathogenicity , we inoculated the susceptible rice cultivar , CO-39 , and barley cultivar , Golden Promise , with uniform spore suspensions of each mutant and Guy-11 ( Figure 5A ) . The Δagl1 and Δgph1 mutants showed lesion numbers reduced by 50 . 44±6% ( P<0 . 0001 ) and 44 . 7±2 . 4% ( P<0 . 001 ) , respectively , whereas the Δagl1Δgph1 strain was reduced in lesion number by 75 . 44±1 . 4% ( P<0 . 0001 ) ( Figure 5A ) . The mutants showed no defects in conidial germination and appressorium formation under nutrient-free conditions and the frequency of appressorium formation was also unaffected ( data not shown ) . It has previously been proposed that glycogen breakdown may be significant in appressorium turgor generation [6] , [7] , [26] . We therefore measured appressorium turgor in the Δagl1 and Δgph1 mutant strains using an incipient cytorrhysis assay [6] . Despite differences in glycogen mobilisation in the mutant strains ( Figure 3 ) , we found no significant differences in turgor between appressoria of the mutant strains and Guy-11 ( not shown ) . We then used an onion epidermis assay [27] to determine whether the observed reduction in virulence could be a consequence of impaired appressorium function . However , no significant differences in epidermal penetration were observed between Guy-11 and the Δagl1 , Δgph1 and Δagl1Δgph1 mutant strains which elaborated penetration pegs normally ( not shown ) . Therefore , the reduction in lesion formation seen for Δagl1 , Δgph1 and Δagl1Δgph1 is not a consequence of reduced turgor pressure or impaired appressorium function . To investigate tissue colonisation by each mutant , a rice leaf sheath assay was performed [28] . We found that Δagl1 , Δgph1 and Δagl1Δgph1 mutants were defective in their ability to grow invasively inside rice cells . Glycogen metabolic mutants formed unusual invasive hyphae that were less bulbous and branched than normal ( Figure 5B ) . We conclude that Δagl1 , Δgph1 and Δagl1Δgph1 mutants are able to penetrate rice cells normally but cannot proliferate within rice tissue effectively , leading to reduced disease symptoms . Given the significance of glycogen mobilisation to virulence , we set out to make a strain of M . oryzae that was depleted in glycogen reserves . A putative glycogen synthase-encoding gene was identified from the M . oryzae genome database ( http://www . broadinstitute . org/annotation/fungi/magnaporthe/ ) using the BLASTX algorithm [29] based on its predicted amino acid homology to yeast and mammalian glycogen synthases . One sequence , MGG_07289 . 6 aligned with yeast and mammalian glycogen synthases [30] , [31] . The predicted glycogen synthase-encoding gene , which we named GSN1 , is present as a single copy gene in the M . oryzae genome and predicted to contain five exons , interrupted by four introns [29] . The predicted gene encodes a 708 amino acid protein and showed high sequence similarity to Gsy1p and Gsy2p of yeast [30] and Gys1p and Gys2p of the human muscle and liver specific forms of glycogen synthase [31] . Figure S6 shows the result of a ClustalW alignment [32] . S . cerevisiae Gsy1 and Gsy2 showed 63% and 64% identity to M . oryzae GSN1 , respectively whereas Homo sapiens Gys1 and Gys2 showed 60% and 55% identity , respectively . Phosphorylation sites putatively involved in post-translational regulation of Gsy2p were conserved in M . oryzae Gsn1p . In Gsy2p , three phosphorylation sites are present at S650 , S654 and T667 in the C-terminus [33] whereas in Gsn1p the predicted residues are present at S632 , S636 and T645 positions at the C-terminus , as shown in Supplemental Figure S5 . To generate a Δgsn1 null mutant of M . oryzae , targeted gene replacement of GSN1 was carried out using the split marker gene deletion method ( Figure S6A ) [25] . Gene deletion constructs were transformed into Guy11 and putative transformants screened based on their resistance to hygromycin B ( Figure S7 ) . In Δgsn1 mutants , glycogen deposits could not be clearly observed ( Figure 6A ) , in contrast to Guy11 ( Figure 1 ) . To investigate this further , we measured glycogen using an enzymatic assay and observed a significant reduction ( P<0 . 01 ) in glycogen within cells of the Δgsn1 mutant ( Figure 6C ) . We conclude that glycogen synthesis is severely compromised in both conidia and appressoria in the absence of the GSN1 glycogen synthase gene . To investigate the role of GSN1 in disease progression of rice blast disease , seedlings of rice CO-39 and barley cv . Golden Promise were each inoculated with the glycogen synthase mutant Δgsn1 and the wild type strain Guy11 . The Δgsn1 mutant was able to cause rice blast disease symptoms , which were identical to the wild type strain from which it may be inferred that this gene is dispensable for rice blast infection in both rice and barley ( Figure 6B ) . We conclude that the synthesis of glycogen as a storage product in spores in M . oryzae is not essential for pathogenicity . We were intrigued to find that impairing glycogen synthesis had no effect on virulence of M . oryzae , given that both Δagl1 and Δgph1 mutants showed such pronounced defects in their ability to cause rice blast disease . We therefore set out to identify whether other biochemical differences were apparent in either of the glycogen metabolic mutants . Interestingly , we observed that Δagl1 and Δgph1 mutants both showed reduced trehalose levels during vegetative growth on CM ( Figure 7A ) . This suggests that in M . oryzae , the precursors of trehalose , at least during mycelium growth , may be derived by glycogen degradation , or are affected by loss of these enzyme activities . We therefore analyzed expression of TPS1 and TPS3 in Guy11 , Δagl1 , Δgph1 and Δagl1Δgph1 mutants . TPS1 encodes trehalose-6-phosphate ( T6P ) synthase and TPS3 encodes a subunit of the T6P synthase/trehalose phosphatase complex [34] . Expression of TPS1 was reduced 5-fold in Δagl1 , Δgph1 and Δagl1Δgph1 mutants compared to Guy11 ( Figure 7B ) , whereas expression of TPS3 was reduced 6-fold in Δagl1 , 9-fold in Δgph1 and 50-fold in the Δagl1Δgph1 mutant ( Figure 7C ) . We also investigated expression of the RSY1 gene involved in the biosynthesis pathway of melanin [35] , because TPS1 is known to regulate the expression of several virulence-associated genes , including melanin biosynthetic genes via the TPS1/NMR genetic switch [11] . Expression of RSY1 was also down-regulated in the independent single and the double mutants ( not shown ) . It is known that M . oryzae Tps1 acts as a G6P-sensing protein , directly binding to G6P and NADPH to control expression of a set of virulence-associated genes expressed during plant infection [11] . We hypothesized that if glycogen metabolic genes ( AGL1 and GPH1 ) regulate expression of TPS1 during infection , then deletion of one of the transcriptional inhibitor genes that have been shown to interact with Tps1p might partially complement phenotypic defects associated with a Δagl1Δgph1 double mutant [11] . Deletion of NMR1 , NMR2 or NMR3 , for instance , has been shown to be sufficient to restore pathogenicity to a Δtps1 mutant [11] . We therefore carried out targeted deletion of the NMR3 gene in a Δagl1Δgph1 double mutant , because Δtps1Δnmr3 double mutants showed maximum restoration of pathogenicity [11] . We employed a split marker strategy to target the NMR3 locus for gene deletion , as shown in Figure S8A [25] . Following transformation , M . oryzae transformants were initially screened for resistance to bialaphos [36] and the deletion confirmed by DNA gel blot ( Figure S8B ) . To test the virulence of Δagl1Δgph1Δnmr3 on the susceptible rice cultivar CO-39 , a pathogenicity assay was carried out using Guy11 , an Δagl1Δgph1 mutant and the Δagl1Δgph1Δnmr3 mutant . We found that NMR3 gene deletion restored the ability of the Δagl1Δgph1 mutant to cause blast disease as shown in Figure 8 . No significant difference in lesion number was observed between Guy11 and Δagl1Δgph1Δnmr3 in rice infection ( P<0 . 05 ) , or barley infections ( Fig . 8 ) . We conclude that AGL1 and GPH1 may influence the operation of Tps1-mediated gene regulatory mechanism that is necessary for rice blast disease .
In this study , we set out to investigate the role of glycogen metabolism in establishment of rice blast disease by M . oryzae . Previous reports have established that glycogen mobilization occurs during appressorium development and may depend on cAMP-dependent protein kinase A and the Pmk1 MAP kinase signalling pathways [7] . Glycogen reserves in the appressorium have also been proposed as a potential substrate for glycerol production in the appressorium , facilitating generation of turgor to rupture the plant cuticle [6] , [7] , [26] . Glycogen reserves in appressoria are , for instance , clearly utilized during appressorium maturation , prior to plant infection [26] . When considered together , the results presented in this study provide evidence that cytosolic glycogen degradation in M . oryzae , which requires glycogen phosphorylase and amyloglucosidase , plays an important role in the virulence of this fungus . This conclusion is based on the mutant phenotypes of Δagl1 , Δgph1 and Δagl1Δgph1 double mutants , which are impaired in their ability to cause blast disease . However , it seems likely that the effect on virulence is pleiotropic , rather than specifically associated with utilization of glycogen reserves , because Δgsn1 mutants that are significantly reduced in glycogen reserves due to absence of glycogen synthase activity , are unaffected in virulence . This result suggests that glycogen reserves in the conidium and appressorium are dispensable for appressorium turgor generation and plant infection . Data presented here also suggests that loss of GPH1 and AGL1 reduces expression of TPS1 which is known to integrate control of carbon and nitrogen metabolism in M . oryzae in response to G6P and NADPH/NADP levels [11] . At the outset of this study , we were particularly interested in understanding how glycogen mobilization occurred during appressorium development by M . oryzae . We therefore initially carried out both cytological and biochemical analysis to study the movement and breakdown of glycogen stores . The results of these experiments were consistent in showing that glycogen is rapidly degraded during conidial germination and that the process is retarded in ΔcpkA mutants lacking the catalytic sub-unit of protein kinase A . Conversely , the Δmac1 sum1-99 mutant , a bypass suppressor of an adenylate cyclase null mutant that carries a mutation in the cAMP-binding pocket of the regulatory sub-unit of protein kinase A [16] , shows mis-regulation in glycogen storage in the spore . This is consistent with earlier suggestions , based solely on microscopy [7] , that Δmac1 sum1-99 mutants are accelerated in glycogen and lipid breakdown . Taken together , these observations suggest that the cAMP response pathway is significant in regulating cytosolic glycogen metabolism . To investigate the role of glycogen breakdown in plant infection by M . oryzae , we generated gene deletion mutants lacking amyloglucosidase and glycogen phosphorylase enzyme activities . Phenotypic analysis revealed that Δagl1 and Δagl1Δgph1 double mutants were severely defective in degrading glycogen within the conidium and , after 24 h , glycogen rosettes remained visible in conidia of both mutant strains . Hyper-accumulation of glycogen has previously been observed in Δagl1 and Δgph1 mutants of S . cerevisiae [37] , [38] , demonstrating the significance of these enzymes to utilization of glycogen reserves . A previous report has shown that a vacuolar glucoamylase , Sga1 , in M . oryzae may contribute to glycogen utilization during conidiogenesis [39] . Mutants lacking Sga1 activity showed reduced conidiation , consistent with a role for vacuolar , autophagic hydrolysis of glycogen during spore development . Interestingly , Δsga1 mutants did not , however , show any effect on appressorium development or pathogenicity [39] , suggesting that M . oryzae may degrade glycogen through a vacuolar autophagic mechanism during spore development , but a largely cytosolic route during spore germination and appressorium development , even though a large burst of autophagy is associated with conidial germination , programmed cell death and appressorium maturation [3] , [5] . Agl1-GFP and Gph1-GFP gene fusions , for instance , confirmed that both enzymes are localized in the cytoplasm within conidia , germ tubes and appressoria . The phenotypes of Δagl1 and Δgph1 mutants are consistent with glycogen mobilization and utilization having a significant effect on the ability of M . oryzae to cause rice blast disease . However , deletion of GSN1 which encodes glycogen synthase resulted in a significant reduction in glycogen levels within cells , but had no effect at all on pathogenicity of M . oryzae . . Lipid , glycogen and trehalose are the major nutrient reserves used by M . oryzae conidia for germination and appressorium differentiation ( reviewed in [2] ) . Based on results presented here , it seems that glycogen utilization may not be as significant a factor in appressorium maturation as triacylglycerol hydrolysis [9] . Triacylglycerol lipase activity is , for instance , induced during appressorium development in M . oryzae [7] . The most significant conclusion we can make regarding the role of AGL1 and GPH1 in pathogenicity of M . oryzae is that the absence of these enzyme activities reduces expression of the TPS1 gene . In M . oryzae , it is known that Tps1 activity is essential for rice blast disease and forms an NADPH-dependent genetic switch [11] . Mutants lacking Tps1 are able to produce appressoria , but they are non-functional and the fungus cannot colonize rice tissues [40] . The Δtps1 mutant does not produce trehalose or its intermediate trehalose-6-phosphate ( T6P ) , but impairment of pathogenicity is not due simply to loss of trehalose biosynthesis . Mutations in the G6P-binding pockets of Tps1 , for instance , rendered the fungus non-pathogenic , whereas mutations in the catalytic sites necessary for T6P synthesis did not affect its role in rice blast disease , suggesting that loss of G6P binding is associated with loss of pathogenicity [12] . Tps1 integrates carbon and nitrogen source utilization and Δtps1 mutants are unable to grow on nitrate due to effects on nitrate reductase ( NR ) activity [34] . NADPH is a co-factor for nitrate reductase and provides reducing power . Sufficient NADPH pools are necessary for the function of NR and are produced from the oxidative pentose phosphate pathway , via activation of G6P dehydrogenase ( G6PDH ) . Both NADPH levels and G6PDH activity have been shown to be reduced in Δtps1 mutants , indicating that G6PDH expression is under the control of Tps1 [34] . Over-expression of G6PDH restores the pathogenicity defect of Δtps1 mutants and Tps1 has recently been shown to bind directly to NADPH , suggesting a wider regulatory role on carbon and nitrogen utilization [11] . Tps1 has also been shown to be a positive regulator of three GATA transcriptional factors in a process that is negatively regulated by three NADP-dependent Nmr inhibitor proteins . The GATA transcriptional factors positively regulate expression of virulence-associated gene expression in a manner modulated by the activity of the NMR inhibitor proteins and are , ultimately , dependent on cellular NADPH/NADP and G6P levels [11] . Strikingly , we observed that Δagl1 and Δgph1 mutants are reduced in trehalose accumulation . Moreover , the trehalose biosynthesis enzyme-encoding genes , TPS1 and TPS3 , are down-regulated in glycogen metabolic mutants . We therefore hypothesized that Agl1 and Gph1 might regulate expression of Tps1 and Tps3 . Deletion of NMR3 was sufficient to restore virulence to a Δagl1Δgph1 double mutant , albeit with small lesions . This suggests that glycogen metabolism may contribute to control of the NADPH-dependent Tps1 genetic switch . This is interesting because AGL1 and GPH1 gene expression , and glycogen turnover , has been shown to be affected in Δtps1 strains [34] , suggesting a negative-feedback mechanism might exist in wild type to regulate glycogen turnover in response to G6P sensing . When considered together , our study has therefore revealed fundamental new information regarding the role of glycogen metabolism in the rice blast fungus Magnaporthe oryzae . Results presented here suggest that glycogen metabolism exerts a wider regulatory role than was initially predicted but there is unlikely to be a direct requirement for glycogen degradation to fuel appressorium-mediated cuticle penetration . Instead the effect of Agl1p and Gph1p on the Tps1-dependent genetic switch suggests an important and unexpected regulatory role during plant infection .
Isolates of Magnaporthe oryzae used in this study are stored in the laboratory of N . J . T . Targeted gene replacement mutants described are isogenic to the wild-type rice ( Oryza sativa ) pathogenic Magnaporthe strain Guy-11 . Media composition and procedures for fungal culture and maintenance , nucleic acid extraction and DNA-mediated transformation were described previously [41] . Glycogen within developing appressoria was visualised using an iodine stain , consisting of 60 mg KI and 10 mg I2 mL−1 in distilled water [42] . Freshly harvested conidia at a concentration of approximately 1×105 mL−1 were incubated on hydrophobic plastic cover slips in either standard growth media ( CM ) with 2% yeast extract added or in distilled water for defined lengths of time . After a few seconds contact with the stain yellow/brown glycogen deposits could be observed microscopically . To measure the amount of glycogen present at stages of M . oryzae development a protocol was developed based on experiments performed in S . cerevisiae [17] , [43]–[46] . Sporulating cultures of 12-day-old M . oryzae were flooded with sterile distilled water and scraped with a glass spreader . The resulting suspension was filtered through sterile miracloth ( Calbiochem ) and conidia recovered by centrifugation at 20°C ( 5 , 000 g , 10 minutes ) . After washing in distilled water the conidia were concentrated by centrifugation and diluted to approximately 1×106 conidia mL−1 . Samples were divided and incubated on hydrophobic plastic coverslips for time periods of 0 , 2 , 4 , 6 , 8 , 24 or 48 hours before being transferred to microfuge tubes , heated to 100°C for 10 minutes to halt enzymatic action and cooled on ice . Conidia were then disrupted by 10 minutes of sonication , performed in 10-second bursts using a Vibracell sonicator ( Sonics and Materials Inc . , Danbury , U . S . A ) . Cell fragments were precipitated by centrifugation and 760 µL of the supernatant was removed to a fresh tube . To this , 100 µL 50 mM CaCl2 and 100 µl 0 . 5 M NaOAc pH 5 . 0 were added along with the following enzymes which convert glycogen to glucose: 29 µL amyloglucosidase ( Roche , Germany ) , 10 µL α-amylase ( Sigma ) . In a control experiment these enzymes were replaced by sterile distilled water . Reactions were incubated overnight at 57°C with constant rotation before centrifugation at 5 , 000 g for 3 minutes . For each time point the supernatant from twelve 50 µL aliquots was analysed with a Glucose oxidase assay kit ( Sigma diagnostics ) in which the final colour intensity is proportional to the glucose concentration . Sample absorbance values were measured at 450 nm using a Dynex Technologies MRX II plate reader ( Jencons-PLS ) . The control with no added enzyme was used to measure the background level of glucose . The complete experiment was performed in biological triplicate on separate days and mean glycogen levels calculated for each time point . Glycogen levels were analysed by linear mixed effects models , for each mutant strain individually , with culture as a random effect and time ( hrs ) as a fixed effect . All pair-wise comparisons were then performed by Tukey HSD tests to maintain an overall test level of 5% . Means and differences in means are reported with 95% family wise confidence intervals and Tukey adjusted p-values . The nlme and multcomp packages in R v2 . 15 were used to perform statistical analyses . Genomic DNA from M . oryzae was used for polymerase chain reaction ( PCR ) amplification with the primers AGL1 ( 5′-GCGTACATGGCCTGCTTTGAACGTA-3′ ) and AGL2 ( 5′-TGCTCTGTGTGGAGTAGGCACGAGA-3′ ) , designed to conserved regions of an AGL1 Expressed Sequence Tag ( EST ) sequence from the Clemson University Genetics Institute ( CUGI ) database ( http://www . genome . clemson . edu/ ) . The PCR was performed by applying 35 amplification cycles . After an initial denaturation for 5 minutes at 94°C the following amplification conditions were used: 45 seconds of denaturation at 94°C , 1 minute of annealing at 55°C , 1 minute of elongation at 72°C . This was followed by a final 10 minute extension at 72°C . The resulting 513 bp genomic DNA fragment was then used to screen a Guy-11 conidial cDNA library [15] . The true start codon and 5′ sequence of the identified AGL1 cDNA were determined using a 5′ Rapid Amplification of cDNA ends ( 5′ RACE ) strategy [47] . Screening of a λGEM-11 genomic library [41] with the AGL1 cDNA clone identified a 6 . 5 kb Kpn I lambda genomic clone , named pLH1 . The genomic sequence of this clone was obtained using a Genome Priming System ( GPS ) strategy . This Tn7 transposon-based system uses a transposase complex to randomly insert a transprimer into the dsDNA target [48] , [49] . A population of products is produced each with the transprimer inserted at a different location , and unique priming sites on the end of each transprimer allow sequencing from both strands of the target DNA at the position of insertion . For gene replacement a 6 . 0 kb amplicon , consisting of the AGL1 ORF with a 1 . 0 kb 5′ flank , was amplified from M . oryzae Guy-11 genomic DNA with primers AGLvectorF ( 5′-CCTGACGGATAATGGTGGGGTG-3′ ) and AGLvectorR ( 5′-CTTCGTCCGCAT CCATGTAGAG-3′ ) , designed to the genome sequence of M . oryzae strain 70-15 ( http://www . genome . wi . mit . edu/annotation/fungi/magnaporthe/ ) . The PCR was performed with the following conditions: an initial 1 minute of denaturation at 95°C , then 30 cycles of 1 minute denaturation at 95°C , 1 minute annealing at 60°C and 6 minutes elongation at 72°C . This was followed by a final 10 minute extension at 72°C . The resulting amplicon was cloned into the pGEM-T vector ( Promega ) to create plasmid pLH3 . pLH3 was then digested with HindIII to release a 2 . 8 kb fragment of the AGL1 ORF and the linearised plasmid ligated to a 1 . 4 kb HindIII-linkered hygromycin B phosphotransferase gene cassette to create the gene deletion vector pLH3H . The M . oryzae GPH1 gene was identified in a similar manner to AGL1 . Primers GPH1 ( 5′-CTTCCTCCAGTCAGTAGAGCG-3′ ) and GPH2 ( 5′-TTTGAGGAAGTCATAGCTACCG-3′ ) were designed from conserved regions of GPH-encoding genes using an EST sequence from the CUGI database showing a high degree of similarity to glycogen phosphorylase genes from S . cerevisiae [50] , [51] and other organisms . A 319 bp fragment of M . oryzae GPH1 was amplified by PCR amplification with the following conditions: an initial denaturation for 5 minutes at 94°C , then 35 amplification cycles comprising: 45 seconds of denaturing at 94°C , 1 minute of annealing at 54°C , 1 minute 30 seconds of elongation at 72°C . This was followed by a final 10 minute extension at 72°C . The resulting amplicon was gel-purified , cloned into the pBluescript vector ( Stratagene ) and used to screen a Guy-11 conidial cDNA library [15] . 5′ RACE [47] was used to determine the 5′ sequence and start codon of GPH1 using a gene specific primer designed to the sequence of the longest cDNA clone . To construct the GPH1 gene replacement vector a 5 . 1 kb amplicon , containing the GPH1 ORF with approximately 1 . 0 kb flanks , was amplified from M . oryzae genomic DNA using the primers GPHnestedF ( 5′-GTTGCACTGAACCTCGAGTCTAGA-3′ ) and GPHnestedR ( 5′-TTCGCCAAGG ATGCTGGGCTCAAG-3′ ) , designed using the genome sequence of the M . oryzae 70-15 strain . The standard PCR protocol was adapted to incorporate a TaqPlus® Long PCR system ( Stratagene ) , designed for the amplification of long products . Samples were heated to 94°C for 5 minutes then the following conditions applied for 35 cycles: 30 seconds denaturation at 94°C , 30 seconds annealing at 55°C , 9 minutes elongation at 72°C . This was followed by a final 10 minute extension at 72°C . The resulting 5165 bp amplicon was gel-purified and cloned into the pGEM-T vector ( Promega ) to create plasmid pLH4 . Digestion of pLH4 with BstBI released a 2 . 2 kb fragment of the gene which was replaced with a 1 . 4 kb BstBI-linkered Hph gene cassette to create the gene deletion vector pLH4H . Plasmids pLH3H and pLH4H were transformed into M . oryzae strain Guy-11 . Single-copy insertion of plasmid DNA was confirmed for all transformants using DNA gel blot analysis . Targeted gene replacement of AGL1 in Δgph1 strain with sulfonylurea resistance marker , ILV1 , to generate Δagl1Δgph1 using split marker gene strategy ( Catlett et al . , 2003 ) . The replacement was achieved by replacing 1 kb of 5′ open reading frame of AGL1 with 2 . 8 kb ILV1 gene . In the first round of PCR , 1 kb of the flanking sequence from each side of the gene coding region ( named LF and RF ) was amplified using Agl1-50 . 1/Agl1-m13F and Agl1-30 . 1/Agl1-m13R primers , in combination , using genomic DNA of Guy11 . Agl1-m13F and Agl1-m13R have additional 5′ sequences , corresponding to the M13F/M13R primers . This additional sequence facilitated overlapping fusion PCR to be carried out in the second round . In a parallel reaction , the selectable marker , ILV1 was amplified in two halves using primers M13F/SUR-F and M13R/SUR-R from the sulfonylurea resistance cassette gene [36] , and then cloned into pBluescript ( Stratagene ) . These two products were named IL and LV from the ILV1 gene . The first PCR was carried out under the following cycle conditions in an Applied Biosystems GeneAmp PCR System gradient cycler: an initial denaturation step at 94°C for 5 min , followed by 35 cycles of PCR cycling parameters , 94°C for 30 sec , 62°C or 30 sec and 72°C for 1 min , followed by final extension at 72°C for 10 min . Each 25 µL PCR reaction contained 1 µL of DNA template ( 50 ng ) , 20 pmol of each primer , 2 to 4 mM MgCl2 , 2 . 5 µl of thermophilic 10× reaction buffer without MgCl2 ( 500 mM KCl , 100 mM Tris-HCl [pH 9 at 25°C] and 1% Triton X-100 ) , 2 mM of all four deoxynucleoside triphosphates ( dNTPs , Amersham BioSciences ) , 0 . 3 µL of Taq and 0 . 2 µL Pfu DNA polymerase . The resulting product was analysed by gel electrophoresis and gel purified . In the second round of PCR , each flank was fused with one half of the selectable marker ( LF/IL and RF/LV ) using nested primers Agl1-nesF/IL split and Agl1-NesR/LV split . The conditions were set as initial denaturation at 94°C for 5 min , followed by 35 cycles of PCR cycling parameters , 94°C for 30 sec , 62°C for 30 sec , 72°C for 3 min , followed by final extension at 72°C for 10 min . The resulting product was analysed by gel electrophoresis and gel purified ready for fungal transformation . Homologous recombination between the overlapping regions of the selectable marker and chromosomal DNA results in targeted deletion of the gene [25] . The transformants were screened by DNA blot analysis . The sequences of the primers used in the study are shown in Table S1 . Loss of amyloglucosidase activity in Δagl1 strains , and loss of glycogen phosphorylase activity in Δgph1 strains , was confirmed enzymatically using proteins extracted from Δagl1 and Δgph1 strains , compared to Guy11 . Strains were grown in GMM for 16 hr , followed by freeze drying and sample preparation , as described previously [12] . Briefly , 10 mg of dried mycelia per strain , in triplicate , was finely ground and re-suspended in 1 ml of sterile distilled H2O . Each sample was snap-freezed in liquid nitrogen to break the mycelial cells and liberate total cell protein , and centrifuged for 3 mins . After centrifugation , samples were maintained on ice . 50 µl of this solution , containing total cell protein , was aspirated , in triplicate , from each sample and added to 1 ml of each enzyme assay . All enzymatic assays were performed at 22°C . All assay components were purchased from Sigma . Enzyme activities were determined spectrophotometrically in triplicate and are given as the concentration of product formed in one minute by total cell protein from 1 mg of mycelium . Amyloglucosidase activity [52] was determined by incubating 50 µl of each total cell protein solution with a 1% starch solution , excess purified hexokinase and excess purified glucose 6-phosphate dehydrogenase . Agl1 in the protein sample liberates glucose from starch , which is phosphorylated by hexokinase and the resulting glucose 6-phosphate used to generate NADPH from NADP using glucose 6-phosphate dehydrogenase . The rate of increase in NADPH as the enzyme assay progressed , measured spectrophotometrically at A340 and compared to the rate of NADPH production in assays containing no starch or boiled mycelial protein extract , was measured to calculate the activity of Agl1 in each sample . For glycogen phosphorylase activity [52] , 50 µl of protein extract from each sample was added to 1 ml of enzyme assay , in triplicate . Each assay contained 4% glycogen , excess purified phosphoglucomutase and excess purified glucose 6-phosphate dehydrogenase . Gph1 liberated glucose-1-phosphate from glycogen , which was converted to glucose-6-phosphate by phosphoglucomutase . Glucose 6-phosphate dehydrogenase converted NADP to NADPH using glucose 6-phosphate , and the rate of increase in NADPH as the enzyme assay progressed , measured at at A340 and compared to control reactions lacking glycogen or protein extract , was used to calculate the activity of Gph1in each sample . Protocols were obtained from the Sigma website ( www . sigma . com ) . Trehalose concentrations were measured as described previously [12] . Briefly , samples were treated with trehalase to liberate glucose from trehalose , and glucose was measured using excess purified hexokinase to generate glucose 6-phosphate , which was used by excess glucose 6-phosphate dehydrogenase to generate NADPH from NADP . The amount of NADPH generated in samples treated with trehalase , compared to the same samples not treated with trehalase , was measured at A340 and the amount of trehalose in each sample was calculated . The AGL1 gene was amplified from genomic DNA of isogenic wild type strain Guy11 with primers Agl1-Pro-U/Agl1-Fus-D , the 2 . 8 kb acetolactate synthase gene allele ( ILV1 ) , conferring resistance to sulfonylurea , was amplified with primers Sur-U/Sur-D from pCB1532 [36] and 1 . 4 kb sGFP:trpC was amplified with primers GFP-U/TrpC-D from pNOX1sGFP . The sequences of the primers are shown in Supplemental Table S1 . The PCR fragments were transformed with SpeI/HindIII digested pNEB1284 yeast into S . cerevisiae . The gene fusion was constructed by yeast gap repair cloning , based on homologous recombination in yeast [53] . Resulting plasmid was transformed into the isogenic wild type strain Guy11 . A 4 . 1 kb amplicon consisting of the M . oryzae GPH1 promoter and open reading frame was PCR amplified from Guy11 genomic DNA with primers Gph1-P-SacII and Gph1-SpeI-D and cloned into pGEM®-T ( Promega ) . The fragment was excised with SacII/SpeI restriction sites and cloned into pCB1532 , consisting of sulfonylurea resistance gene allele ( acetolactate synthase or ILV1 ) , to generate pGPH1 . A 1 . 4 kb fragment of the sGFP:trpC cassette ( Chiu et al . , 1996 ) was PCR amplified using primers GFP-SpeI-U and TrpC-ClaI-D ( Supplemental table S1 ) and subsequently cloned into pGPH1 with SpeI restriction sites to create pGPH1sGFP fusion . The fusion plasmid was transformed in isogenic Guy11 of M . oryzae . To determine vegetative growth rates colony diameter was measured after 12 days growth on complete medium [41] . The frequency and nature of conidial germination and appressorium formation was determined microscopically by counting the numbers of germ tubes and/or appressoria formed from conidia after incubation on hydrophobic plastic cover slips for varying lengths of time . Glycogen mobilisation was visualised microscopically as described above . The effect of each gene deletion on fungal pathogenicity was determined by spraying 5×104 spores/ml onto susceptible CO-39 rice and susceptible Golden Promise barley lines as described previously [41] . To investigate appressorium turgor generation an incipient cytorrhysis ( cell collapse ) method [54] , [55] was used . Conidia at a concentration of 2×104 ml−1 were incubated on hydrophobic plastic coverslips to induce appressorium formation . The surrounding water was then removed and replaced with an equal volume of glycerol solution ranging in concentration from 0 . 5 to 5 . 0 M . After 10 minutes incubation the number of appressoria that had collapsed was recorded . Cuticle penetration by appressoria was investigated using onion epidermal peels as described by Chida and Sisler ( 1987 ) . Conidia at 5×104 spores/ml were allowed to germinate on onion epidermis and the rate of successful appressorium-mediated penetration of the epidermis assayed by light microscopy . All experiments were performed in triplicate . Appressorium-mediated penetration of rice leaf sheaths was assessed using a procedure based on that of Kankanala et al , 2007 [28] . A conidial suspension at a concentration of 1×105 spores mL−1 was harvested in 0 . 25% gelatin and inoculated onto leaf sheaths above the mid vein . The inoculated sheaths were left in the moist petri dishes for 24 , 36 and 48 h in a horizontal position to keep the suspension stayed on the mid vein section . When ready for microscopy , the sheaths were hand trimmed to remove the sides and bottom surface to expose a single cell epidermal layer of the mid vein ( 2–3 cell thick ) . This epidermal layer was mounted on a glass slide and observed using the Zeiss LSM510 Meta confocal-light scanning microscope ( CLSM ) system . Targeted gene replacement of GSN1 and NMR3 was carried out in Guy11 and Δagl1Δgph1 , respectively , using split marker strategy as explained earlier , where GSN1 was replaced with the hygromycin B resistance selectable marker hph and NMR3 was replaced with the bialaphos resistance selectable marker bar . The sequences of the primers used in the study are shown in Supplemental table S1 . Following transformation , transformants were screened in the presence of hygromycin B ( 200 mg mL−1 ) for hygromycin resistance and glufosinate ammonium ( 100 mg mL−1 ) for baialaphos resistance . Gene replacements were confirmed by DNA blot analysis . Quantitative RT-PCR was used to characterise the gene expression of TPS1 and TPS3 in Guy-11 , Δagl1 , Δgph1 and Δagl1Δgph1 . To analyse gene expression of mycelia grown in complete medium , the isogenic wild type Guy-11 , Δagl1 , Δgph1 and Δagl1Δgph1was inoculated into CM 48 hr followed by RNA extraction [41] , cDNA synthesis and QRT-PCR analysis . cDNA synthesis was performed using AffinityScript QPCR cDNA synthesis kit ( Stratagene ) , following the manufacturer's protocol . Gene expression analysis was performed using the Brilliant Green QPCR Master Mix kit from Stratagene and the MX3005P instrument ( Stratagene ) following the manufacturer's protocol . Expression of TPS1 , TPS3 and RSY1 was analysed using the primers Tps1-qRT-5′ ( 5′-CTTATCGTCAACCCCTGGAAC-3′ ) and Tps1-3′ ( 5′-TCCCTCCGTCTTGTTGTCGC-3′ ) , Tps3-qRT-5′ ( 5′-CACATCAACGACGCCTGCGA-3 ) and Tps3-3′ ( 5′-ATGTAGCTCTTGCCTCTGTGTT-3′ ) and Rsy1-qRT-5′ ( 5′- CTACCACCAGCTGCGCGTC-3′ ) and Rsy1-qRT-3′ ( 5′-TTATTTGTCGCCAAAGGTCTCC-3′ ) . Expression was normalised relative to β-tubulin ( MGG_00604 . 6 ) gene expression using the primers QRT . PCR . BTubF ( CGCGGCCTCAAGATGTCGT ) and QRT . PCR . BTubR ( GCCTCCTCCTCGTACTCCTCTTCC ) . PCR conditions were 94°C for 30 seconds , 62°C for 30 seconds and 72°C for 30 seconds . | The fungus Magnaporthe oryzae causes a devastating disease of rice called blast . Each year , rice blast disease destroys almost a quarter of the potential global rice harvest . The fungus infects rice plants by elaborating a special infection structure called an appressorium , which physically breaks the tough outer cuticle of a rice leaf . Magnaporthe can develop appressoria in the absence of a nutrient source . We are therefore studying how the fungus utilizes energy stores in its spores to fuel its initial growth and development . Glycogen is a key storage compound in fungi and in this study we have investigated how glycogen breakdown occurs in the rice blast fungus . We have shown that the two major enzymes that degrade cellular stores of glycogen are important in rice blast disease . However , we also found that a strain of the fungus which is severely impaired in its ability to synthesize its own glycogen can still infect plants normally . To explain these apparently contradictory findings we explored the regulatory role of glycogen breakdown and provide evidence that glycogen metabolism is a key regulator of a recently described , virulence-associated genetic switch in Magnaporthe that is operated by an enzyme called trehalose-6-phosphate synthase . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Glycogen Metabolic Genes Are Involved in Trehalose-6-Phosphate Synthase-Mediated Regulation of Pathogenicity by the Rice Blast Fungus Magnaporthe oryzae |
Virulence factors generally enhance a pathogen's fitness and thereby foster transmission . However , most studies of pathogen fitness have been performed by averaging the phenotypes over large populations . Here , we have analyzed the fitness costs of virulence factor expression by Salmonella enterica subspecies I serovar Typhimurium in simple culture experiments . The type III secretion system ttss-1 , a cardinal virulence factor for eliciting Salmonella diarrhea , is expressed by just a fraction of the S . Typhimurium population , yielding a mixture of cells that either express ttss-1 ( TTSS-1+ phenotype ) or not ( TTSS-1− phenotype ) . Here , we studied in vitro the TTSS-1+ phenotype at the single cell level using fluorescent protein reporters . The regulator hilA controlled the fraction of TTSS-1+ individuals and their ttss-1 expression level . Strikingly , cells of the TTSS-1+ phenotype grew slower than cells of the TTSS-1− phenotype . The growth retardation was at least partially attributable to the expression of TTSS-1 effector and/or translocon proteins . In spite of this growth penalty , the TTSS-1+ subpopulation increased from <10% to approx . 60% during the late logarithmic growth phase of an LB batch culture . This was attributable to an increasing initiation rate of ttss-1 expression , in response to environmental cues accumulating during this growth phase , as shown by experimental data and mathematical modeling . Finally , hilA and hilD mutants , which form only fast-growing TTSS-1− cells , outcompeted wild type S . Typhimurium in mixed cultures . Our data demonstrated that virulence factor expression imposes a growth penalty in a non-host environment . This raises important questions about compensating mechanisms during host infection which ensure successful propagation of the genotype .
The ability to infect a host and elicit disease is dictated by the virulence factors expressed by a given pathogen . This may include , but is not limited to , protective factors neutralizing antibacterial defenses , enzymes involved in nutrient acquisition within the host , regulators of virulence factor expression and toxins or secretion systems for subverting host cell signal transduction . The coordinated expression of such virulence factors enhances colonization , growth/survival within the host and transmission . However , most studies of virulence factor function and pathogen fitness have been performed in bulk assays , averaging the phenotypes over large pathogen populations of genetically identical cells . In contrast , little is known about the potential advantages , costs or burdens arising from virulence factor expression by an individual cell of the pathogen population . Therefore , single cell analyses might be of significant interest , in particular if virulence factors , which are expressed in a bistable fashion by some but not all members of a pathogen population , e . g . the ttss-1 system of S . Typhimurium [1] , [2] , [3] , [4] , [5] , as described in this paper . Bistable gene expression is genetically encoded . In most cases , one particular genotype expresses one predictable phenotype in a given environment . However , in some cases , two different phenotypes are expressed by isogenic organisms living in the same environment . This is termed phenotypic variation , bimodal gene expression or bistability and represents a special case of gene expression [6] . The importance of bistability for pathogenic bacterial fitness and evolution is just beginning to be understood . Like other cases of gene expression , bistability is generally observed in response to particular environmental cues . The response is driven by a dedicated ( set of ) regulator ( s ) , which responds to environmental signals ( operon model of Jacob [7] ) . This response is subject to stochastic fluctuations . In particular in the case of regulators expressed in a few copies per cell , this can significantly affect the active regulator concentration thus randomizing the corresponding phenotype in a population [8] , [9] . In combination with non-linear responses ( e . g . regulator multimerization , feedback loops ) , this can lead to formation of phenotypically distinct and stable subpopulations of isogenic bacteria [6] , [8] , [9] , [10] , [11] . In terms of evolution , two models may explain the advantage of bistability: i . in “bet hedging” , the optimally adapted phenotype will prevail and ensure the survival of the shared genotype in a changing environment [12] . ii . in “division of labor” , both phenotypes cooperate to ensure survival of the shared genotype [4] . In either way , the bistable expression of certain genes is thought to promote the survival of the genotype . However , it has remained poorly understood whether/how bistability may affect the lifestyle of pathogenic bacteria . Salmonella enterica subspecies 1 serovar Typhimurium ( S . Tm ) is a pathogenic Gram-negative bacterium causing numerous cases of diarrhea , worldwide . Its' type III secretion system 1 ( TTSS-1 ) was recently identified as an example for bistable gene expression [1] , [3] , [5] , [13] . TTSS-1 is a well-known virulence determinant of S . Tm required for eliciting diarrheal disease [14] , [15] , [16] . The needle like TTSS-1 apparatus injects effector proteins into host epithelial cells , thus triggering host cell invasion and pro-inflammatory responses [17] , [18] , [19] . TTSS-1 is encoded on a genomic island ( Salmonella pathogenicity island 1 ( SPI-1 ) ) , which also harbors genes for effector proteins and for several regulators of ttss-1 expression , e . g . hilA , hilC and hilD [20] , [21] . The bistable ttss-1 expression is controlled by a complex regulatory network , which includes coupled positive feedback loops , controls the threshold for ttss-1 induction and amplifies ttss-1 expression [5] , [22] . Bistable ttss-1 expression is observed in “ttss-1 inducing” environments , i . e . the gut lumen of infected mice or in non-host environments , e . g . when S . Tm is grown to late logarithmic phase in LB [1] , [2] , [4] , [5] . This yields mixed populations of isogenic S . Tm cells that express ttss-1 ( TTSS-1+ phenotype ) , or do not ( TTSS-1− phenotype ) , in a bimodal fashion . In the mouse gut , only the TTSS-1+ cells can actively invade the mucosal tissue and efficiently trigger inflammation [4] , [18] . This inflammatory response may help to overcome the commensal microflora , thus enhancing Salmonella growth and transmission [23] , [24] , [25] , [26] , [27] , [28] , [29] . Experimental data indicate that bistable ttss-1 expression might represent an example of “division of labor” [4] , but further data is required to settle this point . At any rate , ttss-1 expression seems to be instrumental for eliciting diarrheal disease and enhancing pathogen transmission . But the functional properties of the TTSS-1+ phenotype are not well understood . The complex setting of the infected animal gut has hampered the analysis of the TTSS-1+ phenotype . In vitro experiments are essential for gaining detailed mechanistic insights . Here , we have analyzed the induction of ttss-1 expression and its effects on the growth rate of the TTSS-1+ phenotype by single cell reporter assays , competitive growth experiments and mathematical modeling . In such non-host environments , expression of the ttss-1 virulence system expression imposed a growth penalty on the TTSS-1+ cells . This may have important implications with respect to compensatory mechanisms during the infection of animal hosts .
We started our analysis of the TTSS-1+ phenotype by probing ttss-1 expression at the single cell level . For this purpose , we chose the sicA promoter ( PsicA ) , which controls expression of the chromosomal sicAsipBCDA operon ( Fig . S1C ) . This operon encodes key parts of the TTSS-1 virulence system . On the one hand , we employed a transcriptional sipA-tsrvenus reporter gene cassette placing the reporter downstream of the sicAsipBCDA operon ( Fig . S1; [2] , [3] ) . Due to its localization at the bacterial poles , the tsrvenus reporter allows detecting <10 proteins per cell [30] . Thus , sipA-tsrvenus provides a highly sensitive reporter for the TTSS-1+ phenotype . Next , we verified the performance of the sipA-tsrvenus reporter . sipA-tsrvenus expression was bistable and TTSS-1− and TTSS-1+ individuals were distinguishable by the presence/absence of Tsrvenus spots at the bacterial poles ( [30]; Fig . 1A; Fig . S1D ) . TTSS-1 expression and virulence were not compromised ( Fig . 1B ) . The accurate response of sipA-tsrvenus to Salmonella signaling cascades was established by disturbing known elements of the TTSS-1 gene regulation network and FACS analysis of sipA-tsrvenus expression ( Fig . 1C , D ) . In line with the published work on ttss-1 regulation ( Fig . 1D ) : i . Over-expression of positive TTSS-1 regulators increased the abundance of tsrvenus-expressing individuals ( Fig . 1C; Fig . S1D ) . In particular , hilA , hilC and hilD over-expression increased the fraction of sipA-tsrvenus expressing individuals from ∼20% to 80–100% . ii . The median signal intensity per sipA-tsrvenus expressing cell increased when positive regulators were over-expressed ( philA: 3 . 8±0 . 3-fold; philC: 4 . 0±0 . 1-fold; philD: 4±0 . 1-fold; median ± s . d . ) . iii . Control experiments in a ΔhilA mutant verified that expression of the TTSS-1+ phenotype depended on the ttss-1 master-regulator , HilA ( Fig . 1C; open bars ) and iv . The average HilA protein levels of the analyzed strains correlated positively with the fraction of tsrvenus-expressing individuals ( r2 = 0 . 78; quantitative Western blot; Fig . 1E ) . These data verified the accurate performance of the sipA-tsrvenus reporter and demonstrated that hilA-dependent regulation affects both , the fraction of TTSS-1+ individuals and the level of ttss-1 expression per cell . In addition , we employed psicA-gfp , a reporter plasmid expressing gfp under control of the sicA promoter . This construct yielded brighter fluorescence than the chromosomal sipA-tsrvenus and was better suited for FACS analysis . Again , this reporter yielded a bistable expression pattern ( Fig . 1F ) . Using wt S . Tm psicA-gfp we separated TTSS-1+ and TTSS-1- subpopulations by FACS . Western blot analysis of the FACS-sorted subpopulations verified coincident expression of psicA-gfp and the TTSS-1 protein SipC ( Fig . 1F , G ) . This indicated that our fluorescent reporter constructs are faithful reporters of the bistable expression of the TTSS-1+ phenotype . During our experiments , we observed that hilA , hilC and hilD over-expression led to reduced culture densities ( e . g . OD600 for wt sipA-tsrvenus: 3 . 4±0 . 3 vs . wt sipA-tsrvenus philA: 2 . 0±0 . 3; mean ± s . d . ) . This was a first hint suggesting that retarded growth might be a general feature of the TTSS-1+ phenotype . However , it remained to be shown whether growth retardation occurs in wild type cells expressing normal levels of hilA , hilC and hilD . The growth rate of the TTSS-1+ individuals was analyzed by time-lapse microscopy . Wild type S . Tm harboring gfp- or tsrvenus-reporters for ttss-1 expression were placed on an agar pad ( LB , 1 . 5% agarose ) , the TTSS-1+ individuals were identified by fluorescence microscopy and growth was analyzed by time-lapse phase contrast microscopy ( 1 frame/30 min; Fig . 2A ) . Imaging did not impose detectable photo damage to the bacteria , as indicated by the unaltered growth rate ( Fig . S2 ) . Strikingly , TTSS-1+ individuals grew slower than TTSS-1− individuals ( wt S . Tm sipA-tsrvenus ( M2001 ) ; µT1+ = 0 . 90 h−1 vs . µT1− = 1 . 30 h−1; p = 0 . 027 for the factor ‘phenotype’ in a two-way ANOVA; Fig . 2B ) . The negative control strain ΔhilA sipA-tsrvenus yielded only TTSS-1− individuals , which grew at the “fast” rate ( µT1− = 1 . 16 h−1; Fig . 2B ) . Thus , TTSS-1+ individuals seemed to grow at a reduced rate . To exclude potential artifacts attributable to the sipA-tsrvenus reporter , we analyzed unmodified wild type S . Tm not harboring any reporter ( Fig . 2C; Fig . S2 ) . Using a maximum likelihood approach , we identified two populations with distinct growth rates ( likelihood ratio test for two populations versus one population , p<0 . 001 , µslow = 0 . 66 h−1 vs . µfast = 1 . 27 h−1; Fig . 2C ) , very similar to the ones described above ( Fig . 2B ) . Furthermore , unmarked mutants lacking the entire SPI-1 region ( Δspi-1 ) or the positive ttss-1 regulator hilD yielded exclusively fast growing cells , while deletion of the negative ttss-1 regulator hilE yielded only slow growing cells ( Fig . 2C ) . Finally , wild type S . Tm harboring psicA-gfp or a chromosomal gfp-reporter for the TTSS-1 gene prgH [1] yielded slow growing TTSS-1+ and fast growing TTSS-1− cells ( µT1+ = 0 . 51 h−1 vs . µT1− = 1 . 2 h−1; p = 0 . 006 for the factor ‘phenotype’ in a two-way ANOVA; Fig . 2D ) . Bacteria expressing the psicA-gfp or prgH-gfp reporters grew even slower than the TTSS-1− sipA-tsrvenus bacteria or the slow-growing wt S . Tm subpopulation ( Fig . 2BC ) . Presumably , this was attributable to the additional “burden” conferred by the GFP expression , as described , before [31] . Thus , the time-lapse microscopy experiments verified bistable ttss-1 expression and revealed that the TTSS-1− phenotype has a reduced growth rate , even at wild type HilA and TTSS-1 levels ( µT1+ in the range of 0 . 7 h−1 vs . µT1− in the range of 1 . 3 h−1 ) . This was confirmed in a dye dilution assay ( Fig . S3 ) . Our data suggested that ttss-1 expression represents a “cost” to the bacterial cell . However the mechanism explaining this growth retardation had remained unclear . We speculated that expression of the TTS apparatus itself or the sheer load of the proteins transported by the TTSS-1 ( effectors , translocon proteins ) might play a role . To test these hypotheses , we analyzed two additional S . Tm mutants . In the first mutant , termed Δprg-orgΔinv-spa , we deleted most apparatus-encoding genes ( Table S1 ) . This mutant formed two populations with distinct growth rates ( likelihood ratio test for two populations versus one population , p<0 . 001 , µslow = 0 . 72 h−1 vs . µfast = 1 . 36 h−1; Fig . 2E ) , very similar to those described for wild type S . Tm ( Fig . 2C ) . The second mutant , termed Δ8Δsip , was lacking the genes for most TTSS-1 effector proteins and the secreted translocon components including sipB , sipC , sipD , sipA , sptP , sopE , sopE2 , sopB and sopA ( Tab . S1 ) . In contrast to wild type S . Tm , we could not distinguish two subpopulations in this mutant ( likelihood ratio test for two populations versus one population , p = 0 . 73; Fig . 2E ) . Instead , this mutant displayed a median growth rate of µ = 1 . 10 h−1 , similar to the fast growing subpopulation of S . Tm wt and the mutants Δspi-1 and hilD ( Fig . 2C ) . This data suggests , that expression of the effector proteins and translocon components is “costly” and provides at least in part a mechanistic explanation for the growth retardation of wild type S . Tm cells of the TTSS-1+ phenotype . When monitoring growth and bistable ttss-1 expression in a wt S . Tm ( psicA-gfp ) culture , the fraction of TTSS-1+ individuals began to rise after 2 . 5 h as soon as the culture entered the late logarithmic phase , increased in a linear fashion , and reached approx . 60% after 7 h once the culture entered the stationary phase ( Fig . 3A ) . Our results implied that two different parameters affect the fraction of TTSS-1+ individuals and the overall growth progression in the late logarithmic phase: i . Competitive growth . TTSS-1+ individuals are steadily outgrown by the fast-growing TTSS-1− individuals ( µT1+<µT1−; Fig . 2 ) ; this constantly reduces the size of the TTSS-1+ subpopulation . ii . ttss-1 induction . Presumably , initiation of ttss-1 expression in TTSS-1− individuals compensates the “TTSS-1+ losses” attributable to competitive growth and explains the increasing fractions of TTSS-1+ individuals during the late logarithmic phase . To infer the dynamic initiation rate ri of ttss-1 expression in the late logarithmic phase from our experimental data , we devised a mathematical model describing the growth of the TTSS-1+ ( NT1+; growth rate µT1+ ) and the TTSS-1− population ( NT1−; growth rate µT1− ) as a function of time ( t ) : ( 1 ) ( 2 ) It should be noted that the model does not include a term for “switching off” ttss-1 expression . This was justified by our failure to observe “off switching” in the experiments shown in Fig . 2 and further supported by other data ( Fig . S2 and data shown below ) . During the late logarithmic phase , the relative abundance of the TTSS-1+ individuals increased , and the fraction α of TTSS-1− individuals ( NT1- ) decreased in a linear fashion ( Fig . 3A ) : ( 3 ) Equation ( 2 ) can be rearranged to calculate ri ( t ) ( see Text S1 for details ) : ( 4 ) With the data from Fig . 3A and by using equation ( 3 ) we could determine NT1− ( t ) and , after fitting an empirical function to NT1− ( t ) , also dNT1−/dt . Using equation ( 4 ) , this allowed calculating ri ( t ) during the late logarithmic phase ( see Text S1 for details ) . We found that the mean initiation rate ( ri ) of ttss-1 expression increased continuously during the late logarithmic phase , e . g . from 0 . 28 h−1 at 3 . 5 h to 0 . 54 h−1 at 5 . 5 h ( SEM = 0 . 03 h−1; Fig . 3B ) . The initiation rate of ttss-1 expression seemed to increase upon entry into the late logarithmic growth phase ( Fig . 3A ) . Therefore , it might be induced by growth-related environmental signals ( e . g . oxygen depletion , quorum signals , nutrient depletion , metabolite accumulation ) . To address this , we analyzed the partial oxygen pressure ( pO2 ) during growth . As expected , pO2 declined to <30% relative aeration during the first three hours ( Fig . 3C ) . After approximately 3 . 5 h , we detected a transient rebound of the oxygen pressure followed by a steady decline to <3% relative aeration during the next hour . This undulation of oxygen pressure is indicative of a change in the growth physiology at 3 . 5 h and was in line with the reduced growth rate ( Fig . 3A , shaded area ) . The data suggested that altered metabolism , nutrient availability , waste product accumulation , the reduced growth rate or the low oxygen pressure might represent cues inducing ttss-1 expression . As a first approach to test the role of pO2 , we performed batch culture growth experiments in identical 250 ml culture flasks filled with the indicated volumes of media ( wt S . Tm psicA gfp grown in 5 , 10 , 25 , 50 or 100 ml LB; Fig . 3D ) . This setup allowed analyzing the effect of reduced pO2 ( i . e . in larger , poorly aerated culture volumes ) at equivalent growth rates . We observed that the fraction of ttss-1 expressing cells increased in larger culture volumes . Therefore , low oxygen tension might represent one environmental cue directly or indirectly inducing bistable ttss-1 expression . However , the evidence is merely circumstantial at this moment and other cues might well be involved . Identification of these cues will benefit from the strategies for determining ri as described above . In liquid culture , the initiation of ttss-1 expression occurred in the late logarithmic phase . However , our initial time lapse microscopy data for bacteria sampled from this growth phase did not show initiation of ttss-1 expression ( Fig . 2 ) . We reasoned that this might be attributable to the lack of inducing environmental signals , as these experiments had been performed on agar pads soaked with fresh LB medium . To test this hypothesis , we modified the time lapse microscopy experiment and imaged bacteria ( S . Tm psicA-gfp ) placed on agar pads soaked with filter-sterilized spent medium taken from a culture at the same growth phase ( OD600 = 0 . 9 , see Materials and Methods ) . We analyzed growth of 191 micro colonies . At the beginning , 135 did not express ttss-1 . But remarkably , we observed 15 of 135 initially TTSS-1− micro colonies , in which individual bacteria induced ttss-1 expression during the course of our imaging experiment ( e . g . Fig . 4A , Fig . S4; Video S1 ) . After induction , the TTSS-1+ cells grew at a slower rate than their TTSS-1− siblings . In addition , we observed numerous TTSS-1+ bacteria ( 56 micro colonies ) and TTSS-1− bacteria ( 120 micro colonies ) which did not “switch” their ttss-1 expression status . In line with the results above , ttss-1 expression and the interval between two cell divisions was negatively correlated ( Fig . 4A , B , C , Spearman's rho = −0 . 747 , p<0 . 0001 , N = 29 ) . These experiments support the stochastic initiation of ttss-1 expression . But the initiation rate of ttss-1 expression ( <0 . 04 h−1 ) was lower than that predicted from the batch culture experiment shown in Fig . 3 ( ri = 0 . 18−0 . 45 h−1 ) . This might be attributable to the lack of some environmental cue , e . g . low oxygen pressure , as time lapse microscopy was performed at ambient atmosphere . Only two micro colonies showed a decrease in fluorescence as expected for “off-switching” . Hence , the rate of off-switching is not substantial . This indicated that our mathematical model , which assumed that “switching off” the ttss-1 expression would be negligible , was justified ( equation ( 1 ) did not include ri ( t ) NT1- ( t ) ) . These experiments verified that ttss-1 expression is initiated in a stochastic fashion under “inducing” environmental conditions and that the TTSS-1+ phenotype exhibits a growth defect . Finally , we confirmed the growth penalty attributable to ttss-1 expression in the late logarithmic phase in competition experiments . Wt S . Tm expresses ttss-1 in a bistable fashion and forms a significant fraction of slow-growing TTSS-1+ cells during the late logarithmic phase ( Fig . 3 ) . This slows down the apparent growth of the total wild type population ( see above ) . In contrast , hilA or hilD mutants , which do not express ttss-1 , yield a pure population of fast-growing TTSS-1− cells ( Figs . 1 and 2 ) . Thus , in a mixed culture , hilA or hilD mutants should outgrow wt S . Tm . Indeed , both mutants out-competed the wt strain during the late logarithmic phase of the mixed culture ( ΔhilA , ΔhilD; Fig . 5A , B ) . In contrast , a hilE mutant , which forms a larger fraction of TTSS-1+ cells than wt S . Tm ( Fig . 2 ) , was outcompeted by wt S . Tm in this type of assay ( ΔhilE , Fig . 5C ) . This verified the growth penalty of TTSS-1+ cells in LB batch cultures .
The effect of virulence factor expression on the fitness of an individual pathogen cell has remained unclear . We have analyzed the fitness costs associated with the expression of ttss-1 , which encodes a key virulence function of S . Tm . An in vitro system was chosen for a detailed analysis of the growth phenotype of TTSS-1+ cells . We found that these cells have a reduced growth rate . This established that ttss-1 expression represents a burden ( and not an advantage ) at the level of the individual cell , at least in the non-host environment of our assay system . The growth penalty affects the fraction of TTSS-1+ individuals and the overall growth progression in a S . Tm culture . Mathematical modeling and experimental data demonstrated that this growth penalty and an increasing initiation rate of ttss-1 expression during the late logarithmic growth phase were sufficient to explain the dynamic abundance of TTSS-1+ and TTSS-1− individuals in a clonal S . Tm batch culture . Evidence for bistability of ttss-1 expression has only recently been accumulated . Under inducing conditions , single cell reporters for expression of ttss-1 or effector proteins yielded cells in the “on” and cells in the “off” state [1] , [2] , [3] , [5] , [32] . The regulatory network controlling ttss-1 expression includes at least three positive feedback loops and this architecture is thought to set the threshold for initiating ttss-1 expression and to amplify the level of expression [5] , [32] , [33] . The TTSS-1+ phenotype can persist for several hours , even if the bacteria are shifted into environments normally not inducing ttss-1 expression ( histeresis; shift to fresh LB , Fig . 2; Fig . S2 ) . However , it should also be noted that it has not been possible to define unequivocally where stochasticity is introduced . In fact , stochastic initiation of ttss-1 expression might hinge on different regulators in different environments . TTSS-1+ cells have at least two important characteristics . First , they express the virulence factors enabling host manipulation and elicitation of disease [13] , [17] , [18] . Second , as we have found here , they grow at a reduced rate . ttss-1 expression may represent a “burden” in itself . The mechanism explaining the growth defect of TTSS-1+ cells is of significant interest . A partial disruption of the proton gradient by “leaky” TTSS assembly-intermediates and/or the metabolic energy required for biosynthesis of the TTSS may offer plausible explanations . Typical TTSS-1+ cells are estimated to express 20–200 TTS apparatuses and approx . 3−10×104 effector proteins , amounting to a significant fraction of the total cellular protein [2] , [3] . Indeed , deleting the translocon and most effector proteins significantly increased the growth rate of the TTSS-1+ cells ( Δ8Δsip; Fig . 2E ) , indicating that these proteins account at least in part for the cost of ttss-1 expression . However , the growth rate of Δ8Δsip ( µ = 1 . 10 h−1 ) was still lower than that of the TTSS-1− subpopulation of wt S . Tm ( µfast = 1 . 27 h−1 ) , suggesting that other factors do also contribute to growth retardation . An alternative explanation for the reduced growth rate of TTSS-1+ cells might reside in coordinated expression of a complex regulon . This might be reminiscent of the prf virulence regulon of Listeria monocytogenes , which coordinates metabolism and virulence gene expression thus controlling environment-specific fitness phenotypes in vitro and in vivo [34] . Several global regulators ( e . g . crp , mlc , fur; [7] , [35] , [36] ) and silencing proteins ( hns , hha; [37] , [38] ) can control ttss-1 expression . Moreover , HilA may control multiple loci apart from ttss-1 ( 25 ) . And we have observed co-expression of ttss-1 and of fliC , which encodes a key structural component of the flagella , in the late logarithmic phase ( Fig . S5 ) . Accordingly , ttss-1 expression might be one feature of a “differentiated” state which also includes adaptations reducing the growth rate . It is tempting to speculate that this state might be particularly adapted for mucosal tissue invasion . This would be an important topic for future research . Interestingly , similar phenomena have been observed in other ttss-expressing pathogens . In Pseudomonas aeruginosa , growth in suboptimal media was shown to result in bistable ttss expression [39] . But it remained unclear whether growth might be affected . In contrast , the plasmid-encoded TTSS of Yersinia spp . is well known to cause growth retardation in response to host cell contact or low calcium environments [40] , [41] . However , in this case , ttss induction seems to be uniform even in suboptimal media [42] . Thus , bistability and growth retardation do occur in other ttss expressing bacteria , but specific adaptations may exist for each pathogen . Which environmental cues induce ttss-1 expression in S . Tm ? ttss-1 is expressed in the lumen of the host's intestine and in the late logarithmic phase in LB-batch culture . Low oxygen pressure is common to both environments and may represent an inducing signal ( see Fig . 3C ) . In line with this hypothesis , Shigella flexneri , a closely related gut pathogen , can modulate the activity of its TTSS in response to low oxygen pressures typically observed at the gut wall [43] . Similarly , HilA-mediated ttss-1 expression is known to respond to oxygen pressure [21] , [44] . In addition , numerous other internal and external cues are known to affect ttss-1 expression , including osmolarity , pH , growth rate , or the presence of short chain fatty acids like acetate [45] , [46] , [47] , [48] , [49] , [50] , [51] . The sum of these environmental cues seems to determine the level of ttss-1 induction . This might explain our observation of a low , but detectable initiation rate of ttss-1 expression on agar pads soaked with spent medium ( Fig . 4 ) . This environment should harbor most cues present in the late log culture medium , but lacks low oxygen pressure , which could not be established in the real time microscopy setup . In summary , our findings indicate that the TTSS-1+ phenotype is more complex than previously anticipated . Currently , we can only speculate how this affects the real infection and transmission in vivo . Our results suggest that the TTSS-1+ subpopulation is constantly drained by the burdens inflicted by immune defenses within the infected gut mucosa [4] and by the reduced growth rate ( this work ) . The latter should represent a competitive disadvantage against all other bacteria ( commensals and TTSS-1− S . Tm cells ) present in the gut lumen . Moreover , this burden should materialize even before invading the gut tissue and may explain why ttss-1 defective mutants are sometimes ( though rarely ) found in infected animal flocks and isolated in one case of a human outbreak [52] , [53] . In order to explain the evolution and mainentance of bistable ttss-1 expression and the successful propagation of the ttss-1 genotype , one has to predict that the TTSS-1+ phenotype must confer some type of advantage . According to the “division of labor” model , the advantage might emanate from a “public good” , i . e . the TTSS-1 induced gut inflammation fostering Salmonella growth in the gut lumen and enhancing transmission . Alternatively , the TTSS-1+ phenotype might include ( unidentified ) features enhancing the survival and growth of the ttss-1 expressing bacteria themselves , e . g . in permissive niches of the host's intestine or by enhancing the chances of chronic infection and long-term shedding . Identifying these mechanisms will represent an important step for understanding the evolution of bistable ttss-1 expression .
All strains were derivatives of Salmonella Typhimurium SL1344 or ATCC14028 ( see Tab . S1 and Text S2 for references ) . All plasmids and primers are shown in Tab . S2 and S3 . Bacteria were inoculated ( 1∶100 in LB ) from 12 h overnight cultures ( LB , supplemented with the appropriate antibiotics ) and grown under mild aeration for 4 h at 37°C , if not stated otherwise . In Fig . 1C , E , the medium included 0 . 01% arabinose . The mutants were constructed using the lambda red recombination system [54] . The chloramphenicol or kanamycin resistance cassette of pKD3 ( cat ) resp . pKD4 ( aphT ) were amplified by PCR using the primer pairs ÄhilA::kan-fw and ÄhilA::kan-rev , ÄhilD::kan-fw and ÄhilD::kan-rev , ÄhilE::cat-fw and ÄhilE::cat-rev and electroporated into SL1344 harboring pKD46 to generate the regulator mutants M2005 ( ÄhilA::aphT ) , M2007 ( ÄhilD::aphT ) and M2008 ( ÄhilE::cat ) . Mutants were selected by plating on LB-Agar ( 50 µg/ml kanamycin or 30 µg/ml chloramphenicol ) . M2072 ( termed Δprg-orgΔinv-spa in this paper ) was also generated using the lambda red system using the primers invG-fw and spaS-rev as well as prgH-fw and orgC–rev and the plasmids pKD3 and pKD4 to generate prgHIJKorgABC::aphT , invGEABCIJspaOPQRS::cat , a mutant lacking most genes of the TTS apparatus . For construction of strain M2532 ( termed Δ8Δsip in this paper ) , we transduced the ÄsipBCDA-sptP::aphT allele from SB245 ( SL1344 , ÄsipBCDA-sptP::aphT fliGHI::Tn10; K . Kaniga and J . E . Galan , unpublished data ) via P22 into M2400 ( SL1344 , ÄsopE , ÄsopE2 , ÄsopB , ÄsipA , ÄsptP , ÄsopA , ÄspvB , ÄspvC ) , which has been previously described [55] . M2532 fails to express most TTSS-1 effector proteins and the translocon components . To create the suicide plasmid pM2002 , pVS152Tsr [30] was digested with the restriction endonucleases Eco47III and XmaI . The tsrvenus encoding fragment was ligated into pM1300 ( digested with MslI and XmaI , [56] ) downstream of a truncated sipA fragment ( nt 1156–2058 of the orf ) , to finally create pM2002 and introduced by homologous recombination into the genome of ATCC14028 to generate the reporter strain M2001 . To obtain the tsrvenus reporter for hilA ( M2076 ) , the c-terminal region of hilA ( nt 114 to 1661 of the orf ) was amplified using the primer pair hilA-fw-XmaI-NcoI and hilA-rev-NheI-XbaI and cloned into pBluescriptII ( Invitrogen ) using the restriction endonucleases XmaI and XbaI , yielding pM2090 . This plasmid was digested with NheI and NotI to introduce the tsrvenus encoding PCR fragment ( template pM2002 , primers: venus-NheI-fw and venus-NotI-rev , digested with NheI and NotI ) to obtain pM2095 . The entire region ranging from hilA to tsrvenus was cloned into pSB377 using the restriction enzymes NotI and XmaI yielding the suicide plasmid pM2080 . This plasmid was used to generate the hilA reporter strain M2076 by homologous recombination into the genome of ATCC14028 . To obtain the tsrvenus reporter for fliC , tsrvenus was amplified by PCR ( primers: tsr-XmaI-fw and venus-XbaI-rev ) and cloned into pBluescriptII using XmaI and XbaI thus yielding pM2533 . After amplification of fliC by PCR using SL1344 chromosomal DNA as template and primers fliC-XhoI-fw and fliC-HindIII-rev , the fliC encoding fragment was cloned via XhoI and HindIII upstream of the tsrvenus gene into pM2533 , thus yielding pM2539 . Subsequently , the construct was moved via XhoI and XbaI into the suicide plasmid pGP704 , thus yielding pM2819 . This plasmid was used to create the fliC-tsrvenus reporter strain M2821 by homologous recombination into the genome of SL1344 . All over-expression plasmids from pM2010 to pM2042 were obtained by digesting the indicated PCR fragments ( Table S2 and S3 for plasmids and primers ) with EcoRI and XbaI into pBAD24 . All mutations were verified by PCR or DNA sequencing . HilA expression was analyzed by quantitative Western blot using an affinity-purified rabbit α-HilA antiserum ( Fig . 1E ) . Recombinant HilA was used for normalization . SipC was detected using an α-SipC serum ( Fig . 1G ) . For invasion , MDCK cells were grown in MEM ( Invitrogen ) , infected for 30 min ( MOI = 5; [57] , washed and incubated in MEM ( 400 µg/ml gentamicin; 1 h ) . Intracellular bacteria were enumerated by plating . Prior to analysis , fluorophore formation was ensured ( 2 h , RT , 30 µg/ml chloramphenicol ) . Tsrvenus and Gfp emission was analyzed at 530 nm ( supplement; FACSCalibur 4-color , Becton Dickinson ) . Bacteria were identified by side scatter ( SSC ) . Data were analyzed with FlowJo software ( Tree Star , Inc . ) . For Tsrvenus ( Fig . 1 ) , ln-transformed fluorescence values for 40000 events were median-normalized ( subtraction ) and compared to the similarly normalized data from the reporterless control strain , thus yielding the fraction of TTSS-1+ individuals . For sorting bacterial cells , S . Tm ( psicA-gfp ) cells were sorted by FACS ( Aria Becton Dickinson , FACSDiva Software ) . Bacteria were placed on a 1 . 5% agarose pad equilibrated with LB , sealed under a glass coverslip and mounted ( 37°C temp . control; Axioplan2; Plan-APOCHROMAT 63x/1 . 4 oil; Zeiss or IX81 , UPlanFLN 100x/1 . 3 Oil , Olympus ) . Reporter fluorescence ( Exc . 470/20 nm; BP 495 nm; Em . 505–530 nm ) and micro colony growth ( phase contrast ) were monitored and evaluated using Axiovision software ( Zeiss ) . The slope of the ln-tranformed bacterial numbers ( t ) , as determined from the logarithmic growth phase , yielded the growth rate µ . For sipA-tsrvenus and prgH-gfp , the micro colonies were scored visually as TTSS-1+ or TTSS-1− . To analyze differences in growth rates between TTSS-1+ and TTSS-1− micro colonies , we performed a full-factorial analysis of variance with the two factors phenotype ( fixed ) and experiment ( random ) . Variance was analyzed in SPSS 17 . 0 ( SPSS Inc . - Chicago , IL ) . Growth rates w/o reporter were analyzed via a maximum likelihood approach to test for two subpopulations with different growth rates . The growth rate measurements from five independent experiments ( 87 micro colonies ) were combined . Using maximum likelihood , we fitted a bi-modal distribution ( the sum of two normal probability density functions ) and a unimodal ( normal ) distribution , and compared the two fits with a likelihood ratio test using R software [58] . In Fig . 4 , cell growth and ttss-1 expression were analyzed using a modified version of the cell tracking software described in [9] . The first cell in each micro colony that could be observed over a whole division was used to analyze the statistical association between ttss-1 expression and the interval between two divisions ( by non-parametric correlation analysis using PASW Statistics 18 . 0 . 0 ) . 157 micro colonies were analyzed to estimate the fraction of micro colonies in which all cells , none of the cells , and a fraction of the cells expressed ttss-1 . These groupings were based on visual inspection of each micro colony . | Pathogenic bacteria require virulence factors to foster growth and survival of the pathogen within the host . Therefore , virulence factor expression is generally assumed to enhance the pathogen's fitness . However , most studies of pathogen fitness have been performed by averaging the phenotypes over large pathogen populations . Here , we have analyzed for the first time the fitness costs of virulence factor expression in a simple in vitro culture experiment using the diarrheal pathogen Salmonella enterica subspecies I serovar Typhimurium ( S . Typhimurium ) . TTSS-1 , the cardinal virulence factor for eliciting Salmonella diarrhea , is expressed by just a fraction of the clonal S . Typhimurium population . Surprisingly , time lapse fluorescence microscopy revealed that ttss-1-expressing S . Typhimurium cells grew at a reduced rate . Thus , the pathogen has to “pay” a significant “price” for expressing this virulence factor . This raises important questions about compensating mechanisms ( e . g . benefits reaped through TTSS-1 driven host-interactions ) ensuring successful propagation of the genotype . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"bacteriology",
"microbial",
"evolution",
"gene",
"expression",
"genetics",
"microbial",
"pathogens",
"biology",
"microbiology",
"microbial",
"growth",
"and",
"development",
"genetics",
"and",
"genomics"
] | 2011 | The Cost of Virulence: Retarded Growth of Salmonella Typhimurium Cells Expressing Type III Secretion System 1 |
The impacts of vaccination on the transmission of Rift Valley fever virus ( RVFV ) have not been evaluated . We have developed a RVFV transmission model comprising two hosts—cattle as a separate host and sheep and goats as one combined host ( herein after referred to as sheep ) —and two vectors—Aedes species ( spp ) and Culex spp—and used it to predict the impacts of: ( 1 ) reactive vaccination implemented at various levels of coverage at pre-determined time points , ( 2 ) targeted vaccination involving either of the two host species , and ( 3 ) a periodic vaccination implemented biannually or annually before an outbreak . The model comprises coupled vector and host modules where the dynamics of vectors and hosts are described using a system of difference equations . Vector populations are structured into egg , larva , pupa and adult stages and the latter stage is further categorized into three infection categories: susceptible , exposed and infectious mosquitoes . The survival rates of the immature stages ( egg , larva and pupa ) are dependent on rainfall densities extracted from the Tropical Rainfall Measuring Mission ( TRMM ) for a Rift Valley fever ( RVF ) endemic site in Kenya over a period of 1827 days . The host populations are structured into four age classes comprising young , weaners , yearlings and adults and four infection categories including susceptible , exposed , infectious , and immune categories . The model reproduces the 2006/2007 RVF outbreak reported in empirical surveys in the target area and other seasonal transmission events that are perceived to occur during the wet seasons . Mass reactive vaccination strategies greatly reduce the potential for a major outbreak . The results also suggest that the effectiveness of vaccination can be enhanced by increasing the vaccination coverage , targeting vaccination on cattle given that this species plays a major role in the transmission of the virus , and using both periodic and reactive vaccination strategies . Reactive vaccination can be effective in mitigating the impacts of RVF outbreaks but practically , it is not always possible to have this measure implemented satisfactorily due to the rapid onset and evolution of RVF epidemics . This analysis demonstrates that both periodic and reactive vaccination ought to be used strategically to effectively control the disease .
Rift valley fever ( RVF ) is a mosquito-borne viral zoonosis that causes periodic outbreaks accompanied by low-level virus activity during inter-outbreak periods mainly in sub-Saharan Africa [1] . The disease mainly affects sheep and goats , cattle and camels [2] . Humans can be exposed following a bite from an infected mosquito and or through direct contact with tissues from infected animals [3] . The disease was initially reported in restricted regions in Africa but has progressively spread to almost the whole continent , the island of Madagascar and the Arabian Peninsula [1][4] . The disease outbreaks often occur when favourable environmental drivers such as elevated and widespread rainfall and flat topography that promotes flooding [5] develop in areas where there are susceptible hosts [2] and competent mosquito vectors [6] and predisposing socio-economic practices such as herd replacement patterns [7] . Climatic factors seem to play a more dominant role as almost all historical outbreaks have been associated with cyclical patterns of the El Niño/Southern Oscillation ( ENSO ) phenomenon , which results in elevated and widespread rainfall over the Greater Horn of Africa ( GHA ) [5] . The outbreaks are often associated with adverse public health and economic impacts [8][9][10][11] as well as social impacts . Specifically , on a macroeconomic scale , Rich and Wanyoike [11] estimated that the 2006/2007 RVF outbreak in Kenya generated losses of over Ksh 2 . 1 billion ( US$32 million then ) on the Kenyan economy . The continued occurrence and geographical spread of RVF outbreaks points toward the need to understand the dynamics of the outbreaks as well as explore the approaches to their control . Following the 2006/2007 RVF outbreak in Kenya , a retrospective analyses of the implemented responses revealed systematic delays due to the failure of the relevant institutions to recognize risk factors , act on early warnings messages ( until the initial human cases were confirmed approximately two months after cases were observed in livestock ) [12] , and identify appropriate interventions . Consequently , stakeholders and decision-makers from the GHA region developed a risk-based Decision Support Framework ( DSF ) [12] that could be used to guide responses to similar emergencies in the future [12] . Livestock vaccination is one of the measures that were identified in the framework given that it has a good potential to reduce the impacts of the disease in livestock , contamination of the environment and subsequent exposure to humans [13] . There are many challenges that affect successful utilization of vaccines in the management of RVF outbreaks . First , the inter-outbreak period of the disease ( approximated at 3–7 years [14] ) is much longer than the shelf life of the currently available vaccine ( Smithburn vaccine; 4 years ) [12] . This discourages vaccine manufacturers from maintaining large stocks of these products given the risk of losing a large proportion of them through expiry . Most of these vaccines are often manufactured on order , for example , when the risk of an outbreak heightens . Secondly , the heavy rains and flooding that characterizes the high risk periods limit access and hence the delivery of vaccines to the rural areas . Thirdly , livestock species that are highly susceptible to the disease and hence would benefit from vaccination ( such as goats and sheep ) have a high population turn-over rates , limiting the maintenance of herd immunity especially in the pastoral areas . These challenges indicate an urgent need for policies that can guide utilization of RVF vaccines . Mathematical models for simulating RVF epidemics have been developed [15] [16] [17] [18] . However , most of them are not suitable for evaluating vaccination strategies because they do not incorporate ( i ) climate variability ( mainly precipitation changes ) which greatly influences the timing of vaccination and other reactive interventions , and ( ii ) livestock population dynamics which influence the duration of herd immunity . We develop a model comprising two hosts—cattle as a separate host and sheep and goats as one combined host—and two vectors—Aedes species ( spp ) and Culex spp . Consequently , the model incorporates these components and use it to address policy-relevant questions on the effectiveness of reactive and periodic vaccination strategies including: ( 1 ) How can various vaccination coverages ( VCs ) implemented at different times before an outbreak affect the size of an outbreak in livestock ? ( 2 ) To what extent is it possible to reduce outbreak size in both livestock species by focusing vaccination on one species ? ( 3 ) How can periodic vaccination be used together with reactive vaccination particularly in the high risk areas ? We incorporate two hosts with the recognition that pathogens such as RVFV that can infect multiple host species have different dynamics than single-host pathogens . Faced with scarcity of host-specific transmission parameters , this study sets the stage for the understanding of pathogen transmission dynamics and cost-effective control of RVF in multihost disease systems .
In developing the model , we make the following assumptions:
Two probability distributions generated using the fuzzy and logistic regression models based on TRMM rainfall values are successfully used to drive Aedes and Culex mosquito populations , respectively . Fig 2 shows the temporal relationship between these probability distributions and the respective vector:host ratios . In general , peaks in vector:host ratios lag those of fuzzy and logistic probability distributions by approximately 8–17 days and 30 days , respectively . Between days 9256 and 9450 when there was heavy/persistent rainfall , the fuzzy and logistic regression models generated high probability values which led to an upsurge in the mosquito populations , hence high vector:host ratios ( Fig 2 ) . The other wet seasons before this had short-lived precipitation events that were not adequate to support an upsurge of the Culex mosquito population though that of Aedes mosquitoes responded positively . In the simulated outbreak , Aedes adults that emerge from infected eggs , last for a total of 148 days and peak at day 80 . Susceptible Aedes mosquitoes also develop at the same time peaking on day 87 . Culex mosquito population appears 36 days after the emergence of Aedes population . Culex mosquitoes gain RVFV infection from viraemic hosts from day 69 after initial transmissions by Aedes spp . The maximum FoI exerted to Aedes spp from cattle and sheep are 0 . 016 and 0 . 006 respectively . The maximum FoI exerted to Culex spp from cattle and sheep are 0 . 015 and 0 . 0057 respectively . Predicted RVFV incidence in hosts is shown in Fig 3 . These predictions show five transient RVFV transmissions associated with seasonal rains and one main outbreak associated with heavy and persistent precipitation . In general , seasonal transmission events fail to result in full-blown outbreaks given that no amplification of populations of Culex spp occurs ( Fig 2 ) . The outbreak curve has a characteristic shape–RVFV activity begins slowly until Culex spp population surges , resulting in the amplification of the virus . The predicted peak outbreak incidence of RVFV in cattle is 12% on day 112 of the outbreak while that for sheep is 8% on day 123 . The predicted duration of the outbreak is 184 days . The maximum force of infection exerted to cattle and sheep are 0 . 24 and 0 . 06 , respectively . A simulated RVF outbreak in this study was defined by noting the predicted peak endemic incidence in hosts . The peak endemic incidence was used as the threshold for definition of an outbreak . By comparing endemic verses epidemic patterns predicted in the model , it appears that the number of cases predicted during the outbreak captured is 80% more than those predicted for the endemic periods . We use the 80% threshold for evaluating impacts of the various vaccination scenarios being studied . Sensitivity analyses showed that the infectious period and infectivity in both hosts and vectors ( particularly Culex spp ) were sensitive to the cumulative incidence of RVF . Others included survival and mosquito biting behavior of Culex spp ( Table 2 ) .
We present a deterministic model that combines precipitation patterns , mosquito population dynamics and host demographics to simulate RVFV transmission . The model predicts elevated RVFV activity during the wet seasons as well as a full-blown RVF outbreak following periods with excessive and persistent precipitation . Elevated and persistent rainfall is a risk factor for RVF outbreaks—all the 11 reported RVF outbreaks in Kenya occurred in years when the average annual rainfall increased by more than 50% in the affected districts [14] . The novelty of the model is in the bridging of separate probability distributions that uses satellite-derived daily precipitation for the study area that ensure temporal succession of separate vector species population growths . Since we are not interested in the importance of trans-ovarial transmission and its implications on the generation of the outbreak [17] , we exclude these detailed dynamics in Aedes mosquitoes . Adult Aedes mosquito emergence events are dependent on water ( rainfall ) that inundates breeding habitats [21] . We , therefore , base the dynamic distribution of Aedes species on accumulated rainfall amounts using a fuzzy distribution model similar to that employed by Emert et al . ( 2011 ) [20] . The fuzzy distribution model computes dynamic suitability conditions of hatching of Aedes eggs that mimic the reported strong relationship between Aedes mosquito emergence and weather ( rainfall ) variability [21] . The assumptions driving the fuzzy distribution model , as described in the Methodology section , seem rational and might denote a qualitatively plausible relationship of Aedes egg hatching process than a simple linear function of rainfall . Culex mosquito population dynamics are driven using an approach of obtaining parameters from a statistical analyses of reports of livestock cases and a particular pattern of rainfall during the 2006/2007 outbreak . We used this function based on empirical studies that reported that the mosquito breeding sites were colonized by massive swarms of Culex ( and other species ) if they remained flooded for at least 28–42 days [21] . Additionally , livestock keepers in the study area reported a mean average of 23 days between the start of heavy rains and the appearance of mosquito swarms during the 2006/2007 RVF outbreak [8] , though most likely these included both primary and the secondary species . Our model accurately captures this temporal relationship between cumulative rainfall and secondary mosquito species emergence . A different approach of growing seasonal vector populations in modelling RVFV transmission in West Africa was implemented by Soti et al . [37] using a hydrology model . Their hydrological model uses daily rainfall as input to simulate variations of water pool surface areas . We have not used this approach as the epidemiology of RVF occurrence in West Africa and GHA is different . Whereas in GHA RVF outbreaks are known to be closely associated with ENSO phenomenon [5] , periods of RVF outbreaks in West Africa do not necessarily coincide with years of highest total rainfall [38] . Indeed , RVF epidemiological landscape in West Africa is influenced by the generation of temporary ponds and a particular rainfall temporal distribution ( populations of Aedes and Culex spp depend on the alternation of rainy and dry periods ) [38] . Although rainfall , just as in GHA , is the main driver of hydrologic dynamics of water pools in West Africa , the mechanistic vector productivity of specific habitats and RVFV transmission and the consequent epidemiological inference in the two ecologies can be substantially different . Empirical studies are needed in the two distinct ecologies to accurately quantify the amount and distribution of rainfall regimes ( and how they interact with soil infiltration rates ) required for hatching of primary vectors . We implement the legendary assumption which considers primary and secondary vectors playing a synergistic role in generation of RVF outbreaks . Innovative ways of empirically examining these assumptions are needed to answer questions such as whether primary vectors alone [17] or whether secondary vectors alone ( for example , if augmented with movements of animals ) [7] can drive RVF full-blown outbreaks . In addition , this model hypothesizes that water availability may play a more dominant role in driving the vectors population dynamics . Future model refinements should incorporate not only the effects of temperature and humidity , vegetation and nutrient competition on vector population dynamics but also on the extrinsic incubation periods of RVFV in vectors [39] . For RVF control to be evaluated and optimum control strategies devised , an increased understanding of the transmission dynamics among hosts and vectors is paramount . In this way , we apply the model to identify the key factors driving the number of potentially averted RVF cases in a simulated outbreak . The analyses show that vaccination , as a sole intervention , can be effective in mitigating the impacts of RVF outbreaks . The success of RVF vaccination is predicted to be defined by the targeted vaccination coverage and the time to the outbreak . The proportion of cases averted is related to the targeted vaccination coverage , particularly for low levels . The policy implication of this prediction is that resources and planning required to achieve a given VC corresponds to the number of cases expected to be averted . For a given VC , higher herd immunity at the outbreak onset is predictably highly beneficial . Vaccinating early reduces herd immunity , over time , through removal of immune animals via expected mortality and offtake and birth of susceptible animals . The model predicts that 3–6% more cases can be averted if , for the simulated VCs , vaccination is implemented close to the outbreak . Averting 3–9% more cases can lead to large numbers of deaths being averted particularly in the more RVF-induced mortality susceptible species such as sheep . For greater effectiveness , this prediction implies that a careful balance between a given VC and optimal timing is critical . These predictions concur with recent modelling study predictions that a higher rate of vaccination may help to reduce the epidemic size and a maximal attempt of vaccination just before an outbreak is highly beneficial [18] . In sub-Saharan Africa , vaccination against RVFV has been used for many years either to prevent disease occurrence [40] or to mitigate disease impacts [41] . Our model predictions clearly demonstrate the usefulness of effective implementation of this intervention . Ideally , however , all members of a population need not be vaccinated because as the number of susceptible hosts in the population is reduced , the efficiency with which a pathogen is transmitted is greatly reduced ( the concept of herd immunity ) [42] . The model predicts that this indirect protection is accelerated as vaccination coverage is increased and , moreover , it is experienced more in sheep relative to cattle . Early and optimal timing , in turn , depends upon a sensitive and functioning RVF surveillance and prediction system and a rapid response capacity by the national veterinary authorities [40] . One such surveillance system integrates ENSO related climate anomalies including elevated sea-surface temperatures and satellite-derived normalized difference vegetation index data ( NDVI ) [5] . During the 2006/2007RVF outbreak , this system retrospectively provided a 2 to 4 month period of warning in the GHA region [5] . However , the RVF DSF estimates the lead-time to order , produce , deliver sufficient vaccine to the field and attain herd immunity in livestock to be approximately 5 months [12] . This implies that vaccine orders need to be placed prior to the first RVF early warning . Currently , this is impractical unless the lead time for prospective predictions of RVF outbreaks is lengthened . Still , even if the latter were achieved ( to , e . g . 5 months ) , mobilizing adequate resources to procure the vaccines within the short period is a difficult task in resource-scarce countries in the GHA . Moreover , by this time , the co-occurrence of heavy rains and flooding in the rural areas coupled with the absence of all-weather roads can present huge logistical challenges in vaccine delivery . Innovative strategies are clearly needed as part of outbreak preparedness plan . To overcome some of these challenges , the RVF DSF proposes a strategic regional vaccine shared bank which could be rapidly deployed in times of need [12] . To supplement this proposition , we modeled a periodic vaccination strategy implemented under different vaccination coverage biannually or annually for 2 years in advance of an outbreak . The objective was to assess the impacts of these strategies in not only reducing the outbreak size but also the possibility of complementing them with a reactive strategy close to the outbreak onset . Complementing very low VCs biannually for two years and low reactive VCs is highly effective , e . g . a VC of 10% is predicted to completely avert an outbreak when integrated with a reactive VC of 35% . Annual vaccination is equally effective though at a lower scale . In a large livestock population , averting an outbreak could mean avoiding morbidity and mortality of thousands of animals , reducing vulnerability of local livestock-dependent livelihoods and national economies and , more importantly , reducing chances of virus exposure to humans . Rift Valley fever vectored vaccines are currently being developed [43] and evaluated [44] and this might change ( i ) the way these vaccines are administered in the field , i . e . , some could be given at biannual intervals and or others annual . These combinations can influence the efficacy of the RVFV component of the vaccine . As earlier highlighted in this paper , the shelf-life of current vaccines [12] is shorter than the average inter-outbreak period [14] which presents an economic disincentive to vaccine manufacturers in situations where reactive vaccination campaigns are planned . Similarly , resource-constrained governments are not keen on funding periodic vaccination campaigns partly due to unpredictability of occurrence of the outbreaks . Periodic vaccination campaigns are also a disincentive in situations where livestock population-turn over due to offtakes and expected mortality temporally leads to lower herd immunity . Our analysis is therefore well placed to give policy directions on how vaccination can be used to meet these challenges . Further evaluation of the response impact of integrating periodic and reactive vaccination strategies in preventing the occurrence of a RVF outbreak is an important area for future research and policy development . Multihost pathogens are more likely to have ecologically different dynamics than pathogens that infect only a single host species . In a host population , multiple host species can be viewed as a form of heterogeneity that partitions the total host population into subpopulations between which the FoI experienced by each host species and the FoI exerted by each host species varies [45] . Based on the assumptions we make in the model , the FoI experienced by cattle is larger due to their higher vector: host ratio arising from their lower number in the population and their assumed larger surface area relative to sheep . The assumptions are qualitatively realistic given the differences in host species’ exposed surface area which is obviously higher in cattle relative to sheep , all other parameters ( e . g . blood meal preference ) being constant . Similarly , the FoI experienced by vector species from cattle are higher than that from sheep . Consequently , based on our assumptions , the model predicts that cattle dominate the bi-directional RVFV transmission process between hosts and vectors . With such a pathogen ecological framework where a host species may dominate the virus transmission , we examined the possibility of directing control against either of the host species . Targeting cattle alone provided major protection to cattle and sheep . This benefit arises from the reduction of transmission of RVFV . However , targeting sheep alone provided protection to sheep alone . This prediction , if empirically validated has important policy implications for the implementation of both periodic and reactive vaccination strategies for two reasons: ( 1 ) cattle have longer lifespan and lower population turnover relative to sheep and , therefore , would be able to sustain herd immunity for longer , ( 2 ) in our case study area ( and indeed in all pastoral areas ) , cattle are fewer relative to sheep ( and goats ) and ( 3 ) in the pastoral communities , cattle are likely to be moved long distances translating to potential spatial spread of RVF compared with sheep and goats . These reasons can greatly influence the cost-effectiveness of a strategy that focuses control against cattle in the population . Our model , by necessity , includes a number of simplified assumptions about reality in a number of ways that have a bearing on the predictions . We have assumed transmission-related parameters in Table 1 to be similar in both host species . This implies that our outputs were based on two main parameters ( i ) the use of temporally varying FoI arising from seasonal growth of vectors , and ( ii ) the different numbers of host species in the population . However , model sensitivity analyses found that RVF cumulative incidence may be influenced most by infectious periods and infectivity in both hosts and vectors ( particularly Culex spp ) . Other sensitive parameters include survival and mosquito biting behavior of Culex species . The same parameters have been reported to be sensitive to similar outcomes in RVF modelling , e . g . Chitnis et . al [17] reported that an outbreak size was sensitive to vector-to-host ratio , mosquito biting rate and the infectivity of hosts . Other models [15][16][35] reported adequate contact rates between vectors and hosts and the rate of recovery livestock as sensitive to the basic reproduction number . However , their definition of adequate contact rates between vectors and hosts considered a composite term whereas in our model , we disaggregated the term into its individual components including the vector biting rate , host infectivity , blood meal index and vector host ratio . These findings suggest that apart from RVF vaccination , reducing the the probability of transmission from the vector to the host can be effective in RVF outbreak control . In addition , given the importance of understanding RVFV transmission processes , the lack of knowledge about the processes make gathering of relevant field and experimental data on these biological processes an urgent research priority . Further simplifying assumptions that we make in the analyses of vaccination impacts ignore the individual components that constitute the actual proportion of susceptible hosts vaccinated ( herein referred to as an ideal VC ( iVC ) ) . This can be obtained as a product of the proportion of vaccinations properly applied ( efficiency of vaccination ) and the probability that the vaccine would provide protection from infection ( the efficacy of the vaccine ) [46] , both of which limited data are available for RVF . Naturally , these two proportions are each less than 100% in most cases . Multiplicatively , the further the values are from 100% , the less the iVC . A recent study evaluated the effectiveness of RVF Clone 13 vaccine and reported that 67% of vaccinated cattle and between 91% and 97% of vaccinated sheep and goats develop protective antibodies to the vaccine [44] . Applying an efficiency of vaccination of 80% based on the performance of mass vaccination teams as assessed by the Pan African Rinderpest Campaign in pastoral areas in GHA [47] , vaccinating an entire population in our study would result to an iVC of approximately 54% in cattle and 76% in sheep . To achieve our simulated VCs , therefore , call for high levels of both the efficiency of vaccination and high efficacy of RVF vaccines such as that reported in sheep and goats [44] . Further explorations required include cost-effectiveness analyses taking account of integrating VCs and time to outbreak , integrating periodic and reactive strategies and directing interventions to one host species under different scenarios of efficiency of vaccination and efficacy of vaccines . In conclusion , our results suggest that targeted vaccination can be effective in mitigating the impacts of RVF outbreaks . However , it is not always possible to have this measure implemented satisfactorily due to the rapid onset and evolution of RVF epidemics . The analyses further demonstrates that both periodic and reactive vaccination ought to be used strategically to effectively control the disease . In addition , challenges associated with prediction of the outbreak , availability and delivery of vaccines need to be addressed . Factors driving the number of potentially averted cases include the targeted VC and timing of vaccination in relation to the time to the outbreak . Based on our assumptions , cattle appear to dominate RVF transmission between hosts and vectors . Predictably , directing vaccination against cattle , whether in a periodic and/or a reactive vaccination startegy , may be more effective as it confers herd immunity to both species . The work presented here advances our understanding of impacts of different vaccination strategies . We consider that these predictions provide a first step of information needed by policy makers to plan effective periodic and reactive strategies for mitigating the effects of RVF outbreaks . However , detailed cost-benefit analysis should be integrated with these findings to support decision-making and prioritize these strategies . | Evaluation of the relative impacts of RVF vaccination has not been previously carried out . We present a model that simulates RVFV transmission between two livestock hosts ( cattle as a separate host and sheep referring to both sheep and goats ) and two mosquito species ( Aedes and Culex species ) . We then apply the model to evaluate policy-relevant impacts of vaccinating ( 1 ) different proportions of animals at different times to the simulated outbreak , ( 2 ) either of the host species , and ( 3 ) different proportions of animals in a periodic biannual or annual vaccination preventative strategy . Vector population growth is dependent on rainfall extracted from the Tropical Rainfall Measuring Mission ( TRMM ) for an RVF endemic site in Kenya over a period of 1827 days . The model reproduces the 2006/2007 RVF outbreak reported in empirical surveys in the target area and other seasonal transmission events that occur during the wet seasons . Consistent with anecdotal evidence , mass livestock vaccination can greatly reduce the potential for a major outbreak . The model predicts that the effectiveness can be improved by increasing the proportion of vaccinated animals , targeting vaccination against cattle and strategically augmenting periodic preventative strategies with reactive strategies once a RVF outbreak is predicted . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"death",
"rates",
"invertebrates",
"livestock",
"medicine",
"and",
"health",
"sciences",
"rift",
"valley",
"fever",
"virus",
"pathology",
"and",
"laboratory",
"medicine",
"ruminants",
"demography",
"pathogens",
"immunology",
"vector-borne",
"diseases",
"microbiology",
"... | 2016 | Modelling Vaccination Strategies against Rift Valley Fever in Livestock in Kenya |
The relationship between parasite fitness and virulence has been the object of experimental and theoretical studies often with conflicting conclusions . Here , we provide direct experimental evidence that viral fitness and virulence , both measured in the same biological environment provided by host cells in culture , can be two unrelated traits . A biological clone of foot-and-mouth disease virus acquired high fitness and virulence ( cell killing capacity ) upon large population passages in cell culture . However , subsequent plaque-to-plaque transfers resulted in profound fitness loss , but only a minimal decrease of virulence . While fitness-decreasing mutations have been mapped throughout the genome , virulence determinants—studied here with mutant and chimeric viruses—were multigenic , but concentrated on some genomic regions . Therefore , we propose a model in which viral virulence is more robust to mutation than viral fitness . As a consequence , depending on the passage regime , viral fitness and virulence can follow different evolutionary trajectories . This lack of correlation is relevant to current models of attenuation and virulence in that virus de-adaptation need not entail a decrease of virulence .
The relationship between fitness and virulence is an unsettled question , and sometimes fitness is considered a component of the virulence phenotype of parasites . RNA viruses are ideal systems to address this important question because of their high mutability and fecundity , which result in a potential for rapid evolution , and also because of the availability of quantitative fitness and virulence assays . RNA viruses replicate as complex and dynamic mutant spectra , termed viral quasispecies . Key to quasispecies dynamics are mutation rates in the range of 10−3 to 10−5 substitutions per nucleotide copied , and competition among continuously arising variant genomes [1–4] , which prompt rapid movements in sequence space , with corresponding changes of position in the fitness landscape [5] . Indeed , large population passages of RNA viruses in cell culture permit competitive optimization of mutant distributions that generally result in fitness gain [6 , 7] , while repeated bottleneck events ( experimentally realized as plaque-to-plaque transfers ) lead to random accumulation of deleterious mutations ( operation of Muller's ratchet [8] ) and result in average fitness decreases [9–13] . Fitness recovery of low fitness foot-and-mouth disease virus ( FMDV ) clones occurs mainly with introduction of mutations along the genome , with very few true reversions . An understanding of the consequences of fitness variation for viral virulence is a key question for viral pathogenesis and evolution . Here , we approach this issue with FMDV , an important viral pathogen in veterinary medicine [14] , and one that fully participates of quasispecies dynamics . Our laboratory has characterized multiple FMDV variants that derive from one original biological clone , and that occupy widely different fitness levels when replicating in a defined environment in cell culture . We define fitness as the replication capacity of a mutant FMDV , relative to a reference FMDV , in direct growth-competition upon coinfection of baby hamster kidney 21 ( BHK-21 ) cells [15–17] . Fitness of FMDV in BHK-21 cells is a multigenic trait [7] . In the present study , we define virulence of FMDV as the capacity of the virus to kill BHK-21 cells under a standard set of cell culture conditions [18] . Thus , the FMDV–BHK-21 system offered a means to investigate in a direct and comparative fashion the relationship between fitness and virulence of a virus , measured in the same biological environment provided by BHK-21 cells . We describe the behavior of an FMDV clone ( ) , which has a history of repeated serial plaque-to-plaque transfers in BHK-21 cells [11] , that attained a very low fitness value relative to its parental reference virus ( C-S8c1 ) , and yet , its virulence for BHK-21 cells was significantly higher than that of C-S8c1 . A comparative study of the capacity to kill BHK-21 cells of chimeric FMDVs constructed with cDNA copies of the two parental FMDVs indicates that the enhanced virulence for BHK-21 cells of the low fitness clone is a polygenic trait , with participation of the regions encoding capsid proteins and non-structural proteins 2A , 2B , and 2C as virulence determinants . Three specific amino acid replacements in 2C have been identified as redundant virulence determinants of FMDV for BHK-21 cells . Thus , while large population passages of the virus resulted in a gain of both fitness and virulence , subsequent bottleneck passages resulted in a decrease of fitness but not of virulence . The results suggest that fitness is very vulnerable to mutation in any genomic region . In contrast , because of the involvement of several ( but not all ) viral genes in virulence , and the redundant effect of three 2C substitutions , virulence is a more robust phenotypic trait than fitness , and less vulnerable to accumulation of mutations . Therefore , we provide direct evidence that viral fitness and capacity to kill cells can ( in some cases ) be unrelated traits . Furthermore , the relationship between fitness and virulence , of being either linked or unrelated traits , depends on the evolutionary history of the virus . This observation has implications for viral pathogenesis , and sheds light on models of virulence proposed on the basis of theoretical and experimental studies with cellular organisms .
Several biological clones and populations were obtained by passaging FMDV biological clone C-S8c1 [19–22] in BHK-21 cells , either as large population passages or plaque-to-plaque transfers ( Figure 1 ) . The biological clones and populations differed up to 236-fold in relative fitness ( Table 1 ) . The fitness differences found are expected from previous results on fitness gain upon large population passages of RNA viruses [6 , 7] and fitness decrease upon plaque-to-plaque ( bottleneck ) transfers [9–13] . The initial experiment was aimed at testing whether , because of its low fitness ( 0 . 11 times that of its parental C-S8c1 [12 , 23 , 24] [Table 1] ) , had an advantage in establishing a persistent non-cytopathic infection in BHK-21 cells as compared with its parental clone , C-S8c1 ( Figure 1 ) . A persistent FMDV infection is established by growing the cells that survive a standard cytolytic infection with FMDV [25] . Confluent monolayers of BHK-21 cells were infected either with C-S8c1 or with at a multiplicity of infection ( MOI ) of 0 . 02–0 . 1 plaque-forming units ( PFU ) /cell ( 2 × 106 cells infected with 4 × 104 −2 × 105 PFU ) . Unexpectedly , at 24 h postinfection , the cells infected with showed extensive cytopathology , and at 48 h postinfection , no surviving cells were observed . The frequency of surviving cells in parallel infections with C-S8c1 was 5 × 10−3–9 × 10−3 , which is consistent with previous determinations [25] . No persistently infected BHK-21 cell cultures could be established with , despite several attempts . Thus , C-S8c1 , which displays a 9-fold higher relative fitness than in BHK-21 cells , showed a capacity to kill BHK-21 cells that was at least 103-fold lower than the killing capacity of in the infectivity assay intended to establish a persistent FMDV infection . The capacity of to kill BHK-21 cells despite its low fitness in BHK-21 cells led us to quantitatively examine the relationship between fitness of FMDV and its capacity to kill BHK-21 cells . To this aim , FMDV clones or populations were compared in a cell killing assay , consisting in determining the time required to kill 104 BHK-21 cells as a function of the PFU added ( described in Materials and Methods ) . The results ( Figure 2A ) indicate that over the time range of 12 h to 48 h postinfection , the number of PFUs needed to kill 104 BHK-21 cells varied logarithmically as a function of time . Similar quantifications of relative virulence were obtained by measuring the PFU needed to kill 104 cells in 24 h , and then by extrapolating the PFU values to 0 h postinfection ( Tables 1 and S1 ) . Virulence of was 29 to 35 times higher than virulence of C-S8c1 , despite the latter displaying a 9-fold higher fitness ( Tables 1 and S1 ) . The high virulence of was not due to the plaque-to-plaque transfers , since a high virulence was also quantitated for its parental clone , , and for population C-S8p113 ( Figure 2B; Tables 1 , 2 , and S1 ) . deviated from a line that correlated relative fitness of FMDV and the logarithm of cell killing capacity , as reflected in the decrease of the regression coefficient ( R2 ) ( inset in Figure 2A ) . Probably , this deviation is due to the fact that lost fitness due to plaque-to-plaque transfers , and the other viruses were not subjected to plaque-to-plaque transfers . On the other hand , virulence determinants were acquired during the large population passages done between C-S8c1 and C-S8c1p113 . The 29- to 35-fold higher virulence of with respect to C-S8c1 ( Tables 1 and S1 ) , despite its low fitness , indicates that viral fitness and virulence can be two unrelated traits . The comparison of the consensus nucleotide sequence of the genome with that of C-S8c1 revealed a total of 47 mutations ( Table S2 ) , leading to 21 amino acid replacements affecting structural and non-structural proteins ( Figure 3 ) . To identify the genomic regions associated with the increased virulence of with respect to C-S8c1 , we measured the BHK-21 cell killing capacity of nine chimeric viruses rescued from constructs obtained by introducing fragments of cDNA of the genome into plasmid pMT28 , which encodes infectious C-S8c1 RNA [21] ( Figure 4 ) . The results ( Figure 5; Tables 2 and S1 ) show that several genomic regions contribute to the virulence of for BHK-21 cells , and that the major contributors map within genomic positions 2046 to 3760 ( residues encoding part of VP2 , VP3 , and part of VP1 , Figure 5A ) and 3760 to 5839 ( residues encoding 2A , 2B , 2C , and 3A , Figure 5B ) . The results exclude the internal ribosome entry site and the 3C- and 3D-coding regions as significant virulence determinants of for BHK-21 cells ( virulence of the relevant chimeric viruses ≤ 2 . 5 , relative to C-S8c1; Tables 2 and S1 ) . Infectious progeny production by each chimeric virus was intermediate between the production of the parental viruses pMT28 and , with no significant differences that could be correlated with virulence ( Table 2 ) . Amino acid substitutions in human rhinovirus protein 2C promoted cytopathology for mouse L cells [26] . Remarkably , shares with other FMDV clones and populations , notably , MARLS and C-S8p260p3d ( the two viruses showing the highest virulence for BHK-21 cells; Figure 2A; Tables 1 and S1 ) , three amino acid substitutions in 2C: S80N , T256A , and Q263H . In addition , MARLS and CS8p260p3d include replacement M283V in 2C , relative to C-S8c1 [27 , 28] . To test whether any ( or a combination ) of the three shared amino acid substitutions in 2C contributed to the increased virulence of FMDV , each of the mutations was introduced individually into plasmid pMT28 by site-directed mutagenesis , as described in Materials and Methods . Transcripts of the three mutants , termed pMT28 ( SN ) , pMT28 ( TA ) , and pMT28 ( QH ) ( Figure 4 ) , were used to transfect BHK-21 cells , and the viruses obtained were tested with the BHK-21 cell killing assay . Viruses having any of the substitutions in 2C have a virulence intermediate between that of C-S8c1 and ( Figure 6A ) . To test whether the combination of the three substitutions in 2C could produce an additional increase of virulence , the three mutations were introduced in pMT28 to rescue the triple mutant pMT28 ( SN , TA , QH ) ( Figure 4 ) . The results ( Figure 6B ) show that the virulence of the triple 2C mutant is similar to the virulence of the individual 2C mutants . The 2C mutations did not significantly affect the infectious progeny production ( Table 2 ) . A testable prediction of this result is that the introduction of the wild-type 2C-3A-coding region in the genetic background of should produce a virus with lower virulence than . Indeed , the results with such a chimeric virus ( Figure 5D ) indicate that the presence of the 2C- and 3A-coding region as the only genetic region of the pMT28 in the genetic background of resulted in an FMDV with a 2 . 4- to 4 . 8-fold lower virulence than . We conclude that mutations in 2C contribute to virulence of FMDV for BHK-21 cells . Thus , a virus that evolves towards low fitness levels due to the operation of Muller's ratchet may nevertheless maintain its capacity to kill the same cells in which it displays low fitness . In FMDV , the enhanced capacity to kill BHK-21 cells was multigenic , including participation of non-structural protein 2C with three amino acid substitutions acting in a redundant fashion . In conclusion , the results provide a molecular interpretation of why fitness and virulence of an animal virus can follow disparate evolutionary trajectories , culminating in two unrelated traits .
The results with FMDV clones H5 have documented that both fitness-enhancing and virulence-enhancing mutations can be incorporated in the viral genome in such a fashion that subsequent fitness-decreasing mutations associated with bottleneck ( plaque-to-plaque ) transfers produce only minimal effects on virulence ( Figure 2 ) . The dissection of accompanying molecular events , achieved through quantification of virulence of recombinant and mutant genomes ( Tables 2 and S1 ) , provides an interpretation of the lack of positive correlation between virulence and fitness . Multiple fitness-decreasing mutations occur in the course of plaque-to-plaque transfers , distributed throughout the FMDV genome [11] . In contrast , determinants of virulence for BHK-21 cells are multigenic , but concentrated mainly in some FMDV genomic regions . Similar multigenic but discrete virulence determinants have been described also in other virus–host systems [42 , 43] . To decrease virulence , mutations occurring randomly in the course of plaque-to-plaque transfers should affect specific genomic sites , and this will occur with a lower probability than fitness-decreasing mutations , which can hit any of the multifunctional picornaviral proteins and regulatory regions [11] . This model is reinforced by the observation that three amino acid substitutions in 2C ( S80N , T256A , and Q263H ) had a similar effect in enhancing FMDV virulence , and the three mutations in the same genome had an effect comparable to each mutation individually ( Figure 6; Tables 2 and S1 ) . It is not clear what the basis of the contribution of 2C to virulence for BHK-21 could be . 2C is involved in RNA synthesis and contains a nucleotide-binding domain , although none of the substitutions found in and lie within such a domain . An unlikely triple reversion would be required to eliminate the virulence-enhancing effect of the three mutations in 2C . We propose that a higher robustness of the FMDV genome with regard to virulence for BHK-21 cells , rather than to replicative fitness in the same cells , underlies the different trajectories followed by fitness and virulence upon subjecting the virus to repeated bottleneck transfers . Obviously , we cannot exclude that parameters of the virus life cycle , other than fitness as measured in our experiments , could correlate with virulence for BHK-21 . The comparative analysis of FMDV clones and populations shows that shifts in virulence can occur even through the evolution of a single viral clone ( C-S8c1 ) , with its restricted genetic diversity prompted by different replication regimes in the same host cells , which also have a clonal origin ( see Materials and Methods ) . We conjecture that the demonstration that fitness and virulence can follow different evolutionary courses has been possible thanks to the consequences of the extreme passage regimes to which the viral populations were subjected: competitive evolution of an ample mutant spectra during repeated large population passages , and accumulation of deleterious ( with regard to fitness , but not with regard to virulence ) mutations upon plaque-to-plaque transfers ( predominance of genetic drift and operation of Muller's ratchet ) [12 , 15] . It must be emphasized that fitness and virulence are relative values that pertain to a defined physical and biological environment . Virulence determinants of FMDV , identified here for BHK-21 cells , need not apply to virulence for the natural animal hosts of FMDV [44] . However , the observation of a lack of correlation between fitness and virulence in a FMDV clone is relevant to current models of attenuation and virulence , since it shows that more virulent forms of a virus need not have a reproductive advantage , and that viral virulence is not necessarily a byproduct of viral fitness . Even if virulence is regarded as an unavoidable consequence of parasite adaptation [45] , virus de-adaptation ( fitness loss ) need not entail a decrease of virulence . Most current definitions of virulence include both the ability of the pathogen to multiply and to cause harm to its host; some authors , however , assume a direct relationship between fitness and capacity to produce disease [46–48] . In relating the results with FMDV to general models of virulence in host–parasite systems , it must be considered that in the FMDV system , evolution of the host BHK-21 cells could not influence FMDV evolution , because clonal cell populations with a controlled passage history were supplied in constant numbers at each infection event ( see Materials and Methods ) . Therefore , changes in host density , or mobility , as well as pathogen survival in the external environment , all of which are relevant parameters in virulence models [48 , 49] , cannot play a role in our system . A consistent finding in serial passage experiments is that virulence of a parasite increases with passage number in a new host [50] . The results with FMDV infecting BHK-21 cells cytolytically imply that the increase of virulence can be conditioned to the history of passage regimes undergone by a virus . The invariance of BHK-21 cells in the course of serial cytolytic passages of FMDV is in contrast with the parallel system consisting of BHK-21 cells persistently infected with FMDV C-S8c1 [25] , in which the cells are passaged and coevolve with the resident virus [51] . Host–virus coevolution has generally favored a decrease of viral virulence in the field , a classical example being myxoma virus and myxomatosis in rabbits [52] . Our comparison of FMDV clones did not provide evidence of clones with high fitness and low virulence , which , with regard to natural hosts , is an aim of biomedicine to obtain vaccine strains . Yet , the existence of specific mutations that differentially affect fitness and virulence opens the way to engineer candidate vaccine strains unable to kill the host , while maintaining replicative competence . Virulence is , however , a feature of the host–parasite relationship [46] , and the mutations needed to impair virulence are expected to be host-dependent [53 , 54] .
The BHK-21 cells used in the present study were cloned by end-point dilution , followed by preparation of a cell stock from a single cell; they were passaged a maximum of 30 times before being used for FMDV infection [25 , 51] . Procedures for cell growth , infection of BHK-21 cell monolayers with FMDV in liquid medium , and plaque assays in semi-solid agar medium were carried out as previously described [11 , 19 , 25 , 27] . Mock-infected cells were handled in parallel in all infectivity and plaque assays to monitor absence of viral contamination . The FMDVs used in the present study ( Figure 1 ) are ( i ) the reference clone C-S8c1 , which has been assigned a relative fitness of 1 . 0 [11] . ( ii ) MARLS , a monoclonal antibody escape mutant isolated from population C-S8c1p213 [55]; MARLS has a fitness of 25 relative to C-S8c1 [24] . ( iii ) C-S8p260p3d , a standard FMDV virus rescued by low MOI passage of C-S8p260 . The latter is a virus that evolved by passage of C-S8c1 at a high MOI , which resulted in dominance of two defective FMDV genomes ( both including internal deletions ) that were infectious by complementation , in the absence of standard virus [22 , 24 , 28]; C-S8p260p3d has a relative fitness of 20 [24] . ( iv ) REDpt60 , obtained after 60 successive plaque-to-plaque transfers of RED ( a monoclonal antibody escape mutant isolated from population C-S8c1p100 ) [20]; REDpt60 has a fitness of 1 . 9 relative to C-S8c1 . ( v ) C-S8c1p113 , a viral population obtained after 113 serial cytolytic passages of C-S8c1 at a high MOI in BHK-21 cells ( 2 × 106 BHK-21 cells infected with the virus contained in 200 μl of the supernatant from the previous infection ) . ( vi ) Clone , a biological clone isolated from population C-S8c1p113 [11]; its relative fitness is 26 ( unpublished data ) . ( vii ) Clone , obtained after 95 successive plaque-to-plaque transfers of [11]; its relative fitness is 0 . 11 [23] . The capacity of FMDV to kill BHK-21 cells was measured as previously described [18 , 22] . The assay consists in determining the minimum number of PFU required to kill 104 BHK-21 cells after variable times of infection . The assay was performed in M96 multiwell plates with monolayers of 104 BHK-21 cells per well infected with serial dilutions of virus . At different times postinfection , cells were fixed with 2% formaldehyde and stained with 2% crystal violet in 2% formaldehyde . Results are expressed as the logarithm of the number of PFUs needed for complete cell killing ( as judged by cell staining with crystal violet , with series of control wells with known numbers of cells ) as a function of time postinfection [18 , 22] . The relative fitness of FMDV was determined by growth competition in BHK-21 cells as previously described [7 , 10 , 11 , 24 , 56] . FMDV was mixed with appropriate proportions of , which was used as reference virus . This virus has a fitness 8 . 5-fold higher than that of the reference clone of C-S8c1 in BHK-21 cells [7 , 10 , 11 , 24 , 56] . Four serial infections were carried out at MOI of 0 . 1 PFU/cell . The proportion of the two competing genomes at different passages was determined by real-time reverse transcription ( RT ) –PCR , employing primers 5531wtnew and , which are able to discriminate FMDV RNA from RNA . The nucleotide sequences of the primers will be provided upon request . The fitness vector obtained for corresponded to the equation y = 0 , 0206e1 , 1074x; R2 = 0 . 9507 . The antilogarithm ( base e ) of the vector slope is the fitness of the assayed virus relative to the reference virus [56] . Viral RNA was extracted by treatment with Trizol as previously described [57] . Reverse transcription of FMDV RNA was carried out with avian myeloblastosis virus reverse transcriptase ( Promega , http://www . promega . com ) or Transcriptor reverse transcriptase ( Roche , http://www . roche . com ) , and PCR amplification was performed by using either Ampli-Taq polymerase ( PerkinElmer , http://las . perkinelmer . com ) or an Expand High Fidelity polymerase system ( Roche ) , as instructed by the manufacturers . The FMDV genome-specific oligonucleotide primers used have been previously described [22 , 58] . In all RT-PCR amplifications , negative amplification controls , including all reaction components except template RNA , were run in parallel to monitor absence of contamination . Chimeric viruses containing selected regions of in the genetic background of C-S8c1 ( Figure 4 ) were obtained by replacing the corresponding DNA fragment of pMT28 by a cDNA copy of RNA , using specific restriction sites . To obtain pMT28/ ( 436-2046 ) , a chimera that included nucleotides 436 to 2046 of ( the residue numbering of the FMDV genome is as in [11] ) , RNA was amplified by RT-PCR using primers NR2 and JH2 , and the cDNA was digested with Hpa I ( position 436 ) and Xba I ( 2046 ) , and ligated into pMT28 DNA digested with the same enzymes . To obtain pMT28/ ( 2046–3760 ) , RNA was amplified by RT-PCR using primers 2R1New and pU , and then the cDNA was digested with Xba I ( 2046 ) and Avr II ( 3760 ) . To obtain pMT28/ ( 3760–5839 ) , RNA was amplified by RT-PCR using primers 3R2New and 3CD1 , and then the cDNA was digested with Avr II ( 3760 ) and Rsr II ( 5839 ) . To obtain pMT28/ ( 5839–7427 ) , RNA was amplified by RT-PCR using primers 5531 wt new and C-Not-Pol , and then the cDNA was digested with Rsr II ( 5839 ) and Bam HI ( 7427 ) . To obtain pMT28/ ( 436-3760 ) , RNA was amplified by RT-PCR using primers NR2 and JH2 , and then the cDNA was digested with Hpa I ( position 436 ) and Xba I ( 2046 ) , and ligated into pMT28/ ( 2046–3760 ) DNA digested with the same enzymes . To obtain pMT28/ ( 3760–7427 ) , RNA was amplified by RT-PCR using primers 3R2New and 3CD1 , and then the cDNA was digested with Avr II ( 3760 ) and Rsr II ( 5839 ) , and ligated into pMT28/ ( 5839–7427 ) DNA digested with the same enzymes . To obtain pMT28/ ( 2046–7427 ) , RNA was amplified by RT-PCR using primers 2R1New and pU , and then the cDNA was digested with Xba I ( 2046 ) and Avr II ( 3760 ) , and ligated into pMT28/ ( 3760–7427 ) DNA digested with the same enzymes . To obtain pMT28/ ( 436-7427 ) , RNA was amplified by RT-PCR using primers NR2 and JH2; the cDNA was digested with Hpa I ( position 436 ) and Xba I ( 2046 ) , and ligated into pMT28/ ( 2046–7427 ) DNA digested with the same enzymes . To obtain /2C-3A ( pMT28 ) , pMT28 was digested with Bgl II ( 4201 ) and Rsr II ( 5839 ) , and a DNA fragment including wild-type 2C-3A-coding region was inserted into pMT28/ ( 436-7427 ) DNA digested with the same enzymes . DNA ligation , transformation of Escherichia coli DH5α , isolation of DNA from bacterial colonies , and characterization of DNA by restriction enzyme digestion were performed by standard procedures [59] . The primers used for molecular cloning and site-directed mutagenesis are described in Table S3 . To obtain FMDV C-S8c1 containing the mutations found in gene 2C of , plasmid pMT28 was subjected to site-directed mutagenesis using an oligonucleotide including the required nucleotide replacement , and 3R2New or 3CD1 as external oligonucleotide primer ( Table S3; Figure 4 ) . A DNA fragment termed A was obtained by subjecting plasmid pMT28 to site-directed mutagenesis using primers ( reverse ) mutSNu , mutTAu , and mutQHu ( to introduce mutations S80N , T256A , and Q263H , respectively ) and an external oligonucleotide primer ( 3R2New , forward ) . A DNA fragment termed B was obtained amplifying pMT28 with primers ( forward ) mutSNd , mutTAd , and mutQHd ( to introduce mutations S80N , T256A , and Q263H , respectively ) and an external oligonucleotide primer ( 3CD1 , reverse ) . DNA fragments A and B , including the desired mutations , were recombined by shuffling PCR using equimolar amounts of DNA fragments and two external primers ( 3R2New and 3CD1 ) . The DNA with the desired mutation ( s ) in the 2C gene was digested with Avr II ( genomic position 3760 ) and Rsr II ( position 5839 ) , and cloned into pMT28 to generate pMT28 ( SN ) , pMT28 ( TA ) , and pMT28 ( QH ) . To obtain pMT28 ( SN , TA , QH ) , plasmid pMT28 ( SN ) was subjected to site-directed mutagenesis to introduce mutation T256A in a similar way as described above , and then , plasmid pMT28 ( SN , TA ) was subjected to site-directed mutagenesis to introduce mutation Q263H . All chimeric viruses and mutants were analyzed by nucleotide sequencing using Big Dye Terminator Cycle Sequencing kit ( Abi Prism; PerkinElmer ) and sequencer ABI373 as previously described [58] . Sequences were analyzed using DNASTAR 4 . 0 ( http://www . dnastar . com ) , GeneDoc , and GCC ( University of Wisconsin ) . Each sequence was determined at least twice , with products obtained using different oligonucleotide primers . DNA from pMT28 or its recombinant and mutant derivatives was linearized with Nde I and transcribed with SP6 RNA polymerase as previously described [22 , 27] . Transcript RNA integrity and concentration were estimated by agarose gel electrophoresis , in parallel runs with known amounts of standard C-S8c1 RNA . BHK-21 cell monolayers ( 70% confluent , about 1 × 106 cells ) were transfected with RNA transcripts ( 1 μg RNA ) using lipofectin as previously described [59] . Virus was collected from the culture supernatant at 72 h post-transfection . The virus obtained by transfection was passaged twice before using it in biological studies . RNA was extracted and sequenced to ascertain that the virus maintained the genomic structure and mutations of the initial transcript . Consensus genomic nucleotide sequences of FMDV clones were obtained by RT-PCR amplification of virion RNA using specific primers [7 , 22 , 28] .
The GenBank accession numbers for the C-S8c1 , , , CS8p260p3d , and MARLS genomic sequences are AJ133357 , AM409190 , AM409325 , DQ409185 , and AF274010 , respectively . Nucleotide and amino acid sequences for picornaviruses can be found at http://www . iah . bbsrc . ac . uk/virus/picornaviridae/SequenceDatabase/3Ddatabase/3D . HTM . | Virulence expresses the harm that parasites inflict upon their hosts . Many studies have addressed the basis of virulence and its effect on host and parasite survival . It has generally been accepted that one of the components of parasite virulence is fitness , or the capacity of the parasite to multiply in its host . Some models have equated virulence with fitness . In the present study , we use foot-and-mouth disease virus ( FMDV ) to document that virulence and fitness—measured in the same biological environment provided by cells in culture—can be unrelated traits . This has been achieved by multiplying the virus in a manner that mutations accumulated in its genome . Mutations decreased fitness dramatically , but not virulence . Chimeric and mutant viruses were constructed to show that virulence is influenced by only some of the FMDV genes , while fitness is influenced by the entire genome . For this reason , virulence is more robust ( “resistant” ) than fitness to the effects of deleterious mutations . The fact that virulence can be unrelated to fitness has implications for the design of anti-viral vaccines because it suggests that it may be possible to design high fitness , low virulence strains to stimulate the host immune response . Furthermore , in modelling studies it cannot be assumed that virulence is equal to fitness . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"virulence",
"passage",
"regime",
"recombinant",
"virus",
"foot-and-mouth",
"disease",
"virus",
"virology",
"viral",
"quasispecies",
"molecular",
"biology"
] | 2007 | Molecular Basis for a Lack of Correlation between Viral Fitness and Cell Killing Capacity |
Assessment of attitudes of health care professionals is important as negative attitude could constitute a major deterrent to care-seeking by persons affected by neglected tropical diseases ( NTDs ) such as leprosy . Leprosy continues to pose a major disease burden in India with an annual new case detection rate of 10 . 17 per 100 , 000 population . This paper reports on the development and validation of a culturally appropriate scale to measure attitude of health care providers ( HCPs ) towards persons affected by leprosy in Tamil Nadu , India . The Affective , Behavioural and Cognitive ( ABC ) model of attitudes guided the development of the scale . Steps in scale development included qualitative interviews and focus group discussions with medical officers and paramedical staff selected from high prevalence districts in Tamil Nadu , India which informed the development of the draft scale . Reviews of existing attitude questionnaires in related areas further contributed to scale development and together helped to generate a large pool of items which was then subjected to Thurston’s scaling method for selection of items from this pool . Face and content validity were obtained , following which internal consistency and test , re-test reliability were assessed . Scaling exercise resulted in 11 items being discarded from an initial pool of 38 , owing to the poor agreement among experts regarding relevance . Face and content validity were good with experts endorsing relevance and applicability of items . The intra-class correlation coefficient ( ICC ) for test re-test reliability of the 27 item scale was 0 . 6 ( 95% CI: 0 . 20–0 . 78 ) indicating marginal intra-class correlation . The overall Cronbach’s alpha was 0 . 85 while the alphas for each of the affective and behavioural components was good at 0 . 78 and 0 . 69 respectively indicating a good degree of consistency and homogeneity between items but the alpha for the cognitive component was low at 0 . 53 . The ABC model of attitudes guided the development of the scale , ensured a mix of 27 items tapping into the three domains of Affect , Behaviour and Cognition which best explained the attitude construct . With good validity and alphas for each of the affective , behavioural components and overall alpha estimates , this scale can be a valuable tool to provide accurate estimates of the true attitudes held by HCPs . This , in turn , would be useful to obtain insights for appropriate intervention programmes that would help change negative attitudes of HCPs towards persons affected by leprosy . With some adaptations , the scales can be validated for other NTDs as well .
The Affective , Behavioural and Cognitive ( ABC ) model of attitudes guided the development and selection of items for this scale [15] . According to this model , attitude is comprised of 3 components- affective , behavioural and cognition . The affective component refers to emotional reactions individuals have towards the attitude object; the behavioural component refers to the way individuals behave when exposed to the attitude object and the cognitive component refers to an individual’s beliefs and knowledge about the attitude object . A web search revealed 11 studies that assessed the knowledge , attitudes and practices of health care providers towards persons affected by leprosy . Most of these studies had been carried out in African countries [8 , 9 , 12 , 16–20] with three that had been undertaken in India [21–23] . Based on our review , we discovered that barring a few items , the majority of the items in the different scales were more representative of knowledge and practice rather than attitude thereby justifying the development of such a scale . The attitude items from these different scales used in the 11 studies were listed and served as the item pool from which- following information generated from the qualitative work- we made decisions on the ones to retain . The initial pool of items from these studies were examined for equivalence . Beaton et al . ( 2000 ) guidelines on cross cultural adaptation ( to look for semantic , conceptual , experiential and idiomatic equivalence ) of self-report measure guided us in this process . For example , one of the items that we did not include was “cause of leprosy is because of witches” . This was because the concept of ‘witches’ is not integral to the Indian culture [24] .
Permission to conduct the study was obtained from the Directorate of Public Health & Preventive Medicine ( DPH&PM ) and the research protocol was approved by the Institutional Review Board of GLRA , India . Willing participants were required to sign an informed consent form . The study was carried out during the period April 2015 –March 2016 . A list of districts in the state of Tamil Nadu , which showed a high number of cases of leprosy were obtained from the DPH&PM , Chennai . The districts of Villupuram , Kancheepuram , Tiruvannamalai and Tiruvallur were some of the districts located relatively close to Chennai that was selected from this list . For the year 2013–2014 , a total of 353 new cases of leprosy were reported in the district of Villupuram . Similarly , for the districts of Kancheepuram , Tiruvannamalai and Tiruvallur it was 160 , 176 and 140 new cases respectively for the same time period . A list of primary health centers ( PHCs ) in each of these selected districts was also obtained from the DPH&PM and three PHCs were purposively chosen for the study . The Medical Officers ( MOs ) , Health Inspectors ( HIs ) and Village Health Nurses ( VHNs ) in each of these PHCs were then contacted and their permission sought to participate in the study . Each of these cadres of HCPs , play important roles in the management of persons affected by leprosy . Thus , while both the HIs and VHNs assist in case detection in the community and in educating people about leprosy , etc . , the MOs are involved in providing direct clinical services to the persons affected by leprosy . We also carried out interviews with persons affected by leprosy in order to understand their experiences about accessing care and any discriminatory behaviors they may have faced from HCPs . For the qualitative component we included three PHCs from two districts namely Villupuram and Kancheepuram while for questionnaire validation we contacted as many HCPs from the above list who were willing to participate in the study . The scale development process was divided into two parts . In part 1 , we carried out the qualitative Semi Structured Interviews ( SSIs ) and Focus Group Discussions ( FGDs ) for the purpose of item generation , followed by the scaling exercise and in part 2 we undertook the validation of the questionnaire .
Five MOs and five HIs participated in the qualitative interviews . While three MOs and three HIs were from Villupuram , two MOs and two HIs were from Kancheepuram . One MO and three HIs were women; all the rest were men . The MOs had all completed their MBBS degree . While four of the HIs held post-graduate degrees , one had only completed schooling ( 12 years ) . Two FGDs were carried out with the VHNs , one in Kancheepuram and one in Villupuram . The VHNs were all women who had completed their schooling and were trained in community health services . As far as the MOs were concerned , they regarded leprosy like any other disease and were knowledgeable about its modes of transmission and management . They spoke of setting an example to others by being good role models so that other cadres of HCPs would learn from them and behave accordingly . The VHNs and HIs also had good knowledge about the disease but expressed the need for better understanding of its modes of spread and the side effects of treatment . They mostly spoke of the need to care for these patients and categorically denied any discriminatory behavior by any category of HCPs . Some said that such behavior had been present earlier but was no longer evident while others spoke of harboring some fears in the beginning which they got over once they started caring for patients . All the HCPs however , believed that stigma towards leprosy patients was very much present in the community . This was evident by the fact that patients did not want to be seen taking treatment and tried to hide their condition . They would specifically request the VHN and HIs to keep their disease a secret as they feared being isolated by the community . The general opinion was that as the disease was identified early both because of better awareness of its symptoms by the community and because of proactive efforts made by VHNs and HIs , the disease rarely progressed to the stage of severe ulceration . Consequently , patients were diagnosed and treated early . Very few patients with severe ulceration and disabilities—which could give rise to both revulsion and fear—were seen by HCPs . A few VHNs opined that sometimes steps taken by them towards protecting others were misinterpreted and led to people affected by leprosy feeling offended and believing that they were being discriminated against . Several issues emerging from the qualitative interviews were reflected in the items drawn from the scales used in the 11 studies . For example , one MO said that when he initially started working with leprosy patients he feared contracting the disease but over time he overcame the fear , “Earlier I used to fear whether I too would develop leprosy . Now one year , I have been working in leprosy . Now I feel confident . In the beginning , I had this fear but later it was not there” ( PHC MO- Kancheepuram ) . The corresponding attitude item we selected was , ‘I am concerned with getting infection from patients with leprosy when I treat them’ . Similarly , an HI indicated that seeing patients present with pus and ulcerative wounds was distressing , “if they are having pus then we would feel like a kind of something . But what I can do about that ? I would just tell them how to prevent from getting pus” ( PHC HI- Kancheepuram ) . The corresponding item we selected was , ‘There is a sense of revulsion when seeing a leprosy patient with ulcers’ . The VHNs during the FGD said , “As we treat any other patient , we also treat the leprosy patients in the same way . We don’t isolate leprosy patients and treat them separately” ( FGD VHN-Kancheepuram ) . One MO echoed the same issue when he said , “I am seeing the leprosy patient as normal as how I see other patients . We practice precautionary measures , like wearing gloves ( PHC MO 3 , Villupuram ) . The corresponding item we selected was . ‘It is possible to manage leprosy like any other disease in the general health service’ . An HI described feeling a sense of satisfaction working and helping persons affected by leprosy , “I feel good because when we identify cases and provide them treatment , it gives a good feeling until now , there were many people , who had no knowledge about this disease at all , and they had remained without taking any kind of treatment . So , when we identify cases and put them on treatment , it gives a satisfactory sort of feeling” ( HI Kancheepuram ) . The corresponding item we selected was , ‘I get a sense of satisfaction when I treat patients with leprosy’ . Thus the information derived from these qualitative interviews and FGDs served to corroborate the items we selected to constitute this attitude scale . We also took care to bring in both positive and negatively worded statements . All three persons affected by leprosy said that hospital staff treated them well , were supportive and had guided them to seek care early . They all said that the stigma they faced was mostly from the community and from family members as people feared contracting the disease . They therefore tended to isolate themselves to avoid being hurt and humiliated . The experts’ group judged the 27 items to measure attitudes towards leprosy at face value thereby establishing face validity . In terms of the content validity , the researchers found that the items in each of the affective , behavioural and cognitive domains were an adequate representation of the construct of attitude . As regards reliability a total of 54 HCPs participated in the reliability exercise ( Table 4 ) . While 54 persons participated in the test , thirty-eight persons ( 70% ) completed the re-test . Unfortunately , despite scheduling a date and time for the re-test in consultation with the HCPs , many of them either did not come or else were pre-occupied and could not fill in the questionnaire . The ICC for test re-test reliability of the 27 items scale was 0 . 6 ( 95% CI 0 . 20–0 . 78 ) indicating marginal intra-class correlation while the overall Cronbach’s alpha was 0 . 85 The alphas for each of the affective and behavioural components was good at 0 . 78 and 0 . 69 respectively , but the alpha for the cognitive component was low at 0 . 53 .
Despite some limitations , this scale developed to measure the attitudes of HCPs towards people affected by leprosy presents a first step in this direction . It could be used to evaluate provider attitudes and could aid in identifying positive or negative attitudes held by providers towards persons affected by leprosy because negative attitudes may impede leprosy control activities . It may also serve as a viable tool to assess changes in the attitudes of HCPs following an intervention . However , further research is required , in terms of using the scale on a much larger sample of HCPs across different parts of India so as to fully substantiate its relevance and cultural appropriateness . With some adaptations , the scales can also be validated for other NTDs which are endemic in India . | Leprosy is an infectious disease caused by the Mycobacterium leprae and is one of the major causes of preventable disability . Early diagnosis and prompt treatment of all new cases of leprosy remain the key strategies for leprosy control as it would prevent nerve damage , disability and reduce the transmission of the disease . People affected by leprosy often experience severe stigmatization because of an adverse social judgment about the disease or its disabling consequences . This neglected tropical disease continues to pose a major disease burden in India . Despite the availability of health facilities there continue to be barriers towards leprosy diagnosis and early treatment . Assessment of attitudes of health care professionals is important as negative attitude could constitute a major deterrent to care-seeking by persons affected by leprosy . Researchers developed and validated a culturally appropriate scale to measure attitudes of health care providers towards persons affected by leprosy in Tamil Nadu , India . The scale would be useful to obtain insights of attitudes of health care professionals to plan appropriate programmes that would help to promote positive attitudes of healthcare providers towards persons affected by leprosy . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"behavioral",
"and",
"social",
"aspects",
"of",
"health",
"sociology",
"tropical",
"diseases",
"geographical",
"locations",
"india",
"social",
"sciences",
"health",
"care",
"bacteri... | 2018 | Development and validation of a scale to assess attitudes of health care providers towards persons affected by leprosy in southern India |
Borrelia recurrentis is the causative agent of louse-borne relapsing fever , endemic to the Horn of Africa . New attention was raised in Europe , with the highest number of cases ( n = 45 ) reported among migrants in 2015 in Germany and sporadically from other European countries . So far only one genome was sequenced , hindering the development of specific molecular diagnostic and typing tools . Here we report on modified culture conditions for B . recurrentis and the intraspecies genome variability of six isolates isolated and cultured in different years in order to explore the possibility to identify new targets for typing and examine the molecular epidemiology of the pathogen . Two historical isolates from Ethiopia and four isolates from migrants from Somalia ( n = 3 ) and Ethiopia ( n = 1 ) obtained in 2015 were cultured in MPK-medium supplemented with 50% foetal calf serum . Whole DNA was sequenced using Illumina MiSeq technology and analysed using the CLC Genomics Workbench and SPAdes de novo assembler . Compared to the reference B . recurrentis A1 29–38 SNPs were identified in the genome distributed on the chromosome and plasmids . In addition to that , plasmids of differing length , compared to the available reference genome were identified . The observed low genetic variability of B . recurrentis isolates is possibly due to the adaptation to a very conserved vector-host ( louse-human ) cycle , or influenced by the fastidious nature of the pathogen and their resistance to in vitro growth . Nevertheless , isolates obtained in 2015 were bearing the same chromosomal SNPs and could be distinguished from the historical isolates by means of whole genome sequencing , but not hitherto used typing methods . This is the first study examining the molecular epidemiology of B . recurrentis and provides the necessary background for the development of better diagnostic tools .
Borrelia recurrentis is the causative agent of human louse-borne relapsing fever ( LBRF ) . It is the only relapsing fever Borrelia species transmitted by the body louse Pediculus humanus [1] , and–apart from humans—has no known animal reservoir hosts , nor other vectors [2] . There are 22 different currently known tick-borne relapsing fever ( TBRF ) Borrelia species that are transmitted by different soft and hard ticks , and that have different natural animal reservoir hosts [3 , 4] . The classification and diagnosis of TBRF historically relied often on the vector identification and geographical region [5] . These host and vector specificities of RF borreliae have an impact on human infections , their public health importance and their geographic distribution [3 , 5 , 6] . Due to the cosmopolitan nature of the human body louse , B . recurrentis was once present worldwide and affected millions of people , causing major outbreaks during times of war and crisis [2 , 6] . Improvement of hygiene and living conditions has led to a decrease of body lice infestations in the industrialized world , and the pathogen is nowadays endemic to the Horn of Africa [7 , 8] , where it is still a major public health concern . It is the seventh most common cause ( up to 27% ) of hospital admission and the fifth most frequent cause of death in the highlands of Ethiopia [9–11] . Limited data are available on the prevalence of the disease in neighboring countries such as Sudan , the Republic of South Sudan or Somalia [7 , 12 , 13] . LBRF and TBRF differ not only in the mode of transmission , but also in the severity and the outcome of the disease . LBRF is characterized by a three to ten day incubation period , one to five relapses of fever that typically last three to five days , with afebrile intervals that lengthen as the disease progresses . It has a poorer outcome compared to TBRF [3] . If left untreated mortality of LBRF can reach up to 40% , while for TBRF it is below 5% . Even if treated , LBRF mortality can be as high as 10% [3] , mostly due to a severe reaction caused by pro-inflammatory cytokines called Jarisch-Herxheimer ( JH ) reaction [14] . Whether the severity of the relapsing fevers and development of JH reaction is somehow influenced by the genetic and antigenic make-up of B . recurrentis is not clear and to the best of our knowledge was not studied so far . Molecular typing tools such as PCR of protein coding genes , intergenic spacer typing , or the sequencing of housekeeping genes developed for TBRF Borrelia species [15] showed limited success when applied to B . recurrentis . It proved especially difficult to differentiate the two closely related spirochaetes B . recurrentis and B . duttonii [16] . This high similarity casted doubt on the species separation , and it was suggested that they are ecotypes of the same species , recently adapted to , and evolving in , different vectors [3 , 15 , 17] . The distinction of these closely related relapsing fever spirochaetes is important due to the fact that antibiotic treatment of LBRF leads more often to JH reactions and requires intensive care [14] . Up to now , only one whole genome sequence is available for B . recurrentis [17] . The B . recurrentis strain A1 contains eight linear genome fragments , the longest with 930 kb was designated as the chromosome and the remaining seven as plasmids ( pl124 , pl53 , pl37 , pl35 , pl33 , pl23 , pl6 ) named according to the identified length in kb . Although this provided first insights into the genetic composition , the availability of only one genome does not provide insights into the population variation and , thus , hampered the development of more specific typing tools . It was , for example , not possible to address questions such as genome variability of the species and the potentially differing virulence or the molecular epidemiology of different isolates . Further complications for genome comparison and development of typing and/or diagnostic tools arise from the fact that B . recurrentis are relatively rare and difficult to culture in axenic medium or animal models [18] . Until the 1990s the species was deemed non-cultivable due to their fastidious and slow growing nature [19 , 20] , and even nowadays success in cultivation is limited [3] . B . recurrentis today still bears a considerable potential to occur and spread among vulnerable populations that are exposed to unfavorable conditions , with limited access to sanitation and personal hygiene facilities and in overcrowded circumstances where louse infestations are common [21–23] . The most recent example of B . recurrentis infections have been observed in Europe during 2015 [8 , 24–29] . All cases were imported , and associated with newly arrived migrants , with a possible autochthonous transmission in Italy [24] . Interestingly , the majority of cases reported in Bavaria , Germany ( 92% ) were among people from Somalia , while 3% were from Ethiopia and 5% from Eritrea [30] . Similar observations were made in other European countries , the majority of reported cases were among Somalian refugees , while in the Netherlands and Switzerland reported cases were only from Eritrea . Here , we report on conditions that enabled us to culture a number of isolates from refugees from Somalia , Eritrea and Ethiopia . This provided the opportunity to explore the intraspecies genome variability of six B . recurrentis isolates . To the best of our knowledge , this is the first study examining the genomic variability of B . recurrentis with state-of-the-art next generation sequencing techniques and it provides valuable information necessary for the development of new diagnostic and typing tools and a better understanding of the molecular epidemiology of this pathogen .
Two isolates were historical , isolated from subjects infected in Ethiopia in 1985 [31] and 2004 , and four isolates were isolated in 2015 from subjects arriving from Somalia ( n = 3 ) and Ethiopia ( n = 1 ) . Due to long migration routes that were only vaguely described the exact place of infection is unclear and can only be speculated for the four isolates from 2015 ( Table 1 ) . Blood samples were sent to the Bavarian Health and Food Safety Authority for diagnostic purposes . 20 μl of the samples were placed on a microscope slide , covered with a cover slip and viewed using dark field microscopy Zeiss Axioscope Microscope , objective 40X , ocular 10X . To aid diagnostics , 500 μl of blood sample were placed in to Modified-Kelly-Pettenkofer ( MKP ) culture medium [32] that was additionally supplemented with 50% foetal calf serum ( FCS ) . Cultures were kept in screw cap glass tubes ( 7 ml volume ) at 33°C . Cultures were examined on a regular basis ( 2 x per week ) to check for growth of bacteria . Actively growing bacterial cultures were subcultured or diluted using MKP-50% FCS medium . Aliquots of well growing cultures were frozen according to standard methods with 15% Glycerin [32] . Total DNA was isolated from 35 ml of culture with a cell density of appr . 5X106 - 1X107 using the Maxwell 16 LEV Blood DNA Kit according to the manufacturer's instruction on the Maxwell Instrument ( Promega , Germany ) . DNA libraries for whole genome sequencing were constructed using the Nextera XT DNA Library Preparation Kit ( Illumina , San Diego , USA ) . Samples were sequenced using Illumina MiSeq technology and a V2 Reagent Kit ( Illumina ) to produce 2x250 bp paired-end reads according to the manufacturer's instructions . Analysis was performed using the CLC Genomics Workbench 9 . 0 . 1 software . Briefly , reads were trimmed using quality scores: the limit was set to 0 . 05 , and allowing for 2 ambiguous nucleotides , with a minimum length of 50 . Contaminations with human DNA were removed by mapping all reads to a human reference genome ( GCA_000001405 . 23 ) , with the following mapping options: match score 1 , mismatch cost 2 , insertion and deletion cost 3 , similarity fraction 0 . 8 and length fraction 0 . 5 . The remaining reads were both , de novo assembled and mapped to the reference genome B . recurrentis A1 ( GCA_000019705 . 1 ) . Read mapping was performed using the integrated CLC mapper , with the same conditions as mentioned above for mapping to the human genome . Variant calling was performed with the CLC tool “Basic Variant Detection” , with ploidy set to 1 , ignoring positions with coverage above 100 . 000 , broken pairs and non-specific matches . The minimum read length was set to 20 , minimum coverage 10 and minimum frequency of 90% . Quality filters to the neighboring 5 nucleotides included a minimum central quality of 20 , and neighborhood quality 15 . For de novo assembly two different software were used: the de novo assembler provided in CLC and the SPAdes assembler version 3 . 9 . 0 [33] . In CLC , the same mismatch , insertion and deletion costs as for mapping were applied , with an automatic bubble ( 50 ) and k-mer ( 20 ) size . Alignment mode was set to local and match mode to random , minimum contig length was 500 bp . De novo assembly in SPAdes was performed using the following settings: the “careful mode” was chosen in order to reduce mismatches and short indels , k-mer sizes included 21 , 33 , 55 , 77 , 99 , 127 . De novo assembled contigs were aligned to the reference genome A1 using the CLC Microbial finishing module , MAUVE [34]; and QUAST [35] , and visualised using BRIG [36] . MEGA software version 7 . 02 [37] was used for phylogenetic tree construction using the identified SNP positions with the maximum likelihood method , and genetic distance matrixes corrected using the Tamura-Nei substitution model [38] . The topology of the trees obtained was assessed by bootstrapping , with 1000 replications . A large scale synteny analysis of de novo assembled contigs and the reference genome B . recurrentis A1 was performed using MUMmer 3 . 0 [39] . Primers specific for the 5’ end of the plasmid pl6 and 3’ end of the pl165 were designed ( Table 2 ) . PCR was performed with HotStarTaq PCR from QIAGEN with the following conditions: 15 min at 95°C followed by 30 cycles 1 min at 94°C , 1 min at 50°C , 1 min extension at 72°C with a final extension step at 72°C for 10 min .
During 2015 , blood samples from 38 patients diagnosed with LBRF were available at the Bavarian Health and Food Safety Authority/National Reference Centre for Borrelia , thereof 21 could be adapted to the modified culture conditions described in this study . Initially , culture was attempted in BSK , MPK and MPK supplemented with 50% FCS . This was the medium of choice in our laboratory for tick-borne relapsing Borrelia , and proved to be the most successful for B . recurrentis . No growth was observed in BSK , nor MPK medium . B . recurrentis strains from 9 samples had grown to acceptable densities ( 106−107/ml ) and subjected to DNA extraction . Four isolates from 2015 yielded DNA of sufficient quantity and quality for sequencing . In addition , the two historical isolates that were well adapted and brought to culture in 1985 ( A17 ) and 2004 ( PBek ) were included . The isolate A17 had been recovered from a 15 year old male in Ethiopia and the isolate PBek had been isolated at the German National Reference Centre for Borrelia from a traveler returning from Ethiopia , and was suspected to belong to the B . recurrentis species . We were able to confirm this species designation in this study . Total reads per sample ranged from 736 , 000–2 , 203 , 405 . Read mapping to the A1 reference genome available in GenBank resulted in an average coverage between 52 and 1094 for all plasmids and the main chromosome , and >99% of the genome had a coverage of ≥10x ( Table 3 ) . In isolates obtained from migrants , a total of 17 single nucleotide polymorphisms ( SNPs ) , identical in all four isolates , compared to the main chromosome of A1 were identified . The two historical isolates A17 and PBek had 12 and 16 SNPs , respectively ( with a min . coverage of 10 and a min . frequency 90% ) ( see S1 Table and Fig 1A ) . Six SNPs were common to all isolates , thereof two in non-coding regions , one was a silent mutation , and three nonsynonymous mutations in a serine protease , a permease and in the apparent leading methionine in an UDP-N-acetylmuramoyl-tripeptide—D-alanyl-D-alanine ligase . In addition to the common 6 SNPs , the isolates recovered from migrants in 2015 and the historical isolate PBek shared 10 more SNPs compared to the chromosome of the reference strain A1 . All were located in coding regions , three synonymous changes in UvrC , chemotaxis protein CheW and a trigger factor , while seven were nonsynonymous and would lead to amino acid changes in a transcript cleavage factor , dipeptide/oligopeptide/nickel ABC transporter ATP-binding protein , GyrB , 30S ribosomal protein S5 , chemotaxis protein , tRNA methyltransferase TrmD , transcription termination/antitermination protein NusA . The isolates from 2015 had one more SNP causing an amino acid change in the acriflavin resistance protein . The isolate A17 differed the most compared to the other six sequenced genomes . It contained one SNP in a noncoding region , one change in the 23S rRNA , an additional SNP in the aforementioned serine protease and 3 SNPs causing amino acid changes in the genes coding for the flagellar motor switch protein FliG , peptidase M16 , peptide ABC transporter permease . Performing SNP calling with less stringency ( min . coverage 10 and min frequency 75% or 50% ) , identified two additional SNPs . At 75% a deletion in a poly-A region was detected in all refugee isolates and in the PBek historical isolate and at 50% in the PBek isolate one more SNP was called in a coding region . Both SNPs would lead to a frameshift mutation . Notably , all identified SNPs were outside of the 14 loci developed and used for Borrelia typing [16 , 40 , 41] . This includes the 8 housekeeping genes used for the MLST typing of other Borrelia species [41] , the 5 loci used in MST [16] and the intergenic spacer typing ( IGS ) [40] method developed for typing of B . recurrentis and B . duttonii ( Fig 2 ) . Mapping the reads to B . duttonii ( GCF_000019685 . 1 ) , identified 185–260 single nucleotide differences over the main chromosome , highlighting potential loci for the development of specific molecular diagnostic tools . All seven plasmids present in the A1 reference genome were identified via read mapping in the newly sequenced isolates . Between 17 and 21 SNPs on five of the seven plasmids ( pl124 , pl23 , pl33 , pl35 and pl53 ) were identified in all samples compared to the reference B . recurrentis A1 genome ( Table 1 in supplementary data ) . Majority of SNPs ( n = 14 ) were present at the furthest ends ( 250 bp ) of the linear plasmids . In two of the seven plasmids no SNPs were identified ( pl37 and pl6 ) . The plasmid SNPs from isolates recovered in 2015 were not identical as was the case for the chromosomal SNPs , nevertheless , those isolates were more related than the historical isolates as illustrated by the SNP phylogenetic tree ( Fig 1B ) . Employing the de novo assembly approach , it was possible to assemble draft genomes of the six B . recurrentis isolates in 25–54 contigs with SPAdes and 41–70 contigs with the de novo assembler of CLC ( Tables 4 and 5 ) . At least 98 . 96% of the contigs assembled had a coverage ≥ 10x , and the draft genome of the isolate PBek had 99% of the contigs covered ≥ 20-fold ( Tables 4 and 5 ) . The main chromosome could be reconstructed in 1–2 contigs , and no large gene rearrangements , deletions , insertions , translocations , could be observed ( see supplementary data S1 Data ) . Plasmid pl124 of B . recurrentis strain A1 was described as being co-linear to plasmid pl165 of B . duttonii strain Ly , but was 40 kb shorter [17] . A highly similar plasmid was detected also in a recently sequenced B . crocidurae [42] ( Accession number: NC_017778 . 1 ) . Evidence for a larger version of pl124 was present in all isolates examined in this study . Contigs assembled de novo from reads that did not map to the reference sequence ( approx . 3% , Table 3 ) as well as de novo assembled contigs from raw reads matched the 40 kb on the left 5’ end of the pl165 from B . duttonii and B . crocidurae that was lacking in B . recurrentis strain A1 ( Fig 3 ) . The opposite was true for plasmid pl6 . All sequenced isolates lacked approximately 1 kb at the 3´ end of the plasmid sequence ( Fig 4 . ) . The lacking sequence region contains two hypothetical proteins in the reference genome [17] . In order to rule out assembly errors , PCR primers were designed for the aforementioned regions in both plasmids . For pl6 no PCR products were obtained while for pl124/165 , PCR products of the expected size were obtained in all samples ( S1 Fig ) . These results confirmed the presence/absence of these regions in the plasmids of the B . recurrentis isolates investigated here . Furthermore , it was noticed that the reference genome of B . duttonii available in GenBank does not contain a counterpart of the smallest B . recurrentis plasmid , pl6 , but that similar plasmids are present in the genomes of B . crocidurae [42] ( NC_017775 . 1 ) , B . miyamotoi [43] ( KT355574 . 1 ) and B . hermsii ( CP005739 . 1 ) ( Fig 4 ) .
The variability at the SNP level found in our samples compared to the B . recurrentis A1 reference genome was low , and mostly located on the chromosome . High conservation was already observed among RF borreliae based on sequencing short segments of different genetic loci: 16S , flaB , glpQ , IGS [15] , MST [44] and whole DNA-DNA hybridization studies [31] . The species B . duttonii and B . recurrentis are often indistinguishable by these methods with > 99% sequence identity , that was exemplified again by the low number of SNPs when mapping B . recurrentis reads to B . duttonii reference ( n = 185–260 ) . It has been hypothesized that they are in fact the same species with the latter having a shorter , decaying genome due to an inactive recA [17] involved in DNA doublestrand break repair as a result of the adaptation to new hosts ( louse/human ) . Indeed , similar gene loss was also observed in other louse-borne bacterial species compared to their tick-borne counterpart , such as Bartonella quintana and Bartonella henselae [45] , Rickettsia prowazekii and Rickettsia conorii [46] . Nevertheless , the identified SNPs could be further explored as possible targets for specific diagnostic and/or typing tools . The identified 12–17 SNPs on the chromosome of the six sequenced isolates investigated here were not present in the 8 housekeeping genes used for the MLST typing of Borrelia species . Therefore , by using MLST typing all newly identified isolates would be indistinguishable from the B . recurrentis A1 reference strain , and would belong to the same MLST sequence type ST-669 . This method still has the discriminatory power to distinguish B . recurrentis from B . duttonii and B . crocidurae . Employing other methods used to differentiate closely related B . duttonii and B . recurrentis would not provide more discriminatory power in the investigation of intraspecies variability in this study , as all isolates examined contained same sequences as the reference genome on the loci used for the MST and IGS , and therefore facing the same limitations as already noted for these methods [16 , 40] . Interestingly , applying the IGS and MST typing methods to a higher number of samples from different sources did in fact find variability among B . recurrentis , separating them into two clades albeit by a single SNP difference [16 , 40] . The isolates came from a limited number of cultured patient samples and DNA isolates obtained directly from lice [40] or DNA isolates obtained directly from patient blood samples [16] . Similar observations were made with the IGS method and B . duttonii isolates . In total , by means of IGS typing , four different clades could be identified among B . duttonii isolates from cultured isolates , DNA isolated directly from patient blood , or from ticks [40] . However , all cultured B . duttonii isolates belonged to the same clade implying that there might be a bias introduced with adaptation to culture , masking the real variability present in host and vector . A similar effect could have occurred for the six B . recurrentis isolates examined here . Although cultivable B . recurrentis isolates from two IGS subtypes were described in literature [40] , all isolates sequenced in this study belong to only one IGS subtype and thus may not reflect the potential full extent of diversity . Furthermore , one might , speculate that the variability is higher amongst the bacterial population that is not cultivable from patient blood . The observed difficulties to adapt the pathogen to in vitro culture conditions would support this hypothesis: Out of 38 blood samples available in 2015 at the Bavarian Health and Food Safety Authority , 21 adapted to culture conditions . One needs to consider , that some patients were already treated with antibiotics prior to blood sampling and that this might have hampered recovery of B . recurrentis from blood . Only nine had grown to meaningful densities and four strains yielded sufficient amounts of DNA for sequencing . It is possible that culture adaptation acts as selective pressure and it may be that only specific variants/clones readily adapt to culture as observed for B . duttonii [40] . Therefore , the observed low variability in the currently investigated strains would be the variability of the variant/clone that grows readily under in vitro culture conditions . An alternative hypothesis to explain the low variability is that due to its adaptation to a new host and vector B . recurrentis underwent a strong population bottleneck that is still noticeable in the population today . There is no natural animal reservoir available to study the infection and virulence of this pathogen . In order to address these issues , it would be very interesting to collect lice from infected people and examine the variability of B . recurrentis in lice . Additionally , culture independent sequencing methods directly from patient blood and from vectors could be employed , to circumvent possible bias introduced by infection and/or culture [47] . The stringency of SNP calling in our study was high , in order to exclude SNPs resulting from sequencing error . Performing SNP calling with less stringency ( minimum frequency 75% and 50% ) identified two additional SNPs . At 75% a deletion in a poly-A region was detected in all isolates from 2015 and in the PBek historical isolate and at 50% in the PBek isolate one more SNP was called in a coding region . Both SNPs led to a frameshift and would hamper the expression of the given gene , therefore , we consider it highly likely that these SNPs were artefactual as they would render the resulting protein non-functional . Out of the six SNPs that were common to all newly sequenced isolates compared to the reference genome , one was in the leading methionine AUG start codon of an UDP-N-acetylmuramoyl-tripeptide-D-alanyl-D-alanine ligase , changing it to a valine ( GUG ) . However , this leading methionine codon was identified via automated gene prediction method , while the homologous genes in B . duttonii and B . crocidurae started with an alternative start codon UUG that is present three base pairs upstream of this hypothetical start codon . The sequence coding for the same alternative start codon was present also three base pairs upstream in the isolates examined here , therefore active transcription of this gene can be assumed in spite of the change in the alleged start codon identified via automated gene prediction method . Even though the genomes of the different B . recurrentis isolates were highly conserved , the phylogenetic trees based on the SNPs ( both only chromosomal , as well as total genomic SNPs ) showed a clear distinction between the historical isolates and the isolates recovered in 2015 ( Fig 1 ) . The latter were all identical , bearing the same chromosomal SNPs in comparison to the reference strain and differing also from the two historical isolates A17 and PBek in 18 and 1 SNPs , respectively . The epidemiological investigations pointed to a common infection source along the Mediterranean migration route for refugees from East Africa [8 , 28] . The identical SNP pattern in refugee isolates supports this hypothesis , even in the light of the low variability of the B . recurrentis genome . Using the WGS approach and SNP analysis we were able to find a limited number of differences between the two isolates obtained at the same time and kept in culture ( A1 and A17 ) . These isolates were otherwise indistinguishable with other methods [15] . This underlines once more the value of WGS as a tool to examine phylogenetic relationships , especially when investigating pathogens with low genome variability . Compared to the seven linear plasmids of the A1 reference genome , the majority of SNPs identified through mapping were located at the ends of the plasmids and might have been introduced through sequencing errors either in the reference genome or the newly sequenced isolates . Interestingly , the highest number of SNPs ( n = 10–14 ) was at the left 5’ end of plasmid pl53 in the coding sequence for a hypothetical protein ( locus tag: BRE_RS05500 ) . This hypothetical CDS showed high similarity to other genes coding for surface proteins ( vlp , vsp , vmp ) from relapsing fever Borrelia . The high SNP density at this particular site would support the already noted antigenic variation of these surface proteins , that occurs through gene conversion in the genomes of relapsing fever Borrelia in order to escape the host immune system[48 , 49] . The proportion of reads that remained unmapped in our sequence analysis suggested that the reference genome deposited at GenBank , might not have been complete . De novo assemblies of both , unmapped reads and performed with all reads , revealed contigs similar to the longest plasmid , pl165 , of TBRF Borrelia ( B . duttonii and B . crocidurae ) . The assembled contigs matched to the left end of the plasmid , that was initially not identified in the B . recurrentis reference genome of strain A1 . There are several possible explanations to this: ( i ) it may be the result of long adaptation to culture conditions , as already noted for B . hermsii and B . turicatae that spontaneously lost part of the largest plasmid after 100 serial passages in vitro without impeding their infectivity [50] , ( ii ) it could also be attributed to sequencing and/or assembly error . We favour the last hypothesis for the following reasons: Initial studies performed on B . recurrentis cultivable isolates ( A1-A18 ) and plasmid analysis with PFGE did suggest the presence of a large , approximately 180–190 kb long plasmid [20 , 31] . In addition , the presence and functionality of a gene designated cihC , coding for an outer membrane lipoprotein that binds major inhibitors of the human complement activation system ( C4b-binding protein and C1 esterase inhibitor ) was shown to be present in both isolates ( A1 and A17 ) and located on a ~200 kb long linear plasmid [51] . Orthologous genes are present on the left termini of the longest plasmids of B . duttonii , B . hermsii and B . turicatae genomes [51] . PCR with primers designed to overlap the border between pl124 and pl165 of B . duttonii resulted in PCR products of the expected size in all six here examined isolates . These data provide evidence in support of B . recurrentis possessing a longer plasmid of approximately 160 kbp than present in the currently available strain A1 in GenBank . Although B . recurrentis is hypothesized to be a louse-adapted clone of B . duttonii , interestingly , there is no counterpart of the smallest plasmid pl6 in the B . duttonii genome . However , a highly similar plasmid is present in other tick-borne relapsing fever borreliae , i . e . B . crocidurae , B . miyamotoi and B . hermsii . To overcome the difficulties of distinguishing B . duttonii and B . recurrentis [15] , it might be of interest to consider this plasmid as a potential target for diagnostic and/or typing tools . However , it would be necessary to investigate the stability and retention of this specific plasmid in vivo and in vitro . It is known that RF Borrelia retain infectivity and majority of their plasmids with minimal rearrangements on the largest plasmid even after 100 cycles of in vitro cultivation [50] . Furthermore , compared to the reference genome of strain A1 , all of the currently examined isolates contained a shorter version of the plasmid of approximately 5 kb . The length of the de novo assembled contigs representative of this plasmid did show some differences depending on the assembly algorithm employed ( as exemplified in Fig 4 with PUfA_CLC and PUfA_spades ) . However , irrespective of the assembly method employed ( mapping , de novo assembly with CLC or SPAdes ) the right 3’ end of the plasmid was not detected . PCR reactions with primers designed to this region were negative as well . We propose that it may have resulted from a contamination and/or sequencing error in the reference genome A1 , as this region has a much higher GC % content compared to the rest of the plasmid ( Fig 4 ) . Likely due to the low passage number the isolates investigated in this study contained in the raw reads a proportion of human DNA ( 30–60%; Table 3 ) which interfered with de novo assembly . Therefore all reads were first mapped to the human genome and the remaining unmapped reads were assembled using SPAdes or the CLC de novo assembler . This improved the assembly significantly and produced longer contigs , in some cases even covering the whole chromosome ( PBek in Table 4 ) . Although the chromosome was easily identified and assembled even with de novo assemblers , the assembly of plasmids was more tedious , similar as recently published for Lyme disease borrelia [52] . The contigs matching to the main chromosome were in synteny with the published reference genome ( see supplementary data S1 Data ) . When comparing contigs representing plasmids from the two employed assemblers , it is unclear which performed better . CLC produced more and shorter contigs that aligned more unambiguously to the reference genome and had less identified misassemblies ( Tables 4 and 5 ) . On the other hand , SPAdes produced fewer and longer contigs which matched to regions in more than one plasmid of the reference A1 genome , thus rendering their location inconclusive ( Tables 4 and 5 ) . RF borreliae have a smaller number of plasmids than Lyme borreliosis group spirochetes [17] . However , RF borreliae contain several copies of the so called vlp and vsp genes present on different plasmids . B . recurrentis A1 strain has 17 intact vlp genes and 29 pseudogenes and 10 vsp genes [17] . These genes code for surface proteins that facilitate the evasion of the human immune system , and are responsible for the febrile relapses of the clinical picture of the disease [48 , 49] . Due to both biological constraints and inherent methodological limitations of short-read sequencing and assembly , the construction of complete/whole plasmids in RF Borrelia could not have been unambiguously achieved with only de novo assembly approaches . The high sequence similarity and number of vlp and vsp genes present in the genome impeded the design of specific PCR primers necessary for the resolution of the position of ambiguous contigs matching to the location of two or more plasmids , as well as the size of the plasmids . Interestingly , when mapping the reads directly to the reference genome , a discrepancy in coverage of the regions of the genome containing vlp and vsp genes was observed . The coverage of the vlp and vsp genes was higher than the surrounding regions ( 10-20x higher ) , which would be suggestive of the presence of even more copies of the vlp and vsp genes , than previously reported in the reference genome , similar as already observed for the differing size of plasmid pl124 [51] . New and powerful emerging technologies that exploit long-read sequencing and hybrid assemblers combining long and short reads could offer more insights and allow the elaboration of the exact number and location of vlp and vsp genes , as well as the length of the plasmids in RF borreliae , as already done in a proof-of-concept manner for Borrelia burgdorferi [52] . Nevertheless , the currently obtained sequences are a valuable contribution in the field of RF Borrelia research and could present the basis for the development of molecular tools to facilitate the diagnostics and characterization of LBRF Borrelia . | Louse-borne relapsing fever , as the name suggests , is the only relapsing fever transmitted by lice , and caused by the spirochaete Borrelia recurrentis . Today it is endemic to the Horn of Africa , but due to the cosmopolitan nature of the vector , the pathogen still bears epidemic potential to spread globally among vulnerable populations . The most recent account of that has been observed among migrants arriving to Europe in 2015 . Up to date , only one strain was sequenced , thus hampering the development of species-specific typing tools . We employed state-of-the-art high-throughput sequencing to six B . recurrentis isolates obtained at different time-points and currently available in culture . Our aim was to address the question of genome variability of this pathogen at the highest resolution and provide information necessary for the development of specific typing tools . B . recurrentis has highly conserved genomes , differing in 29–38 SNPs compared to the reference genome B . recurrentis A1 , all identified outside the loci currently developed and used for relapsing fever Borrelia typing . Therefore , applying these typing methods would render them indistinguishable , while at the SNP level we found a distinction between isolates obtained in 2015 from migrants and the two historical isolates . Our data provide first insights in the genome variability and baseline information necessary for further studies of the molecular epidemiology of the pathogen and for the development of improved diagnostic tools . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"tropical",
"diseases",
"microbiology",
"geographical",
"locations",
"relapsing",
"fever",
"bacterial",
"diseases",
"genome",
"analysis",
"neglected",
"tropical",
"diseases... | 2017 | First insights in the variability of Borrelia recurrentis genomes |
The modern synthetic view of human evolution proposes that the fixation of novel mutations is driven by the balance among selective advantage , selective disadvantage , and genetic drift . When considering the global architecture of the human genome , the same model can be applied to understanding the rapid acquisition and proliferation of exogenous DNA . To explore the evolutionary forces that might have morphed human genome architecture , we investigated the origin , composition , and functional potential of numts ( nuclear mitochondrial pseudogenes ) , partial copies of the mitochondrial genome found abundantly in chromosomal DNA . Our data indicate that these elements are unlikely to be advantageous , since they possess no gross positional , transcriptional , or translational features that might indicate beneficial functionality subsequent to integration . Using sequence analysis and fossil dating , we also show a probable burst of integration of numts in the primate lineage that centers on the prosimian–anthropoid split , mimics closely the temporal distribution of Alu and processed pseudogene acquisition , and coincides with the major climatic change at the Paleocene–Eocene boundary . We therefore propose a model according to which the gross architecture and repeat distribution of the human genome can be largely accounted for by a population bottleneck early in the anthropoid lineage and subsequent effectively neutral fixation of repetitive DNA , rather than positive selection or unusual insertion pressures .
The present-day human genome arose from the prosimian ancestor through a series of complex chromosomal and local rearrangements . An important feature of our genome , used frequently to understand the adaptive forces that have led to its present-day topology , is the common prevalence of repetitive sequences . Analyses of the Alu family , a 300-bp , primate-specific retrotransposon that represents the most abundant class of repeats [1] , have indicated that they underwent a seemingly rapid proliferation at two major evolutionary junctions: the prosimian-anthropoid split some 37–55 million years ago ( mya ) and the platyrrhine/catarrhine split thereafter [2] . Some studies have pointed to a correlation between retrotransposon expansion and speciation [3 , 4] and have suggested that the unidirectional proliferation of more than ten copies of the retrotransposon [1 , 5] might provide a useful marker for tracing phylogeny [6 , 7] . Despite the apparent importance of repeat expansion to understanding the origins of the human genome , the mechanisms of repeat proliferation are poorly understood . For Alu repeats , a model of increased retrotransposition activity has been proposed [8] , but the underlying evolutionary forces behind such a mechanism are unclear . To investigate the evolutionary forces that might govern the acquisition and retention of repetitive elements in the human genome , we selected an entirely different class of repeat whose mechanisms for insertion , deletion , and selection are so fundamentally different from Alu that any commonality in their evolutionary dynamic is probably due to the fact that they share the same population size , rather than any underlying biological mechanism . We focused on numts ( nuclear mitochondrial sequences/pseudogenes ) , partial copies of the mitochondrial genome found abundantly in chromosomal DNA . Since the first demonstration of organellar sequence embedded in nuclear DNA [9] , numts have been described in several mammalian species , as well as over 70 other eukaryotes [10–12] . The varying level of homology between these sequences and the present-day mitochondrial genome , as well as population and family polymorphisms , indicates that the nuclear transfer of mtDNA is an ongoing process [13–21 , 28] . In contrast to plants and fungi , in which numts have arisen from both RNA- and DNA-mediated mitochondrial DNA ( mt-DNA ) transfers [22] , the origin of numts in metazoans has been proposed to be DNA- rather than RNA-mediated [23–25] . As such , the numts family of repeats represents a useful tool for evolutionary analysis since its proliferation mechanism is distinct from Alu elements , in that it does not rely on retrotransposition .
We first used the assembled human genomic sequence ( Build 36 ) to investigate the prevalence and distribution of numts in the human genome . Using default sequence alignment selection criteria ( e-value <10 ) , we identified 2 , 329 numts fragments that range in size from <100 bp to 16 kb ( Figure 1 ) , a number consistent with previous studies [19 , 23 , 26] . Fine-mapping of numts showed many instances in which multiple , seemingly independent , fragments map in close proximity to one another , suggesting a higher-order organization , whereby each numts does not represent an independent integration , but is rather a fossil of a single ancestral integration ( Table S1 ) . Clustering of such numts blocks indicated that the human genome likely contains in excess of ∼1 , 200 numts elements ( Table S2 ) . A similar analysis of the mouse and rat assembled genomes showed a marked numts paucity , with 636 and 529 numts fragments , respectively . By contrast , the recent draft of the chimp genome contains numbers comparable to humans , ≥1 , 280 numts , suggesting that these elements might have undergone a dramatic expansion in the primate lineage ( Table S2 ) . These observations are unlikely to be due to inappropriate exclusion of numts sequences from the draft genome assemblies , since analysis of the raw trace data ( i . e . , all individual preassembly sequence reads ) showed a similar percent identity distribution of putative numts , with both sequence collections peaking at 82%–88% identity with the present-day mitochondrial sequence ( data not shown ) . Prior to further analysis , we corroborated our computational data in two ways . First , we performed flourescent in situ hybridization ( FISH ) with mtDNA as a molecular probe on interphase and metaphase nuclei of mtDNA-depleted cells as target DNA . Consistent with the predicted abundance of numt in the nuclear genome , we detected fluorescence signals scattered along each chromosome ( Figure 2 ) . We observed a similar pattern on chromosomes of mtDNA-depleted lymphoblast cells from chimp , gorilla , and orangutan ( Figure 2 ) . These data indicate that the numts element is distributed widely in the genomes of these species and that the actual numts population is probably larger than our computational predictions , potentially reflecting our criteria for numts identification . In addition , amplification from a monochromosomal hybrid panel and subsequent sequencing of 24 randomly selected nucleo–numts junctions , showed that in each case the amplification and sequence data matched exactly with the computationally predicted sequence of each numts ( data not shown ) . We next investigated numts proliferation . Previous studies have indicated that the mechanism of integration of these repeat elements into the genome is distinct from retroviral insertion or recombination [10] , thus enabling us to study the acquisition characteristics of exogenous DNA in a genome context-independent fashion . To identify a subpopulation of numts that arose by independent integrations , rather than a single integration followed by subsequent segmental duplication , we first correlated the positions of all identified numts with the segmental duplication map . In agreement with previous studies founded on numts base substitution rates [13] , we determined that although some numts proliferated through chromosomal rearrangements , the majority of numts acquisition of the genome reflects independent integration; some 3%–5% of build 36 has been identified as segmental duplication [27] , and only 4% of all numts map to these regions . To further confirm these observations , we compared 500 bp of nuclear sequence on either side of each putative integration and found no similarities among the nuclear junction sequences ( data not shown ) . We next asked whether numts integration is likely to be genome sequence independent by evaluating the sequence characteristics of nucleo–numts junctions . First , we asked whether there is any observable enrichment for a recognizable element at repeat junctions . A comparison of 1 kb of flanking nuclear junction sequence surrounding 266 numts with the entire human genome showed an initial deficit of repeats , returning to genome-wide levels 500–600 bp past the insertion boundary ( Figures 3 and S1 ) . This suggested that: ( a ) there is no repeat excess at the boundary and ( b ) the true boundary probably lies 500–600 bp away from our initial prediction . In addition , the possibility of a TE ( transposable element ) insertional mechanism was also deemed unlikely , since we found no evidence of sequence duplication anywhere within the 1kb region that flanks the boundaries of each numt . Our data suggest that the human genome has probably acquired a minimum of several hundred numts , most of which arose in an ancestor as independent events , in a process that is still ongoing [28] and can have detrimental effects to gene function [29] . Even though the mechanism of insertion of numts is clearly different from that of Alu elements , especially since numts cannot mediate their own proliferation , similarities or differences in the fitness consequences of those insertions are less obvious . Although numts are unlikely targets for unequal exchange events , they might contain potentially functional genes that could be co-opted into some nuclear role . Thus , we assessed for possible fitness effects of numts insertion by examining their positional preference in the genome , as well as their transcriptional and translational potential . To interrogate whether numts have positional preference , we determined the relative distribution of all large numts arisen by independent integrations with respect to the coding sequence distribution of the genome . We conducted two tests , one for numts >1 kb ( n = 99 ) and one for numts >500 bp ( n = 121 ) . None of the numts considered for the two experiments occurred in exons . In build 36 , the fraction of the intronic human genome is ∼28 . 85% . The percentage of intronic numts is 22 . 3% ( 22/99; binomially p = 0 . 086 ) for numts >1 kb and 21 . 5% ( 26/121; p = 0 . 042 ) for the those >500 bp . Thus , numts appear to be distributed relatively randomly in the genome ( Figure 4 ) , but a slight statistical tendency towards intergenic intervals was observed , probably underlying the higher potential of intragenic insertions for a deleterious effect . Overall , we conclude that numts position within the genome provides little evidence of its use for transcriptional control . Next , we considered the possibility that numts might have functionality at the mRNA level . We first examined whether numts are transcribed , by interrogating each numts against dbEST . To reduce the incidence of matches with dbEST due to short segments of sequence , we restricted our queries to numts with length greater than 1 kb and numts longer than 500 bp . Of the 99 numts >1 kb evaluated , ( 23/99 ) 23 . 23% were represented in dbEST , also from the 121 numts >500 bp considered , ( 33/121 ) 27 . 27% were found in dbEST . Reverse-transcriptase PCR ( RT-PCR ) of 24 randomly selected , nonoverlapping ESTs also indicated that the majority of these sequences represent bona fide transcription , since in 22 instances we amplified successfully the correct fragment from a panel of eight adult human RNA samples by RT-PCR ( data not shown ) . However , we found no positional preference for putatively transcribed numts , suggesting that numts mRNA is unlikely to exert a cis-acting regulatory role . Finally , we considered the possibility that the introduction of numts into the genome provided the template for new protein sequence , despite the fact that the nuclear and mitochondrial genome have different genetic codes . We therefore examined the translational potential of each numts in all six reading frames ( Figure 5 ) . Translating with the nuclear code results in a distribution of open reading frame ( ORF ) lengths indistinguishable from random sequence ( 3/64 codons are stop , therefore random sequence will generate ORF sizes with a mean size of ∼20 codons ) . Although there is a slight excess of long ORFs ( suggesting that a small fraction of numts might be translated ) , the overall distribution of ORF lengths is approximately exponential with a mean length of 5–15 codons . Cumulatively , our data suggest that there is little evidence for overt functionality for the majority of numts , and although we cannot formally exclude the possibility that some individual repeats have a biological role ( and may thus be obvious targets for positive selection ) , the overall population of this repeat is likely to be on average evolutionarily neutral or deleterious . To gain a better understanding of the evolutionary dynamics of numts , we sought to determine the most likely time of integration of each numts into the nuclear genome . To do so , we aligned each numts to a collection of complete modern mtDNA sequences spanning the primate radiation . The time of each integration was inferred independently with multiple fossil calibration points [30] under an overdispersed model of molecular evolution , accounting for variation in evolutionary rates within and between numts and the extant mitochondria ( Figure 6A ) [31] . In contrast to an expectation of progressive numts accumulation during evolutionary time , we were surprised to find an apparent burst of numts integrations at approximately 54 mya . Focusing first on numts >1 kb in length , we found that ∼76% out of the 99 unique integration events , have an estimated time of insertion within 10 mya of 54 mya ( Figure 6C ) . Next , we considered the numts >500 bp , and from 121 unique integration events ∼75% also occurred within 10 mya of 54 mya ( Figure 6E ) . Thus , 75%–80% of all numts integrations appear to have occurred within a relatively narrow window of time around 54 mya , between the New World Monkey and Old World monkey transition ( Figure 6B and 6D ) . Importantly , this estimate is likely to remain true irrespective of assumptions regarding the nucleotide substitution rate of numts versus mtDNA , as judged by a confidence interval plot of the 121 500-bp+ numts ( Figure S2 ) .
Most numts appear to have accumulated in a 10-millon-year window centered around 54 mya . Importantly , other repetitive elements show a similar pattern , including Alu repeats [2 , 32] and processed pseudogenes [33] , suggesting a period of intense DNA acquisition in the ancestral genome . Given that numts are markedly distinct from Alu repeats and other retrotransposons in both their mechanism of integration , as well as proliferation ( especially since numts lack the ability to self propagate ) , the force behind the expansion of repeats is likely independent of genome structure . This notion is further supported by the fact that the boundaries of numts integration show no marked enrichment for any sequence elements ( Figure 3 ) . It will always remain a formal possibility that numts integration was primarily driven by positive selection for the accumulation of these elements . However , the absence of overt functionality of numts in the present-day genome , and the fact that numts integration is a continuing process [10] , principally detected because of its disease phenotype , argues against this hypothesis . Thus , we arrive at three important questions concerning the evolutionary history of numts: ( 1 ) Why did so many numts accumulate approximately 54 mya ? ( 2 ) Why did they stop accumulating ? ( 3 ) Why does this time period correspond temporally with accumulation of other entirely unrelated genetic elements ? The theory that governs the evolutionary dynamics of TEs can provide important clues about the mechanism of acquisition and retention of numt , Alu , and other repeat elements in the human genome . In an infinite sized population , the change in the mean number of TEs per individual , , is approximately where Vn is the variance in copy number between individuals , μ is the rate of new insertions , ν is the rate of new deletions , and is population mean fitness [34 , 35] . Thus , in an infinite-sized population , TE copy number is governed by a balance between the effects of new insertion , new deletion , and selection . By contrast , in a finite population , Equation 1 will approximately hold whenever is much bigger than 1/N , where N is the effective size of the population . If 1/N > , TE copy number will rise ( if the insertion rate is greater than the deletion rate ) or fall ( if deletion is more frequent than insertion ) . Thus , a sudden change to TE copy number could reflect a sudden decrease in population size , shifting the balance between selection and mutation forces to one where genetic drift ruled and allowed for unbounded increase in TEs . The Liu et al . hypothesis [8] , on the other hand , suggests that the increase in Alu copy number may have resulted from a sudden increase in μ , the rate of insertion . If we assume that numts integrations are principally weakly deleterious on average ( a notion supported by their ongoing contribution to disease ) , an examination of Equation 1 suggests that a simple population size hypothesis can provide an answer to all three of our questions . We begin by assuming that prior to 54 mya , the effective population size of the primate ancestor was relatively large , leading to an insertion/deletion/selection equilibrium with numts count being few and held stable at that low value ( which is consistent with the relative paucity of numts in the mouse and rat lineages ) . However , if we further assume that at approximately 54 mya , effective population sizes declined dramatically , to a point where 1/N > , then numts would for evolutionary purposes become effectively neutral , and , during their period of effective neutrality , they would accumulate with little selective check , at a rate proportional to μ−ν ( the difference between the insertion and deletion rates of an element ) . Since population size changes affect everything in the genome , elements with high insertion rates ( such as Alu elements ) would be expected to accumulate in great abundance ( which they do ) , whereas elements with relatively low insertion rates ( such as numts ) also accumulated , albeit in fewer numbers . Finally , a subsequent increase in effective population size would shift the population back into an insertion/deletion/selection equilibrium , and the period of accumulation would end . Clearly , the assumptions of relative numts neutrality and of a population bottleneck at ∼54 mya cannot be proven definitively . Nonetheless , based on observations of the landscape of the present day genome of humans and other species , our proposed evolutionary model has many attractive features . First , it provides a common mechanism ( decline in effective population size ) for the increase in numbers of unrelated repetitive elements . Second , it explains both the sudden increase in repetitive DNA , and the later cessation of the increase . Third , the timing of the event , occurring immediately prior to the adaptive radiation of monkeys , is highly evocative , reminiscent of a Wrightian/Simpsonian view of speciation: a large population of stem anthropoids splintered into multiple demes . One or more such small deme accumulated repetitive DNA in abundance , which in turn may have served as a post-zygotic reproduction barrier with the original population . This isolated deme ultimately speciated and underwent an adaptive radiation into the anthropoid primates . It is notable ( and unlikely to be coincidental ) that the timing of the repeat-inferred bottleneck at ∼54 mya coincides with a major environmental disturbance at the Paleocene–Eocene boundary ( ∼55 mya ) , which strongly effected global mammalian faunas and corresponds to the first appearance of primates in the fossil record of the northern hemisphere [36] . This hypothesis suggests that human and primate genomic architecture , with its abundance of repetitive elements , arose primarily by evolutionary happenstance; although it remains plausible ( and indeed , probable ) that some integrons were subsequently co-opted into an interesting use such as X inactivation [37] or perhaps gene regulation [38] , these complicated hypotheses do not explain satisfactorily the bulk of human genomic architecture . A simple explanation states that the population that gave rise to primates was quite small , and as a result the genomic architecture of primates may have resulted from effectively neutral integrations of repetitive DNA .
Human mitochondrial genome sequence was compared against genomic sequence with BLAST ( NCBI Build 36 ) . The process was repeated for the mitochondrial sequence of chimp , mouse and rat against the following draft builds: chimp Build 2 ( October 2005 ) , mouse Build 33 ( May 2004; mm5 ) , and rat Version 3 . 1 ( June 2003; rn3 ) . In each case , hits that scored with an expected value <10 were retained . All annotations ( repeat classes , gene boundaries , etc . ) were taken from the University of California Santa Cruz genome browser , http://genome . ucsc . edu/ . Blast hits were sorted by genomic position , and the differences ( “gaps” ) between consecutive hits on both the genomic and mitochondrial scales were calculated . Pairs of hits that had a ratio of mitochondrial gap size to genomic gap size between 0 . 9 and 1 . 1 were assigned to be in the same block ( hand picked ) . The numts distribution plots were created using Circos ( http://mkweb . bcgsc . ca/circos/ ) . We used high-molecular-weight genomic DNA and highly purified mt-DNA from HeLa cells ( kindly provided by Samuel E . Bennett , Oregon State University , Corvallis , Oregon , United States ) for PCR . For generating molecular probes in FISH experiments , we used two different PCR products: the complete mitochondrial genome ( 16 . 3 kb ) amplified with the TaKaRa PCR kit ( Fisher Scientific , https://new . fishersci . com/ ) , using conditions as described [39] . Alternatively , we designed seventeen PCR primer sets and amplified overlapping ∼1-kb fragments , covering the entire mt-DNA sequence . Primers and detailed PCR conditions are available upon request . The nonhuman primate immortalized Epstein–Barr virus–stimulated cell lines of common chimpanzee ( Pan troglodytes ) , lowland gorilla ( Gorilla gorilla , CRL 1854 ) , and orangutan ( Pongo pygmaeus ) , were purchased from the American Type Culture Collection ( ATCC , http://www . atcc . org/ ) . The pygmy chimp ( Pan paniscus ) lymphoblast sample was kindly provided by D . Nelson at Baylor College of Medicine , Houston , Texas , United States . Human and primate lymphoblasts were depleted of mt-DNA according to the slightly modified protocol of King and Attardi [40] . Cells were grown for 5–6 d in DMEM enriched with 10% FCS glucose ( 4 , 500 mg/ml ) , sodium pyruvate ( 1 mM ) , uridine ( 50 μl/ml ) , and ethidium bromide ( 50 μl/ml ) . Normal and mt-DNA-depleted lymphoblasts were harvested using standard methods . FISH was performed on metaphase and interphase cells as described [41] . Briefly , PCR products were labeled with biotin ( Life Technologies-GibcoBRL , http://www . invitrogen . com/ ) or digoxigenin ( Boehringer Mannheim , http://www . roche . com/ ) by nick translation . Biotin was detected with FITC-avidin DCS ( fluoresces green; Vector Labs , http://www . vectorlabs . com/ ) and digoxigenin was detected with rhodamine-anti-digoxigenin antibodies ( fluoresces red; Sigma , http://www . sigmaaldrich . com/ ) . Chromosomes were counterstained with DAPI diluted in Vectashield antifade ( Vector Labs ) . Cells were viewed under a Zeiss Axioskop fluorescence microscope ( http://www . zeiss . com/ ) equipped with appropriate filter combinations . Monochromatic images were captured and pseudocolored using MacProbe 4 . 2 . 2/Power Macintosh G4 system ( Apple , http://www . apple . com/; Perceptive Scientific Instruments , http://www . perceptive . co . uk/ ) . The flanking sequence composition of 266 numts was compared to 50 , 000 randomly chosen sequences drawn uniformly from the human genome . For each flanking sequence , and each randomly drawn sequence , the proportion of the sequence covered by various repeat families ( Alu , L1 , MALR , etc . ) and repeat classes ( SINE , LINE , LTR , etc . ) was calculated and the repeat composition of each category was evaluated with a t-test . Once the composition and distribution of numts blocks was established , we designed primers to amplify 250–400-bp junction fragments whereby one primer was anchored at unique nuclear sequence and the other primer was situated at the edge of a numts block . We performed PCR using standard condition on human–rodent monochromosomal hybrids as described [42] . We designed primers from ESTs that matched human numts with >98% identity over 200 bp of sequence . To ascertain their expression patterns , we generated amplicons from eight adult human cDNAs ( Clontech , http://www . clontech . com/ ) according to manufacturer's instructions . Each numts was translated in all six possible reading frames . An ORF was defined as the sequence between two stop codons , and the frame with the longest mean ORF length was chosen for inclusion in the analysis . Numts were translated using the nuclear genetic codes ( stop codons TAA/TAG/TGA ) . Each numts was aligned individually with ClustalW ( http://www . ebi . ac . uk/clustalw/ ) to a collection of complete modern mtDNA sequences spanning the primate radiation , rooted by a carnivore outgroup . All pairwise per-site divergences were calculated with the PHYLIP program ( http://evolution . genetics . washington . edu/phylip . html ) dnadist , using a Kimura 2-parameter substitution model to correct for multiple hits . For each numt , the evolutionary tree was inferred by both parsimony ( using the PHYLIP dnapars program ) and neighbor-joining ( using the PHYLIP program neighbor ) . In all cases the expected phylogeny [2] of the primate and outgroup was recovered , but the exact position of the numts varied slightly ( see below ) . Once the tree was inferred for each numt , the number of substitutions per branch was estimated by least-squares minimization using the PHYLIP program fitch with default parameters . To account for any potential uncertainty in the divergence time between extant primates , nonconstancy of evolutionary rates within and among different functional portions of the extant mtDNA , and perhaps vastly different rates of evolution among nuclear pseudogene copies of mtDNA and extant functional mtDNA , the time of each integration was inferred with dating [31] , under a stationary substitution model with multiple fossil calibration points [30] . In all cases , the stationary model fit better than the constant rate Poisson model by several orders of magnitude . Confidence intervals for each integration were also calculated [31] .
The National Center for Biotechnology Information ( NCBI ) Genbank ( http://www . ncbi . nlm . nih . gov/sites/entrez ? db=Nucleotide ) accession number for the human mitochondrial genome sequence discussed in this paper is NC_001807 . | Throughout evolutionary history , fragments of the mitochondrial genome , known as numts ( for nuclear mitochondrial sequences ) , have been inserted into the nuclear genome . These fragments are distinct from all other classes of repetitive DNA found in nuclear genomes , not least because they are incapable of mediating their own proliferation . Taking advantage of their unique evolutionary properties , we have used numts to improve our understanding of the architecture of the human genome with special emphasis on the mechanism of acquisition and retention of repeat sequences , which comprise the bulk of nuclear DNA . We find that numts are unlikely to have any evolutionary benefit driving their retention . Moreover , numts are not acquired randomly during evolutionary time . Instead , their rate of acquisition spikes dramatically around pronounced population bottlenecks , in a manner reminiscent of other repeat classes . Therefore , we propose that the primary driving force of repeat acquisition in the genome is not selection , but random genetic drift , whose force becomes pronounced during profound reductions of population size . Our findings support the theory of neutral evolution , according to which random genetic drift exerts an influence on the acquisition of DNA changes that far outweighs the power of positive selection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"primates",
"mammals",
"computational",
"biology",
"evolutionary",
"biology",
"homo",
"(human)",
"genetics",
"and",
"genomics",
"mus",
"(mouse)"
] | 2007 | Population Bottlenecks as a Potential Major Shaping Force of Human Genome Architecture |
The current genetic makeup of Latin America has been shaped by a history of extensive admixture between Africans , Europeans and Native Americans , a process taking place within the context of extensive geographic and social stratification . We estimated individual ancestry proportions in a sample of 7 , 342 subjects ascertained in five countries ( Brazil , Chile , Colombia , México and Perú ) . These individuals were also characterized for a range of physical appearance traits and for self-perception of ancestry . The geographic distribution of admixture proportions in this sample reveals extensive population structure , illustrating the continuing impact of demographic history on the genetic diversity of Latin America . Significant ancestry effects were detected for most phenotypes studied . However , ancestry generally explains only a modest proportion of total phenotypic variation . Genetically estimated and self-perceived ancestry correlate significantly , but certain physical attributes have a strong impact on self-perception and bias self-perception of ancestry relative to genetically estimated ancestry .
Understanding the basis of a variation in human physical appearance has been a topic of long-standing research interest . However , little is known about the genetic basis of most of this variation . An exception is pigmentation , which has been the focus of considerable research , particularly in Europeans [1]–[4] . Refining our knowledge on the genetics of physical appearance in human populations is of considerable evolutionary , biomedical and forensic importance . This research is also of broad social interest due to its bearing on debates around notions of self-identity , ethnicity and race . Latin America provides an advantageous setting in which to examine the impact of genetic variation on physical appearance . The region has a history of extensive admixture between three continental populations: Africans , Europeans and Native Americans [5] , [6] . Latin America also provides an informative context in which to explore the perception of variation in physical appearance . The region has a unique history relating to the social and cultural politics of ethnicity , race and nation [7]–[9] . A considerable number of genetic studies have examined admixture in Latin America [10]–[14] . However , these analyses have mostly been based on relatively small samples and focused mainly on describing patterns of variation in admixture proportions between individuals and countries/regions . Few studies have examined the impact of genetic ancestry on physical appearance or the relationship of these to individual notions of ethnicity and ancestry [15] , [16] . In this paper we present the first phase of a research program focused on the genetics of physical appearance in Latin Americans . We base this program on a sample of over 7 , 000 individuals ascertained in five countries: Brazil , Chile , Colombia , México and Perú . Information was obtained for a range of socio-demographic variables , physical attributes and self-perception of ancestry . Here we report analyses based on individual mean genome admixture proportions . Coordinate-based spatial analyses illustrate the significant variation in ancestry existing across Latin America , in agreement with demographic history and census information . Significant effects of ancestry were detected for most of the phenotypes examined , and the direction of these effects agrees with the phenotypic differentiation of Africans , Europeans and Native Americans . Finally , we observe that certain phenotypes have a strong impact on self-perception and that these phenotypes bias self-perceived relative to genetically estimated ancestry .
We estimated individual African/European/Native American admixture proportions with data for 30 highly informative SNPs using the ADMIXTURE program [17] . These markers were chosen from the 5 , 000 proposed by Paschou et al ( 2010 ) [18] as highly informative for continental ancestry estimation ( see Methods ) . The selected set of markers produced individual ancestry estimates in 372 Colombians , included in a recent genome-wide association study [19] , with correlations of ∼70% ( for the three continental ancestries ) compared to estimates obtained with 50 , 000 markers ( LD-pruned ) , and identical sample means . Although we estimated individual ancestry with a relatively small number of markers , we verified that the inferences drawn are robust to the level of uncertainty of the estimates obtained ( see below ) . Consistent with previous studies , we observe extensive variation in ancestry between countries ( Table 1 ) as well as between individuals within countries ( Text S1 ) and between socioeconomic strata ( Text S2 ) [12] , [13] , [20]–[22] . In order to obtain a spatial representation of variation in ancestry we obtained interpolated maps based on the geographic coordinates for the birthplaces of research volunteers . The geographic distribution of these birthplaces ( Figure 1 and Figure S4 ) overlaps with regional population density from national census data ( Figure S5 ) . Consistent with this pattern , the number of volunteers for each birthplace correlates with census size for these localities: Brazil ( r = 0 . 32 , p-value <10−5 ) , Chile ( r = 0 . 51 , p-value <10−4 ) , Colombia ( r = 0 . 54 , p-value <10−13 ) , Mexico ( r = 0 . 44 , p-value <10−8 ) , Perú ( r = 0 . 41 , p-value <10−4 ) . Few volunteer birthplaces were thus located in sparsely populated regions ( e . g . Amazonia ) and geographic interpolation of ancestry in those regions should be regarded with special caution . The Brazilian sample ( Figure 1A ) shows widespread European ancestry with the highest levels being observed in the south . African ancestry is also widespread ( except for the south ) and reaches its highest values in the East of the country . Native American ancestry is highest in the north-west ( Amazonia ) . The Chilean sample ( Figure 1B ) shows the least regional variation , with low levels of African ancestry throughout the country . European and Native American ancestry are relatively uniform , although somewhat higher European ancestry is seen around the main urban areas of the north and centre , Native ancestry predominating elsewhere , particularly in the south . The Colombian sample ( Figure 1C ) shows highest African ancestry in the coastal regions ( particularly on the Pacific ) and highest European ancestry in central areas . Native ancestry appears highest in the south-west and in the east of the country ( Amazonia ) but interpolations in these areas are based on few data points . In the Mexican sample ( Figure 1D ) Native American ancestry is highest in the centre/south of the country with the north showing the highest proportion of European Ancestry . African ancestry is generally low across Mexico except for a few coastal regions . The Peruvian sample ( Figure 1E ) shows substantial Native American ancestry throughout the country , particularly in the south , European ancestry appears highest around northern/central areas . African ancestry in Peru is generally low , except for parts of the northern coast . To evaluate the statistical significance of the observed spatial variation in ancestry we calculated Moran's Index ( I ) of association between each individual ancestry component and spatial location . These were significant for the three ancestries in all countries ( p-values <0 . 02 ) . Since the three ancestry components are not independent , we also calculated canonical correlation coefficients between ancestry and geographic location . These were also significant for all countries ( p-values <0 . 001 ) . The variation in ancestry seen in the admixture maps of Figure 1 also result in highly significant correlations of the three ancestries with altitude of birthplace ( p-values <2×10−16 for the three ancestries ) : African and European ancestry decreases with altitude ( r of −0 . 24 and −0 . 39 , respectively ) , while Native American ancestry increases ( r = 0 . 48 ) . The Kriging interpolation scheme used in building the maps of Figure 1 uses the mean ancestry at each birthplace and does not provide information on the extent of individual variation in ancestry at each map location . In the 102 birthplaces with 10 or more individuals sampled we observe that the standard deviation in the three individual ancestry estimates extends over a wide range: African ( 0 . 012–0 . 022 ) , European ( 0 . 046–0 . 273 ) and Native American ( 0 . 039–0 . 274 ) . We evaluated the correlation of this variation in individual ancestry with the census size of these localities and found a significant positive correlation for all ancestries ( r>0 . 3 , p-values <0 . 01 ) . Regression of phenotypic variation on genetic ancestry ( taking Native American as reference ) demonstrates a significant effect for most of the traits examined ( p-value <10−3 using a conservative Bonferroni multiple testing correction , Table 2 ) . Among the non-facial phenotypes ( accounting for sex , country , age , educational attainment and wealth ) higher European ancestry is associated with: increased height , lighter pigmentation ( of hair , skin and eyes ) ( Figure S6 ) , greater hair curliness and male pattern baldness . Hair graying approaches statistical significance ( p-value 10−2 ) . Higher African ancestry is associated with: increased height , higher skin pigmentation and greater hair curliness . The proportion of phenotypic variance explained by ancestry is highest for skin pigmentation ( 19% ) followed by hair shape ( 8% ) and color of eyes and hair ( 4% and 5% , respectively ) but at most 1% for the other phenotypes . We also observed highly significant effects of educational attainment ( p-value 3 . 87×10−13 ) and age ( p-value <2×10−16 ) on height , with height increasing for individuals born more recently at a rate ∼1 cm every 10 years ( Text S3 ) . Genetic ancestry also has a range of effects on facial features , both in terms of size and shape , after accounting for height and BMI ( in addition to the other covariates ) . Higher European ancestry is associated with reduced eye fold and an overall smaller face ( centroid size ) . Face size and shape effects were also evaluated through the analysis of all pair-wise inter-landmark distances ( Table S3 ) . Amongst these distances , 133 and 2 show significant effects of European and African ancestry , respectively ( p-values 10−6 assuming a conservative Bonferroni multiple testing correction; Table S3 ) . The most significant effects of European ancestry ( P<10−10 ) involve mainly distances between landmarks placed on the lips and nose . Face shape variation , independent of size , was assessed via Principal Components ( PCs ) of procrustes 3D coordinates . Significant effects of European ancestry were detected for PCs 1 and 3–5 , while African ancestry impacts on PCs 1 , 2 and 4 ( Table 2 , Text S5 and Figure S3 ) . These 5 PCs account for ∼55% of the variation in face shape captured by the 36 landmarks placed on the facial photographs , with ancestry explaining up to 5% of the variance in PC scores ( for PC4 ) . Examination of the correlation between inter-landmark distances and facial PCs , indicates that the highest correlation of distances between landmarks of the lips and nose is with PC4 ( results not shown ) , consistent with this PC showing the largest proportion of variance explained by ancestry ( Table 2 ) . Four ethno/racial categories ( “Black” , “White” , “Native” and “Mixed” ) are commonly used across Latin America in national censuses and other population surveys . We contrasted genetic ancestry and skin pigmentation ( as measured by the melanin index ) across these four self-estimated categories for the countries sampled ( Figure 2 and Table S4 ) . Within each country there is a gradient of decreasing European ancestry ( and increasing pigmentation ) for the “White” , “Mixed” and “Native/Black” categories . Across countries , skin pigmentation is relatively uniform within ethnicity categories , except for “Black” . For “White” , “Native” and “Mixed” the mean melanin index across countries varies within ∼2 units , while the range for “Black” is ∼25 units . By contrast , genetic ancestry varies greatly between countries for all ethnicity categories . For example , European ancestry varies across countries by about 40% for “White” , “Mixed” and “Native” and about 20% for “Black” ( Figure 2; estimates for African and Native American ancestry are shown in Table S4 ) . Contrasting self-perceived ( ranked into five bands at 20% increments ) and genetically estimated continental ancestry we observe a moderate , but highly significant , correlation: America: r = 0 . 48 , P<2 . 2×10−6 , Europe: r = 0 . 48 , P<2 . 2×10−6 , Africa: r = 0 . 32 , P<2 . 2×10−6 . However , there is a trend for higher self-perceived Native American and African ancestry to exceed the genetic estimates ( Figure 3 ) . Similarly , there is a trend for lower self-perceived Native American and European ancestry to underestimate the genetic ancestry ( Figure 3 ) . To explore these trends further we performed a multiple linear regression of the difference between self-perceived and genetically estimated ancestry ( i . e . the bias , see Methods ) , using genetic ancestry and covariates as predictors ( Table 3 ) . As expected , we observe that genetic ancestry has a highly significant effect ( <2×10−16 for all ancestries ) and the negative sign of the regression coefficients reflects the orientation of bias seen in Figure 3 . At increasing European genetic ancestry , there is greater underestimation in self-perception ( a more negative bias ) . By contrast , with increasing African genetic ancestry there is less overestimation ( less positive bias ) . For Native American ancestry , there is an overestimation ( positive bias ) at low levels , and an underestimation at high levels of ancestry ( negative bias ) . Most of the phenotypic traits that show ancestry effects ( Table 2 ) also have a significant effect on self-perception bias ( Table 3 ) . There is a particularly strong effect of pigmentation: individuals with lower skin pigmentation tend to overestimate their European ancestry while individuals with higher pigmentation overestimate their Native American and African ancestries . Similarly , lighter eye and hair color lead to an overestimation of European ancestry and an underestimation of Native American ancestry ( but not African ancestry ) . Hair type is strongly associated with an overestimation of African ancestry . Marginally significant associations are seen with other phenotypes , including facial features such as eye fold ( leading to an underestimation of European ancestry ) and landmark coordinate PCs ( Table 3 ) . An effect of social factors on perception bias is evidenced by the observation that greater wealth is significantly associated with an overestimation of European ancestry and that there is significant variation in bias between countries ( Table 3 ) . We examined the impact on these results of the uncertainty associated with the ancestry estimates by repeating the regression analyses using ancestry estimates obtained with a subset of 15 markers ( Methods ) . We found that the same covariates had significant effects and that the regression coefficients were not significantly different in the two sets of regression analyses .
Since the late 15th century , the population of what is now called “Latin America” has undergone major demographic changes within the context of a highly diversified physical and social environment [6] , [23] . These changes include the occurrence of waves of immigration from various parts of Africa and Europe , the resulting decline of the Native populations most exposed to the immigrants and a variable admixture between these groups . There have also been a number of noticeable population movements in the region . For example , in recent generations there has been an extensive migration to the cities , Latin America now being the most urbanized region of the world ( about 80% of its population is currently considered urban ) [24] . Three of the countries we sampled ( Brazil , Mexico and Colombia ) are the most populous in the region and the combined population of the five countries examined here account for ∼70% of Latin Americans . Although ours is a convenience sample , the individuals studied show considerable variation in birthplace and for a range of biological and social variables , illustrating the extensive heterogeneity of Latin Americans . The interpolated ancestry maps obtained ( Figure 1 ) are consistent with other genetic studies [20] , [21] , [25] , [26] and with census information on the distribution of the main ethnicity groups within each country ( available at www . ine . cl; geoftp . ibge . gov . br; www . igac . gov . co; www . censo2010 . org . mx , www . indepa . gob . pe ) . Altogether , these data underline the extensive genetic structure existing within and between Latin American countries . It is possible to relate this genetic heterogeneity to well documented historical factors [6] , [23] , [27] . Broadly , Native American ancestry is highest in areas that were densely populated in pre-Columbian times ( particularly Meso-America and the Andean highlands ) as well as in regions that received relatively little non-native immigration and which currently have relatively low population densities ( e . g . Amazonia ) . During the colonial period Africans were brought to Latin America as forced labour mainly to coastal tropical areas , particularly in the Caribbean and Brazil [28] . That country was the main recipient of African slaves in the region ( representing about 40% of all African slaves brought to the Americas [29] ) . Early ( mostly male ) Iberian immigrants settled across the continent , admixing extensively with Native Americans and Africans [5] . These were followed by further currents of European immigration , including individuals from various parts of Europe ( often arriving as a result of governmental initiatives ) and resulting in the dense settlement of specific geographic regions ( such as the south of Brazil ) . The larger variance in individual ancestry observed for larger urban centres is consistent with the increasing urbanization of Latin America seen recent generations , the cities absorbing immigrants with diverse genetic backgrounds . Other than demographic history , it is possible that assortative mating has also contributed to shaping population structure across Latin America . The Iberian “Conquest” ( i . e . the first century of settlement ) was characterized by extensive admixture between Natives and immigrants ( driven by the highly predominant immigration of males ) [5] . However , during the subsequent colonial period society became increasingly stratified , including the instauration during the 18th century of a caste system regulating marriages [6] , [27] . These restrictions were mostly abolished with the establishment of republican governments in the 19th century [6] . However , a number of studies have documented continuing assortative mating in Latin America , in relation to genetic ancestry , physical appearance and a range of social factors [30]–[34] . The pattern of variation we observe between physical appearance and genetic ancestry is consistent with information on the variation in frequency of the traits examined in Native Americans , Europeans and Africans . Constitutive skin pigmentation ( i . e . in areas not exposed to light ) , hair and eye color and hair type are traits with little environmental sensitivity and show large differences between continental populations [35] . As expected , increased European ancestry shows a highly significant association with lighter skin , hair and eye pigmentation . A number of allelic variants impacting on these traits have been identified in Europeans and certain of these show large allele frequency differences between Europeans and non-Europeans [1] , [2] , [36] . We also found a highly significant effect of ancestry on hair type , individuals with higher Native American ancestry showing greater frequency of straight hair , a phenotype that is essentially fixed in Native Americans . Recent studies in East Asians implicate a p . Val370Ala substitution in the EDAR gene in hair morphology [37]–[39] . One of the ancestry informative markers typed here ( rs260690 ) is located in the first intron of EDAR , is in high linkage disequilibrium with the p . Val370Ala variant in the HapMap dataset and is strongly associated with hair type in our sample , after accounting for ancestry ( Text S4 ) , suggesting that variants at EDAR could be impacting on hair morphology in Latin Americans . Greater European ancestry also correlates significantly with higher rates of male balding and ( marginally ) with hair greying ( our sample is perhaps underpowered to detect these effects due to its relatively young age; Table 1 ) . Although no thorough comparative data is available , classical population studies indicate that hair greying and androgenetic alopecia are rarer , less severe and of later onset in Native Americans than in other continental populations [40] and our data points to the existence of loci influencing the continental distribution of these traits . Studies in Europeans have recently identified loci associated with androgenetic alopecia [41] , [42] , but no similar analyses have been performed for hair greying . Recent genome-wide association analyses in Europeans have implicated loci for variation in height and related anthropometric traits [43] , [44] . However , these traits are also strongly influenced by environmental factors , including nutrition [45] . In the sample studied here we find that Native American ancestry correlates significantly with lower height and we also detect a significant effect of socioeconomic position ( Text S3 ) , lower socioeconomic position correlating with decreased height . The significant effect of age on height , with younger individuals tending to be taller than older ones suggests that the two socioeconomic indicators examined here ( education and wealth ) capture only part of the environmental variation impacting on height . The rate of increase in height for individuals born more recently ( ∼0 . 1 cm/year ) estimated here is similar to that obtained from extensive longitudinal surveys in Latin America ( ∼1 cm per decade in the last century ) , an observation that has been interpreted as resulting from the historical improvement in living standards across the region [45] , [46] . It is thus possible that the ancestry effect on height that we detect could be influenced by environmental factors that correlate with ancestry that are not captured by the socioeconomic variables examined here . The ancestry effects that we detect for facial features ( eye fold , face shape and size ) , but not for head circumference , agree with the notion of a greater developmental and evolutionary constraint on neuro-cranium than on facial variation . This is also in line with proposals that human facial features include a range of environmental adaptations [47]–[49] . Aspects of face shape variation captured by principal components analysis that are influenced by genetic ancestry include mainly , width and height of the face , facial flatness , position of the glabella and fronto-temporal points , extent of eye fold and the relative size and position of lips and nose ( a fuller description of face shape variation associated with each PC is presented in Text S5 and Figure S3 ) . Two genome-wide association scans in Europeans have identified a few loci associated with aspects of face shape [50] , [51] but these results are pending confirmation by further studies . No genetic variants have yet been implicated in intercontinental differentiation for facial features . Our joint analysis of genetic , phenotypic and self-perception variation emphasizes the strong impact of physical appearance on self-perception . Comparison of skin pigmentation across self-perceived ethno/racial categories shows remarkable consistency between countries , underlining the weight given to this trait in self-perception [52] . The large variation in genetic ancestry between countries for each ethnicity category illustrates the relatively low predictive power of physical appearance for genetic ancestry . Although we detected highly significant effects of ancestry on many of the phenotypes examined , the observed correlations are relatively low ( Table 2 ) . The poor reliability of physical appearance as an indicator of genetic ancestry likely relates to the impact of environmental variation on some of these traits , and to their specific genetic architecture . Particularly , a few genetic variants could have relatively large phenotypic effects ( as documented for pigmentation [2] , [36] ) . The impact of physical appearance on self-perception of ancestry likely relates to admixture in Latin America largely occurring many generations ago and the frequent unavailability of reliable genealogical information . The contrast between self-perceived and genetically estimated admixture proportions confirms the impact of physical appearance on self-perception and shows how certain traits , particularly but not exclusively related to pigmentation , can bias self-perception of ancestry . This biased perception of physical attributes is likely to be influenced by social and individual factors shaping the interpretation of phenotypic variation . The effect of such factors is illustrated by the observation of differences in bias across countries and the positive correlation between wealth and European ancestry ( Table 2 ) . An effect of wealth on self-perception of ancestry has also been the subject of study in the sociological literature on Latin America [52] . In conclusion , our study sample illustrates the extensive geographic variation in genetic ancestry seen across Latin America , reflecting the heterogeneous demographic history of the region . The highly significant impact of genetic ancestry on physical appearance is consistent with some of the phenotypic variation seen in Latin Americans stemming from genetic loci with differentiated allele frequencies between Africans , Europeans and Native Americans [53] . Further analysis of the study sample collected here should enable the identification of such loci . The significant correlation between self-perceived and genetically estimated ancestry is consistent with the observed effects of genetic ancestry on physical appearance . However , self-perception is biased , possibly due to non-biological factors affecting the perception of phenotypic variation and to the genetic architecture of physical appearance traits . Our findings exemplify the informativity of Latin America for studies encompassing genetic , phenotypic and sociodemographic information and the interest of a multidisciplinary approach to human diversity studies .
Recruitment took place mainly in five locations: México City ( México ) , Medellín ( Colombia ) , Lima ( Perú ) , Arica ( Chile ) and Porto Alegre ( Brazil ) . With the exception of Chile , most subjects recruited in these cities were students and staff from the universities participating in this research . In Chile about 2/3 of the subjects recruited were professional soldiers . In Brazil ∼10% of samples were collected in smaller towns of the states of Rio Grande do Sul , Bahia and Rondonia . Adult subjects of both sexes were invited to participate mainly through public lectures and media presentations . Maps showing the number of volunteers in each unique birthplace are presented in Figure S4 . Being a convenience sample , the main collection sites are overrepresented on these maps for each country . We obtained ethics approval from: Escuela Nacional de Antropología e Historia ( México ) , Universidad de Antioquia ( Colombia ) , Universidad Perúana Cayetano Heredia ( Perú ) , Universidad de Tarapacá ( Chile ) , Universidad Federal do Rio Grande do Sul ( Brazil ) and University College London ( UK ) . All participants provided written informed consent . Blood samples were collected by a certified phlebotomist and DNA extracted following standard laboratory procedures . A physical examination of each volunteer was carried out by the local research team using the same protocol and instruments at all recruitment sites . We obtained: height , weight , head , hip and waist circumference , cheilion-cheilion width and sellion-gnation height . All measurements were obtained in duplicate and the mean of the two measurements retained for further analyses . We recorded eye colour into five categories ( 1-blue/grey , 2-honey , 3-green , 4-light brown , 5-dark brown/black ) , and natural hair colour into four categories ( 1-red/reddish , 2-blond , 3-dark blond/light brown or 4-brown/black ) . Balding in males was recorded using a modified Hamilton scale as: 0 ) no hair loss , 1 ) frontal baldness only , 2 ) frontal hair loss with mild vertex baldness , 3 ) frontal hair loss with moderate vertex baldness , and 4 ) frontal hair loss with severe vertex baldness . Similarly , graying was recorded along a five point scale: 0 ) for no greying , 1 ) for predominant non-graying , 2 ) for ∼50% graying , 3 ) for predominant greying and 4 ) for totally white hair . Due to the small number of individuals in categories 1–4 for male pattern balding and greying , we pooled these categories so as to contrast only two categories ( presence or absence of the trait ) . Macroscopic hair type was categorized by visual inspection as 1-straight , 2-wavy , 3-curly or 4-frizzy . A quantitative measure of constitutive skin pigmentation ( the Melanin Index ) was obtained using the DermaSpectrometer DSMEII reflectometer ( Cortex Technology , Hadsund , Denmark ) . Measurements were obtained from both inner arms and the mean of the two readings used in the analyses . Five digital photographs of the face: left side ( −90° ) , left angle ( −45° ) , frontal ( 0° ) , right angle ( 45° ) , right side ( 90° ) were taken from ∼1 . 5 meters at eye level using a Nikon D90 camera fitted with a Nikkor 50 mm fixed focal length lens . The frontal facial photographs were used to score ( by visual inspection ) the presence of an eye fold along the upper eye lids using a three point scale: 0 ) absence 1 ) partial ( interior , middle or outer fold ) and 2 ) full ( along the entire eye lid ) . All photographs were annotated manually with 36 anatomical landmarks and 3D landmark coordinates extracted using the software Photomodeler ( http://www . photomodeler . com/ Eos Systems Inc , Vancouver , Canada ) ( Figure S1 ) . Landmark configurations were superimposed by Generalized Procrustes Analysis [54] and Principal Components ( PCs ) of the 3D landmark coordinates extracted using the software MORPHOJ [54] . To ease visualization of the 3D shape changes associated with each PC we obtained deformation surfaces via a thin plate spline algorithm . A structured questionnaire was applied to each volunteer . We obtained information on two indicators of socioeconomic position ( Text S2 ) . The first indicator is highest education level attained , categorized as: ( 1 ) none/primary/technical , ( 2 ) secondary and ( 3 ) university and post-graduate . The second indicator is a wealth index obtained from a list of items used to assess living standards . These items were: home ownership , number of bathrooms at the place of residence , ownership of household items ( cars , bicycles , fridge/freezer/dishwasher , TVs , radios , CD/DVD players , vacuum cleaner , washing machine ) and availability of domestic service . We used polychoric principal component analysis to examine the variability of each country sample and retained the first principal component as an indicator of wealth . To allow comparisons across countries we converted an individual's wealth score to decile within each country . The questionnaire included items exploring self-perception of ethnicity in the categories: “Black” , “Native” , “White” and “Mixed” , and self-perception of African , European , and Native American ancestry proportions . This was explained as a personal estimation of the proportion of ancestors that had a particular continental origin . We proposed a five point scale , expressed in 20% per cent brackets ( and in words ) : 1 ) 0–20% ( none or very low ) , 2 ) 20–40% ( low ) , 3 ) 40–60% ( moderate ) , 4 ) 60–80% ( high ) and 5 ) 80–100% ( very high or total ) . The questionnaire also recorded information on the place of birth of the volunteer . In order to select 30 markers highly informative for estimating African/European/Native American ancestry , we started from the list of 5 , 000 markers , highly informative for world-wide continental ancestry estimation , identified by Paschou et al ( 2010 ) [18] using the approach of Rosenberg et al . ( 2003 ) [55] based on the worldwide CEPH-HGDP cell panel genotyped with Illumina's Human610-Quad beadchip ( including data for about 600 , 000 SNPs [56] ) . The full list of these 5 , 000 markers is at: http://www . cs . rpi . edu/~drinep/HGDPAIMS/WORLD_5000_INFAIMs . txt . Of these , allele genotype data is available in Native Americans for 3 , 848 markers [57] , of which 2 , 392 have been placed on subsequent Illumina bead-chip products . This subset of markers was retained for selection of those to be typed here so as to facilitate subsequent data comparison and integration . We ranked these 2 , 392 markers based on allele frequency differences in European-Native American or European-African samples . Amongst markers with the highest inter-continental allele-frequency differences we selected those with lowest heterozygosity in Native Americans ( so as to reduce the effect of variable allele frequencies between Native Americans on ancestry estimation ) . Of the final set of 30 markers retained , 13 are monomorphic in 408 Native Americans ( from 47 populations from México Southwards [57] ) , the rest have minor allele frequencies ranging from 0 . 01 to 0 . 15 ( median 0 . 06 ) in that group of populations . The list of markers typed is provided in Table S1 . Genotyping was carried out by LGC genomics ( www . lgcgenomics . com/ ) . In a sample of Colombians recently included in a genome-wide association study that used Illumina's 610 chip [19] , this set of 30 markers produced individual ancestry estimates with correlations of ∼0 . 7 ( for all the three ancestries ) compared with ancestry estimates obtained using an LD-pruned set of 50 , 000 markers from the chip data , and identical mean estimates . We compared the accuracy of these estimates with estimates obtained using markers from the list of 446 proposed by Galanter et al . ( 2012 ) [58] , specifically for studying admixture in Latin Americans . From this list , 152 markers are present on Illumina's 610 chip ( i . e . ∼5 times the number of markers that we used ) and produced estimates with correlations of ∼0 . 85 with the ancestry estimates from the 50 , 000 marker set . By contrast , when the set of markers we selected was reduced to 15 , the resulting ancestry estimates had a correlation of ∼0 . 6 with the 50 , 000 marker set estimates , again showing that there is a diminishing return in accuracy when one increases the numbers of SNPs used in ancestry estimation . Individual African , European and Native American ancestry proportions were estimated using the ADMIXTURE program [17] using supervised runs where African , European and Native American reference groups ( K = 3 ) were provided ( see below ) . Unsupervised runs at K = 3 produced very similar estimates ( Figure S2 ) , confirming our choice of ancestry-informative markers and parental populations . Standard errors of the individual ancestry estimates were obtained by bootstrap using the program's default parameters ( 200 replication runs ) . Data from a total of 876 individuals sampled in putative parental populations were used in ancestry estimation and specified in the supervised ADMIXTURE runs . These were selected from HAPMAP , the CEPH-HDGP cell panel [56] and from published Native American data [57] as follows: 169 Africans ( from 5 populations from Sub-Saharan West Africa ) , 299 Europeans ( from 7 West and South European populations ) and 408 Native Americans ( from 47 populations from México Southwards ) . The full list of the putative parental population samples ( and their sizes ) is provided in Table S2 . The birthplace names of all individuals was consolidated into a list of unique locations organized into three fields: city/municipality , region/state and country . Geographic coordinates ( and altitude ) for each placename were obtained via the Google Maps Geocoding API ( https://developers . google . com/maps/documentation/geocoding/ ) . The GeodesiX software ( http://www . calvert . ch/geodesix/ ) was used for the geocoding query . We used the Global Rural-Urban Mapping Project version 1 data set ( GRUMPv1; http://sedac . ciesin . columbia . edu/data/set/grump-v1-settlement-points ) to attribute census size to these localities ( see Supp . Text S6 ) [59] . We use census data for 1990 , as the median age in our country samples ranges between 20 and 25 . Geographic maps displaying spatial variation in individual admixture were obtained with Kriging interpolation using the software ArcGis 9 . 3 ( http://www . esri . com/software/arcgis ) . The cartographic database was geo-referenced to the SIRGAS geodesic system ( Geocentric Reference System for the Americas , www . ibge . gov . br/home/geociencias/geodesia/sirgasing/index . html ) using a Universal Transverse Mercator projection . Corel-DRAW X3 ( Corel Corporation , Ottawa , Canada ) was used to edit the map images . When a geographic location had multiple data entries ( i . e . volunteers ) , the Kriging interpolation scheme uses the mean ancestry at that location . The correlation between the standard deviation of individual ancestry variation ( at locations with more than 10 samples ) and census size was tested using Spearman's rank correlation ( as population sizes are generally non-normal but rather distributed exponentially ) . Statistical significance was obtained via permutation of individual birthplaces . We tested the null hypothesis of ancestry being spatially uniformly distributed using two approaches . Firstly , we obtained Moran's ‘I’ index for each ancestry component ( African , European , Native American ) separately . This index tests for spatial uniformity of a variable using standard autocorrelation models and we evaluated significance by permuting birthplace locations for every individual maintaining constant the number of individuals sampled per location . To assign a single value to each location we used the average ancestry , recalculating this average after every permutation . Secondly , we used canonical correlation analysis . A disadvantage of Moran's method is that the three ancestry variables are not independent , complicating the interpretation of p-values . Canonical correlation allows one to combine the three ancestries into a single variable: it is the maximal correlation between two sets of linear combinations of multiple variables . In our case , the three ancestries constitute one set and the geographical coordinates ( latitude & longitude ) constitute the second set . Adding quadratic and cubic powers of the geographic coordinates improved the fit , consistent with the curved shape of the ancestry gradients and the existence of regions with markedly different ancestry . Adding a fourth power did not improve the fit any further . P-values were obtained by permutation as above . To evaluate the effect of ancestry on phenotype we used multivariate regression models including basic covariates ( age , sex , country , education , wealth , and optionally BMI and height ) . Depending on the trait we used multiple linear ( for continuous and ordinal traits ) or logistic ( for binary traits ) regression . The categorical traits in Table 2 were considered ordinal variables ( converted into four or five integer levels as specified in Table 1 ) . The justification for doing so is the convention that for an ordinal variable with several categories there is little difference in fitting a linear regression model or an ordered probit model [48] . This is true because for these traits we can assume an underlying continuous variable ( for eye or hair colour it could be the amount of pigment , for hair shape it could be the curvature of hair ) . Since an underlying continuous variable converted into ordered categories is the main assumption for the development of a probit model , this similarity in the two analysis holds . We verified this by examining both models and verifying that the results are similar . Regression results corresponding to the ancestry variables are presented in Table 2 along with R2 from this full model . A baseline regression model with only the covariates was also performed , leaving out ancestry , and the difference in R2 in the two models was taken to be the proportion of variance in the phenotype explained by ancestry . Standard errors of the individual ancestry estimates ( provided by the ADMIXTURE software ) were incorporated in the multivariate regressions via the errors-in-variables model [49] . This adjusts the estimated regression coefficients and p-values for all covariates . The error in estimating a variable generally leads to an underestimation of the regression coefficients . However , the p-value still approaches zero under the alternative hypothesis , provided samples sizes are sufficiently large . For the ancestry estimates , the error in estimation was relatively low ( ∼1–5% ) , consequently for our large sample sizes the reduction in effect size for each variable was modest ( ∼5–10% ) . To evaluate the relationship between self-perceived and genetically estimated ancestry we performed a bias analysis . This bias was defined as self-perception minus the estimated genetic ancestry . Overestimation therefore means that self-perception exceeds the genetic estimate , while underestimation indicates that self-perception is lower than the genetic estimate . Each genetic ancestry estimate was obtained as a percentage ( proportion ) , while self-perception was recorded into five bands at intervals of 20% . The bias in self-perception was therefore considered zero if the percentage of genetic ancestry fell within the chosen self-perception interval . Otherwise , bias was measured to be the distance of the closest boundary of the self-perception interval to the genetic ancestry percentage . We then performed multivariate linear regression of the bias on the genetic ancestry estimates and other variables ( Table 3 ) . The advantage of analysing the bias is that the regression model is easily interpretable . If self-perception was accurate ( bias of zero ) all the regression coefficients would be non-significant . If the bias is non-zero and some variables show significant effects , the signs of the coefficients are interpretable as leading to overestimation ( positive coefficients ) or underestimation ( negative coefficients ) of ancestry , as indicated above . All statistical analyses were performed using R ( www . r-project . org ) [60] or MATLAB [61] . | Latin America has a history of extensive mixing between Native Americans and people arriving from Europe and Africa . As a result , individuals in the region have a highly heterogeneous genetic background and show great variation in physical appearance . Latin America offers an excellent opportunity to examine the genetic basis of the differentiation in physical appearance between Africans , Europeans and Native Americans . The region is also an advantageous setting in which to examine the interplay of genetic , physical and social factors in relation to ethnic/racial self-perception . Here we present the most extensive analysis of genetic ancestry , physical diversity and self-perception of ancestry yet conducted in Latin America . We find significant geographic variation in ancestry across the region , this variation being consistent with demographic history and census information . We show that genetic ancestry impacts many aspects of physical appearance . We observe that self-perception is highly influenced by physical appearance , and that variation in physical appearance biases self-perception of ancestry relative to genetically estimated ancestry . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"biology",
"and",
"life",
"sciences",
"human",
"genetics"
] | 2014 | Admixture in Latin America: Geographic Structure, Phenotypic Diversity and Self-Perception of Ancestry Based on 7,342 Individuals |
The aim of the study is to examine the spatiotemporal pattern of Japanese Encephalitis ( JE ) in mainland China during 2002–2010 . Specific objectives of the study were to quantify the temporal variation in incidence of JE cases , to determine if clustering of JE cases exists , to detect high risk spatiotemporal clusters of JE cases and to provide evidence-based preventive suggestions to relevant stakeholders . Monthly JE cases at the county level in mainland China during 2002–2010 were obtained from the China Information System for Diseases Control and Prevention ( CISDCP ) . For the purpose of the analysis , JE case counts for nine years were aggregated into four temporal periods ( 2002; 2003–2005; 2006; and 2007–2010 ) . Local Indicators of Spatial Association and spatial scan statistics were performed to detect and evaluate local high risk space-time clusters . JE incidence showed a decreasing trend from 2002 to 2005 but peaked in 2006 , then fluctuated over the study period . Spatial cluster analysis detected high value clusters , mainly located in Southwestern China . Similarly , we identified a primary spatiotemporal cluster of JE in Southwestern China between July and August , with the geographical range of JE transmission increasing over the past years . JE in China is geographically clustered and its spatial extent dynamically changed during the last nine years in mainland China . This indicates that risk factors for JE infection are likely to be spatially heterogeneous . The results may assist national and local health authorities in the development/refinement of a better preventive strategy and increase the effectiveness of public health interventions against JE transmission .
Japanese encephalitis ( JE ) is a mosquito-borne disease , which primarily occurs in rural and suburban areas of Southeast Asia and the Western Pacific region [1]–[3] . Japanese encephalitis virus ( JEV ) is transmitted in an enzootic cycle among mosquitoes and vertebrate amplifying hosts , primarily in domestic pigs and ardeid birds naturally [4] . Vectors , primarily Culex tritaeniorhynchus , are abundant in rural areas where their larvae breed in rice paddies and pools of stagnant water . Humans are a dead-end host and get infected when bitten by infected mosquitoes [5] . A recent literature review estimated that annually a total of 68 , 000 JE cases are reported worldwide [6] . JE cases present a wide spectrum of clinical manifestations , which vary from non-specific febrile flu-like illness to severe clinical manifestations including behavioral abnormality , alteration in sensorium and respiratory paralysis [7] . Historically , large human outbreaks of JE were observed in China in the 1960s ( up to 15 , 000 JE cases were reported ) and 1970s ( up to 17 , 000 JE cases were notified ) [8] . The number of JE cases has declined substantially from 20 . 92/100 , 000 in 1971 to 0 . 23/100 , 000 in 2008 since the beginning of a nationwide immunization programme in the early 1970's [9] , [10] . However , JE still remains a significant public health problem in mainland China , with approximately 50% of global cases annually [6] . In recent years , evidence suggests that JEV has expanded its geographic limits within China . For example , JEV has been identified in mosquitoes and pigs in Tibet , where was previously believed to be free of JE because of its altitude [11] . Because of the high mortality and protracted severe sequelae , JE still causes a severe health burden [7] , [12]–[14] . In the past decades , spatiotemporal analysis techniques have been widely used in infectious disease surveillance and outbreak investigation [15]–[17] . It is used to visualize epidemiological data , detect and evaluate hotspots or clusters and improve surveillance and efficient vector control programmes . Studies using spatiotemporal analysis have been widely used in the field of disease mapping such as Sleeping Sickness [18] , dengue fever [19] , [20] , and malaria [21] , [22] , however its application in JE has been minimal [23] . Few studies have explored the spatiotemporal patterns of JE cases in China [24] , [25] . In the absence of specific treatment for the disease and ineffective and unskilled vector control and management of the amplifying hosts in resource-poor regions [26] , interventions need to target vaccination to areas most in need . To inform the efficient targeting of vaccination programmes , it is important to characterize the spatiotemporal pattern of JE cases . Our study was designed to partly address these gaps in knowledge and aimed to describe the nationwide JE epidemic status throughout China , to explore the presence of spatial and seasonal patterns of JE cases , to identify the spatiotemporal clusters of JE cases at the county level and hence to provide evidence-based suggestions for policy-makers and service providers for disease control and prevention .
The study was approved by the Ethics Committee of Beijing Institute of Disease Control and Prevention . In this study , all the data analyzed were anonymized for the confidentiality . Data on monthly JE cases from January 2002 to December 2010 were collected through the China Information System for Diseases Control and Prevention ( CISDCP ) . In this study , all JE cases were confirmed according to the unified diagnostic criteria issued by the Ministry of Health of the People's Republic of China [27] . The case definition for JE consists of an individual who lived in an epidemic area during the vector-biting season or travelled to an epidemic area within 25 days prior to illness onset , showing clinical manifestations such as abrupt onset of fever , headache , vomiting , convulsions or drowsiness or movement and consciousness disorders and with one of the following: JEV-specific IgM antibody in a single sample of cerebrospinal fluid ( CSF ) or serum , JE virus antigens , a 4-fold rise in JE virus-specific antibody , JE virus genome in samples by PCR , or isolated JE virus . Demographic information for each county was collected from the National Bureau of Statistics of China . For the purpose of performing spatial analysis , the county was considered as the spatial unit of analysis and a county-level vector map was acquired from National Administration of Surveying , Mapping and Geoinformation . Local Indicators of Spatial Association ( LISA ) was used to describe the spatial pattern of JE incidence clusters on the county level during the study periods . LISA was used to identify significant hot spots ( High-High ) , cold spots ( Low-Low ) , and spatial outliers ( High-Low and Low-High ) by calculating local Moran's I between a given location and the average of neighboring values in the surrounding locations[23] , [28] , [29] . Significance of clusters was measured by Z score , based on the randomization null hypothesis computation . A high positive Z score indicates that the surroundings have spatial clusters ( High-High: high values spatial clusters or Low-Low: low values spatial clusters ) and a low negative Z score indicates spatial outliers ( High-Low: high values surrounded with low values or Low-High: Low values surrounded with high values ) [30] . To identify the spatial patterns of JE , LISA analysis were performed independently for the annual average incidence of JE on the county level in each period using ArcGIS software ( version 9 . 3 , ESRI , Redlands , CA ) . We also evaluated the dataset for presence of high risk space-time clusters using SaTScan software ( version 9 . 1 . 1 ) , which implements Kulldorff's spatiotemporal scan statistic [31] . Cases files , population files , and coordinate files ( centroids of counties ) were generated in comma delimited format for analysis . We fitted a discrete Poisson model and using a maximum temporal cluster size of 10% of the study period in the temporal window and the maximum spatial cluster size of 5% of the population at risk in the spatial window to identify space-time clusters . The variable-sized elliptic window scanned for clusters with high rates noting the number of observed and expected inside the window . The primary cluster and secondary clusters were detected through the log likelihood ratio ( LLR ) test [32] . Significance of clusters was evaluated using Monte Carlo simulation , generating 999 random simulations to obtain P-values . The null hypothesis of a spatiotemporally random distribution was rejected when the P-value was<0 . 05 .
A total of 48 , 892 cases were reported in 1992 counties during 2002–2010 in mainland China in a total of 2 , 922 counties . Figure 1A describes the monthly distribution of JE cases and its linear trend during 2002–2010 . JE cases had a significant seasonal peak with 80 . 14% cases occurring in July-August ( Figure 1B ) . The annual incidence varied from 0 . 65/100 , 000 in 2006 to 0 . 21/100 , 000 in 2010 . JE showed seasonality with lower peaks: a decreasing trend from 2002 to 2005 , peaking in 2006 and then fluctuating then onwards ( Figure 1 A ) . Considering two peaks each 3 or 4 years during the time span of 9 years , we explore the spatiotemporal pattern of the years with large outbreaks ( 2002 and 2006 ) , the larger peaks from 2003–2005 , and the small number of cases ( 2007 onwards ) . Figure 2 describes the spatial distribution of annual average JE incidence at the county level in China over the study period . JE incidence varied among different counties ranging from 0 to 6 . 41 per 100 , 000 persons . Group A included non-epidemic areas , with 31 . 83% of counties covering 62 . 81% of the total land and 19 . 16% of the total population; Group B represented low-epidemic areas ( with an annual average incidence <0 . 1/100 , 000 ) , including 19 . 30% counties covering 9 . 04% of the total land and 24 . 73% of the total population; Group C included low/moderate-epidemic areas ( with an annual average incidence ranging from 0 . 1–0 . 5/100 , 000 ) , with 29 . 77% of counties covering 15 . 09% of the total land and 34 . 68% of the total population; Group D included high/moderate-epidemic areas ( with an annual average incidence ranging from 0 . 5–1/100 , 000 ) , including 9 . 21% of counties covering 5 . 78% of the total land and 10 . 40% of the total population; and Group E being high-epidemic areas ( annual average incidence >1/100 , 000 ) , including 9 . 89% counties covering 7 . 29% of the total land and 11 . 03% of the total population in China . The at-risk population per county varied from 9 , 649 to 7 , 048 , 255 among the counties with JE cases , while in the non-epidemic areas the at-risk population per county varied from 2 , 123 to 1 , 241 , 857 . LISA analysis of JE epidemics identified foci mainly concentrated in Southwestern China , with an expanding trend to Central China ( Figure 3 ) . The shift of hot spots ( High-High ) and outliers ( High-Low ) can be observed during the four periods ( Table 1 ) . Annual average incidence of High-High cluster had decreased from 3 . 41/100 , 000 in 2002 to 1 . 16/100 , 000 in 2007–2010 . However , the proportion of counties in High-High cluster had increased from 11 . 09% counties in 2002 to 15 . 26% counties in 2007–2010 . The increasing trend of the proportion of potential infection population was also observed in hotspots clusters , with 12 . 39% in 2002 to 16 . 33% in 2007–2010 . High-Low outliers were sporadically distributed in Southeastern China in 2002 with 7 counties , increased to 20 counties in 2003–2005 , then decreased to 13 counties in 2006 , and increased to 21 counties in 2007–2010 ( Figure 3 ) . Figure 4 shows the distribution of annual average JE incidence and spatiotemporal clusters in China . During the four periods under analysis , the primary cluster of JE occurred in Southwestern China , where the geographic extent was 119 , 125 , 133 , and 144 counties respectively . The relative risk ( RR ) of JE infection for population inside the primary cluster compared to those outside the cluster ranged from 38 . 59 in July–August 2007 to 63 . 17 in July 2002 . The highest RR of the primary cluster was identified during July 2002 , including 92 counties in Guizhou Province , 9 counties in Sichuan Province , 5 counties in Yunnan Province ( Figure 4 ) . In addition , the primary cluster was temporally concentrated during July 2002 , 2003–2005 , August 2006 , and July–August 2007–2010 ( Table 2 ) . The identified primary clusters covered 35 . 05% , 77 . 59% , 35 . 81% , and 30 . 17% of cases and accounted for 3 . 88% , 4 . 20% , 4 . 79% , and 4 . 97% of the total population in each clustering time frame respectively ( Table 3 ) . Secondary clusters of JE cases were mainly located in Southern and Central China . The RR of the secondary clusters varied from 7 . 97 to 69 . 02 . The highest RR of secondary clusters was identified in Donggang County , Liaoning Province , where 11 cases were unexpectedly notified in September 2007 . The 2nd RR of secondary clusters was identified during August 2002 , including 46 counties in Sichuan Province , 19 counties in Shaanxi Province , and 7 counties in Gansu Province .
The results of this study indicate that there was a significant spatiotemporal heterogeneity of JE throughout mainland China during 2002–2010 , in that JE reported cases were clustered during the four periods under analysis . Both LISA and spatial scan statistics analysis identified similar clusters , mainly concentrated in Southwestern China and the geographical range of JE transmission increased over the study period . This concentration maybe associated with the distribution of rice-planting areas , the extent of pig rearing and the proportion of the rural population [13] , [33] , [34] . Rice-planting areas are man-made vectors breeding sites that are preferred by JE vectors [35] . The proportion of the rural population is usually associated with levels of poverty and the availability of funds for JE control and prevention [36] . Non-epidemic areas were mainly distributed in the North and Western regions , where the disease was probably not autochthonous , although researchers have isolated JEV from mosquito samples and detected JEV antibody in local population and pigs [37] . Spatial scan statistics , considering both spatial and temporal features , are commonly used in disease surveillance for geographical cluster detection and evaluation , for which they have been shown to have good statistical power [38] . This allows health officials to investigate disease outbreaks whether due to environmental factors , differences in behavioural factors , the transmission of an infectious agent , and the genetic makeup of the population or not [39] . The methods used in the study were previously used in Nepal in a study which showed that the distribution of JE cases had shifted , with clusters found in the central hill areas [23] . Local indicators of spatial association identify patterns in geographic units which deviate from the average of neighboring values in the surrounding locations [40] . LISA , a spatial cluster analysis , has a better chance of detecting true cluster areas with low false-positive rates especially performing well on outlier detection [41] . Elliptic version of the spatial scan statistic rather than commonly used circular version was selected in our study because it has proven to be more appropriate at detecting whether regular shapes or irregular shapes for most of situations [31] , [40]–[42] . Our study identified circular clusters ( e . g . primary clusters in 2002 , 2003–2005 , and 2006 ) and irregular shapes , indicating elliptic scan window should be preferentially used when the shape of the cluster is not known . The results of LISA and spatial scan statistics were consistent which indicates that these methods are reliable and could have wider applications in the fields of disease surveillance and management in China , in particular to the surveillance and monitoring of other vector-borne diseases [19]–[22] . Overall , our results show that JE cases were widespread but relatively concentrated in rural areas in China , especially in Southwestern China . Our findings indicate that preventative strategies for JE , including boosting existing surveillance , financial support from central and provincial governments , infrastructure redevelopment and in-house workforce training should be particularly focused to counties in Southwestern China identified in this study . In addition to contributing to the scientific advancement of JE epidemiology in China , the results from this study also provide important evidence to health authorities , policy-makers and public health practitioners and other service providers to improve JE control . For example , prevention and control measures including service guideline establishment and refinement , disease and vector/host surveillance , immunization program implementation , local vector control , health education and promotion campaigns , community engagement and environmental management , should focus on the high risk areas identified in our study . Targeting of prevention strategies at high-risk clusters is likely to increase the program's effectiveness . Individuals in highest risk areas should be informed of the risk and the possibilities for risk reduction . The results of this study have to be interpreted in light of the studies' limitations: firstly , the data are from a passive surveillance system , which means that some cases of JE might go underreported because of their sub-clinical symptoms [12] , [13] , [43] . Secondly , we used an elliptic scan window in the spatial scan statistics . Although the elliptic scan algorithm searches more zones than a circular scan window , it requires more computational time . In addition , the elliptic scan window has been reported to perform poorly compared to the circular window at detecting for large clusters , as it may select long and narrow string of noncontiguous areas [39] .
This study described the spatiotemporal patterns of JE in mainland China and identified spatial and temporal high risk clusters at the county level over the last 9 years , with important public health implications for targeting JE control in the country . Further studies are needed to explore the role of the physical environment ( e . g . landscape and climate ) and social environment ( e . g . human movement , farming activities , housing conditions and personal health behavior environment ) , population immunity , mosquito control measures in driving the spatiotemporal distribution identified in this study . | Japanese encephalitis ( JE ) is a mosquito-borne disease , which primarily occurs in rural and suburban areas of Southeast Asia and the Western Pacific region . JE still remains a significant public health problem in mainland China , with approximately 50% of global cases annually . Few studies have explored the spatiotemporal patterns of JE cases in China . Here we reported the results of Local Indicators of Spatial Association and spatial scan statistics of JE cases in mainland China at the county level during the four periods: 2002; 2003–2005; 2006; 2007–2010 . The primary spatiotemporal cluster of JE was detected in Southwestern China between July and August , with the geographical range of JE transmission increasing over the past years . The results of LISA and spatial scan statistics were consistent which indicates that these methods are reliable and could have wider applications in the fields of disease surveillance and management in China , particularly in the surveillance and monitoring of other vector-borne diseases . These findings may assist in informing prevention and control strategies and increase the effectiveness of public health interventions against JE transmission . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"disease",
"mapping",
"spatial",
"epidemiology",
"epidemiology"
] | 2013 | Spatiotemporal Patterns of Japanese Encephalitis in China, 2002–2010 |
Dengue virus ( DENV ) infection is a major emerging disease widely distributed throughout the tropical and subtropical regions of the world affecting several millions of people . Despite constants efforts , no specific treatment or effective vaccine is yet available . Here we show a novel design of a DNA immunisation strategy that resulted in the induction of strong antibody responses with high neutralisation titres in mice against all four viral serotypes . The immunogenic molecule is an engineered version of the domain III ( DIII ) of the virus E protein fused to the dimerising CH3 domain of the IgG immunoglobulin H chain . The DIII sequences were also codon-optimised for expression in mammalian cells . While DIII alone is very poorly secreted , the codon-optimised fusion protein is rightly expressed , folded and secreted at high levels , thus inducing strong antibody responses . Mice were immunised using gene-gun technology , an efficient way of intradermal delivery of the plasmid DNA , and the vaccine was able to induce neutralising titres against all serotypes . Additionally , all sera showed reactivity to a recombinant DIII version and the recombinant E protein produced and secreted from mammalian cells in a mono-biotinylated form when tested in a conformational ELISA . Sera were also highly reactive to infective viral particles in a virus-capture ELISA and specific for each serotype as revealed by the low cross-reactive and cross-neutralising activities . The serotype specific sera did not induce antibody dependent enhancement of infection ( ADE ) in non-homologous virus serotypes . A tetravalent immunisation protocol in mice showed induction of neutralising antibodies against all four dengue serotypes as well .
Dengue is a mosquito-borne viral infection caused by dengue virus ( DENV ) , one of the most important human pathogens worldwide [1] . The infection produces a systemic disease with a broad spectrum of outcomes , ranging from non-symptomatic/mild febrile illness ( Dengue Fever , DF ) to severe plasma leakage and haemorrhagic manifestations ( Dengue Haemorrhagic Fever , DHF ) that can further evolve into potentially fatal conditions ( Dengue Shock Syndrome , DSS ) [2 , 3] . DENV , which is spread by Aedes spp . mosquitoes , is widely distributed throughout the tropical and subtropical regions of the world [2] . About 3 billion people , in over 100 countries , are estimated to be at risk of infection , with over 300 million infections , 500 , 000 episodes of DHF manifestations and 20 , 000 deaths reported each year [1 , 4] . The remarkable spread and impact of the disease led the World Health Organization to classify it as the “most important mosquito-borne viral disease in the world” [5] . Four different serotypes of dengue viruses ( DENV1 , DENV2 , DENV3 and DENV4 ) have been identified , all of which are pathogenic in humans [6] . Infection with any one serotype induces lifelong immunity against that specific serotype , with only transient cross-protection against the three other serotypes [7–9] . In fact , severe manifestations of dengue infection are generally associated with secondary infections involving different viral serotypes; this happens through a mechanism known as antibody-dependent enhancement of infection ( ADE ) [8 , 10] . In ADE , recognition of viral particles by cross-reacting , but weakly or non-neutralising antibodies , leads to an increased Fc receptor-mediated uptake of immature or incompletely neutralised viruses by monocytes , macrophages , and dendritic cells , which represent the primary targets of dengue virus infections in humans , resulting in increased infectivity and deterioration of the patient’s clinical condition [11] . This is critical in dengue vaccine development , since an immunogen that does not elicit fully neutralising antibodies to all four serotypes may contribute to disease , rather than prevent infection [12] . Given the lack of efficient treatment against the infection and the risk to human health , in particular ( but not only ) in developing countries , research to develop an efficient vaccine has become an increasing but yet unsuccessful task . DENV is an enveloped virus of the Flaviviridae family , with a single-stranded , positive-sense RNA genome of around 11 Kb that encodes 10 mature viral proteins [13 , 14] . Among these , the envelope glycoprotein ( E ) is the major constituent of the viral membrane envelope , which plays essential roles during the endosomal-mediated virus internalisation , by promoting attachment to the host cell and fusion to the cellular membranes [15–17] . The E protein is formed by three structural domains ( DI , DII and DIII ) , separated by a stem region from the two trans-membrane domains that anchor the protein to the virus membrane envelope [16 , 18] . While DII plays a major role in E dimerisation and harbours the hydrophobic fusion loop , DIII is an Ig-like domain that has been implicated in binding to cellular receptors [16 , 19 , 20] . The mature infective viral particle has a relatively round , smooth surface with a highly ordered icosahedral scaffold formed by 180 E molecules , that are distributed in 30 rafts of 3 parallel dimers organized in a herringbone pattern [19 , 21] . Additionally , E is the most important target of the antibody immune response during infection [8 , 22] . Antibodies against the upper lateral surface of DIII have been described as the ones with high neutralising capacity; coincidently , this region shows also the highest variability among serotypes , which accounts for the high specificity of these antibodies [12 , 23 , 24] . Thus , DIII of protein E has been widely considered as the antigen of choice for vaccine development . Genetic vaccination , based on the delivery into cells of the host of DNA or RNA constructs capable of expressing and secreting the encoded selected protein [25 , 26] , has been extensively tested and found to be highly effective in inducing immune responses against a wide range of pathogens and conditions [27] . It offers the possibility to raise virus-neutralising activity in a simple and effective way due to the very low costs of production and excellent stability , which can be critical when dealing with the conditions found in some developing countries [27 , 28] . In this study we present four DNA constructs , each one encoding the DIII of a different DENV serotype that were engineered to enhance expression and secretion in mammalian cells and tested as genetic vaccines in mice . Our results indicate that they were able to drive strong neutralising responses against all four serotypes , with the potential of further development .
HEK293 and HEK293T/17 cells ( ATCC , Rockville , MD , USA , numbers CRL-1573 and CRL-11268 , respectively ) were cultured in Dulbecco's modified Eagle's medium ( DMEM , Life Technologies , Paisley , UK ) supplemented with 10% heat-inactivated foetal calf serum ( FCS ) ( Life Technologies ) , 50 μg/ml gentamycin and 2 mM L-glutamine . To select HEK293 stable clones , 0 . 4 mg/ml Geneticin ( G418 , Life Technologies ) was added . Mouse myeloma Sp2/0-Ag14 cells ( ATCC CRL-1581 ) were cultured in RPMI 1640 medium ( Life Technologies ) supplemented with 10% heat inactivated FCS , 50 μg/ml gentamycin , 2 mM L-glutamine , 1 mM sodium pyruvate . Sp2/0 stable clones were grown in selective media containing 0 . 4 mg/ml Geneticin . BHK-21 ( ATCC CCL-10 ) and Vero ( ATCC-CCL-81 ) cells were grown in DMEM medium supplemented with 10% heat inactivated FCS , 50 μg/ml gentamycin and 2 mM L-glutamine . Vero FM cells ( Vero E6 derivate , kindly provided by Dr . Toni Rieger , BNI , Hamburg , Germany ) were maintained in the same conditions with 1% non-essential amino acids . Aedes albopictus C6/36 cells ( ATCC CRL-1660 ) were grown at 28°C in RPMI medium supplemented with 10% heat inactivated FCS , 50 μg/ml gentamycin , 2 mM L-glutamine , 1 mM sodium pyruvate and 1% non-essential amino acids . THP-1 human monocytic cells were grown in RPMI 1640 medium ( Life Technologies ) supplemented with 10% heat inactivated FCS . DENV1 Hawaii A strain , DENV2 NGC strain , DENV3 3140/09 isolate and DENV4 TC25 strain ( kindly provided by Dr . Alessandro Marcello , ICGEB , Trieste , Italy ) were used for plaque reduction neutralisation test ( PRNT ) . All DENV strains were propagated in Vero ( DENV3 in Vero FM ) , BHK-21 and C6/36 cells in complete medium containing 2% heat inactivated FCS . Viral neutralisation titres were determined by plaque assay on Vero cells . Unless indicated differently , all cell cultures were grown at 37°C with 5% of CO2 . Sequences coding for the envelope ectodomains were obtained from DENV1 Nauru Island strain ( GenBank accession number U88535 . 1 ) , DENV2 New Guinea C strain ( GenBank accession number AF038403 ) , DENV3 3H87 strain ( GenBank accession number M93130 ) , and DENV4 Dominica strain ( GenBank accession numbers AF326573 . 1 ) . The original and codon optimised DIII sequences of all DENV serotypes were obtained as synthetic fragments in pUC57 vectors from GenScript ( Piscataway , NJ , USA ) . Each DIII sequence was fused to an immunoglobulin leader sequence ( sec ) at the N-terminus [29] and to the SV5 tag ( GKPIPNPLLGLD ) [30] at the C-terminus . DIII-CH3 constructs contained , in addition , the human IgG heavy chain constant domain 3 ( γCH3 ) downstream of the SV5 tag . Both DIII-SV5 and DIII-SV5-γCH3 were cloned into a pcDNA3 vector in which the neomycin resistance gene was deleted and in vector pVAX ( Life Technologies ) . DENV3 sE ( 3sE ) coding region was also obtained as a synthetic gene and cloned in the same vectors . The DIII aminoacidic sequences from all DENV serotypes are shown in S1 Fig . The DIII-εCH4 constructs were obtained by replacing SV5-CH3 with the human εCH4 [31] , followed by either the biotin acceptor peptide ( BAP ) sequence ( GLNDIFEAQKIEWHE ) [32 , 33] to obtain the secretory biotinylated DIII-εCH4 molecule , or a glycosyl-phosphatidylinositol ( GPI ) anchor signal [34] to obtain the membrane-bound DIII-εCH4-GPI . 3sE was also engineered , fused to BAP and roTag [35] at the C-terminus and cloned into a bigenic vector containing the gene for a secretory E . coli biotin ligase [33] . From this construct , 3DI/DII-BAP-roTag was derived after deletion of the DIII domain . An additional DIII-SV5-γCH3 construct for DENV4 TC25 strain , which contains 3 aminoacid changes with respect to the DENV4 Dominica strain DIII sequence ( L357F , N360Y and N384D ) , was obtained by site-directed mutagenesis ( QuikChange XL Site-Directed Mutagenesis Kit , Agilent Technologies , La Jolla , CA , USA ) on the Dominica strain construct following the instructions of the manufacturer . Transient transfections of HEK293T/17 cells and stable transfections of HEK293 cells were performed essentially as described by Sambrook et al . [36] , using circular or linearized plasmids respectively . HEK293T/17 cells were seeded in 6-well plates at 5x105 cells/well density and transfected using standard calcium phosphate method with 0 . 5–5 μg of plasmids . 24 h after transfection the culture medium was replaced with a serum-free medium , when required supplemented with 100 μM biotin ( Sigma-Aldrich , St . Louis , MO , USA ) , and after 24 more hours cellular extracts and supernatants from transiently transfected cells were collected . To remove free biotin , the supernatants of samples containing biotinylated molecules were extensively dialyzed against PBS . Cellular extracts were prepared in 100 μl of TNN lysis buffer ( 100 mM Tris-HCl , pH 8 , 250 mM NaCl , 0 . 5% NP-40 ) at 4°C , supplemented with Protease Inhibitor Cocktail ( Sigma ) according to manufacturer's instructions . The expression and secretion of the recombinant dengue molecules was confirmed by western blot . To produce large amounts of biotinylated antigens , recombinant biotinylated DIII-εCH4-BAP of all four serotypes , 3sE-BAP-roTag , 4sE-BAP-roTag and 3DIDII-BAP-roTag proteins were expressed in stably-transfected HEK293 cells . HEK293 cells were transfected with 15 μg of BglII-linearized DNA using calcium phosphate technique . Stably-transfected clones were screened by ELISA and secretion of biotinylated proteins confirmed by western blot . Supernatants from the selected clones were collected after 72 h of culture in serum-free medium supplemented with biotin and dialyzed against PBS . In experiments involving the use of denatured biotinylated DIII-εCH4-BAP , the dialyzed supernatants were denatured in presence of 0 . 5% SDS ( Sigma-Aldrich ) and 2 . 5% 2-Mercaptoethanol ( Sigma-Aldrich ) and boiled for 10 min . N-Ethylmaleimide ( NEM , Sigma-Aldrich ) was then added and samples were extensively dialyzed against PBS before using . To generate cell lines expressing membrane-bound DIII , Sp2/0 cells were stably-transfected with recombinant DIII-εCH4-GPI proteins . Transfections were performed by electroporation as previously described [37] . Clones were analysed after staining with FITC-labelled anti-human IgE antibodies ( KPL , Gaithersburg , MD , USA , 1:500 in PBS with 3% BSA ) in a FACSCalibur flow cytometer ( BD Biosciences , San Jose , CA , USA ) . Samples of cell lysates and supernatants were separated by 10% SDS-PAGE gels in reducing conditions and then transferred onto polyvinylidene difluoride ( PVDF ) membranes ( Millipore , Temecula , CA , USA ) . Membranes were blocked for 1 h with PBS with 5% Milk and 0 . 1% Tween-20 , probed for 1h with an anti-SV5 monoclonal antibody ( 1:10 , 000 dilution ) and incubated for 1h with HRP-linked goat antibodies anti-mouse IgG , ( Jackson ImmunoResearch , Newmarket , UK , 1:5000 ) . As loading controls mouse mAb anti-tubulin ( clone DM1A , Millipore ) and rabbit antibodies anti-actin ( Sigma ) were used . Signals were visualized by ECL ( ThermoFisher-Pierce , Rockford , IL , USA ) . 5–6 weeks old , female Balb/c mice were purchased from Harlan ( Milan , Italy ) . All mice were immunised three times at fifteen days intervals ( Days 1 , 15 and 30 ) by biolistic delivery of 1 μm gold particles coated with 1 μg of DNA using Gene Gun technology ( Bio-Rad , Hercules , CA , USA ) ; blood samples were collected at days 45 and 60 by sub-mandibular puncture . In the case of the tetravalent formulation , each animal was vaccinated with 2 μg of DNA ( two 1 μg shots applied at different body sites ) following the same vaccination protocol . Serum samples were collected and stored at -20°C until use . Mouse sera were tested on mono-biotinylated DIII-εCH4-BAP and sE-BAP-roTag proteins . The relative concentrations of biotinylated proteins collected from the stably transfected clones were normalized by western blot and comparable amounts of biotinylated protein were used in coating the ELISA plates . Nunc Maxi Sorp Immuno-Plates ( ThermoFisher-Nunc , Roskilde , Denmark ) were pre-coated with 100 μl/well of 5 μg/ml avidin ( Sigma ) in 50mM Na2CO3/NaHCO3 buffer pH 9 . 5 and incubated overnight at 4°C . Plates were washed in PBST buffer ( 0 . 05% Tween 20 in PBS pH 7 . 4 ) , blocked with 1% BSA in PBST for 1 h 30 min . at RT , and second-coated with the dialyzed biotinylated-antigen diluted in PBS ( 100 μl/well ) , at 4°C overnight . After washing , different 100 μl dilutions of sera from immunised mice were added to plates and incubated for 2h at RT . After washing , 100 μl/well of HRP-linked goat antibodies anti-mouse IgG γ-chain ( Jackson ImmunoResearch , 1:50000 ) were added and incubated for 1h at RT . The bound conjugate was detected using TMB substrate ( Sigma ) for 10 min . The reaction was stopped with H2SO4 1M and measured at 450 nm ( OD450 ) on a Bio-Rad iMark microplate reader . The anti-dengue IgG titres were determined as the reciprocal of the dilution at which the OD450 was 3 times higher than that of the negative control serum . Negative control sera obtained from animals immunised with a construct containing an irrelevant protein fused to γCH3 showed the same performance as pre-immune sera or sera from animals immunised with empty vector . The concentration of anti-DIII specific antibodies in sera from vaccinated animals was determined by creating a calibration curve obtained by plotting the OD450 values from an ELISA on biotinylated 3sE with different amounts of a previously quantified and purified sample of anti-dengue envelope mAb 4G2 ( Millipore ) . mAb 4G2 recognises a conformational epitope on the fusion loop of all DENV serotypes [38 , 39]; in our case , the antibody reacts strongly against the 3sE protein with almost 100% avidity , and was therefore used to generate the calibration curve . OD450 resulting from different dilutions of each pool of sera on its homologous DIII were interpolated into this calibration curve to obtain approximate concentrations of specific anti-DIII antibodies in sera . The concentrations are reported as the arithmetic means ± standard deviations of all the dilutions with OD450 included within the calibration curve . Plates were coated with the immunoglobulin fraction from a human serum cross-reactive with all 4 serotypes ( 15 mg/ml in 50mM Na2CO3/NaHCO3 buffer pH 9 . 5 ) ( kindly provided by Dr . Vivian Huerta , Centre for Genetic Engineering and Biotechnology ( CIGB ) , Habana , Cuba ) and incubated overnight at 4°C . Plates were washed , blocked and second-coated with 4x104 PFUs/well of each viral serotype for 2h at RT . After washing , plates were incubated for 1h at 36°C with 100 μl/well of the different anti-DIII sera ( diluted to a concentration of 100 ng/ml ) or negative control sera at an equivalent dilution . mAb dengue 1–11 ( AbD Serotec , Kidlington , U . K . ) , specific for DENV1 envelope protein ( used at 1 μg/ml ) and a DENV panreactive serum ( a kind gift of Dr . Vivian Huerta , Centre for Genetic Engineering and Biotechnology ( CIGB ) , Habana , Cuba ) were used as positive controls . For detection , HRP-conjugated goat anti-mouse IgG γ-chain ( Jackson ImmunoResearch ) was used . Serum avidity index was measured by a modified ELISA protocol with urea washes [40 , 41] . Briefly , the different sera were tested at dilutions corresponding to an OD450 value of 0 . 6–0 . 8 . After incubation with serum , plates were washed two times ( 3 min each ) in PBST , with or without 6M urea , and incubated with secondary antibody as described above . The avidity index was calculated as the ratio between the OD450 obtained after the urea treatment and the OD450 without urea , multiplied by 100 . Serotype-specific immune sera ( diluted 1:1000 in PBS with 3% BSA ) were incubated with the stable Sp2/0 DIII-εCH4-GPI transfectants followed by Alexa488-conjugated goat antibodies anti-mouse IgG ( Jackson ImmunoResearch , 1:1000 ) and analysed in a FACSCalibur ( BD Biosciences ) . Vero cells were infected with DENV1 , DENV2 , DENV3 and DENV4 at multiplicity of infection ( MOI ) of 0 . 1 . 36 h post-infection cells were fixed with 3 . 7% paraformaldehyde ( PFA ) in PBS for 20 min and quenched with 100 mM PBS glycine . After washing with PBS , cells were permeabilized with 1% Triton in PBS for 15 min and blocked with 0 . 1% BSA PBS-Tween 0 . 1% for 1h . Viruses in infected cells were detected using serotype specific mouse anti-DIII sera ( dilution 1:50 ) , mAb 4G2 ( dilution 1:400 ) and mouse control sera followed by Alexa488-conjugated goat anti-mouse IgG ( diluted 1:1000 ) . Images were acquired using a Zeiss ( Goettingen , Germany ) LSM510 META microscope . PRNT was carried out on Vero cells seeded at a density of 160 , 000 cells/ well 24h before infection in 24 multi-well plates . De-complemented mouse sera samples ( 30 min . at 56°C ) were 2 fold serially diluted from 1:12 . 5 to 1:400 in DMEM serum-free medium . Then an equal volume of DMEM-diluted dengue virus containing 50 PFU was added and incubated for 1 . 5h at 36°C in a final volume of 60 μl . Vero cells were then washed with DMEM serum-free media , infected in duplicate with 25 μl of the neutralisation mixture and incubated for 1h at 36°C . Afterwards , the viral inoculum was removed and cells were overlaid with 1 ml of DMEM with 2% FCS and 3% carboxymethylcellulose ( Sigma ) . Plates were incubated at 36°C for 7–8 days depending on serotype ( 7 days for DENV2 and DENV3 , 8 days for DENV1 and DENV4 ) . After this period , cells were washed twice with PBS , fixed for 20 min . with PFA 3 . 7% and stained with 1% crystal violet for 30 min . Plaques were counted and percentage of plaque reduction against control serum was calculated . Neutralising antibody titres were expressed as the serum dilution yielding a 50% plaque number reduction ( PRNT50 ) . The ADE method used was as previously described [42] . Briefly , serial two-fold dilutions of each serotype-specific anti-DIII sera were incubated with the virus for 1 hour at 37°C before added to THP-1 cells at a MOI of 10 . At 72 h after infection , the culture was clarified by centrifugation , and the infectious titer of dengue virus in the culture supernatant was quantified with plaque assay . All data shown were calculated from at least four independent experiments done in duplicate or triplicate . Except for the avidity data ( showed in boxplots ) , all data are represented as arithmetic means ± standard deviations and were analysed using GraphPad Prism ( version 6 . 0 ) software . Unpaired two-tailed t test was used to analyse sets of data between two groups . P values of <0 . 05 were considered significant . All animal procedures were approved by the Italian Ministry of Health ( Ministero della Salute ) and the ICGEB Animal Welfare Board ( protocol DGSAF0024706 ) in compliance to laws and policies established in the legislation D . L . vo 26/2014 of the Italian Government .
Activation of B cells for effective induction of antibody responses in the framework of DNA-based immunisations requires optimal antigen expression and secretion from the cells of the mammalian host [43] . DIII from all four serotypes , initially derived from viral cDNAs ( from amino acid 298 to 416 for DENV1 and DENV2 , 296 to 415 for DENV3 and 298 to 416 for DENV4 ) , were engineered adding a secretion leader peptide ( sec ) at the N-terminus ( to allow translocation into the endoplasmic reticulum ( ER ) and the secretory pathway ) , and with or without the dimerising CH3 domain from the human IgG H-chain ( γCH3 ) at the C-terminus ( Fig 1A ) . To facilitate detection , the 11 aminoacid-long SV5-tag was also included . Expression and secretion of the proteins encoded in the different constructs were then tested in transiently transfected HEK 293 T cells . As shown in Fig 1B , DIIIs from the four serotypes were produced very poorly or not secreted and mostly retained intracellularly . In contrast , when fused to the γCH3 domain enhanced production and active secretion was obtained in all cases . We therefore adopted the DIII-CH3 format as the antigenic design to be used in plasmid-DNA immunisations . However , large differences in the efficiency of DIII-CH3 secretion were observed among the four serotypes ( Fig 1C ) . With the exception of DIII from serotype 1 ( 1DIII ) , which was secreted at an acceptable level ( 1 μg/ml ) , all other serotypes showed much reduced secretion . In order to increase expression , DIII nucleotide sequences corresponding to serotypes 2 , 3 and 4 were codon-optimised for expression in mammalian cells . Improved and comparable secretion levels for all serotypes were thus obtained ( Fig 1D ) . As expected , improved antigen secretion resulted in a higher induction of antibody levels ( S2 Fig ) . Groups of 10 animals for each serotype were immunised by intradermal gene-gun delivery of the DIII-CH3 plasmid DNA construct , with a total of three shots with 1μg of DNA each , at 15 days intervals . Mouse sera were then tested for anti-DIII antibodies in a conformational ELISA ( Fig 2A ) using in vivo mono-biotinylated recombinant proteins secreted from mammalian cells , containing either DIII ( for all four serotypes ) or the full soluble E ectodomain ( sE , E sequence without the C-terminal amino acids from the last half of the stem and anchor regions ) , of the best secreted serotypes 3 and 4 ( 3sE and 4sE , respectively ) . DIII were fused to the CH4 domain of human IgE H-chain ( εCH4 ) , which also supports secretion , but is not cross-reactive with anti-γCH3 antibodies . Both , DIII-εCH4 and sE proteins were fused at the C-terminus to the biotin acceptor peptide ( BAP [32] ) . sE molecules contained at the C-terminus also a tag for detection ( roTag [35] ) ( S3 ( A ) Fig ) . These proteins were expressed and secreted by stably transfected clones ( in HEK293 cells ) co-expressing the biotin ligase sec-BirA , which catalyses the covalent ligation of biotin into the single Lys residue within BAP , as previously reported [33] . The biotinylated proteins were captured on avidin-coated plates ( schematically shown in S3 ( B ) Fig ) . Sera were tested in plates containing comparable amounts of recombinant DIII . Anti-DIII antibodies from all serotype groups were elicited at high titres ( expressed as the dilution producing an optical density at 450nm ( OD450 ) threefold higher than the negative control serum ) , with values ranging from 1:19500 ( for anti-1DIII ) to 1:41300 ( for anti-2DIII ) ( Fig 2B ) , corresponding to a range of antigen-specific antibody concentrations between 16–35 μg/ml , respectively ( Table 1 ) . This was true for sera collected at both time points ( days 45 and 60 ) . These antibodies were also able to recognise DIII in the full-length sE protein in ELISA ( shown in Fig 2A for 3sE and 4sE ) . In addition , immunofluorescence microscopy revealed that all four different serotype-specific sera recognised E protein in virus-infected Vero cells ( Fig 3A ) . Further confirmation of reactivity with protein E was obtained by ELISA on infective viral particles that were captured on plates coated with a serum reacting against the four different viral serotypes . All sera were used at an antibody concentration of 100 ng/ml ( Fig 3B ) . These results indicate that a substantial antibody response was directed against DIII epitopes exposed in the complete envelope ectodomain . The different pools of sera were also able to bind to DIII domains displayed on the cell surface membrane through a glycosyl-phosphatidylinositol ( GPI ) anchor in Sp2/0 stably transfected clones , detected by cytofluorimetry ( S4 Fig ) . These results suggested a conformational nature of the induced antibodies . In fact , all four sera lost most of their reactivity towards the denatured antigen , as compared to the native one , when tested by ELISA with the same amount of coating proteins ( Fig 4A ) . The comparative antibody titres determined on native and denatured antigens are shown in Fig 4B . The avidity index of each serum was also determined performing ELISA in stringent dissociating conditions with 6M urea . All four sera showed an avidity index close or well above 30 , indicating the presence of a substantial concentration of high affinity antibodies [41] ( Fig 4C ) . The avidity index of the control anti-dengue mAb 4G2 was also determined on 3sE and 4sE proteins which resulted in indexes close to 100 in both cases . In order to determine the capacity of each serotype in inducing cross-reactive antibody responses against other serotypes , sera from each group were tested against all serotypes by conformational ELISA on DIII or the 3sE and 4sE ectodomains . As shown in Fig 5A , each pool of serotype-specific anti-DIII showed some degree of cross-reactivity against the different serotypes . As expected , all sera showed the highest reactivity against the homologous antigen . Yet , the cross-reactivity profiles were not the same . While serotype 1 antibodies were mostly cross-reactive with serotype 4 DIII , they reacted much less with serotypes 2 and 3 . Serotype 2 antibodies instead , showed lower cross-reactivity ( the most significant with 4DIII ) . A similar low cross-reactivity was observed for serotype 4 sera . In contrast , serotype 3 antibodies were highly cross-reactive with 4DIII and 2DIII and somehow less with 1DIII . The antibody titres of each pool of serotype-specific sera determined on all four different DIII are plotted in Fig 5B . The data in Table 2 summarises the reactivity of each serum relative to the value of the homologous serum . For instance , anti-3DIII serum showed a titre against 1DIII of around 43% of the corresponding homologous serum , while it showed a higher value for 4DIII ( 94% ) , suggesting that immunisation with 3DIII could significantly contribute to increase the anti-4DIII response . Neutralising activity of the serotype specific sera was tested implementing the plaque reduction neutralisation test ( PRNT ) for each virus serotype in Vero cells . Fig 6A shows a representative set of plates with plaques for DENV2 , not treated or treated with mouse anti-2DIII serum or a mouse negative control serum . Neutralisation titres ( taken as fold dilution producing 50% reduction of plaques , PRNT50 ) of each group of animal sera were first determined against the homologous DENV serotype . All animals within each group showed neutralisation titres ( Fig 6B right panels ) , albeit with some differences . The neutralisation curves for the four different pools ( Fig 6B , left panels ) indicated high titres for serotypes 1 ( 300 ) , 2 ( 1600 ) and 3 ( 300 ) and a lower value for serotype 4 ( 65 ) . Antibody titres are summarised in Fig 6C . Thus , all four DIII-CH3 constructs were efficient in inducing neutralising responses against the homologous DENV serotype . This high efficiency was mostly due to the level of secretion of the DIII domain , as constructs that were not codon-optimised , and therefore produced and secreted at much lower levels , induced lower immune responses ( S2 Fig ) . We then set out to compare the immune responses elicited by four different constructs encoding DIII from serotype 3 in distinct contexts: the preferred codon-optimised 3DIII-CH3 , the same one with non codon-optimised DIII ( 3DIIINOp-CH3 ) , the non codon-optimised DIII alone ( 3DIIINOp ) and the 3sE . The secretory phenotype , tested in transfected HEK293T/17 cells , showed that 3DIII-CH3 was clearly the one secreted at highest levels ( Fig 7A ) . Sera from groups of 5 animals vaccinated with each construct were tested by ELISA , determining anti-DIII antibodies ( plates coated with biotinylated 3DIII-εCH4 ) , anti-E antibodies ( plates coated with biotinylated 3sE ) and anti-DI/DII antibodies ( plates coated with 3DI/DII , an sE version with DIII deleted ) . As shown in Fig 7B , the responses of animals vaccinated with DIII-CH3 were higher than those obtained with DIIINOp-CH3 , DIIINop alone or with sE when tested on DIII ( topmost panel ) and on the sE protein ( middle panel ) . As expected , only sE induced antibodies strongly reacting with DI/DII protein ( bottommost panel ) . Antibody titres are summarised in Fig 7C . As shown in Fig 7D the reactivity of the anti-sE antibodies towards DI/DII was higher than to DIII and comparable to sE ( Fig 7D , insert ) . Conversely , despite its focused reactivity towards DIII , the neutralisation titre ( on DENV3 ) of sera induced with DIII-CH3 was significantly higher than the one obtained with sE , further highlighting the crucial role of DIII as a target for neutralising responses ( Fig 7E ) . Moreover , the avidity indexes of anti-DIII antibodies induced with DIII-CH3 and DIIINOp-CH3 were significantly higher ( 60 ) than the one induced with DIIINOp alone , thus indicating the importance of the CH3 domain ( Fig 7F ) ( 24 ) . To normalize conditions , these assays were all performed with equal amounts of coated proteins . In order to investigate cross-neutralisation activity towards the non-homologous serotypes , each pool of sera was also tested against the other DENV serotypes using the classical PRNT50 in Vero cells . The results obtained are summarised in Table 3 . Similarly to the data obtained on the ELISA cross-reactivity , serotype 1 , serotype 2 and serotype 4 antibodies did not show significant cross-neutralising activity ( <10 for serotypes 2 and 4 and <25 for serotype 1 against all others ) , while serotype 3 antibodies did show significant neutralisation of DENV4 ( titre ≈30 ) and DENV2 ( titre ≈135 ) but not to DENV1 ( titre <25 ) despite being cross-reactive . To further assess the functional specificity of the antibody responses and to determine if the antibody response to our vaccine would cross react with and enhance infection of heterologous DENV serotypes , we also performed an antibody-dependent infection assay . Each serum sample was serially diluted and incubated with each of the four DENV serotypes before inoculating each reaction onto the human monocytic cell line THP-1 that expresses FcγRI and FcγRII [9] . Enhancement of DENV titres was observed only to the homologous but not to the heterologous serotypes of DENV ( Fig 8 ) , indicating functional specificity in antibody binding . As expected , the control serum samples did not enhance DENV titres in THP-1 cells . Based on the responses obtained with each of the different immunogens , we then tested a tetravalent formulation . The tetravalent vaccine contained a mix of the four serotype-specific genetic constructs . Using the same vaccination protocol , the total amount of DNA per dose was increased to 2 μg ( two shots of 1 μg per shot , containing a mix of serotypes 1 and 2 and a mix of serotypes 3 and 4 , respectively ) ; thus each construct was present at 50% of what delivered alone ( 0 . 5 μg ) . After three immunisations , a pool of sera from 5 vaccinated animals was tested by ELISA and PRNT . As for the individual constructs , the tetravalent vaccine was able to induce DIII-specific antibody titres ( Fig 9A left panels ) with high avidity indexes and a balanced neutralising activity ( PRNT ) against all four serotypes ( Fig 9A right panels , and Fig 9C ) . Despite the reduction in antibody concentration and neutralisation titres in comparison with monovalent vaccines ( Fig 9A right panels and Fig 9B ) , this is a first attempt of a tetravalent formulation and represents a proof of principle , with potential for further optimisation .
In this paper we show that immunisation with DNA plasmid constructs encoding properly engineered DIII domains of the four DENV serotypes can induce strong antibody responses in mice . In the context of the immune response against DENV E protein , the Ig-like DIII has been shown to be one of the main targets for protective neutralising antibodies . Highly neutralising epitopes have also been found in regions involving DI , DII [24 , 44 , 45] and , more recently , dimer-dependent epitopes at the interface between the opposing E monomers [46 , 47] were described . However , antibodies reactive with DI/DII were shown to be more cross-reactive , with lower neutralisation potency and consequently implicated in enhanced severity of infection [39 , 48] . Recent studies have shown that in virus infected individuals the antibody response is dominated by highly cross-reactive antibodies , while antibodies directed against the more specific DIII represent only a minor component [24] . Because of the increased risk of ADE due to the presence of such cross-reactive antibodies , several attempts for an efficient anti-dengue vaccine have focused on the use of the highly specific DIII as antigen [49] . In designing our DIII DNA-based vaccine we took into consideration the biochemical properties of the antigen , as it has to be expressed in and secreted from the cells of the host . A good level of protein expression and secretion is required to induce a strong immune response via DNA vaccination [25 , 26 , 43] . A leader signal peptide was fused to the DIII N-terminus to direct its translocation to the ER lumen and the secretory pathway . However , robust secretion was only achieved when DIIIs were fused at the C-terminus to the dimerising γCH3 and further increased upon codon optimisation for expression in mammalian cells of the encoded DIIIs . The CH3 domain played a crucial role in allowing intracellular transport of the recombinant proteins leading to their efficient secretion . This was essential for serotypes 2 and 4 , which were otherwise essentially not secreted , and very important for serotypes 1 and 3 , whose secretion was strongly increased . Interestingly , by fusing to CH3 the quality of the immune response ( in terms of avidity index ) was also improved . We immunised mice with plasmid DNA delivered through gene-gun technology . Our results confirm that this is a very efficient way to induce , with very low amounts of total DNA , highly specific antibody responses , which are mainly directed against conformational epitopes exposed on the infective viral particle . This was indeed shown in experiments where either the virion or the same amounts of native and denatured coated antigen were used to determine sera reactivity . This was also reflected , in part , in the relatively low cross-reactivity of each serotype specific serum . We took special care in the design of the ELISA to test antibody responses . The proteins used were all exclusively produced and secreted from mammalian cells in a mono-biotinylated form that did not require any further purification . Dialysed culture supernatants were used as the source of proteins that were then captured on plates coated with avidin . Thus , antibodies detected in this assay corresponded to those reacting mainly with conformational epitopes on the folded antigen . In fact , the anti-DIII sera elicited with each serotype were able to react also with the whole sE ectodomain ( shown for the DENV3 and DENV4 ) expressed in mammalian cells and with the whole infective viral particle ( for all four serotypes ) . Given the impossibility of comparing ELISA data across dengue vaccine-related studies , we decided to translate our ELISA reactivity data into estimated antibody concentrations , in an attempt to promote the use of measurements and methodologies that allow the establishment of parallelisms between different vaccine candidates . All four serotypes induced high antibody concentrations; in particular serotype 2 was the one producing the highest responses , both in antibody concentrations as well as in neutralisation titres . This was in part due to the contribution of the optimised secretion levels , which otherwise induce very low titres , as well as the site of immunisation . Gene-gun technology delivers DNA intradermally , transfecting mainly keratinocytes that produce and secrete the antigen in an immunologically favourable environment [50] . Availability of antigen , as reflected by the secretion levels , was important . In fact , optimised DIII-CH3 elicited stronger responses than the one with non-codon optimised DIII , which was in turn stronger than the DIII alone . This was also reflected in the neutralisation titres . Additionally , the dimeric structure of the immunogen as a result of its fusion to the γCH3 domain , could also be in part relevant , as it would favour engagement of the B-cell receptor ( BCR ) and subsequent activation of naive B cells . In addition , the xenogeneic nature of the CH3 domain contributes to an increased activation of T helper cells [51] . Thus , when analysing the contribution of the modifications introduced during the design of the antigen , we confirmed that codon optimisation improved antibody titres by increasing antigen secretion , while fusion to CH3 improved immunogenicity and secretion levels of the antigen as well as the avidity index and neutralising capacity of the induced antibodies . It has been shown that neutralisation of virions in flaviviruses follows a “multi-hit” requirement model , in which the number of bound antibodies must surpass a required threshold [10 , 12] . This threshold is different for each epitope and is mainly determined by the combination of two biochemical factors: antibody avidity and accessibility of the epitope on the virus [52] . Antibody avidity and in vitro neutralising activity were shown to positively correlate for anti-Flavivirus monoclonal antibodies [10 , 53–55] , and for antibodies to other viral infections [56 , 57] . Recently , this correlation was also demonstrated in sera of DENV-infected patients [58] . Our results confirm these observations as the DIII-CH3 DNA immunisations induced polyvalent antibody responses in which the neutralising capacity ( determined as PRNT50 titres ) correlates better with the respective avidity indexes than with the sera reactivity measured in ELISA . These data thus support not only the use of specific avidity indexes , but also the idea of introducing these measurements into the evaluation of vaccine candidates , especially for DENV [58] . Several DIII-based dengue vaccines have been reported , using different vaccination strategies including recombinant protein subunit vaccines [59–68] , DNA vaccines [69–71] or viral-vectored live vaccines [72 , 73] . In our case , we have emphasised the design and evaluation of the antigen’s biochemical properties necessary to improve immunogenicity . As shown here , our DIII-CH3 DNA vaccine was able to induce stronger neutralising responses against all four serotypes as compared to other DIII-based vaccines [60 , 69] , even those based on protein immunisation without or with DNA boosting . In addition , the cross-neutralisation profile obtained in the PRNT50 in Vero cells was similar to the cross-reactivity profile obtained by ELISA , suggesting that most of the cross-reacting antibodies were also cross-neutralising . This is in agreement with recent data showing that cross-reactive antibodies contribute to neutralisation during acute DENV infections [58] . In our case , serotype 3 anti-DIII antibodies were the most cross-reactive and showed the highest cross-neutralisation towards the other serotypes . Noteworthy , the ADE assay revealed that each of the DIII-specific sera were not able to enhance infection of heterologous DENV serotypes in the monocytic cell line THP-1 , demonstrating serotype specificity in our DIII-CH3 DNA vaccine . Without antibodies that enhance heterologous DENV serotype infection , ADE would hence only occur when homologous antibodies decay to sub-neutralising levels . For clinical administration of the tetravalent formulation , further optimisation should be performed to ensure that the antibodies are produced at levels that do not result in ADE . Various studies have proposed that genotype differences within each serotype could affect vaccine efficacy [54 , 55 , 74–76] . In a recent phase 2b study conducted in Thailand , failed protection against DENV2 ( 9 . 2% ) was hypothesised to be due to differences in the circulating genotype [77] , while in a more recent phase 3 study conducted in Latin America vaccine efficacy for DENV2 was reported to be higher [78] . Interestingly , the neutralisation titres we obtained for DENV2 were the highest among all four serotypes in the monovalent immunisations and remained high in the tetravalent one . If genotype differences represent an important issue to obtain wide protection against defined serotypes , DNA vaccines are particularly adapted to easily introduce appropriate changes , what represents a significant advantage when compared to other vaccination strategies . Despite the fact that antibodies against DIII have greater neutralising capacity , most of the vaccine candidates for DENV use the whole E protein ectodomain ( with or without PrM ) as an antigen [79] . In this regard , recent data indicate that the antibody response against dengue is dominated by highly cross-reactive antibodies that are mainly focused on antigenic determinants around DII [45 , 48] . Specifically , the main neutralising targets in the response against the E protein , involve epitopes located in or around the fusion loop and the DI/DII hinge region [39 , 44 , 48 , 80] . Considering this , we compared the immune response elicited by our DIII-CH3 construct with that of the whole E ectodomain . Our data proved that the DIII-CH3 construct was able to induce a stronger antibody response and also confirmed that , when using the E ectodomain as antigen , the antibody response was shifted towards DI/DII with a significantly weaker response against DIII . As a consequence , neutralising responses induced by our DIII-CH3 were significantly higher than the ones elicited in animals vaccinated with the E ectodomain ( sE ) . In fact , when compared to other DNA vaccines against dengue that use the complete E protein as their main antigen [81–85] , our DIII-based vaccine , despite delivering significantly lower amounts of plasmid DNA per immunisation , was able to elicit higher neutralising immune responses in mice . Moreover , the DIII-CH3 candidate still shows higher efficiency when compared to other DNA vaccines where the E protein was further modified to enhance immunogenicity [86–91] . The only exception was the DENV4 neutralising response , which was lower than the ones reported by others [82–84 , 88] . As a proof of principle we attempted to formulate a tetravalent immunisation protocol . The results showed an efficient anti-DENV activity elicited against all four virus serotypes , despite the reduced amount of DNA used for each serotype . As with the monovalent formulation , the lowest neutralising activity in the tetravalent formulation was against DENV4 . This appears to be a characteristic of the whole 4DIII domain , as immunisations with 4DIII variants derived from two different genotypes ( strain Dominica from genotype II and strain TC25 from genotype I ) with three amino acid differences ( L357F , N360Y and N384D ) produced similar results and comparable neutralisation titres when tested on strain TC25 ( S5 Fig ) . This indicates that further development of the tetravalent formulation is needed to reach comparable levels of neutralisation activities against all four serotypes . Some aspects to consider include the total amount of DNA , the relative proportion of each plasmid and the sites of administration . In conclusion , we think that at least four key points in our protocol contributed to the high responses observed . i ) First , the genetic nature ( DNA ) of the immunisation , which is capable of inducing long-term humoral and cellular immune responses by effectively surrogating the viral infection process . ii ) Second , engineering of the antigen molecule to make it available to the immune system in significant amounts , since our DIII-CH3 codon-optimised version is actually secreted at much higher levels as a dimeric molecule . iii ) Third , the use of DIII as immunogen instead of the whole E ectodomain , thus reducing the level of cross-reactive non-neutralising antibodies and consequently , the risk of ADE; and iv ) Fourth , the intradermal delivery of the plasmid DNA that elicits a balanced Th1/Th2 response , as opposed to intramuscular delivery which mainly activates the Th1 pathway [92] . As live vaccines candidates against dengue remain disappointing in clinical trials , next-generation vaccines have emerged as new alternatives with the potential to succeed where the classical strategies have failed . Since the first clinical trial for a DNA vaccine against HIV-1 virus in 1998 [93] , several other DNA vaccine candidates developed against infectious diseases have been tested in Phase 1 studies [94–98] . To date there has been only one published dengue DNA vaccine clinical trial involving a Phase 1 study of a plasmid expressing the PrM and E proteins of DENV1 ( D1ME100 , [99] ) . In all cases , the studies showed that the vaccines were well-tolerated and safe in humans [100] , although low immunogenicity remains a main concern associated with DNA vaccines [100 , 101] . Compared to other vaccination strategies , genetic vaccines are safer , more stable , easier to manipulate and have a relatively low production cost . These represent important aspects to consider when designing vaccines for developing countries [100 , 102] . We believe that accurate design of the antigen and the ability to induce the right antibody response avoiding the undesirable non-neutralising cross-reactive ones are key points to develop for a successful DNA vaccine . | Dengue disease is a mosquito-borne viral infection caused by Dengue virus ( DENV ) , one of the most important human pathogens worldwide . DENV infection produces a systemic disease with a broad symptomatic spectrum ranging from mild febrile illness ( Dengue Fever , DF ) to severe haemorrhagic manifestations ( Dengue Haemorrhagic fever and Dengue Shock Syndrome , DHF and DSS respectively ) . To date there is no vaccine available to prevent dengue disease . We show here a strategy of immunisation , tested in mice , that elicits a strong immune response against the four different DENV serotypes . The novelties presented in our work open the way to the development of an efficient vaccine accessible to developing countries . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Dengue E Protein Domain III-Based DNA Immunisation Induces Strong Antibody Responses to All Four Viral Serotypes |
Protein-protein interactions are the cornerstone of numerous biological processes . Although an increasing number of protein complex structures have been determined using experimental methods , relatively fewer studies have been performed to determine the assembly order of complexes . In addition to the insights into the molecular mechanisms of biological function provided by the structure of a complex , knowing the assembly order is important for understanding the process of complex formation . Assembly order is also practically useful for constructing subcomplexes as a step toward solving the entire complex experimentally , designing artificial protein complexes , and developing drugs that interrupt a critical step in the complex assembly . There are several experimental methods for determining the assembly order of complexes; however , these techniques are resource-intensive . Here , we present a computational method that predicts the assembly order of protein complexes by building the complex structure . The method , named Path-LzerD , uses a multimeric protein docking algorithm that assembles a protein complex structure from individual subunit structures and predicts assembly order by observing the simulated assembly process of the complex . Benchmarked on a dataset of complexes with experimental evidence of assembly order , Path-LZerD was successful in predicting the assembly pathway for the majority of the cases . Moreover , when compared with a simple approach that infers the assembly path from the buried surface area of subunits in the native complex , Path-LZerD has the strong advantage that it can be used for cases where the complex structure is not known . The path prediction accuracy decreased when starting from unbound monomers , particularly for larger complexes of five or more subunits , for which only a part of the assembly path was correctly identified . As the first method of its kind , Path-LZerD opens a new area of computational protein structure modeling and will be an indispensable approach for studying protein complexes .
Many biological processes involve protein complexes with multiple subunits . Insights into the molecular mechanisms of the functions of these multimeric complexes can be gleaned from their quaternary structures , which are determined by experimental methods including X-ray crystallography [1] , nuclear magnetic resonance ( NMR ) [2 , 3] , small-angle X-ray scattering ( SAXS ) [4] , and electron microscopy [5] . Computational methods have been also used for modeling protein complexes [6–10] . Although an increasing number of protein complex structures have been revealed , there has been relatively less work conducted to elucidate the mechanisms of protein complex assembly: in particular , the assembly order of protein complexes . Tompa and Rose [11] discussed in the context of interactions of the whole proteome that the assembly of the interactome has an enormous number of combinations , which makes random exploration unrealistic , analogous to the Levinthal paradox [12] . They concluded that there must be hierarchical assembly pathways that make the correct formation of individual complexes possible . Ordered pathways allow efficient assembly and may reduce the possibility of forming incorrect topology . Many protein complexes have evolved to assemble in a defined order as shown in gene fusion events [13 , 14] or conserved gene orders [15] . Some complexes are required to follow an ordered assembly pathway for realizing their biological functions . For example , in ATP synthase , the proton channel forms as the last step which avoids the negative consequences of futile proton transport [16] . Thus , the order of subunit assembly can offer critical clues to the function and evolution of a multimeric complex . From a practical standpoint , knowledge of assembly order is helpful for in vitro reconstitution of multimeric protein complexes . When solving the entire complex is difficult , knowledge of the assembly order can allow reconstruction of a subcomplex which may be easier to solve . Assembly order needs to be taken into account when designing artificial protein complexes [17] . In addition , designing drugs that target protein-protein interactions is of increasing interest [18] , and knowledge of the assembly order is indispensable for creating and evaluating drugs that prevent a critical step in a protein complex assembly pathway [19] . The assembly order of a multimeric complex can be experimentally determined by reconstructing stable intermediates of two or more subunits [20] , which are detected , for example , by gel electrophoresis [21 , 22] or co-immunoprecipitation [23] . Real-time mass spectrometry can identify stable subcomplexes that appear in the assembly time course [24 , 25] . Deletion mutants were constructed to examine if deletions affect complex assembly [26] . Pulse-chase monitored by quantitative mass spectrometry ( PC/QMS ) was applied to investigate assembly pathways for the 30s ribosomal subunit , which detects an assembly order by measuring the ratio of labeled and unlabeled proteins that are added later in the time course [27] . Recently , single-particle electron microscopy was used to determine subcomplex structures stained at different time points of assembly in combination with mass spectrometry [28 , 29] . Alternatively , assuming that assembly and disassembly proceed via the same pathway in opposite directions , electrospray ionization mass spectrometry ( ESI-MS ) can be used to determine disassembly pathways [30] . Previous works by Teichmann and her colleagues used the buried surface area ( BSA ) of each subunit to predict the complex assembly order [13 , 14] . BSA is the difference between the solvent-accessible surface area ( SASA ) of a subunit in the complex and in the isolated state . From thermodynamics principles , the assembly order of a multimeric protein complex is determined probabilistically by the population sizes of various subcomplexes that appear during the assembly process , where the population size of each subcomplex is determined by its binding free energy . BSA was used in their works as a rough approximation of binding free energy , where large buried surface area corresponds to lower ( more favorable ) binding free energy and thus earlier assembly . The BSA method agreed with experimentally determined assembly order in thirteen out of sixteen ( 81 . 3% ) homomeric protein complexes [13] and seven out of nine ( 77 . 8% ) heteromeric cases [14] . However , the BSA method has two primary limitations . BSA has been found to have only moderate correlation to binding free energy [31 , 32] . More fundamentally , the BSA method requires a complete protein complex structure; thus , it cannot be applied to cases where all subunits of the complex have been solved separately but a complete structure is not available . In this work , we used a multiple protein docking method , Multi-LZerD [33] , developed in our group , which can simulate the assembly process of protein complexes to predict the docking order of protein complexes [34] . Multi-LZerD builds structure models of a multimeric protein complex from the structures of its individual subunits . The complex structure model is assembled by combining pairwise docking models of subunits , which are predicted by a pairwise protein docking program , LZerD [35 , 36] . Complex models are refined in many generations of a genetic algorithm , which finally produces about 200 models . The entire Multi-LZerD algorithm of producing multimeric complex structure models somewhat mimics the actual complex assembly procedure; in fact , it was found that the assembly order of complexes can be well predicted by analyzing the assembly pathways of models produced and refined in the Multi-LZerD model building process . The key observation that led to the path prediction is that the binding energy of pairwise docking models assembled to construct a complex indicates the docking order . The method to predict complex assembly order with Multi-LZerD is called Path-LZerD . A strong advantage of Path-LZerD is that , unlike the BSA method , the assembly order can be predicted even when the complex structure is not determined yet , because with Path-LZerD the assembly order of a multimeric complex is predicted by simulating the assembly process of the complex . The binding free energy of subunits was estimated using knowledge-based statistical contact potentials , which can evaluate the energy more accurately than simply considering BSA and are successful in protein-protein docking [37] . Interestingly , in many cases the assembly order was correctly predicted even when the predicted structure models from Multi-LZerD were not entirely correct . Using 21 protein complexes with between three and seven subunits , the complex structure and/or topology was well predicted in nine cases and the assembly order was correctly predicted for ten cases . When homology models or unbound structures were used for building complex structures , the assembly order was similarly well predicted for small complexes of 3 and 4 subunits; however , the predictions deteriorated for larger complexes . The ability to predict assembly order from the structures of the subunits can offer additional insights into the biological function of multimeric protein complexes .
The dataset of multimeric protein complexes includes 21 complexes of 3-7 chains , which have evidence of their complex assembly order ( Table 1 ) . The set was manually collected from literature and from the Protein Data Bank ( PDB ) [38] . The assembly pathway of each protein complex is listed in Table 1 . For example , the assembly order of BG> BGP for 1a0r indicates that chain B and G form a subcomplex first , to which chain P docks to construct the complex . A chain ID with superscript prime ( ′ ) or a number , for example , B′ or B4 , indicates that it has the same sequence as chain B . The evidence for assembly pathways is classified into four categories shown in the last column of the table: experimental evidence ( E ) , biological inference ( B ) , structural inference ( S ) , and model of assembly ( M ) . “Experimental evidence” includes co-immunoprecipitation of subcomplexes and ESI-MS . “Biological inference” indicates that the order of the assembly can be reasonably inferred from the function of each subunit . For example , if the complex is between a protein dimer and its inhibitor , the dimer is expected to form first and the inhibitor to bind later . Another type of evidence , “structural inference , ” is from structural information of a complex , which indicates that features of the structure , such as which chains are in contact , restrict the possible assembly orders . “Model of assembly” indicates that the assembly pathway has been proposed in a publication . A detailed explanation of the evidence for the assembly order of each complex is provided in S1 Appendix . The assembly order prediction was performed for bound and unbound/computationally modeled cases of this set of proteins . The assembly pathway prediction by Path-LZerD uses the Multi-LZerD [33] algorithm at the core of its protocol . Multi-LZerD predicts the structure of a multimeric protein complex from the structures of the subunits of the complex ( Fig 1 ) . In the first step , all pairwise combinations of subunits are docked using a pairwise protein docking method , LZerD [6 , 35 , 39 , 40] . LZerD represents protein surface shape using 3D Zernike descriptors ( 3DZD ) [39 , 41 , 42] , which are based on a mathematical series expansion of a 3D function ( in this case , protein surface shape ) . The 3DZD are a soft representation of the surface shape , conferring tolerance to the conformational changes associated with binding . Typically , over 100 , 000 docking models ( decoys ) are generated for a pair of protein structures and the 54 , 000 decoys with the best shape complementarity score ( described below ) are kept . The decoys are clustered to reduce redundancy with a cutoff of 10 Å , which usually yields between 2 , 000 and 6 , 000 decoys . Next , models of the entire complex are built by combining pairwise docking decoys and the models are refined using a genetic algorithm ( GA ) , a combinatorial search algorithm . Multi-LZerD represents a multimeric protein complex as a spanning tree ( i . e . a connected graph with no cycles ) , where nodes are proteins and edges are pairwise docking decoys . The initial population of M ( set to 200 ) complex models are generated by random combinations of pairwise decoys . Then complex models in the population undergo iterative modification to search for better models that have better fitness scores . Complexes are modified using a GA operation called mutation , where a model has one random edge ( pairwise docking decoy ) removed and one random edge replaced ( Fig 1 ) . The modified complexes are assessed for atomic clashes ( atom pairs closer than 3 Å ) and discarded if the number exceeds a threshold ( 200 ) . Clustering of the complexes was also performed to reduce redundancy . If clustering decreased the number of complexes below the initial population size M , complexes were added back at random to fill the population ( except for the final population , which is allowed to have fewer than M complexes ) . Complex fitness was evaluated using a molecular mechanics scoring function ( described below ) and the population was reduced to the M complexes with the best fitness score . This procedure of exploring model conformations with better fitness scores is called a generation . 400 mutations were performed in each generation . Each complex was run for 2000 generations and evaluated for convergence based on the fitness score . If the complex had not converged , the GA was run for an additional 1000 generations . The previous paper showed that most of the cases converged within 1000 generations [33 , 43] . The overall procedure mimics a population of protein assembly process . Finally , 200 or fewer models were generated . The algorithm of Multi-LZerD and parameters used were not modified from its original work [33] . Refer to the original paper for more details . The ranks of the pairwise decoys that comprise a Multi-LZerD complex model were used to predict assembly order ( described in detail below ) . The pairwise decoys for each subunit pair were ranked by a scoring function , which evaluates the binding energy of the decoys . Here we briefly describe the eight scoring functions used in this work . Multi-LZerD is originally equipped with two scores , a shape-based score and a molecular mechanics-based function . In addition , we benchmarked six knowledge-based statistical scores , DFIRE [44] , Dligand [37] , ITScorePro [45] , GOAP [46] , OPUS-PSP [47] , and SOAP-PP [48] . These statistical scoring functions have been very successful in various problems in protein structure prediction , such as single protein structure prediction , protein-protein docking , and model quality assessment . The general approach of constructing a knowledge-based statistical scoring function is to use the observed distribution of some feature ( e . g . atom pair distance or angles ) in a set of known protein structures and normalize the distribution by a reference state . Scoring functions typically differ in the choice of features and the reference state considered . We predict the assembly order of a complex by comparing the ranks of the pairwise decoys that were assembled by Multi-LZerD to obtain the whole complex model . For example , if a model of an A-B-C complex is made up of the A-B decoy with rank 1 and the B-C decoy with rank 125 , where pairwise decoys are ranked by a scoring function , the complex is predicted to assemble A-B first , followed by AB bound with C ( denoted as AB> ABC ) . Using the ranks of pairwise decoys is rationalized by the thermodynamic consideration that the population of a decoy is determined by its binding free energy . The free energy of a binding pose i between two subunits , A and B , is defined as the difference between the free energy of the complex and the free energy of the subunits: Δ G b i n d A B , i = G A B i - ( G A + G B ) ( 1 ) where G A B i is the free energy of the ith binding pose of the AB complex and GA and GB are the free energies of subunits A and B , respectively . Thus , the probability that A and B take the binding pose i is p A B i = e - Δ G b i n d A B , i / k T ∑ n e - Δ G b i n d A B , n / k T ( 2 ) where k is the Boltzmann constant , T is the temperature , and n is the index of the binding poses . The probability of a binding pose j of B and C , p B C j , is computed in the same way . For a complex with three subunits , ABC , assuming the subunits have equal concentrations , binding pose i of AB is more populated than binding pose j of BC if p A B i > p B C j and the more populated binding pose will statistically assemble first . Moreover , if we assume that different pairwise complexes have normalization factors ( denominator of Eq 2 ) of the same order and similar energy distributions of binding poses , the ordering of the probabilities , p A B i > p B C j , follows from the ordering of the ranks , rank ( Si ) < rank ( Sj ) . Here , Si is the score that estimates the binding free energy for binding pose i of AB and rank ( Si ) is the rank of the score , where the lowest ( i . e . best ) score has rank 1 . Since the assembly order is predicted by considering pairwise decoys , the binding free energy of a subunit to a subcomplex is approximated by a pairwise interaction , i . e . Δ G b i n d A B : C ≈ Δ G b i n d B C or Δ G b i n d A C , where Δ G b i n d A B : C is the binding free energy of the AB subcomplex binding with the C subunit . The binding free energies of pairwise decoys are estimated by the scoring functions introduced above , which have been successfully used for protein structure prediction and docking [36 , 37] . The thermodynamic rationale of the assembly order prediction is made under reasonable assumptions in the same spirit as protein structure prediction . As we demonstrate later , the algorithm shows successful prediction in many cases . The next choice to make in the prediction procedure is which complex models built by Multi-LZerD to use for the binding energy rank comparison . We used two methods that require knowledge of the native structure ( non-blind ) and two methods that do not ( blind ) . Both non-blind methods use the metric of root mean square deviation ( RMSD ) to the native complex structure . We used and compared the following four methods , which use different complex models: Low RMSD decoy combination method: In this method , low RMSD pairwise decoys from LZerD are combined to form a low RMSD complex structure . To generate the model , for each pair of subunits , five decoys with the lowest RMSD are selected . The selected pairs are exhaustively combined to create fully assembled complexes . To assemble complexes , concretely , first we examined the native complex and recorded each contacting pair of subunits using the binding interfaces shown in PISA [49] and visual inspection . These interfaces were treated as edges and all possible spanning trees using those edges were constructed . For each spanning tree , the five lowest RMSD pairwise decoys for each edge were exhaustively combined . The lowest RMSD model out of all combinations of pairwise decoys was selected . Lowest RMSD method: From the final GA generation of Multi-LZerD ( 200 or fewer models ) , the model with the lowest RMSD to the native structure was selected . Final generation method: This method belongs to the blind strategy , which does not use the tertiary structure of the target protein complex . All models from the final generation of Multi-LZerD were used to predict the assembly pathway . Each model was given one vote , the assembly pathways were tallied , and the most frequently occurring assembly pathway was predicted . Thus , unlike the first two methods , this method uses many models in the final generation of the GA and does not refer to the native structure . For example , for a complex of three chains , A , B , and C , if the final generation has 200 models , among which 160 models indicate an assembly order of AC> ABC based on their score rank of pairwise decoys , 30 models indicate an AB> ABC order , and the rest indicate BC> ABC , the resulting prediction is AC> ABC because it has the majority of votes . Consensus across generations method: This method belongs to the blind strategy . This method is equivalent to the final generation method , except that votes are tallied from the generation 1000 through the final generation . Since each generation produces up to 200 models , the total number of votes will be up to 200 , 000 . Low RMSD decoy combination and lowest RMSD are non-blind strategies which require knowledge of the native complex structure to compute RMSD . In contrast , final generation and consensus across generations are blind strategies which do not require the native complex structure . Given a model of the complete complex selected by the methods above , the pairwise decoys that make up the complete complex were noted and the ranks and Z-scores of their binding scores were compared . The pairwise decoys were sorted by the score rank and ties were resolved using the Z-score . Assembly was predicted to begin with the pair with the lowest score rank and proceeded in ascending order of score rank . In addition to each individual score , the score ranks were summed to form an additional score . The whole procedure of predicting assembly order ( path ) using Multi-LZerD models is named Path-LZerD . In order to compare with the four methods in Path-LZerD , we also predicted assembly order using buried surface area ( BSA ) in two ways , based on previous work [13 , 14] . For all cases , solvent-accessible surface area ( SASA ) was computed using Naccess [50] considering all atoms in the crystal structure . In the first BSA approach , the pairwise buried surface area is computed for each pair of contacting chains: BSAAB = SASAA + SASAB − SASAAB , where SASAA and SASAB are the SASAs of chains A and B alone and SASAAB is the SASA of the pairwise complex . The pairs were sorted in descending order of BSA and assembled into a spanning tree , where each node is a protein chain and each edge is a contacting pair . N − 1 edges were added where N is the number of subunits , but an edge was skipped if it forms a cycle . In the second BSA approach , instead of computing BSA for pairs of subunits , SASA was computed for all possible subcomplexes ( e . g . combinations of [2 . . N − 1] chains that form a connected subgraph ) . Then , BSA was computed for every possible transition between subcomplexes as the difference between the SASA of the component subcomplexes and the SASA of the new subcomplex , e . g . if A combines with BC , BSAA:BC = SASAA + SASABC − SASAABC . The complex was then assembled from its components in descending order of BSA . The former approach will be referred to as pairwise BSA while the latter will be referred to as subcomplex BSA .
The overall assembly order prediction results on bound docking cases using Path-LZerD and BSA are summarized in Table 2 . In the table , target protein complexes are classified by the number of chains in the complex . On the left , the lowest RMSD of the complex structure models in the final GA generation of Multi-LZerD is shown . Then , the assembly order prediction results are shown for three non-blind strategies: subcomplex BSA , low RMSD decoy combination , and lowest RMSD . The non-blind methods need the native structure of the complex to predict the assembly order . The right columns show the results of two blind strategies , final generation and consensus across generations , which do not need the native structure of the target . Multi-LZerD successfully predicted the structure for many of the complexes but was not able to model the larger 6 and 7 subunit complexes within a small RMSD to the native ( Table 2 ) . A model with an RMSD under 2 . 0 Å was constructed for 5 cases: 1a0r , 1vcb , 2aze , 1es7 , and 1gpq . Multi-LZerD usually builds at least a subcomplex structure correctly , even when the overall complex was not correctly assembled [51] . In parentheses is the number of subunits that are assembled within an RMSD of 4 . 0 Å . Particularly , in two cases , all but one subunit is well predicted ( 2e9x and 1w88; Fig 2 ) . For the four-chain complex 2e9x , a three-chain subcomplex was assembled with RMSD 1 . 6 Å and for the five-chain complex 1w88 , a four-chain subcomplex was assembled with RMSD 1 . 3 Å . In another two cases ( 1ikn and 1hez ) , the topology was correct or almost correct ( Fig 3 ) . The diagram next to each complex illustrates interactions between subunits . Pairwise interfaces are defined as having at least one contacting residue pair ( at least one pair of atoms is closer than 5 . 0 Å . ) A solid line in the diagram indicates that there are more than 20 interacting residue pairs between subunits while a dotted line indicates that there are fewer than 20 interacting residue pairs . The lowest RMSD model of 1ikn has the correct topology . The model of 1hez contains all the native interactions with an extra interaction between chain B and E . These nine cases where Multi-LZerD was correct or mostly correct will be referred to as the well-predicted target subset . This subset was also separately analyzed to investigate correlation between the assembly order prediction accuracy and the complex structure prediction accuracy . Now we turn our attention to the assembly order prediction . A prediction for a target complex is evaluated by the number of correctly predicted assembly steps ( X ) over the total number of steps ( Y ) denoted as X/Y . Each step corresponds to a correct subcomplex; for example , if the correct assembly pathway is AB> ABC> ABCD , the predicted pathway AC> ABC> ABCD has a score 1/2 because the second subcomplex is correct . The known assembly steps of each target are shown in Table 1 and a description of the evidence is in S1 Appendix . For each Path-LZerD strategy , the results are shown for the best single score ( e . g . shape , OPUS-PSP , GOAP ) and for the sum of score ranks . Results for each individual score are shown in supplementary tables , from S2 to S5 Tables . As for prediction using BSA , results for the subcomplex BSA method are shown in Table 2 and the results of the pairwise BSA method are provided in S1 Table . Overall , the subcomplex BSA method made the largest number of correct predictions when the number of correct full assembly orders was concerned . It correctly predicted the assembly pathway for 13/21 ( 61 . 9% ) complexes . This is understandable because it is the only method that directly uses the interfaces shown in the native complex structure ( note that the two non-blind Path-LZerD methods , the low RMSD and the lowest RMSD methods , refer to the native structure but it is only to identify the lowest RMSD models , which may have substantially different from the native structure ) . The second was the lowest RMSD decoy and the lowest RMSD strategy by Path-LZerD predicting eleven cases correctly followed by the final generation and the consensus strategy with ten correct predictions . On the other hand , when the number of partially correctly predicted assembly orders were counted , interestingly , the lowest RMSD method with DFIRE ( S3 Table ) and the final generation and consensus methods with the molecular mechanics score ( S4 and S5 Tables ) have the best performance with 19/21 ( 90 . 5% ) . This is interesting because the final generation and the consensus methods do not use the native structure to select complex models but still achieved the best performance . In a close look ( S4 and S5 Tables ) , these two Path-LzerD blind strategies made at least partially correct predictions for all but two targets , while the BSA method made three completely wrong predictions . The two blind strategies made partially correct predictions for all the targets with five to seven chains , even though the structure models have large RMSD values . Among the Path-LZerD strategies , the best performance was observed for the low RMSD decoy combination strategy using the shape score ( S2 Table ) and the lowest RMSD strategy using OPUS-PSP ( S3 Table ) when perfect prediction was considered ( 11/21 , 52 . 4% ) . If partial correct predictions were counted , the top performing methods were the final generation ( S4 Table ) and the consensus strategy using molecular mechanics score ( S5 Table ) ( 19/21 , 90 . 5% ) . While the low RMSD decoy combination method has the highest number of perfectly predicted pathways ( 11/21 using the shape score ) , it also has the lowest ( 4/21 using GOAP ) . The blind methods using Multi-LZerD had a smaller range of numbers: 8-10 perfectly predicted pathways . This suggests that simply choosing pairwise decoys based on RMSD can be either very effective or very ineffective , and that the pairwise decoys that survive the Multi-LZerD genetic algorithm more consistently predict assembly orders probably taking advantage of the voting strategy . Comparing the BSA method to the Path-LZerD strategies , the assembly pathway of five complexes not predicted by the BSA method were rescued by some of the Path-LZerD strategies: 1ikn , 2e9x , 1kf6 , 4hi0 , and 4igc . There was only one case where BSA was successful and no LZerD strategy made a perfect prediction: 3vyt . Finally , the assembly pathway of two complexes had no perfect predictions by any method: 3uku and 4gwp . The success rate of the assembly order prediction of the well-predicted subset of nine complexes ( PDB IDs in bold ) was higher than for the entire dataset . If only the well-predicted subset of nine complexes is considered , the lowest RMSD strategy with sum of score ranks is more successful than the BSA method with eight out of nine correct predictions . For 4hi0 and 4igc , the BSA method failed to predict their assembly order while Multi-LZerD made successful prediction although the complex structure was not well predicted . The last two rows of Table 2 evaluate prediction accuracy in a different way by counting the number of correctly predicted subcomplexes that appear during the assembly process . Out of 58 total subcomplexes in all the targets , the BSA method identified 42 ( 72 . 4% ) . The second was the consensus method with GOAP with 37 subcomplexes correctly identified . Interestingly , when methods among Path-LZerD are considered , blind strategies ( i . e . the final generation and the consensus method ) perform better than the non-blind strategies , having 35 and 37 subcomplex hits . Also , when subcomplexes in the nine target subsets are considered ( the last row ) , Path-LZerD obtained 15 correct subunits , which was better than BSA ( 14 subunits ) . When the results by the final generation method were examined ( S4 Table ) , which involves a voting step by structure models generated in the final generation of Multi-LZerD , it seems that a higher number of votes correlates weakly with assembly path prediction accuracy ( S6 Table ) . For cases with more than 75% of the votes ( e . g . 150/200 votes ) , on average 85 . 4% of the assembly steps were correctly predicted , while for cases with fewer than 75% of the votes , the average accuracy dropped to 53 . 1% . We also examined which steps were better predicted in Fig 4 . From the plots , the earlier assembly steps , particularly the first step , seem to be better predicted . For the five-chain targets ( 1hez and 1w88 ) , five methods predicted the first step of both targets correctly , but the second and the third step were predicted for both targets by only two methods . The tendency is clearer for the six-chain targets since there are more assembly steps and targets in this class . The first step of all the targets were correctly predicted by five methods while subsequent steps were less well predicted . For the seven-chain targets , the first step of one out of two targets were correctly predicted by seven methods . We also predicted the assembly path for unbound cases where individual subunit structures are determined in an isolated condition and cases where subunit structures were computationally modeled . Modeller [52] was used for modeling individual structures from structures of homologous proteins . Homologous proteins to the 21 multimeric protein complexes in Table 1 were searched using BLAST ( BLASTP 2 . 2 . 31+ ) runs against protein sequences from the PDB obtained from the Modeller website ( https://salilab . org/modeller/supplemental . html ) . The E-value cutoff used was 0 . 01 . The search found unbound structures for three complexes ( 1es7 , 1rlb , and 3vyt ) and template structures for modeling for eight complexes . The results are summarized in Table 3 . Only the blind strategies , the final generation method and the consensus across generation method , were used . In the sequence identity ( Seq . Id . ) column , U indicates that the prediction was made with a complex built from unbound structures while the rest used homology models . For all but three targets , templates with different sequence identity levels were used to see how the quality of models influences assembly path prediction . Detailed information about the individual unbound structures and homology models is provided in S7 Table . Compared with the prediction results in Table 2 , the path predictions showed little to no deterioration for 1a0r , 1es7 , and 3fh6 , three complexes with up to four chains . For larger complexes with five chains or more , the number of correctly identified subunits decreased , but for most of the cases and still identified a part of the assembly steps correctly . For 1a0r and 1es7 , the quality ( i . e . RMSD ) of the predicted complex structure was significantly worse than the bound cases in Table 2 , but interestingly , the assembly path predictions remained almost perfect . For the 1a0r case , a close look at the assembly paths of individual docking models in the final generation found that the successful prediction was possible because the topology of the structure models were correct despite their large RMSD . For the 1es7 case , the first step of the assembly process , the interaction between chain A and C , was well identified due to their large interface area , which led to the correct path prediction . Thus , for these cases , similar to what was observed in Table 2 , the assembly path were correctly predicted even in cases that complex structure itself was not well predicted . Path prediction is often not very sensitive to the quality of individual structure or complex structure models , because the underlying docking simulation often captures affinity of subunits that appears from more coarse-grained features of subunit structures . On the other hand , we also observed in Table 3 that modeled structure cases were substantially worse than the bound cases . Prediction for 1w88 identified one correct subcomplex in the bound case ( Table 2 ) , which decreased to 0 in six out of eight results shown in Table 3 . A close examination of the unbound predictions found that the RMSD of the pairwise decoys was worse than the bound cases for 1w88: in the bound case , the average of the best RMSD for 10 pairwise decoys was 1 . 53 Å , while it worsened to 4 . 39 Å and 4 . 71 Å for the two modeled cases ( sequence identity ranges of 40 . 8-50 . 9% and 40 . 8-48 . 7% , respectively ) . Similar situations were observed for 1rlb and 3uku where path prediction for modeled cases did not identify any correct subcomplexes by some strategies . The average best RMSD of pairwise decoys of the 1rlb bound case was 5 . 96 Å while it was 10 . 87 Å for the modeled case . For 3uku , pairwise decoys of the bound case had the average best RMSD of 4 . 41 Å but it deteriorated to 6 . 23 Å and 6 . 58 Å in the two sets of modeled cases . Thus , for these cases , the inaccuracy of the individual models negatively affected the quality of the pairwise decoys . This is one of the fundamental problems of current pairwise protein docking prediction [36]—for improvement , the core pairwise docking algorithm , in this case LZerD , needs to be improved in order to achieve better unbound docking performance . We will discuss in detail several complexes with notable results in Table 2 . The first complex is 1gpq , which is the complex structure of inhibitor of vertebrate lysozyme ( Ivy ) from E . coli bound to hen egg white lysozyme C . Ivy forms a homodimer ( denoted as A and A′ in Fig 5 ) and binds to lysozyme C ( denoted as C and C′ ) . Since Ivy is functional as homodimer , it needs to be formed first . Thus , the known assembly order is AA′> AA′C> AA′CC′ ( Fig 5 ) . The structure is predicted correctly at an RMSD of 1 . 74 Å by Multi-LZerD . The assembly order is predicted perfectly by the BSA method and all Path-LZerD methods . Thus , this is an example where all predictions were correct . On the other hand , there is one complex that was perfectly predicted by the BSA method but not by Path-LZerD: 3vyt , which is a hexamer comprised of two HypCD heterodimers bound to a central HypE homodimer . HypC ( denoted as C and C′ in Fig 6 ) , HypD ( D and D′ ) , and HypE ( E and E′ ) are proteins required for the maturation of [NiFe] hydrogenase , which is involved in microbial hydrogen metabolism [53] . Since this complex is an assembly of two HypCD dimers and a HypE homodimer ( S1 Appendix Fig 6 ) , the correct assembly order is CD+C′D′+EE′> CD+C′D′EE′> CC′DD′EE′ . A problem for this target was that the complex structure was not modeled correctly . The best model had an RMSD of 36 . 8 Å and no pair of subunits is predicted with RMSD <4 . 0 Å . Partly due to the incorrect structure model , only part of the order was correctly predicted . For example , the final generation/consensus strategy with the GOAP and the rank sum score correctly predicted the HypE homodimer and one copy of the HypCD heterodimer; however , for higher order subcomplexes , errors emerged . The pathways from the final generation method with the GOAP scoring function are shown in S1 Fig . GOAP gave votes in the final generation to the correct pathway; however , it was a minority ( seven ) . As shown in the diagram , the majority of the pathways correctly identified the EE′ and CD subcomplexes , but D was docked to EE′ without the presence of C for most of the cases , which resulted from underestimation of the strength of the interaction between HypC and HypD . 4gwp is an example where Path-LzerD did not perform well . It is the structure of the mediator head module from yeast [54 , 55] . Mediator is an essential protein complex that regulates transcription in eukaryotes , connecting activators and repressors that are bound to promoters with RNA polymerase II ( Pol II ) . In yeast , mediator is organized into three modules: head , middle , and tail . The head module plays key roles , including messenger RNA synthesis and interaction with promoters , transcription factor TFIID , and Pol II . The head module is comprised of seven subunits , Med11 , Med17 , Med8 , Med22 , Med18 , Med20 , and Med6 . In the PDB file , these subunits correspond to chain A , B , C , D , E , and F , respectively . It is known that the assembly begins with a subcomplex with Med17 , Med11 , and Med22 ( chain A , B , and D ) , forming a helix bundle . Subsequently , Med8 and Med6 are added ( C , G ) , followed by docking of the Med20–Med18 heterodimer ( E , F ) . Thus , the assembly order is ABD> ABCDG+EF> ABCDEFG ( Fig 7 ) . The best output of Multi-LZerD has an RMSD of 34 . 25 Å and no pair of subunits is predicted within an RMSD below 4 . 0 Å . It was not trivial for Multi-LZerD to obtain the correct complex structure partly because many subunits have non-compact , elongated conformations and the pairwise decoys do not form tightly packed interactions during the assembly pathway . The assembly order is predicted almost perfectly by the BSA method , which predicts BD> ABD> ABD+EF> ABD+CG+EF> ABCDG+EF> ABCDEFG . The fourth subcomplex , ABD+CG+EF , is incorrect because chains C and G do not form a dimer before binding [54] . On the other hand , using Path-LZerD , only the low RMSD decoy combination method with some scoring functions obtained partially correct prediction ( S2 Table ) . S2 Fig shows the pathways predicted by the low RMSD decoy combination strategy using DFIRE , GOAP , and the molecular mechanics score . GOAP had both ABD and EF subcomplexes , but also the incorrect CG subcomplex . Both DFIRE and molecular mechanics had the correct steps EF and ABCDG+EF , but not the ABD subcomplex . Although the complete path was not successfully predicted for this complex , it is interesting that the ABD subcomplex , the first subcomplex that appear in the assembly path , and the EF complex , tended to be better captured by the prediction methods . This is consistent with experimental observation that failure of the ABD assembly leads to disassembly of the head [55] . Also , the detection of the EF ( Med18–Med20 ) subcomplex is consistent with their stable hydrophobic interaction , which was detected by various experimental techniques [56–58] . Thus , the path prediction is capturing stable subcomplexes during the assembly process . In several cases , the BSA method and Path-LZerD’s non-blind strategies were more successful than the blind strategies . One such case is 1w88 , a tetramer of pyruvate dehydrogenase E1 bound to the peripheral subunit binding domain of dihydrolipoyl transacetylase ( E2 ) . This complex is part of the pyruvate dehydrogenase multienzyme complex that converts pyruvate into acetyl-CoA , and consists of three enzymes , pyruvate dehydrogenase ( E1 ) , dihydrolipoyl transacetylase ( E2 ) , and dihydrolipoyl dehydrogenase ( E3 ) . The tetramer of E1 with two chains of the α subunit and two β subunits is expected to form before binding to the E2 subunit ( chain I ) , making the assembly order AA′BB′> AA′BB′I ( Fig 8 ) . Multi-LZerD built the tetramer AA′BB′ correctly with an RMSD of 1 . 3 Å , but misplaced Chain I , which resulted in an overall RMSD of 4 . 8 Å . Despite the incorrectly placed subunit , the lowest RMSD model method predicted the entire assembly order perfectly with many of the scoring functions ( S3 Table ) including the sum of score ranks ( Table 2 ) . On the other hand , the blind strategies did not predict the entire pathway correctly . Examining the pathways predicted using the final generation strategy and the sum of score ranks revealed that the majority of models , 174/200 , predict the assembly pathway BB′> AI+BB′> A′BB′+AI> AA′BB′I ( S3 Fig ) partly because many of the models in the final generation have large RMSD values with incorrect topologies . Only one model in the final generation , i . e . the lowest RMSD model , was consistent with the correct assembly pathway ( S3 Fig ) . Placing E2 ( chain I ) in the correct position was difficult in the docking because chain I is very small ( 49 residues ) relative to the other subunits ( E1 α , A and A′: 368 residues; E1 β , B and B′: 324 residues ) . On the other hand , for some complexes , the overall complex structure was not correctly predicted by Multi-LZerD , but the assembly order prediction was nonetheless successful . 1s5b is such an example . 1s5b is the structure of cholera holotoxin , formed of a homopentamer ring composed of B subunits with the A subunit bound to its face [59] . The B pentamer binds to gangliosides on the surface of target cells while A is an enzyme component , which permanently activates adenylate cyclase . The resulting elevation of cAMP causes ion efflux , leading to severe dehydration . Both components are necessary for in vivo toxic activity . Surprisingly , the homopentamer ring does not form completely prior to the A subunit binding; in fact , if the homopentamer ring is fully assembled in vitro , the A subunit cannot bind [60] . The A subunit binds to a B subunit trimer and forms major contacts with those three subunits [61]; thus , the assembly order is B1B2> B1B2B3> AB1B2B3> AB1B2B3B4> AB1B2B3B4B5 ( Fig 9 ) . The extensive contacts between the A subunit and the B subunit trimer can be seen in the third subcomplex in Fig 9 . The homopentamer ring has C5 symmetry formed by five identical homomeric heterologous interfaces . Thus , the pairwise homomeric interfaces have very similar buried surface area ( BSA ) and therefore , by definition , the pairwise BSA method must predict that all of the interactions form sequentially , e . g . B1B2> B1B2B3> B1B2B3B4> B1B2B3B4B5> AB1B2B3B4B5 . The assembly pathway exhibited by the cholera holotoxin , in which the ring formation is interrupted by a heteromeric binding step , could never be predicted by the pairwise BSA method ( S1 Table ) . However , the subcomplex BSA method was able to detect that the B subunit trimer forms a larger surface area with the A subunit than the pairwise B interface ( Table 2 ) , suggesting that the subunit BSA method is more biologically relevant than the pairwise BSA method . Despite the fact that the lowest RMSD model of 1s5b produced by Multi-LZerD was 22 . 09 Å , Path-LZerD , including the blind strategies , predicted the assembly order perfectly using many scoring functions ( Table 2 , S4 and S5 Tables ) . Although the correct structure was not predicted for this complex according to the RMSD , 199 out of 200 models in the final generation of docking prediction had almost correct topology with two or fewer additional incorrect subunit interactions , which probably is the main reason of the correct path prediction by the blind strategies . Using OPUS-PSP , the plurality of models , 77 , predict the correct assembly pathway in the final generation method ( S4 Fig ) . This suggests that the statistical scoring functions are able to detect the more major contacts that the A subunit makes with three of the B subunits . Knowing the importance of the interaction between chain A and a trimer of B in the assembly process , this protein-protein interaction may be an effective target to block [62] for preventing formation of the whole complex . The importance of obtaining models with correct topology was also observed for 2qsp , a complex of bovine hemoglobin . The best RMSD of complex models was 18 . 41 Å; however , the assembly order was correctly predicted by the blind strategies ( Table 2 ) . For this target , out of 200 models constructed in the final generation , 147 had almost correct topology with two or fewer extra incorrect subunit interactions , among which 75 ( 51 . 0% ) voted to the correct assembly order . Finally , we discuss two interesting cases where the blind strategies are correct but all of the non-blind strategies including BSA failed . 1ikn is a heterotrimer consisting of p65 ( RelA ) -p50 ( an NF-κ-B heterodimer ) bound to the inhibitor I-κ-B . Because the inhibitor binds to the heterodimer , the correct assembly order is AC> ACD ( Fig 10 ) . However , the BSA method predicts AD> ADC because the BSA of the AD interface has a larger BSA than the AC interface . The lowest RMSD structure ( 14 . 51 Å ) has the inhibitor bound primarily to p50; thus , the lowest RMSD methods predict CD> ACD . In contrast , both blind strategies predicted the correct assembly order ( Table 2 ) . Using the sum of ranks , the correct pathway had 149 votes in generation 1000 and declined slightly to 124 votes by generation 3000 ( Fig 11 ) . Nevertheless , the correct pathway maintained the majority of votes for both the final generation and consensus across generations strategies . In the final generation , 135 out of 200 models show the correct topology ( correct connections between subunits using a 5 . 0 Å cutoff distance ) , which likely improved prediction accuracy . In this case , despite the lowest RMSD structure being incorrect , the population of multiple models was able to collectively select the correct assembly pathway . It was possible because the knowledge-based scores successfully identified strong interaction between chain A and C despite their smaller interface . 4hi0 is a structure of urease accessory complex from Helicobacter pylori , which is involved in maturation of urease . Urease enables the use of urea as the sole nitrogen source is and essential for H . pylori to survive in acidic gastric environment . The complex is a hexamer consisting of a dimer of UreF/UreH heterodimers with a UreG homodimer bound . The known assembly order is forming of a dimer of the FH dimer followed by recruiting the G dimer as summarized in S1 Appendix , i . e . FH+F′H′+GG′> FF′HH′+GG′> FF′GG′HH′ ( Fig 12 ) . For this complex , the BSA method wrongly predicted that the FF′ interaction occurs first because the FF′ interface is larger than the FH interface . In contrast , for the blind methods , GOAP recognized the strength of the FH interface and perfectly predicts the assembly order ( Table 2 ) . In the final generation using GOAP , the plurality of models ( 44 ) voted for the correct assembly pathway ( S5 Fig ) . To further explore the process of assembly order prediction with Path-LZerD , we also looked at the number of votes for the correct and incorrect assembly pathways across multiple generations ( Fig 13 ) . At generation 1000 , the correct pathway had only 2 votes; however , the number of votes increased steadily across 2000 more generations until it achieved plurality . This indicates that the genetic algorithm recognized and rewarded the pairwise interfaces that were consistent with the correct assembly order . To disrupt the formation of this important protein complex for H . pylori by a small chemical compound , the target would be the F homodimer , because they have the largest interface and this interaction is prerequisite for recruiting the G dimer , whose GTPase function is essential for urease maturation [63] . These examples demonstrate that there are cases where the largest interface is not the first to form . In such cases , the BSA method will generally fail to predict the correct assembly order , while using scoring functions can lead to successful predictions . Computational time of the assembly pathway prediction for several examples are shown in Table 4 . Path-LZerD has three computational steps: pairwise subunit docking , multimeric complex construction , and path prediction from files from the complex construction process . For a three-chain complex or a small four-chain complex , the total computation was roughly 300 to 500 CPU hours , which is 1 day or less if 20 CPUs ( or cores ) are used , which is nowadays commonly available . For a five to six chain complex , the time can go up to about 1500 to 2000 CPU hours , which is 3-4 days with 20 CPUs . The time for pairwise docking is essentially proportional to the number of subunit pair combinations ( e . g . 3 for a three-chain complex and 15 for a six-chain complex ) , but the actual time is reduced if there are identical subunits in a complex . The size of proteins is another factor that influences the computational time because in general larger proteins have larger surface area to explore in docking .
The assembly order of a protein complex provides not only important insights of the molecular mechanism of complex formation but also useful practical information for obtaining subcomplexes as well as drug and protein design . This is the first systematic study of predicting protein complex assembly order that employed several different approaches . Predicting the assembly order without looking at the experimentally determined structure of the complex is totally new , and is possible by the use of a multimeric protein docking method . For example , it will take too long time for molecular dynamics to simulate an assembly process . As a core of the algorithm of the assembly pathway prediction , we used our multimeric protein docking program , Multi-LZerD . There are several multimeric docking methods developed in the past . Wolfson and his colleagues pioneered multimeric protein docking with their development of CombDock [64 , 65] and their more recent development of DockStar [66] . Methods were also reported that are specific to symmetric multimeric assembly [67–69] and homology-based modeling [70] . Protein docking has extended its applications from structure modeling , which is the original purpose , to other related topics including prediction of protein interactions in a proteome [71 , 72] and prediction of protein binding affinity [73] . The current work shows a novel application of the multimeric protein docking algorithm to complex assembly order prediction demonstrating that Multi-LZerD output can be informative even if the quaternary structure and topology are unknown or incorrect . Multi-LZerD generates a pool of models by GA , which turned out to be particularly useful for blind prediction , where a prediction is made without knowing the native structure of the target complex . In contrast to the BSA method , which needs the experimentally determined quaternary structure of the query protein complex , the current work ( Path-LZerD ) does not need the structure of the query complex because the method builds the complex structure in the course of assembly order prediction . Rather , it only needs structures of subunits as input for the assembly path prediction , because it performs multimeric protein docking prediction , and examines energy ranking of assembled pairwise decoys ( the blind strategies ) . The results show that the blind strategies worked perfectly for 3-chain targets , and performed well even in some cases where structure of the complexes were not correctly predicted . There are also some cases where the blind prediction worked better than the BSA method . The key observation that led to the development of Path-LZerD was that the pairwise decoy rankings by a binding energy scoring function can indicate docking order of a complex . Ultimately , it is desired that the approach predicts both the structure and assembly order of a multimeric complex correctly starting from the structures of subunits . Multi-LZerD successfully predicted the structure of a 6-chain complex previously [33] but none of the 6-chain complexes in the assembly order prediction dataset were well predicted ( Table 2 ) . Although there were cases where the assembly orders were correctly predicted despite incorrect complex structure modeling , in general the assembly order prediction tends to be more successful when the complex structures are well predicted , as shown for the well-predicted target subset in Table 2 . The importance of correctly predicting complex structures is also highlighted in the unbound and the modeled structure cases where the assembly order prediction accuracy deteriorated for larger complexes . Thus , a key for improving the accuracy of docking order prediction is to improve the complex structure prediction . Currently , work is ongoing to improve the performance of Multi-LZerD using more accurate scoring functions [36] and more efficient conformational search methods . | Protein-protein interactions , particularly those involving multiple proteins , are the cornerstone of numerous biological processes . Although an increasing number of multi-chain protein complex structures have been determined , fewer studies have been performed to determine the assembly order of complexes . Knowing the assembly order of a complex provides insights into the process of complex formation . Assembly order is also practically useful for reconstructing and determining the structure of a subcomplex of a large protein complex . It also has important applications including designing artificial protein complexes and drugs that prevent the assembly of protein complexes . We present a computational method , Path-LZerD , which predicts the assembly order of a protein complex by simulating its assembly process . This is the first method of this kind . A strong advantage of Path-LZerD is that the assembly order can be predicted even when the overall complex structure is not known . Path-LZerD opens a new area of computational protein structure modeling and will be an indispensable approach for studying protein complexes . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"dimers",
"(chemical",
"physics)",
"protein",
"interactions",
"molecular",
"mechanics",
"protein",
"structure",
"prediction",
"protein",
"structure",
"thermodynamics",
"protein",
"structure",
"determination",
"protein-protein",
"interactions",
"proteins",
"chemistry",
"molecu... | 2018 | Modeling the assembly order of multimeric heteroprotein complexes |
Helicobacter pylori persistently colonizes the human stomach , with mixed roles in human health . The CagA protein , a key host-interaction factor , is translocated by a type IV secretion system into host epithelial cells , where its EPIYA tyrosine phosphorylation motifs ( TPMs ) are recognized by host cell kinases , leading to multiple host cell signaling cascades . The CagA TPMs have been described as type A , B , C or D , each with a specific conserved amino acid sequence surrounding EPIYA . Database searching revealed strong non-random distribution of the B-motifs ( including EPIYA and EPIYT ) in Western H . pylori isolates . In silico analysis of Western H . pylori CagA sequences provided evidence that the EPIYT B-TPMs are significantly less associated with gastric cancer than the EPIYA B-TPMs . By generating and using a phosphorylated CagA B-TPM-specific antibody , we demonstrated the phosphorylated state of the CagA B-TPM EPIYT during H . pylori co-culture with host cells . We also showed that within host cells , CagA interaction with phosphoinositol 3-kinase ( PI3-kinase ) was B-TPM tyrosine-phosphorylation-dependent , and the recombinant CagA with EPIYT B-TPM had higher affinity to PI3-kinase and enhanced induction of AKT than the isogenic CagA with EPIYA B-TPM . Structural modeling of the CagA B-TPM motif bound to PI3-kinase indicated that the threonine residue at the pY+1 position forms a side-chain hydrogen bond to N-417 of PI3-kinase , which cannot be formed by alanine . During co-culture with AGS cells , an H . pylori strain with a CagA EPIYT B-TPM had significantly attenuated induction of interleukin-8 and hummingbird phenotype , compared to the isogenic strain with B-TPM EPIYA . These results suggest that the A/T polymorphisms could regulate CagA activity through interfering with host signaling pathways related to carcinogenesis , thus influencing cancer risk .
Helicobacter pylori , a spiral-shaped , microaerophilic gram-negative bacterium , persistently colonizes the human gastric mucosa [1 , 2] . H . pylori is carried by about 50% of the world’s population , and it exhibits extensive genetic diversity and distinct phylogeographic features [3 , 4] . Colonization increases risk of peptic ulcer disease and gastric carcinoma [5 , 6] , and has been associated with diminished risk for esophageal inflammatory and neoplastic lesions [7 , 8] , and childhood-onset asthma [9 , 10] . In 1995 , the cytotoxin-associated gene A ( CagA ) protein of H . pylori was first associated with increased risk of gastric cancer [11] , and since then , its pathogenic effects have been intensely studied [1 , 12] . The 120–145 kDa CagA protein is encoded by the cagA gene , located within the ∼40 kb H . pylori cag pathogenicity island ( cagPAI ) [13 , 14] , along with a type IV secretion system that injects it into host gastric epithelial cells [15] . The carboxy-terminal region of CagA has several Glu-Pro-Ile-Tyr-Ala ( EPIYA ) motifs which are strongly correlated to gastric disease outcomes [16 , 17] . The carboxy-terminal region of CagAs exhibit geographical , structural , and functional diversity , which is the result of the evolution of this protein through various modes of recombination mechanism [18] . In host cells , CagA molecules are associated with the inner surface of the plasma membrane and are dimerized via the carboxy-terminal EPIYA motif-containing regions [19 , 20] . CagA molecules undergo tyrosine phosphorylation ( pCagA ) at the EPIYA motifs by host Src-family kinases ( SFKs ) and Abl kinase [21–23] . CagA interacts with multiple host signaling factors through its EPIYA TPMs in a phosphorylation-dependent or -independent manner [24 , 25] , affecting cell proliferation , motility , polarity , apoptosis , inflammation and nuclear responses , which may promote gastric carcinogenesis [26–28] . By mimicking host substrates through its C-terminal sequence , CagA inhibits PAR1/MARK family kinase pathways [29] , and by association with the human tumor suppressor apoptosis-stimulating protein of p53-2 ( ASPP2 ) through its N-terminal sequence , CagA inhibits apoptosis of host cells co-colonized with H . pylori [30] . Through phosphorylated EPIYA TPMs , pCagA binds to the Src homology 2 ( SH2 ) domains of host signaling factors [26 , 28] . In this way pCagA activates the tyrosine phosphatase Shp2 , which affects cell proliferation by inducing the ERK MAP kinase cascade [31–33] , and also leads to cell elongation ( producing the hummingbird phenotype ) by inhibition of focal adhesion kinase ( FAK ) [34–36] . Phosphorylated TPMs also facilitate CagA interactions with C-terminal Src kinase ( CSK ) , which inhibits SFK activity and negatively regulates CagA-Shp2 interaction [37] . The phosphorylated CagA TPMs directly bind other tyrosine phosphatases such as Shp1 , phosphatidylinositide 3-kinase ( PI3-kinase ) and GTPase activating protein Ras GAP1 , as well as adaptor proteins Crk-I , Crk-II , Crk-L , Grb2 , and Grb7 [12 , 26] . Transgenic mice expressing wild-type CagA but not tyrosine-phosphorylation-resistant CagA developed gastric and small intestinal epithelial hyperplasia and neoplasia and B cell lymphomas and myeloid leukemias [38] , supporting a critical role of CagA tyrosine phosphorylation in H . pylori-induced oncogenesis . In addition to phosphorylation-dependent effects , CagA also associates with the polarity-regulating kinase partitioning-defective 1 ( PAR1 ) protein through its C-terminal CagA-multimerization motif ( CM ) , which overlaps with the EPIYA C- or D-TPM sequences [33 , 39] . The interaction between CagA and Par1 disrupts gastric epithelial cell tight junctions and apical-basal polarity [33] , and enhances CagA TPM-phosphorylation-dependent interactions by stabilizing complex structures such as CagA-Shp2 [20] . In total , both phosphorylation-dependent and -independent [33–35] interactions affect host signaling pathways . H . pylori has extensive genetic diversity [40–42]; isolates from different populations exhibit distinct biogeographic features , reflecting ancient human migrations [43]; cagA also possesses population-specific polymorphisms with major East Asian and Western groupings [44 , 45] . Four distinct CagA EPIYA TPMs ( A , B , C or D ) , have conserved flanking sequences [46] . East Asian CagA include A- , B- , and D-TPMs , while Western CagA has A- , B- , and C-TPMs [47] . The East Asian CagAs are more interactive with host cells than Western CagAs , largely due to the higher affinity of the strongly phosphorylated D-TPM to Shp-2 than the Western C-TPMs [47 , 48] . Western CagA includes one or multiple C-TPMs , while East Asian CagA only has one D-TPM [12] . Regardless of C/D type , most CagA molecules include single A- and B-TPMs that undergo later and not simultaneous tyrosine phosphorylation [25] . The phosphorylated A- or B-TPMs have distinct host interaction partners from C- or D-TPMs and from each other [26] , suggesting unique signaling functions . Here we report and characterize the functional importance of a specific A/T polymorphism present only within the Western CagA EPIYA B-TPMs .
Based on the CagA sequences published in Genbank , we investigated the variability within the C-terminal EPIYAs . A total of 2 , 561 complete or partial H . pylori CagA protein sequences were analyzed for polymorphisms within the EPIYAs . Our analysis indicated that the CagA B-TPM exhibits the highest variability ( Table 1 ) . EPIYA represents only 72 . 6% of 2 , 617 B-TPM sequences , with 23 alternative sequences present , including EPIYT , ESIYT , ESIYA and GSIYD; EPIYT is the most frequent alternative . In contrast , very few alternative sequences are identified in the A- and C-TPMs ( Table 1 ) , and none in the 1 , 196 type D-TPMs; only a single alternative sequence ( EPIYV ) is observed in the 1 , 620 C-TPMs ( Table 1 ) . All 25 independent CagA sequences with the EPIYV C-TPM show the same rare B-TPM ( GSIYD ) . Only one low frequency ( 0 . 2% ) alternative sequence ( EPIYT ) is observed in the A-TPMs ( Table 1 ) . In total , these results indicate specific polymorphisms in the CagA tyrosine phosphorylation ( EPIYA ) sequences , especially involving the B-TPMs . Among the 35 fully-sequenced H . pylori genomes and 81 ongoing partially-sequenced H . pylori genomes possessing the cagPAI ( based on the NCBI Bioproject Database ) , EPIYA represents 52 . 6% , while EPIYT represents 34 . 5% of the B-TPMs . Next , analyzing the B-TPMs in East Asian CagA possessing D-TPMs or Western type CagA possessing C-TPMs , we found that the distributions of the identified alternative sequences were significantly different ( Table 2 ) . In the Western CagA EPIYA B-TPMs , there are two major sequences ( EPIYA; 55 . 5% and EPIYT; 32 . 9% ) . Other alternative TPMs comprise about 11 . 6% of the sequences . In contrast , among East Asian CagA B-TPMs , EPIYA is the major sequence ( 91 . 1% ) , EPIYT is at low ( 1 . 7% ) frequency , and ESIYA is the alternative TPM with the highest frequency ( 6 . 0% ) ( Table 2 ) . In total , Western CagAs have more alternative B-TPM EPIYA sequences than do East Asian CagAs . Since Western CagAs may have more than one C-TPM , we asked whether the presence of the alternative EPIYA B-TPMs is related to C-TPM number . There was no link between the alternative B-TPM sequences and the number of C-TPMs , except for ESIYT , which co-appears with 2 C-TPMs on the same Western CagAs at significantly high frequency ( Table 3 ) . These findings indicate a common previously unrecognized TPM polymorphism with strongly non-random distribution in the available census of strains . To assess the relationship of B-TPM sequence and clinical outcome , we analyzed a total of 364 Western CagAs , which were reported to be present in patients with defined gastrointestinal pathology , according to the descriptions from Genbank and the indicated publications ( Table 4 ) . Compared with gastritis alone , gastric cancer was significantly associated with the EPIYA B-TPMs , whereas duodenal ulcers were significantly associated with the EPIYT B-TPM ( Table 4 ) . That these polymorphisms in B-TPM are associated with different diseases suggest that EPIYT and EPIYA may differentially regulate the CagA pathophysiologic roles in Western H . pylori strains that interact with host cells; while the CM and CRPIA motifs are commonly present in both forms of CagA . To assess whether the EPIYA/T polymorphisms at the protein level were random , we compared the codons in which the A/T polymorphisms were present ( S1 Table in S1 Text and Fig . 1 ) . Only one major ( 91 . 1% ) set of codons ( TAT ACT ) encodes the YT of EPIYT B-TPM . In contrast , there are two major sets of codons ( TAC GCT and TAT GCT ) that encode the YA of the EPIYA B-TPMs with similar frequencies ( 59 . 4% and 40 . 4% , respectively ) . Compared with YA and YT codons in the other loci in the H . pylori 26695 genome , this distribution is significantly non-random ( p<0 . 001 ) . This non-random distribution suggests that the polymorphisms have been selected rather than being stochastic , potentially providing for different CagA functional roles . To investigate how the major ( EPIYT ) alternative TPM affects CagA functions , we created a series of isogenic H . pylori mutants that express a Western CagA with variant TPMs . The Western-type cagA gene from H . pylori strain 147C , originally isolated from a human antrum corpus [44] , was used as the CagA parent gene for the constructions , created in a site-directed manner using a recombinant PCR technique ( S1A Fig . in S1 Text ) . This 147C cagA gene possesses one A- , B- and C-TPM . The gene product CagA147C has been previously shown to induce both AGS cell hummingbird phenotype and IL-8 production [44 , 49] . We replaced the H . pylori 26695 cagA ORF with isogenic cagA genes in its native genetic locus via transformation . The set of 6 isogenic H . pylori 26695 mutants all possess cagA with EPIYA , EPIYT , or EPIAT ( as a control: presumed to be inactive ) B-TPMs , and with EPIYA or EPIAA forms of the A- or C-TPM ( S1A Fig . in S1 Text ) . These mutants exhibit the same physiological characters , transformation frequency and growth rates in vitro as the parental 26695 strains . Western blotting confirmed that each isogenic mutant expressed a CagA molecule ( S1B Fig . in S1 Text ) , and sequencing of each cagA ORF confirmed that the sequences surrounding the target site ( s ) were identical . To investigate CagA B-TPM phosphorylation status and function during co-culture , we first generated phospho-specific and non-phospho antibodies , α-pCagA-EPIYT-918 ( phospho ) and α-CagA-EPIYT-918 ( non-phospho ) against the EPIYT B-TPM motif of CagA , corresponding to the amino acid residues derived from strain 26695 ( S2 Fig . in S1 Text ) . Our analysis indicated that the α-pCagA-EPIYT-918 ( phospho ) or α-CagA-EPIYT-918 ( non-phospho ) antibodies recognize the B-TPM including both EPIYT B-TPM and EPIYA B-TPM , but not the A-TPM or C-TPM ( control ) peptides ( S2 Fig . in S1 Text ) . To investigate whether this CagA EPIYT B-TPM can be phosphorylated during co-culture , AGS cells were co-incubated for 6 h with a set of clinical H . pylori strains , which vary in their CagA carboxy-terminal TPM sites . Phosphorylation of CagA at B-TPM was examined using the now-confirmed phospho-specific α-pCagA-EPIYT-918 antibody . Results indicated that the EPIYT-motif can be phosphorylated during co-culture , but to varying extents ( Fig . 2 ) . To investigate how this alternative B-TPM affects the role of CagA in host signaling pathways , we co-cultured the isogenic H . pylori mutants with AGS cells for 24 h and analyzed cell lysates for CagA-mediated protein binding and signal activation by immunoblotting and co-immunoprecipitation . Co-culture with the EPIYT isogenic strain induced the phosphorylation of serine/threonine kinase AKT ( also called protein kinase B , PKB ) at threonine residue 308 ( T-308 ) 2 . 4 ± 0 . 10 fold , compared with 1 . 7 ± 0 . 14 fold for the EPIYA strain ( Fig . 3 ) , indicating intensified induction of PI3-kinase/AKT activation . Neither co-culture with the isogenic H . pylori strain possessing a B-domain with an abolished tyrosine phosphorylation site , nor co-culture with the cagA knockout strain significantly increased AKT phosphorylation at T-308 ( Fig . 3 ) , suggesting that CagA-positive H . pylori can activate the PI3-kinase/AKT pathway with activity dependent on B-domain TPM phosphorylation . To investigate the interaction between the CagA B-TPM and PI3-kinase , we first used α-CagA antibodies to perform immunoprecipitation after co-culture of AGS cells with isogenic H . pylori 26695 strains with cagA variations . Western blotting using α-PI3-kinase antibody revealed that only wild-type CagA can bind to PI3-kinase but not EPIYA-ABCY>F or EPIYT-BY>F mutants ( S3 Fig . in S1 Text ) . These data indicate that EPIYT B-TPM , but not EPIYA A- or EPIYA C-TPMs , is necessary for this interaction . The α-pY-99 control blot shows that the wild-type CagA is phosphorylated , while phosphorylated CagA with the EPIYT-BY>F mutation cannot interact with PI3-kinase ( S3 Fig . in S1 Text ) . In a similar experiment , AGS cells were co-cultured with several isogenic CagA-expressing H . pylori strains including the EPIYT-ACY>F and EPIYT-BY>F mutants , followed by α-CagA immunoprecipitation . For both CagA variants ( EPIYT-ACY>F and EPIYT-BY>F ) , the phosphorylation signal was as expected . Western blotting using α-PI3-kinase antibody revealed that only CagA ( wt ) and EPIYT-ACY>F ( with intact B-TPM ) can bind to PI3-kinase , but not the EPIYT-BY>F mutant ( Fig . 4 ) . These findings indicate that EPIYT B-motif can be phosphorylated , which is necessary for the PI3-kinase-CagA B-TPM interaction . To further confirm our observation , CagA presence and phosphorylation at the EPIYT-site was examined using phospho-specific α-pCagA-EPIYT-918 and α-CagA antibodies when all samples contained similar amounts of PI3-kinase ( Fig . 5A and B ) . Only the lane with H . pylori expressing wild-type CagA revealed a signal for CagA and phosphorylation at EPIYT B-TPM in the immunoprecipitation . These findings provide further evidence that phosphorylated EPIYT B-TPM is necessary for the interaction with PI3-kinase . Co-immunoprecipitation assays indicated that in AGS cells , CagAs possessing the EPIYA or EPIYT B-TPM interacted with PI3-kinase ( p85 ) , while CagA possessing EPIAT did not ( Fig . 5C ) , suggesting that CagA activates PI3-kinase/AKT signaling pathways by interacting with the kinase via a functional B-motif . In AGS cells , CagA molecules possessing the B-domain EPIYT had higher affinity for PI3-kinase than those with EPIYA ( Fig . 5C ) . These results suggest that the CagA interaction with PI3-kinase has activating , rather than inhibiting effects on its major downstream effector , AKT . The CagA molecules without a C-TPM sequence ( as in CagA147A ) did not induce AKT phosphorylation at T-308 , indicating that the C-TPM sequence is necessary for expression of the B-domain function . The C-domain enhancement of B-domain activation of the PI3-kinase/AKT pathway is not dependent on the C-domain tyrosine phosphorylation since the B-domain EPIYT ( in HPXZ1066 ) and B-domain EPIYA ( in HPXZ1067 ) without C-domain activity activated the PI3-kinase/AKT pathway . This suggests the importance of maintaining the CagA dimerization state via the CagA multimerization ( CM ) sequence [39] . Molecular modeling of the CagA B-TPM EPIYTQVA sequence in complex with the N-terminal PI3-kinase SH2-domain reveals that the threonine residue at the pY+1 position forms a side chain hydrogen bond with an asparagine residue ( N-417 ) of PI3-kinase ( Fig . 6A ) . The respective hydrogen bond cannot be formed for the “EPIYAQVA” motif , because the alanine present at the respective sequence position lacks the hydroxyl group side chain required for an interaction ( Fig . 6B ) . Therefore , a T>A substitution at the pY+1 position is expected to significantly decrease binding affinity of the B-TPM motif to PI3-kinase . This model is also supported by experimental peptide binding studies , that show that threonine at the pY+1 position forms a stronger interaction than alanine with the respective SH2-domain [50] . Notably , PI3-kinase also contains a second SH2 domain , in which the asparagine required for ligand binding is conserved ( N-707 ) , suggesting that both PI3-kinase SH2-domains possess similar binding specificity for the pY+1 position . CagA-positive H . pylori co-cultured with AGS cells induce an elongated cell morphology known as the hummingbird phenotype which is associated with effects on host cell polarity , migration , and adhesion [36 , 51] . Next we evaluated whether the major alternative B-domain TPM EPIYT affected CagA-induced hummingbird cell formation by co-culturing AGS cells with the isogenic H . pylori 26695 cagA mutants based on a comparable bacterial/host cell population level . We found that the isogenic cagA+ H . pylori strains induced significantly more hummingbird-type AGS cells than did an H . pylori ΔcagA mutant , but abolishing the A- and C- ( EPIYA ) TPMs significantly decreased hummingbird phenotype , indicating their involvement in hummingbird induction ( Fig . 7A and B ) . In the presence of functional A and C TPMs , CagA with the B-domain EPIYA induced significantly more hummingbird cells than the CagA possessing the B-domain EPIYT or EPIAT ( Fig . 7A and B ) . This observation suggests that the B-domain EPIYT has functional differences as compared to EPIYA . In the presence of non-functional ( EPIAA ) A- and C-TPMs , strains with CagA possessing EPIYT or EPIYA B-TPM induced significantly more hummingbird cells than the strain possessing the EPIAT B-TPM ( p<0 . 05 ) ( Fig . 7B ) . B-TPM functions may be affected by the A- and C-TPMs since the significant differential effects of EPIYT and EPIYA B-TPM were lost when we abolished those tyrosine phosphorylation sites . Interleukin-8 ( IL-8 ) , a neutrophil-activating chemokine [52] , may play an important role linking chronic inflammation and carcinogenesis [53] . The IL-8 induction effect is also associated with the number of C domains in Western CagA+ strains [49 , 54] . Here , we evaluated whether the major alternative B-domain TPM sequence , EPIYT , affects CagA-induced IL-8 production by co-culturing AGS cells with the isogenic H . pylori 26695 cagA mutants based on a comparable bacterial/host cell population level . At 24 h , AGS cells co-cultured with H . pylori ΔcagA mutants had the same IL-8 level as the AGS cells without H . pylori co-culture ( control ) , but co-culture with the isogenic H . pylori cagA variants induced significantly higher IL-8 levels . Under these conditions , the isogenic cagA mutant containing the EPIYA B-TPM induced significantly more IL-8 induction than the mutant with the EPIYT B-TPM ( Fig . 7C ) , a finding indicating differential EPIYA- and EPIYT-B TPM protein functions . The isogenic mutant with an EPIAT B-TPM had significantly decreased IL-8 induction ( Fig . 7C ) . These results confirm that CagA can induce AGS cell IL-8 production via its B-domain TPM , and indicate the differential EPIYT and EPIYA functions .
A key host interaction factor of H . pylori , the CagA protein , has multiple polymorphisms which differ in their affinities to host interaction partners and in their regulation of gastric cell signaling cascades . EPIYA TPMs are critically important for CagA regulation of host signaling pathways [28 , 55] , and the four types ( A , B , C and D ) have different host interaction partners [26] and/or varying affinities to the same partners [31 , 46] , suggesting differential roles in regulation of host signaling pathways . Matsunari et al . first reported there are three most common types of EPIYA sequences including the EPIYA , EPIYT and ESIYA , and that EPIYT of B-TPM is more predominant in Western CagAs [56] . In this study , we further reveal the A/T polymorphism that specifically occurs within the Western type B-domain , and we provide evidence for the first time that this polymorphism significantly affects CagA functions in host cells . Indeed , the B-domain polymorphisms of the Western strains differed in their correlation with upper GI tract diseases ( Table 4 ) , suggesting that a single SNP in a major bacterial interactive factor could decide disease outcome . The isogenic H . pylori cagA mutants expressing from the native genetic locus created for the present investigations may be valuable for further studies . Through its B-domain , CagA interacts with host partners including the Shp2 phosphatase , Csk and PI3-kinases , as well as adaptor proteins Grb2 and Crk and the Shp1phosphatase , all of which carry SH2-domains [28] . Among these , Shp1 and Shp2 also interact with the A- or C-domains , Csk with the A-domain , Grb2 with C-domain , while PI3-kinase and CrkII only with the B-domain in a tyrosine-phosphorylation-dependent manner [26 , 28] . Different TPMs have both shared and specific host interaction partners suggesting that different EPIYA motifs could have specific roles in regulating host signaling pathways . Moreover , other motifs aside from the EPIYA TPMs could also be involved in CagA interactions with these host factors . CagA activation of PI3-kinase/AKT appears dependent on the CRPIA sequences [24] , but activation of PI3-kinase/AKT also may be CagA-independent [37] . Our finding that H . pylori CagA with functional EPIYA or EPIYT B-domains binds with PI3-kinase confirms and extends the observations by Selbach et al . [28] . Recently , Lind et al . developed a novel strategy to systematically analyse phosphotyrosine antibodies recognizing single phosphorylated CagA EPIYA-motifs utilizing synthesized phospho- and non-phosphopeptides [57] . With this strategy , by generating and analyzing a novel phospho-specific CagA B-motif antibody ( anti-pCagA-EPIYT-918 ) and isogenic CagA mutants with abolished TPMs , we further confirmed the CagA EPIYT-B domain tyrosine-phosphorylation status during H . pylori co-culture with host cells and revealed that tyrosine phosphorylation of the B-domain is necessary for the interaction between CagA and PI3-kinase . We observed that the EPIYT B-domain has higher affinity to PI3-kinase and greater AKT activation than the EPIYA B-domain . Our analysis by structural modeling of the CagA EPIYA and EPIYT B-motifs interacting with the SH2 domain of PI3-kinase further revealed the nature of differential interaction effects caused by the A/T polymorphism . During gastric colonization , the host tyrosine kinases Src and Abl phosphorylate H . pylori CagA EPIYA motifs [21–23 , 25] , which differ from the classical consensus phosphorylation sites in eukaryotic target factors ( E-E-I-Y-E/G-X-F and I/V/L-Y-X-X-P/F of two tyrosine kinases , respectively ) [58] . This suggests that the CagA EPIYA motif phosphorylation level in host cells may not be maximal . In that case , we propose that the A/T polymorphism/switch in the EPIYA motif could be important in regulation of the TPM phosphorylation efficiency and stability . Strain-specific CagA sequence variation involves both conserved and non-conserved regions . CagA147C used in our studies and CagA26695 used by Suzuki et al . share only 87 . 1% identify and have numerous SNPs flanking each EPIYA/T TPM as well as differing tagging . Considering the complex interactions between CagA with host protein partners as well as the complex signaling network , use of transfected protein-expressing systems and both technical and structural differences may affect signaling . The number of H . pylori CagA molecules within host cells in different assays ( e . g . CagA transfection and co-culture with H . pylori with native expressing CagA ) could markedly vary with differing kinetics of CagA phosphorylation , leading to different outcomes . Increased AKT activation and decreased IL-8 secretion of AGS cells [59] , and PI3-kinase/AKT pathway repression of IL-8 production during Salmonella co-culture with intestinal epithelial cells [60] have been described . Strain- and time-dependent H . pylori CagA-mediated IL-8 induction in AGS cells occurs through the Erk and NF-κB pathways [49 , 61–64] . CagA-mediated PI3-kinase/AKT activation attenuates IL-8 induction by repressing Erk/NF-κB including through the Shp-2/Erk pathway [49] , the Shp2-independent Ras/Raf/Mek/Erk pathway [65] , and by Ras-independent Erk activation [66] . Inactivating B-TPM abolished PI3-kinase/AKT activation , but decreased IL-8 secretion . Through B-TPM , CagA also may interact with multiple other proteins in a site-competition and/or time-dependent manner . For example , CagA TPMs interact with Shp2 through phosphorylated EPIYAs [28 , 34] enhancing Shp2 activity and Erk phosphorylation [31] . The B-TPM could positively regulate IL-8 production through activating the Shp2/Erk/ NF-κB pathway [49] . Abolishing the isogenic single B-TPM inactivated the IL-8-repressing PI3-kinase/AKT , but also inactivated IL-8-stimulating Shp2/Erk . Repression of IL-8 production by the B-TPM-mediated PI3-kinase/AKT effect reflects cross-talk between the PI3-kinase/AKT and Shp2/Erk pathways . Consistent with the overall differential signaling , Western cagA+ H . pylori strains with EPIYT or EPIYA B-TPMs are associated with different patterns of clinical outcomes ( Table 4 and Fig . 8 ) , an observation that needs to be confirmed . The clinical outcome of H . pylori colonization results from long-term processes , and therefore , how the B-TPM-mediated PI3-kinase/AKT effect alters host gastric cancer development in long-term H . pylori colonization deserves further investigation . The PI3-kinase/AKT signaling pathway controls many of the hallmarks of cancer , and many tumor tissues have enhanced PI3-kinase/AKT activities [67 , 68] . However , PI3-kinase/AKT and their effectors are pleiotropic and have complex crosstalk and feedback behaviors in the signaling network , which are not fully known [67 , 68] . We studied the regulation of CagA on PI3-kinase/AKT pathway in vitro at an early time of bacterial interaction with host cells , while long-term studies in mice or observations in patients at risk for gastric cancer will help resolving the clinical significance of the polymorphism . H . pylori colonization induces AGS cell scattering and elongation ( hummingbird phenotype ) through multiple CagA-related mechanisms; CagA binds to Csk through its A- or B-TPMs , inhibiting SFK activity , and binds to Shp2 through its A- , B- or C-TPMs , leading to FAK dephosphorylation [35 , 69] . Attenuated hummingbird phenotype induction present in the cagA mutant with the EPIYT B-TPM ( vs . EPIYA ) reflects differential B-domain functional roles , possibly through modulating direct interactions with Csk or Shp2 . Inhibition of AKT activation by the PI3-kinase inhibitor LY294002 has no effect on the hummingbird phenotype [36] . However , LY294002 inhibits PI3-kinase catalysis by competing for ATP binding [70] , but does not directly affect the p85 binding activity with tyrosine-phosphorylated motifs such as the CagA B-TPM . Such findings suggest that hummingbird induction by the CagA B-TPM relates to the competition between PI3-kinase and Csk or Shp2 for binding at the B-TPM , but is not directly related to PI3-kinase activity . The A- and B-domains have unique host interacting partners , such as Csk , which do not interact with C- or D-TPMs . These domains could possibly attenuate the C-domain-Shp2-interaction by binding with Csk to inactivate SFK members [71] . In this model , the C- and D-TPMs serve as the primary phosphorylation motifs interacting with host signaling partners , and the A- and B-TPMs serve as secondary sites , phosphorylated after C- or D-TPMs [25] , suggesting a potential regulatory role through competition between different TPMs . The EPIYT/EPIYA B-TPM polymorphism that we studied provides a new level of complexity in H . pylori colonization and pathophysiology .
H . pylori strain 26695 was used to construct a series of isogenic cagA mutants , which express CagA variants from the native CagA genetic locus [72] . H . pylori strains 147C and 147A , a pair of naturally occurring isogenic cagA strains with EPIYA ABC and AB motifs , respectively [49] , were used as isogenic cagA sequence templates . H . pylori CagA-expressing strains P227 , Oki-61 , P341 , 2002-370 and 26695 , also were used for AGS co-culture and evaluation of CagA phosphorylation [73] . The H . pylori strains were grown at 37°C in 5% CO2 on trypticase soy agar ( TSA ) plates with 5% sheep blood ( TSA , BBL Microbiology Systems , Cockeysville MD ) or Brucella agar plates ( BA , Difco Laboratories , Detroit MI ) supplemented with 10% newborn calf serum ( NBCS; Serologicals Corporation , Norcross GA ) and suitable antibiotics [74] . Antibiotic-resistant isogenic H . pylori strains were selected with kanamycin ( Km; 10 μg/mL ) or chloramphenicol ( Cm; 30 μg/mL ) , as appropriate . Alternatively , H . pylori strains were grown in thin layers on horse serum GC agar plates supplemented with vancomycin ( 10 μg/mL ) , nystatin ( 1 μg/mL ) , and trimethoprim ( 5 μg/mL ) , and for defined mutants with Cm ( 6 μg/mL ) and/or Km ( 8 μg/mL ) at 37°C for 2 days in an anaerobic jar containing a Campygen gas mixture of 5% O2 , 10% CO2 , and 85% N2 ( Oxoid , Wesel , Germany ) [75] . E . coli DH5α was grown in Luria-Bertani ( LB ) medium at 37°C [76] . Ampicillin ( Ap; 100 μg/mL ) , Cm ( 30 μg/mL ) or Km ( 50 μg/mL ) were used for selecting vectors or the constructs in E . coli during cloning . Western type H . pylori strain 147C cagA has an EPIYT B-TPM , as well as one EPIYA -A and -C TPM ( cagA147C B:EPIYT A&C:Y ) [49] . To evaluate B-TPM A/T polymorphism effects on CagA functions , the threonine site of EPIYT B-TPM was replaced with alanine by recombination-PCR mediated site-directed mutagenesis , leading to the generation of the isogenic cagA , cagA147C B:EPIYA A&C:Y . To evaluate B-TPM A/T polymorphism effects on tyrosine phosphorylation of CagA B-TPM , the two tyrosine phosphorylation sites , A- and C-TPMs of the wild-type cagA ( cagA147C B:EPIYT A&C:Y ) and the isogenic cagA ( cagA147C B:EPIYA A&C:Y ) , were replaced with alanine , leaving the B-TPM as the only functional tyrosine phosphorylation site . This resulted in two isogenic cagAs: cagA147C B:EPIYT A&C:Y>A and cagA147C B:EPIYA A&C:Y>A . For controls , the tyrosine of B-TPMs of the wild-type and the isogenic cagAs were further replaced with alanine to abolish B-TPM tyrosine phosphorylation function , leading to cagA147C B:EPIAT A&C:Y and cagA147C B:EPIAT A&C:Y>A . Each of the wild-type and isogenic cagA genes was first fused at the 3’ end to the hemagglutinin ( HA ) tag and then linked to a 617 bp cagA downstream region sequence based on the 26695 genomic sequence [72] . An aphA ( KmR ) cassette [77] was inserted between the cagA-HA fusion gene and the cagA downstream region sequence as a selection marker for the H . pylori mutant construction . Each construction was cloned into the vector pGEM-T easy ( Promega , Madison WI ) , creating plasmids pXZ476 , pXZ465 , pXZ468 , pXZ471 , pXZ472 , and pXZ475 , which carry cagA147C B:EPIYT A&C:Y , cagA147C B:EPIYA A&C:Y , cagA147C B:EPIAT A&C:Y , cagA147C B:EPIYT A&C:Y>A , cagA147C B:EPIYA A&C:Y>A , and cagA147C B:EPIAT A&C:Y>A , respectively ( S2 Table in S1 Text ) . To replace the cagA26695 sequence of H . pylori strain 26695 with the isogenic cagA147C sequences , a truncated cagA147CN ( 2445 bp ) lacking C-terminal A- B- or C-TPMs was cloned and linked with a cat ( CmR ) cassette [77] and then with the 617 bp cagA downstream region sequence based on the 26695 sequence using pGEM-T easy , creating pXZ478 ( S2 Table in S1 Text ) . To construct the H . pylori ΔcagA control , the cagA upstream ( 839 bp ) and downstream ( 1076 ) region sequences of H . pylori 26695 were linked and inserted with an intervening sacB-cat ( CmR ) cassette to replace the entire cagA26695 ORF on the same vector , creating plasmid pXZ083 ( S2 Table in S1 Text ) . To express the series of mutant cagA genes from the cagA native genetic locus in the 26695 genetic background , we first replaced the cagA26695 sequence on the 26695genome with a cagA147Cs via homologous recombination ( S1 Fig . in S1 Text ) . The wild-type H . pylori strain 26695 was transformed by plasmid pXZ478 to CmR to create strain HPXZ1043 with isogenic cagA147CN replacing native cagA26695 . DNA sequencing confirmed the replacement of cagA26695 with the cagA147CN sequence in the 26695-derived CmR/KmS strain HPXZ1043 , and western blot confirmed the expression of the truncated CagA147CN protein from the native locus and the CagA promoter . Strain HPXZ1043 CmR/KmS was then transformed to KmR/CmS with plasmids , pXZ476 , pXZ465 , pXZ468 , pXZ471 , pXZ472 , or pXZ475 , to create mutants HPXZ1061 ( cagA147C B:EPIYT A&C:Y ) , HPXZ1062 ( cagA147C B:EPIYA A&C:Y ) , HPXZ1065 ( cagA147C B:EPIAT A&C:Y ) , HPXZ1066 ( cagA147C B:EPIYT A&C:Y>A ) , HPXZ1067 ( cagA147C B:EPIYA A&C:Y>A ) and HPXZ1070 ( cagA147C B:EPIAT A&C:Y>A ) , respectively . The wild-type H . pylori strain 26695 was transformed to CmR with pXZ083 to create the cagA-negative mutant HPXZ1146 ( ΔcagA::sacB-cat ) ( S2 Table in S1 Text ) . To confirm each construction , sequencing of related regions was performed at Macrogen ( Rockville MD ) , and all sequence analysis was performed using Sequencher ( Gene Codes , Ann Arbor MI ) . To analyse the EPIYA- and EPIYT-motifs in the CagA protein , the complete cagA gene of H . pylori strain 26695 ( accession number: AAD07614 ) containing its promoter was amplified by PCR , cloned into the pCR2 . 1 vector ( Invitrogen ) and sequenced [73] . For construction of a complementation vector , this cagA fragment was cloned in the E . coli/H . pylori shuttle vector pHel3 containing the oriT of RP4 and a kanamycin resistance gene cassette ( Aph-A3 ) as a selectable marker , resulting in vector pSB19 [73] . Site-directed mutagenesis of tyrosines Y-899 , Y-918 and Y-972 in the CagA sequence was done using the Sculptor mutagenesis kit , as directed ( Amersham Pharmacia Biotech ) and resulting plasmids were transformed into H . pylori isogenic ΔcagA mutant [78] , as described [79] . The C-STEPIYAKVNK ( EPIYA-A ) , C-STEPI ( pY ) AKVNK ( phospho-EPIYA-A ) , C-PEEPIYTQVAK ( EPIYT-B ) , C-PEEPI ( pY ) TQVAK ( phospho-EPIYT-B ) , C-PEEPIYAQVAK ( EPIYA-B ) , C-PEEPI ( pY ) AQVAK ( phospho-EPIYA-B ) , C-SPEPIYATIDD ( EPIYA-C ) and C-SPEPI ( pY ) ATIDD ( phospho-EPIYA-C ) amino acid sequences were synthesized by Jerini AG ( Berlin , Germany ) . These 11-mer peptides were chosen because prior studies have shown that α-phosphotyrosine antibodies typically recognize short phosphopeptides , and 11-mer and 9-mer sequences are both necessary and sufficient [57] . Commonly , 11-mer peptides also are used for immunizations to generate phospho-specific antibodies , which then recognize the corresponding phosphopeptides bound to affinity columns and in ELISA ( Biogenes , Berlin , Germany ) . All above EPIYA and EPIYT peptides were purified by HPLC , and full-length synthesis as well as purity of each peptide was confirmed by mass spectrometry by Jerini AG . The peptides were resolved at a concentration of 1 mg/mL in DMSO and stored at −20°C . Twenty μg of each CagA peptide were mixed in 1 mL of TBST blotting buffer ( 140 mM NaCl; 25 mM Tris-HCl , pH 7 . 4; 0 . 1% Tween-20 ) . These peptide samples were spotted onto Immobilon-P membrane ( Merck Millipore , Darmstadt , Germany ) using the BioDot SF apparatus ( Bio-Rad , Munich , Germany ) . The resulting Dotblots were dried and subjected to antibody detection as described below for Western blots [57] . Human gastric epithelial ( AGS; ATCC CRL 1739 ) cells ( obtained from American Type Culture Collection ) were cultured at 37°C in a humidified atmosphere with 5% CO2 in RPMI 1640 ( Invitrogen , Carlsbad CA ) with 10% fetal bovine serum ( FBS; Invitrogen ) with antibiotic-antimycotic mixture ( 1X; Life Technologies , Grand Island NY ) [80] . Before co-culture experiments , AGS cells ( 2×105cells/well ) were transferred to a new 6-well plate and incubated in fresh RPMI 1640 with 10% FBS and antibiotic-antimycotic for 24 h . The attached AGS cells were washed and incubated in the RPMI 1640 media without serum or antibiotic for 16 h . The AGS cells were then co-cultured with PBS-prewashed H . pylori cells , which were collected from 24-h TSA plates at a multiplicity of infection ( MOI ) of 100:1 and from fresh serum- and antibiotic-free RPMI 1640 for 8–24 h . AGS cultures and the co-culture of AGS and H . pylori were grown on coverglass ( Fisher Scientific , Pittsburgh PA ) in 6-well plates [81] . After 24-h co-culture , 10 random fields of view for each coverglass were examined at a magnification of ×200 using a Leica DMI 6000B microscope . Alternatively , the AGS cells were co-cultured with H . pylori cells at MOI of 50 for 6 h , when the cells were harvested in ice-cold PBS containing 1 mmol/L Na3VO4 ( Sigma-Aldrich ) . Elongated AGS cells in each experiment were quantitated in 10 different 0 . 25-mm2 fields using an Olympus IX50 phase contrast microscope [82] . All experiments were performed in triplicate . After 24-h incubation , co-culture media were sampled and centrifuged at 16 , 000 g , and supernatants collected . AGS cell IL-8 secretion was measured by an enzyme-linked immunosorbent assay using the Human IL-8 ELISA Kit II ( BD Biosciences , San Jose CA ) , in accordance with the manufacturer’s instructions . Phospho-specific and non-phospho polyclonal rabbit CagA antibodies were raised against peptides corresponding to the following amino acid residues derived from the B-TPM motif of strain 26695: C-PEEPIYTQVAK ( non-phospho-EPIYT-B ) and C-PEEPI ( pY ) TQVAK ( phospho-EPIYT-B ) . For this purpose , both peptides were conjugated to Limulus polyphemus haemocyanin carrier protein and two rabbits each were immunized by Biogenes GmbH ( Berlin , Germany ) , according to standard protocols . The resulting phospho-specific antibodies ( α-pCagA-EPIYT-918 ) were affinity-purified against the corresponding non-phospho peptide bound to a column . The resulting non-phospho antibodies ( α-CagA-EPIYT-918 ) were affinity-purified against the corresponding phospho-peptide . Both antibodies were prepared and purified by Biogenes GmbH ( Berlin , Germany ) . Their specificity was confirmed by dot blotting against the phospho- and non-phospho peptides ( S2 Fig . in S1 Text ) . To prepare whole cell extracts for immunoblotting , media were removed after 24-h incubation and AGS cells were washed with ice-cold PBS 5 times to remove H . pylori cells . The whole cell lysates for western blotting were prepared with RIPA lysis buffer ( Thermo Scientific Pierce , Rockford IL ) with Halt Protease and Phosphatase Inhibitor Cocktail ( Thermo Scientific Pierce ) . Lysates were separated by SDS-PAGE ( Expedeon Inc . San Diego CA ) and transferred to Immobolin-P PVDF Transer Membrane ( Fisher Scientific ) . Membranes were blocked in TBST with 3% BSA or 5% skim milk for 1 or 2 h at room temperature . Membranes were incubated with the following antibodies according to the instructions of the manufacturer . Immunodetection of CagA peptides and the various proteins of interest were performed using horseradish peroxidase–conjugated anti-mouse or anti-rabbit polyvalent sheep immunoglobulin secondary antibodies and using chemiluminescence reagents , West Femto Chemiluminescent Substrate ( Thermo Scientific Pierce ) or Amersham ECL Western Blotting Detection Reagents ( GE Healthcare , Piscataway NJ ) in accordance with the manufacturers’ instructions [83] . After exposure to X-ray film , the target band intensities were quantified using ImageJ software ( NIH , Bethesda MD ) . For immunoprecipitation , the whole cell lysates were prepared with IP Lysis/Wash buffer ( Thermo Scientific Pierce , Rockford IL ) with Halt Protease and Phosphatase Inhibitor Cocktail ( Thermo Scientific Pierce ) . Immune complexes were prepared using Pierce Crosslink Immunoprecipitation kit ( Thermo Scientific Pierce ) in accordance with the manufacturers’ instructions . The immunoprecipitates were subjected to SDS-PAGE , as described above . Anti-actin , anti-AKT , anti-phospho-AKT ( pThr308 ) , and anti-phospho-AKT ( pSer473 ) ePI3-kinase were obtained from Cell Signaling Technology ( Danvers MA ) . Anti-HA , anti-Shp2 , anti-Crk II , and anti-phospho-CrkII ( pTyr221 ) were obtained from Thermo Scientific Pierce , anti-phosphotyrosine was obtained from EMD Millipore Inc . ( Billerica MA ) , anti-Csk , anti-phospho-Csk ( pSer364 ) , anti-GAPDH and pan-phosphotyrosine antibody pY-99 were obtained from Santa Cruz Biotechnology , Inc . ( Santa Cruz CA ) , and anti-CagA was produced as described [49] . Anti-PI3-kinase ( p85 ) antibodies were obtained from Cell Signaling Technology ( Danvers MA ) or Santa Cruz Biotechnology , Inc . ( Santa Cruz CA ) . Rabbit polyclonal and mouse monoclonal α-CagA antibodies were from Austral Biologicals , or from Emd Millipore Corporation ( Billerica MA ) . The phosphorylation status of CagA and bound PI3-kinase also was verified by immunoprecipitation experiments as described [73 , 79] . Briefly , co-cultured or control AGS cells were washed with cold PBS and lysed for 30 min at 4°C in lysis buffer ( 20 mM Tris pH 7 . 2 , 150 mM NaCl , 5 mM EDTA , 1% Triton X-100 , 10% glycerol , 1 mM Na3VO4 , COMPLETE™ inhibitor mix from Roche ) . Lysates were pre-cleared with protein G-Sepharose ( Pharmacia , Uppsala , Sweden ) for 2 h at 4°C . Two micrograms of the monoclonal α-CagA antibody ( Austral Biologicals , San Ramon CA ) or polyclonal antibody against the p85 subunit of PI3-kinase ( Santa Cruz Biotechnology , Dallas TX ) were added to the supernatant and incubated overnight at 4°C on a shaker . Immune complexes were precipitated by addition of protein G-sepharose for 2 h , washed three times in 0 . 5× PBS and then mixed with equal amounts of 2× SDS-PAGE buffer . Precipitates were analyzed by SDS-PAGE and immunoblotting . Spot or band intensities on blots probed with the different α-phosphotyrosine antibodies were quantitated with the Lumi-Imager F1 ( Roche Diagnostics , Mannheim , Germany ) . Densitometric measurement of signal intensities revealed the percentage of phosphorylation per sample [84] . The CagA-PI3-kinase interaction was modeled using the crystal structure of the N-terminal PI3-kinase SH2-domain in complex with a C-kit phosphotyrosyl peptide ( PDB: 2IUH ) as a template [85] . Modeling of the Cag-A B-TPM motif was performed using SwissModel [86] and included the sequence stretches “EPIYTQVA” or “EPIYAQVA” . Structural analysis and visualization was performed using RasMol [87] . A total of 2561 H . pylori complete or partial CagA protein sequences available at GenBank on August 8th 2013 were collected . The cagA EPIYA A- , B- , C- , and D-TPM types were defined as described [46] . The numbers of each type of EPIYA TPMs and the polymorphisms within the five specified amino acids were tabulated independently three times . A Chi-square test was performed to evaluate the variance in the representation of each EPIYA TPM . | As the dominant bacterium living in the human stomach , Helicobacter pylori has mixed roles in host health . One significant pathogenic risk factor is the CagA protein , which interferes with multiple host cell signaling pathways through its EPIYA tyrosine phosphorylation motifs ( TPMs ) . Through database searching and silico analysis , we reveal a strong non-random distribution of the EPIYA B motif polymorphisms ( including EPIYT and EPIYA ) in Western H . pylori isolates , and provide evidence that the EPIYT are significantly less associated with gastric cancer than the EPIYA . By constructing a series of H . pylori cagA isogenic mutants and isogenic complementation plasmids , generating specific antibodies , co-culturing with human AGS cells , performing biochemical and modeling analysis , we demonstrate that CagA B-motif phosphorylation status is essential for its interaction with host PI3-kinase during colonization and that CagA with an EPIYT B-motif had significantly attenuated induction of interleukin-8 and the hummingbird phenotype , had higher affinity with PI3-kinase , and enhanced induction of AKT compared to the EPIYA . These findings provide insight into how Western H . pylori CagA regulates cancer-related activity inside host cells through the A/T polymorphisms at the functionally important B motif . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | A Specific A/T Polymorphism in Western Tyrosine Phosphorylation B-Motifs Regulates Helicobacter pylori CagA Epithelial Cell Interactions |
The circadian clock plays a vital role in monarch butterfly ( Danaus plexippus ) migration by providing the timing component of time-compensated sun compass orientation , a process that is important for successful navigation . We therefore evaluated the monarch clockwork by focusing on the functions of a Drosophila-like cryptochrome ( cry ) , designated cry1 , and a vertebrate-like cry , designated cry2 , that are both expressed in the butterfly and by placing these genes in the context of other relevant clock genes in vivo . We found that similar temporal patterns of clock gene expression and protein levels occur in the heads , as occur in DpN1 cells , of a monarch cell line that contains a light-driven clock . CRY1 mediates TIMELESS degradation by light in DpN1 cells , and a light-induced TIMELESS decrease occurs in putative clock cells in the pars lateralis ( PL ) in the brain . Moreover , monarch cry1 transgenes partially rescue both biochemical and behavioral light-input defects in cryb mutant Drosophila . CRY2 is the major transcriptional repressor of CLOCK:CYCLE-mediated transcription in DpN1 cells , and endogenous CRY2 potently inhibits transcription without involvement of PERIOD . CRY2 is co-localized with clock proteins in the PL , and there it translocates to the nucleus at the appropriate time for transcriptional repression . We also discovered CRY2-positive neural projections that oscillate in the central complex . The results define a novel , CRY-centric clock mechanism in the monarch in which CRY1 likely functions as a blue-light photoreceptor for entrainment , whereas CRY2 functions within the clockwork as the transcriptional repressor of a negative transcriptional feedback loop . Our data further suggest that CRY2 may have a dual role in the monarch butterfly's brain—as a core clock element and as an output that regulates circadian activity in the central complex , the likely site of the sun compass .
In insects , circadian clocks regulate the timing of numerous biological events [1] . Some examples of critical circadian rhythm outputs in holometabolous insects include the time of day of adult eclosion , the seasonal timing of reproductive diapause , and time-compensated sun compass navigation . The molecular clock mechanism has been the subject of intense investigation in Drosophila [2 , 3] , while less attention has been directed at the clockwork mechanism in other , non-drosophilid insects . In the fruit fly , the central clock is driven primarily by a negative transcriptional feedback loop that involves the products of the period ( per ) , and timeless ( tim ) genes , and the transcription factors Clock ( Clk ) and cycle ( cyc ) . CLK and CYC heterodimers drive per and tim transcription through E-box enhancer elements . The resultant PER and TIM proteins form heterodimers that translocate back into the nucleus to repress their own transcription via inhibitory effects on CLK and CYC . Drosophila CRYPTOCHROME ( CRY ) is co-localized in clock cells with PER and TIM and functions as a blue-light photoreceptor involved in photic entrainment [4–6] . CRY disrupts PER and TIM heterodimers by directly interacting with TIM in a light-dependent process [7–9] , and it also participates in its own light-dependent degradation [10] . The eastern North American monarch butterfly ( Danaus plexippus ) is well known for its long-distance fall migration [11] . We have been developing this species as a model to examine the role of the circadian clock in time-compensated sun compass orientation and in the seasonal induction of the migratory generation [12] . Using clock protein expression patterns , we previously identified the location of circadian clock cells in the dorsolateral protocerebrum ( pars lateralis [PL] ) of the butterfly [13] , which expresses PER , TIM , and a Drosophila-like CRY ( designated CRY1; see below ) . We also identified a CRY1-staining neural pathway that may connect the circadian ( navigational ) clock to polarized light input entering brain , which is important for sun compass navigation [14 , 15] . A CRY1 pathway also may connect the circadian clock to neurosecretory cells in the pars intercerebralis ( PI ) for the initiation of the migratory state [12 , 13] . A direct clock-to–sun compass pathway has also been postulated [13] . In the course of our molecular investigations of the circadian clock mechanism in monarchs , we have discovered that these butterflies , like all other non-drosophilid insects so far examined , express a second cry gene that encodes a vertebrate-like protein designated insect CRY2 [16] . Functional studies in Drosophila Schneider 2 ( S2 ) cells show that monarch CRY2 is light insensitive , but potently inhibits CLOCK:CYCLE-mediated transcription , whereas monarch CRY1 is light sensitive , but does not show transcriptional repressive activity . However , the mechanistic details of CRY2′s actual function within a clockwork have not been defined in any insect . Further molecular evolutionary studies have shown that gene duplication and loss have led to three modes of cry gene expression in insects , giving rise to three types of circadian clocks [17]: two derived clocks , in which only cry1 ( e . g . , Drosophila ) or cry2 ( e . g . , the honey bee Apis mellifera and red flour beetle Tribolium castaneum ) is expressed , and an ancestral clock in which both cry1 and cry2 are expressed ( e . g . , the monarch butterfly ) . The expression of two functionally distinct crys in monarchs suggests that the butterfly clock may use a novel clockwork mechanism that is not yet fully described in any organism . In the studies discussed here , we have therefore used in vivo approaches , a monarch cell line that contains a light-driven molecular clock , and Drosophila carrying monarch cry1 or cry2 transgenes to elucidate the monarch clockwork mechanism and its photic entrainment . Our results define many characteristics of a CRY-centric clock in the monarch butterfly with CRY1 functioning potentially as a blue-light photoreceptor for photic entrainment , whereas CRY2 functions , without PER , within the clockwork as the major transcriptional repressor of the core transcriptional feedback loop . We also present evidence of a CRY2-positive neural pathway that oscillates in the central complex , the apparent site of the sun compass [18 , 19] . CRY2 may thus function as both a core clock element and as an output-regulating circadian activity in the central complex .
If a negative transcriptional feedback loop underlies the circadian clock in monarch butterflies , it should drive the rhythmic expression of per and tim in vivo . We thus used quantitative real-time polymerase chain reactions ( qPCRs ) to examine the temporal expression patterns of the clock gene homologs in monarch butterfly heads at 3-h intervals in a 12 h light:12 h dark cycle ( LD ) and during the first day in constant darkness ( DD ) . Monarch per RNA levels exhibited a daily rhythm in LD with peak levels at Zeitgeber time ( ZT ) 18 and low levels at ZT 0–3 , and the rhythm persisted in DD ( p < 0 . 0001 , one-way analysis of variance [ANOVA] ) ( Figure 1A ) , as previously described [20] . We found that a rhythm of similar phase was manifested by monarch tim RNA levels in both LD and DD ( p < 0 . 0001 ) ( Figure 1A ) . We also examined cry1 and cry2 RNA levels; although each RNA profile showed a similar trend , neither exhibited a significant daily rhythm ( p > 0 . 05 ) ( Figure 1A ) . Monarch-specific antibodies against PER , TIM , CRY1 , and CRY2 were used to examine the temporal profiles of clock protein abundance in monarch head extracts by Western blot analysis . Indeed , PER and TIM showed significant temporal oscillations in abundance in LD ( p < 0 . 001 ) , with peak levels occurring from ZT 18–24/0 ( Figure 1B ) . There were also temporal changes in PER electrophoretic mobility; the changes in mobility were due to changes in phosphorylation , as phosphatase treatment converted >90% of the high–molecular weight forms of PER to a single , lower–molecular weight band ( Figure S1 ) . The more highly phosphorylated forms of PER were predominant at 3 h after lights-on . In DD , the oscillation in PER abundance persisted ( p < 0 . 01 ) , while the oscillation in TIM abundance was markedly blunted , to the point that there was no longer a significant daily rhythm ( p > 0 . 05 ) . Thus , the daily TIM abundance oscillation in the head is mainly light driven . There was no significant daily change in either monarch CRY1 or CRY2 abundance in whole head extracts in either LD or DD ( p > 0 . 05 ) ( Figure 1B ) . We evaluated a monarch butterfly cell line designated DpN1 [21] , which was originally derived from embryos , for expression of circadian clock RNAs and proteins , because such a cell line might be useful for helping us delineate the molecular clock mechanism in the butterfly . In DpN1 cells , we in fact found that the RNAs for per , tim , cry1 , cry2 , Clk , cyc , vrille , Pdp1 , slimb , doubletime , CKIIα , CKIIβ , and shaggy were all expressed ( see Table S1 ) . We focused our studies of the temporal dynamics of clock gene expression in DpN1 cells on per , tim , cry1 , and cry2 to parallel our in vivo analyses . Remarkably , when studied at 4-h intervals under LD , we found cycling in clock gene RNA levels ( by qPCR ) and clock protein abundance ( by Western blot analysis ) . At the level of gene expression , we found that monarch per , tim , and cry2 exhibited near-synchronous daily rhythms in RNA levels , with peak levels between ZT 16 and 24 , and trough levels between ZT 4 and 8 ( p < 0 . 001 ) ( Figure 1C ) . There was no significant daily oscillation in cry1 levels in LD ( p > 0 . 05 ) . In DD , no clock gene RNA oscillation was apparent on the first day . This lack of a circadian oscillation was consistently observed in repeated experiments . At the protein level , monarch PER and TIM showed robust temporal oscillations in abundance in DpN1 cells in LD , with highest protein levels at the end of the dark period ( ZT 24/0 ) ( p < 0 . 05 for PER and p < 0 . 001 for TIM; Figure 1D ) . CRY2 also showed temporal changes in abundance , with highest levels 4 h later at ZT 4 ( p < 0 . 001; Figure 1D ) . For PER , there was not only a diurnal change in protein abundance but also in electrophoretic mobility , as found in head extracts ( Figure S1 ) . Phosphorylated PER was the dominant form at ZT 4 , which correlated with the highest level of CRY2 abundance , and the rapidly declining per , tim , and cry2 RNA levels . This temporal increase in CRY2 abundance in DpN1 cells contrasts with the lack of rhythmicity in CRY2 abundance over the 24-h day in LD in monarch heads ( compare Figure 1B with 1D ) . The reason for this discrepancy is because CRY2 is more widely expressed in the monarch brain than the other clock proteins examined , and CRY2 is not under robust circadian control in most areas ( see below ) . The temporal profiles of clock gene RNA and protein expression in DpN1 cells are consistent with PER and/or CRY2 being involved in negative feedback repression of CLK:CYC-mediated transcription in the cell line , which is further explored below . Similar to what we found for RNA expression in DpN1 cells , we were unable to identify a circadian oscillation of the clock proteins in the cells in DD ( Figure 1D ) . Although it is unclear why we were not able to detect a functional circadian clock in DpN1 cells , the close correlation of clock gene RNA and protein expression patterns between DpN1 cells and heads in LD , makes the cell line a useful system in which to study the molecular and biochemical details of the monarch clock transcriptional feedback loop in LD ( focusing on the role of CRY2 ) , as well as its intracellular light input pathway ( focusing on the role of CRY1 ) . We first used DpN1 cells to examine whether monarch CRY1 mediates the light-induced decrease in TIM abundance , providing a light-resetting pathway into the molecular clock . By using RNA interference induced by double-stranded RNAs ( dsRNAs , ) we supply evidence that the light-induced decrease in TIM abundance in DpN1 cells is mediated through CRY1 ( Figure 2 and Figure S2 ) . Once lights were turned on to initiate the normal light period in LD-cultured control cells ( those treated with double-stranded RNA [dsRNA] targeting the green fluorescent protein [GFP] gene ) , there was a transient increase in CRY1 abundance at 15 and 30 min ( Figure 2A , black line ) , followed by a rapid decrease by 60 min , reaching constant low levels by 120 min; the light-induced decrease in CRY1 abundance in LD-cultured cells was unexpected ( see below ) . With lights on , there was a rapid decrease in TIM abundance at 15 min , reaching constant low levels by 60 min ( Figure 2B , black lines ) . Light induced a slower decrease in PER abundance starting at 120 min , with a steady decline throughout the light period ( Figure 2C , black line ) . The light-induced decline in CRY2 abundance was even slower and only apparent at 540 min ( Figure 2D , black line ) . The time course of light-induced protein decrements from TIM to CRY2 was similar to that seen after lights on ( ZT 12 ) in LD without dsRNA treatment ( Figure 1D ) and is consistent with a series of protective protein:protein interactions in which TIM:PER interactions protect PER from degradation , whereas PER:CRY2 interactions protect CRY2 from degradation ( see below ) . A surprising aspect of the control experiment was that the initiation of the light period now caused a decrease in CRY1 abundance in cells treated with dsRNA targeting GFP , rather than CRY1 levels remaining at constant dark-like levels in the light , as seen in untreated cells cultured under LD ( Figure 1D ) . This light-induced CRY1 decrease was found to be secondary to a 5-h serum starvation of the medium that is necessary for efficient transfection of dsRNA into DpN1 cells ( unpublished data ) ; serum starvation likely induces the expression of a kinase that is important for monarch CRY1′s proteasomal degradation by light ( see Figure S2 ) . Nonetheless , pretreatment of cells maintained in LD with dsRNA targeting cry1 , which caused a ∼60% reduction in CRY1 abundance in darkness just prior to ( time 0 ) and throughout light exposure ( Figure2A , red line ) , greatly reduced the decrease in TIM abundance in response to light ( Figure 2B , red lines ) . Pretreatment also greatly reduced the subsequent decreases in PER and CRY2 abundance ( Figure 2C and 2D , red lines ) , compared with controls ( cells treated with dsRNA targeting GFP ) . The lack of a complete block of the light-induced reduction of TIM appeared to be secondary to the partial CRY1 knockdown ( see Figure 2A ) . The dsRNA data strongly suggest that CRY1 mediates the light-induced TIM degradation in DpN1 cells ( see also Figure S2 ) . Consistent with CRY1-mediating photic entrainment in the butterfly [22] , we found that blue light is the spectral component that degrades CRY1 and TIM in DpN1 cells and also synchronizes the timing of behavior ( the adult eclosion rhythm ) to the 24-h day ( Figure S3 ) . Next , we examined the location of light-sensing clock cells in monarch brain by immunocytochemistry , using our newly-developed monarch-specific anti-TIM antibodies . Monarch TIM-like immunoreactivity was detected by the new antibodies in the cytoplasm of cells in the PI and PL ( Figure 3A–3G and unpublished data ) , as previously described using an anti-TIM antibody against Drosophila TIM [13] . Each of the monarch-specific antibodies gave prominent staining patterns in the cytoplasm ( compared with weak staining with the Drosophila antibody , see [13] ) , with ∼25 large cells stained in the PI and four cells consistently stained in the PL . In addition , approximately eight cells were identified near the lobula region of the optic lobe ( OL ) , and approximately eight cells were found in the suboesophageal ganglion ( SOG ) . Double-labeling studies showed that the cytoplasmic TIM staining was localized in the PL to the four cells that co-express corazonin ( Figure 3B and 3C ) , a neuropeptide that marks clock cells in the PL of lepidopteran brains [23 , 24] , including monarchs , in which two of the four cells also stain for PER and CRY1 [13] . Moreover , direct comparison confirmed that the cytoplasmic staining of CRY1 and TIM were colocalized in two of the four cells in PL ( Figure S4 ) . We were unable to determine whether CRY1 and TIM were colocalized in the PI , however , because of weak staining for CRY1 in this structure; because there were twice as many TIM-positive cells as CRY1-positive cells in PI , only half of those TIM-positive cells would be expected to be colocalized with CRY1 . The anti-TIM antibodies also stained a group of cells in the dorsal region of the OL ( Figure 3A and Figure S5 ) in close vicinity , but not identical to the CRY1-positive group of cells previously described there [13] . These TIM-positive cells projected into the same glomerular structure as the adjacent CRY1-staining cells ( Figure S5 ) . We did not observe detectable TIM staining in the nuclei of any of the cell groups . All of the cyptoplasmic staining in TIM-positive cells in the brain appeared to be light sensitive in LD . As previously noted , Western blot data showed a large light-driven daily oscillation of TIM in heads under LD conditions , with the daily oscillation of TIM abundance substantially blunted on placement in DD ( Figure 1B ) . A similar pattern was found for the TIM-positive cells in the brain . In LD , all TIM-positive regions exhibited significantly lower levels of TIM staining at ZT 6 , compared to ZT 15 , including all four TIM-positive cells in PL ( Figure 3H ) . In DD , on the other hand , there was a significant oscillation in PL only ( p < 0 . 05 ) , with lower staining at circadian time ( CT ) 9 and higher staining at CT 15 ( Figure 3D , 3E , and 3I ) . In all other areas ( PL , OL , and SOG ) , there was no significant difference between CT 9 and CT 15 ( p > 0 . 05 ) . When subjected to a 1-h light pulse from ZT 14–15 ( ZT 15L ) , a significant light-induced decrease in TIM levels was detected in the PL only ( p < 0 . 01 ) , affecting all four TIM-positive cells , compared with brains kept in the dark ( Figure 3F , 3G , and J ) . In all other areas ( PI , OL , and SOG ) , there was a clear trend for a decrease in TIM staining with the light pulse ( Figure 3J ) , but it did not reach significance ( p > 0 . 05 ) . Collectively , the data show that there is a good correlation in the different lighting schedules between TIM abundance changes in heads detected by Western blots and TIM staining patterns in brain regions detected by immunocytochemistry . TIM staining in the PL was the area most consistently regulated ( by light and in DD ) . These data show a complex relationship between CRY1 and TIM degradation in the monarch brain . Wherever CRY1 and TIM are colocalized , CRY1 likely mediates TIM degradation , based on our studies in DpN1 cells ( Figure 2 ) . In the other TIM-positive areas , either CRY1 is present below the level of antibody detection or TIM in those cells is degraded in a CRY1-independent manner , perhaps by local interactions ( as may occur in PL ) , by opsins expressed in brain , and/or by neural pathways from eye and/or stemmata to TIM-positive cells . We showed previously by immunocytochemistry that CRY1 levels in the PI and PL are not altered by light exposure [13] . It thus appears that the light-induced decrease in TIM in TIM/CRY1 colocalized cells is not necessarily accompanied by a measurable decrease in monarch CRY1 abundance , which has also been shown by Western blot analysis in LD ( Figure 1B and 1D ) and with short-term light exposure at night both in DpN1 cells and in whole-head extracts ( Figure S6 ) , as well as in Drosophila [7] . It thus appears that light may induce a conformational change in monarch CRY1 , leading to TIM degradation , but without necessarily inducing its own degradation . Because there are no genetic approaches yet available in monarch butterflies [12] , we asked whether monarch CRY1 can function as a circadian photoreceptor by expressing monarch transgenes in Drosophila . We used the GAL4-UAS system , with tim-GAL4 as the driver , which drives transgene expression in clock neurons that generate the circadian locomotor activity rhythm [25] . For these studies , we took advantage of the cryb mutation in Drosophila , because it induces severe light-input defects; circadian phase does not shift in response to a light pulse , and TIM does not cycle in LD [4–6] . We attempted to rescue these phenotypes by expressing UAS-monarch cry1 or UAS-monarch cry2 transgenes in the cryb background . We first examined the ability of the monarch cry1 transgene to restore the ability of discrete light pulses at night to phase-shift the circadian clock that drives locomotor activity in cryb mutant flies . We used two light pulses; a 1-h light pulse at ZT 15 , which normally causes phase delays , or a 1-h light pulse at ZT 21 , which normally causes phase advances [5] . The light-pulse experiments using four independent UAS-cry1 lines showed a partial rescue of the cryb phenotype . With a light pulse at ZT 21 , the phase advances in the UAS-cry1 lines 1a , 15b , and 22b were as robust as the y w control ( no significant differences ) , and the phase advance of line 6b was only slightly less than that of y w ( p < 0 . 05 ) ( Figure 4A ) . With a light pulse at ZT 15 , the rescue was still evident , but not as robust; all four UAS-cry1 lines had a statistically smaller phase change than y w ( p < 0 . 001 for each ) , but they also had a statistically larger phase change than the cryb line ( p < 0 . 01 for 1a and 6b; p < 0 . 001 for 15b and 22b ) ( Figure 4A ) . When the same phase shift experiment was performed with three UAS-cry2 lines—19a , 18b , and 125a—at both ZT 15 and ZT 21 , the phase changes were minimal and not significantly different from the cryb line without transgene expression ( p > 0 . 05 ) ( Figure 4B ) . Next , the four UAS-cry1 lines were examined for their ability to rescue the light-induced , CRY-dependent TIM oscillations in heads of the cryb background . In cryb flies , TIM levels do not cycle in LD . It is known that the light-induced TIM oscillation can be rescued by expressing Drosophila CRY under the tim-GAL4 driver [5] . Each of the UAS-Cry1 lines partially rescued TIM cycling in fly heads ( Figure 4C ) . Note that although TIM does not normally degrade in cryb flies , some degree of cycling is occasionally observed , as seen in this set of experiments ( Figure 4C , lanes 1 and 2 ) . When TIM cycling was examined in the three UAS-cry2 lines in LD , TIM cycling was not restored , indicating that monarch CRY2 cannot rescue this cryb defect ( Figure 4D ) . The results of these behavioral ( light pulse ) and biochemical ( TIM degradation ) experiments strongly suggest that monarch CRY1 can function as a circadian photoreceptor in Drosophila , whereas monarch CRY2 cannot . Having provided several lines of evidence suggesting that CRY1 functions as a photoreceptor for the butterfly clock , we next used DpN1 cells to construct the primary gear of the circadian clock , a negative transcriptional feedback loop , by examining the ability of monarch PER , TIM , CRY1 , or CRY2 to inhibit monarch ( dp ) CLK:dpCYC–mediated transcription . Previous studies in S2 cells have shown that monarch CRY2 is a potent repressor of dpCLK:dpCYC–mediated transcription [16 , 17] , but it has also been shown in S2 cells that both the Drosophila and Antheraea pernyi PER proteins alone potently repress Drosophila ( d ) CLK:dCYC–mediated transcription [26–29] . The DpN1 cell line was ideal for the current study because it allowed for the exogenously expressed monarch proteins to be examined in a homologous cell-based system . We used luciferase reporter gene assays with a reporter construct containing a tandem repeat of the proximal CACGTG E-box enhancer from the monarch per gene promoter [16 , 17] . Cotransfection of the reporter with monarch CLK and CYC caused a 100-fold increase in transcriptional activity ( Figure 5A ) . As expected , monarch CRY2 potently inhibited dpCLK:dpCYC–mediated transcription in a dose-dependent manner , yet neither monarch PER nor monarch TIM inhibited transcription ( Figure 5A ) ; transfected monarch PER is >90% nuclear in DpN1 cells ( unpublished data ) . The same result was found with independent PER constructs obtained from cDNA from different sources of monarch head RNA ( unpublished data ) . Monarch PER does have the potential to inhibit transcription in other cellular contexts , because it robustly inhibited dCLK:dCYC–mediated transcription in a dose-dependent manner in Drosophila S2 cells ( unpublished data ) . These data suggest that the monarch clock homologs can participate in a negative transcriptional feedback loop . A novel aspect of this feedback loop is that monarch CRY2 has the major inhibitory role for repressing dpCLK:dpCYC–mediated transcription from a monarch per E box enhancer , while PER was ineffective ( either alone or in combination with TIM or sub maximal inhibitory doses of CRY2 , Figure S7 ) . Next , a repressive effect of endogenous monarch CRY2 was examined on dpCLK:dpCYC–mediated transcription using dsRNAs to knock down endogenous clock gene expression in DpN1 cells . For one dsRNA approach , the monarch per E box luciferase reporter and monarch CLOCK and CYC were cotransfected to elevate reporter activity . The ability of endogenous PER , TIM , CRY1 , or CRY2 to inhibit CLK:CYC–mediated transcriptional activity was then evaluated using dsRNA directed against each clock gene RNA to determine whether knockdown elevated ( de-repressed ) luciferase activity and what effect knockdown had on the levels of all four clock proteins . The luciferase value obtained with dsRNA against GFP was the control for comparison of clock protein levels and knockdown-induced de-repression ( Figure 5B , lane 1 ) . Double-stranded RNA directed against per caused a substantial reduction in both PER and CRY2 abundance , and luciferase activity was elevated ( de-repressed ) by ∼3-fold ( Figure 5B , lane 2 ) . The decrease in CRY2 abundance with dsRNA against per did not appear to be the result of a decrease in cry2 transcription ( Figure S8 ) , but was due to a post-transcriptional process , likely involving direct PER:CRY2 interactions , which protect CRY2 from degradation ( see below ) . Double-stranded RNA against tim knocked down TIM abundance , and also caused a modest decrease in PER and CRY2 abundance , while luciferase reporter activity was elevated ( de-repressed ) 2-fold ( Figure 5B , lane 3 ) . Double-stranded RNA against cry1 substantially reduced CRY1 abundance only , and did not cause an elevation in luciferase reporter activity compared to GFP control ( Figure 5B , lane 4 versus lane 1 ) . Double-stranded RNA against cry2 caused a ∼70% reduction in CRY2 abundance only , while reporter activity was elevated ( de-repressed ) to a level comparable to the value with dsRNA against per ( Figure 5B , lane 5 versus lane 2 ) . Collectively , these data strongly suggest that endogenous CRY2 alone ( not PER ) is a dominant repressor of dpCLK:dpCYC–mediated transcription in DpN1 cells . The dsRNA knockdown results are also consistent with PER stabilizing CRY2 and TIM stabilizing PER ( see also the temporal order of light-induced clock protein degradation , Figure 1D , Figure 2 , and Figure S2A ) , and show that the de-repression following knockdown of PER or TIM is due to secondary reductions in CRY2 levels . These biochemical data suggest that TIM , PER , and CRY2 are in the same protein complex . We therefore examined endogenous protein interactions by incubating DpN1 cell or brain extracts with clock protein antisera and probing the resulting immune complexes for each of the three clock proteins by Western blot analysis . Immunoprecipitated PER pulled down TIM and CRY2 , immunoprecipitated TIM pulled down PER and CRY2 , and immunoprecipitated CRY2 pulled down PER and TIM in both DpN1 cells and in brains ( Figure 5C ) . These results are consistent with the existence of endogenous clock protein complexes containing PER , TIM , and CRY2 . The data are also consistent with the protective protein interactions ( TIM protects PER from degradation and PER protects CRY2 from degradation ) suggested in previous experiments ( see Figure 2 , Figure S2A and Figure 5B ) . In our second dsRNA approach , dsRNA against cry2 was transfected into DpN1 cells to knock down CRY2 , and per RNA levels were monitored at 4-h intervals over 24 h in LD , and dsRNA against GFP served as the control . We could not use dsRNA against per for this approach , because of the secondary effect of PER knockdown decreasing CRY2 levels , as documented above ( Figure 5B , lanes 2 ) . With GFP dsRNA , the normal daily oscillation of per RNA in LD was clearly apparent and unaltered with high levels from ZT 20–24 ( Figure 5D ) . With CRY2 knockdown , on the other hand , per RNA levels remained at peak values throughout the 24-h period , with no oscillation ( Figure 5D and Figure S9A ) . This result confirms that endogenous CRY2 is the major repressor of dpCLK:dpCYC–mediated transcription for this light-driven clock , because without substantial CRY2 , per transcription remains constantly high over the 24-h period in LD . Moreover , the increase in PER levels with CRY2 knockdown again shows that endogenous CRY2 is the major repressor; there is no evidence for a role of PER in CRY2′s repressive ability in DpN1 cells . If CRY2 is the transcriptional repressor of the diurnal clock in DpN1 cells , then its cellular localization should change over the day , being mainly nuclear at the time of maximal repression of dpCLK:dpCYC–mediated transcription . We thus examined the temporal profile of nuclear CRY2 in DpN1 cells and compared the time course to the normal daily rhythm in per RNA levels depicted in Figure 5D ( solid lines ) , as a measure of dpCLK:dpCYC–mediated transcriptional readout . When the temporal profiles were examined at 4-h intervals over 24 h in LD , we found a clear daily change in the cellular location of CRY2 ( Figure 5E and Figure S9B ) . The amount of CRY2 in the nucleus began to increase at ZT 16 and peaked at ZT 4 , the predicted time of CRY2 maximal repression , when per RNA levels had dropped to near low values ( Figure 5D ) . Because the low levels of per RNA persisted with increasing time in the light period of LD ( ZT 8 and 12; Figure 5D ) , the amount of CRY2 in the nucleus began to decline ( Figure 5E ) . These data show an oscillation in nuclear CRY2 abundance that is consistent with its role as the major transcriptional repressor of the light-driven clock in DpN1 cells . Perhaps in LD , only a portion of CRY2 in DpN1 cells—the portion translocated from cytoplasm to nucleus—is functionally relevant for inhibition of dpCLK:dpCYC–mediated transcription . But what about CRY2 function in the monarch brain ? We first used in situ hybridization to map cry2 RNA expression in the monarch brain . The brain distribution revealed RNA staining in ∼16 cells in the PI , four cells in the PL , ∼six cells in the central protocerebrum ventrally from the central body and dorsally from the oesophageal foramen , and ∼four cells in the SOG ( Figure S10A–S10C ) . There was also extensive staining in the OLs , which included cells in the dorsal and ventral OL , and several hundred small cells that were found between the lobula and medulla , between the medulla and lamina , and between the lamina and retina ( Figure S10A ) . Using our newly developed monarch-specific anti-CRY2 antibodies , the anatomical location of CRY2 staining by immunocytochemistry was very similar to the RNA expression pattern ( Figure 6A ) . CRY2-like immunoreactivity was detected in the cytoplasm of ∼16 cells in the PI and four cells in the PL ( Figure 6B and 6C ) . Double labeling studies showed that the CRY2 staining was localized in the PL to the four cells that co-express corazonin ( Figure 6D and 6E ) and TIM . Direct comparison confirmed co-localization of CRY2 and TIM in the same four cells in the PL ( Figure S11 ) . There were ∼25 CRY2-positive cells in the dorsal OL , ∼35 in the ventral OL , and ∼500 small CRY2-positive cells between the lobula and medulla and medulla and lamina . The main discrepancy between the RNA and protein patterns was that CRY2 staining was not detected in the RNA-expressing cells between the lamina and retina ( Figure S10A versus Figure 6A ) . When the temporal profile of CRY2 staining in the PI , PL , and dorsal and ventral OL ( the CRY2-positive cell groups in which signal intensity allowed for semiquantitative assessment ) was analyzed over the circadian cycle , we found a significant circadian oscillation of cytoplasmic CRY2 staining in PL ( p < 0 . 05 ) , PI ( p < 0 . 01 ) , and OL ( p < 0 . 01 ) , which was most pronounced in OL ( Figure 6F ) , with peak staining at CT 15 . Importantly , there was no detectable circadian oscillation in the ∼500 small cells in OL between lobula and medulla and between medulla and lamina , which compose over 90% of CRY2 staining in brain . This staining pattern accounts for our inability to detect a daily CRY2 oscillation in either head extracts ( Figure 1B ) or brains dissected away from photoreceptors ( unpublished data ) . These CRY2-positive cells in OL overlap with those detected as expressing cry2 RNA by in situ hybridization ( Figure S10A ) ; therefore , these cells in OL also likely account for the lack of a detectable cry2 RNA rhythm in heads ( Figure 1A ) . CRY2 nuclear staining should be observed in the PL at the time of transcriptional repression . Such evidence of nuclear translocation is expected based on the transcriptional feedback loop model of the Drosophila circadian clock [2] and on what we found for CRY2 in DpN1 cells ( Figure 5E ) . Until now , we have not been able to find an obvious rhythmic nuclear accumulation of any clock protein so far examined ( PER , TIM , CRY1 , as well as CRY2 ) in the PL or in any other monarch brain region . One possible explanation for not finding nuclear clock proteins is that each protein is heavily expressed in cytoplasm of PI and PL and , by comparison , there might be a relatively small amount of functionally relevant clock protein that does cycle into the nucleus to alter transcription , as appears to occur for phosphorylated nuclear PER bound to chromatin in Drosophila [28] . With this in mind , we initially examined CRY2 staining in thin ( 5 μm ) sections throughout the entire monarch brain focusing on nuclear occurrence of CRY2 at 2-h intervals from ZT 18 to ZT 6 , which covered seven points over the time interval in which we would expect to find CRY2 in the nucleus ( Figure 6G and 6H ) , based on our studies of DpN1 cells ( see Figure 5D and 5E ) . We compared the temporal pattern of nuclear CRY2 to the per RNA rhythm in monarch brain ( Figure 6H , upper panel ) , because the per RNA rhythm is the most consistent clock gene rhythm in monarchs ( Figure 1A ) , and it is the same assay we used as a transcriptional readout of dpCLK:dpCYC–mediated transcription for comparison with the temporal profile of nuclear CRY2 in DpN1 cells ( Figure 5D , solid line ) . In the PL , the nuclei are large ( 10 μm in diameter ) , and counterstaining with three specific fluorescent DNA probes revealed that these cells are unique in that most of the chromatin is distributed around the inner edge of the nuclear envelope and in small patches in the nucleus . In addition , the amount of DNA staining detected in the nucleus per se is minute , compared with nuclear staining in surrounding cells ( Figure S12A ) . Nonetheless , using high-power microscopy in combination with a sensitive charge-coupled device ( CCD ) camera , we found clear evidence of temporal control of CRY2 staining in the nucleus of PL cells , which was limited to the four cells in PL and was not found in any other CRY2-positive cells in brain ( Figure 6G and 6H ) . Specifically , over the 12-h period of study in LD , we identified nuclear CRY2 staining at ZT 2 and 4 only; no nuclear staining was detected at ZT 18 , 20 , 22 , 24 , or ZT 6 ( Figure 6G , left column; Figure 6H , middle panel ) . The CRY2 nuclear staining in the PL co-localized with the chromatin detected in the nucleus by the DNA probes ( Figure S12B ) . We next examined four time points over the circadian cycle and found CRY2 nuclear staining in PL only at CT 3 , and not at CT 9 , 15 , or 21 ( Figure 6G , right column; Figure 6H , lower panel ) . The timing of CRY2 nuclear occurrence correlated well with the time of maximal transcriptional repression of the per RNA oscillation in monarch brain ( Figure 6H , upper panel ) , similar to the temporal profiles described in DpN1 cells ( see Figure 5D and 5E ) . It is likely that CRY2 is present in the nucleus of relevant PL cells starting several hours before the peak , with the peak being what we are detecting for nuclear CRY2 in Figure 6G and 6H , based on our studies in DpN1 cells . We thus conclude that the cyclic presence of CRY2 in the nucleus of PL cells closes the circadian transcriptional feedback loop in vivo in the monarch butterfly . We also looked at 5-μm sections for PER staining in the nuclei of PL cells over the circadian cycle using an antipeptide antibody that we previously used to characterize PER staining in monarch brain [13] . However , high background staining gave inconclusive results and no clear nuclear staining was detected above background at any of the Zeitgeber or clock times examined ( unpublished data ) . Nonetheless , because of the strong evidence presented for CRY2 as a major transcriptional repressor of a clock feedback loop in monarchs ( data in Figure 5 ) , the detection of temporally controlled , nuclear CRY2 in putative clock neurons in butterfly brain helps resolve a puzzle that has existed for the last 10 y of work on lepidopteran clocks [26 , 30 , 31] . The site of the sun compass in insects now appears to be the central complex [18 , 19] . The central complex is a midline structure consisting of the dorsally positioned protocerebral bridge and the more ventrally situated central body , which has upper and lower subdivisions . Recent studies in locusts and Drosophila have shown that the central complex is not only a control center for motor coordination but is also the actual site of the sun compass ( for polarized skylight integration from both eyes and probably all skylight information ) [18] , as well as being involved in visual pattern learning and recognition [32] . Finding a clock connection with the central complex in the monarch butterfly would be a major advance for beginning to understand its remarkable navigational capabilities . Both CRY2 arborizations and projections were identified in the brains of monarch butterflies ( Figure 7A ) . The strongest and most dense arborization of CRY2 staining was found in the central body , just ventral from the protocerebral bridge ( Figure 7B ) . This staining in the central complex was specific for CRY2 , because staining for PER , TIM , or CRY1 was not detected in the central body . Another CRY2 arborization was found in the superior medial and lateral protocerebra , which are connected via the protocerebral bridge just above ( dorsal to ) the central body . In addition to these two arborizations , there are three CRY2-staining projections that could be traced . The first projection was coming from the protocerebral bridge and PI laterally toward the four cells in the PL ( Figure 7C–7E ) . The second projection was extending from the superior lateral protocerebrum toward the OL ( but it was not seen in the OL ) ( Figure 7F and 7G ) . The third projection traveled from the superior medial protocerebrum ventrally , likely to the corpora cardiaca/corpora allata complex , because CRY2 staining was detected in both these neurohemal organs ( Figure 7H ) . It appeared that the CRY2 pathways arise from cells in PL and/or PI . The CRY2-positive arborizations were under circadian control with strong staining in all areas at CT 15 and little to no staining detectable in those areas at CT 9 . Dramatic CRY2 cycling was especially apparent in central body ( Figure 7I–7K ) . These data provide evidence for a potential dual role for CRY2: as a core clock element and as an output that regulates circadian activity in the central complex . Collectively , our results provide several lines of evidence suggesting that monarch CRY1 functions in vivo as a circadian photoreceptor , whereas CRY2 functions as a transcriptional repressor for the butterfly clockwork . This novel clock mechanism has aspects of both the Drosophila and mouse circadian clocks rolled into one , as well as unique aspects of its own ( Figure 8A ) . The CRY1-TIM pathway for light-induced resetting of the monarch clock is similar to that found in the fruit fly , and the butterfly is the only other animal , outside of Drosophila , in which a photoreceptive function of CRY1 for clock entrainment has been shown in vivo . What is different between photoreceptive CRY function in fruit fly and monarch is that the cascade of protein degradation events ends with CRY2′s degradation in the butterfly , rather than with PER's , as occurs in Drosophila . We propose that it is the ultimate decrease in CRY2 levels that resets the CLK:CYC–driven transcriptional feedback loop in monarch butterflies ( see temporal protein decay patterns in the light periods in Figure 1D ) . Then what is the function of monarch PER ? We have shown that PER is important for stabilizing CRY2 , and PER:CRY2 heterodimers may also be involved in translocating CRY2 into the nucleus , as occurs in mammals [33] , although we could not detect PER in the nucleus of PL cells using currently available antibodies . It is also still possible that PER has a minor role in repression of CLK:CYC–mediated transcription , although the dominant repressor in monarchs is CRY2 . The role of monarch CRY2 as a transcriptional repressor is similar to the role of the CRYs in the mouse clockwork [33] . The existence of CRY2 and its repressive function , independent of PER , are major distinguishing features of the monarch clock mechanism from that of Drosophila . Drosophila CRY has been suggested to function in the peripheral clockwork as a transcriptional repressor [34–36] , but only when overexpressed with PER [37] , and no such clock-like function driving behavior has been detected for fruit fly CRY overexpressed within the central clock of Drosophila [4] . We have been able to track monarch CRY2′s movement into the nuclei of PL cells at clock times appropriate for its role as a major transcriptional repressor of the butterfly clock feedback loop ( Figure 6G and 6H ) —no previous nuclear translocation of clock proteins has been reported in any other non-dipteran species . Our studies set the stage for more careful examination of this issue in other insects , as also suggested by a recent study in the housefly Musca domestica [38] . It is likely that monarch CRY2 exerts its inhibitory function on transcription by directly interacting with CLK:CYC heterodimers , which can now be assessed in DpN1 cells . DpN1 cells are also an important reagent for examining CRY1 signaling mechanisms , as it is the only insect cell line reported that has all the endogenous machinery from CRY1 light sensing through the degradation of CRY2 . The CRY-centric ancestral circadian clock we have defined in monarch butterflies may be common in those non-drosophilid invertebrates that express both cry1 and cry2 . The CRY-centric clock of the monarch may also hold a key to understanding the regulation of critical migratory behaviors , including time-compensated sun compass navigation [20 , 39 , 40] . The relatively intense staining of the CRY proteins in cytoplasm suggests output roles for the proteins distinct from those involved in the circadian clock mechanism and its entrainment by light ( Figure 8A ) . Indeed , previous work has shown that a CRY1-staining neural pathway may connect the circadian clock to polarized light input entering brain that may ultimately impinge on the sun compass ( Figure 8B; [13] ) . The results presented here further show that a CRY2-staining neural pathway may more directly connect the circadian clock to the central complex ( Figure 8B ) , the likely site of a sun compass [18 , 19] , and that the pathway communicates circadian information to the sun compass ( Figure 7I and 7J ) . CRY2 may simply be marking a circadian pathway to the sun compass or it may be directly involved in rhythmic synaptic activity in that region . The elucidation of a novel central clock mechanism in monarch butterflies and the finding of CRY-staining neural pathways to aspects of sun compass integration provide a solid cellular , molecular , and biochemical foundation for further functional and genetic studies into the remarkable navigational capabilities of the monarch butterfly .
Monarch butterflies were purchased from commercial sources . The butterflies were housed in the laboratory in glassine envelopes in Percival incubators with controlled temperature ( 21 °C ) , humidity ( 70% ) , and lighting . The butterflies were fed 25% honey every third day . cDNA fragments were cloned by degenerate PCR ( see Table S1 ) . cDNA templates for PCR were prepared from RNA purified from monarch butterfly whole heads or brains . The ends of the coding regions were obtained by rapid amplification of cDNA ends ( RACE; Clontech kits ) . Complete open reading frames were obtained by PfuTurbo ( Stratagene ) PCR from cDNA . Clones were sequenced at core facilities at University of Massachusetts Medical School . Sequences were analyzed with MacVector ( Accelrys ) and the National Center for Biotechnology Information website ( http://www . ncbi . nlm . nih . gov/BLAST/ ) . Total RNA was extracted using Trizol ( Invitrogen ) . For head RNA extraction , an additional charcoal purification step was added before isopropanol precipitation to remove eye pigments and other factors that interfere with reverse transcription . The quantifications of clock gene expression were done using real-time quantitative PCR by TaqMan probes with an ABI Prism 7000 SDS ( Applied Biosystems ) . Total RNA was treated with RQ1 DNase ( Promega ) , and random hexamers were used ( Promega ) to prime reverse transcription with Superscript II ( Invitrogen ) , all according to manufacturers' instructions . PCR reactions were assembled by combining two master mixes . The first mix contained approximately 1 μg of cDNA template and 13 μl Platinum Quantitative PCR SuperMix-UDG w/ROX ( Invitrogen ) per reaction and was aliquoted into a PCR plate . The second mix contained forward and reverse primers ( 0 . 9 μM final concentration of each ) , probe ( 0 . 25 μM final concentration ) and the water needed to bring each reaction to a final volume of 25 μl , and was subsequently aliquoted into the PCR plate . The monarch per and control rp49 primers and probes were identical to those reported previously [20] . The other primers and probes were as follows ( F , forward primer; R , reverse primer; P , probe; all 5′-3′ ) : monarch timF , CCAAACAGAGGACCAACAACAA; timR , CCTCGTTTGACGATCTTCTTTCTC; timP , FAM-TCGCGCTGGCGTAACGCTTCA-TAMRA; monarch cry1F , AAAGATGGTGGGCTACAATCGT; cry1R , CCTGAACTGCTGGTCCAAATC; cry1P , FAM-TGCGATACCTGCTGGAGGCGCT-TAMRA; monarch cry2F , CTGGAGCGACATTTGGAGAGA; cry2R , CAAGAGTGATTCTGGCGTCATCT; cry2P , FAM-AGGCTTGGGTCGCTTCGTTCGG-TAMRA . All primers and FAM-TAMRA labeled probes were purchased from Integrated DNA Technologies ( Coralville ) . The efficiency of the amplification and detection by all primer and probe sets were validated by determin-ing the slope of Ct versus dilution plot on a 3 × 104 dilution series . Individual reactions were used to quantify each RNA level in a given cDNA sample , and the average Ct from duplicated reactions within the same run was used for quantification . The data for each gene were normalized to rp49 as an internal control and normalized to the average of all time points within a set for statistics . DpN1 cells were cultured in Grace's insect medium ( Gibco 11605–094 ) supplemented with 10% fetal bovine serum ( Gibco 26140–079 ) . The cells were maintained at 28 °C in 25-cm2 plug seal flasks ( Corning 430168 ) and split every 4 d . The high-efficiency DpN1 cell expression vector , pBA , was derived from pIE/153A ( V4+ ) vector ( Cytostore ) , where the IE1 activator gene was removed by PCR . pBA-FLAG was generated by cloning the FLAG tag into the NotI site of the multiple cloning site of the vector . DpN1 expression plasmids that were used in luciferase reporter assays ( Figure 5A ) were generated by subcloning monarch per , tim , cry1 , and cry2 into pBA-FLAG , and monarch clk and cyc into pBA . The luciferase reporter dpPer4Ep-Luc was reported previously [16] . The normalization control was generated by subcloning β-galactosidase into the pBA vector . In addition , monarch clk and cyc were subcloned into a relatively low efficiency expression pIB5 . 1 vector ( modified from the Invitrogen pIB vector , see [26] ) to bolster the luciferase reading for experiments depicted in Figure 5B . Transient transcription assays were done using 50 ng/well of dpPer4Ep-Luc as reporter and 50 ng/well pBA-β-galactosidase as normalization control . The cells were co-transfected with 50 ng/well of pBA-clk , pBA-cyc , and varying amounts of pBA-FLAG–per , –tim , –cry1 , and –cry2 . DpN1 cells were split into 12-well dishes and incubated at 28 °C for 2 d so the cultures were ∼50% confluent . Cells were then incubated in 300 μl serum-free Grace medium ( Invitrogen ) premixed with plasmids and 5 μl/well Cellfectin ( Invitrogen ) for 5 h . Grace's medium supplemented with 10% fetal bovine serum ( 700 μl ) was added at the end of transfection . The cells were then incubated for 2 d before harvesting for luciferase assay , real-time quantification , and Western blot analysis . For RNA interference ( RNAi ) experiments , dsRNAs were synthesized using the Megascript T7 transcription kit ( Ambion ) from PCR templates between 500–900 bp . Primers to generate PCR templates contain a T7 promoter at their 5′ ends , and the amplified regions correspond to cDNA locations in base pairs as: GFP ( 94–658 ) , per ( 445-1374 ) , tim ( 423–1 , 356 ) , cry1 ( 787–1 , 520 ) , and cry2 ( 311–924 ) . 20 μg of dsRNA with 10 μl of Cellfectin per well were used to transfect 12-well culture dishes as above . For dsRNA treatment of DpN1 cells kept in LD , dsRNA and Cellfectin were incubated with the cells in serum-free Grace's media for 5 h during the light phase on the second day after the cells were split . At the end of transfection , serum-containing Grace's media was added to the dish . Cells were then incubated in an LD cycle for 2 d and harvested throughout the dark-to-light transition on day four . We generated antibodies against monarch PER , TIM , and CRY2 . Purified proteins containing the C-terminal 197 amino acids of PER , or amino acids 251–450 of TIM were used as immunogens in rats and guinea pigs [41] . For monarch CRY2 , purified proteins containing the N-terminal 218 residues , the C-terminal 209 residues , or the full-length protein were used . Both affinity-purified and unpurified sera were used in Western blot , immunoprecipitation , and immunocytochemistry experiments . Representative antisera to each clock protein were affinity purified and designed as follows: PER-GP40 ( “GP” indicates raised in guinea pigs ) for the antibody against PER; TIM-GP47 for TIM; and CRY2-GP51 and CRY2-R41 ( “R” indicates raised in rats ) for CRY2 . The non–affinity purified antisera used included PER-R33 , TIM-R38 , CRY2-R42 , and CRY2-GP50 . Specificity of the affinity-purified antibodies was evaluated by the size of immunoreactive bands as determined by Western blot of extracts from heads , brains , and DpN1 cells , compared with the exogenously expressed protein in S2 cells . Specificity was further verified by showing that the band intensity of endogenous protein was reduced by specific RNAi knockdown in DpN1 cells ( see Figure 5B ) . DpN1 cells were incubated in the dark for 4 d and homogenized in extraction buffer ( 20 mM HEPES , pH 7 . 9 , 5% Glycerol , 100 mM KCl , 0 . 1% Triton X100 , 1X Complete Protease Inhibitor [Roche] ) . Monarch brains were dissected from animals frozen at ZT 18 . The photoreceptor layers of the eyes were removed , and the brains were then homogenized in the same extraction buffer . Insoluble cell debris was removed by centrifugation . Protein G sepharose beads ( GE Healthcare ) were prepared for immunoprecipitation by washing three times in the extraction buffer . The beads were then incubated with rat anti-monarch PER ( R33 ) , TIM ( R38 ) , CRY2 ( R41 ) antibodies for 1 h at room temperature . Normalized rat immunoglobulin G ( IgG ) ( Santa Cruz Biotechnology ) was used as control . The unbound antibodies were removed with an additional wash . Protein extracts were added to the beads and incubated overnight at 4 °C . The beads were washed three times with extraction buffer at 4 °C and then protein sample buffer was added . A Western blot was probed with guinea pig anti-PER ( GP40 ) , anti-TIM ( GP47 ) , and anti-CRY2 ( GP51 ) antibodies . Brain-suboesophageal ganglion complexes were dissected from CO2 anesthetized adult monarchs and processed immediately for immunocytochemistry as described earlier [30] . For examining nuclear localization of CRY2 , the sections were counterstained with specific fluorescent DNA probes ( DAPI , 1 μg/ml , 10 min at room temperature; Propidium iodide , 0 . 5 μg/ml , 10 min at room temperature; or YOYO-1 [Molecular Probes] 0 . 1 μM , 10 min at room temperature , respectively ) . Stained and mounted sections were examined using a Zeiss Axioplan 2 microscope equipped with Nomarski ( DIC ) optics , epifluorescence , and a CCD camera . The following primary antibodies and their corresponding dilutions were used: TIM-R38 ( 1:500 ) ; TIM-GP47 ( 1:1 , 000 ) ; rabbit anti-CORAZONIN ( from Makio Takeda , 1:1 , 000 ) ; CRY1-R31 ( 1:500 ) ; CRY1-GP37 ( 1:500 ) ; CRY2-R42 ( 1:200 ) ; CRY2-GP50 ( 1:200 ) ; and CRY2-GP51 ( 1:500 ) . To visualize the primary antibody binding , the following secondary antibodies were used: horseradish peroxidase-conjugated goat anti-rat ( 1:1 , 000 ) ; horseradish peroxidase-conjugated goat anti-guinea pig ( 1:1 , 000 , both from Jackson ImmunoResearch Laboratories ) ; Alexa Fluor488-conjugated goat anti-rat ( 1:200 ) ; Alexa Fluor555-conjugated goat anti-rat ( 1:400 ) ; Alexa Fluor488-conjugated goat anti-guinea pig ( 1:200 ) ; Alexa Fluor555-conjugated goat anti-guinea pig ( 1:400 ) , Alexa Fluor488-conjugated goat anti-rabbit ( 1:200 ) ; Alexa Fluor488-conjugated goat anti-mouse ( 1:200 ) ; Alexa Fluor555-conjugated goat anti-mouse ( 1:400 , all from Molecular Probes ) . To verify the specificity of immunological reactions , primary antibodies were replaced with normal goat serum . In an additional control of binding specificity , the anti-CORAZONIN antibody was pre-incubated with 100 molar excess of the original antigen prior to immunocytochemical staining . In all cases , no significant staining above background was observed . For scoring of immunoreactive intensities , stained sections were coded and viewed under a microscope . Levels of staining were subjectively scored with an intensity scale from 0–5 . The time of collection was decoded after scoring . For immunocytochemistry of CRY2 location in DpN1 cells , cells were seeded on cover slips and entrained in LD at 28 °C for 2 d . The cells were fixed at the times indicated . The cellular localization of CRY2 was assayed by immunocytochemistry using anti-CRY2 ( GP51 ) antibody and Alexa594 conjugated anti–guinea pig secondary antibody ( Invitrogen ) . The cells were also stained with SYTOX Blue ( Invitrogen ) to visualize the nucleus . For monarch cry2 , the methods were similar to those above for immunocytochemistry except that after fixation in paraformaldehyde , the tissue was embedded in paraplast and sectioned ( 10 μm ) . In situ hybridization was carried out using the mRNA locator kit ( Ambion ) . The riboprobes were localized by incubation with an alkaline phosphatase-conjugated anti-digoxigenin antibody ( Boehringer and Mannheim; 1:500 dilution overnight at 4 °C ) , and visualized with BCIP and NBT ( Perkin Elmer ) . DIG-labeled sense RNA probes were used in control experiments . In all cases , sense probes produced no signal . For generating UAS-cry1 transgenic lines , the 1 , 605-bp monarch cry1 ORF was amplified from cDNA . To generate the untagged construct , the cry1 product was cloned into the pUAST vector [42] . We created an N-terminal , myc-tagged monarch cry1 construct by cloning the cry1 PCR product into a myc-pUAST vector; the myc-pUAST vector was generated by cloning a BamHI-myc-BglII fragment , created using two oligos followed by primer extention , into the BglII site of pUAST . All constructs were sequenced . Both cry1 constructs were injected into y w;Ki pp[ry+Δ2–3]/+ embryos . For generating UAS-cry2 transgenic lines , the 2 , 229-bp monarch cry2 ORF was amplified from cDNA . To generate the untagged construct , the cry2 product was cloned into the pUAST vector . To generate the N-terminal , myc-tagged cry2 construct , the cry2 cDNA was cloned into the myc-pUAST vector . All clones were sequenced . Both cry2 constructs were injected into w1118 embryos by Genetic Services . During balancing , the w1118 X chromosome was replaced with the y w-containing chromosome . Flies were reared and experiments were conducted at 25 °C . For monarch transgene expression in cryb flies , the driver line was tim-GAL4/CyO [4] . The following lines were used: cryb ( y w; tim-GAL4/+; cryb ) , y w ( y w ) , 1a ( y w , UAS-cry1#1a/Y; tim-GAL4/+; cryb and y w , UAS-cry1#1a/y w; tim-GAL4/+; cryb ) , 6b ( y w; UAS-myc-cry1#6b/ tim-GAL4; cryb ) , 15b ( y w; UAS-myc-cry1#15b/ tim-GAL4; cryb ) , 22b ( y w; UAS-myc-cry1#22b/ tim-GAL4; cryb ) , 19a ( y w; UAS-cry2#19a/ tim-GAL4; cryb ) , 18b ( y w; UAS-myc-cry2#18b/ tim-GAL4; cryb ) , and 125a ( y w; UAS-myc-cry2#125a/ tim-GAL4; cryb ) . For light pulse/phase shift experiments , 16 males per genotype per light pulse were entrained in 12:12 LD for three full days in 120–220 lux before receiving a 1-h light pulse at 1 , 000–1 , 400 lux ( for CRY1 experiments ) or 1 , 200–1 , 600 lux ( for CRY2 experiments ) at ZT 15 or ZT 21 . [This small difference in light intensities between these two experiments was unfortunately unavoidable; we were unable to use the same incubator for both experiments , and there are enormous technical challenges in producing equivalent lux readings between incubators . Both of these experiments were performed at saturating light intensities and , thus , this difference should not affect the results . ] A “no-pulse” control group was also included . Flies were then placed in DD for 6 d . Data were collected using the TriKinetics Drosophila Activity Monitor ( DAM ) system . To identify and exclude arrhythmic flies , 5 d of activity in DD were analyzed starting 12 h after the last “lights off” using the Fly Activity Analysis Suite ( FaasX ) CYCLE_P software ( Michel Boudinot; michel . boudinot@iaf . cnrs-gif . fr ) under the following parameters: no filter for high frequencies , chi-square significance 0 . 01 . Matlab with the Signal Processing Toolbox and the FlyToolbox [43] was used to plot behavior peaks of pulsed versus nonpulsed flies . Phase shifts were determined for each genotype by taking the average delay or advance of the three peaks of activity after the light pulse . The first peak of activity directly after the light pulse was not included in the average . For Western blot samples , eight males and eight females per sample were entrained in 12:12 LD for at least two full days before collecting on dry ice . Frozen fly heads were collected into Eppendorf tubes and homogenized in 30 μl lysis buffer [150 mM NaCl , 50 mM Tris-Cl , pH 7 . 4 , 0 . 5% NP40 , 100 mM NaF , Complete Protease Inhibitor Tablet ( Roche ) ] with Kontes pestles . After centrifugation , 25 μl of the homogenate was transferred to fresh tubes . Protein concentrations were normalized by Coomassie Reagent ( Pierce ) , and either 5 or 10 μg of protein was loaded per lane ( depending on well size ) . Tubulin and dTIM/dpCRY2 were separated on the same gel and the filter cut at 75kDa . Primary antibodies were rat anti-dTIM ( 1:5 , 000 ) [6] , and monoclonal mouse anti-Alpha Tubulin ( Sigma ) ( 1:8 , 000 or 1:16 , 000 ) . Secondary antibodies ( Santa Cruz ) were goat anti-rat IgG HRP conjugated , goat anti-guinea pig IgG HRP conjugated , and goat anti-mouse IgG2a HRP conjugated . Films and chemiluminescent blots were imaged with the FUJIFILM LAS-1000 , and bands were quantified using the ImageGauge V4 . 22 software .
The GenBank ( http://www . ncbi . nlm . nih . gov/Genbank/ ) accession numbers for the monarch genes discussed in this article are period ( AY237279 ) , timeless ( AY367059 ) , Clock ( AY364477 ) , cycle ( AY364478 ) , crytochrome1 ( AY860425 ) , cryptochrome2 ( DQ184682 ) , casein kinase II α ( EF554579 ) , casein kinase II β ( EF554578 ) , shaggy ( EF554581 ) , double-time ( EF554580 ) , vrille ( AY576272 ) , Pdp1 ε ( EF649714 ) , and slimb ( EF649713 ) . | During their spectacular fall migration , eastern North American monarch butterflies ( Danaus plexippus ) use a time-compensated sun compass to help them navigate to their overwintering sites in central Mexico . The circadian clock plays a critical role in monarch butterfly migration by providing the timing component to time-compensated sun compass orientation . Here we characterize a novel molecular clock mechanism in monarchs by focusing on the functions of two CRYPTOCHROME ( CRY ) proteins . In the monarch clock , CRY1 , a Drosophila-like protein , functions as a blue-light photoreceptor for photic entrainment , whereas CRY2 , a vertebrate-like protein , functions within the clockwork as the major transcriptional repressor of the self-sustaining feedback loop . An oscillating CRY2-positive neural pathway was also discovered in the monarch brain that may communicate circadian information directly from the circadian clock to the central complex , which is the likely site of the sun compass . The monarch clock may be the prototype of a clock mechanism shared by other invertebrates that express both CRY proteins , and its elucidation will help crack the code of sun compass orientation . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Methods",
"Supporting",
"Information"
] | [
"neuroscience",
"cell",
"biology"
] | 2008 | Cryptochromes Define a Novel Circadian Clock Mechanism in Monarch Butterflies That May Underlie Sun Compass Navigation |
The sterol regulatory element-binding protein ( SREBP ) family member SREBP1 is a critical transcriptional regulator of cholesterol and fatty acid metabolism and has been implicated in insulin resistance , diabetes , and other diet-related diseases . We globally identified the promoters occupied by SREBP1 and its binding partners NFY and SP1 in a human hepatocyte cell line using chromatin immunoprecipitation combined with genome tiling arrays ( ChIP-chip ) . We find that SREBP1 occupies the promoters of 1 , 141 target genes involved in diverse biological pathways , including novel targets with roles in lipid metabolism and insulin signaling . We also identify a conserved SREBP1 DNA-binding motif in SREBP1 target promoters , and we demonstrate that many SREBP1 target genes are transcriptionally activated by treatment with insulin and glucose using gene expression microarrays . Finally , we show that SREBP1 cooperates extensively with NFY and SP1 throughout the genome and that unique combinations of these factors target distinct functional pathways . Our results provide insight into the regulatory circuitry in which SREBP1 and its network partners coordinate a complex transcriptional response in the liver with cues from the diet .
The transcription factor SREBP1 is a key regulator of the transcription of numerous genes that function in the metabolism of cholesterol and fatty acids [1] . Alternative promoter usage gives rise to two nearly identical isoforms of SREBP1—SREBP1a and SREBP1c—that differ from each other at a few N-terminal amino acids encoded by different first exons [2] . Both isoforms begin as ER membrane-bound precursors and require proteolytic cleavage to release N-terminal fragments that function in the nucleus [3] . Levels of nuclear SREBP1a increase in response to cholesterol depletion through a sterol-sensing pathway that determines the rate of cleavage [3] . In contrast , levels of nuclear SREBP1c increase in response to insulin signaling [4] , [5] and stimulation of the nuclear receptor LXRα [6]—inputs that upregulate the transcription of SREBP1c . SREBPs are weak transcriptional activators on their own and interact with their target promoters in cooperation with additional regulators , most commonly including one or both of the transcription factors NFY and SP1 [7]–[9] . Excessive accumulation of fatty acids in the liver is associated with a number of diet-related diseases such as insulin resistance , metabolic syndrome , and type 2 diabetes . The de novo synthesis of fatty acids ( lipogenesis ) in the liver is induced by the insulin signaling pathway . Numerous studies based on the analysis of single genes have demonstrated that SREBP1 upregulates the transcription of lipogenic genes , primarily through the insulin-responsive SREBP1c isoform which predominates in the liver [1] . Although these studies have identified a number of important lipogenic genes that are regulated by SREBP1 , our understanding of the targets of SREBP1 involved in lipid metabolism is incomplete and the role of SREBP1 in other pathways that may contribute to diet-related disease states is poorly understood . Furthermore , our understanding of the manner in which SREBP1 selects and regulates its target genes in combination with other transcriptional regulators is limited . In the present study , we sought to further understand the function of SREBP1 and how it operates with other regulators by globally mapping its target genes and those of its associated factors NFY and SP1 in human hepatocarcinoma ( HepG2 ) cells using ChIP-chip . In addition , we sought to correlate SREBP1 occupation with changes in gene expression induced under conditions in which SREBP1 is activated and repressed . The overall scheme is depicted in Figure 1A .
The physiological levels of hepatic SREBP1c are downregulated in the fasted state and dramatically upregulated upon refeeding with high carbohydrates [10] . Therefore , to identify the complete repertoire of hepatic SREBP1 target genes using ChIP-chip , we treated HepG2 cells with insulin and glucose—conditions that mimic the high-carbohydrate refeeding response . We then performed chromatin immunoprecipitation with a well-characterized antibody that recognizes both SREBP1 isoforms due to their nearly identical primary structure . Since both isoforms are present in the nucleus under these conditions , the binding profiles generated include targets of both SREBP1a and SREBP1c . After amplification and fluorescent labeling , the immunoprecipitated DNA was hybridized to NimbleGen human promoter tiling arrays . These arrays contain 50-mer oligonucleotide probes tiled at 100 bp resolution from 1200 bp upstream to 300 bp downstream of the transcriptional start sites of approximately 16 , 700 annotated human genes . Examples of the enrichment profiles generated at representative SREBP1 target promoters are shown in Figure 1B . Statistical analysis of four replicate experiments using a hidden Markov model approach [11] provided evidence for SREBP1 occupation at the promoters of 1141 genes ( Tables S1 and S2 ) . The interaction dataset exhibited a striking enrichment ( approximately 9-fold ) for known targets of SREBP1 ( P = 1 . 4×10−29 , hypergeometric test ) —37 genes occupied by SREBP1 in our analysis have been previously characterized as SREBP1 targets in the liver ( Table S3 ) . This subset includes 18 out of 19 represented SREBP1 target genes in the cholesterol biosynthesis pathway ( Figure 2A ) as well as known target genes involved in fatty acid biosynthesis and the pentose phosphate pathway ( a pathway that supplies the NADPH required for lipid synthesis ) . To determine the accuracy of our ChIP-chip results , we selected a random set of 19 putative SREBP1 binding sites and performed quantitative PCR using site-specific primers ( Table S4 ) . Binding was confirmed for 17 of these sites ( Figure S1 ) . We next performed an unbiased search for DNA sequence motifs representing protein-DNA interaction sites in the set of SREBP1 target sequences using the program MDScan [12] . The top-scoring motif discovered in this search ( Figure 2B ) was 10 bp in length and contained an E-box consensus CANNTG known to bind basic helix-loop-helix transcription factors such as SREBP1 [13] . Indeed , comparison of the discovered motif to the TRANSFAC [14] database using a motif comparison and alignment algorithm [15] identified the known SREBP1 consensus E-box motif as the most similar match ( Figure 2B and Figure S2 ) . To investigate the evolutionary sequence conservation of SREBP1 binding sites , we scanned the set of SREBP1-occupied promoters and a set of randomly selected promoters for matches to the discovered motif . The matching 10 bp DNA sequences identified in SREBP1 target promoters exhibited significantly higher levels of conservation across vertebrate species ( P = 3 . 0×10−3 ) based on their phastCons conservation scores [16] than matching sequences in the randomly selected promoter set ( Figure 2C ) . Thus , the binding sites and motifs identified are likely to have a functional role in vivo . To determine the fraction of SREBP1-occupied genes that is potentially transcribed in HepG2 cells , we performed ChIP-chip with an antibody that recognizes the initiating form of RNA polymerase II . Our results revealed that 43% ( n = 7070 ) of the genes on the promoter array were occupied by RNA polymerase II ( Figures S1 and S2 ) . As expected , the vast majority of SREBP1 target genes ( 88% , n = 1007 ) were also occupied by RNA polymerase II ( Figure 2D ) . To explore the cellular processes that are regulated by SREBP1 , we examined the functional categories associated with SREBP1 targets genes based on annotations in the Gene Ontology ( GO ) [17] database and manual curation and identified enriched GO categories using the BiNGO program [18] ( Figure S3 and Table S5 ) . Genes associated with lipid metabolism were highly enriched in the interaction dataset ( P = 1 . 5×10−4 ) , consistent with the known functions of SREBP1 . Inspection of these genes revealed 41 novel SREBP1 targets ( Table S6 ) , considerably expanding the repertoire of SREBP1-regulated genes involved in lipid metabolism . One important target that we identified in this category , ADIPOR2 , encodes the primary liver receptor for the insulin-sensitizing adipocytokine adiponectin [19] . Intriguingly , ADIPOR2 expression is inversely regulated by insulin in hepatocytes [20] , suggesting that SREBP1 may play a repressive role at the ADIPOR2 promoter . Other important targets include endothelial lipase ( LIPG ) , a key player in HDL metabolism [21] , ethanolamine kinase 1 ( ETNK1 ) , an enzyme that catalyzes the first committed step in phosphatidylethanolamine synthesis , and sterol O-acyltransferase 2 ( SOAT2 ) , an enzyme responsible for cholesterol esterification in the liver [22] . Further investigation of the functions of novel SREBP1 target genes suggested that SREBP1 plays a more expansive role in a number of other processes in which it has been previously implicated . For example , SREBP1 is known to regulate the expression of IRS2 [23] and PIK3R3 [24] , components of the insulin signaling pathway . We identified promoter interactions with several additional components of this pathway , including AKT1S1 , AKT2 , GRB2 , GSK3A , MEKK2 , PRKCQ ( also known as PKCθ ) , and S6K1 . These interactions suggest that SREBP1 is involved in complex feedback regulation of insulin signaling and implicate SREBP1 in additional branches of the insulin signaling pathway such as the regulation of protein synthesis . Additionally , recent studies have identified SREBP1a as a transcriptional activator of the cyclin-dependent kinase inhibitor gene CDKN1A ( also known as p21 ) —an interaction that likely provides a crucial link between the synthesis of membrane lipids and cell proliferation [25] . Our findings support the role of SREBP1 in linking these processes , as 60 genes associated with cell cycle regulation based on GO annotation were identified among SREBP1 targets . Important cell cycle regulators occupied by SREBP1 include ABL1 , CCNG2 , CDC25C , CDK2 , and E2F4 . Finally , the targets of SREBP1 were enriched ( P = 1 . 1×10−3 ) for genes involved in cellular respiration ( e . g . CYCS , IDH3G , SUCLA2 , SDHB , and MDH1 ) , providing insight into the role of SREBP1 in energy metabolism and expanding on the known regulatory interaction of SREBP1 with the mitochondrial aconitase gene ACO2 [26] . The list of SREBP1 target genes was also enriched for genes in a variety of GO categories that have not previously been associated with SREBP1 , including RNA processing ( P = 1 . 1×10−11 ) , protein transport ( P = 8 . 1×10−6 ) , and DNA metabolism ( P = 3 . 0×10−2 ) . Taken together , our results suggest that SREBP1 may be involved in coordinating the activity of a wide range of cellular pathways with nutrient availability at the transcriptional level . In the liver , the expression of many known SREBP1c target genes is downregulated in the fasted state when SREBP1c levels are reduced and upregulated upon high carbohydrate refeeding when SREBP1c levels increase [10] . To determine whether the putative direct targets of SREBP1 identified by ChIP-chip are differentially expressed in the fasted and refed states , we performed gene expression analysis . Levels of mRNA were compared between HepG2 cells treated with insulin and glucose ( in the same manner as cells used for ChIP-chip analysis ) or with forskolin and dexamethasone—molecules that reproduce the counterregulatory effects of glucagon and cortisol in the fasted state . The gene expression changes induced by these treatments were consistent with those observed in the fasted and refed liver in vivo ( Table S7 ) . To compare the list of SREBP1 targets with the gene expression dataset , the expression dataset was rank-ordered by fold change such that the most highly-induced genes in the refed state were at the top of the ranked list . We observed a marked enrichment of SREBP1 targets among the top ranking genes within this ranked list ( Figure 3A ) . To determine whether this biased distribution of SREBP1 targets was statistically significant in comparison to the distribution expected at random , we used Gene Set Enrichment Analysis ( GSEA ) [27] . GSEA revealed that the enrichment of SREBP1 targets at the top of the ranked list was statistically significant ( P<10−3 ) and identified a subset of 555 SREBP1-occupied genes that accounted for this effect ( Figure 3A and Table S8 ) . To highlight the most strongly induced SREBP1 targets in this subset , all SREBP1-occupied genes exhibiting greater than 1 . 75-fold induction in the refed state ( 71 total ) are shown in Figure 3B; notably , 31% of these genes are known targets of SREBP1 . These observations strongly indicate that transcriptional activation is the predominant effect of SREBP1 binding . Although SREBP1 binding sometimes results in gene repression , as previous studies have demonstrated for a few genes such as IRS2 [23] and PCK1 [28] , repressed targets are less frequent than induced targets and fail to exhibit significant enrichment in our dataset . Importantly , our results demonstrate that a substantial proportion of the direct targets of SREBP1 identified by ChIP-chip are activated by insulin and glucose in the refed state . To begin to investigate the regulatory networks in which SREBP1 operates , we performed location analysis with antibodies that recognize the transcription factors NFY and SP1—two key partners involved in the control of many SREBP1 target genes [7]–[9] . These ChIP-chip experiments were also performed in HepG2 cells treated with insulin and glucose as described above for SREBP1 and revealed that NFY and SP1 occupied the promoters of 1707 and 1641 genes , respectively ( Tables S1 and S2 ) . Comparison of these genes with the set of SREBP1-occupied targets revealed a high degree of overlap ( Figure 4A ) ; 32% of SREBP1 targets were occupied by NFY and 34% were occupied by SP1 . In total , 48% of SREBP1 targets were occupied by at least one other transcription factor . The regulatory circuitry among SREBP1 , NFY , and SP1 was highly interconnected ( Figure 4B ) . For example , SP1 participates in multicomponent loops with both SREBP1 and NFY , an arrangement wherein two factors occupy each other's promoters . In addition , all three factors exhibited autoregulation , a hallmark of master regulators of key cellular processes [29] . Finally , these factors bind upstream of many other transcription factor genes , including the SREBP family member SREBP2 , suggesting that SREBP1 , NFY , and SP1 are involved in transcriptional regulatory cascades that can operate on additional target genes and pathways in an indirect manner . We next used Gene Ontology to investigate the functions of genes targeted by different combinations of SREBP1 and its associated factors . This type of analysis is important for expanding our currently limited understanding of how transcriptional regulatory networks are organized into functional modules in mammalian genomes . Identification of enriched GO categories using BiNGO [18] revealed that different combinations of these regulators display distinct preferences for genes involved in certain biological pathways ( Figure 4C and Table S9 ) . In some cases , these pathways were targeted by a unique combination of regulators . For example , genes involved in lipid metabolism , including important lipogenic genes such as stearoyl-CoA desaturase 1 , were preferentially occupied by the combination of SREBP1 and NFY ( P = 2 . 4×10−3 ) . Likewise , genes involved in carbohydrate metabolism ( P = 1 . 5×10−2 ) and cellular respiration ( P = 2 . 5×10−2 ) , key nutrient-regulated pathways , were enriched among targets of SREBP1 alone , indicating that SREBP1 may cooperate with regulators other than NFY and SP1 at the promoters of these genes . In some cases , however , one biological pathway was targeted by multiple combinations of regulators . For example , genes encoding cholesterol biosynthetic enzymes were enriched among targets of all three regulators ( P = 3 . 0×10−6 ) and among targets of both SREBP1 and NFY ( P = 3 . 9×10−4 ) . This type of configuration may reflect differences among genes in the same pathway , such as their timing or responsiveness to different stimuli . Additional representative enriched categories for each combination of regulators are displayed in Figure 4C . It will be of interest to determine how SREBP1 participates in the control of different biological pathways in combination with a wider array of transcriptional regulators in the liver by performing location analysis for additional factors , especially those that participate in nutrient-regulated pathways , such as SREBP2 , LXRα , PPARα , and ChREBP . In conclusion , these studies present the first global analysis of targets of an important set of regulators involved in mammalian lipid metabolism . The targets of SREBP1 are involved in a wide array of cellular processes suggesting new and expanded roles for this transcription factor . In addition , many of these targets are differentially regulated in the fasted and refed states , suggesting that they respond to changes in the diet . Finally , we find that SREBP1 and its associated factors NFY and SP1 form an interconnected regulatory circuit and bind in a combinatorial manner to functionally distinct sets of target genes .
HepG2 cells were routinely cultured in DMEM supplemented with 10% FBS and 100 U/ml penicillin/streptomycin . For ChIP-chip and gene expression profiling , cells were cultured in glucose-free DMEM with 0 . 5% BSA for 12 h then treated with either insulin ( 100 nM ) and glucose ( 10 mM ) or with forskolin ( 1 μM ) , dexamethasone ( 1 μM ) , and pyruvate ( 2 mM ) for 6 h in glucose-free DMEM with 0 . 5% BSA . ChIP was performed as described with modifications [30] . Briefly , for each ChIP reaction , ∼5×107 HepG2 cells were crosslinked for 15 min at room temperature , washed twice in cold PBS and swollen for 15 min on ice in hypotonic buffer . Nuclei were then pelleted and resuspended in lysis buffer for 30 min on ice followed by sonication using a Branson 450 Sonifier set at 50% amplitude . Samples were sonicated 8 times for 20 sec with 1–2 min on ice between pulses . Extracts were clarified by centrifugation at 14 , 000 rpm for 15 min at 4°C , pre-cleared with 50 μl of Protein A/G beads ( Pierce ) for 1 hr at 4°C , then incubated overnight with 10 μg of the appropriate specific or negative control antibody at 4°C . Antibodies used were rabbit anti-SREBP1 ( sc-8984 ) , rabbit anti-NFYA ( sc-10779 ) , rabbit anti-SP1 ( sc-59 ) , and normal rabbit IgG ( sc-2027 ) obtained from Santa Cruz , and mouse anti-RNA Pol II ( MMS-126R ) obtained from Covance . Antibody-bound complexes were then captured by incubation with 50 μl of Protein A/G beads . Beads were washed once with RIPA buffer ( 150 mM NaCl ) , once with high-salt RIPA buffer ( 500 mM NaCl ) , twice with LiCl detergent , and once with TE . Antibody-bound complexes were then eluted by incubation with 100 μl of elution buffer ( 100 mM NaHCO3/1%SDS ) for 15 min with gentle vortexing followed by a second 15 min elution with 150 μl of elution buffer . The eluates were combined and treated with RNaseA ( 20 μg/ml ) for 1 hr at 37°C then incubated overnight at 65°C to reverse crosslinks . Whole cell extract DNA was also treated for crosslink reversal . Samples were then digested with Proteinase K ( 20 μg/ml ) for 2 h at 37°C and DNA was purified by one extraction with phenol , one extraction with phenol/chloroform/isoamyl alcohol and one extraction with chloroform/isoamyl alcohol followed by ethanol precipitation and Qiagen PCR purification . Purified DNA was blunt-ended , ligated to universal linkers , and amplified using ligation-mediated PCR . Amplified DNA was then labeled with Cy3 or Cy5 dyes and hybridized to NimbleGen human promoter tiling arrays ( array design “2005-04-18_HG17_min_promoter_set” ) according to the manufacturer's protocol ( NimbleGen Systems of Iceland ) . Four biological replicate hybridization datasets ( provided in Dataset S1 ) were generated for transcriptional regulators ( ChIP ) , negative control rabbit IgG ( control ) , and unenriched ( input ) DNA using an unpaired reference design . For SREBP1 , NFY , and SP1 , log2-transformed probe intensities were quantile normalized across the set of IP , control , and input replicate groups . For RNA Pol II , log2-transformed intensities were quantile normalized within groups then scaled to have a median feature intensity of 10 . Normalized intensities were analyzed using a two-state Hidden Markov Model ( HMM ) in the program TileMap [11] . Briefly , TileMap first computes a test-statistic for every probe based on a hierarchical empirical Bayes model then uses HMM to combine information from neighboring probes and calculate the posterior probability that a probe is in the enriched state [11] . The enriched state was defined as: ( ChIP>control ) AND ( ChIP>input ) . Enriched regions were selected by taking a posterior probability cutoff of 0 . 90 and merging probes separated by less than 500 bp . Regions less than 100 bp in length or containing fewer than 5 probes were discarded . A summary enrichment score was calculated for each region by taking the maximum probe value of mean ( ChIP ) -mean ( input ) . Regions were matched to the closest RefSeq transcripts within 1500 bp on each strand based on the refGene table ( hg17 , May 2004 assembly ) downloaded from the UCSC Table Browser ( http://genome . ucsc . edu/ ) [31] . Entrez GeneIDs were assigned to transcripts via the refLink conversion table ( hg17 , May 2004 assembly ) . Quantitative PCR with site-specific primers was performed in duplicate on a BioRad MyiQ real-time PCR cycler with BioRad iQ SYBR Green supermix . Primers were designed for 19 randomly selected SREBP1 target promoters and for a negative control unbound region in exon 4 of the GAPDH gene . Normalized Ct ( ΔCt ) values for each sample were calculated by subtracting the Ct value obtained using input DNA from the Ct value obtained using SREBP1 ChIP DNA ( ΔCt = CtChIP–Ctinput ) . Fold enrichment was then calculated using the formula 2− ( ΔCt[target]−ΔCt[GAPDH] ) . For the purpose of estimating the specificity of the SREBP1 ChIP-chip dataset , promoters with an average qPCR fold enrichment greater than 2 were considered enriched . Primer sequences are listed in Table S4 . The motif discovery program MDScan [12] was used to identify candidate protein-DNA interaction motifs in the SREBP1 dataset . Input to the MDScan program consisted of 200bp sequences centered on the probe exhibiting maximum enrichment in each bound region . The top-scoring motif from MDScan was queried against the TRANSFAC database using the STAMP web server ( http://www . benoslab . pitt . edu/stamp ) [15] to identify the best matching known motifs . Results are presented in Figure S2 . The DNA consensus sequence logos in Figure 2B were generated with the WebLogo tool ( http://weblogo . berkeley . edu/ ) . Statistical analysis of the enrichment of Gene Ontology categories ( ‘Biological Process’ branch ) was performed using BiNGO ( http://www . psb . ugent . be/cbd/papers/BiNGO/ ) [18] . Enrichment was determined in reference to all Entrez GeneIDs annotated in the Biological Process branch ( 13 , 460 genes total ) . P-values are derived from a hypergeometric test followed by Benjamini and Hochberg false discovery rate correction . A P-value cutoff of 0 . 05 was used to identify significantly enriched categories . Total RNA from 5×106 fasted or refed HepG2 cells was collected using Trizol ( Invitrogen ) and further purified using a Qiagen RNeasy kit . Purified total RNA from two biological replicate experiments was then submitted to the Affymetrix Resource at the Yale W . M . Keck Foundation Biotechnology Resource Laboratory for labeling and hybridization to Affymetrix human genome U133 Plus 2 . 0 arrays . The robust multiarray averaging method [32] in the Bioconductor affy package [33] was used to generate expression measures from probe-level data . Updated probe set definitions were used to map expression data to the EntrezGene database [34] . For heatmap display , log2-transformed gene expression measures were ranked by expression difference ( mean[refed]-mean[fasted] ) , mean centered , and visualized using Java Treeview ( http://jtreeview . sourceforge . net/ ) . The complete expression dataset is summarized in Table S7 . To visualize the distribution of SREBP1 target genes within the ranked gene expression dataset , we generated a running overlap plot by sliding a 2000 gene window down the ranked list and calculating the percent overlap between the two datasets in each window . The significance of the association of SREBP1 target genes with the top of the ranked list was assessed using Gene Set Enrichment Analysis ( GSEA ) software [27] . Briefly , a maximum enrichment score ( ES ) was derived for the SREBP1 target gene set by taking the maximum deviation from zero of a running-sum statistic computed by walking down the ranked list and increasing the running-sum statistic when an SREBP1 target gene was encountered and decreasing it otherwise ( as detailed in ref . [27] ) . The magnitude of the maximum ES reflects the degree to which a gene set is overrepresented at the extremes of the ranked list . The ranked list was then randomized 1000 times , calculating the maximum ES for each permutation in order to derive a null distribution . The nominal P-value for the SREBP1 target gene set ( P<10−3 ) represents the fraction of scores in the null distribution that are at least as high as the observed maximum ES . The set of SREBP1-bound DNA sequences ( 500 bp regions centered at the peak of ChIP enrichment ) was scanned for matches to the PWM discovered by MDScan using the PatSearch web program ( http://www . ba . itb . cnr . it/BIG/PatSearch/ ) with a similarity threshold of 0 . 80 . The same procedure was performed for an equally-sized set of 500 bp regions selected at random from the promoters represented on the promoter array . The maximum 8-way phastCons conservation score [16] of each match in SREBP1-bound and random regions was then obtained using the Galaxy web server ( http://main . g2 . bx . psu . edu/ ) . Scores from the SREBP1-bound and random regions were compared using a Wilcoxon rank-sum test . | Transcription factors ( TFs ) are DNA-binding proteins that regulate the transcription of their target genes . TFs typically bind in proximity to the transcription start sites of their target genes in a region called the promoter . SREBP1 is a TF that increases the transcription of numerous genes involved in cholesterol and fat metabolism and has been linked to diet-related diseases such as insulin resistance and type 2 diabetes . Using microarray technology , we identified all of the promoters in the human genome that are bound by SREBP1 and two associated TFs called NFY and SP1 in a human liver cell line . Our findings greatly expand the number of genes and biological pathways that may be regulated by SREBP1 and reveal that different combinations of SREBP1 and its partners preferentially target genes involved in different pathways . Thus , in contrast to traditional studies that focus on individual genes , we have used a genomics approach to provide a novel global view of the regulatory circuitry in which SREBP1 and its partners coordinate a transcriptional response in the liver with cues from the diet . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/functional",
"genomics",
"computational",
"biology/transcriptional",
"regulation",
"genetics",
"and",
"genomics",
"genetics",
"and",
"genomics/bioinformatics",
"diabetes",
"and",
"endocrinology/type",
"2",
"diabetes"
] | 2008 | Genome-Wide Occupancy of SREBP1 and Its Partners NFY and SP1 Reveals Novel Functional Roles and Combinatorial Regulation of Distinct Classes of Genes |
Dengue is a mosquito-borne viral illness that causes a variety of health outcomes , from a mild acute febrile illness to potentially fatal severe dengue . Between 2005 and 2010 , the annual number of suspected dengue cases reported to the Passive Dengue Surveillance System ( PDSS ) in Puerto Rico ranged from 2 , 346 in 2006 to 22 , 496 in 2010 . Like other passive surveillance systems , PDSS is subject to under-reporting . To estimate the degree of under-reporting in Puerto Rico , we built separate inpatient and outpatient probability-based multiplier models , using data from two different surveillance systems—PDSS and the enhanced dengue surveillance system ( EDSS ) . We adjusted reported cases to account for sensitivity of diagnostic tests , specimens with indeterminate results , and differences between PDSS and EDSS in numbers of reported dengue cases . In addition , for outpatients , we adjusted for the fact that less than 100% of medical providers submit diagnostic specimens from suspected cases . We estimated that a multiplication factor of between 5 ( for 2010 data ) to 9 ( for 2006 data ) must be used to correct for the under-reporting of the number of laboratory-positive dengue inpatients . Multiplication factors of between 21 ( for 2010 data ) to 115 ( for 2008 data ) must be used to correct for the under-reporting of laboratory-positive dengue outpatients . We also estimated that , after correcting for underreporting , the mean annual rate , for 2005–2010 , of medically attended dengue in Puerto Rico to be between 2 . 1 ( for dengue inpatients ) to 7 . 8 ( for dengue outpatients ) per 1 , 000 population . These estimated rates compare to the reported rates of 0 . 4 ( dengue outpatients ) to 0 . 1 ( dengue inpatients ) per 1 , 000 population . The multipliers , while subject to limitations , will help public health officials correct for underreporting of dengue cases , and thus better evaluate the cost-and-benefits of possible interventions .
Dengue is a mosquito-borne viral illness that represents a major public health problem in all tropical and subtropical countries . Dengue incidence has increased an estimated 30-fold from 1962 to 2012 , and two-fifths of the world’s population is thought to be at risk for dengue [1 , 2] . It has been recently estimated that the global incidence of dengue is between 50 and 100 million cases per year [3] . Approximately 15% of the worldwide burden of dengue occurs in the Americas [4] . Dengue is endemic in Puerto Rico and has been a reportable condition for several decades [5] . Incidence of medically attended , clinically suspected cases of dengue has been monitored in Puerto Rico , since the late 1960s , by the Passive dengue surveillance system ( PDSS ) . This system is operated by the Puerto Rico Department of Health ( PRDH ) and Centers for Disease Control and Prevention Dengue Branch ( CDC-DB ) [5–7] . In recent years , PDSS recorded dengue epidemics in 1998 , 2007 , 2010 and 2012–2013; the latter with 29 , 386 suspected dengue cases , of which approximately 54% were laboratory confirmed as dengue [8 , 9] . However , like all passive surveillance systems , PDSS is subject to under-reporting of medically attended cases . Previous estimates indicated that for every suspected medically attended dengue case reported to PDSS , 10–27 additional cases occurred but were not reported [10–13] . Reasons for under-reporting included the need to obtain an acute and convalescent serum specimen to make an accurate laboratory diagnosis . This was difficult to implement and resulted in a substantial portion of suspected dengue cases being classified as indeterminate . In addition , non-hospitalized medically attended ( i . e . , outpatient ) dengue cases were reported less frequently than hospitalized cases , and the frequency of reporting varied by region , type of healthcare facility and healthcare provider . Estimating the degree of dengue under-reporting from standard passive surveillance systems is critical for making robust estimates of the true disease burden . Several approaches have been used to estimate the degree of dengue under-recognition and/or under-reporting . Wichmann et al . used childhood cohort studies to estimate the degree of under-reporting for national surveillance systems in Thailand and Cambodia [14] . A capture-recapture approach was used by Dechant et al to estimate the degree of under-reporting for inpatients by comparing passive surveillance data to data submitted by infection control nurses in Puerto Rico during 1991–1995[10] . Similarly , others have compared inpatient and emergency room hospital administrative data to surveillance data [15 , 16] or used expert opinion to estimate under-reporting of hospitalized dengue cases [17 , 18] . Since 2007 , reporting by infection control nurses in Puerto Rico has ceased , and there is a need to update the multiplier estimates . In this analysis , we estimate the degree of under-estimation of medically attended dengue cases in Puerto Rico . Our model gives both point estimates of the multiplier , the uncertainty around them , and includes the effect on underreporting due to test sensitivity and number of indeterminate cases . We also provide separate estimates for inpatients and outpatients . Our estimates of multipliers will help public health officials correct for underreporting of dengue cases and thus enable them to evaluate better the cost-and-benefits of possible interventions .
This study underwent institutional review at the CDC and was determined to be public health practice ( evaluation of an ongoing surveillance system ) and not research; as such , we obtained a waiver from CDC’s Institutional Review Board approval . We estimated the degree of under-reporting of medically attended dengue cases in Puerto Rico by building a spreadsheet-based probability ( Monte Carlo ) multiplier model . The probability simulations were run using a spreadsheet add-in program ( @Risk 7 . 0 , Palisade , Ithaca , NY ) . We adapted models used to calculate the under-recognized impact of foodborne illness and pandemic influenza in the United States [19 , 20] . We used data from two different surveillance systems in Puerto Rico: the Passive Dengue Surveillance System ( PDSS ) , and Enhanced Dengue Surveillance System ( EDSS ) ( see description under Data Sources ) . The model had separate modules for inpatients and outpatients ( Fig 1 ) . Using our model , for each module we calculated adjustments of the medically attended dengue case counts to account for five factors that could affect the degree of underreporting . These factors are: the sensitivity of the dengue diagnostic tests ( Multiplier A ) ; the number of submitted specimens not tested due to inadequate volume or incomplete information ( used to derive Multiplier B ) ; an estimate of the proportion of indeterminate dengue diagnostic test results considered as positives ( Multiplier C ) ; and differences in rates of reported laboratory-positive dengue cases between PDSS and the EDSS ( Multiplier D ) . In the outpatient module , we also adjusted for the less than 100% submissions from medical providers of diagnostic specimens collected from suspected cases ( Multiplier E; Fig 1 ) . We list the values for each multiplier in Table 1 ( see later for full description ) . We further allowed for the fact that two different laboratory tests are used to test serum specimens from patients suspected of having dengue ( see later for details ) . We corrected for different probabilities of false negative readings ( i . e . , which causes underreporting ) associated with each type of test . For those patients tested using both tests , if one test result was positive and the other negative , we considered the patient as being “positive . ” We used , to obtain corrected estimates of the number of medically attended cases , the previously described multipliers in the following general formulae: Corrected hospitalized positive casesYear X , Test Y = Number of reported hospitalized positive casesYearX , TestY x ( ( 1/Multiplier A ) x ( 1/Multiplier B ) x ( Multiplier C ) x Multiplier D ) Corrected outpatient casesYear X , Test Y = Number of reported outpatient positive casesYearX , TestY x ( ( 1/Multiplier A ) x ( 1/Multiplier B ) x ( Multiplier C ) x Multiplier D x ( 1/Multiplier E ) ) Where Year X refers to the number of reported cases in a given year , and Test Y refers to the type of laboratory-based test ( see later for details ) . The value for Multiplier A depends upon which test was used , while the values for Multipliers C and D depended upon which patient classification/ ( initiation ) sub-model was used ( Table 1 and see later for details ) . We defined indeterminate cases as specimens from both hospitalized and outpatient medically attended cases that had either no detected dengue RNA and/or no available specimen collected a week or later after illness onset to test for presence of anti-dengue virus antibodies . We used Multiplier C to estimate the percentage of indeterminate specimens that would likely have tested positive if they had sufficient dengue RNA and/ or had been collected a week after illness onset . We calculated the value of Multiplier C using data from CDC-DB databases indicating the percentage of submitted specimens ( with sufficient serum to allow complete testing ) that tested positive from both inpatients and outpatients ( see details later ) . Given this methodology for indeterminate specimens ( only ) , we set Multipliers A and B each to a value of 1 . The remainder of the calculations to determine the corrected number of positives for the indeterminate specimens was the same as for the other specimens ( i . e , used Multipliers D for inpatient specimens , and Multipliers D and E for outpatient specimens–see earlier ) . To clarify , for specimens with sufficient dengue RNA and/ or had been collected more than a week after illness onset , Multiplier C was set to 1 ( i . e , no correction for indeterminate cases ) . We conducted a sensitivity analysis to determine which of the individual multipliers ( A-D ) were relatively the most important in determining the overall multiplier ( as defined in the equations , earlier ) . We used the software to create a tornado graph of the 2010 inpatient multiplier ( MA sub-model ) . The tornado graph plots out the partial correlation coefficients between the Overall Multiplier and the probability distributions defining the individual multipliers . To examine the potential impact of laboratory-based false positive test results , we also altered the assumed specificity values of both the RT-PCR and IgM ELISA laboratory tests . We used the previously described data from 2010 concerning reported cases , inpatients and outpatients . We recalculated the multipliers after we reduced the specificity of both types of tests to an arbitrary 80% , well below the reported specificities of 100% and 96 . 7% ( see earlier ) .
The tornado graph shows that the relatively most important individual multiplier is rate of reported dengue: PDSS vs EDSS ( Multiplier D ) ( Fig 3 ) . The next most important individual multiplier was Multiplier C ( Percent positive dengue among cases reported as intermediates ) . However , the partial correlation coefficient between Multiplier C and the overall Multiplier was much smaller than that for Multiplier D , indicating a relatively small effect between changes in Multiplier C and the Overall Multiplier . When we tested for the potential impact of false-positive test result on our estimates of multipliers , we reduced the specificities of the laboratory-based test to 80% ( from the baseline values of 100% for RT-PCR and a mean of 96 . 7% for the anti-DENV IgM ELISA ) . For outpatients , the estimated multipliers for 2010 increased from 74 . 55 for the MA model and 21 . 41 for the DO model ( Table 3 ) to 76 . 27 and 21 . 92 , respectively . For inpatients , the estimated multipliers for 2010 increased from 5 . 25 for the MA model , and 5 . 01 for the DO model ( Table 2 ) to 5 . 35 and 5 . 10 , respectively . We concluded that all of these results illustrate that false positive test results are unlikely to increase notably the size of our estimated multipliers .
We estimated that a multiplication factor of between 5 ( for 2010 data ) to 9 ( for 2006 data ) must be used to correct for under-reporting the number of laboratory-positive dengue inpatients reported to the Puerto Rican PDSS . Multiplication factors of between 21 ( for 2010 data ) to 115 ( for 2008 data ) must be used to correct for under-reporting of laboratory-positive dengue outpatients . We also estimated that the mean annual rate of medically attended dengue in Puerto Rico to be between 2 . 1 ( for dengue inpatients ) to 7 . 8 ( for dengue outpatients ) per 1 , 000 population . These estimated rates compare to the reported rates of 0 . 4 ( dengue outpatients ) to 0 . 1 ( dengue inpatients ) per 1 , 000 population . The estimates were most sensitive to the difference between PDSS and EDSS in numbers of reported , laboratory-positive cases , and how we identified potential clinical cases of dengue . Use of the more permissive Medically Attended ( MA ) classification resulted in notably higher estimates of overall multipliers for outpatient dengue cases than estimates obtained when using the more restrictive DCIF-Only ( DO ) patient classification . Previously published estimates of multipliers for dengue inpatients ranged from 1 . 4–3 . 4 for four countries in the Americas ( including Puerto Rico ) and 1 . 8–2 . 5 for five South East Asian countries [17 , 18] . In comparison , our estimates for dengue inpatients were higher , ranging from 5 to 9 . Interestingly , our inpatient estimate is comparable with a multiplier found for laboratory-positive dengue deaths in Puerto Rico found using an enhanced surveillance system . In that study [32] the mortality rate was 2–3 times higher than detected previously under the passive surveillance system . One would expect fatal cases to be more readily identifiable and reported than for inpatient dengue cases . For outpatients , the previously published estimates ranged between 1–28 for four countries in the Americas ( including Puerto Rico ) and 5 . 0–29 . 8 for Southeast Asian countries [17 , 18] . The upper end of that range is similar to our range of estimates of 21–33 for outpatient cases , which we calculated using the DO sub-model ( Table 3 ) . Our estimates from the outpatients MA sub-model , however , were notably higher , at 75–115 ( Table 3 ) . We are only aware of one study that had similar multiplier estimates for outpatient cases . That a study was from Malaysia , and was based on expert opinion that there were 65 . 6 additional cases for each identified outpatient case [18] . This wide variability of estimated underreporting indicates that there is little meaningful correlation between the relative amount of resources available to public health and clinical health care systems and the ability of such systems to accurate capture the number of outpatient and inpatient dengue case . The reason for such lack of definitive correlation remains to be explained . This study has several limitations . We assumed that the rate of positivity among those with both acute and convalescent specimens was the same as those who had only an acute specimen submitted ( i . e . , indeterminate cases ) . We also assumed that the rate of dengue in EDSS areas ( Guayama and Patillas ) was representative for all of Puerto Rico , which may not be accurate . When calculating rates of dengue from the PDSS system , we used all cases reported in the PDSS system for the entire island of Puerto Rico . This method may mask a degree of heterogeneity in rates of cases on the island . Further , the health care seeking behaviors of patients for whom two serum specimens were submitted ( one while acutely ill , another while convalescent , necessary for complete diagnosis using the IgM-based test ) may be different than behaviors of those for whom had only one specimen submitted . Correcting for these limitations would require an analysis in nearby municipalities , with high and low incidence of dengue , or cases from within the EDSS municipalities that were not reported through the EDSS . The resources for such a level of surveillance were not available . The estimated expansion factors for 2005–2010 were quite consistent year-to-year . Thus , we have no reason to believe that estimates of multipliers for later years would be notably different than those shown in Table 2 . The EDSS in Patillas ended in 2012 . Dengue inpatient EDSS in Guyama continued , and enhanced surveillance was implemented in 2012 at Saint Luke's Episcopal Hospital in Ponce municipality . However , following the arrival of chikungunya and Zika viruses to Puerto Rico in 2014 and 2016 , respectively [30 , 31] , it is unclear how case reporting or rates of laboratory-confirmation of dengue may be affected . Additional investigation may therefore be necessary to define the rates of both under-reporting of clinically-apparent illness caused by these pathogens . There is also the potential for correlation between some of the multipliers . For example , doctors that reported cases of dengue may work in areas where there were more laboratory-tested cases of febrile illnesses . There thus could be a negative correlation between Multipliers D and E . We do not have any evidence , however , that doctors who reported cases of dengue in the areas studied were more likely to be practicing in areas that have a higher likelihood of dengue cases among patients with acute febrile illnesses . In addition , in a given locale dengue incidence varies over time , due to the buildup of naturally acquired immunity in individuals and thus increased herd immunity . For example , towns and municipalities that were severely impacted in the 2007 dengue outbreak were less likely to be impacted in the 2010 dengue outbreak . Further , many of the physicians in Puerto Rico work at least part-time in large referral hospitals and see patients from several municipalities with varying dengue incidence that changes over time . Thus , doctors who report dengue are actually likely to encounter patients from a variety of locales with varying risks of acquiring dengue . Finally , as mentioned earlier , there is not a complete concurrence in which dengue patients were entered into the two systems . We know of no way to correct this problem , or to correct for the problems related to having only two EDSS sites . These problems are not unique to Puerto Rico . Many countries that have passive dengue surveillance systems with non-specific case definitions , and have often lacked comprehensive dengue-based surveillance ( 14 , 15 , 18 ) . We maintain that , even with these problems , the multipliers presented here are sufficient to allow public health officials to usefully correct for under-reporting . This study augments existing evidence showing that reporting of severe cases of dengue , which are more likely to be admitted to hospitals , is more accurate than for less severe outpatient cases . Our results demonstrates the benefits of an enhanced dengue surveillance system , which can supplement passive dengue surveillance systems in countries with resource constraints . As noted earlier , in situations where there are insufficient data to construct probability distributions for each multiplier , our methodology can be applied using only point estimates ( single values ) for each multiplier . Our results illustrate the need for , and thus potential benefits of , frequently using our methodology to estimate the degree of under-reporting in passive dengue systems during epidemic and non-epidemic years .
The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention | The number of global cases of dengue has increased an estimated 30-fold from 1962 to 2012 , and two-fifths of the world’s population are thought to be at risk for dengue . It has been recently estimated that the global incidence of dengue is between 50 and 100 million cases per year . These estimates of burden and impact are , however , are not considered very reliable . It has been previously established and reported that there is notable under-reporting of clinical cases of dengue , even those who sought medical treatment . This includes under-reporting of those hospitalized with laboratory-confirmed dengue . This lack of reliable estimates hampers efforts of public health officials in determining the of burden of disease and the costs-and-benefits of potential interventions . We estimated that multiplication factors ranging from 5 to 9 must be used to correct for under-reporting of laboratory-positive dengue inpatient cases reported to public health officials in Puerto Rico . Multiplication factors ranging from 21 to 115 must be used to correct for the underreporting of laboratory-positive dengue outpatients . Our results illustrate the need for , and thus potential benefits of , using our methodology to estimate the degree of under-reporting in passive dengue systems during epidemic and non-epidemic years . | [
"Abstract",
"Introduction",
"Methods",
"and",
"materials",
"Results",
"Discussion",
"Disclaimer"
] | [
"medicine",
"and",
"health",
"sciences",
"enzyme-linked",
"immunoassays",
"geographical",
"locations",
"tropical",
"diseases",
"health",
"care",
"north",
"america",
"probability",
"distribution",
"inpatients",
"mathematics",
"neglected",
"tropical",
"diseases",
"infectious"... | 2018 | Estimating dengue under-reporting in Puerto Rico using a multiplier model |
Although a previous study predicted that Japanese encephalitis virus ( JEV ) originated in the Malaysia/Indonesia region , the virus is known to circulate mainly on the Asian continent . However , there are no reported systematic studies that adequately define how JEV then dispersed throughout Asia . In order to understand the mode of JEV dispersal throughout the entire Asian continent and the factors that determine the dispersal characteristics of JEV , a phylogenetic analysis using Bayesian Markov chain Monte Carlo simulations was conducted on all available JEV E gene sequences in GenBank , plus strains recently isolated in China . Here we demonstrate for the first time that JEV lineages can be divided into four endemic cycles , comprising southern Asia , eastern coastal Asia , western Asia , and central Asia . The isolation places of the viruses in each endemic cycle were geographically independent regardless of years , vectors , and hosts of isolation . Following further analysis , we propose that the southernmost region ( Thailand , Vietnam , and Yunnan Province , China ) was the source of JEV transmission to the Asian continent following its emergence . Three independent transmission routes from the south to north appear to define subsequent dispersal of JEV . Analysis of JEV population dynamics further supports these concepts . These results and their interpretation provide new insights into our understanding of JEV evolution and dispersal and highlight its potential for introduction into non-endemic areas .
Japanese encephalitis ( JE ) is arguably one of the most serious viral encephalitic diseases worldwide [1] , [2] . According to the latest report of the World Health Organization , JE is endemic in 24 Asian and Oceanian countries , with an estimated 67 , 900 JE cases annually ( the total morbidity rate is 1 . 8/100 , 000 population ) . An estimated 3 billion people live in countries where JE is endemic . Additionally , with increased international travel to JE endemic areas , more people are at risk of JE infection . Therefore , JE is not only an endemic disease in Asian and Oceanian countries , it could also potentially cause significant public health issues in non-endemic countries or regions and has the realistic possibility of becoming a serious global public health problem [1]–[4] . JEV is the prototype member of the JEV serogroup within the genus Flavivirus , family Flaviviridae . The viral genome is a positive-sense , single-stranded RNA that is approximately 11 kb in size . The genome carries a single open reading frame ( ORF ) encoding a polyprotein that is processed into three structural proteins [capsid ( C ) , membrane ( M ) , and envelope ( E ) ] and seven nonstructural proteins ( NS1 , NS2A , NS2B , NS3 , NS4A , NS4B , and NS5 ) [5] , [6] . Phylogenetic analysis of JEV has shown that based on the E gene or the complete genome , JEV can be divided into five genotypes ( G1–G5 ) [7] , [8] , [9] . JEV is maintained in nature in a cycle involving vertebrate hosts ( including pigs and waterbirds ) and Culex mosquitoes . Culex tritaeniorhynchus is the primary vector [10] . Pigs are important reservoir hosts of JEV . Migrating birds are thought to be an important factor in the dispersion of JEV to new geographical areas [11] . During the virus transmission cycle , mosquitoes become infected with JEV when they feed on infected pigs and birds . They replicate the virus and subsequently feed again , in some cases transmitting the virus to humans or horses which are incidental hosts of JEV . JEV was predicted to have originated from the tropical Indonesia/Malaysia region because there is evidence that this region had all genotypes of JEV circulating , whereas only the more recent genotypes circulate in other areas [12] . However , the virus is currently known to circulate throughout Asia . These observations raise several questions . Firstly , what is the pattern and direction of JEV dispersal from the Indonesia/Malaysia region to the entire Asian continent ? Secondly , what are the primary factors that determine the dispersal characteristics of JEV ? Thirdly , what is the contribution to virus dispersal of migratory birds , seasonal winds , mosquitoes and other factors , such as temperature and rainfall ? Resolution of these intriguing questions will not only inform science but will also provide guidance for public health authorities in the development of prevention and control strategies for JE . Previous reports on transmission patterns showed that JEVs in East Asia were introduced from South East Asia [13] , and JEVs circulating in Japan were introduced from South East Asia and continental East Asia [14] . However , these studies were limited to local areas . Therefore , in the present study , together with the new JEV strains isolated in China , we analyzed all available sequences of the currently predominant genotype ( G1 ) of JEV isolates that are widely dispersed over the Asian continent .
The E protein gene of 22 JEV strains newly isolated in China from 2005 to 2010 was sequenced . Among these 22 strains , two were isolated from Yunnan Province in 2005 and 2006; four were isolated from Guangxi Province in 2006 , Henan Province in 2006 , Shanxi Province in 2006 , and Jiangxi Province in 2009 , respectively; two were isolated from Liaoning Province in 2006 and 2007; three were isolated from Shandong Province in 2008 and 2009; three were isolated from Chongqing Municipality in 2008 and 2009; and eight were isolated from Hubei Province during 2008–2010 . The isolation protocols have been described elsewhere [8] . Briefly , the viruses were amplified once by infecting Aedes albopictus C6/36 mosquito cells . After development of cytopathic effects ( CPE ) , culture supernatants were harvested , and viral RNA was extracted using the QIAamp Viral RNA Mini Kit ( Qiagen , Hilden , Germany ) . The purified RNA was used as the template for cDNA synthesis using Ready-to-Go You-Prime First-Strand beads ( Amersham Biosciences , Piscataway , NJ , USA ) . The complete E gene was amplified using the following primers: JEV-Ef [5′-TGYTGGTCGCTCCGGCTTA-3′ ( 955–973 ) ] and JEV-Er [5′-AAGATGCCACTTCCACAYCTC-3′ ( 2516–2536 ) ] . Amplified products were examined by agarose gel electrophoresis ( 1% ) , purified using a QIAquick Gel Extraction Kit ( Qiagen ) and then sequenced directly . The envelope sequences of these 22 newly isolated JEVs determined in the present study were deposited in GenBank under Accession Numbers JQ937333 and JQ937336–JQ937356 ( Table S1 ) . The most recent study showed that in evolutionary terms , the G1 JEV genotype is the youngest of five genotypes and is the dominant genotype circulating in Asia [15] . For studying the dispersing patterns of G1 JEV , totally 656 E sequences of JEV , with information regarding the isolation time and place , were downloaded from GenBank ( as of May 1 , 2012 ) . ClustalX version 2 . 0 . 9 [16] was used to generate sequence alignments of 678 E gene sequences ( including 22 newly contributed sequences ) . The dataset was screened for recombination using RDP3 ( Recombination Detection Program3 ) and GARD ( genetic algorithm for recombination detection ) [17] , [18] . No recombination events were identified ( data not shown ) . Subsequently , in order to differentiate G1 from the other four genotypes , the Neighbor-joining method in Mega Version 5 . 05 [19] was applied for phylogenetic analysis . Finally , 359 E sequences of the G1 JEV genotype were obtained and comprised a sequence database for phylogenetic analysis ( Table S1 ) . The sequence database constructed above was analyzed using Bayesian Markov chain Monte Carlo ( MCMC ) method . The GTR+I+G was selected as the optimal nucleotide substitution model by MrModelTest [20] . The nucleotide substitution rate and divergence time of the most recent common ancestor ( TMRCA ) were estimated using the relaxed ( uncorrelated lognormal ) molecular clock model under a coalescent model of constant population size in the BEAST software package [21] . Demographic histories were inferred by Bayesian skyline reconstruction . The analysis was run through 600 , 000 , 000 generations to ensure sufficient mixing . Convergence of parameters was checked using TRACER and was indicated as effective sample size ( ESS>200 ) , and the maximum clade credibility ( MCC ) tree was built using TreeAnnotator with 10% burn-in ( http://beast . bio . ed . ac . uk/ ) . Statistical uncertainty was expressed for nodal support by 95% confidence intervals of the highest posterior density ( HPD ) . In order to infer the history of geographical dispersion of JEV , the Bayesian stochastic search variable selection ( BSSVS ) was used to provide evidence for statistically supported diffusion between state variables under BEAST v1 . 7 . 5 [22] . This method estimates the most probable state at each node in the MCC trees , allowing us to reconstruct ancestral positions for ancestral viral lineages along the tree . For phylogeographic reconstructions , each region was coded as a discrete trait . BSSVS output and surfaces representing uncertainty for continuous diffusion processes were formatted as KML using the SPREAD utility [23] . Determination of each locality was coordinated and performed using Google Earth v . 6 . 2 . 2 . Mapinfo was finally used to display the dispersal pattern of JEV based on the phylogeographic analysis .
Based on the isolation sources of 359 G1 JEV strains , by 2010 , the distribution of the G1 JEV genotype had shifted northward to a latitude of 45° ( Japan ) and westward to a longitude of 75° ( India ) and had covered almost all JEV endemic areas , including Australia , Thailand , Vietnam , Cambodia , India , Japan , Korea , Taiwan , and most regions of China ( 15 provinces ) . The Malaysian G1 JEV sequence was not included in the present study because only the PrM sequence was available . The 22 newly isolated JEVs contributed to knowledge regarding the geographical distribution of the G1 genotype in central Asia , a highly endemic area for JEV and added to strains isolated in China after 2005 , but especially after 2009 . A time span of 44 years was found for the isolation time of the 359 strains studied , from 1967 , when the first G1 JEV genotype was isolated , to 2010 . Additionally , the viral strains used were from various sources , including insect vectors ( various mosquito species and midges ) and host animals ( pigs and human patients ) . The maximum clade credibility ( MCC ) tree of the 359 E sequences of JEV is shown in Figure 1 . The tree showed that the G1 JEV genotype isolates can be divided into seven clusters according to their geographic isolation sites ( designated clusters 1–7 ) . They were further grouped into four lineages , lineage I ( clusters 1 and 2 ) , II ( clusters 3–5 ) , III ( cluster 6 ) and IV ( cluster 7 ) . According to the geographic locations of the principal JEV strains in each lineage , four lineages were also designated as the southern Asia endemic cycle , the eastern coastal Asia endemic cycle , the western Asia endemic cycle , and the central Asia endemic cycle , respectively . Virus strains in the southern Asia endemic cycle were mostly derived from Vietnam , Thailand , Cambodia , Australia , and Yunnan Province in China . Strains in the eastern coastal Asia endemic cycle were mainly derived from Shanghai , Zhejiang , Liaoning , Shandong , Taiwan , Japan , and Korea; strains from southernmost Yunnan Province and those isolated after 2008 in Chongqing , Hubei , and Jiangxi of central China ( such as JX0939 , SZ18 , JL18 , ES57 , HBZG0907 , and HBZG0809 ) are also included in this endemic cycle . Most of the strains in the western Asia endemic cycle were obtained from Tibet and India in western Asia , and other strains were isolated in southernmost regions such as Thailand and Vietnam or Japan . The dominant strains in the central Asia endemic cycle were from the inland provinces in China , including Guizhou , Sichuan , Chongqing , Henan , Hubei , Gansu , and Shanxi . This latter cycle also contains a few strains from the southernmost part of Asia such as Vietnam and Yunnan Province in China , and the eastern coastal regions of Asia such as Shandong , Taiwan , and Guangxi in China , Japan , and South Korea . The most recent common ancestor ( TMRCA ) for the G1 JEV genotype is estimated to have diverged approximately 78 years ago based on the Bayesian MCMC approach using E gene sequences . The resultant G1 JEV genotype first appeared in southernmost Asian regions , such as Thailand , Vietnam , and Yunnan Province in China and established endemic cycles in those regions . Detailed analysis of the geographical locations of strains in the four endemic cycles showed that at least one strain of JEV isolated from southernmost regions of Asia , including Thailand , Vietnam , and Yunnan Province in China was present in eastern coastal Asia , western Asia , and central Asia endemic cycles ( Figure 1 , Table 1 ) . For example , isolates from Yunnan Province in China were included in the eastern coastal Asia endemic cycle , isolates from Thailand and Vietnam were included in the western Asia endemic cycle , and isolates from Vietnam and Yunnan Province in China were included in the central Asia endemic cycle . Moreover , based on the chronological order of evolution , the strains isolated from the southernmost regions mostly occurred earlier than others and rooted those in each endemic cycle . However , those strains from other regions of Asia were found in endemic cycles consistent with their places of isolation . Furthermore , homology analysis revealed that nucleotide homology of strains isolated from the southern Asia endemic region was about 94% . However , strains from endemic regions of the eastern Asian coast and central Asia shared more than 96% and 97% homology . Thus , all data above indicate that viruses isolated from southernmost regions of Asia maintained the diversity of virus populations . In addition , strains isolated from Thailand , Vietnam , and Yunnan Province in China in the same endemic cycle , have a wide span of isolation time . For example , a virus strain isolated in 1979 in Thailand and strains isolated later in 2005 in Thailand were included in the southern Asia endemic cycle . Moreover , the eastern coastal Asia endemic cycle contained strains isolated in 1982 in Yunnan Province in China and others isolated later in 2005 in Yunnan Province; also the western Asia endemic cycle contained strains isolated in Thailand in 1992 and later in 2005; and the central Asia endemic cycle contained strains isolated in Vietnam in 2001 and others isolated in 2007 . Thus , in addition to the diversity of virus populations , viruses isolated from the southernmost regions of Asia also maintained the stable genetic characteristics of JEV . This implies that the southernmost region of the Asian continent plays a key role in transmission of JEV from its origin to the Asian continent , providing a source for continental dispersal of JEV strains . The estimated history of JEV dispersal in endemic regions is shown in detail in Figure 2 . The maps in Figure 2 display dispersal characteristics over time . According to our reconstructions , G1 JEV was initially introduced to Thailand and Vietnam located in southernmost Asia during the 1970s after originating in Malaysia/Indonesia ( Figure 2 , 1970 ) . It then dispersed to Japan and Shanghai , located in east coastal Asia around 1980 ( Figure 2 , 1978 , 1981 ) . Subsequently , the virus was introduced to Sichuan province located in Central China from east coastal Asia around 1990 ( Figure 2 , 1990 ) . Simultaneously , it spread to India located in the West part of Asia . In 2000 , two lineages were dispersed to east coastal Asia ( Zhejiang , Liaoning , Taiwan , South Korea ) and to the Chinese inland provinces ( Henan , Hubei , Gansu and Guizhou ) , respectively ( Figure 2 , 2000 ) . After 2000 , in addition to continuing its dispersion in east coastal Asia and in Chinese inland provinces , a lineage from Yunnan was introduced to the eastern coastal areas ( Figure 2 , 2003 , 2006 ) . On the other hand , a lineage from Vietnam was introduced to inland China and also from Thailand to Yunnan ( Figure 2 , 2006 ) . Around 2009 , the G1 genotype appears to have dispersed from India to Xizang ( Figure 2 , 2010 ) . This virus dispersal history supports our contention that the southernmost regions of the Asian continent acted as the source of JEV prior to its transmission throughout the Asian continent . Furthermore , by combining the divergence time and geographical distribution characteristics of each cluster ( Table 1 ) within the MCC tree , three different routes from southern to northern Asia were postulated for dispersal of G1 JEV ( Figure 3 ) Through the eastern route , JEV dispersed from Thailand and Yunnan Province in China , to Japan , South Korea , and Shanghai , Zhejiang , and Liaoning in China generating the eastern coastal Asia endemic cycle . Through the western route , JEV dispersed from Thailand and Vietnam to the western regions of Asia , reaching India and Tibet , establishing the western Asia endemic cycle . Through the central route , JEV was transmitted from southern countries such as Vietnam , to central Asia , and reached inland provinces of China including Sichuan , Chongqing , Guizhou , Hubei , Henan , Gansu , and Shanxi . In addition , JEV was introduced to inland China provinces from bordering east coastal regions and established the largest central Asia endemic cycle . A skyline plot of the G1 JEV genotype population dynamics is shown in Figure 4 . There was minimal fluctuation during the first half of the plot . This was followed by a major population increase from 1980 to 1990 , a relatively stable period from 1991 to 2003 , a marked decrease during 2004–2007 , and then a relatively stable period after 2008 . According to the time nodes of change in population dynamics identified in the skyline plot , the isolation times of JEV in the four endemic cycles were analyzed ( Table 2 ) . Viruses were all isolated from the southernmost regions of Asia ( Thailand , Cambodia , and Yunnan Province in China ) before 1990 during the major period of population increase . Subsequently , from 1991 to 2003 , a period of population stability , G1 JEV was found in eastern coastal and southernmost regions of Asia , such as Shanghai and Liaoning in China , Japan , and South Korea . Although there was a virus population decline from 2004 to 2007 , the dominant G1 JEV genotype continued its expansion to the central Asian areas . From 2008 , the G1 JEV genotype dispersed to all endemic regions in the entire Asian continent and maintained a relatively stable population . JEV continued to be isolated in the southernmost regions of Asia during the entire fluctuation process of virus populations from the first occurrence . In conclusion , all the data are consistent with the concept that the southernmost regions of the Asian continent played a key role as the source for evolution and dispersal of the G1 JEV genotype .
Based on phylogenetic and phylogeographic analysis of the envelope gene of G1 JEV , the following conclusions were drawn from this study . 1 ) Southernmost Asia , particularly Thailand , Vietnam , and Yunnan Province in China , appear to represent the source for the continental dispersion of JEV which appears to have originated from the Indonesia/Malaysia region; 2 ) During the dispersal of G1 JEV , limited introductions were observed among difference geographic locations; and 3 ) Reverse dispersion from north to south occurred during the more recent years . The rationale behind the concept that southernmost Asia might be the source of JEV G1 can be explained as follows . Lying in the tropics and subtropics , these southern regions have a high annual average temperature and heavy annual rainfall both of which are particularly suitable conditions for high population density breeding of a wide range of mosquitoes [24] . C . tritaeniorhynchus , the predominant transmission vector for JEV , is widely distributed in Thailand and Vietnam with estimated coverage of 80 . 9% and 60% , respectively [10] . Additionally , following the remarkable increase in the acreage of irrigated rice in recent years , the distribution of C . tritaeniorhynchus has further expanded in Thailand and Vietnam which are traditional rice exporting countries [1] . Furthermore , swine husbandry in these regions has developed rapidly with the number of farmed pigs increasing by 100% from 1990 to 2005 [1] . In conclusion , the combination of a suitable climate year-round , wide distribution of the primary vector and abundant pig farming , provide the ideal prerequisite for the emergence , maintenance and reproduction of JEV in these southernmost regions of Asia . In addition to these factors , examination of the recognized flight paths of migratory birds reveals a remarkable coincidence between the eastern , central , and western routes of JEV dispersal patterns and the recognized eastern , central , and western flight paths of migratory birds in Asia ( Figure 3 ) . In previous studies [11] , [12] , migratory birds were shown to be important hosts for introducing JEV to new territories . The black-crowned night heron ( Nycticorax nycticorax ) , plumed egret ( Egretta intermedia ) , and little Egret ( Egretta garzetta ) are the main birds that carry and spread JEV [11] , [25] . These observations are consistent with the opinion that migratory birds carrying JEV migrate from south to north annually whereas non-migratory or resting birds are fed upon by mosquitoes in their local habitats . JEV is then further dispersed by local mosquitoes acquiring JEV from the infected birds and transmitting it to domestic pigs which then amplify the virus and provide a local source of infection when local mosquitoes feed on the infected pigs . After a period of evolution and dispersal , dominant JEV populations establish endemic cycles in mosquitoes , pigs and other hosts providing the opportunity for JEV epidemics in local areas . However , taking into account current expert opinion , global climate change could influence the migration patterns or routes of birds , resulting in JEV dispersal into new local areas or even new continents . It is therefore essential to extend JEV surveillance in migrating birds , mosquitoes and pigs in areas currently considered to be free of JEV , including Europe and the New World , where closely related viruses such as West Nile virus ( Europe and the Americas ) and Usutu virus ( Europe ) are already known to have established endemic cycles [25] , [26] . It was also reported that mosquitoes can be carried very long distances on the wind , especially during the typhoon season [27] , [28] . It is therefore conceivable that wind-blown mosquitoes may play a significant role in the dispersal of JEV . Also , based on studies of the G1 JEV genotype , it is evident that although changes in JEV population dynamics throughout the Asian continent were recorded , a relatively stable state of JEV populations , and wide distribution throughout the Asian continent , was also observed . It therefore seems reasonable to propose that in the absence of adequate control strategies for JEV , the maintenance and widespread distribution of current dominant JEV populations provides a foundation for further expansion of JEV into traditionally non-endemic areas . In the early and mid-20th century , JEV has caused many pandemics in Asian countries , such as Japan , China and India [2] , [29] , [30] , resulting in huge health , social and economic burdens . With global warming and the increase in acreage of irrigated rice , more Asian regions provide habitats ideal for breeding of Culex tritaeniorhynchus , the primary vector of JEV [9] . Additionally , with the changes in pig farming practices and increasing international trade and personnel exchanges , JEV is being provided with numerous opportunities to disperse northwards and westwards to regions outside Asia [31] . For example , a range of Culex species is distributed in Europe , such as Culex pipiens pallens and Culex bitaeniorhynchus . These are recognized transmission vectors of JEV [32] . Moreover , pigs are raised in large areas of north-western European countries [33] . Therefore , if JEV is constantly being introduced to these regions , via infected migratory birds or transportation of infected mosquitoes , during the summer season , the possibility exists for JEV to become established and to form an endemic cycle among the local Culex mosquitoes and pigs . Additionally , whilst many viral encephalitic cases in Europe and Asia are known to have been caused by tick-borne encephalitis virus in regions adjacent to Europe [34] , JE incidence is currently considered to be extremely low or non-existent in these regions [3] . However , JEV has been detected in specimens collected in the 1990s in the Wooded Steppe region of Northern Eurasia [35] and in avian tissues collected in Russia ( GenBank Accession Number AF501313–15 ) . Therefore , it is worth considering the possibility that at least in some cases of encephalitis in Russia and western and central Europe , the aetiological agent may be JEV rather than tick-borne encephalitis . Whilst such a proposal would possibly have been ridiculed a few years ago , several pathogenic mosquito-borne flaviviruses have emerged in Europe during the past decade and recent reports suggest the possibility of the presence of a G3 JEV genotype in birds in Italy [36] , [37] . In conclusion , there is clearly a need for a coordinated system of surveillance throughout Europe and Asia in the hope of identifying potential threats of emergence of JEV in new territories . It is worth noting that , although questions on the transmission pattern of G1 JEV were answered to a certain degree through this study , other questions still require answers . For example , what factors were primarily responsible for the dispersion from Malaysia/Indonesia to Thailand/Yunnan ? Why is the JEV G1 genotype no longer detectable in Malaysia/Indonesia ? What is the situation regarding the other JEV genotypes ? Do they have similar transmission patterns ? Further studies will be needed to answer these questions . Our study was based on the available strains . It remains to be seen if , when more data are available , subsequent studies support or revise our conclusions . | Japanese encephalitis virus ( JEV ) probably originated in the Malaysia/Indonesia region . Currently , there are no systematic studies that adequately define how it subsequently dispersed throughout Asia . In this study , we demonstrate that JEV lineages can be divided into four endemic cycles , comprising southern Asia , eastern coastal Asia , western Asia , and central Asia . In each endemic cycle the source of virus was geographically independent regardless of year , vector , and host of isolation . The southernmost region ( Thailand , Vietnam , and Yunnan Province , China ) was identified as the most likely source of JEV transmission from its origin to the Asian continent . Based on the evidence , we identified three probable JEV dispersal routes from south to north . Analysis of JEV population dynamics further supports this view . Our results provide new insights into the understanding of JEV evolution and dispersal and highlight its potential for introduction into non-endemic areas . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2013 | Southernmost Asia Is the Source of Japanese Encephalitis Virus (Genotype 1) Diversity from which the Viruses Disperse and Evolve throughout Asia |
Cyclin Y family can enhance Wnt/β-catenin signaling in mitosis . Their physiological roles in mammalian development are yet unknown . Here we show that Cyclin Y-like 1 ( Ccnyl1 ) and Cyclin Y ( Ccny ) have overlapping function and are crucial for mouse embryonic development and mammary stem/progenitor cell functions . Double knockout of Ccnys results in embryonic lethality at E16 . 5 . In pubertal development , mammary terminal end buds robustly express Ccnyl1 . Depletion of Ccnys leads to reduction of Lrp6 phosphorylation , hampering β-catenin activities and abolishing mammary stem/progenitor cell expansion in vitro . In lineage tracing experiments , Ccnys-deficient mammary cells lose their competitiveness and cease to contribute to mammary development . In transplantation assays , Ccnys-deficient mammary cells fail to reconstitute , whereas constitutively active β-catenin restores their regeneration abilities . Together , our results demonstrate the physiological significance of Ccnys-mediated mitotic Wnt signaling in embryonic development and mammary stem/progenitor cells , and reveal insights in the molecular mechanisms orchestrating cell cycle progression and maintenance of stem cell properties .
Stem cell self-renewal is tightly associated with cell cycle progression . In particular , active stem cells rapidly divide and coordinate with cell fate choices [1] . Active stem cells can be replenished by quiescent stem cells over time or upon injury [2] . Distinct from other somatic cell proliferation , the process of self-renewal needs to ensure stemness maintenance . The molecular mechanism overseeing stemness maintenance during division is poorly understood . Wnt/β-catenin signaling plays a prominent role in adult stem cell self-renewal in many tissues [3] . The level of Wnt/β-catenin signaling in stem cells is meticulously regulated , while different activation levels of the signaling result in distinct fate decisions by stem cells [4–6] . Wnt signaling is initiated upon the binding of Wnt ligands to Frizzled and lipoprotein receptor-related proteins 5 and 6 ( LRP5/6 ) receptors . Consequent abolishing or titrating away of the APC/Axin/GSK3 complex leads to β-catenin accumulation and translocation to the nucleus where it binds to Tcf/Lef family members , activating target gene expression [3] . A key step for stoichiometric regulation of Wnt signaling happens at the membrane , where Lrp6 receptor activation occurs in sequential steps before reaching its full competence . It is well established that Lrp6 is phosphorylated on PPPSP motif by GSK3 and on CK1 site by CK1γ [7 , 8] . Recently , an additional phosphorylation event of Lrp6 has been identified to precede Wnt ligand stimulation [9] . During mitosis , cyclin Y ( Ccny ) localized at the plasma membrane recruits Cyclin-dependent kinase 14 ( Cdk14 ) for the phosphorylation of Lrp6 on PPPSP motif , which sensitizes Lrp6 for upcoming Wnt signals [9 , 10] . The finding of Lrp6 phosphorylation by Ccny/Cdk14 reveals a cell cycle dependent Wnt signaling activation mechanism , adding a new level of complexity to stoichiometric Wnt signaling activation [9 , 11] . Although enhancing the Wnt-receptor Lrp6 competence by Ccny is important for Xenopus development [9] , whether a similar Ccny/Lrp6 regulatory event is physiologically significant in mammals and in stem cell biology is unknown . The mammary gland is a bi-layered epithelial organ consisting of an inner layer of luminal cells and an outer layer of basal cells ( myoepithelial cells ) . Mammary stem/progenitor cells are Wnt-responsive cells resided in the basal layer [12–14] . The mammary gland develops mostly in the postnatal stage . At the onset of puberty , at around 3 weeks of age in mouse , a rapid expansion of the rudimentary ductal tree begins . Branching ductal morphogenesis proceeds across the entire mammary fat pad , and is completed at approximately 7 weeks [15 , 16] . During puberty , at the growing tips of the ducts are the highly proliferative terminal end buds ( TEBs ) , which are believed to house active mammary stem cells ( or transient amplifying cells ) , actively cycling to fuel the growth spurt [17 , 18] . This notion is further supported by a lineage tracing study in which labeled TEB cells undergo massive clonal expansion , giving rise to ample progeny cells in development [13] . In this study , we applied a genetic approach to investigate the roles of Ccny and its paralogue Ccnyl1 in active stem/progenitor cells and to examine their functions in mammary development and regeneration . Our study establishes that mitosis-induced and Ccny family-mediated Wnt signaling activity is essential for keeping the developmental potential of dividing mammary stem/progenitor cells , shedding light in the molecular mechanisms coordinating cell cycle progression and undifferentiated state maintenance .
We first investigated the expression patterns of Ccny and Ccnyl1 , hereafter referred to collectively as Ccnys . We found that both Ccnys , which share high similarity in amino acid sequence ( S1A Fig ) , are expressed in many tissues , including the mammary gland ( Fig 1A ) . We generated Ccny and Ccnyl1 polyclonal antibodies and validated their specificity ( S1B–S1D Fig ) . Cell fractionation and Western analyses indicated membrane localization of Ccnyl1 , similar to that of Ccny ( Fig 1B ) [10] . To investigate the function of Ccny , we generated Ccny conditional mutant mice , with two loxP sites inserted to flank exon 4 ( Fig 1C and see Methods for details ) . To create Ccny+/- mutant mice , male Ccnyflox/+ mice were bred with female EIIa-cre mice , inducing recombination in germ cells and transmitting the Ccny deletion to progeny . The resulting Ccny+/- mice went through additional cross to generate Ccny-/- mice ( Fig 1C ) . We confirmed the deletion of Ccny in various tissues by western analyses , and the deletion of Ccny did not affect the level of Ccnyl1 ( Fig 1D ) . The Ccny-/- mice were grossly normal and their mammary glands displayed no discernable phenotypes ( S2 Fig ) , suggesting a possible compensation by Ccnyl1 . To address this , we utilized a Ccnyl1 knock-in mouse line ( Ccnyl1lacZ/+ ) generated in EUCOMM , in which a LacZ cassette was inserted into the intron between exon 4 and 5 ( Fig 1E ) . Although the insertion disrupted the Ccnyl1 transcription , Ccnyl1lacZ/lacZ mice were viable and also exhibited normal mammary gland morphology ( S2 Fig ) . The deletion of Ccnyl1 was also validated in multiple tissues while the level of Ccny was not affected ( Fig 1F ) . Interestingly , the Ccnys double knockout mice ( Ccny-/-;Ccnyl1lacZ/lacZ; DKO ) were embryonic lethal ( Fig 1G ) . At E14 . 5 , Ccnys DKO embryos appeared smaller in body size yet alive ( Fig 1H ) . At E16 . 5 , the DKO embryos harvested were lethal , infiltrated with blood and partially absorbed by the uterus ( Fig 1I ) . Together , these data suggest that Ccny and Ccnyl1 have overlapping functions in development . As neither single mutant displays discernable mammary gland phenotype , functional redundancy likely persists during mammary development . We examined the expression of Ccnyl1 in the mammary gland using Ccnyl1lacZ/+ mouse . Mammary glands were isolated from pubertal mice ( 5-week and 6-week old ) for whole mount X-gal staining . At this stage , mammary epithelium undergoes active extension . Interestingly , Ccnyl1 expression was enriched at the forefront of the pubertal mammary epithelium extension where TEBs are located ( arrows in Fig 2A and 2B ) . Ccnyl1 expression appeared mostly in basal cells and surrounding stromal cells , but rarely in the inner layer body cells ( Fig 2C ) . It has been reported that several members of the Wnt family are expressed in the mammary gland at this stage [19–21] , which could contribute to the proliferative state of TEBs . We examined the Wnt-responsiveness in pubertal mammary glands using Axin2lacZ/+ reporter mouse [22] . We found that Axin2-expressing cells are also enriched in the TEB area ( Fig 2D and 2E ) , with robust staining in the basal cells and surrounding stromal cells ( Fig 2F ) , exhibiting a similar pattern to the Ccnyl1-expressing cells . To address whether Ccnyl1 is expressed in Axin2+ cells , double in situ hybridization were performed , revealing that Axin2 and Ccnyl1 frequently co-localized in basal cells of the TEBs ( S3A Fig ) . We next investigated whether Ccny expression also has a TEB enriched pattern . We harvested mammary glands from 5-week-old Actin-GFP mice , in which the forefront of the epithelium has extended slightly past the lymph node . Guided by the green fluorescence of GFP , we separated the TEB region from the ducts ( illustrated in Fig 2G ) . Basal ( Lin- , CD24+ , CD29hi ) and luminal ( Lin- , CD24+ , CD29lo ) cells were isolated by FACS from the two compartments for quantitative PCR ( qPCR ) analysis . We found that Ccny was evenly expressed in the ducts and TEBs , with little difference between luminal and basal cells ( Fig 2H ) . By contrast , Ccnyl1 exhibited a higher expression in TEBs , especially in the basal cell of TEBs ( Fig 2H ) , consistent with the observation in the Ccnyl1lacZ/+ reporter mice ( see Fig 2A–2C ) . Double colored RNA in situ hybridization was then performed to validate Ccny and Ccnyl1 expression in TEBs . We found that , consistent with the qPCR results , Ccny mRNA was detected in both basal and luminal cells , whereas Ccnyl1 mRNA was predominantly distributed in basal cells ( Fig 2I ) . In 8-week-old nulliparous mice , the mammary gland has ceased rapid proliferation and the TEB structure has vanished . At this stage , we detected very rare Ccnyl1 expression in mature mammary ducts ( S3B Fig ) , similar to the Axin2-lacZ expression pattern at this stage ( S3B Fig ) [12] . Thus , Ccnyl1 is robustly expressed in the basal cell of TEBs , coinciding with Wnt/β-catenin signaling activation . In light of the overlapping expression of Ccnyl1 and Axin2 in pubertal mammary gland , we set to address whether the expression of Ccnyl1 is induced by Wnt/β-catenin signaling . We cultured the basal cells in 3D matrigel as previously described [12] and found that neither Wnt3A nor Wnt4 ( the endogenous Wnt in the mammary gland ) was sufficient to induce Ccnyl1 or Ccny expression , while either treatment successfully increased Axin2 mRNA levels ( Fig 3A ) . A gradient of lithium chloride ( LiCl ) was also used to activate Wnt signaling , yet it failed to stimulate Ccnyl1 or Ccny expression ( Fig 3B ) . Thus , Ccnys are likely not Wnt signaling targets . Previous study indicates that Ccny is enriched in G2/M phase [9] . We thus examined whether Ccnyl1 level is regulated in a cell cycle dependent manner . We monitored endogenous Ccnyl1 level in drug synchronized mammary epithelial Eph4 cells and observed a prominent elevation of Ccnyl1 in M phase ( Fig 3C ) . For insights into the upstream regulation of Ccnyl1 , we examined possible binding sites for transcriptional factors in the Ccnyl1 promoter region . The cell cycle related transcription factors E2Fs and proliferation related transcriptional factors , Egr1 and Elk1 , were predicted to associate with the Ccnyl1 promoter . We tested the potency of various transcription factors in activating Ccnyl1 transcription by luciferase reporter assays . The -2 . 3kb to +0 . 7kb genomic region of Ccnyl1 was cloned to drive a luciferase reporter gene . We found that E2F1 is able to activate the Ccnyl1 promoter in Eph4 cells ( Fig 3D ) . When similar luciferase assays were performed using the Ccny promoter , however , E2Fs could not activate Ccny-luciferase activities ( Fig 3E ) . Furthermore , chromatin immunoprecipitation ( ChIP ) analysis confirmed direct association of E2F1 with the Ccnyl1 genomic region containing a putative E2F1 binding site ( Fig 3F ) . Together , these results suggest that Ccnyl1 expression in mammary cells is cell cycle regulated . Axin2+ cells are enriched for basal mammary stem/progenitor cells [2] . In light of the similar expression patterns of Ccnyl1+ and Axin2+ cells , we next investigated the function of Ccnys on mammary stem/progenitor cells . We utilized the Ccny mutant mice ( Ccny-/- ) and knocked down the expression of Ccnyl1 by shRNA in the Ccny-/- background . The knockdown efficiency of the shRNA ( Sh-Ccnyl1 ) was validated by Western analysis ( S4 Fig ) . Basal cells were FACS-isolated from Ccny-/- mammary glands and infected with control or Sh-Ccnyl1 lentivirus in suspension . Infected mammary cells were then cultured in 3D Matrigel to allow colony formation . We found that the colony sizes were significantly smaller when the Ccnys expression was inhibited ( Fig 4A ) , with decreased cell proliferation as shown by reduced EdU incorporation ( Fig 4B ) . Next we dissociated the primary colonies to single cells and replated them to examine serial colony formation . We found that knockdown of Ccnyl1 in Ccny-/- mutant cells results in drastically decreased colony numbers in each passage , while the control colony numbers continuously expanded ( Fig 4C ) . We also examined whether Ccnys affect luminal cell colony formation using the same approach in the Ccny-/- background . No differences in luminal colony sizes were observed when Ccnyl1 was knocked down ( S5 Fig ) . Together , these results show that Ccnys are important for the expansions of basal , but not luminal , colonies in vitro . We next investigated the influence of Ccnys on the regenerative capacity of mammary stem/progenitor cells . We followed the in vivo knockdown method established in previous studies [23 , 24] . Mammary cells were isolated and infected with lentivirus to express both shRNA and GFP . Infected cells were FACS-isolated using the GFP tag , followed by transplantation into the cleared fat pad of immunocompromised recipient mice . The mammary outgrowths with GFP were examined at 8 weeks post transplantation ( Fig 4D ) . In control experiments , we observed that Ccny+/-;Ccnyl1lacZ/+ mammary cells ( loss of 2 copies of Ccnys ) infected with a scramble shRNA readily generated outgrowths ( Fig 4E ) . Ccny-/-;Ccnyl1lacZ/+ mammary cells ( loss of 3 copies of Ccnys ) infected with the scramble shRNA also generated new mammary glands though their outgrowths were smaller ( Fig 4E ) . By contrast , Ccny-/-;Ccnyl1lacZ/+ cells infected with Sh-Ccnyl1 ( loss of 3 copies of Ccnys plus RNAi ) completely lost the regeneration capabilities and were not able to reconstitute any outgrowths ( Fig 4E ) . Together , these results suggest that Ccnys are critical for mammary stem/progenitor cell self-renewal and regeneration capacity . To investigate the impact of Ccnys loss in normal development , we generated K14-Cre;Ccnyflox/flox;Ccnyl1lacZ/lacZ;mTmG mouse model to delete both Ccnys using a basal cell specific K14-Cre [25] , at the same time tracking the fate of the Ccnys-deficient cells using the mTmG reporter [26] . K14 is activated as early as E15 . 5 . The expressed Cre recombinase in basal cells would result in excision of the stop cassette in the reporter , thus marking the basal cells and their progeny , including luminal cells , with membrane-bound GFP ( mG ) . On the other hand , intact cells would express membrane-bound tdTomato ( mT ) , a red fluorescence protein . In control mice ( K14-Cre;Ccnyflox/+;Ccnyl1lacZ/+;mTmG ) ( Fig 5A ) , the mammary glands ( 8-week ) were largely labeled with GFP , as evidenced by whole mount imaging as well as histological sections ( Fig 5B ) . FACS analysis indicated that 58% of basal cells and 88% of luminal cells expressed GFP ( Fig 5C and 5D ) , suggesting that the K14-Cre did not induce 100% recombination . By sharp contrast , the mammary glands of the Ccnys-deficient mice ( K14-Cre;Ccnyflox/flox;Ccnyl1lacZ/lacZ;mTmG ) ( Fig 5E ) were mostly positive for tdTomato , as indicated by whole mount imaging and sections ( Fig 5F ) . FACS analysis confirmed that GFP+ cells comprises of only 2 . 5% of basal cells and 2% of luminal cells ( Fig 5G ) . In light of the partial efficacy of K14-Cre ( 58% ) used in this study , we propose that stem cells that had successfully recombinated and lost all copies of Ccnys failed to compete with the normal stem cells in the developing mammary gland ( Fig 5H ) . Since the mammary glands of K14-Cre; Ccnyflox/flox;Ccnyl1lacZ/lacZ;mTmG mice exhibited normal morphology ( S6 Fig ) , we postulated that the remaining normal stem cells are able to generate the whole mammary gland during the development and thus render the animals phenotypically silent . In fact , the replacement of cell population through cell competition can often be phenotypically silent as previous reported [27–29] . Together with the in vitro and transplantation data , these results suggest that loss of Ccnys impairs the function of mammary stem/progenitor cell , thereby affecting their contribution to differentiation during development . We next investigated the molecular mechanism by which Ccnys control the activities of mammary stem/progenitor cells . Ccny has been implicated in Lrp6 phosphorylation at Ser1490 ( pS1490 ) in HEK293T cells [8 , 30] . In basal colonies , we observed enriched pS1490 in M phase cells ( Fig 6A ) , suggesting that the cell cycle induced Lrp6 phosphorylation occurs during basal stem/progenitor cell expansion . To address the contribution of Ccnys in this event , we knocked down Ccnyl1 by shRNA in Ccny-/- mammary cells in order to completely inhibit the expression of Ccnys . We observed that the levels of Lrp6 phosphorylation at S1490 and active form of β-catenin were reduced in these cells ( Fig 6B ) . In embryonic fibroblasts ( MEFs ) isolated from the Ccnys DKO mice , similar reduction in pS1490 was also observed ( S7 Fig ) . Thus Ccnys are important for the S1490 phosphorylation of Lrp6 and may function in mammary basal stem/progenitor cells through Lrp6 activation . We tested whether activation of Wnt/β-catenin signaling can restore the phenotypes induced by the loss of Ccnys . As described in Fig 4 , knockdown of Ccnyl1 in a Ccny-/- background resulted in decreased basal colony sizes in 3D culture . We found that inhibiting GSK3β by LiCl restore the size of basal colonies to a normal range , while Wnt3A is ineffective ( Fig 6C ) , supporting the notion that Ccnys function at the level of Lrp6 phosphorylation [9 , 11] . Next , we assessed the rescue capability in regeneration assays . We induced constitutive Wnt/β-catenin signaling activation using the Ctnnb1flox ( ex3 ) /+ allele [31] . Mammary cells from Ctnnb1flox ( ex3 ) /+ mice were infected with Cre adenovirus to induce recombination ( Fig 6D ) . Simultaneously , we used mCherry-tagged Sh-Ccny and GFP-tagged Sh-Ccnyl1 lentivirus to knockdown expression of both Ccnys ( Fig 6D; S4 Fig ) . The appearance of the N-terminal truncated form of stabilized β-catenin ( β-cateninΔex3 ) ( Fig 6E ) indicated that Wnt signaling is constitutively activated in the Cre-expressed mammary cells . By contrast , no β-cateninΔex3 band was detected in the cells infected with a control adenovirus ( Fig 6E ) . In order to assess their capability in regeneration , the infected cells were FACS-isolated and transplanted into the recipient cleared fat pads . We found that Ctnnb1flox ( ex3 ) /+ mammary cells infected with the shRNAs were not able to reconstitute any outgrowth ( Fig 6F ) , due to loss of Ccnys . When Ctnnb1flox ( ex3 ) /+ mammary cells were infected with both the shRNAs and Adeno-Cre , however , they efficiently generated outgrowths , often with hyperplasia formation ( Fig 6F ) , a phenotype reminiscent of activation of Ctnnb1flox ( ex3 ) /+ alone [32] , indicating that the constitutively active β-catenin restores the regenerative capability in the Ccnys-deficient stem/progenitor cells . Together , these data suggest that Ccnys’ regulation of Lrp6 activation facilitates Wnt/β-catenin signaling in mammary stem/progenitor cells .
Our study demonstrates the physiological significance of Ccnys in embryonic development and developing mammary gland . While the detailed cause of the embryonic lethality requires further investigation in the future , our results establish that mitosis-induced Wnt signaling enhancement is essential for keeping the properties of dividing mammary stem/progenitor cells . Although Ccny and Ccnyl1 showed different expression patterns in developing mammary gland ( Fig 2 ) , they are functionally redundant . In addition to single knockout ( S2 Fig ) , female mice with only one allele of either Ccny or Ccnyl1 were not only viable ( Fig 1G ) but also produced functional mammary glands . Only upon full knockout did the mammary cell fail to contribute to the mammary gland formation ( Fig 5 ) . Consistently , Ccny-/-;Ccnyl1+/lacZ basal cells still reconstituted mammary outgrowths upon transplantation until the residual Ccnyl1 was further depleted by RNAi ( Fig 4D and 4E ) . These results indicate that in mammals not only the gene numbers ( two genes ) but also the copy numbers ( four alleles ) are redundant for Ccnys , at least in most events of the development . This possibly serves as a way to safeguard this important layer of Wnt regulation . The interconnections between cell cycle progression and cell fate specification have been explored in embryonic stem cells ( ESCs ) owing to the robust in vitro culture systems . The pluripotent status and differentiation propensity of ESCs are determined by specific cell-cycle profiles [1 , 33–35] . The importance of the cell cycle towards adult stem cells has also been documented in a variety of organs [2 , 36 , 37] . Notably , CDKs , together with their regulatory subunits cyclins , are involved in coupling cell cycle with stemness . G1 Cyclins , such as cyclin D , can impact the tendency and capacity of neural stem cells and hematopoietic stem cells to differentiate [37 , 38] , and CDK6 regulates hematopoietic stem cell quiescence exit [39] . Our study adds a new episode by showing that the Ccnys-enhanced Wnt signaling activities in M phase is essential for dividing mammary stem/progenitor cell to maintain their competitiveness and developmental potential . In adult stem cells , Wnt signaling activation can impact the adult stem cells by promoting cell-cycle entry through induced expression of the G1 factor , c-Myc and cyclin D1 , which function as switches between quiescence and division ( in intestine ) [40 , 41] . Our data indicate that the reverse is also true , particularly the cell cycle can impact the adult stem cells by enhancing Wnt signaling activities during mitosis , which is crucial in maintaining the stem/progenitor cell properties and the progeny cell fate decision ( in mammary gland ) . Our data highlight a cell cycle and Wnt signaling feed forward mechanism in active stem/progenitor cells for their expansion . In many in vitro expansion systems of adult stem cell , both Wnts and mitogenic growth factors are required for sustaining the culture [12 , 42–45] . Wnt proteins are not required for proliferation in these contexts , indicating that inhibition of differentiation is its main function in self-renewal [12 , 46] . This leads to a simplified view that mitogenic growth factors are responsible for pushing the cells into division . In light of the feed forward mechanism , mitogenic growth factors also partake in keeping the stem/progenitor cell properties by influencing the output of Wnt signaling during cell division . Many parameters can trigger cell competition , including differences in protein synthesis rates , growth factor receptivity and the expression level of Myc [27] . The replacement of cell population through cell competition is phenotypically silent , because the competitor cells conform to size-control mechanisms [27–29] . In this study , we found that deletion of Ccnys in a subset of mammary stem/progenitor cells diminishes their capability in generation of progeny cells and contribution to development , due to loss in competition with wild type stem cells . Hence , the Wnt signaling enhancement mediated by Ccnys is critical for dividing stem/progenitor cells to retain their competitiveness and full potency . This process may occur naturally , as it might provide a mechanism for elimination of suboptimal stem/progenitor cells during development . In conclusion , our findings here establish the importance of Ccnys in keeping the stem/progenitor cell properties and contribute to a better understanding of cell cycle control of Wnt signaling activation in cycling mammary stem/progenitor cells .
All procedures were carried out in accordance with the Chinese guidelines for the care and use of laboratory animals . Experimental procedures were approved by the Animal Care and Use Committee of Shanghai Institute of Biochemistry and Cell Biology , Chinese Academy of Sciences ( SIBCB-NAF-15-002-S335-005 ) . Experimental procedures were approved by the Animal Care and Use Committee of Shanghai Institute of Biochemistry and Cell Biology , Chinese Academy of Sciences . All mice were maintained in the specific-pathogen-free animal facility . The Ccny flox mice were constructed in Shanghai Research Center For Model Organisms . The Ccny flox mice were maintained on a 129/B6 mixed background . The Ccnyl1 knockout-first ( kof ) mice were kindly provided by EMMA ( Stain ID , EM:04396 . Stain name , C57BL/6NTac-Ccnyl1<tm1a ( EUCOMM ) Wtsi>/H ) , and maintained on a C57BL/6 background . The Axin2lacZ/+ mice , K14-Cre mice , Rosa26mTmG/+mice and Ctnnb1fl ( ex3 ) /+ mice were described previously [22 , 26 , 31 , 47] . The EIIa-Cre mice were maintained on a FVB background . Genotyping analyses were performed by PCR with genomic DNA extracted from tail tips . For embryos studies , pregnancies were obtained by natural mating and were timed from the day of the vaginal plug , which was defined as embryonic day ( E ) 0 . 5 . Embryos were dissected from uterus and then photographed under a dissection microscopy ( Olympus SZX16 ) . Primary mouse embryonic fibroblasts ( MEFs ) were isolated from E14 . 5 embryos and cultured as previously described [48] . The full lengths of mouse Ccny ( NM_026484 . 3 ) and Ccnyl1 ( ENSMUSG00000070871 ) cDNAs were amplified by PCR . Ccny and Ccnyl1 cDNA was cloned into pcDNA-HA to express HA-Ccny and HA-Ccnyl1 fusion proteins . To express GST or 6×His tagged Ccnyl1 , its cDNA was cloned into pGEX-4T-1 or pET28a vector . For lentiviral shRNA constructs , the annealed oligonucleotides were inserted into pLKO . 1 , modified by replacing the puromycin-resistance gene with a cDNA encoding GFP or mCherry . The sequences of shRNAs were as follows: Scramble shRNA: 5'-TCCTAAGGTTAAGTCGCCCTCG-3'; Ccnyl1 shRNA: 5'-GCTCATGCTCAACAATATTTC-3'; Ccny shRNA: 5'-GCAAGAGTCTCTTCATTAACCC-3' . GST tagged Ccnyl1 ( 77–367 aa ) and 6×His tagged Ccnyl1 ( 77–367 aa ) were expressed in E . Coli BL21 codon plus strain for anti-Ccnyl1 antibody generation and purification . Anti-Ccny antibody was as prepared previously described [10] . Other antibodies used were as follows: anti-β-catenin antibody ( C-terminus ) ( BD , 610153 ) ; anti-β-catenin antibody ( N-terminus ) ( Abclonal , A2064 ) ; anti-active β-catenin antibody ( Millipore , 05–665 ) ; anti-Cre antibody ( Novagen , 69050–3 ) ; anti-GAPDH antibody ( Proteintech , 10494-1-AP ) ; anti-Flag antibody ( Sigma , F3165 ) ; anti-β-actin antibody ( Sigma , 5316 ) ; anti-Lrp6 antibody ( Cell Signaling Technology , 3395S ) ; anti-phospho-Histone H3 Ser10 antibody ( Cell Signaling Technology , 3377S ) ; anti-phospho Lrp6 Ser1490 antibody for western blot ( Cell Signaling Technology , 2568 ) ; anti-phospho Lrp6 Ser1490 antibody for immunofluorescence ( gift from Christof Niehrs lab ) ; anti-E2F1 IgG ( Santa Cruz , sc-193 ) . Mammary glands were dissected and stained with X-gal as described [49] . Briefly , mammary glands were dissected and washed once with PBS . Mammary glands were fixed at room temperature for 2 h with β-galactosidase fixative ( 0 . 2% glutaraldehyde , 1 . 5% formaldehyde , 5 mM EGTA , 2 mM MgCl2 in PBS ) , and then washed 3 times in wash buffer ( 2 mM MgCl2 , 0 . 01% sodium deoxycholate and 0 . 02% Nonidet P-40 in PBS ) for 15 min each time . Finally , mammary glands were stained overnight in staining solution ( 1 mg/ml X-gal , 5 mM K4Fe ( CN ) 6 , 5 mM K3Fe ( CN ) 6 and 2 mM MgCl2 in PBS ) at 30°C . After X-gal staining , mammary glands were washed with PBS for several times , stained with carmine alum , and then dehydrated in 50% , 70% , 95% , 100% , 100% ethanol and Histoclear . Whole mount analyses were performed under a dissection microscope ( Leica ) . To localize lacZ+ cells easily , mammary glands embedded in paraffin were sectioned for 10 μm thick . The sections were de-waxed in Histoclear , rehydrated in 100% , 95% , 85% , 75% , 50% , 30% ethanol , and counterstained with nuclear fast red , and followed by a serial of dehydration in 30% , 50% , 75% , 85% , 95% , 100% ethanol , and cleared in Histoclear before sealed by coverslip . The stained samples were photographed with an Olympus BX51 microscope equipped with an Olympus DP71 cooled CCD camera . For whole mount analyses in conditional knockout mice ( cKO ) , observation was made under a fluorescence dissection microscopy ( Leica ) . After analysis , the mammary glands were processed for frozen sections or FACS analysis . The terminal end-buds of mammary glands from 5 week-old female mice were fixed with 10% neutral buffered formalin at room temperature for 36 h and then prepared for paraffin sections of 7 μm thick . In situ hybridization was performed using the RNAscope kit ( Advanced Cell Diagnostics ) following the manufacturer’s instructions . Axin2 , Ccny and Ccnyl1 probes were ordered from Advanced Cell Diagnostics . Mammary glands were isolated from 8- to 12-week-old virgin or other specified-stage female mice . The minced tissue was placed in culture medium ( RPMI 1640 with 25 mM HEPES , 5% fetal bovine serum , 1% penicillin-streptomycin-glutamine , 300 U ml−1 collagenase III ( Worthington ) ) and digested for 2 h at 37°C . After lysis of the red blood cells in NH4Cl , a single-cell suspension was obtained by sequential incubation with 0 . 25% trypsin-EDTA at 37°C for 5 min and 0 . 1 mg ml−1 DNase I ( Sigma ) for 5 min with gentle pipetting , followed by filtration through 70-μm cell strainers . For cell labeling , the following antibodies were used: FITC-conjugated CD31 , CD45 , TER119 ( BD PharMingen ) ; CD24-PE/cy5 , CD29-APC ( Biolegend ) . Antibody incubation was performed on ice for 15 min in HBSS with 10% fetal bovine serum . All sortings were performed using a FACSJazz ( Becton Dickinson ) . The purity of sorted population was routinely checked and ensured to be more than 95% . FACS-sorted cells were infected with lentivirus overnight , and resuspended at a density of 4 × 105 cells ml−1 in chilled 100% growth-factor-reduced Matrigel ( BD Bioscience ) . The mixture was allowed to polymerize before being covered with culture medium [DMEM/F12 , ITS ( 1:100; Sigma ) ] which was changed every 24 h . Primary colony numbers were counted and the diameters were measured after 5–7 days in culture . The colonies were mostly spherical , if colony was oval , the long axis was measured . LiCl ( Sigma ) was added to the culture medium from day 3 . For Wnt treatment , 200 ng ml−1 Wnt3A or Wnt4 conditional medium ( 1:50 conditional medium from Wnt4-expressing EpH4 stable cell line ) was added from day 1 . For passaging colonies , the medium was aspirated , and Matrigel was digested by incubation in 500 μl of Matrigel recovery solution ( BD Bioscience ) for 1 h on ice . Colonies released from Matrigel were harvested after pelleting . Single cells were obtained through incubation in 0 . 25% Trypsin-EDTA for 5 min at 37°C followed by gentle pipetting . Single cells were then replated in Matrigel as described above . Mammary cells were prepared , infected by indicated virus , and then cultured in monolayer . After 5–7 days , cells were digested with trypsin , sorted by FACS , and resuspended in 50% Matrigel , PBS with 20% FBS , and 0 . 04% Trypan Blue ( Sigma ) , to be injected in 10-μl volumes into the pre-cleared fat pads of 3-week-old female nude mice . Reconstituted mammary glands were harvested after 8–10 weeks post surgery . Outgrowths were detected under a fluorescence dissection microscope ( Leica ) . Outgrowths with more than 10% of the host fat pad filled were scored as positive . HEK293T cells , AD-293 cells , and EpH4 cells were cultured in DMEM ( high glucose , Hyclone ) supplemented with 10% fetal bovine serum , 100 units/ml streptomycin , 100 units/ml penicillin , and 0 . 3 mg/ml L-glutamine at 37°C and 5% CO2 . Plasmids were prepared using UNIQ-500 ( Sangong Biotech ) . Lentivirus was packaged in HEK293T cells as described [50] . Briefly , HEK293T cells were co-transfected with vesicular stomatitis virus G , packaging plasmid Delta8 . 9 , and transfer vector , using the conventional calcium phosphate method . At 48 h post-transfection , the culture medium was harvested and prepared for ultracentrifugation . The pellet of lentivirus was resuspended in PBS . Adenovirus particles were prepared in AD-293 cells and concentrated using CsCl gradient centrifugation as described [51] . Purified virus particles were stored at -80°C . EpH4 cells were synchronized to specific cell cycle stage by drug treatment . Briefly , cells were synchronized to late G1 phase by 800 μM mimosine treatment for 18 h or to S phase by administration of 2 mM thymidine for 18 h . Mitotic cells were harvested by shake-off after 200 ng ml−1 nocodazole treatment for 4 hrs . Cell fractionation was performed using the Membrane and Cytosol Protein Extration Kit ( Beyotime , P0033 ) . Briefly , unsynchronized EpH4 cells were lysed , and nucleus was depleted by centrifugation for 10 min at 700 g at 4°C . The post-nuclear supernatant of EpH4 cells was fractionated by ultracentrifugation for 30 min at 14 , 000 ×g at 4°C into cytosol ( C ) and membranes ( M ) . Equal-volume fractions of C and M were analyzed by western blotting . Total RNA was extracted using RNAiso plus ( Takara ) , and the PrimeScript RT Master mix kit ( Takara ) with oligo ( dT ) primers was used for the reverse transcription reaction . Quantitative RT-PCR ( qPCR ) was performed using an Applied Biosystems 7500HT sequence detection system with a FastStart Universal SYBR Green Master Mix kit ( Roche ) . Gapdh served as internal control . The reaction mixtures were incubated at 95°C for 10 min , followed by 40 cycles of 15 s at 95°C and 1 min at 60°C . qPCR primers used in this work: Ccny-F: 5'-TCTCTTCATTAACCATCATCCTCC-3' , Ccny-R: 5'-AATTTGCTTCTGTTCTGGGT-3'; Ccnyl1-F: 5'-AGTGACGTTGGTTTACTTAGAG-3' , Ccnyl1-R: 5'-GCCTTTCCATCTCATTCATGTC-3'; Axin2-F: 5’-AGCCTAAAGGTCTTATGTGGCTA-3’ Axin2-R: 5’- ACCTACGTGATAAGGATTGACT-3’ Gapdh-F: 5'-AGGTCGGTGTGAACGGATTTG-3' , Gapdh-R: 5'-TGTAGACCATGTAGTTGAGGTCA-3' . Tissues or cells were lysed with ice-cold RIPA buffer [50 mM Tris-HCl ( pH7 . 5 ) , 150 mM NaCl , 1% Nonidet P-40 , 0 . 5% deoxysodium cholate , 0 . 1% SDS , 5 mM EDTA , 10 mM NaF . Before use , add 1 mM PMSF , 3 mM dithiothreitol , 1 mM sodium vanadate , and protease inhibitors ( Merck ) ] . Proteins were resolved by SDS-PAGE and transferred to nitrocellulose membranes or polyvinylidene fluoride membranes . Immunoblots were developed in chemiluminescence reagent ( PerkinElmer Life Sciences ) and exposed in a Fujifilm LAS 4000 imager . Antibodies were diluted in PBS containing 2% BSA . Primary colonies were fixed with 4% paraformaldehyde at room temperature for 10 min . Colonies were then blocked in PBS containing 0 . 2% Triton X-100 and 10% goat serum for 1 h , followed by incubation with rabbit anti-phospho-Lrp6 Ser1490 antibody overnight at 4°C . The samples were washed with PBS containing 2% BSA for 3 times and incubated with secondary antibody for 1 hr . The samples were finally washed with PBS for 3 times and stained with DAPI . Whole mount fluorescent images of mammary glands were obtained using a Leica MZFLIII dissection microscope . RNA in situ images were acquired using a Zeiss A1-AXIO upright microscope . For confocal imaging , mammary glands were minced and then coverslipped . Confocal Images were acquired through a 40× or 63× oil immersion objective on a Leica TCS SP8 confocal microscope . EpH4 cells were plated into 24 well tissue culture dishes . After 16–20 h , EpH4 cells were co-transfected with 0 . 33 μg firefly luciferase reporter plasmid ( pGL4 . 17 ) containing the promoter region of Ccnyl1 , 0 . 033 μg Renilla luciferase plasmid ( pRL-TK ) and 0 . 67 μg pcDNA3-HA-E2Fs or the other transcription factors in per well of the 24-well plate . After 48 h , the cells were lysed and luciferase activities were measured with Dual-Luciferase Reporter Assay System ( Promega , Madison , USA ) . ChIP analysis was performed as previously described [52] . Briefly , EpH4 cells were cross-linked with 1% formaldehyde for 10 min at room temperature . For each group , a 10-cm dish of EpH4 cells with 80% confluency were lysed and chromatin was sonicated in a sonicator for 30 min ( with 7-sec sonication and 7-sec rest alternatively ) . Sonicated chromatin was then diluted and immunoprecipitated with anti-E2F1 IgG or rabbit normal IgG ( Santa Cruz , sc-2027 ) . Immunoprecipitation products and input were analyzed by quantitative PCR using the following specific primers: Site a , forward:5’-CAGCTCGAGATGAATGGAAACC-3’ , reverse: 5’-TAGCCAATCAGACCCGGACTTC-3’ . Site b , forward: 5’-CAGCAATGTCTCCATGTCACAT-3’ , reverse: 5’-CCCATGAGCACAACACAATTTC-3’ . Results are presented as mean ± s . d . , unless otherwise stated . Differences were considered significant when P<0 . 05 in an unpaired Student’s t-test . Three independent experiments were carried out for statistic results unless specified otherwise . | Stem cell self-renewal has two essential elements , cell division and at least of one of the daughter cells retaining stem cell properties , so-called stemness . The interconnections between cell cycle and cell fate specification have been explored in embryonic stem cells . However , less is known about how cell cycle affects the cell fate decision in tissue stem cells . In this study , we explore the function of particular mitotic factors Ccny and Ccnyl1 in regulating the dividing tissue stem cells . The development of the mammary gland occurs mostly in postnatal pubertal stage . At the time , the robustly dividing stem/progenitor cells reside at the forefront of the mammary epithelium extension , underlining the mammary gland as a good model to study the interconnection of cell cycle and tissue stem cells . In this study , we show that in dividing mammary stem/progenitor cells , Ccny and Ccnyl1 enhance Wnt signaling activities in mitosis . The signaling enhancement in this time window is essential for the stem/progenitor cell property maintenance during division . Deletion of Ccnys results in diminishing their competitiveness and developmental potential . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"phosphorylation",
"medicine",
"and",
"health",
"sciences",
"keratinocytes",
"reproductive",
"system",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"epithelial",
"cells",
"animal",
"models",
"developmental",
"biology",
"model",
"organisms",
"embryos"... | 2016 | Essential Roles of Cyclin Y-Like 1 and Cyclin Y in Dividing Wnt-Responsive Mammary Stem/Progenitor Cells |
The elimination of blinding trachoma focuses on controlling Chlamydia trachomatis infection through mass antibiotic treatment and measures to limit transmission . As the prevalence of disease declines , uncertainty increases over the most effective strategy for treatment . There are little long-term data on the effect of treatment on infection , especially in low prevalence settings , on which to base guidelines . The population of a cluster of 14 Gambian villages with endemic trachoma was examined on seven occasions over five years ( baseline , 2 , 6 , 12 , 17 , 30 and 60 months ) . Mass antibiotic treatment was given at baseline only . All families had accessible clean water all year round . New latrines were installed in each household after 17 months . Conjunctival swab samples were collected and tested for C . trachomatis by PCR . Before treatment the village-level prevalence of follicular trachoma in 1 to 9 year olds ( TF%1–9 ) was 15 . 4% and C . trachomatis was 9 . 7% . Antibiotic treatment coverage was 83% of the population . In 12 villages all baseline infection cleared and few sporadic cases were detected during the following five years . In the other two villages treatment was followed by increased infection at two months , which was associated with extensive contact with other untreated communities . The prevalence of infection subsequently dropped to 0% in these 2 villages and 0 . 6% for the whole population by the end of the study in the absence of any further antibiotic treatment . However , several villages had a TF%1–9 of >10% , the threshold for initiating or continuing mass antibiotic treatment , in the absence of any detectable C . trachomatis . A single round of mass antibiotic treatment may be sufficient in low prevalence settings to control C . trachomatis infection when combined with environmental conditions , which suppress transmission , such as a good water supply and sanitation .
Trachoma is the leading infectious cause of blindness worldwide [1] . Repeated infection of the ocular surface by Chlamydia trachomatis , provokes conjunctival inflammation ( active trachoma ) and scarring . This may progress in some people to entropion ( in-turned eyelid ) , trichiasis ( TT , lashes scratching the cornea ) and blinding corneal opacification . The World Health Organization's ( WHO ) most recent estimates indicate approximately 40 million people have active trachoma , 8 million have un-operated trichiasis and 1 . 3 million are blind [1] , [2] . The burden of this disease falls disproportionately on poor rural communities , predominantly in Sub-Saharan Africa . In a major WHO-led effort to control trachoma , the Alliance for the Global Elimination of Blinding Trachoma by 2020 ( GET2020 ) was formed which focuses on the development and implementation of the SAFE Strategy: Surgery for trichiasis , Antibiotics for infection , Facial cleanliness , and Environmental improvements to suppress transmission [3] . Trachoma was endemic in many parts of Europe in the 19th century; however , it gradually declined in the absence of any trachoma-specific interventions such as antibiotic treatment . This was attributed to general improvements in living conditions and hygiene , and underlines their importance in controlling trachoma . Unfortunately , such socio-economic transformation is only arriving slowly in many of today's trachoma endemic regions . Through implementing control programmes some encouraging reductions are being achieved . The Gambia , for example , has seen a marked decline [4] . Outside Sub-Saharan Africa , Morocco , Iran , Oman and Mexico have reported elimination of blinding trachoma . However , other countries have yet to experience similar reductions , probably because of higher initial levels of disease and variations in environmental factors such as water availability [2] . Typically , control programmes rely on the clinical signs of active trachoma to determine whether treatment should be given or stopped . Unfortunately , there is often a mismatch between disease and infection: infection may detected without clinical disease and disease may be found without infection . The correlation between disease and infection is not particularly strong and tends to weaken as the community prevalence drops [5] , [6] . This led the WHO to recommend the use of mass drug administration ( MDA ) with either single –dose oral azithromycin ( which has therapeutic tissue concentrations for several days ) or topical tetracycline to entire communities or districts [7] . Revised recommendations were published in 2004 with emphasis placed on the prevalence of follicular trachoma in children aged 1–9 years ( TF%1–9 ) as the primary clinical indicator [8] . Where TF%1–9 exceeds 10% , three annual MDAs followed by re-assessment , to determine whether to continue treating , is recommended [8] . This strategy increases the chance that infected persons are treated . However , in low prevalence settings where infection and disease are poorly correlated it may not represent the most efficient way to eliminate trachoma . The WHO also published specific Ultimate Intervention Goals ( UIG ) ; targets that need to be reached for the national certification of elimination of blinding trachoma [8] . For the surgery component of the SAFE strategy the UIG is to bring the population prevalence of TT below 0 . 1% in people aged 15 years and over . For the “AFE” components the UIG is to bring TF%1–9 below 5% . At this level it is expected that C . trachomatis infection will be too infrequent to drive scarring , although , there are no published data demonstrating this . The efficient and timely elimination of blinding trachoma is a major challenge for prevention of blindness programmes . It presents various practical difficulties , which need to be overcome if , for example , regular high coverage with antibiotic treatment is to be achieved . In addition , there are significant areas where the evidence base which guides policy , particularly around the use of antibiotic treatment , needs to be strengthened . In particular , there are limited published long-term data from low prevalence settings on C . trachomatis infection and disease to inform policy makers and programme managers in the development of guidelines for handling the end stage of trachoma elimination . Several key questions need to be answered in a range of endemic settings . Who should be treated ? How often should treatment be given ? What indicators can be used to guide the treatment decision ? When can treatment be discontinued ? What form of surveillance is needed after discontinuing treatment ? What environmental changes reinforce the reduction achieved by antibiotic ? We investigated the long-term effect of a single MDA on both the signs of trachoma and ocular C . trachomatis infection in a low prevalence setting in The Gambia . Other components of the SAFE strategy were also implemented , without any additional antibiotic treatment . Here we present results to five years; the first 17 months have previously been reported [9] .
Approval for this study was given by the Gambian Government/Medical Research Council Joint Ethics Committee ( SCC Number: 856 ) and by the London School of Hygiene & Tropical Medicine Ethics Committee . Informed consent at village , family and individual level was required before enrolment . A rapid assessment of active trachoma in children was conducted in 31 villages in Upper and Lower Saloum Districts , The Gambia . A geographically defined study area was selected , which contained a group of communities with endemic trachoma . This cluster of villages was chosen as it was typical of rural Gambia , where there is marked variation in the prevalence of active disease between neighbouring communities . The study area population was enumerated and information was collected on latrine access , water supply , livestock , house construction and other socio-economic indicators . Individuals normally resident in the study area for at least 6 months of the year were enrolled . Visitors were excluded . New residents were included in the study population if they lived in the study area for at least six months . For the first 17 months of this study each household was visited weekly by a project fieldworker to update the census . Inward and outward travel was recorded . Episodes of illness and the treatment , especially antibiotics , were also recorded . The weekly visits stopped after the first 17 months . The census was updated again just prior to the 30 and 60-months follow-ups . At five years household access to a functional pit latrine was reassessed . The study population was assessed on seven occasions over five years: baseline ( April 2001 ) , 2 , 6 , 12 , 17 , 30 and 60 months . The previously reported 6 and 12 month data are not presented here [9] . On each occasion all individuals normally resident within the study area were eligible for assessment . The assessments were conducted within a 6 to 10 day period . Individuals who had travelled out of the study area for this period were not seen . All clinical observations were made by the same ophthalmologist . Faces were assessed for fly-eye contact and the presence of nasal and ocular discharge . The left eye of each subject was examined for trachoma and classified using the WHO Trachoma Grading System [10] . Clinically active trachoma was defined as the presence of either a follicular score of 2 or 3 ( F2/3 ) or a papillary score of 3 ( P3 ) , equivalent to Trachomatous Inflammation , Follicular ( TF ) and Trachomatous Inflammation , Intense ( TI ) of the WHO Simplified Trachoma Grading System , respectively [11] . A Dacron swab sample was collected from the left upper tarsal conjunctiva , in a standardised manner , for the detection of C . trachomatis [9] . Care was taken to minimise the risk of contamination , by carefully handling only the far end of the swab ( away from the Dacron head ) and washing of the examiner's hands between subjects . All samples were kept on ice packs until transfer to a −20°C freezer later the same day for storage until processing . Mass antibiotic treatment of the entire population of the study area was conducted by the study team immediately after the baseline clinical assessment in April 2001 . Children ( <16 years ) were offered a single oral dose of Azithromycin ( suspension , 20mg/kg , up to a maximum of 1g ) . Infants under six months were given tetracycline eye ointment ( 1% , twice daily for six weeks ) . Adults were treated with a single oral dose of Azithromycin ( tablets , 1g ) , except for women of childbearing age who received oral erythromycin , as per the Gambian National Treatment Guidelines at that time ( 500mg , twice daily for two weeks ) . During the follow-up period no further treatment was given for active trachoma by the study team or the National Eye Care Programme ( NECP ) . The final five year follow-up took place after the introduction of the current WHO trachoma treatment guidelines , therefore , at the end of the study all villages with a TF%1–9 of 10% or more received mass antibiotic treatment as outlined above [8] . All individuals with trachomatous trichiasis were offered surgery , which was provided free of charge within the community . After the 17-month follow-up new concrete pit latrines were installed in each household . The project team provided health education about trachoma to the communities involved throughout the course of the study . C . trachomatis was detected using the Amplicor CT/NG kit ( Roche Molecular Systems , Branchburg , NJ ) . For samples in which C . trachomatis was detected by Amplicor , the infection load was estimated by quantitative real-time PCR of the C . trachomatis ompA gene using a previously described methodology [12] . The quantitative PCR method used was based on the ompA gene sequence of C . trachomatis serovar A ( Forward Primer: 5′-GCTGTGGTTGAGCTTTATACAGACAC-3′ , Reverse Primer: 5′-TTTAGGTTTAGATTGAGCATATTGGA-3′ ) . For the 30 and 60 month ompA quantitation an additional forward primer ( 5′-TCTGTTGTTGAGTTGTATACAGATAC-3′ ) was used which was optimised for estimating serovar B . Quantitation was done on two 4 µL replicate samples , first with the serovar A forward primer and then with that for serovar B . Here we focus on the prevalence of TF and C . trachomatis infection in 1 to 9 year olds due to the importance placed on this group in the WHO guidelines [8] . Because of the duration of this study , many of those who were initially in the 1 to 9 year old group were above this range by the end . As the probability of being infected is strongly influenced by age , it is necessary to compare the same age group rather than the same individuals over the duration of the study . Hence data were analysed according to each person's contemporaneous age at the different study time points; e . g . we compared all children aged 1 to 9 years at baseline with all those who were aged 1 to 9 years at five years . The estimated number of copies of ompA/swab was the geometric mean of two replicate assays . If infection was detected by Amplicor but no ompA could be detected by quantitative PCR , a maximum likelihood copy number estimate was made [12] . Where no C . trachomatis was detected a copy number of zero was assigned . The estimated copy numbers in individual samples were combined into an estimate of the burden of chlamydia infection for the study population as a whole [13] . An adjusted geometric mean ( the Williams Mean ) was calculated by adding 1 copy/swab to each estimated ompA copy number , calculating the geometric mean and then subtracting 1 copy/swab from the result . This measure can be used where one or more zero data values make the true geometric mean zero . To compare the prevalence of TF%1–9 and infection at the final five year follow-up with those at baseline , the within-village differences in prevalence were bootstrapped with 100 , 000 replicates [14] . For this last analysis , villages were given equal weight irrespective of numbers sampled . However , results based on minimum variance weights , taking into account both numbers sampled and intra-village correlation , were identical at the number of decimals places used [15] . Data were analysed in STATA version 10 and R version 2 .
The study population was comprised of all the individuals living in a geographically defined area 3km by 4km in size in Upper Saloum District , The Gambia . This area contains a cluster of 14 small villages , which have previously been described in detail [6] , [9] . At baseline in April 2001 the study area had a total recorded population of 1595 people . The median age was 14 . 2 years ( Interquartile range: 6 . 2–30 . 2 years ) and there were 542 children aged 1 to 9 years . The mean village size at baseline was 115 people; the individual village population sizes are shown in Table 1 . One village ( No . 12 ) of 123 people withdrew from the study after the 17-month follow-up . During the course of the five years we recorded 400 births , 57 deaths , 405 new residents , and 567 people moving away . At five years the total population of the study area ( excluding village 12 ) had risen slightly to 1667 people . The age structure was comparable: median age 13 . 5 years ( Interquartile range 5 . 5–30 . 5 years ) . The villages in the study area are typical of rural Gambia: the family compounds are grouped closely together and surrounded by farmland . The main employment is subsistence farming . All the households in the study area had access to clean water all year round within a few minutes' walk from wells situated in the middle of each village . At baseline 683/1595 ( 42 . 8%; 95% CI: 40 . 4–45 . 2 ) of the study population had access to a functional latrine in the family compound , although this varied considerably from village to village ( Table 2 ) . We have previously reported that at baseline there was a strong association between C . trachomatis infection and not having access to a latrine ( OR 13 . 3 , p<0 . 0001 ) [9] . Therefore , shortly after the 17-month follow-up visit , new pit latrines were installed in all 114 households . At five years 1154/1667 ( 69 . 2%; 95% CI: 67 . 0–71 . 4 ) of the population had access to a functional latrine ( Table 2 ) . These communities had not received any mass antibiotic treatment for trachoma control prior to this study . Antibiotic treatment coverage at baseline was 83 . 3% ( 95% CI: 81 . 4–85 . 1 ) of the study area population and 91 . 3% ( 95% CI: 89 . 0–93 . 7 ) of 1 to 9 year olds: oral azithromycin 1079 ( 81% ) , oral erythromycin 226 ( 17% ) , topical tetracycline 23 ( 2% ) . Coverage by village is shown in Table 1 for all ages and those aged 1 to 9 years at baseline . The follow-up rates of the 1 to 9 year old children were around 90%; individual village rates at each visit are shown in Table 3 . During the first 17 months of the study 153 episodes of illness were recorded . Forty-five of these episodes involved treatment with systemic antibiotics , which were usually obtained from the district health centre 10km away . This was equivalent to 20 antibiotic treatments/1000 people/year , during that 17 month period . Co-trimoxazole ( Septrin ) was the commonest antibiotic used ( 61% ) and tetracyclines the next commonest ( 16% ) . At baseline , the village-level average TF%1–9 was 15 . 4% ( Table 4 ) . There were marked differences in TF%1–9 between villages , ranging from 0% ( in four ) to 46% . It was greater than 10% in seven villages ( Table 4 ) . Overall , TF%1–9 declined during the study: reaching a minimum of 2 . 6% at 30 months and rising to 4 . 8% at five years ( Figure 1 ) . Comparing the baseline with five years , the village-level TF%1–9 decreased ( in absolute terms , Table 4 ) by an average of 9% ( 95% CI −18 to −2 , p = 0 . 002 ) . Most villages had a reduction in this indicator within the first few months following MDA ( Table 4 ) . In contrast , in villages 1 and 3 it took longer to decline . By 30 months all villages had 10% TF%1–9 or less . At five years , six villages had cases of TF; two villages had a TF%1–9 of greater than 10% and were therefore retreated ( Table 4 ) . There was a marked reduction in observed fly-eye contacts at the time of examination in 1 to 9 year olds from 18 . 0% ( 95% CI 14 . 7 to 21 . 5 ) at baseline to 1 . 5% ( 95% CI 0 . 5–2 . 5 ) at 5 years . Similarly there was a large reduction in the proportion of 1 to 9 year olds with unclean faces ( nasal or ocular discharge present ) : 38 . 4% ( 95% CI 34 . 1 to 42 . 7 ) at baseline and 5 . 0% ( 95% CI 3 . 1 to 6 . 8 ) at 5 years . Trichiasis ( TT ) was found in 9/592 ( 1 . 5% ) people 15 years and over at baseline . Six of these individuals accepted surgery by 12 months . At five years TT was found in 6/456 ( 1 . 3% ) aged 15 years and over . Two were incident cases ( compared with baseline ) , three had TT at baseline ( one of which had received surgery ) and one was a new resident at five years . Before treatment , the village-level average prevalence of C . trachomatis infection in 1 to 9 year olds was 9 . 7% ( Table 5 ) . There was marked variation in prevalence between the study villages despite their close proximity , with 8 villages having none . There were 7 villages where the TF%1–9 was greater than 10% , but only 4 of these had any cases of chlamydial infection . Following MDA the prevalence of C . trachomatis infection declined gradually , dropping below 1% by five years ( Figure 2 & Table 5 ) . Comparing the baseline with five years , the village-level prevalence of C . trachomatis infection ( in absolute terms , Table 5 ) decreased by 10% ( 95% CI −22 to −1 , p = 0 . 001 ) . We have previously reported the village level heterogeneity in the initial response to treatment [9] . In most villages ( 12 ) all cases of infection identified before treatment had cleared by two months ( Table 5 ) , and over the subsequent five years we detected only 12 isolated cases of infection in those villages . In contrast , in villages 1 and 3 the prevalence of infection increased at two months ( Table 5 ) . This increase was strongly associated with the travel of most village residents to a religious festival in Senegal one month after receiving treatment , which has been previously reported ( OR 12 . 2 , p<0 . 0001 , 95% CI 4 . 0–44 . 1 ) [9] . After the 2-month assessment , infection prevalence in these two villages declined progressively to zero at five years ( Table 5 ) . The risk of infection at any follow-up time point did not differ between those who had received treatment at baseline and those who had not or who were new residents during the follow-up ( data not shown ) . After the two-month follow-up TF%1–9 was found to exceed the 10% treatment threshold in several villages on one or more occasions ( Table 4 ) . However , in only 4/12 ( 33% ) of these occasions were cases of chlamydial infection detected in the same village ( Table 5 ) . Conversely , after the two-month follow-up chlamydial infection was detected in samples from several villages on one or more occasions ( Table 5 ) . However , TF%1–9 exceeded 10% on only 6/16 ( 37 . 5% ) of these occasions ( Table 4 ) . The geometric mean load of C . trachomatis infection amongst infected individuals ( all ages ) increased after treatment ( Table 6 ) . This burden of infection was concentrated in villages 1 and 3 . At 30 and 60 months we found very few infected individuals , most of whom had low infection loads . The estimated community C . trachomatis burden ( Williams mean ) declined throughout the five-year follow-up period ( Table 6 ) .
In this study a single round of mass antibiotic treatment was given to a cluster of 14 Gambian villages with low prevalence trachoma ( TF%1–9 15 . 4% ) . In addition , the other components of the SAFE strategy were promoted: free surgery was offered for trichiasis , some limited health education on trachoma was provided and latrines were installed at 18 months . The government had previously improved the water supply . Prior to treatment there were marked variations in village-level prevalence of disease and infection , despite their close proximity and environmental similarity . Six villages had cases of infection before treatment . The initial response to treatment was heterogeneous . In four villages with pre-treatment infections , no infection was detected at 2-months , seven villages had no infection at baseline or at 2-months and one village which had no baseline infection had a single new case at 2-months . In these 12 villages only a few isolated cases were subsequently detected during the five years of follow-up . In the other two villages there appeared to be a failure to respond to treatment [9] . This was attributed to possible re-infection , because it was associated with en masse travelling to a large religious festival in Senegal . Thereafter , the prevalence of infection in these two villages gradually declined in the absence of any additional antibiotic treatment for trachoma , reaching zero by five years . In the absence of additional antibiotic treatment , this decline is likely to be attributable to favourable environmental factors . It is interesting to note that there was less Ct strain diversity two-months after the mass treatment , which may have also contributed to the decline in infection [16] . Overall , these observations suggest that: ( 1 ) a single MDA was effective in most of the study area , although re-infection seemed to have occurred in two village; ( 2 ) the environmental conditions and/or the susceptibility of the population did not favour the transmission of C . trachomatis , leading to its decline and failure to re-emerge . These findings contrast those from highly endemic regions such as Ethiopia where C . trachomatis infection is not easily controlled or readily re-emerges . For example , a study of a single MDA to eight Ethiopian villages documented an initially good response to treatment , but subsequently , several villages experienced re-emergent infection [17] . Even where MDA was given every six months for two years and the village prevalence of infection dropped to very low levels or zero , re-emergence of infection was reported two years after treatment stopped [18] . However , even in these highly endemic parts of Ethiopia local elimination may be possible with more prolonged treatment [19] . The Gambia is currently considered to have low levels of endemic trachoma , although individual districts have TF%1–9 above the threshold for MDA [20] . Historically the levels were somewhat higher , with a well-documented downward trend during recent decades [4] , [21] . This reduction has been largely in the absence of a MDA programme . The Gambia became a beneficiary of the azithromycin donation programme through the International Trachoma Initiative in 2007 . Prior to this , tetracycline eye ointment was used to opportunistically treat individuals with active disease and their families . Therefore , it seems more likely that the decline in the prevalence of trachoma is mainly due to other factors . Various environmental factors may favour this change . However , it is difficult to determine their relative contributions . All households in the study area had access to clean water within a few minutes' walk . Access to water improved markedly throughout The Gambia during the 1990's through the provision of covered wells with hand-pumps by the government and coincided with a general reduction in trachoma [4] . A similar transition was observed in Malawi , which also coincided with marked improvements in water supply and hygiene programme [22] . Easier access to water combined with hygiene promotion probably leads to improved facial cleanliness and washing of bed linen and clothes . Numerous cross-sectional studies have found an association between active trachoma and unclean faces [23] , [24] . This led to the hypothesis that improving facial cleanliness may help to control trachoma by suppressing transmission of C . trachomatis in ocular secretions . However , the causality in this relationship could go both ways , as active conjunctival inflammation can produce secretions . This question was assessed in a trial of promoting facial cleanliness , which found a reduction in severe trachoma ( TI ) but not “any trachoma” ( TF and/or TI ) [25] . In our study we observed a marked reduction in the prevalence of unclean faces as the prevalence of TF%1–9 declined . We do not know whether this promoted or was secondary to the decline in active trachoma or whether it was related to health education about facial cleanliness . Eye seeking flies ( principally Musca sorbens ) are thought to contribute to the transmission of C . trachomatis in The Gambia . Measures to suppress the fly population by insecticide spraying have been associated with a drop in the prevalence of active trachoma [26] , [27] . As this fly preferentially breeds in exposed human faeces , the construction and use of pit latrines may indirectly reduce the transmission of infection [28] . However , in a cluster randomised trial from the Gambia the provision of pit latrines was only associated with a non-significant reduction in the prevalence of active trachoma [27] . In our study latrine coverage improved from 43% at baseline to 69% at five years . At the same time fly-eye contacts dropped significantly . However , we do not know whether this was due to a reduction in the overall fly population secondary to the latrines or a reduction in unclean faces , which attract flies . Whatever the explanation , reducing the frequency of fly-eye contact probably reduces the probability that C . trachomatis ( if present ) will be transmitted by this route in this particular environment . In contrast , in a study from Tanzania , intensive insecticide spraying was not associated with a reduction in trachoma [29] . This may reflect differences in the transmission ecology of this infection in different regions . The study population did not have ready access to antibiotics with anti-chlamydial activity . Antibiotic use was monitored during the first 17 months of the study , and we have no reason to think that it was different during the rest of the follow-up period . Background antibiotic use probably contributed little to the decline in C . trachomatis , in contrast to a study from Nepal , where a much higher background use of antibiotics may have contributed to a decline in trachoma [30] . During our study there were no significant demographic changes , such as fewer young children , favouring trachoma control . The relatively small village size in this part of The Gambia may make trachoma less stable . In a smaller community children probably come into contact with a smaller absolute number of potentially infected people than in a larger village , possibly reducing their chance of becoming infected . A difficult issue facing trachoma control programmes as the prevalence of disease declines is the increasing uncertainty over the infection status of formerly endemic communities , because of a mismatch between disease signs and chlamydial infection [5] . We have previously reported that prior to MDA in these villages a minority of the people with C . trachomatis had TF and conversely many with clinically active trachoma did not have infection [6] , [9] . At the five year follow-up two villages had a TF%1–9 of >10% , triggering intervention with MDA , under current WHO guidelines [8] . However , neither village had a single case of chlamydial infection . These observations are consistent with those from a survey conducted elsewhere in The Gambia in the same year ( 2006 ) as our five year follow-up [20] . With the increasing effectiveness of trachoma control programmes other countries are likely to reach the low prevalence situation current in The Gambia . The implication is that , applying the current intervention algorithms , many uninfected communities will continue to receive unnecessary rounds of MDA on multiple occasions . It remains uncertain , on clinical grounds , when MDA can stop without re-emergent infection . The mismatch of disease and infection probably arises for several reasons in low prevalence settings . Firstly , there is a difference in the time course of episodes of C . trachomatis infection and active disease [31] , [32] . There is probably an initial subclinical “incubation” period prior to the development of inflammatory signs in the conjunctiva; subsequently the inflammatory signs can linger for many weeks after the infection becomes undetectable . In contrast , in high prevalence settings this asynchronous time course is less evident , as the spacing between separate infection and disease episodes is shorter and may indeed overlap , increasing the correlation . Secondly , in low prevalence settings the disease is probably milder , with some infected individuals having conjunctival inflammatory signs , which do not reach the TF diagnostic threshold . Thirdly , follicular conjunctivitis is not exclusively caused by C . trachomatis . In low prevalence settings a greater proportion of “TF” will be due to other agents , such as adenovirus , causing outbreaks , which mimic endemic trachoma . In a number of studies the prevalence of TF after the successful control of C . trachomatis infection with MDA has taken several years to decline [9] , [33] , [34] , [35] . The reason for this is not known , as most clinical episodes last only for a few weeks; however , it is likely that in many situations MDA will inevitably continue some time after the infection has been successfully cleared from formerly endemic communities . Finally , clinical observations and laboratory tests may both produce false negative and false positive results . The observation that incident TT developed after the prevalence of C . trachomatis dropped is important . This indicates the potential for cicatricial disease progression after the chlamydial infection is controlled . The reasons for this are unknown , although other pro-inflammatory stimuli such as non-chlamydial bacterial infection and ocular dryness may contribute [36] , [37] . The implication for control programmes is that surveillance and treatment for TT will need to continue for some years after the chlamydial infection has been successfully controlled . This study has a number of limitations . Firstly , in the absence of an untreated control group , it is not possible to determine what proportion of the reduction seen in these villages was due to the treatment at baseline and what was due to a secular trend . However , it seems likely that both contributed to the decline . Secondly , there was a long gap between the last two time-points . It is unknown what occurred during that time . Thirdly , we do not routinely run parallel assays for human nucleic acid in swab samples to check specimen adequacy . However , in a separate study using the same swab type collected in the same standardised manner by the same ophthalmologist from subjects in this study area we found that human RNA could be extracted from all swabs collected [38] . Field control swabs were not collected , as this was not standard practice at that time . Finally , one village withdrew from the study after the first 17 months . At enrolment this village had one of the higher rates of TF%1–9 and on the last occasion it was assessed it had four cases of TF but no cases of infection . We do not know what happened in this village subsequently and what affect this might have had on the surrounding villages . In summary , our findings suggest that a single round of mass antibiotic treatment in association with the other components of the SAFE strategy may be all that is required to clear C . trachomatis infection in low prevalence settings , if reasonably high antibiotic coverage is achieved . We also observed an unusual re-infection episode soon after treatment , which was associated with travel . However , the conditions were such that this infection did not persist or appear to transfer to neighbouring villages , indicating the critical importance of F&E interventions that suppress transmission , in addition to antibiotic treatment . Coordinated attempts to control trachoma across wide geographical areas may also reduce the risk of re-infection . The current guidelines will probably lead to widespread overtreatment of communities in similar settings . This might be preventable if a role for suitable field-based diagnostic tests for C . trachomatis can be established in confirming elimination of infection . With the implementation of the SAFE Strategy , many previously highly endemic regions will eventually reach this position . Therefore , a clear evidence-based approach for handling the trachoma “End Game” is still needed . | Trachoma is the most common infectious cause of blindness worldwide . Mass antibiotic treatment with azithromycin is used to control ocular Chlamydia trachomatis infection . There is uncertainty over how frequently and for how long treatment is needed , particularly in low prevalence settings . This study examines the effect of a single round of treatment on clinical disease and infection in a cluster of trachoma endemic Gambian villages over a five-year period . These villages had good water supplies and sanitation improved part way through the study . We found treatment was followed by a marked decline in infection prevalence ( by PCR ) to less than 1% . The decline in prevalence of active disease in children was less marked . Several villages had a prevalence of active trachoma in 1 to 9 year old children of greater than 10% during the follow-up period , mostly in the absence of detectable infection . The implication of this study is that a single , high coverage mass treatment may be sufficient to control C . trachomatis infection in a low prevalence setting , particularly when combined with environmental measures to limit transmission . However , relying on clinical signs to guide treatment decisions is likely to lead to significant amounts of over treatment where current guidelines are implemented . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"ophthalmology/eye",
"infections"
] | 2010 | Profound and Sustained Reduction in Chlamydia trachomatis in The Gambia: A Five-Year Longitudinal Study of Trachoma Endemic Communities |
Adaptive plasticity allows organisms to cope with environmental change , thereby increasing the population’s long-term fitness . However , individual selection can only compare the fitness of individuals within each generation: if the environment changes more slowly than the generation time ( i . e . , a coarse-grained environment ) a population will not experience selection for plasticity even if it is adaptive in the long-term . How does adaptive plasticity then evolve ? One explanation is that , if competing alleles conferring different degrees of plasticity persist across multiple environments , natural selection between genetic lineages could select for adaptive plasticity ( lineage selection ) . We show that adaptive plasticity can evolve even in the absence of such lineage selection . Instead , we propose that adaptive plasticity in coarse-grained environments evolves as a by-product of inefficient short-term natural selection: populations that rapidly evolve their phenotypes in response to selective pressures follow short-term optima , with the result that they have reduced long-term fitness across environments . Conversely , populations that accumulate limited genetic change within each environment evolve long-term adaptive plasticity even when plasticity incurs short-term costs . These results remain qualitatively similar regardless of whether we decrease the efficiency of natural selection by increasing the rate of environmental change or decreasing mutation rate , demonstrating that both factors act via the same mechanism . We demonstrate how this mechanism can be understood through the concept of learning rate . Our work shows how plastic responses that are costly in the short term , yet adaptive in the long term , can evolve as a by-product of inefficient short-term selection , without selection for plasticity at either the individual or lineage level .
Organisms that live in variable environments are often subject to opposing selective pressures , either temporal or spatial , such that intermediate generalist phenotypes have decreased fitness across all environments . Rather than evolving a generalist phenotype , populations can keep adapting to each environmental condition as they encounter them , a process known as adaptive tracking [1 , 2] . Populations that evolve via adaptive tracking need time to adapt to each new environment . As a result of this adaptation , the population experiences reduced fitness after each environmental change . Both populations that evolve a generalist phenotype and those that evolve by adaptive tracking thus have reduced fitness in the long term . By contrast , adaptive phenotypic plasticity allows individuals to maintain an adaptive fit between phenotype and environment: plastic individuals produce only high fitness phenotypes by responding appropriately to environmental cues . Populations evolving adaptive plasticity thus avoid both the fitness loss arising from trade-offs of generalist phenotypes and the fitness loss that tracking populations suffer after environmental change . Within this framework , the question of whether plasticity evolves can be interpreted as the comparison between the average fitness across all environments for populations which evolve plastic responses , evolve generalist phenotypes or evolve via tracking [3 , 4] . As such , a considerable amount of effort has been invested in characterizing the conditions that determine the fitness of plastic rather than non-plastic solutions , and to document if plasticity itself incurs a fitness cost [5–7] . While adaptive plasticity is common in nature and demonstrably superior to non-plastic solutions for a wide range of conditions , the process by which it evolves remains a matter of debate . The standard assumption that natural selection favours the best available solution is problematic , since natural selection only discriminates between phenotypes that are expressed . Natural selection is thus unable to detect that a plastic organism is adapted to more environments than a non-plastic one unless individuals encounter multiple environments within their life spans , a condition known as environmental fine-grain [8] . Even when individuals experience more than one environment per lifetime , each individual may express only a single phenotype if plastic responses are irreversible [9–11] , too slow ( e . g [12] ) or too costly ( e . g . [5] ) relative to the fitness advantage of producing the right phenotype for the current conditions [2 , 7] . This creates an evolutionary dilemma: adaptive plasticity maximizes fitness in the long-term , but natural selection favours non-plastic phenotypes in each short-term environment . In other words , experiencing one environment per lifetime ( environmental coarse grain ) does not allow individual selection for plasticity , so that if plastic responses incur any cost compared to non-plastic phenotypes they will be selected against in the short-term . Since costly plastic responses in coarse-grained environments provide fitness benefits only when individuals are selected over multiple generations , we refer to those responses as long-term adaptive plasticity . While long-term adaptive plasticity is selected against in the short-term , adaptive responses to coarse-grained environments commonly evolve , and include environmental determination of resistance and dispersal phenotypes [13 , 14] and seasonal morphs of short-lived species [9 , 15] . How can we explain the process by which costly adaptive plasticity evolves in such coarse-grained environments ? While individual-level selection does not favour plasticity in coarse-grained environments , alleles that determine an organism’s plasticity are transmitted between generations , and their fixation or loss will depend on their fitness across the set of environments they encounter [16 , 17] . Natural selection may therefore discriminate between plastic and non-plastic alleles if both are maintained long enough to be selected across multiple environments , even if each individual organism experiences only a single environment . Plastic adaptations to coarse-grained environments could therefore evolve if multiple alleles ( genetic lineages ) persist long enough to be subject to natural selection across multiple generations and environments , a process known as lineage selection [4 , 18 , 19] . More precisely , we define lineage selection as a specific type of natural selection acting on multiple alleles which persist for multiple generations ( see [20] ) . This is in contrast with Strong Selection and Weak Mutation regimes ( SSWM ) in which each new allele is either lost or fixed before more genetic variation can arise . Under SSWM genetic variation is provided only by new mutations ( rather than standing genetic variation ) , so that repeatedly comparing multiple alleles is impossible . The availability and persistence of standing genetic variation on plastic responses is thus a key requirement for the evolution of adaptive plasticity in coarse grained environments ( e . g . [4 , 17 , 19] ) . This implies that plasticity will not evolve in populations that are small or under strong selection , since these conditions remove the genetic variation lineage selection requires to operate ( e . g . [21] ) . Because small population size and strong selection are representative for populations experiencing rapid environmental change , evolution of plasticity appears unlikely to play a role in evolutionary rescue or successful colonization [22 , 23] . The evolution of costly adaptive plasticity will only be possible if genetic diversity is available , but high genetic diversity will also cause rapid removal of costly plastic variants in favour of non-plastic short-term solutions , so that costly adaptive plasticity should only evolve as an intermediate step towards non-plastic solutions . We apply a core concept of learning theory—learning rate—to propose an alternative mechanism for the evolution and maintenance of costly adaptive plasticity without lineage selection . In machine learning , learning rate measures the amount of change a system accumulates with each example shown . Existing literature demonstrates that the process of learning by trial and error is mechanistically analogous to evolution by natural selection [24] . In the context of adaptation , genetic learning rate measures the ability of a population to change in response to new environments by accumulating adaptive mutations . More specifically , we can define genetic learning rates as the amount of genetic change fixed by a population in each new environment . Genetic learning rate ( henceforth just learning rate ) depends both on the ability to generate variation ( mutation rate and effect size , population size ) and to fix particular variants ( strength of selection ) . Since both the processes that produce and fix variants require time to operate , increasing the time spent in each environment will allow populations to accumulate more adaptive change . Thus , the more generations a population spends in a single environment the higher its learning rate will be . As we show in our simulations , populations initially produce phenotypes matching their current environment by accumulating both mutations that change the mean phenotypic value and mutations that change plasticity . Populations with high learning rates find optimal phenotypes for the current environment and remove costly plasticity before each new environmental shift: When populations can quickly reach current optima in each current environment , plastic adaptations to past environments cannot evolve . Populations with low learning rates cannot reach current optima before the next environmental shift , and pass on to the next environment all genetic changes which brought them closer to the previous phenotypic optimum , whether or not these genetic changes cause phenotypes to be plastic . Selection in the new environment thus starts from a population which already accumulated adaptively plastic changes , so that the overall plastic responses can be further refined over time . In evolutionary terms , low learning rates maintain directional selection for plastic development , with the end result of directing evolution towards the production of long-term adaptive plastic responses . Unlike the lineage selection explanation , the learning theory explanation does not require the prolonged co-existence of alleles with different effects on plasticity: adaptive plastic responses will evolve even in populations which exhibit only a single reaction norm at any given time . Rather , learning theory only requires that the population accumulates limited genetic change per environment , so that the average genotype retains some of the adaptive plasticity accumulated in past environments . Learning theory thus predicts that , as long as natural selection is inefficient in bringing about genetic change , long-term adaptive plasticity should evolve even in the extreme case when only one lineage is present in the population at any given time ( strong selection weak mutation ) and plasticity is selected against in each current environment . In this paper , we provide a first exploration of the evolution of adaptive plasticity from a learning theory perspective . To do so , we employ a classic linear reaction norm model [25 , 26] to simulate the evolution of costly adaptive plasticity in temporally coarse-grained scenarios . This allows us to contrast the predictions made by learning theory and lineage selection regarding when and how plasticity should evolve . First , we demonstrate that plasticity can evolve in coarse-grained environments , showing that individual-level selection for plasticity is not necessary to evolve adaptive plasticity . Second , we demonstrate that adaptive plasticity evolves in coarse-grained environments even in the absence of multiple lineages , counter to the predictions of lineage selection . Third , we show that limiting mutation rates biases populations towards adaptive plasticity rather than adaptive tracking , in accordance with the predictions of learning theory . These results reveal that long-term adaptations can evolve even when each current environment selects against them , as long as natural selection is inefficient .
We simulate a population that experiences temporal environmental heterogeneity . Each individual receives information from the environment and develops into an adult phenotype , upon which selection can act . We follow standard approaches for the evolution of plasticity [18 , 27 , 28] and model development as a linear reaction norm , whose intercept a represents the mean genetic trait value across environments ( also known as G , e . g . [17] ) and slope b the degree of plasticity , or genotype by environment interaction ( GxE , see Reaction norm model ) . The developed phenotype P is thus P = a + b * C where C is the univariate environmental cue . We model a heterogeneous varying environment with 10 environmental states , so that each environmental state , Ei produces a single , unique value of the cue C E i and requires a single specific univariate phenotype P E i . We model the matching between cues and trait optima as a linear function ( see Environmental variability ) . This implies that a linear reaction norm with appropriate slope and intercept can achieve perfect fit for all environments in our set . We assume non-overlapping generations of individuals with a constant fixed lifespan . This assumption allows us to control the granularity of environmental variability with a single parameter , K . If K ≥ 1 the environment changes every K generations , indicating coarse-grained ( K = 1 ) or slow coarse-grained ( K > 1 ) environmental variability . If instead K < 1 the population encounters on average 1/K environments per generation , indicating fine-grained environmental variability . We evaluate the fitness of each individual based on the distance of its developed phenotype from the optimal target phenotype in the current environment . In case the individuals experience more than one environment , we calculate their fitness as the mean match between the developed phenotypes and the selective environments experienced . We further impose a fitness penalty proportional to the individual’s responsiveness to its environment ( absolute reaction norm slope b , see above ) . This cost of plasticity ensures that plastic individuals will have lower fitness than non-plastic ones regardless of their phenotypes , and effectively represents a trade-off incurred by plastic organisms ( see Evaluation of fitness ) . While few empirical studies have found evidence for costs of plasticity ( see Conclusion ) , including a cost means that plasticity is selected against , and thus serves as a form of conservative bias against plasticity . Since we measure fitness as relative to a pre-specified optimal phenotype , we express it as phenotypic mismatch or lack-of-fit: a measure which decreases quadratically from zero as the phenotype diverges from the optimum ( see section Evaluation of reaction norms ) . Organisms reproduce asexually with a probability proportional to their relative fitness within the population ( see Evolutionary process ) . Every individual inherits the same slope and intercept as their parents , which are then mutated by adding a random value selected from a normal distribution with mean 0 and standard deviation equal to the mutation size ( 0 . 01 unless otherwise specified ) . Thus , both intercept and slope mutate every generation ( effective mutation rate = 1 ) , but most mutations have small effects . Unless otherwise stated , we set a population of 1000 individuals and choose strength of selection ω of 0 . 2 . In addition , we set the associated cost of plasticity , λ , to be 0 . 1 . While we assume that the cost of plasticity is a property of the genotype , the fitness losses caused by adaptive tracking depend on both the frequency of environmental changes and the amount of time required to reach new short-term optima after each environmental change . Thus , if environmental changes are rare or if the population can quickly reach new optima , the cost of adaptive tracking can be lower than the cost of adaptive plasticity . To verify whether or not adaptive plasticity is the optimal long-term strategy , we analytically tested all parameter combinations used in our simulations ( see S1 Appendix ) . Our analysis confirmed that the fitness cost of adaptive tracking is greater than the cost of adaptive plasticity for all parameter combinations used in this paper . Since adaptive plasticity is the optimal strategy across all our simulations , we can rule out that the eventual evolution of adaptive tracking is because of its greater long-term fitness . In other words , lineage selection should select for adaptive plasticity across all our simulations , since adaptive plasticity incurs lower fitness costs compared to adaptive tracking . In this section , we compare the evolution of plasticity in fine-grained environments , which allow individual-level selection for plasticity , with coarse-grained ones , which do not . We initially assess the evolution of phenotypic plasticity when individuals encounter multiple environmental states per life-time ( i . e . , a fine-grained environment; here 10 , K = 0 . 1 ) . We further assume that the phenotype can change during individuals’ lifespan ( reversible plasticity ) , and this change is both immediate and incurs in no fitness costs . In fine grained environments , the evolved reaction norms converge the optimal intercept and slope in less than 3000 generations ( Fig 1A , inset ) . This means that individuals produce trait values that perfectly match the optimal trait value of all environmental states they encountered during their lifetime , as we can see from the fact that the distance between realised and optimal phenotypes decreases to zero for all environments in our set ( Fig 1A ) . We find minimal residual genetic variation on both the slope and intercept terms of the reaction norm ( Fig 1B ) . This is reflected in the limited differences between the reaction norms of top and mean performing individuals ( Fig 1A ) . Note that the reaction of the average ( yellow dots ) and best individual ( green dots ) are perfectly aligned and match the optimal reaction norm ( red crosses ) . We contrast the previous fine-grained scenario with a slow coarse-grained environment in which conditions change every 4000 generations on average ( K = 4000 ) . As such , each individual experiences only one environment , and environmental change between generations is also slow . In this coarse-grained environment , the population fails to evolve adaptive long-term plasticity ( Fig 2 ) . After each environmental change we observe a drop in fitness to the current environment , followed by a distinctive two-step pattern in their adaptive paths . During the first phase , organisms evolve towards the new target phenotype , as indicated by the steep increase in current fitness ( Fig 2A , inset , green line ) . Crucially , the increase in current fitness during this phase is accompanied by a corresponding increase in fitness to past environments ( Fig 2A , blue line ) , which indicates evolution of adaptive plasticity . During this phase , mutations which increase plasticity can be selected for if they cause the production of fitter phenotypes , offsetting the cost of plasticity ( see S1 Appendix ) . After organisms are able to produce phenotypes which match the current phenotypic optima , we observe a decrease in their fitness to past environments ( Fig 2A , blue curve ) . This indicates that the same organisms would no longer be able to produce adaptive phenotypes when exposed to past environments , consistent with a decrease in costly adaptive plasticity . During this phase plasticity is directly selected against in order to decrease its fitness costs . In other words , the population reaches the optimal phenotype using a combination of slope and intercept ( phenotypic adaptation ) and then minimizes the slope ( plasticity minimization ) . From a fitness perspective , selection during the phenotypic adaptation phase increases fitness by producing the current target phenotype , whereas selection in the plasticity minimization phase increases fitness by maintaining the current target phenotype while removing costly plasticity . It is worth noting that these two phases match those described in the analogous model presented in [17] . After the plasticity minimization phase we still observe some genetic variation in reaction norm slope ( grey lines in Fig 2B ) , but the average slope is 0: adaptive plastic responses are approximately as likely as maladaptive ones . Populations evolving under slow , coarse-grained environments thus fail to evolve adaptive plasticity and instead re-adapt upon each environmental change , consistently with adaptive tracking . Next , we test whether or not direct selection for plasticity is required for its evolution . To do so , we set the environment to change every generation ( K = 1 ) , which is the fastest rate we can set under a coarse-grain scenario: every individual experiences only a single environment , but every generation experiences a different one . Since each individual only experiences one environment , we can rule out direct selection for adaptive plasticity . Furthermore , costly plasticity is selected against within each short-term environment . In this fast coarse-grained environment , populations evolved adaptive plasticity ( Fig 3 ) . We observe that the deviation from the optimal phenotype for both current and past environments decreased to zero , indicating optimal fit to all environments within the range experienced ( Fig 3A ) . In addition , we observe less residual genetic variation compared to the case of slow coarse-grained environmental variability ( Fig 3B ) . This is also indicated by the narrow gap between the top and the mean performance curve in Fig 3A . Looking at the evolutionary trajectory of the population , we can see that while fitness to the current environment ( green line ) fluctuates , fitness to the whole environment set ( past environment; blue line ) gradually increases over time . Moreover , we see no gap between performance in current and past environments . This indicates that increasing fitness to the current environments does not cause loss of fitness in past environments . Instead , the population accumulates responses that are adaptive for all previously experienced environments . These results demonstrate that populations evolving in fast-changing environments produce adaptive plastic responses even when plasticity is costly and environmental change only occurs between generations . At this stage , we have merely confirmed well-known results ( e . g . , [17] ) . We now consider two explanations for the evolution of adaptive plasticity in coarse-grained environments . The standard interpretation is based on a lineage selection model , where faster environmental change will increase the odds that each allele is tested in more than one environment . Adaptive plasticity can evolve since plastic alleles have greater mean fitness than non-plastic alleles when compared across multiple environments , even though the latter have higher fitness within each current environment . The learning theory interpretation instead is based on the prediction that decreasing the number of generations in each environment will decrease the genetic change accumulated within each environment ( i . e . , the learning rate ) , ensuring that the changes accumulated during the phenotypic adaptation phase are not lost because of optimization to current environments . While both mechanisms cause a shift from short to long-term adaptation , each has distinct requirements: lineage selection relies on the transmission of genetic variants in order to compare the fitness of multiple alleles; learning theory requires that populations accumulate little genetic change in each environment , so that the system retains some information from the past . In contrast with lineage selection , learning theory does not require that past information is stored in separate lineages . Rather , past information can also be stored in developmental parameters , such as the slope of plasticity . As long as plasticity does not revert to zero , the system retains some information about past adaptive plasticity and can be progressively improved after each environmental change , regardless of the presence of trans-generational genetic variation . In the next two sections , we make use of this key difference to determine which of the two processes can better explain the evolution of plasticity in coarse environments . To test the need for lineage selection , we repeat the scenarios for the evolution of plasticity in fine-grained ( K = 0 . 1 ) , coarse-grained ( K = 1 ) and slow coarse-grained ( K = 40000 ) environments enforcing strong selection and weak mutation ( SSWM ) . Under SSWM , the speed at which mutations arise is much slower compared to the speed at which they are fixed or lost , driving standing genetic variation to zero . Comparing the fitness of alleles across different environments is therefore impossible . We model SSWM using a hill-climber algorithm: each evolutionary step produces only one mutation . If the new mutation is fitter than the previous one it is fixed , otherwise it is lost ( see Hill-climbing model ) . SSWM leads to a constant effective population size of 1 and makes lineage selection impossible . Therefore , if the lineage selection hypothesis is correct , we expect that adaptive plasticity will fail to evolve in all coarse-grained environments . To rule out that the potential failure to evolve plasticity is due to insufficient time , we verify the results under an extended simulation time of 2*107 generations . Contrary to the predictions of the lineage selection explanation , we find that the results from the above simulations are qualitatively and quantitatively similar to those obtained using a population size of 1000 , despite the SSWM selection regime ( Fig 4 ) . That is , populations fail to evolve plasticity when environments change every 40000 generations ( Fig 4A ) , and succeed in doing so when provided with either fine environmental grain ( Fig 4B ) or a rapid coarse-grained ( i . e . , trans-generational ) change ( Fig 4C ) . The evolutionary trajectory of populations under SSWM also remains remarkably similar to that of populations with standing genetic variation ( compare Fig 4 with Figs 1 , 2 and 3 ) . Populations evolving in fine-grained and fast coarse-grained environments both show a gradual increase in fitness to past environments , which remains comparable to fitness in the current environment . This indicates that they adapt to all previously seen environments rather than just the current one . Populations in slow coarse-grained environments instead perform consistently better in current environments compared to past ones , showing the repeated evolution of phenotypes adapted to current conditions , or adaptive tracking . Their evolutionary trajectory also displays the same two-step cycle after each environmental change: fitness increase in both current and past environments ( phenotypic adaptation ) followed by fitness decrease in past environments only ( plasticity minimization ) ( Fig 4A ) . Taken together , these findings demonstrate that both the final results and the evolutionary trajectories of our simulations are largely unaffected by the lack of standing genetic variation . Since standing genetic variation is required for adaptation via lineage selection , these results falsify the hypothesis that plasticity needs to evolve by averaging the fitness benefits of alternative variants across multiple environments . In the next section , we make further predictions based on the learning theory explanation and try to falsify them . Using a learning theory framework , we can define the conditions that allow evolution in coarse-grained environments to approximate evolution in fine-grained ones . The two scenarios will produce the same outcome only as long as the average of evolutionary changes in coarse-grained environments is the same as the evolutionary changes that would happen in fine grained environments . In our specific example , individuals selected in slow coarse-grained environments evolve non-plastic solutions after each environmental change . On average , evolutionary changes in slow coarse-grained environments decrease plasticity until it reaches zero . This is in contrast with fine-grained environments , which evolve plasticity towards the optimal adaptive slope . Since the average change in plasticity in coarse-grained environments is different from the change in plasticity under fine-grained environments , the two scenarios have different outcomes . Conversely , individuals selected in fast coarse-grained environments retain some plasticity between environments . Furthermore , on average , the change in plasticity induced by each new environment points towards optimal adaptive plasticity: inherited maladaptive plasticity will be selected against , and inherited adaptive plasticity will be conserved . Therefore , as long as plasticity does not reach zero before the environment changes , evolution in coarse-grained environments will follow the same direction as evolution in fine-grained environments . This is the reason why we expect lower learning rates to cause the evolution of adaptive plasticity in coarse-grained environments: lower learning rates ensure that the population does not find short-term , non-plastic optima before the next environmental change , which allows the averaging of plasticity across environments . Since we define learning rates in biological systems as the amount of genetic change accumulated by the population in each new environment , they can be affected by several parameters other than rate of environmental change . Population size , mutation size and mutation frequency will all increase the amount of genetic change produced within each environment and thus increase the population’s learning rates . Stronger selective pressure will speed up the fixation of beneficial variants , and therefore also increase learning rates . If the learning rate explanation for the evolution of adaptive plasticity in coarse-grained environments is correct , these factors should be interchangeable with the rate of environmental change . For example , small populations or populations with low mutation frequency should be able to find long-term plastic solutions even when environmental change is rare . It is important to point out that decreasing population size or mutation frequency would instead hinder the action of lineage selection , which benefits from the maintenance of a large pool of genetic variants to select from . While a full exploration of all possible parameter space is beyond on the scope of this paper , we evaluate the learning theory explanation by testing the specific prediction that adaptively plastic responses can evolve even when environmental changes are slow , provided that mutation sizes are sufficiently small ( and hence learning rate is low ) . This question can be answered using the same model , and in particular the case of slow coarse-grained environments ( environments change every 4000 generations ) with a population size of 1000 individuals . As shown above , adaptive plasticity fails to evolve under these conditions . Learning theory explains this failure with the high learning rates in this population . Rather than decreasing the learning rate by decreasing the number of generation spent in each environment , we lower the standard deviation of mutation sizes from 10−2 to 10−4 . As we can see in Fig 5B , the population eventually evolves an optimally adaptive plastic reaction norm , with negligible amounts of variation around both slope and intercept . Their evolutionary trajectories ( Fig 5A ) are also qualitatively similar to those of populations evolving in fast , coarse-grained environments . In both scenarios , fitness to the current environment ( green ) fluctuates around average fitness to past environments ( blue ) , indicating that the populations are not evolving phenotypes that increase current fitness at the expense of past adaptation . The steady increase in average fitness to past environments instead indicates the evolution and retention of more general , plastic solutions . While the two trajectories are similar in shape , the population experiencing slower environmental changes and smaller mutation rates takes a significantly longer to reach optimal plasticity . An increase in the number of generations required to find solutions is a known consequence of lower learning rates . Intuitively , we can explain the longer time required to adapt as a consequence of the slower rate at which variants become available . While lineage selection is technically viable in this simulation , decreasing mutation sizes would also decrease the amount of available genetic variation , making it even less effective . A potential alternative explanation to our findings is that the reduced amount of genetic change per generation would enable multiple lineages to persist for longer , thus enabling the action of lineage selection . To test for this alternative explanation we run a simulation with K = 40000 and σμ = 10−5 using a hill-climber to model SSWM . The results are both qualitatively and quantitatively similar to those obtained in the previous simulation ( see Fig 6 ) . Since our results are unaffected by the absence of lineages , we can rule out that the observed evolution of plasticity with smaller mutation rates is due to the longer persistence of multiple lineages . Taken together , our simulations provide falsifying evidence for a number of frequent assumptions on the requirements for the evolution of costly adaptive plasticity in coarse-grained environments , which we summarize in Table 1 . The evolution of costly adaptive plasticity has often been framed as a necessity caused by environmental change outpacing the ability of natural selection to generate new adaptations [2 , 3 , 29 , 30] , but the process by which organisms achieve plasticity in these conditions have seldom been clarified . We demonstrate that neither individual nor lineage-level selection for adaptive plasticity are necessary for the evolution of adaptive plasticity . Rather , the speed of adaptation relative to environmental change ( modelled as learning rates ) is by itself a causal factor in the evolution of plastic responses that are adaptive across a range of coarse-grained environments . High learning rates allow optimization of phenotypes in each current environment , at the expense of more general solutions that improve their fitness across all environments experienced . Low learning rates instead make it impossible for phenotypes to chase short-term optima , yet allow individuals to reach long-term optimal plasticity despite the presence of short-term trade-offs . If approached from a purely adaptationist perspective , these results seem counter-intuitive: the conditions which allow natural selection to work most effectively ( high population sizes , high mutation rates , strong selective pressure and rare changes in the environment ) result in an evolutionary outcome ( adaptive tracking ) which has lower fitness than adaptive plasticity across all of our simulations ( see S1 Appendix ) . Conversely , changes in the same parameters that decrease the ability of natural selection to effectively cause phenotypic change result in an evolutionary outcome ( adaptive plasticity ) which maximizes fitness of the population in the long-term . We explain these counter-intuitive findings by using learning rates , a core concept of learning theory . Specifically , we demonstrate that low learning rates prevent populations from reaching short-term optima before a new environmental change occurs . This in turn allows evolved plastic reaction norms to be transferred across environments , so that they are effectively selected across multiple environments . The end result is that , as long as learning rates are sufficiently low , selection in coarse-grain environments converges on the same outcome as selection in fine-grained ones: adaptive plasticity . In learning theory terms , the cumulative effect of testing models sequentially on each individual example ( online learning ) will be the same as testing them on the entire set at once ( batch learning ) only if learning rates are low enough to prevent overfitting to the last example seen [31] . While low learning rates are necessary to evolve general solutions in the presence of trade-offs in performance , none of the factors that affect learning rates is necessary by itself . This is because learning rate is a composite measure , so any given factor may be offset by the others . We demonstrate this by showing that low mutation rate is sufficient to evolve costly adaptive plasticity even in slow , coarse-grained environments . Increasing population size and selection strength should instead decrease the odds of evolving costly adaptive plasticity , as both factors increase learning rates . As a consequence , even populations with no measurable genetic variation in plasticity could evolve adaptive plastic responses as long as ( 1 ) new genetic variation can be produced over time and ( 2 ) short-term optima change before natural selection can reduce plasticity to zero . This observation reverses the suggested causal link between plasticity and the rate of genetic evolution . Current theory proposes that plastic individuals experience weaker selection because they are able to cope with a wider range of environments [4] . Because of the reduced selective pressure , the amount of genetic change that accumulates in the population ( learning rate ) is also reduced . We instead suggest a low learning rate itself may skew populations towards evolving more general solutions , including plastic responses that are costly in current conditions but optimal across the entire set of previously experienced environments . As such , weak selection could facilitate the evolution of plasticity . Since low learning rates promote the evolution of adaptive plastic responses by reducing the relative importance of minimizing plasticity costs , they are irrelevant to the evolution of inexpensive plastic responses . When there are no costs of plasticity , every combination of slope and intercept that generates the optimal short-term phenotype is fitness equivalent within each environment . Because plastic and non-plastic solutions have the same short-term fitness , adaptive plasticity is selected for when the population moves towards the current phenotypic optimum and randomly drift after the optimal phenotype has been reached . The population will thus inevitably find the optimum for all past environments , and learning rates will only determine the speed at which the population reaches the optimum . Learning rates are likewise irrelevant for the evolution of costly adaptive plasticity in fine-grained environments , which are sufficient ( but not necessary ) for the evolution of adaptive plasticity across all our simulations ( see S1 Fig ) . Fine-grained environments allow natural selection to directly compare the fitness of phenotypes across multiple environments at the individual-level within each generation , so that adaptive plasticity is optimal even in the short-term . Direct selection for plasticity is unsurprisingly sufficient to ensure the evolution of adaptive plasticity . Under those conditions , learning rates can only determine the speed of selective process rather than its outcome . Our simulations consider the specific case of maintenance costs for plasticity . That is , we assume that plasticity directly decreases fitness , regardless of whether it is expressed . This assumption has a long history in modelling the evolution of plastic responses , but has been largely unsupported by empirical data which does not find costs of plasticity for the vast majority of traits analysed [32 , 33] . However , several alternative scenarios can create mathematically equivalent trade-offs between selection in current and past environments . A well-studied example is that of inaccurate cues , either due to imperfect perception or noise in the cues themselves [3 , 22 , 34] . Alternatively , the target phenotypes may not perfectly match with the best possible reaction norm . This scenario can happen for any reaction norm which is selected on a set of environments larger than its degrees of freedom ( 3 in the case of linear reaction norms ) [35] or if there are limits to the maximum amount of plastic changes that an organism can evolve [27 , 32 , 33 , 36] . In all of the above mentioned cases , optimal long-term plasticity would cause loss of fitness across current environments and consequently be selected against . Learning rates will thus be relevant for the evolution of plastic responses across all of them . In our simulations , mutations that lead to adaptive plasticity are selected since they increase phenotypic fitness within current environments , eventually causing the evolution of adaptive long-term plasticity . This is in contrast with lineage selection models , in which mutations that cause adaptive plasticity are selected because of their long-term benefits , but are ( at best ) selectively neutral in current environments . Since the evolution of plasticity in our model is driven by a short-term ( rather than lineage ) selection process , we predict it to be both faster and more robust to the presence of trade-offs . Similar dynamics apply to the evolution of modularity as a by-product of short-term phenotypic selection , and are proven to be scalable to arbitrarily complex systems [37] . From a learning theory perspective , low learning rates cause the evolution of adaptive plasticity because they constrain populations to evolve new adaptive solutions starting from previous genetic adaptations of the reaction norm rather than ‘from scratch’ . As a result , evolved reaction norms do more than just ‘remember’ which specific phenotype associated with each specific environment: they capture the logic that connects all cues to all phenotypes . In learning theory terms , organisms learn the regularities of the ( evolutionary ) problem , a process also known as generalization [31] . Therefore , as long as regularities remain the same , each individual will be able to produce adaptive phenotypes even in environments it has never experienced in its evolutionary history ( extrapolation ) , without the need for further adaptation . Conversely , several studies show that systems that learn a problem’s regularities are also able to quickly adapt to new problems which share a similar logic [38 , 39] . This ability to more rapidly evolve new adaptive phenotypes in response to new environments can instead considered as an increase in their evolvability . Our demonstration that organisms can learn regularities between environments even when each organism only ever experiences a single environment opens up the possibility that evolved plastic responses may both prepare organisms for future , more extreme , environments ( via extrapolation ) and enable them to more rapidly evolve new adaptive solutions ( via evolvability ) . This demonstrates that past evolution can shape evolutionary trajectories by biasing the phenotypic variants that are exposed to selection [24 , 40] . In summary , we use a simple reaction norm model to demonstrate that costly adaptive plasticity can evolve even when natural selection is unable to compare competing alleles over multiple environments ( i . e . , lineage selection ) . A learning theory framework helps us interpret this finding: Populations evolving in coarse-grained environments can evolve adaptive plasticity if the amount of adaptive change accumulated per environment—the learning rate—is low . Populations with high learning rates evolve via repeated short-term adaptation even if this pattern is maladaptive in the long term . Low learning rates facilitate adaptation to the entire set of environments experienced over adaptation to just the current environment , favouring adaptive plasticity even in the presence of short-term functional trade-offs . Thus , long-term adaptive plasticity can evolve even when it is not selected for at either the individual nor lineage level . Whether a population evolves phenotypes that optimize fitness in the short or long term instead depends on the amount of adaptive changes it accumulates within each environment .
For plasticity to evolve , the environment needs to fulfill two roles: determining the selective conditions ( selective role ) and providing information about those conditions ( constructive role ) [41] . We simulate the selective role by assigning each environmental state ( current or short-term environment ) a target single trait optimum ϕ , represented by a single real number . We simulate the constructive role by assigning each target optima an environmental cue represented by a real number C sampled from a normal distribution with mean 1 and standard deviation 1 . Each of our simulations cycles between 10 short-term environments , which make up the long-term environment . For simplicity , we consider a linear relationship between phenotypic targets and environmental cues , so that ϕ = g ( C ) = g1 * e + g0 . Hence , the targets are directly proportional to the respective cue . We choose g1 = −2 and g0 = 6 . This ensures that the relationship between selective environment and cues remains constant across environmental states . We assume that the lifespan of the individuals is fixed and the same for all . As a result , environmental grain is solely determined by the parameter K . K < 1 indicates fine-grained environmental variability , where the population encounters an average of 1/K environments per generation . On the other hand , K >= 1 indicates coarse-grained ( K = 1 ) or slow coarse-grained ( K > 1 ) environmental variability where the population encounters a new environment every K generations on average . We choose small K values compared with the total number of generations in our simulations so that each population is able to evolve for multiple environmental cycles . Our simulations were designed with temporal variation in mind , but the conclusions should be applicable to spatial variation as well . In fact , the environmental fluctuations described within our model match those experienced by a population in which all individuals migrate after fitness evaluation and before reproduction , or in which all propagules are dispersed to the same new environment . In this scenario environmental change rates are effectively interchangeable with migration rates , with other findings remaining unchanged . We model plastic responses using a univariate linear reaction norm model [42] . A reaction norm can be defined as the set of phenotypes that would be expressed if the given individual would be exposed to the respective set of environments . Since we consider univariate and linear reaction norms , we can describe the development of an organism’s phenotype as P = a + b * C . Each organism’s genotype can thus be described by the factors a and b . Of those , a determines the organism’s breeding value and b the direction and magnitude of its plasticity . We model the evolution of a population of asexual individuals as follows . First , we select a parent using a fitness proportional criterion [43 , 44] . Each individual can be selected with a probability of f / f ¯ , where f ¯ corresponds to mean fitness in the current population and f to the parent’s own fitness ( see section Evaluation of fitness for details on how we calculate f ) . Then , we generate a new individual with the same genotype ( reaction norm intercept a and slope b ) as the parent . Finally , we independently mutate both the offspring’s intercept and slope by adding a random value sampled from a normal distribution with mean μ = 0 and standard deviation equal to mutation size ( σμ = 0 . 01 unless otherwise specified ) . We repeat this process until we generate a number of offspring equal to the set population size . The parameters a and b are initialized at zero . Following previous work [35 , 37 , 38] , we define an organism’s overall fitness f in terms of a benefit-minus-cost function , which allows us to consider both positive ( benefits ) and negative ( costs ) contributions to its fitness . The benefit of a given genotype , b E i , for each environment , Ei , is determined based on how close the developed adult phenotype , Pa , is to the target phenotype , P E i , of the given selective environment , Ei . Since we deal with an univariate phenotype , we can calculate this amount as b E i = w ( P a , P E i ) = - | P a - P E i | , ( 1 ) where |*| corresponds to the absolute distance between the two phenotypes . Note that the selective advantage of respective genotypes is solely determined by its immediate fitness benefits on the currently encountered selective environment ( s ) . We consider that individuals experience a distribution of selective environments during their lifetime with occurring probabilities , q E 1 , q E 2 , . . , q E N . Each environment contributes to the selection process in proportion to its occurrence [45] . The overall fitness benefits of an individual over all experienced environments in its lifetime , bE is determined by the arithmetic mean of the fitness benefits in each environment , b E i , weighted by the occurrence , q E i , of each environment: b E = ∑ i q E i b E i . ( 2 ) In cases of coarse-grained environmental variability , where each individual encounters a single environment in its lifespan , q E i = 1 for the respective environment , i = j , and q E i = 0 for i ≠ j . On the other hand , in cases of fine-grained environmental variability , we assume a uniform distribution of environments experienced during individual’s lifespan , that is , q E i = 1 / K . The cost represents how maintaining plasticity reduces the organism’s fitness . Unlike the benefit , the cost of plasticity is a property of the genotype and does not change in different environments . Thus , we can calculate the overall performance , d , of a genotype over a range of selective environments as d = b E - λ | b | , ( 3 ) where parameter λ indicates how steeply fitness decreases in proportion to the reaction norm slope b . The final fitness score is calculated with the following formula: f = e x p ( d 2 ω ) , ( 4 ) which penalizes lower performances exponentially and re-scales them to a 0-1 range . ω is a scaling factor on the relation between f and d . Lower ω values cause greater loss of fitness per loss of performance , and correspond to steeper selection gradients . We choose ω = 0 . 2 , which corresponds to a scenario of strong selection ( see [38] ) . We evaluate the adaptive potential of the population due to plasticity by estimating how close the reaction norm of each individual in the population is to the ( theoretical ) optimal reaction norm . The optimal reaction norm here corresponds the function that given any environmental cue , C E i , produces the appropriate target phenotype , P E i , which best matches the current selective environment , Ei ( Evaluation of fitness ) . We evaluate the performance of reaction norms based on how different they are from the optimal reaction norm . The lack of fit , LackD of a given reaction norm , D , is estimated as a function of the phenotypic trait values in each of the past selective environments ( here 10 ) , Ei , whose magnitude increases quadratically with the distance from each phenotypic optimum , P E i: LackD=−ΣEi| D ( eEi ) −PEi |NE ( 5 ) Where NE stands for the number of past selective environments . The evaluation of lack of fit is performed for each individual at the end of each environmental period . We report the average and best performance in the population . A hill-climbing evolutionary model simulates a scenario of strong selection and weak mutation , where each new mutation is either fixed or lost before a new one can arise . Therefore , the entire population shares the same values of a and b . Each evolutionary step introduces a single mutant genotype with parameters a′ and b′ equal to a and b plus a random value sampled from a normal distribution with mean 0 and standard deviation equal to mutation size . We develop both the reference and mutant phenotypes P and P′ ( section Reaction norm model ) and compare their fitness values f and f′ ( section Evaluation of fitness ) . If f′ > f , the mutation is beneficial and therefore adopted so that at+1 = a′ and bt+1 = b′ . Otherwise , the mutation is deleterious and a and b remain unchanged . The code used to generate the results shown in this paper is provided in S1 File . | Organisms respond to different environments by changing how they act , look or function . When these responses improve the chances of survival , we call them adaptive plasticity . But observing adaptive plasticity does not prove that the response evolved because it improved survival . Being plastic is only selected for if individuals experience environmental variation , so that in slow changing environments plasticity may be selected against even if it is adaptive in the long term . Can adaptive plastic responses still evolve under these conditions ? Yes . We use learning theory to describe how genetic changes accumulate when individual lifespan is shorter than the time between environmental changes , and show that adaptively plastic responses can evolve even when they are selected against . This is because adaptive plastic responses can evolve as the by-product of selection for different functions in different environments , as long as organisms retain some plasticity until the next environmental change . Our work demonstrates that evolution can reach general solutions even when each individual is only presented with a simple fraction of a more complex problem . This intuition could explain why plastic responses to past environments can be adaptive even to environments the entire lineage has never seen before . | [
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Methods"
] | [
"organismal",
"evolution",
"evolutionary",
"rate",
"genetic",
"polymorphism",
"phenotypes",
"natural",
"selection",
"genetics",
"biology",
"and",
"life",
"sciences",
"evolutionary",
"adaptation",
"population",
"genetics",
"population",
"biology",
"evolutionary",
"biology",
... | 2019 | How adaptive plasticity evolves when selected against |
Apicomplexans are obligate intracellular parasites that invade host cells by an active process leading to the formation of a non-fusogenic parasitophorous vacuole ( PV ) where the parasite replicates within the host cell . The rhomboid family of proteases cleaves substrates within their transmembrane domains and has been implicated in the invasion process . Although its exact function is unknown , Plasmodium ROM1 is hypothesized to play a role during invasion based on its microneme localization and its ability to cleave essential invasion adhesins . Using the rodent malaria model , Plasmodium yoelii , we carried out detailed quantitative analysis of pyrom1 deficient parasites during the Plasmodium lifecycle . Pyrom1 ( - ) parasites are attenuated during erythrocytic and hepatic stages but progress normally through the mosquito vector with normal counts of oocyst and salivary gland sporozoites . Pyrom1 steady state mRNA levels are upregulated 20-fold in salivary gland sporozoites compared to blood stages . We show that pyrom1 ( - ) sporozoites are capable of gliding motility and traversing host cells normally . Wildtype and pyrom1 ( - ) sporozoites do not differ in the rate of entry into Hepa1–6 hepatocytes . Within the first twelve hours of hepatic development , however , only 50% pyrom1 ( - ) parasites have developed into exoerythrocytic forms . Immunofluorescence microscopy using the PVM marker UIS4 and transmission electron microscopy reveal that the PV of a significant fraction of pyrom1 ( - ) parasites are morphologically aberrant shortly after invasion . We propose a novel function for PyROM1 as a protease that promotes proper PV modification to allow parasite development and replication in a suitable environment within the mammalian host .
Malaria is a pervasive infectious disease that causes one million deaths each year and exacerbates the social and economic instability of endemic areas [1] . Malaria is caused by Plasmodium species , obligate intracellular protozoan parasites in the phylum Apicomplexa . Plasmodium spp have a complex life cycle with multiple differentiated forms that cycle between a sexual stage in the mosquito vector and an asexual stage in the vertebrate host . As obligate intracellular parasites , apicomplexans invade cells through the use of highly specialized secretory organelles , the micronemes and rhoptries [2] . Secretion from the micronemes is concurrent with apical reorientation and attachment of the parasite to the host cell membrane [3] . Tight apposition between the parasite and host cell plasma membrane forms the moving junction through the cooperation of the microneme adhesin AMA1 and the Rhoptry Neck proteins ( RON ) [4]–[6] . As the parasite pushes itself forward into the host cell , the moving junction , a constrictive ring that translocates posteriorly and forms a parasitophorous vacuole ( PV ) forms through the invagination of host cell plasma membrane . PV formation is accompanied by the secretion of rhoptry contents in the form of secretory vesicles that become incorporated into the nascent parasitophorous vacuole membrane ( PVM ) [7]–[9] . The PV is devoid of most host cell proteins and avoids fusion with host lysosomes [10]–[11] . Proper establishment and modification of the nascent PV is critical for the survival of the parasite . Sporozoites and merozoites are the invasive stages of Plasmodium parasites that form a PV as they enter their host cells , the hepatocytes and the red blood cells ( RBC ) , respectively . Until recently , studies of parasite invasion and development in hepatocytes have been limited . Much of what we know about apicomplexan host cell invasion and PV formation comes from studies using the model apicomplexan , Toxoplasma gondii , and from studies using Plasmodium falciparum erythrocytic stages [2] , [12] . It is likely , however , that sporozoites invade host cells in a manner similar to T . gondii tachyzoites as illustrated in a recent study showing the importance of host F-actin polymerization at the site of parasite entry for T . gondii tachyzoites and Plasmodium sporozoites [13] . Transcriptomic and proteomic analysis of Plasmodium sporozoites have enabled a gene-based approach to studying this important stage [14]–[17] . Sporozoites developing within the hepatocyte undergo a radical transformation within the first few hours post invasion . The intracellular sporozoite within its PV settles near the nucleus of host cells and modifies its long and polarized shape to become spherical [18]–[19] . The liver stage parasite grows and replicates within the PV ultimately releasing membrane bound bundles of thousands of daughter merozoites ( merosomes ) that then enter the erythrocytic cycle [20]–[21] . Pre-erythrocytic stages are considered targets for the development of vaccines and prophylactic drugs [22]–[23] . Recently , genetically modified live attenuated sporozoites that confer sterile immunity in rodent models were generated . The deleted genes , UIS3 , UIS4 , and P52 are upregulated in salivary gland sporozoites and are crucial for development within hepatocytes . Their gene products localize to the micronemes/secretory organelles of salivary gland sporozoites and are necessary for early liver stage development [24]–[26] . Host cell invasion by apicomplexans is associated with proteolysis of surface proteins , which include resident surface antigens and apically-secreted adhesion molecules , collectively called adhesins [27]–[30] . Proteases implicated in shedding of adhesins are parasite-encoded and include the subtilase class of serine proteases and the rhomboid class of serine proteases [31] . Rhomboid proteins are polytopic membrane-associated serine proteases that form a catalytic pocket within the lipid bilayer and have the unique characteristic of cleaving substrates within the transmembrane domain [32]–[34] . Rhomboid proteases recognize small residues such as glycine and alanine within the transmembrane domains of substrates [35] . In the Apicomplexa , processing of key microneme adhesins , such as TgMIC2 , TgMIC6 , TgAMA1 and PfEBA175 , has been shown to occur via intramembranous cleavage at sites predicted to be rhomboid-like substrates [36]–[38] , [29] . The specific rhomboid protease ( s ) involved in these processes have not been identified . Based on expression , localization , and substrate specificity , two Plasmodium rhomboids , ROM1 and ROM4 , are predicted to play a role during invasion . PfROM1 is co-localizes with microneme markers has the ability to cleave certain microneme invasion adhesins such as PfAMA1 , PfMAEBL , certain EBLs , and rhoptry proteins of the Rh family in a cell based assay [29] , [39] . In addition , a separate study reported that PfROM1 is localized to a new apical organelle , the mononeme [40] . When expression of the PfROM1 ortholog in Toxoplasma gondii , TgROM1 , is down regulated in tachyzoite parasites , there is a slight growth phenotype and mild invasion defect compared to the wildtype control [41] . Disruption of pbrom1 in Plasmodium berghei led to attenuation ascribed to an invasion defect [42] , however , invasion of parasites into host cells was not directly tested in this study . It is still unclear if ROM1 plays a role during Plasmodium parasite invasion . Using the rodent malaria parasite , Plasmodium yoelii , we investigated the specific role of ROM1 during the parasite lifecycle . We show that ROM1 is not required for entry into host cells but , instead , ROM1 is necessary shortly after the parasite has entered the host cell to promote fitness and parasite survival .
Pyrom1 was cloned from cDNA of Plasmodium yoelii 17XNL mixed blood stages using 5′ and 3′ Rapid Amplification of cDNA Ends ( Smart RACE Clontech ) . The sequence obtained consists of 4 exons and 3 introns , encompassing two annotated genes on PlasmoDB , py00729 and py00728 . Based on topology predictions ( TMHMM and HMMTOP ) , PyROM1 has seven transmembrane domains with the canonical rhomboid catalytic serine motif ( GASTS ) found within transmembrane domain four and a conserved histidine found within transmembrane domain six ( merops . sanger . ac . uk ) . It has an N-terminal tail of 52 amino acids that includes the conserved microneme targeting motif YPHY [43] and a very short carboxy terminal tail ( Figure S1 ) . Based on the P . falciparum DNA microarray data , expression of pfrom1 is similar to genes involved during invasion with a significant upregulation in the sporozoite stage [44]–[45] . We quantified pyrom1 mRNA in several stages of the malaria life cycle using quantitative RT-PCR ( Figure 1A ) . Amplification of Pyrom1 cDNA from synchronized erythrocytic stages shows modest expression , with greatest expression in schizont ( S ) stages . There is a 10-fold increased expression in midgut ( MG ) sporozoites and a 20-fold increased expression in salivary gland ( SG ) sporozoites relative to expression levels of schizont stages . To control for appropriate expression analysis , additional expression profiles of other known genes ( PyAMA1 , PyUIS3 , PyADA , PyTUB1 , and PyCSP ) was carried out using the same cDNA ( Figure S2 ) . To further characterize pyROM1 expression at the protein level , we generated a transgenic line that expresses pyROM1 tagged at the N-terminus with a triple hemaglutinin tag ( 3xHA ) . Single cross-over homologous recombination using hDHFR selection cassette yields the expression of the HA-tagged pyrom1 open reading frame driven by its own 5′ and 3′ regulatory elements ( Figure 1B ) . Successful integration of the knock-in construct was verified by Southern blot analysis of genomic DNA from mixed erythrocytic stage transgenic parasites R1HA ( Figure 1C ) . Western blot analysis of erythrocytic and sporozoite stages of PyROM1-HA reveals a band that migrates at the expected size of ∼30 kDa on SDS-PAGE ( Figure 1D and 1E ) . In synchronized erythrocytic stages , PyROM1-HA protein expression is lowest in ring stages ( 3 hours post invasion ) and increases during the course of development with a peak in schizont stages ( Figure 1D ) . In sporozoite stages , PyROM1-HA is readily expressed in salivary gland sporozoites with very minimal expression in midgut sporozoites . The expression pattern of PyROM1 at the mRNA and protein level implies a function at multiple stages of the parasite life cycle with a particular importance for the zoite stages that mediate host cell invasion . PyROM1-HA co-localizes with the microneme adhesin , pyMAEBL and with the rhoptry neck protein RON4 at the apical end of merozoites within mature schizonts by immunofluorescence ( Figure 2A ) . Although localization overlapped more consistently with pyMAEBL than RON4 , pyROM1 shows partial co-localization with both markers . Some localization overlap is observed between PyROM1-HA and the endoplasmic reticulum ( ER ) marker BiP ( Figure 2A ) . PyROM1-HA is found in a speckled intracellular pattern in salivary gland sporozoites ( day 14 ) that partially overlaps with localization of PyMAEBL and PyUIS4 ( Figure 2B ) . Some co-localization is observed with BiP in salivary gland sporozoites ( Figure 2B ) . During sporozoite invasion into Hepa 1–6 cells , PyROM1 remains prominent with a speckled pattern diffusely dispersed throughout the sporozoite ( Figure 2C ) . Expression of pyROM1-HA is observed in developing EEFs at four hours post invasion with intracellular localization throughout the length of the parasite that is distinct from the localization of the sporozoite surface antigen , circumsporozoite protein ( CSP ) ( Figure 2C ) . To investigate the importance of pyrom1 in the parasite life cycle , we employed a reverse genetics approach to generate loss-of-function deletion mutants . Initially we used a single crossover homologous recombination step to disrupt the endogenous pyrom1 locus . Successful disruption of the pyrom1 gene was confirmed by Southern blot and RT-PCR ( Figure S3 ) . Disrupted parasites ( R1INT ) were used to carry out initial phenotypic screens . Because single crossover disruptants can revert to wildtype [46]–[47] we also created a deletion mutant that would allow us to cycle the parasites between the mammalian host and mosquito vector without the risk of reversion . We used a gene replacement vector to exchange the endogenous pyrom1 gene for a pbDHFR/TS-GFP cassette by double crossover homologous recombination [48] . The targeting vector contains the pbDHFR-TS/GFP selection cassette , which confers pyrimethamine resistance , flanked by DNA fragments from the upstream ( 5′ARM ) and downstream ( 3′ARM ) regions of the predicted open reading frame of pyrom1 . We successfully integrated the replacement cassette , deleting exons 1–3 of the pyrom1 gene that include the region encoding the functional catalytic serine motif and the conserved histidine ( Figure 3A ) . Recombinant parasites were detected by diagnostic PCR of genomic DNA from wildtype and transfected parasites . Cloning of pyrimethamine-resistant parasites yielded two knock out clones ( R1KO-1 and R1KO-2 ) and a wildtype clone ( Wt Ctrl ) . Southern blot analysis of genomic DNA from mixed blood stages confirmed successful cloning of pyrom1 deletion mutants ( R1KO ) ( Figure 3B ) . RT-PCR analysis of mutant clones revealed no detectable pyrom1 transcript expression , whereas pyrom1 expression was detected in the wildtype control clone ( Figure 3C ) . The successful generation of P . yoelii 17XNL parasites deficient in pyrom1 demonstrates that this gene is not essential for proliferation of the intra-erythrocytic stages . To test whether PyROM1 serves an important function during asexual growth within red blood cells , we performed in vivo infectivity assays . Six-week old female BALB/c mice ( n = 5 ) were injected intravenously with 1×104 infected red blood cells ( iRBCs ) and parasitemia was monitored every 24 hours by counting Giemsa-stained blood smears . Since the pyrom1 ( - ) parasites were generated in P . yoellii 17XNL , a non-lethal strain , the parasitemia in mice reached a peak and parasites were then cleared by the mouse immune system . We monitored parasitemia until infected erythrocytes were undetectable . Growth curves revealed a mild attenuation of pyrom1 ( - ) ( R1KO ) with a decrease in peak parasitemia ( 25% versus 43% ) and a decrease in the duration of infection compared to the wildtype control ( Wt Ctrl ) ( Figure 4A ) . Infection of BALB/c mice ( n = 5 ) with salivary gland sporozoites by natural mosquito bite or by intravenous injection ( inoculum of 20–2000 ) revealed that mutant parasites could establish a liver infection leading to blood stage infection with a pre-patent period of three days ( Table S1 ) . Therefore , pyrom1 ( - ) parasites are able to progress through the entire lifecycle . Due to the highly infectious nature of P . yoelii sporozoites in vivo , it is difficult to quantify liver development by relying solely on pre-patency period [49] . Therefore , we analyzed the development of pyrom1 ( - ) sporozoites in vivo using quantitative RT-PCR , a more sensitive assay to quantify hepatic development [50] . BALB/c mice were injected intravenously with equal numbers ( 1×104 ) of salivary gland sporozoites from wildtype control or pyrom1 ( - ) parasites . Livers were harvested at 36 hours and RT-qPCR analysis was carried out . Parasite burden in mice infected with pyrom1 ( - ) parasites was decreased by at least 60% compared to mice infected with wildtype parasites ( Figure 4B ) . Similar results were seen with livers harvested at 24 hours and 42 hours from mice infected with 104 salivary gland sporozoites ( data not shown ) . This difference in liver burden suggests that pyROM1 is important for efficient sporozoite infection in the liver . Since pyrom1 is robustly expressed in sporozoite stages , we investigated the effect of pyROM1 depletion in the mosquito phase of the parasite life cycle . Analysis of blood smears showed no difference in the capacity to produce gametocytes between wildtype and R1KO parasites ( Table 1 ) . No significant difference was seen in the number of oocysts ( day 8 post infection ) per mosquito midgut or in the prevalence of infected mosquitoes between wildtype and R1KO parasites ( Table 1 ) . Furthermore , there was no difference in the number of salivary gland associated sporozoites ( day 14 post infection ) in R1KO mutant parasites compared to wildtype in three independent experiments , testing two R1KO clones ( Table 1 ) . Thus , deletion of pyrom1 does not affect development of oocysts , sporozoite release into the hemolymph , or invasion of salivary glands . A decrease in liver stage infection in vivo can be attributed to a defect in sporozoite motility or cell traversal ability [51]–[55] . Potential rhomboid substrates such as the microneme adhesin , TRAP , bridge the function of gliding motility and host cell invasion [56]–[57] . To begin an in-depth analysis of the function of pyrom1 in salivary gland sporozoites , we tested gliding motility and cell traversal of pyrom1 ( - ) parasites . Gliding trails formed by the deposition of circumsporozoite protein ( CSP ) on glass slides by sporozoites were readily detectable in the pyrom1 disrupted ( R1INT ) parasites ( Figure 5A ) and the quantity and quality of trails produced by wildtype and R1INT sporozoites were similar ( Figure 5B ) . A unique feature of salivary gland sporozoites is their ability to use the motility machinery to traverse cells without productively invading a host cell and forming a PV [58] . When sporozoites traverse cells they glide in and out of the cell puncturing the plasma membrane and causing host cell injury . Host cell traversal can be detected by quantifying host cells that take up non-permeable high molecular weight FITC-dextran , in the presence of migrating sporozoites [59] . Using this assay , there was no difference in cell traversal activity between wildtype or R1KO sporozoites ( Figure 5C ) . To test the importance of pyROM1 during host cell invasion , we performed a double labeling invasion assay that distinguishes parasites that are intracellular versus parasites that are extracellular [60]–[61] . The number of intracellular sporozoites between wildtype and pyrom1 ( - ) parasites did not differ at two hours post invasion ( Figure 5D ) . Invasion rates at 10 , 20 , and 40 minutes after loading onto Hepa1–6 monolayers did not differ significantly between Wt Ctrl and R1KO , illustrating no significant difference in the kinetics of invasion between wildtype and R1KO sporozoites into Hepa1–6 cells ( Figure 5E ) . Together , these results indicate that pyROM1 is not important for efficient host cell entry or for establishment of a successful infection during the early stages of sporozoites in host cells . Analysis and quantification of sporozoite development in Hepa 1–6 cells at 6 , 12 , and 24 hours post invasion was carried out using fluorescence microscopy . Differential red-green staining of pyCSP was carried out to distinguish intracellular versus extracellular parasites at 6 hours post invasion . By 6 hours , exoerythrocytic forms ( EEF ) have begun to round up and change morphology as a prelude to development in hepatocytes . The percent of R1KO developing sporozoites in Hepa 1–6 cells was decreased at 6 hours compared to wildtype sporozoites ( Figure 6A ) but this difference was not statistically significant ( P-value 0 . 17 ) . At 12 hours , the number of developing foci of R1KO exoerythrocytic forms ( EEFs ) had significantly decreased by 45% compared to the number of wildtype EEF foci ( P-value 0 . 008; Figure 6B ) . By 24 hours post infection , R1KO EEF development was decreased by more than 60% compared to wildtype EEF development ( P-value 0 . 001; Figure 6C ) . This time course analysis of EEF development suggests that pyROM1 enhances the survival of developing sporozoites within hepatocytes during the critical stages of early intra-hepatocytic development . Although the number of developing EEFs decreased with each time point , the R1KO EEFs that were observed still exhibited morphological characteristics of parasites that had progressed appropriately during development . For example , these EEFs exhibited rounding up of the elongate sporozoite , increase in cell size , and juxta-nuclear position within host cell ( Figure 6D ) . To determine the fate of the fraction of parasites that survive past the initial 24 hours post infection , we allowed parasites to develop in Hepa1–6 for 40 hours . By this time , the EEF has grown in size and has undergone several rounds of multiplication , becoming a liver schizont ( i . e . mature EEF ) . Liver schizonts were analyzed microscopically by staining with CSP to determine if pyROM1 plays a role during growth and later development stages of EEF maturation . Quantification of EEF area reveals that R1KO parasites that survived the initial stages of development are able to proceed with normal growth and cell division ( Figure 6E ) . Proper establishment and modification of the PV in newly invaded parasites is a prerequisite for intracellular survival and growth . The early transcribed membrane proteins ( eTRAMPs ) are highly charged type I integral membrane proteins that localize to secretory organelles and associate with the PVM shortly after parasite invasion [15] , [62] . The eTRAMP UIS4 ( Upregulated in Sporozoites 4 ) is specifically upregulated in salivary gland sporozoites and is essential for parasite development in host hepatocytes [24] . In salivary gland sporozoites , UIS4 co-localizes with TRAP to secretory organelles [15] and is secreted at some point post invasion ( within 2 hours ) to become incorporated into the PVM [24] , [63] . To assess whether the defect in pyrom1 ( - ) parasite development in host hepatocytes was due to a defect in PV formation , we tested for the presence of PyUIS4 . Sporozoites within hepatocytes were labeled with CSP mAb and UIS4 antisera at 2 hours and 6 hours post infection and quantified microscopically . The total number of parasites ( intracellular and extracellular ) was determined by CSP staining . The percent of developing parasites surrounded by a UIS4-positive PVM was calculated as the number of parasites that showed double staining with UIS4 and CSP . R1KO parasites had a 33% and 43% decrease compared to wildtype parasites of UIS4 positive staining at 2 hours and 6 hours , respectively ( P-value 0 . 010 ) ( Figure 7A ) . Pyrom1 ( - ) parasites negative for the circumferential staining typical of UIS4 localization to the PVM , often times had a punctuate localization of UIS4 within the sporozoite body ( Figure 7B ) . The eTRAMP proteins possess a conserved TMD with sequence similarity to known rhomboid substrates [32] . To test if the TMD of UIS4 is a potential rhomboid substrate we carried out a previously established heterologous cleavage assay where rhomboid and substrate are transiently co-expressed in COS7 cells and analyzed via Western blot [39] . A truncated version of PyUIS4 encompassing the full TMD with an N-terminal GFP tag and an IgK leader sequence ( UIS4TMGFP , Figure 7C , Table S3 ) was co-expressed with HA-tagged rhomboid protease . Proper expression of rhomboid proteases was observed in Western blots probed with anti-HA antibody ( Figure 7D ) . A single band running at ∼40 kDa corresponding to full length UIS4TMGFP was observed in Western blots of cell lysates probed with anti-GFP antibody ( Figure 7D ) . A lower molecular weight band ( ∼34 kDa ) corresponding to the expected size of a rhomboid cleaved product was observed in cell lysates and conditioned media only when UIS4TMGFP was co-expressed with the D . melanogaster rhomboid protease , DmRho-1 ( Figure 7D ) . No cleavage activity was readily observed when UIS4TMGFP was co-expressed with either TgROM5 or its catalytic mutant ( TgROM5SA ) ( Figure 7D ) . Therefore , it can be concluded that PyROM4 , and potentially other eTRAMPs , can serve as substrates to rhomboid proteases . Our analysis of UIS4 in Pyrom1 ( - ) parasites show that there is an inability to properly secrete UIS4 into the PVM . Furthermore , a rhomboid protease may be involved in the proper maturation , processing and targeting of PyUIS4 to the PVM since the TMD of UIS4 serves as a rhomboid substrate . To investigate whether pyrom1 ( - ) parasites have a defect in PVM formation , we carried out ultrastructural analysis of R1KO parasites during early intra-hepatocytic development . Intracellular parasites at 4 hours post infection of Hepa1–6 cells were analyzed by electron microscopy . R1KO parasites ( 19 of 42 ) exhibited a strikingly reduced intravacuolar space with the PVM in close apposition to the plasma membrane of the parasite . In wildtype vacuoles this parasitophorous vacuole space has expanded and appears as a white “halo” surrounding the developing parasite inside a PVM and only one of 38 vacuoles had a closely apposed PVM Figure 8A , Figure S4 , Table S2 ) . To quantify this phenotype we measured the area of the PV space ( halo ) as the difference between the area of the total PV space and the parasite area . This value was then normalized to the area of the parasite to give a ratio that represents the percentage of the PV space area relative to the parasite and is a representation of the extent of PV space expansion . Wildtype parasites had a significantly ( P-value 0 . 0007 ) larger ratio value ( 0 . 17±0 . 02 ) compared to R1KO parasites ( 0 . 09±0 . 01 ) ( Figure 8B ) . To quantitatively show the existence of two populations , we plotted the distribution of parasites as a function of PV space area/parasite area ratio . The distribution plot clearly shows the existence of two parasite populations within the R1KO group ( Figure 8C ) . We hypothesize that the R1KO parasites with a normal PV space go on to develop past the initial 24 hours of development and the R1KO parasites with a reduced PV space abort development within the initial 24 hours . A few parasites in direct contact with the cytoplasm or nucleoplasm were observed in both wildtype ( 4/38 ) and R1KO ( 2/42 ) parasites , which we assume be sporozoites in traversal mode [64] . This ultrastructural analysis indicates that while R1KO parasites are capable of productively invading host cells with the formation of a PVM , a fraction of them have a defect in the subsequent PV expansion and modification in early development .
Successful transmission of malaria into the mammalian host is dependent on the ability of sporozoites to invade and establish a proper PV within the host hepatocyte . To date , only a handful of genes have been identified that play a role during invasion or early development of the sporozoite within hepatocytes . In this study we have characterized the rhomboid protease , ROM1 , throughout the lifecycle of the malaria rodent model , Plasmodium yoelii . Quantitative expression analysis of pyrom1 shows it is expressed at various invasive stages of the malaria life cycle . Expression of pyrom1 follows a pattern similar to genes involved in merozoite invasion with maximal erythrocytic stage expression in schizonts . Relative to schizont stages , expression of pyrom1 is increased by at least 10-fold during the sporozoite stages with pyrom1 transcript levels upregulated by 2-fold from midgut sporozoites to salivary gland sporozoites . Midgut sporozoites are substantially less infectious in the mammalian host compared to salivary gland sporozoites [14] , [65] . This gain of infectivity has been ascribed in part to transcriptional changes that occur when sporozoites invade mosquito salivary glands [14]–[15] . In addition , we observe that , although transcript expression of pyrom1 is elevated in midgut sporozoites , protein expression is barely detectable by Western blot , suggesting regulation of pyrom1 gene expression at the level of transcript translation in midgut sporozoites . The upregulated expression of ROM1 mRNA and protein in salivary gland sporozoites is consistent with a role during infectivity and transmission into the vertebrate host . Supporting our observations , a microarray study looking at gene expression changes triggered by different host environments ( i . e . mosquito host to mammalian host ) reported that pfrom1 is upregulated 4-fold when salivary gland sporozoites are shifted to 37°C in the presence of hepatocytes [66] . We have localized PyROM1 at various stages during the malaria life cycle . In schizonts , PyROM1-HA is localized to the apical end . In salivary gland sporozoites PyROM1-HA has a diffuse granular staining pattern similar to microneme proteins such as PyMAEBL . Microneme localization of PyROM1 agrees with previous studies of ROM1 localization in both Toxoplasma gondii and Plasmodium spp [67]–[68] , [41]–[43] , [29] , but in our study this localization was only partial . Localization of pyROM1-HA in sporozoites during invasion revealed that it remains intracellular and does not seem to be secreted onto to the surface during invasion . There is no consensus on the localization of ROM1 during apicomplexan invasion since some studies report surface expression and posterior translocation [40] , [43] and others report internal localization during invasion [29] , [41] . Because of differences in the experimental conditions and expression constructs used for these localization studies , results are difficult to compare . Using a gene deletion approach , we demonstrate that pyROM1 functions during intracellular growth , within hepatocytes and erythrocytes , to provide a fitness advantage . Quantitative analysis of parasite development within Hepa1–6 cells revealed that PyROM1 is not essential for sporozoite invasion into the host cells . Instead , survival of pyrom1 ( - ) parasites decreases within the first 24 hours post invasion . It is during these first 24 hours of development that Plasmodium sporozoites undergo critical morphological changes such as PV modification , sphericalization of the elongate sporozoite , and increase in cytoplasmic size [18]–[19] , [69] . PyROM1 may facilitate these initial vital steps during liver stages by allowing parasites to reach the critical threshold required to survive past the first 24 hours of differentiation . This in vitro development phenotype was confirmed in vivo when mice infected with pyrom1 ( - ) sporozoites had decreased parasite liver burden at 36 hours post infection . Our results agree with previous studies showing that lack of ROM1 expression causes a decrease in parasite survival [41]–[42] . In a previous study , disruption of pbrom1 resulted in attenuation during erythrocytic stages , a decrease in oocyst formation , and a decrease in liver stage burden [42] . One major difference between the current study and the Srinivasan et al study is that we did not detect a decrease in the number of oocyst during development in the mosquito midgut . This difference in phenotypic observation can be attributed to the different methodologies used to analyze this stage . Oocyst development is biologically variable due to multiple factors such as host immunology ( mosquito and mammalian ) , parasite variability , biological bottle necks , and other environmental conditions . Therefore , it is imperative that careful and repeated quantification and analysis of oocyst development be carried out in order to obtain statistically reproducible results . We have analyzed oocyst development in pyrom1 ( - ) parasites thoroughly through multiple repeated experiments ( at least three independent experiments ) performed simultaneously with wildtype controls , using different batches of mosquitoes , and multiple parasite clones . We have analyzed the development of pyrom1 ( - ) parasites in hepatocytes thoroughly by using well-established in vitro development and invasion assays . Thus our studies have extended our understanding of ROM1 function and have pinpointed more precisely the role of ROM1 during intracellular development . Brossier et al , studied ROM1 in Toxoplasma gondii tachyzoites and found that TgROM1 knockdown parasites had a 50% reduction in the number of daughter tachyzoites per vacuole [41] . Based on measurement of EEF area , we did not detect a defect in pyrom1 ( - ) parasite growth per vacuole , but an overall decrease in the number of developing parasites . Therefore , we predict that the intracellular parasites abort development in the stages preceding multiplication . This difference in phenotype between the TgROM1 knockdown parasites and our pyrom1 ( - ) parasites may be explained by fundamental biological differences in the mode of replication between the two genera . Toxoplasma gondii tachyzoites divide by endodyogeny , a process where two daughter cells are formed within a mother cell , whereas Plasmodium parasites divide by schizogony a process where nuclear division leads to the formation of a multinuclear syncytium followed by the budding off of daughter merozoites at the periphery [70] . Thus , the events that occur post-invasion may have differing effects on the subsequent survival and development of the respective parasites [71] . The malaria parasite has four invasive stages: the ookinete , the midgut sporozoite , the salivary gland sporozoite , and the erythrocytic stage merozoite . Only two of these invasive zoites , the merozoite and the salivary gland sporozoite , form a PV , within which , growth and division of the parasite ensues . The phenotype displayed by pyrom1 ( - ) parasites is only observed during the two stages where PV formation is a prerequisite for growth . The PVM provides a barrier that protects the parasite from host cell defenses such as lysosomal clearance or autophagy [72]–[73] . This barrier is also the portal for nutrient acquisition and communication with the external environment . The intravacuolar parasite must modify and rearrange the PV through the secretion of contents from rhoptries and dense granules , with molecules that are yet to be fully characterized . The exclusivity of the pyrom1 ( - ) phenotype during intracellular development within the mammalian host suggests that ROM1 function is linked to proper PV formation and maturation either during or post invasion . Ultrastructural analysis at four hours post invasion reveals that pyrom1 ( - ) parasites are capable of forming a PVM during hepatic development . But , the PVM in the pyrom1 ( - ) parasites is intimately associated to the plasma membrane of the mutant parasite . In comparison , after invasion , wildtype parasites have substantially expanded their PV space , which is visible as a white ‘halo’ surrounding the intracellular parasite . When parasites first enter host cells , they have a tight fitting vacuole . Shortly after invasion , the vacuolar space expands , representing a modification of the PV . This modification is accompanied by targeting of parasite proteins to the PVM as well as changes in the lipid composition of the specialized vacuolar membrane [12] . Notably , a fraction of pyrom1 ( - ) parasites had a PV space similar to the wildtype parasites . We hypothesize that these are the parasites go on to develop normally into liver schizonts . These parasites may be able to progress due to a redundant function of other proteases such as ROM4 or ROM8 , which are also expressed in sporozoites [44] , [66] . Thus , pyrom1 ( - ) parasites display a partial penetrance phenotype where only a fraction ( 50–70% ) of the clonal population display a mutant phenotype and the remainder fraction go on to develop normally . In addition to redundant function and partial complementation by another protease , this partial penetrance phenotype is explained by a stochastic mechanism where levels of processed substrate ( s ) must reach a threshold level to allow normal parasite development [74]–[76] . In systems biology , such a stochastic mechanism occurs when a gene encodes for a non-essential phenotypic capacitor that serves to buffer various development , environmental , and genetic stresses . Since PyROM1 encodes an enzyme , it is subject to random biomolecular interactions that promote phenotypic variation . A previous study showed that intracellular parasites , deficient in two sporozoite specific proteins , p52 ( p36p ) and p36 , were negative for UIS4 staining at the PVM [63] . Like pyrom1 ( - ) parasites , mutant parasites in p36/p52 have normal gliding , traversal , and hepatocyte invasion . However , unlike pyrom1 ( - ) parasites , which are capable of establishing a PVM , p52 mutant and p36/p52 double mutant parasites failed to develop inside the hepatocyte due to an inability to establish or maintain a PVM during early infection [26] , [63] , [77]–[79] . Interestingly , pyrom1 ( - ) parasites have a 30% decrease in UIS4 staining at 2 hours and 6 hours post invasion , despite their ability to invade cells normally . If the PVM is established in conjunction with invasion , then the absence of UIS4 in the pyrom1 ( - ) parasite vacuoles reflects a defect in the targeting of UIS4 to the PVM shortly after invasion and not necessarily a defect in establishing a PV . The mechanisms by which UIS4 or other eTRAMPS are secreted and associate to the PVM are still unknown . A previously published analysis suggested eTRAMP proteins have a TMD with sequence similarity to the canonical rhomboid substrate , Spitz [32] . Here , we show for the first time that the transmembrane domain of an eTRAMP , PyUIS4 , serves as a rhomboid substrate to DmRho-1 . Unfortunately , we were not able to express a catalytically active pyROM1 ( data not shown ) to test activity against candidate substrates , including known substrates of the PyROM1 homolog , PfROM1 such as AMA1 or Spitz [39] . Therefore , we could not determine directly whether PyROM1 cleaves UIS4 . Little is known about the dynamic processes that occur just after invasion when the parasite has entered a new environment and must undergo dramatic changes . It is well established that cells sense changes in their environment from external cues that activate a signaling cascade through an internal sensor such as a surface receptor . A function in the activation of a signaling molecule has been described for rhomboid proteases in several species [80] . A recent study has attributed such a function to rhomboid protease activity in the parasite Toxoplasma gondii [81] . In this study , TgROM4 dominant negative mutants defective in replication were rescued by the over-expression of the cytoplasmic tail of TgAMA1 or PfAMA1 that resembled the rhomboid cleavage product [81] . Based on these results , intramembrane proteolysis of AMA1 is hypothesized to link the switch from invasion to replication . A biological role for rhomboid processing of AMA1 in Plasmodium spp has not been established since it is a minor event that is enhanced when the subtilisin protease PfSUB2 , is inhibited [30] , [38] , [82] . Given that PfROM1 preferentially cleaved PfAMA1 relative to the TgROM4 ortholog , PfROM4 [39] and our current data , it is possible that ROM1 plays a role in a similar process in Plasmodium species . A separate study , using conditional knockdowns , showed TgROM4 functions as a sheddase of surface adhesins during gliding motility and invasion [83] . Our studies demonstrate that ROM1 does not directly function as a sheddase during cell entry , but instead is important in the stages just after invasion of hepatocytes . Therefore , as proposed for TgROM4 [81] , we hypothesize that ROM1 activity links parasite invasion of host cells with parasite development within host cells . In a second model , ROM1 may function as a secretase that promotes the secretion of proteins such as UIS4 that are targeted to the PV or PVM for modification . A secretase activity may also serve to activate the secretion of a signal for parasite differentiation . In both models , ROM1 processing activity could occur prior to or after invasion . Further studies to identify substrates of ROM1 are needed to understand ROM1 function . Most importantly , this study provides new insights into the function of a rhomboid protease during intracellular development of Apicomplexan parasites .
For routine passage of blood stage Plasmodium yoelii 17XNL parasites and for mosquito feedings , Swiss Webster mice ( female , 4–5 weeks ) were used . Mice were infected by either intraperitoneal or intravenous injections . For parasite infectivity assays such as liver parasite burden or blood stage infectivity , BALB/c mice ( female , 6 weeks of age ) were used . Blood stage parasites were harvested by intraocular bleeding of infected mice . All animals were purchased from either Charles River or Taconic . Handling of mice and rodent malaria infections were conducted in accordance to approved protocols by the institutional animal use committees at Albert Einstein College of Medicine ( protocol 20081001 ) and New York University School of Medicine ( protocol 090809-02 ) , in facilities approved by the Association for Assessment and Accreditation of Laboratory Animals . Protocols followed the recommendations of the Guide for the Care and Use of Laboratory Animals of National Institutes of Health Office of Laboratory Animal Welfare . Anopheles stephensi mosquitoes were reared at 27°C and 80% humidity under a 12/12 h light/dark cycle , and adults were fed on 20% sucrose solution . Three to five- day old mosquitoes were used for feeding on infected mice with P . yoelii 17XNL and maintained at 24°C , 80% humidity . All cells used were grown in Dulbecco's modified Eagle's medium supplemented with 10% heat inactivated fetal bovine serum , 1 mM Glutamine , 100 U Penicillin/ml , 10 µg/ml streptomycin , and maintained at 37°C and 5% CO2 . The gene encoding pyrom1 was cloned by RT-PCR using total RNA from Plasmodium yoelii 17XNL mixed blood stage parasites extracted with TRIzol reagent ( Invitrogen ) followed by DNAse treatment ( Ambion ) . Full length sequence of the pyrom1 cDNA was amplified via 5′ and 3′ Rapid Amplification of cDNA Ends ( SMARTRACE , Ambion BD Biosciences ) using primers pyR1RACE5′R and pyR1RACE3′F . A touchdown PCR was carried out to increase the specificity of the PCR reaction . The PCR reaction products were separated by agarose gel and PCR products of various sizes were gel extracted ( Qiagen ) and cloned to a Topo vector ( Invitrogen ) for sequencing . Gene sequence of the open reading frame was further confirmed via RT-PCR using primers , pyR1 . 1-F and pyR1 . 1-R , specific to the deduced 5′ and 3′ ends of the pyrom1 open reading frame . Exon/intron boundaries were confirmed to follow the conserved splicing motifs . Functional prediction of the translated amino acid sequence was obtained by using the Translate tool on the Expasy website ( www . expasy . org ) . Prediction of transmembrane domains was performed using the TMHMM Server v . 2 . 0 ( http://www . cbs . dtu . dk/services/TMHMM-2 . 0/ ) and HMMTOP ( http://www . enzim . hu/hmmtop/ ) . Primer sequences are listed in Table S3 . Total RNA was prepared from synchronized Plasmodium yoelii parasites at various time points during development . Erythrocytic stages were synchronized by injecting 1×108 purified schizonts into 5 wk old Swiss Webster mice ( Charles River ) by tail intravenous injection [84] . Specific synchronized erythrocytic stages were confirmed by Giemsa stained blood smears and collected at specified time points to obtain a ring stage , trophozoite stage , and schizont stage parasites . Mice infected with synchronized parasites were fully bled into heparinized complete medium . Infected blood was saponin lysed , centrifuged at 4°C , and the resulting parasite pellet was resuspended in 500 µl of TRIzol reagent ( Invitrogen ) . To collect RNA from mosquito stage sporozoites , midguts or salivary glands were dissected from 100 infected ( P . yoelii 17XNL ) A . stephensi mosquitoes at 10 days post blood meal ( midgut sporozoites ) or 14 days post blood meal ( salivary gland sporozoites ) . Midguts or salivary glands were dounce-homogenized in 500 µl of TRIzol reagent . For all of the TRIzol samples , RNA was extracted followed by treatment with DNAseI ( Ambion ) , and passed through an RNeasy Cleanup column ( Qiagen ) . Total RNA ( 1 µg ) was incubated with random hexamers and used to make cDNA using the Superscript III First-Strand Synthesis System ( Invitrogen ) according to manufacturer instructions . Real-time PCR was carried out in a 10 µl volume using SYBR Green PCR master Mix ( Applied Biosystems ) and 1 µM gene specific primers . Real-time PCR was performed using the ABI Prism 7300 qPCR machine ( Applied Biosystems ) . Transcript expression of pyrom1 was normalized to the expression of the control gene , 18s rRNA . The normalized expression for each gene was determined by the ddCt method [85] and values are expressed relative to the expression of the schizont stage . Primers used for real time PCR were designed using Primer Express Software ( Table S3 ) . P . yoelii 17XNL genomic DNA was used to amplify two 600 base pair ( bp ) fragments that are upstream and downstream of the pyrom1 coding region using primers , pyR1 . 5′KpnI-F , pyR1 . 5′XhoI-R , pyR1 . 3′BamHI-F , pyR1 . 3′NotI-R , containing appropriate restriction sites to facilitate cloning of the PCR fragments into the PMD205GFP vector [48] . This vector contains a selection cassette that expresses the P . berghei dihydrofolate reductase-thymidylate synthase ( pbDHFR-TS ) gene fused to the Green Fluorescent Protein ( GFP ) open reading frame under the control of the pbDHFR promoter . The resulting targeting construct contains the pbDHFR-TS/GFP selection cassette flanked by the upstream ( 5′ARM ) and downstream ( 3′ARM ) fragments from the pyrom1 genetic locus . The targeting vector was linearized using restriction enzymes KpnI , NotI , and ScaI prior to transfection . For the transfection , a total of 15 µg of linearized vector DNA was mixed with 5×107 schizonts resuspended in 100 ul of cytomix [86] and transferred to an AMAXA cuvette for electroporation using the AMAXA nucleofector [87] . Pyrimethamine ( 1 mg/kg ) treatment in the drinking water was started 24 hours post transfection and maintained until the appearance of drug resistant parasites . Mixed blood stage parasites were released from host erythrocytes by treatment with 0 . 05% Saponin in ice-cold PBS . Parasite pellet was incubated with 2 mg/ml Proteinase K ( Roche ) at 37 C for 2 hours and genomic DNA ( gDNA ) was extracted using phenol/chloroform , chloroform extraction , followed by ethanol precipitation . Genomic DNA ( 2 µg ) was treated with the restriction enzymes ( New England Biolabs ) overnight for complete digestion . To test for successful gene replacement in R1KO , gDNA was digested with enzymes ScaI/MscI or ScaI alone . To analyze integration of the N-terminal HA tag in R1HA parasite line , gDNA was digested with enzymes BstZ171 and MscI . Restricted DNA was separated by gel electrophoresis on a 0 . 8% agarose gel , transferred via capillary action to a Nylon membrane ( Roche ) . DNA probe specific to the 3′homologous region was amplified using PCR and Digoxigenin ( DIG ) labeled nucleotides ( Roche ) . Probe hybridization and chemiluminescent detection were carried out using manufacturer's instructions ( Roche ) . The spliced open reading frame amplified from P . yoelii 17XNL cDNA was cloned into a vector containing an N-terminal triple HA tag using the EcoRI and NotI restriction sites [33] . The 5′ flanking region ( FR ) was amplified from gDNA ( 1 . 5 kb ) and was cloned into a pBluescript SK ( - ) ( Stratagene ) plasmid using restriction sites ApaI and XbaI to generate pB5′FR . The HA-pyrom1 fragment was subsequently cloned within the XbaI and Not I sites to generate the pB5′FRHAR1 plasmid . The 3′FR ( 1 kb ) was then cloned using NotI and SacII sites into the pB5′HAR1 plasmid to generate pB5′FRHAR13′FR plasmid . The entire insert containing a 5′FR , an N terminally tagged pyrom1 ORF , and a 3′FR , was then cut out with ApaI and SacII and was cloned into the PL0006 vector that contains the hDHFR selectable marker ( Leiden ) to ultimately generate the pR1HA knock-in vector . Prior to P . yoelii transfection the pR1HA vector was linearized with the restriction enzyme Aria in order to promote homologous recombination within the 5′FR . Transfection was carried out as described above . Following pyrimethamine treatment ( 1 mg/kg ) , recombinant parasites were analyzed by PCR and Southern blot analysis of gDNA . Expression of tagged ROM1 was verified by Western Blot . Immunofluorescence microscopy was then carried out to analyze expression and localization of pyROM1 during different stages of the parasite life cycle . Primer sequences are listed in Table S3 . Thin smears of mixed blood stages were fixed in ice cold acetone/methanol ( 1∶1 ) for 3 minutes , blocked with 3% BSA/PBS , incubated with appropriate primary antibody , followed by secondary antibody . Washes were carried out using PBS-Tween 20 ( 0 . 005% ) . Salivary gland sporozoites were centrifuged onto 8-well glass Labtek chambers ( Nunc ) , fixed with 4% Paraformaldehyde , and permeabilized with 0 . 2% TritonX-100 . Fixed sporozoites were then blocked and incubated with antibody as described above . Washes were carried out using 1× PBS . For IFA of liver stages , infected Hepa1–6 cells at specified time points were fixed with ice cold methanol for 15 minutes and incubated with antibodies as described below . To capture invading R1HA sporozoites , Hepa1–6 cells were infected for 10 minutes , fixed with 4% paraformaldehyde , blocked , and stained with α-pyCSP to detect extracellular or invading parasites , permeabilized with ice cold 100% methanol , and incubated with rabbit α-HA to detect pyROM1-HA . Primary antibodies used were 1/1000 rabbit α-HA ( Sigma ) , 1/100 α-pyCSP 2F6 from hybridoma supernatants , 1/500 α-HA rat mAb 3F10 ( Roche ) , 1/1000 rabbit α-pyMAEBL YM2T8 antisera [88]–[89] , 10 ug/ml rabbit α-PfBiP antisera ( MR4 ATCC , MRA-20 ) , 1/500 and rabbit αUIS4 antisera [15] . Secondary antibodies conjugated to Alexa 488 or Alexa 568 ( Molecular Probes ) were used and 0 . 02 µg/ml of 4′6-diamidino-2-phenylindole ( DAPI ) was used for nuclear staining . Images were taken at 100× magnification with the Upright Olympus BX61 microscope using the IP Lab 4 . 0 . 8 software through the Analytical Imaging Facility of the Albert Einstein College of Medicine . An inoculum of 1×104 mixed blood stage parasites from R1KO and wildtype parasites was injected intravenously into five 6-week old female BALB/c mice ( Charles River ) . Parasitemia was monitored by daily blood smears from each infected mouse , stained with Giemsa ( Sigma ) and counted under light microscopy . Percent parasitemia was determined by counting at least 1 , 000 red blood cells per smear and calculated as the percentage of infected red blood cells ( iRBC ) to total number of red blood cells . Swiss Webster mice ( 4 weeks , female ) were injected intravenously with 5×107 infected RBC from either wildtype or R1KO parasites . Two days later , starved Anopheles stephensi mosquitoes were fed on the infected mice harboring mature gametocytes for 15 minutes . A second 5 minute feeding was carried out the following day . Parasite transmission and infectivity was determined by counting the number of oocysts from dissected midguts at day 8 post feeding and the number of salivary gland associated sporozoites at day 14 post feeding under a light microscope . At day 14 post blood meal feeding , mosquito salivary glands were dissected and sporozoites were collected and counted using a hemocytometer . Salivary gland sporozoites ( 3×104 ) were added to 8-well Lab-Tek glass slides ( Nalgene ) that had been pre-coated with 200 µl of 10 ug/ml α-pyCSP ( mAb 2F6 ) antibody in PBS at room temperature , overnight . Loaded Lab-Tek slides were incubated for 1 hr at 37°C to allow sporozoite gliding , after a brief centrifugation ( 500 RPM , 1 minute ) . Media was removed and wells were immediately fixed with 4% Paraformaldehyde at 4°C overnight . Wells were blocked with 1% BSA/PBS solution , incubated with biotinylated mAb 2F6 ( 1 hour , 37°C , 1/100 dilution ) , and stained with streptavidin-Alexa 488 ( Molecular Probes ) . Trails were visualized and counted under a fluorescence microscope . Salivary gland sporozoites ( 3×104 ) were dissected in 1% BSA/DMEM and loaded onto Hepa 1–6 cell monolayers in the presence of 1 mg/ml FITC-dextran ( 10 , 000 MW; Molecular Probes ) . In control wells , sporozoites were pre-treated for 10 minutes with 1 mM of Cytochalasin D on ice before being loaded onto cells . Sporozoites were centrifuged ( 1000 RPM , 3 minutes ) and incubated for 1 hour at 37°C . Cells were washed extensively with 1×PBS to remove excess FITC-dextran that had not been taken up and were fixed with 4% Paraformaldehyde , mounted and visualized under a fluorescent microscope at 40× magnification . The number of FITC-positive cells was counted in at least 30 fields and values represented are FITC-positive cells per field . For invasion assay , semi-confluent Hepa 1–6 cells were loaded with 5×104 P . yoelii WT or R1KO salivary gland sporozoites and centrifuged ( 1000RPM , 3 minutes ) . After 2 hour or 6 hour incubation at 37°C to allow invasion , media was removed and cells were washed with PBS , fixed with 4% paraformaldehyde for 30 minutes , and blocked for 1 hour at 37°C with 1% BSA/PBS . Cells were incubated with the α-pyCSP 2F6 mAb following the Red-Green double staining method to distinguish intracellular versus extracellular parasites [60]–[61] . For liver development assays , Hepa 1–6 cells were loaded with 5×104 P . yoelii Wt or R1KO salivary gland sporozoites , spun down and incubated for 6 , 12 , and 24 hours to allow EEF development . At the end of each time point , cells were fixed with ice-cold methanol and stained with either mAb 2F6 ( α-pyCSP ) or mAb 2E6 ( α-HSP70 ) [90] to visualize and quantify parasite development . At least 50 fields were counted per well and each experiment was done in duplicate or triplicate . For detection of UIS4 positive PVM salivary gland sporozoites were incubated for 2 or 6 hours with Hepa 1–6 cells as describe above . After each time point , cells were fixed with 4% paraformaldehyde for 30 minutes at room temperature followed by permeabilization with 100% ice cold methanol for 10 minutes . Double staining was performed using α-PyCSP 2F6 mouse monoclonal antibody and α-PyUIS4 rabbit polyclonal antibody for 1 hour at 37°C . Staining with secondary antibodies was followed with goat α-mouse IgG–Alexa 488 and goat α-rabbit IgG-Alexa 568 . The percent of parasites double staining ( CSP-UIS4 ) over total number of parasites ( CSP only ) was recorded in duplicate . Statistical analysis was carried out using either One-way ANOVA multiple comparison test or unpaired t-test with the GraphPad Prism software . Analysis of malaria infection in mouse livers was carried out as described previously [50] . Salivary gland sporozoites from wildtype or R1KO infected mosquitoes were collected and counted using a hemocytometer . Female BALB/c mice 6 weeks of age ( purchased from Charles River ) were intravenously injected with 1000 or 10 , 000 sporozoites . Four or five mice were independently injected per parasite line . Livers were harvested 36 hours later and homogenized in 10 ml of TRIzol reagent ( Invitrogen ) . RNA was extracted from 1 . 5 ml of liver homogenate , treated with DNAseI ( Ambion ) and purified with an RNeasy purification column ( Qiagen ) . Total RNA ( 4 µg ) was used to make cDNA using the Superscript III cDNA synthesis system ( Invitrogen ) and random hexamers as primers . Quantitative PCR was carried out with the ABI 7300 apparatus using the POWER SYBR green master mix ( Applied Biosystems ) in a 20 µl reaction volume containing 2 µl of cDNA and 1 µM of primers . Test Primers to detect parasite burden within liver were to the Plasmodium 18s rRNA gene and the internal control primers were specific to the mouse GAPDH gene . Relative transcript quantification was determined using the 2−ddCt method . HA-tagged rhomboid expressing plasmids ( DmRho-1 , TgROM5 , TgROM5SA ) were obtained from Dr . Sinisa Urban as published in [33] . A truncated PyUIS4 that excludes the predicted signal sequence and most of the C-terminal domain was amplified from sporozoite cDNA . The reverse primer was designed to include a 2xMyc tag . This PCR product was fused to GFP preceded by an IgK leader sequence via PCR joining using primers listed in Table S3 . The resulting PCR was inserted into the pcDNA3 . 3 TOPO TA cloning vector . The template for GFP is from the Spitz-GFP expression plasmid used in [33] . The cleavage assay was carried out as described in [33] , [39] with minor modifications . Briefly , COS7 were seeded on 6-well plates and transfected with plasmid DNA for transient expression using Fugene6 ( Roche ) reagent following manufacturer's protocol . For PyUIS4TMGFP , 250 ng of plasmid DNA was used per well and 100 ng of plasmid DNA was used for rhomboid constructs . A pBluescript ( Stratagene ) plasmid was used as filler DNA . At 18 hours post-transfection , media was removed , cells were washed with serum-free media ( SFM ) , and 800 µl of SFM containing protease inhibitor Galardin ( GM_600 , Biomol ) was added . The conditioned media was collected and cells were harvested by lysis with sample buffer at 18–24 hours later . Media fraction was concentrated using Centricon ( Millipore ) centrifugal concentrator with a cut off of 3 kDa . Western blot analysis was carried out for both media fractions and cell lysates . Anti-GFP antibody ( Santa Cruz Biologicals ) was used to detect PyUIS4TMGFP and anti-HA mAb clone 3F10 ( Roche ) was used to detect rhomboid constructs . Salivary gland sporozoites ( 1×106 ) were dissected and loaded onto a confluent monolayer of Hepa1–6 cells at an MOI of 2∶1 ( 5×105 ) , centrifuged ( 1000 RPM , 3 minutes ) , and allowed to invade for two hours . Infected cells were washed with media containing 10× Pen/Strep and 0 . 25 µg/ml of Fungizone ( Invitrogen ) to remove debris and unbound sporozoites . Development was allowed to continue another two hours for a total of four hours of development . Cells were then washed with PBS , trypsinized , and washed one more time with PBS , followed by fixation with 2% Paraformaldehyde/2 . 5% Glutaraldehyde for 1 hour at room temperature . Cells were washed with PBS and postfixed in 1% osmium tetroxide ( Polysciences Inc . , Warrington , PA ) for 1 hour . Samples were then rinsed extensively in dH20 prior to en bloc staining with 1% aqueous uranyl acetate ( Ted Pella Inc . , Redding , CA ) for 1 hour . Following several rinses in dH20 , samples were dehydrated in a graded series of ethanol and embedded in Eponate 12 resin ( Ted Pella Inc . ) . Sections of 95 nm were cut with a Leica Ultracut UCT ultramicrotome ( Leica Microsystems Inc . , Bannockburn , IL ) , stained with uranyl acetate and lead citrate , and viewed on a JEOL 1200 EX transmission electron microscope ( JEOL USA Inc . , Peabody , MA ) . Image J software was used for quantification of the PVM/PV area in the EM images and statistical analysis was carried out with the GraphPad Prism software using unpaired t test . Measurements were made by on blinded images . | Plasmodium parasites are obligate intracellular organisms that invade cells by an active mechanism mediated by the secretion of contents from specialized secretory organelles , the micronemes and rhoptries . Invaded parasites reside and replicate within a membrane-bound compartment called the parasitophorous vacuole ( PV ) . PV formation is exclusive to development within mammalian specific host cells , the erythrocytes and hepatocytes . Proper modification of the PV is important to protect the parasite from host defenses and to serve as a gateway for nutrient acquisition and communication with the environment . The rhomboid proteins , a class of intramembrane serine proteases , are implicated in the invasion process . We studied the microneme rhomboid protease , ROM1 in the rodent malaria parasite , Plasmodium yoelii . We find that pyROM1 is not important for efficient invasion into host cells and instead is important for survival within the host cells . Analysis of parasites developing within hepatocytes reveals a defect in PV development . We propose that pyROM1 provides a fitness advantage to parasites developing within host cells by promoting the proper modification of the PV . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"microbiology",
"host-pathogen",
"interaction",
"parasitic",
"diseases",
"parasitology",
"gene",
"function",
"developmental",
"biology",
"emerging",
"infectious",
"diseases",
"cell",
"growth",
"global",
"health",
"microbial",
"growth",
"and",
"development",
"i... | 2011 | Plasmodium Protease ROM1 Is Important for Proper Formation of the Parasitophorous Vacuole |
Decisions about noisy stimuli require evidence integration over time . Traditionally , evidence integration and decision making are described as a one-stage process: a decision is made when evidence for the presence of a stimulus crosses a threshold . Here , we show that one-stage models cannot explain psychophysical experiments on feature fusion , where two visual stimuli are presented in rapid succession . Paradoxically , the second stimulus biases decisions more strongly than the first one , contrary to predictions of one-stage models and intuition . We present a two-stage model where sensory information is integrated and buffered before it is fed into a drift diffusion process . The model is tested in a series of psychophysical experiments and explains both accuracy and reaction time distributions .
Decision making is of crucial interest in many disciplines such as psychology [1] , [2] , neuroscience [3]–[5] , economics [6] , [7] , and machine learning [8] . Binary decision theories relate to situations where an observer ( or machine ) is confronted with one of two possible noisy stimuli ‘A’ and ‘B’ . A decision has to be made whether ‘A’ or ‘B’ is present . For example , human readers have to decide whether a handwritten character is an or a ; a trader has to decide whether to sell or to keep; a monkey has to decide whether dots on a screen are moving to the left or to the right [9] . While engineering and economical decision theories focus on how to compute optimal decisions [6] , [7] , [10] , psychology and neuroscience investigate the actual decision making process in humans and animals [9] , [11]–[14] . Decision making is usually assumed to be a one-stage process where evidence integration and decision making are identical ( but see [15] , [16] ) . In a standard accumulator model each bit of evidence is integrated and a decision is reached once the accumulated evidence for one of the two response alternatives crosses a threshold [13] , [14] , [17]–[33] . If the evidence itself is noisy , then the accumulation of evidence for each of the two stimulus alternatives leads to a diffusion-like process . For example , in the well-known random motion paradigm [9] , moving dots appear at random moments in time , so that evidence for leftward or rightward moments arrives probabilistically and the accumulator is expected to evolve along a stochastic path that can be approximated by a drift-diffusion process . This is in good accordance with experimental studies where neurons in the macaque lateral intraparietal cortex ( LIP ) increase firing rates along a noisy trajectory up to the moment of decision [9] , [32]–[34] . Since evidence is very noisy in this case , and arrives slowly over time , the decision process is rather slow [9] . Most experimental [9] , [11] , [12] and theoretical work on decision making [5]–[8] focuses on paradigms where noisy stimuli are presented for long durations , e . g . until a response is elicited ( for exceptions see [31] , [32] ) . In other paradigms , where stimuli are less noisy , decisions can be extremely fast . For example , humans only need a fraction of a second to recognize objects such as animals in a picture [35] . This astonishing speed is also evident in sports such as table tennis or soccer requiring rapid reactions to moving balls . In these examples , the brain has to decide rapidly upon visual information available for only a hundred milliseconds or less . Note that even in these scenarios where stimuli are of high contrast ( “low noise” ) , the responses of the observers can still be “noisy” . Here , we first show psychophysically that one-stage models of the noisy accumulator or drift-diffusion type cannot explain the results of feature fusion experiments where two stimulus alternatives are presented in rapid succession for durations in the range of 20–160 ms . Second , we propose , instead , a two-stage model , where evidence integration is separated from a noisy drift-diffusion decision making process . Our results reveal additional aspects of the dynamics of decision making that are hidden in standard experimental paradigms where only one stimulus alternative is presented per trial .
In experiment one , vernier stimulus ‘A’ , offset either to the left or right , was immediately followed by a second vernier stimulus ‘B’ with opposite offset direction ( right or left , respectively ) . The durations and of both verniers were equal , i . e . , but varied from 10 to 80 ms , each . Vernier stimulus ‘B’ dominates the percept the stronger the longer both vernier stimuli ‘A’ and ‘B’ are presented ( Figure 2D ) . For example , when the two vernier stimuli are presented for 20 ms each , observers report a percept corresponding to stimulus ‘B’ in 60% of the trials , while ‘A’ is reported in only 40% of the trials . When the two stimuli are presented for 40 ms each , observers report a percept corresponding to stimulus ‘B’ in 67% of the trials , while ‘A’ reported in only 33% of the trials . We wondered whether the dominance of the second stimulus could be explained by classical noisy accumulator models , also called Drift-Diffusion models . In the standard , one-stage Drift-Diffusion Model [20] , [22] , [23] , [27] , evidence for ‘A’ or ‘B’ translates directly into the drift rate ( upward for ‘A’ , downward for ‘B’ ) of a decision variable ( Figures 2B , C ) . As usually , we added noise to the drift process leading to a random walk of the trajectory . The noise accounts for both noisiness of the evidence itself ( an important aspect in the moving-dot paradigm [9] , [33] , [34] , [40] ) and internal noise in the brain . After presentation of both stimuli , the drift goes back to zero . A decision is made when hits the upper ( for ‘A’ ) or lower bound ( for ‘B’ ) . In this one-stage model , dominance of stimulus ‘A’ is the stronger the longer the presentation times of ‘A’ and ‘B’ , and respectively . This is in striking contrast to the experimental results . We found that the qualitative nature of the results is independent of the specific choice of parameters of the one-stage drift diffusion model: for all tested parameters , the dominance of the second stimulus decreased with increasing duration ( whereas the dominance of the second stimulus increased in the experiments ) . Whereas , for certain , fixed stimulus durations , we could achieve dominance of the second stimulus with specifically optimized parameters , we could never achieve dominance of the second stimulus for the entire range of stimulus durations with one set of parameters . We explored whether minor modifications of the one-stage drift-diffusion model can explain the dominance of the second vernier . For example , we replaced the noisy accumulator by a noisy leaky accumulator . However , this did not change the results qualitatively . We then tested a very basic two stage model . During stimulus presentation , the stimulus served as the drift in a noisy leaky integrator model . After stimulus termination , the leak was artificially set to zero and the integration continued as a free , unbiased noisy diffusion process . In other words , the result of the leaky evidence integration served as initial condition for the leak-free diffusion process . While qualitatively such a drift-diffusion model explains the dominance results well ( Supporting Figure S3B ) , we suggest an alternative model , which accounts very well for both the dominance and the reaction time distributions . In this two-stage model , the evidence integration enters the second stage as a drift rate rather than as a bias in the initial condition . ( a ) During stage one , evidence integration is leaky and dominated by the intrinsic noise of the stimulus . The variable of noisy evidence integration is . ( b ) Stage two starts after a fixed time after stimulus onset and ends when a second variable hits the upper or lower decision threshold . ( c ) The variable of the leaky integrator of stage one sets the drift in the ( leak-free ) drift-diffusion model of stage two . The combination of ( b ) and ( c ) implies that , for long stimuli , stage two is a drift-diffusion model with time-dependent drift set by the momentary value of the integration variable of stage one . In case that the total duration of the stimulus is shorter than the time needed to reach the decision threshold in stage two , the value of the leaky integrator of stage one at the end of the stimulus is written into a buffer and this buffered value serves , during the remaining time , as the ( constant ) drift for the diffusion process in stage two until a decision is reached . In the limit that stimuli are shorter than , stage two has therefore a constant drift . In the limit that stimuli are presented for times much longer than ( so that is negligibly short compared to the stimulation time ) , our two-stage model becomes equivalent to a standard one-stage drift-diffusion model with a time-dependent drift that is given by the low-pass filtered version of the input signal . However , for very short stimuli , the prediction of our two-stage model is remarkably different from that of a standard one-stage model – and these ultra-short stimuli are at the center of our study . The results on stimulus dominance during the feature fusion paradigm with two short Verniers can indeed be explained by the two-stage model ( Figures 2E , F ) . Since our stimuli are comparatively strong ( over 90 percent accuracy for stimli presented separately ) , we consider the limit where the evidence integration in stage one is noise-free . Hence , in the first integration stage , evidence for stimulus ‘A’ and ‘B’ is simply accumulated in a noiseless forgetful ( leaky ) integrator ( see also [30] ) . The time scale of forgetting is related to the time over which an ideal observer expects stimuli to remain constant ( see Materials and Methods ) . The second phase , the decision stage starts at a fixed time and consists of a standard drift-diffusion model without leak ( Figure 2F , bottom panel ) . For a sequence of two short stimuli , the stimulation ends before so that at the termination of the second stimulus ( ) , the output of the evidence integration is written into a buffer and fed later from the buffer as a constant drift rate into stage two . The two-stage model captures the dominance of the second vernier very well ( Figure 2G ) . The critical test for models of decision making is to account for reaction time distributions rather than accuracy [20] . We therefore wondered whether the two-stage model captures the reaction time distributions in the fusion experiments . In experiment two , stimulus ‘A’ ( the first vernier stimulus ) was presented for a duration , immediately followed by stimulus ‘B’ ( a vernier with opposite offset ) of duration with ( Figure 3A ) . Parameters of the two-stage model were adapted individually for each observer and kept fixed across all stimulus conditions . The dominance of the first vernier stimulus increased when increased ( Figure 3B ) . Reaction times for strongly biased situations ( e . g . where the first vernier stimulus is much longer than the second one or vice versa ) are faster ( 75% of decisions made before 560 ms ) than those in conditions with dominance around 50% ( 75% of decisions made before 610 ms ) leading to an inverted-U-shaped curve of the reaction time quantiles ( Figure 3C ) . The same pattern is observed when responses for the first and second vernier stimulus are analyzed separately ( Figure 3D ) . Median response times varied strongly across the 13 observers ( Figure 4A ) . We separated the observers into a group of fast responders ( median reaction time <500 ms ) and one of slow responders ( median reaction time >500 ms ) . While the reaction times of both groups show an inverted U-shape function , the qualitative picture is different between slow and fast responders . If the first vernier stimulus is presented for a short time only , fast responders are particularly fast whereas slow responders are particularly slow . The two-stage model qualitatively reproduces this behavior ( Figures 4B , C , Supporting Figure S1 ) . For each stimulus condition , the outcome of the leaky integration in the first stage serves as a drift of the leakfree drift-diffusion model during the second stage of the two-stage model . For short stimuli , like the ones considered so far , where the stimulus ends before the integration , the result of stage one is written into a buffer and used as a constant value of the drift in the decision stage . In other words , the evidence at stimulus termination serves as drift value , rather than as an initial condition of stage two . As an alternative , we have also analyzed a drift-diffusion model where the drift was taken as a free parameter , optimized for each stimulus condition independently so as to optimally predict the distribution of reaction times . The drift predicted from this model ( which has more degrees of freedom ) is statistically not different ( as determined by a two way repeated measures analysis of variance ) from our two-stage model where the drift is not a free parameter but the result of stage one . This finding suggests that the simple preprocessing by leaky integration correctly determines the drift rate ( Figure 4D ) . Further above we had reported that a qualitative fit of the dominance was possible by a noisy leaky integrator , if the leak was set to zero at the end of the stimulus . In such a model , the result of the leaky evidence integration serves as the initial value for a free diffusion process . The results of Figure 4D , however , indicate that the result of the leaky integration in the first stage should be used as the drift , and not as an initital condition for the diffusion in stage two . The results of Figure 4D can therefore be considered as a strong argument in favor of the two-stage model . In the following , we consider other aspects of the two-stage model . If the writing into the buffer is triggered at stimulus termination , as assumed in the two-stage model , the question arises why the switch from ‘B’ to the background , but not that from stimulus ‘A’ to ‘B’ , triggers the transition from stage one to stage two in the two-stage model . We suggest that the large change from a vernier stimulus to background is “interpreted” as stimulus termination because there is a strong neural off-transient for a change from ‘A’ to a blank screen , whereas there are no on- and off-transients for a change from ‘A’ to ‘B’ , respectively [41] . This is well in accordance with a Bayesian approach ( see Supporting Figure S2 ) suggesting that feature integration should terminate when it becomes unlikely that the momentary stimulus is a continuation of the previous stimulus . The readout in the two-stage model should therefore start when a novelty value of the momentary stimulus crosses a predetermined threshold ( cf . Supporting Text S1 ) . We tested this prediction by the psychophysical experiment in Figure 1 , where the first vernier stimulus was followed by a blank background ( interstimulus interval; ISI ) before the second vernier was presented . With an ISI of 20 ms , the two vernier stimuli , presented for 10 ms each , became individually discriminable . Observers could tell whether the first stimulus was offset to the left or to the right by motion cues [41] , [42] . However , for a sequence of ‘A’ immediately followed by ‘B’ with , verniers are not individually visible even though the total duration is 40 ms as in the sequence with the 20 ms ISI . This suggests that in the condition with the 20 ms ISI the termination signal of the first vernier stimulus stopped evidence integration and wrote the result into a buffer , for later use in stage two , whereas evidence was integrated across the two vernier stimuli in the experiment without the blank , before the final result was written into a buffer . In our experiments with ultra-short stimuli , the time where the read-out from the buffer starts , occurs after stimulus termination ( and is included in the non-decisional time . We also tested a model where the decision process was triggered at stimulus termination , i . e . , at the same moment when the result of evidence integration is written into the buffer ( i . e . ) . Such a model predicts that reaction times increase with total stimulus duration ( data not shown ) , which disagrees with our observation that , for a given level of dominance , the mean reaction times remain largely constant for total stimulus durations of 20 ms , 40 ms , and 80 ms ( Supporting Figure S3 D ) . For most of the stimuli considered so far , the total stimulus duration was below 40 ms . In this case , the two stages of the model are sequential and do not overlap . However , for longer stimuli , evidence integration of stage one is not finished at the moment of when the diffusive decision process in stage two is started . Indeed , a model with fixed drift in stage two works well for stimuli up to a total duration of 80 ms , but breaks down at 160 ms ( data not shown ) . However , our two stage model assumes that as soon as stimuli extend beyond , the momentary value of the evidence integration stage is written into the buffer and immediately used as drift in the diffusion process of stage two . The drift is updated continuously so that the diffusion process becomes time-varying . The fact that a constant drift in stage two fails when the stimulus extends over 160 ms indicates that the parameter of our model is much shorter than 160 ms . We tested this by fitting for individual subjects such that the mean square error in the dominance was minimized across all stimulus durations , including the 160 ms conditions . The optimal values for were indeed smaller than 160 ms ( , , ) . In the model , we explored the situation that the first stimulus becomes much longer than . Obviously , if the first stimulus is made very long , our two-stage model then predicts that the first stimulus dominates .
In our model , we assumed several components which are worth discussion each . First , evidence integration in stage one must be leaky . It is the leak that explains why , when the first and second vernier stimulus are of the same duration ( ) , the second vernier stimulus dominates ( experiment one ) . The leak in our model arises naturally from a Bayesian approach and can be traced back to the fact that stimuli are expected to change in natural environments . Similar to our Bayesian novelty detection approach ( cf . Supporting Text S1 ) , the leaky evidence integration can also be derived in the framework of Kalman filters [47]–[50] . Second , the accumulated evidence must transferred at an appropriate moment and written into a temporal buffer . Such a buffer is necessary since decisions often occur a considerable time after the stimulus has disappeared . We suggest that the precise moment of transfer is set by a novelty score monitored during evidence integration ( see Supporting Text S1 ) . Such a novelty signal and subsequent buffering explains why the two vernier stimuli are perceived individually , if the stimuli are separated by a blank screen ( ISI ) , but fused into a single percept in the absence of the blank . In this sense , feature fusion can be interpreted as a failure to detect the onset of a new stimulus because the new evidence is not sufficiently different to raise a ‘novelty signal’ . In contrast , the switch from stimulus to background creates a sufficiently strong transient to stop the feature integration process ( Supporting Figure S2 ) . Third , the noisy decision process is triggered at a fixed time after stimulus onset . If the decision process were triggered at , reaction times would increase with stimulus duration . This is , however , not the case ( Supporting Figure S3D ) . From the fact that our model assumes a fixed start time of the second stage , it necessarily follows that we have to distinguish two different situations: If the total stimulus duration is shorter than , we need to bridge the time between the end of the stimulus and decision by storing the intermediate result of evidence integration into a buffer . This value is then used in stage two as a fixed mean drift rate . If the total stimulus duration is longer than , the result of stage one is used online as a time-dependent drift for all times until the end of the stimulus ( at which point it is again ‘frozen’ and transferred into the drift-buffer . ) . Our two stage model is similar to previous two stage models in which sensory processing , e . g . motion processing or contrast detection , precedes a decision making stage ( e . g . [15] , [16] , [46] ) . In our model , the sensory integration stage is leaky to account for the dominance of the second vernier . Our two-stage model comprises a leaky integration stage followed by a drift-diffusion stage . The question arises whether or not a one-stage model with leak in the drift-diffusion process can explain the results . However , this is not the case because in such a model always the first stimulus dominates because the leak pushes the decision variable towards the starting point and not across it ( Supporting Figure S3 ) . Another way to integrate the leak into a one-stage model is to directly transform the input by a leaky integrator ( like our stage one ) and to use the outcome of the leaky integrator as a time-variant drift in stage-two ( , ) . However , using stage one only for pre-processing will not change the pattern of results [28] . In such models , the decision variable also moves towards the decision bound for stimulus ‘A’ before dropping back to chance level . Therefore , these models also show a dominance of the first stimulus . The novel features of the two-stage models are observable well only for stimuli in the range of up to about 100 ms . This duration is in line with the duration of visual integration found in other studies [51]–[53] . One of the paradoxical aspects of our model is that the second stage starts at a fixed time . Obviously , if the duration of a stimulus extends beyond , then the stage of evidence integration and that of stochastic decision making ( stage two ) will overlap and the separation into two distinct phases disappears ( see Supporting Text S1 ) . Therefore it is not surprising that for longer stimulus durations standard one-stage models work well [27] , [29] , [54] . In our model , a deterministic filter ( leaky integrator ) is applied in stage one to a step-like input , representing a noiseless stimulus . This is the limiting case where the stimulus is considered to be of high contrast . In a more realistic scenario the stimulus itself is noisy . The stochasticity of stimuli leads , after stage one , to a noisy result of evidence integration , which is written into the buffer and then used as drift for stage two . This noisy result is modeled by the variance of the drift constant of stage two . It is therefore tempting to relate the stochasticity of drift constants to sensory or physical noise . The stochasticity of stage two may be related to internal noise in the brain [32] , [55] . What is the advantage of adding a separate noisy decision process ? It is well known that human observers can manipulate the speed-accuracy trade-off according to instruction or reward scheme by a change in strategy corresponding to a shift of the initial condition , , or the decision thresholds in the drift-diffusion process [13] , [54] . Neurons in the superior colliculus [56] , the LIP [9] , [32] , [33] , the pre-motor cortex [57] , [58] , and the dorsoventral lateral prefrontal cortex [11] , [12] , [59] were shown to be involved in decision making . The firing rate of these neurons increases as long as stimuli are displayed . This ramping activity may relate either to evidence accumulation ( “stage one” ) or to decision making ( “stage two” ) . Future experiments with feature fusion stimuli may be used to decide between these two alternatives . In summary , it is often ( intuitively ) assumed that visual input directly translates into decisions . A stimulus presented first should drive decisions stronger and faster than a later stimulus ( first in , first out ) . This is obviously correct when the two stimuli are long , because a decision may be reached even before the second stimulus can influence decision masking . In this case , we can assume that evidence integration and decision making are the same . However , for short stimuli this is not the case . Evidence integration and decision making can only be disentangled , when the two stimulus alternatives are presented within one trial ( feature fusion ) but not when only one stimulus is presented per trial , as it is usually . The distinction between evidence integration and decision making is described well by our two-stage model , where rapid stimuli are integrated and buffered before the decision process starts .
All participants signed informed written consent . The study was approved by the Commission cantonale ( VD ) d'éthique de la recherche sur l'être humain ( Lausanne , Switzerland ) and conducted according to the principles expressed in the Declaration of Helsinki . A total of 24 observers ( 8 female , aged 21–32 years ) signed informed written consent . Participants had normal or corrected-to-normal visual acuity as measured by the Freiburg visual acuity test [60] . All but two observers ( the first and second author ) were naive to the purpose of the study . Naive observers were paid students from local universities . Stimuli were presented on a Tektronix 608 X-Y display or a HP 1332A X-Y display . Both X-Y displays were equipped with a P11 phosphor and controlled by a PC via a fast 16 bit DA converter . Stimuli were presented at , a 1 MHz dot rate , a 500 Hz refresh rate , and a dot pitch of . Viewing distance was 2 m . The room was dimly illuminated by a background light ( ) to prevent adaptation to scotopic vision . Stimulus contrast was close to 1 . 0 . In each experiment , the conditions have been presented randomly interleaved to reduce the influence of hysteresis , learning , or fatigue in the averaged data . The vernier stimuli were composed of two vertical segments . Each segment was 10′ ( arc min ) long , 0 . 5′ wide , separated by a vertical gap of 1′ . A small horizontal offset was inserted between the upper and the lower segments ( Figure 2A ) . Horizontal offset sizes ranged from 30″ to 40″ ( arc sec ) . Offsets were chosen individually to be at least twice the offset size of the offset discrimination threshold for a single vernier stimulus of 20 ms duration as determined using the adaptive PEST procedure [61] . A sequence of two vernier stimuli with opposite offset directions was presented foveally in rapid succession . The offset direction of the first vernier ( stimulus ‘A’ ) was chosen randomly in each trial ( left or right ) . The second vernier ( stimulus ‘B’ ) had an offset direction opposite to that of the first vernier . If , for example , the first vernier stimulus was offset to the left , the second vernier was offset to the right , and vice versa . Observers perceived only one fused vernier and were asked to report the position of the lower segment with respect to that of the upper segment by pressing one of two push buttons . Observers were instructed to respond as rapidly as possible , but also as accurately as possible . No feedback about performance was given . Naive observers did not know that a sequence of two vernier stimuli was presented . We computed dominance , defined as the proportion of trials on which the response matched the offset direction of the first vernier stimulus . Thus , values above 50% indicate dominance of the first vernier ( stimulus ‘A’ ) ; values below 50% indicate dominance of the second vernier ( stimulus ‘B’ ) . 50% vernier dominance is the point of subjective equality , i . e . first and second vernier stimulus equally contribute to performance . First vernier stimulus ( ‘A’ ) and second vernier ( ‘B’ ) were presented in immediate succession ( Figure 2 ) . Both vernier stimuli had either the same duration or the duration of one of the verniers was four times longer than the other . The total duration of the first and second vernier was 20 ms , 40 ms , 80 ms , or 160 ms . All conditions were presented in a random order . Every condition has been repeated 400 times per observer . As Experiment 1 , except for that the duration of the first vernier was varied in 12 steps between 0 ms and 40 ms . The total duration + always summed up to a total of 40 ms ( Figure 3A ) . Every condition has been repeated 400 times per observer . In Figure 1 , an ISI was inserted between the first and second vernier stimulus . Observers were informed about the experimental design and asked to indicate whether the first or second vernier stimulus was offset to the right . Reaction times below 300 ms or above 1200 ms were excluded from analysis to reduce the impact of motor errors and unattended trials ( less than 3% of the trials ) . We model the stimuli by a time-varying input signal , which is +1 during the presentation of stimulus ‘A’ , −1 for stimulus ‘B’ and 0 otherwise . In the evidence accumulation stage of the two-stage model , the stimulus is subjected to leaky integration: Since our stimuli have high contrast , the evidence integration is modeled as a noise-free process . For times larger than the integrated evidence is fed as the drift into the noisy drift-diffusion model at stage two . We distinguish two different cases . a ) Stimuli are shorter than . At the termination of stimulus ‘B’ ( ) the integrated evidence is stored and written into a buffer . Later , for the buffered value is used as the mean drift rate with a fixed scaling factor for the decision stage , which encompasses a standard drift-diffusion model . b ) Stimuli are longer than . In this case the momentary evidence is used as the mean drift for . Again , at the end of the stimulus , the last value of the evidence is buffered and used as drift henceforth . During stage two , in every trial , a decision variable is initialized at and evolves according to the Langevin equation where is the drift rate and is a Wiener process , which introduces noise to the decision process . A decision is made when the decision variable reaches one of two decision boundaries ( decision ‘A’ ) or ( decision ‘B’ ) . The associated reaction time is the sum of a non-decisional time ( which accounts for sensory delay , and the evidence integration and buffering times as well as motor delays ) and the time when the decision variable reaches the boundary . We used the Ratcliff extension [20] of a standard drift diffusion model , in which the non-decisional time , the initial condition and the drift rate vary stochastically from trial to trial . The non-decisional time is drawn from uniform distributions with mean and width . The initial condition is drawn from uniform distributions with mean and width . The drift rate is drawn from a Gaussian distribution with mean – the output of the first stage – and standard deviation . and represent noise in the evidence accumulation . As a reference , we used a one-stage model , which encompasses a standard drift diffusion model , in which the drift rate depends on time and is given by the input signal: . In this model , the drift becomes zero after the end of the stimulus . We also simulated leaky variants of this one-stage model , for details see Supporting Text S1 . In the first step , the parameters , , , , , and of the decision stage were fitted to the experimentally obtained cumulative reaction time distributions by minimizing the product of the p-values of the Kolmogorov-Smirnov statistic for each stimulus condition [62] , [63] . Responses to stimuli ‘A’ and ‘B’ and different stimulus conditions were fitted simultaneously using the fast-dm software of Voss & Voss [64] . For both experiments , fits were done individually for each observer . In the experiment of Figures 2 , all parameters except the mean drift rate and the drift variability were the same in all stimulus conditions . The drift was calculated from stage one . Drift variability was a function of stimulus duration . In the experiment of Figures 3 and 4 , only the mean drift rate was varied across conditions and calculated from stage one . In order to obtain the parameters and of the evidence integration in stage one we ran a simulation experiment with free drift rates as in Figure 4D . The obtained mean drift rates were then used to fit the time constant and the scaling factor , again separately for each observer . This fit was done using the fit-routine of MATLAB . Finally , to extract the optimal values for , we first used the data of experiment 1 with stimulus durations and fitted the parameters of both stages with the described procedure . Then , we performed a line scan of all values of and identified the value that minimized the mean square error of the measured dominance , now including the long duration of 160 ms . Parameters are different for each observer , i . e . , , , , , , , and for stage one and . | In models of decision making , evidence is accumulated until it crosses a threshold . The amount of evidence is directly related to the strength of the sensory input for the decision alternatives . Such one-stage models predict that if two stimulus alternatives are presented in succession , the stimulus alternative presented first dominates the decision , as the accumulated evidence will reach the threshold for this alternative first . Here , we show that for short stimulus durations decision making is not dominated by the first , but by the second stimulus . This result cannot be explained by classical one-stage decision models . We present a two-stage model where sensory input is first integrated before its outcome is fed into a classical decision process . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cognitive",
"neuroscience",
"psychology",
"social",
"and",
"behavioral",
"sciences",
"biology",
"sensory",
"perception",
"neuroscience"
] | 2012 | Paradoxical Evidence Integration in Rapid Decision Processes |
Synchronizing cell growth , division and DNA replication is an essential property of all living cells . Accurate coordination of these cellular events is especially crucial for bacteria , which can grow rapidly and undergo multifork replication . Here we show that the metabolic protein ManA , which is a component of mannose phosphotransferase system , participates in cell wall construction of the rod shaped bacterium Bacillus subtilis . When growing rapidly , cells lacking ManA exhibit aberrant cell wall architecture , polyploidy and abnormal chromosome morphologies . We demonstrate that these cellular defects are derived from the role played by ManA in cell wall formation . Furthermore , we show that ManA is required for maintaining the proper carbohydrate composition of the cell wall , particularly of teichoic acid constituents . This perturbed cell wall synthesis causes asynchrony between cell wall elongation , division and nucleoid segregation .
The bacterial cell wall is a key determinant of cellular morphology that provides structural support and mechanical protection . The structural dynamics of the bacterial cell wall enable elasticity , growth and division . The cell wall of the Gram positive bacterium Bacillus subtilis ( B . subtilis ) is primarily composed of peptidoglycan ( PG ) , a net-like polymer of glycan strands cross-linked by peptide bridges , and anionic phosphate-rich polymers . Both PG and anionic polymers play critical roles in maintaining the structural integrity and viability of the bacterial cell [1] . PG strands comprise alternating units of N-acetylglucosamine ( GlcNAc ) and N-acetylmuramic acid ( MurNAc ) . PG synthesis initiates in the cytoplasm with fructose-6-phosphate and proceeds through a linear pathway to generate the precursors UDP-GlcNAc and UDP-MurNAc-pentapeptide [2] . The following steps are membrane bound and carried out using a special recycled lipid carrier , undecaprenyl phosphate . Initially , the MraY transferase attaches MurNAc-pentapeptide to the lipid carrier thereby yielding lipid I . Consequently , GlcNAc is added to lipid I by the MurG enzyme producing lipid II . Lipid II carries the basic PG monomer composed of GlcNAc and MurNAc-pentapeptide . Next , lipid II is flipped across the membrane , the PG monomer is cleaved of the undecaprenyl phosphate and incorporated into the growing chain [2]–[4] . Labeling of the newly synthesized PG revealed that it is inserted in a helical pattern along the lateral cell wall [5]–[8] . Accordingly , atomic force microscopy of the B . subtilis cell wall exposed helical PG cabling arrangement with glycan strands up to 5 µm in length , longer than the bacterium cell itself [9] . A protein implicated to play a key role in inserting new PG is the actin homolog MreB . MreB forms a dynamic helical scaffold that serves as a platform onto which the cell wall machinery localizes [5] , [10]–[16] . B . subtilis has three MreB isoforms , called MreB , Mbl and MreBH , which have been demonstrated to colocalize in a single helical structure [17] . Mutations within the genes encoding these isoforms , as well as in other essential PG components , induce severe morphological defects ( e . g . : [6] , [13] , [15] , [18]–[21] ) . In B . subtilis the anionic polymers can be either bound to PG , wall teichoic acid ( WTA ) , or anchored to the cytoplasmic membrane , lipoteichoic acid ( LTA ) . The major form of WTA comprises glycerol phosphate polymer [1] , [22]–[24] and the minor form is a polymer of glucose ( Glc ) and N-acetylgalactosamine ( GalNAc ) [1] , [22] , [25] , [26] . Fluorescence analysis has revealed that WTA enzymes are localized at division sites and along the lateral sides of the bacterial cells [27] . Similar to the PG pathway , WTA biosynthesis begins with formation of nucleotide sugars in the cytoplasm and proceeds with a membrane step that utilizes the same lipid carrier undecaprenyl phosphate . Notably , mutations in the WTA and/or LTA pathways lead to loss of rod shape and non-uniform thickening of the PG layer [28]–[31] , suggesting coordinated biogenesis of the cell wall components . Here we show that the sugar metabolic enzyme ManA ( mannose phosphate isomerase ) , which is part of the mannose phosphotransferase system , is unexpectedly necessary in rich medium , when mannose is not utilized as a carbon source . In the absence of ManA , cells display abnormal morphologies and fail to properly package and segregate their chromosomes . Furthermore , we demonstrate that these abnormal phenotypes are due to a role played by ManA in cell wall construction . We show that the lack of ManA perturbs proper cell wall carbohydrate composition and thereby causes asynchrony between cell growth , division and nucleoid segregation .
To identify new components required for cell division and chromosome segregation , we performed a transposon mutagenesis and screened for B . subtilis mutants exhibiting growth defects ( Materials and Methods ) . The selected mutants were then subjected to a visual microscopy assay . One of the slow growing mutants had a striking phenotype with the cells exhibiting a severe shape defect and atypical nucleoid morphologies ( Figure 1 ) . Evidently , mutated cells lost the characteristic rod shape typical of wild type B . subtilis cells and instead appeared as elongated spheres , which were significantly larger than normal . This spheroid like morphology resembles the phenotype described for mutants defective in cell wall synthesis ( e . g . : [6] , [13] , [15] , [18] , [20] , [21] , [29] , [30] ) . In addition , we observed internal membrane invaginations in many of the mutant cells indicative of inappropriate cell divisions ( Figure 1F ) . Moreover , DAPI staining revealed a variety of abnormal nucleoid structures , which have lost their spatial organization in comparison to the well-organized wild type chromosomes ( Figure 1 ) . Cloning and sequence analyses revealed that the transposon was inserted within the coding region of the manA ( mannose phosphate isomerase ) disrupting its function . Accordingly , deletion of manA was sufficient to confer the observed defects and ectopic expression of manA fully complemented the mutant phenotype ( Figure S1A-S1C ) . manA encodes a conserved enzyme that catalyzes the reversible isomerization of fructose-6-phosphate ( Fru-6-P ) and mannose-6-phosphate ( Man-6-P ) [32] . Introducing point mutations into the predictable ManA active site abolished the ability of the protein to complement the null manA phenotype ( Materials and Methods; Figure S1D and S1E ) . Surprisingly , the manA mutant phenotype was displayed by cells grown in rich LB medium when mannose is not exploited as a carbon source . Nevertheless , ManA was found to be produced at significant levels under such conditions ( Figure S2 ) . Notably , the B . subtilis genome contains a homologue of manA , named pmi ( 56% identity ) , which encodes a second mannose phosphate isomerase . However , a strain bearing a knock out of pmi had no observable phenotype and unlike ManA , Pmi was undetectable in rich LB medium ( Figure S2 ) , suggesting that the two homologues have non-identical roles . Thus , besides its traditional role as a metabolic enzyme , ManA possesses an additional , previously unrevealed , crucial cellular activity that relies on its enzymatic activity . The observation that ΔmanA cells display altered nucleoid morphologies suggested that this mutant is perturbed in organizing and segregating the chromosomes . To examine in more detail the nature of ΔmanA nucleoids , we visualized the replication origin region using GFP fused to Spo0J , a protein that binds near the origins [33]–[35] . Wild type cells producing Spo0J-GFP were observed to contain 2–4 fluorescent foci ( Figure 2A and 2C ) , while the number of origins within the larger ΔmanA cells frequently exceeded 4 ( Figure 2B ) . Figure 2D exemplifies a characteristic ΔmanA cell containing as many as 12 copies of origin . Quantification analysis of the number of Spo0J-GFP foci per cell showed that the wild type population exhibited a narrow distribution with more than 99% of the cells containing 1–4 foci , whereas the number of foci in the mutant cells varied widely from 1 to 12 , with the majority of cells ( 55% ) bearing more than 4 foci per cell ( Figure 2G ) . The increased number of origin foci observed within ΔmanA cells raised two possibilities: either the cells are polyploids containing multiple chromosomal copies , or alternatively these mutant cells erroneously reinitiate replication without the actual completion of chromosome synthesis . To distinguish between these possibilities , we visualized the terminus , which is the last chromosomal region to be replicated . We generated wild type and ΔmanA strains that harbor TetR–GFP and carry repeated tetO units inserted in proximity to the terminus region [36] . Consistent with the idea that ΔmanA cells are polyploids , the ΔmanA strain exhibited more TetR–GFP foci than the typical 1–2 foci observed in the wild type strain ( Figure 2E and 2F ) . Nevertheless , to verify that the multiple copies of origin and terminus represent whole chromosomes , we applied DNA microarray to compare the DNA content of wild type and ΔmanA cells [37] ( Materials and Methods ) . Equal amounts of genomic DNA extracted from wild type and mutant cells were labeled and hybridized to a B . subtilis DNA chip . This analysis indicated no significant difference between the two samples implying that amplification of specific DNA regions is not a feature of ΔmanA cells . Thus , ΔmanA cells most likely contain several copies of fully replicated chromosomes within each cell . Further examination of the ΔmanA mutant strain revealed that the defects in cell shape and chromosome morphology are growth rate dependent . When growth was slowed , either by reducing the temperature or by growing ΔmanA cells in minimal medium , an almost normal phenotype was observed ( Figure 3A; Figure S3 ) . Accordingly , the levels of ManA production were significantly higher in rich medium ( Figure S3 ) . A prominent feature of rapid bacterial growth is the ability to perform multifork replication but still allocate the daughter chromosomes properly into progeny cells [38] . In light of the growth dependence of the ΔmanA phenotype , we speculated that ΔmanA cells are defective specifically in performing this complex task and as a consequence become polyploids . In order to test this premise , we investigated whether reducing the occurrence of multifork replication suppresses the ΔmanA phenotype . To that end , we combined the ΔmanA allele with a temperature sensitive allele of the replication initiation factor dnaB ( dnaBts ) [39] , [40] . Incubating this strain at the restrictive temperature leads to inactivation of DnaB and therefore reduces the rate of multifork replication . In line with our expectation , when the ΔmanA dnaBts strain was grown at the restrictive temperature ( 42°C ) the ΔmanA phenotype was partially suppressed , as manifested by the frequent appearance of almost wild type cell chains ( Figure 3C ) . Importantly , these “wild-type like” cells were absent when the ΔmanA dnaBts strain was incubated at the permissive temperature ( 37°C ) , or when ΔmanA strain was incubated at the restrictive temperature ( Figure 3B ) . Notably , by the time of transfer to the restrictive temperature the majority of the cells already acquired the ΔmanA phenotype and were polyploid , a phenotype which at some point may become irreversible . This occurrence may account for the partial nature of the suppression . To confirm that the appearance of “wild-type like” ΔmanA dnaBts cells correlates with reduced chromosomal copy number , we visualized the number of replication origins in the ΔmanA dnaBts strain by expressing Spo0J-GFP . Indeed , the number of Spo0J-GFP foci observed within the suppressed cells was similar to that in wild type cells ( 2–4 foci , Figure 3D and 3E ) . Thus , we conclude that ManA is crucial for proper DNA organization and segregation during rapid growth . As mentioned above , in association with abnormal nucleoids , ΔmanA cells display morphological defects reminiscent of the phenotypes exhibited by cell wall mutants ( e . g . : [6] , [13] , [15] , [18]–[21] , [29] , [30] ) . This observation raised the idea that ManA is required for cell wall construction . To explore this possibility we visualized the cell wall architecture of ΔmanA cells . Wild type and ΔmanA cells were labeled with a fluorophore ( fluorescein isothiocyanate - FITC ) conjugated to wheat germ agglutinin ( WGA ) , which is a carbohydrate binding protein that recognizes mainly GlcNAc . Living wild type cells stained with WGA-FITC exhibited bright midcell bands and fainter helical sidewall staining ( Materials and Methods; Figure 4A ) . This non-uniform pattern is similar to the patterns reported when fixed B . subtilis cells are stained with WGA [9] , [41] and when nascent PG is labeled with fluorescent vancomycin [5] . In contrast , living ΔmanA cells stained with WGA-FITC displayed a much more homogeneous pattern ( Figure 4B ) . The ΔmanA cell wall appeared significantly thicker than the wild type cell wall and lacked the characteristic helical staining indicative of nascent PG . Examining the effect of ManA on the subcellular localization of the two cell wall components Mbl and TagO corroborated these findings . Mbl that is typically localized in a helical pattern [15] became dispersed , while the localization of the membrane-associated TagO [27] was hardly modified ( Figure S4 ) . To substantiate that ManA is required for cell wall synthesis , we took advantage of the antibiotic tunicamycin to artificially interfere with cell wall construction and compare the resultant phenotypes with that of ΔmanA cells . Tunicamycin is a uridine nucleoside analog that specifically binds to and blocks the first membrane-associated step of both PG and WTA biosynthesis and thus , actively inhibits cell wall synthesis [42] , [43] . When wild type B . subtilis cells were grown in the presence of tunicamycin and subsequently visualized by fluorescence microscopy , a dramatic change in their cell shape was readily detected ( Figure S5 ) . Surprisingly , the tunicamycin treated cells resembled the ΔmanA cells not only in their cell shape but also in their nucleoid morphologies ( Figure 4D and 4E ) . Moreover , when wild type cells carrying the Spo0J-GFP fusion were treated with tunicamycin , a significant increase in the number of origins per cell was observed , indicating the transition to polyploidy ( Figure 4F and 4G ) . Consistent with the finding that tunicamycin reproduces characteristic ΔmanA phenotypes , wild type cells treated with tunicamycin and labeled with WGA-FITC displayed an architecture similar to the one exhibited by ΔmanA cells ( Figure 4C ) . Namely , the cell wall appeared thicker than usual and lacked typical sidewall cylindrical structures . Taken together , our data demonstrate that blocking cell wall synthesis per se by adding tunicamycin is sufficient to recapitulate both the cell shape and chromosome defects characteristic of ΔmanA cells . Therefore , we surmise that ManA plays a significant role in cell wall construction . Since ManA is primarily classified as a sugar metabolic enzyme , we reasoned that it could affect the carbohydrate composition of the cell wall . To test this possibility , cell walls of wild type and ΔmanA cells were isolated , hydrolyzed and their glycosyl composition determined using HPAEC neutral monosaccharide analysis ( Materials and Methods ) . The profile of wild type cell wall material was found to include high amounts of GlcNAc and Glc , medium levels of GalNAc and Gal , and relatively small quantities of Fucose ( Fuc ) ( Figure 4H ) . These carbohydrates , though in different proportions , have been reported as cell wall constituents of pathogenic Bacilli species ( i . e . : B . cereus , B . anthracis , and B . thuringiensis ) [44] , suggesting that these sugars are common to bacilli . The ΔmanA cell wall contained the same carbohydrates , however , a significant decrease in the amounts of Glc ( ∼4 fold ) and GalNAc ( ∼5 fold ) was monitored , with a milder decrease in the levels of Gal and Fuc ( 6-deoxy-L-Galactose ) ; these four carbohydrates are characteristic components of WTA [1] . In contrast , the major PG sugar GlcNAc was found to be present at the same level in wild type and ΔmanA cells . Thus , ManA is required for proper formation of WTA and to a lesser degree for PG synthesis . Nevertheless , the modified architecture of the PG observed in the absence of ManA ( Figure 4A and 4B ) implies a tight coordination between WTA and PG construction . To better understand the connection between cell wall integrity and chromosome morphology we followed both components simultaneously . We took advantage of the observation that ΔmanA cells exhibit almost normal phenotypes at a low temperature ( 23°C ) but gradually evidence defects when shifted to a high temperature ( 37°C ) . Cell wall and nucleoid morphologies were followed by WGA-FITC labeling and DAPI staining , respectively , as the ΔmanA cells were temperature shifted . In wild type cells the chromosome appeared mostly helical , exhibiting morphologies typical of replicating forms ( Figure 5A , 5C and 5E ) [45] . Notably , the helicity of the chromosome in the majority of the cells seemed to follow the sidewall staining of the cell wall . In a way , it seems that the cell wall restricts the nucleoid spatial localization by caging it . This feature was highlighted when stacks of optical sections were deconvolved , and three-dimensional structure was reconstructed ( Figure 5G and Video S1 ) . In comparison , the morphology of the nucleoids in ΔmanA cells altered distinctively upon temperature shift . At the lower temperature the nucleoids appeared similar to those displayed by wild type cells ( Figure 5B ) . However , after the temperature was shifted , concurrently with the cell bulging and loss of helical sidewall staining , the nucleoid lost its helical shape and became less compact and structured ( Figure 5D and 5F ) . Moreover , as rods became spheres , the nucleoid expanded in a pattern associated with the new cell geometry , as if constraints confining its structure were being released ( Figure 5H and Video S2 ) . It is possible that the incapability to confine the chromosome leads to defects in DNA segregation , ultimately resulting in the formation of polyploid cells . Consistent with a putative relationship between cell wall integrity and chromosome morphology , polyploidy was observed in L-form mutant cells , which completely lack a cell wall [46] , [47] . Additionally , an examination of chromosome copy number in the cell wall mutant mreB in E . coli revealed multiple chromosomal copies per cell [21] . In accord , B . subtilis cells harboring mutations in mreB or mbl , or wild type cells treated with tunicamycin showed multiple chromosomal copies ( Figure 4F and 4G; Figure S6 ) . Thus , an intimate connection between cell wall integrity and chromosome morphology exists . Perturbing this connection gives rise to the formation of polyploid cells .
Cell wall is responsible for shape determination and cellular viability for most bacterial species . Here we demonstrate that the carbohydrate metabolic enzyme ManA , which is conserved among prokaryotes and eukaryotes , participates in cell wall construction of the Gram positive bacterium B . subtilis . Several lines of evidence demonstrate a direct connection between ManA and cell wall synthesis: 1 ) the shape defect and perturbed cell wall architecture exhibited by ΔmanA cells , 2 ) the resemblance between the phenotype of ΔmanA cells and that of wild type cells treated with the cell wall synthesis inhibitor tunicamycin and , 3 ) the decreased amount of specific carbohydrates coating the cell surface observed in the absence of ManA . What could be the role of ManA in cell wall synthesis ? ManA is a cytoplasmic enzyme that catalyzes the reversible isomerization of Fru-6-P and Man-6-P [32] . Both products could affect directly cell wall synthesis . Man-6-P is a substrate for generating GDP-mannose , which is an important precursor of many nucleotide sugars , such as GDP-fucose [48] , whereas Fru-6-P is a component of a pathway that leads to the formation of the nucleotide sugar UDP-GlcNAc , basic for PG assembly [2] . Thus , the absence of ManA enzyme could interfere with the equilibrium of several pathways producing nucleotide sugar reservoirs for PG and WTA synthesis . Consistently , deleting the pgi gene encoding an enzyme that produces Fru-6-P [49] resulted in phenotypes similar to ΔmanA ( Figure S7 ) . This notwithstanding , according to our observations the absence of ManA has a specific effect on the composition of cell wall carbohydrates: it causes reduced abundance of Glc and GalNAc without perturbing the levels of the major PG precursor GlcNAc . Since , both Glc and GalNAc are components of the teichoic acid pathway [1] , our data suggest that ManA is specifically involved in the WTA synthesis rather than in the PG pathway . Moreover , although the overall composition of the PG components was unaffected , visualization of the PG revealed a modified architecture implying that the balance between PG and WTA is crucial for proper cell wall construction . Interestingly , in some cases mannose phosphate isomerases have been reported to affect cellular structures in other microorganisms . For example , the ManA enzyme of the bacterium Helicobacter pylori participates in capsular biosynthesis [48] , while the eukaryotic fungus Aspergillus fumigatus displays altered cell wall synthesis and morphogenesis under mannose starvation conditions [50] . We have previously shown that the bacterial nucleoid adopts a helical morphology during DNA replication [45] . Here we demonstrate that this nucleoid architecture is strictly dependent on cell wall integrity ( Figure 5; Video S1 and S2 ) . Interestingly , it has been proposed that the bacterial chromosome structure is largely affected by the transcription of rRNA operons and the transcription-translation-insertion ( transertion ) of membrane proteins that fasten the chromosome to the membrane [51] , [52] . Thus , it is conceivable that the chromosome is anchored to , and coordinated with both components cell wall and membrane . The association between DNA morphology and the cell wall is strongly manifested by the high chromosome copy number observed in L-form [46] , [47] , tunicamycin treated cells ( this work ) , mreB and mbl mutants [21] ( this work ) . We propose a model whereby the insertion of new cell wall material in a helical pattern dictates chromosome helicity ( Figure 6A ) . In wild type cells the nucleoid is attached to the newly synthesized cell wall ( see below ) and its organization and segregation are coordinated with cell wall synthesis and elongation . In the absence of ManA the normal extension of the cell wall is blocked , as indicated by the disappearance of helical sidewall staining . Consequently , the nucleoid is detached from cell wall components and the synchronization is lost between cell growth and DNA replication and segregation , resulting in the formation of polyploid cells . When detached , the nucleoid loses its compact helical structure and adopts a looser conformation that seems to follow the overall cellular geometry . It still remains to be resolved how the nucleoid is attached to the newly synthesized cell wall . Cell wall synthesis initiates on the cytoplasmic side of the membrane and then cell wall precursors are translocated to the outer membrane side [2] . This flip-flop reaction could be an active mechanism for coordinating DNA organization with cell wall elongation . We propose that the DNA is attached to the newly synthesized lipid intermediates on the cytoplasmic side of the membrane . When lipid II flips across the membrane the bound DNA is released , and the detached DNA is now free to bind a new cytoplasmic cell wall precursor ( Figure 6B ) . These attaching/detaching cycles serve to coordinate DNA replication , segregation and cell wall formation . When cell wall construction is blocked the amount of new cytoplasmic cell wall precursors is reduced . Therefore the detached chromosome falls off the membrane and the linkage between cell wall and DNA is lost . The interaction between the cell wall machinery and the DNA could be also mediated by linker membrane proteins that bind to both DNA and the cell wall . An interesting candidate proposed to possess such a capability is RodZ [53] , a bacterial cell morphogenesis protein identified recently in E . coli [54] , [55] , C . crescentus and B . subtilis [56] . RodZ contains a transmembrane domain and a cytosolic helix-turn-helix ( HTH ) DNA-binding motif and was shown to form helical structures and associate with MreB [54]–[56] . An additional membrane protein reported to colocalize with MreB and affect nucleoid morphology is SetB in E . coli [57] . Importantly , the primary role of SetB is metabolic , as it was shown to act as a lactose and glucose efflux transporter [58] . It is possible that metabolic proteins such as ManA and SetB through their role in cell wall synthesis operate as sensors that synchronize cell wall elongation with metabolite availability . Indeed , a similar ‘coupling’ function between cell mass and cell division has been attributed to the metabolic protein UgtP , a sugar transferase in B . subtilis that acts both in WTA biosynthesis and inhibits assembly of the cell division protein FtsZ [59] . The similarity between cell wall biosynthesis in bacteria and protein N-linked glycosylation in eukaryotes , a process which is utilized to determine the rate of protein folding [60] , [61] , raises the possibility that cell wall elongation serves to time cellular activities . In eukaryotes , N-linked glycan chains are added in the endoplasmic reticulum to growing nascent polypeptides and promote proper protein folding [62] . In both N-linked glycosylation and cell wall biosynthesis the initial step is transfer of a sugar nucleotide to a lipid carrier , undecaprenyl phosphate in bacteria and dolichol phosphate in eukaryotes [61] , reactions that are inhibited by the antibiotic tunicamycin [42] , [63] . During N-linked glycosylation the sugar chain , containing mainly mannose , is transferred from the lipid carrier to the unfolded protein . This sugar tree then undergoes cycles of cleavage and synthesis that serve as a timer for reporting the state of protein folding [64] . In a similar manner , we propose that the cell wall elongation rate in bacteria serves as a timer that coordinates cell growth with critical cellular activities such as chromosome segregation and cell division .
Plasmid and primers used for this study are described in Text S1 and Table S1 . B . subtilis strains were derivatives of the wild type strain PY79 [65] and are listed in Table S2 . All general methods were carried out as described previously [66] . Cultures were inoculated at 0 . 05 OD600 from an overnight culture and growth was carried out at 23°C , 37°C or 42°C , in LB medium or in S7 minimal medium [67] as indicated . Mini-Tn10 transposon was inserted into the B . subtilis ( PY79 ) chromosome as described previously [68] , [69] . Out of 2700 colonies that were screened , 260 that showed atypical colony morphologies ( i . e . : smooth colonies , transparent colonies ) and/or growth defect ( i . e . : small colonies ) were selected for further analysis . Using UniProtKB/Swiss-Prot ( http://www . uniprot . org/uniprot/O31646 ) , the active site of ManA was determined largely by comparison with the crystal-solved ortholog Pmi from Candida albicans . Accordingly , two amino acids were chosen to be examined: histidine 97 which is located within a Zinc ion binding site , and the highly conserved arginine 192 , which is predicted to be part of the catalytic domain . Fluorescence microscopy was carried out as described previously [70] . Briefly , samples ( 0 . 5 ml ) were taken during logarithmic phase , centrifuged and resuspended in 10 µl of PBS×1 supplemented with the fluorescent membrane stain FM1-43 or FM4-64 ( Molecular Probes , Invitrogen ) at 1 µg/ml and the DNA stain 4 , 6-Diamidino-2-phenylindole ( DAPI ) ( Sigma ) at 2 µg/ml . For cell wall labeling , cells were harvested , gently centrifuged , resuspended in 100 µl of T-Base×1 supplemented with WGA-FITC ( 5 µg/ml , Sigma ) , incubated for 15 min at room temperature , and washed twice with T-Base×1 before imaging . Stained cell walls and GFP fused proteins were visualized by placing the cells on thin T-Base×1 1% agarose pads . Cells were visualized and photographed using an Axioplan2 microscope ( Zeiss ) equipped with a CoolSnap HQ camera ( Photometrics , Roper Scientific ) or an Axioobserver Z1 microscope ( Zeiss ) equipped with a CoolSnap HQII camera ( Photometrics , Roper Scientific ) . System control and image processing were performed using MetaMorph software ( Molecular Devices ) . For deconvolution microscopy , samples of cells grown in rich LB medium were labeled with WGA-FITC ( 30 µg/ml ) and DAPI ( 2 µg/ml ) , applied to an agarose pad , and then subjected to deconvolution microscopy . Optical sections ( 20–40 ) were collected at a spacing of 0 . 2 µm . Images were deconvolved through 50 iterations and then visualized as SFP volume render by using the Huygens Professional Software ( Scientific Volume Imaging b . v . , Netherlands ) . Cell walls of wild type ( PY79 ) and ΔmanA ( ME37 ) cells ( triplicates of 250 ml late logarithmic phase cultures ) were isolated as described previously [44] . The analysis was performed by GlycoSolutions Corporation ( MA , USA ) . Briefly , each of the lyophilized samples was resuspended in 250 µl sterile double distilled water , rinsed with 250 µl of 4 M trifluoroacetic acid . Of note , by using this method acidic carbohydrates such as MurNAc cannot be indentified . After 3 hours of hydrolysis samples were examined using GlycoSolutions SOP HPAEC Neutral Monosaccharide Analysis . Equal volumes of each sample were injected at various dilutions: undiluted , x10 , x100 . ManNAc and glucose standard curves were included to normalize the values . DNA Microarray analysis was preformed to determine the DNA content of wild type ( PY79 ) and ΔmanA ( ME37 ) . Digested genomic DNA ( 0 . 5 µg , HaeIII ) of each strain was amplified by random primer and then labeled indirectly with cy3 or cy5 dyes . Equal amounts of the labeled samples were mixed and hybridized to a DNA chip containing Sigma B . subtilis OligoLibrary , which represents the entire B . subtilis ORFs . Arrays were scanned using a Genepix 4000B scanner ( Axon Ltd ) . Fluorescence intensities were quantitatively analyzed using GenePix Pro 4 . 1 software ( Axon ) . | The bacterial cell is resistant to extremes of osmotic pressure and protected against mechanical damages by the existence of a rigid outer shell defined as the cell wall . The strength of the cell wall is achieved by the presence of long glycan strands cross-linked by peptide side bridges . The cell wall is a dynamic structure continuously being synthesized and modified to allow for cell growth and division . Damaging the cell wall leads to abnormal cellular morphologies and cell lysis thereby making it a key target for antibiotics . Here we describe the involvement of ManA enzyme in cell wall construction of the rod shaped bacterium Bacillus subtilis . In cells lacking ManA the normal extension of the cell wall is blocked . Consequently , manA mutant cells display abnormal morphologies and fail to properly package and segregate their DNA resulting in the formation of large cells containing disorganized multiple chromosomal copies . Based on these findings , we propose that appropriate cell wall synthesis is necessary for synchronizing chromosome architecture with cell growth . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cell",
"biology/microbial",
"growth",
"and",
"development",
"cell",
"biology/morphogenesis",
"and",
"cell",
"biology",
"microbiology/microbial",
"growth",
"and",
"development"
] | 2010 | The Metabolic Enzyme ManA Reveals a Link between Cell Wall Integrity and Chromosome Morphology |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.